Haematologica, Volume 109, Issue 2 by Haematologica

VOL.

109 Journal of the Ferrata Storti Foundation

haematologica.org

FEBRUARY 2024

ISSN 0390 - 6078

Editor-in-Chief Jacob M. Rowe (Jerusalem)

Deputy Editors Carlo Balduini (Pavia), Jerry Radich (Seattle)

Associate Editors Shai Izraeli (Tel Aviv), Steve Lane (Brisbane), Pier Mannuccio Mannucci (Milan), Jessica Okosun (London), Pavan Reddy (Ann Arbor), David C. Rees (London), Paul G. Richardson (Boston), Francesco Rodeghiero (Vicenza), Gilles Salles (New York), Kerry Savage (Vancouver), Aaron Schimmer (Toronto), Richard F. Schlenk (Heidelberg)

Statistical Consultant Catherine Klersy (Pavia)

AI Consultant Jean Louis Raisaro (Lausanne)

Editorial Board Walter Ageno (Varese), Sarit Assouline (Montreal), Andrea Bacigalupo (Roma), Taman Bakchoul (Tübingen), Pablo Bartolucci (Créteil), Katherine Borden (Montreal), Marco Cattaneo (Milan), Corey Cutler (Boston), Kate Cwynarski (London), Ahmet Dogan (New York), Mary Eapen (Milwaukee), Francesca Gay (Torino), Ajay Gopal (Seattle), Alex Herrera (Duarte), Martin Kaiser (London), Marina Konopleva (Houston), Nicolaus Kröger (Hamburg), Austin Kulasekararaj (London), Shaji Kumar (Rochester), Ann LaCasce (Boston), Matthew J. Mauer (Rochester) Neha Mehta-Shah (St. Louis), Moshe Mittelman (Tel Aviv), Alison Moskowitz (New York), Yishai Ofran (Haifa), Farhad Ravandi (Houston), John W. Semple (Lund), Liran Shlush (Toronto), Sarah K. Tasian (Philadelphia), Pieter van Vlieberghe (Ghent), Ofir Wolach (Haifa), Loic Ysebaert (Toulouse)

Managing Director Antonio Majocchi (Pavia)

Editorial Office Lorella Ripari (Office & Peer Review Manager), Simona Giri (Production & Marketing Manager), Paola Cariati (Graphic Designer), Giulia Carlini (Graphic Designer), Debora Moscatelli (Graphic Designer), Igor Poletti (Graphic Designer), Diana Serena Ravera (Peer Review), Laura Sterza (Account Administrator)

Assistant Editors Britta Dost (English Editor), Rachel Stenner (English Editor), Anne Freckleton (English Editor), Rosangela Invernizzi (Scientific Consultant), Marianna Rossi (Scientific Consultant), Massimo Senna (Information Technology), Luk Cox (Graphic Artist)

Haematologica | 109 February 2024

Brief information on Haematologica

Haematologica (print edition, pISSN 0390-6078, eISSN 1592-8721) publishes peer-reviewed papers on all areas of experimental and clinical hematology. The journal is owned by a non-profit organization, the Ferrata Storti Foundation, and serves the scientific community following the recommendations of the World Association of Medical Editors (www. wame.org) and the International Committee of Medical Journal Editors (www.icmje.org). Haematologica publishes Editorials, Original articles, Review articles, Perspective articles, Editorials, Guideline articles, Letters to the Editor, Case reports & Case series and Comments. Manuscripts should be prepared according to our guidelines (www.haematologica.org/information-for-authors), and the Uniform Requirements for Manuscripts Submitted to Biomedical Journals, prepared by the International Committee of Medical Journal Editors (www.icmje.org). Manuscripts should be submitted online at http://www.haematologica.org/. Conflict of interests. According to the International Committee of Medical Journal Editors (http://www.icmje. org/#conflicts), “Public trust in the peer review process and the credibility of published articles depend in part on how well conflict of interest is handled during writing, peer review, and editorial decision making”. The ad hoc journal’s policy is reported in detail at www.haematologica.org/content/policies. Transfer of Copyright and Permission to Reproduce Parts of Published Papers. Authors will grant copyright of their articles to the Ferrata Storti Foundation. No formal permission will be required to reproduce parts (tables or illustrations) of published papers, provided the source is quoted appropriately and reproduction has no commercial intent. Reproductions with commercial intent will require written permission and payment of royalties. Subscription. Detailed information about subscriptions is available at www.haematologica.org. Haematologica is an open access journal and access to the online journal is free. For subscriptions to the printed issue of the journal, please contact: Haematologica Office, via Giuseppe Belli 4, 27100 Pavia, Italy (phone +39.0382.27129, fax +39.0382.394705, E-mail: info@haematologica.org). Rates of the printed edition for the year 2022 are as following: Institutional: Euro 700 Personal: Euro 170 Advertisements. Contact the Advertising Manager, Haematologica Office, via Giuseppe Belli 4, 27100 Pavia, Italy (phone +39.0382.27129, fax +39.0382.394705, e-mail: marketing@haematologica.org). Disclaimer. Whilst every effort is made by the publishers and the editorial board to see that no inaccurate or misleading data, opinion or statement appears in this journal, they wish to make it clear that the data and opinions appearing in the articles or advertisements herein are the responsibility of the contributor or advisor concerned. Accordingly, the publisher, the editorial board and their respective employees, officers and agents accept no liability whatsoever for the consequences of any inaccurate or misleading data, opinion or statement. Whilst all due care is taken to ensure that drug doses and other quantities are presented accurately, readers are advised that new methods and techniques involving drug usage, and described within this journal, should only be followed in conjunction with the drug manufacturer’s own published literature.

Direttore responsabile: Prof. Carlo Balduini; Autorizzazione del Tribunale di Pavia n. 63 del 5 marzo 1955. Printing: Press Up, zona Via Cassia Km 36, 300 Zona Ind.le Settevene - 01036 Nepi (VT)

Associated with USPI, Unione Stampa Periodica Italiana. Premiato per l’alto valore culturale dal Ministero dei Beni Culturali ed Ambientali

Haematologica | 109 February 2024

Table of Contents Volume 109, Issue 2: February 2024

About the Cover Image taken from the editorial by Ralf Küppers in this issue.

Landmark Papers in Hematology 367

Evolution in treatment of acute promyelocytic leukemia: a major victory M.S. Tallman https://doi:10.3324/haematol.2023.284421

369

The splendor and the tyranny of JAK inhibition E.O. Hexner https://doi.org/10.3324/haematol.2023.283545

371

Molecular measurable residual disease: staring at red herrings A.C. Winters and D.A. Pollyea https://doi.org/10.3324/haematol.2023.283708

374

Expanding treatment options by selectively targeting the cytokine storm with ruxolitinib in primary hemophagocytic lymphohistiocytosis J.A.M. van Laar https://doi.org/10.3324/haematol.2023.283915

376

Distinct t(14;19) translocation patterns in atypical chronic lymphocytic leukemia and marginal zone lymphomas R. Küppers https://doi.org/10.3324/haematol.2023.283975

Editorials

Spotlight Reviews 379

Advances in next-generation sequencing and emerging technologies for hematologic malignancies R. Kwon and C.C.S. Yeung https://doi.org/10.3324/haematol.2022.282442

Review Articles 388

Diffuse large B-cell lymphoma involving the central nervous system: biologic rationale for targeted therapy M. Roschewski and D.J. Hodson https://doi.org/10.3324/haematol.2021.278613

Haematologica | 109 February 2024

Articles 401

411

422

431

444

458

466

479

493

509

Acute Myeloid Leukemia Quantification of measurable residual disease using duplex sequencing in adults with acute myeloid leukemia L.W. Dillon et al. https://doi.org/10.3324/haematol.2023.283520 Acute Myeloid Leukemia miR-30e-5p regulates leukemia stem cell self-renewal through the Cyb561/ROS signaling pathway Y. Ge et al. https://doi.org/10.3324/haematol.2023.282837 Bone Marrow Failure Spontaneous remission and loss of monosomy 7: a window of opportunity for young children with SAMD9L syndrome M. Erlacher et al. https://doi.org/10.3324/haematol.2023.283591 Cell Therapy & Immunotherapy Allogeneic hematopoietic stem cell transplantation in patients aged 60-79 years in Germany (19982018): a registry study J. Frederic Weller et al. https://doi.org/10.3324/haematol.2023.283175 Cell Therapy & Immunotherapy Regulatory T-cell dysfunctions are associated with increase in tumor necrosis factor α in autoimmune hemolytic anemia and participate in Th17 polarization M. Ciudad et al. https://doi.org/10.3324/haematol.2023.282859 Cell Therapy & Immunotherapy Ruxolitinib-based regimen in children with primary hemophagocytic lymphohistiocytosis J. Ge et al. https://doi.org/10.3324/haematol.2023.283478 Chronic Lymphocytic Leukemia IgH 3’RR recombination uncovers a non-germinal center imprint and c-MYC-dependent IgH rearrangement in unmutated chronic lymphocytic leukemia I. Al Jamal et al. https://doi.org/10.3324/haematol.2023.282897 Chronic Lymphocytic Leukemia Global miRNA profiling reveals key molecules that contribute to different chronic lymphocytic leukemia incidences in Asian and Western populations P. Liu et al. https://doi.org/10.3324/haematol.2023.283181 Chronic Lymphocytic Leukemia BCL3 rearrangements in B-cell lymphoid neoplasms occur in two breakpoint clusters associated with different diseases A. Carbó-Meix et al. https://doi.org/10.3324/haematol.2023.283209 Hematopoiesis Prognostic relevance of clonal hematopoiesis in myeloid neoplastic transformation in patients with follicular lymphoma treated with radioimmunotherapy Z. Xie et al. https://doi.org/10.3324/haematol.2023.283727

Haematologica | 109 February 2024

521

533

543

553

567

578

591

604

Myeloproliferative Disorders Phenotypic profiling of CD34+ cells by advanced flow cytometry improves diagnosis of juvenile myelomonocytic leukemia C. Bugarin et al. https://doi.org/10.3324/haematol.2023.282805 Non-Hodgkin Lymphoma Results from a phase I trial of pembrolizumab plus vorinostat in relapsed/refractory B-cell nonHodgkin lymphoma J. Godfrey et al. https://doi.org/10.3324/haematol.2023.283002 Non-Hodgkin Lymphoma Feasibility and outcomes after dose reduction of immunochemotherapy in young adults with Burkitt lymphoma and leukemia: results of the BURKIMAB14 trial J.M. Ribera et al. https://doi.org/10.3324/haematol.2023.283342 Non-Hodgkin Lymphoma Tafasitamab for patients with relapsed or refractory diffuse large B-cell lymphoma: final 5-year efficacy and safety findings in the phase II L-MIND study J. Duell et al. https://doi.org/10.3324/haematol.2023.283480 Plasma Cell Disorders PVT1 interacts with polycomb repressive complex 2 to suppress genomic regions with pro-apoptotic and tumor suppressor functions in multiple myeloma P. Nylund et al. https://doi.org/10.3324/haematol.2023.282965 Plasma Cell Disorders TTK/MPS1 inhibitor OSU-13 targets the mitotic checkpoint and is a potential therapeutic strategy for myeloma L. Valle Guilhen Longo et al. https://doi.org/10.3324/haematol.2023.282838 Plasma Cell Disorders Minor clone of del(17p) provides a reservoir for relapse in multiple myeloma J. Cui et al. https://doi.org/10.3324/haematol.2023.283533 Plasma Cell Disorders Isatuximab plus carfilzomib and dexamethasone in patients with early versus late relapsed multiple myeloma: IKEMA subgroup analysis T. Facon et al. https://doi.org/10.3324/haematol.2023.283073

Letters to the Editor 617

Efficacy and feasibility of pharmacoscopy-guided treatment for acute myeloid leukemia patients who have exhausted all registered therapeutic options J.A. Schmid et al. https://doi.org/10.3324/haematol.2023.283224

622

Vaccine utilization and overwhelming post-splenectomy infection risk factors in two asplenia cohorts M.A. Soderstrom et al. https://doi.org/10.3324/haematol.2023.283419

627

Identification of PSMB4 and PSMD4 as novel target genes correlated with 1q21 amplification in patients with smoldering myeloma and multiple myeloma J. Burroughs Garcia et al. https://doi.org/10.3324/haematol.2023.283200 Haematologica | 109 February 2024

III

632

Quantitative evaluation of the clinical severity of hemoglobin H disease in a cohort of 591 patients using a scoring system based on regression analysis Y. Liu et al. https://doi.org/10.3324/haematol.2023.283211

639

Thiostrepton induces cell death of acute myeloid leukemia blasts and the associated macrophage population I. Weinhäuser et al. https://doi.org/10.3324/haematol.2023.283621

646

Systemic and mucosal adaptive immunity to SARS-CoV-2 during the Omicron wave in patients with chronic lymphocytic leukemia H.M. Ingelman-Sundberg et al. https://doi.org/10.3324/haematol.2023.282894

652

Prognostic value of minimal disseminated disease assessed using digital polymerase chain reaction for 3' ALK assays in pediatric anaplastic lymphoma kinase-positive anaplastic large cell lymphoma R. Fukano et al. https://doi.org/10.3324/haematol.2023.282812

657

Uniform conditioning regardless of donor in bone marrow transplantation for severe aplastic anemia A.E. DeZern et al. https://doi.org/10.3324/haematol.2023.284022

661

Frequent, high density expression of surface CD38 as a potential therapeutic target in adult T-lineage acute lymphoblastic leukemia S. Koslowski et al. https://doi.org/10.3324/haematol.2023.283814

666

Oligosecretory Waldenström macroglobulinemia exhibits excellent treatment response and outcomes W. Xiong et al. https://doi.org/10.3324/haematol.2023.283402

671

676

Clinical outcomes of patients with myelofibrosis after immediate transition to momelotinib from ruxolitinib R. Mesa et al. https://doi.org/10.3324/haematol.2023.283106

Case Reports & Case Series 682

Immune-mediated facial nerve paralysis in a myeloma patient post B-cell maturation antigen-targeted chimeric antigen receptor T cells Y.K. Kathari et al. https://doi.org/10.3324/haematol.2023.283296

689

693

Erratum to: Introduction to the peripheral T-cell lymphoma review series: advances in molecular characterization, classification refinement and treatment optimization K.J. Savage and L. de Leval https://doi.org/10.3324/haematol.2024.285005

Errata

Haematologica | 109 February 2024

LANDMARK PAPER IN HEMATOLOGY

M.S. Tallman

Evolution in treatment of acute promyelocytic leukemia: a major victory Martin S. Tallman Chicago, IL, USA E-mail: martintallman1@gmail.com https://doi.org/10.3324/haematol.2023.284421 ©2024 Ferrata Storti Foundation Published under a CC BY-NC license

TITLE

Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia.

AUTHORS Huang M-E, Ye Y-C, Chen S-R, et al. JOURNAL Blood. 1988;72(2):567-572. PMID: 3165295. Acute promyelocytic leukemia (APL) is the most curable subtype of acute myeloid leukemia (AML) in adults. The leukemia cells express the t(15;17)(q24;q21) leading to the formation of the PML::RARa fusion transcript which inhibits differentiation of the leukemia cells. During the last 50 years, treatment has progressively evolved. Approximately 98% of low-risk patients (presenting white blood cell count [WBC] </= 10x109/L) and > 90% of high-risk patients (WBC > 10x109/L) who survive induction are cured of their disease. The first important observation addressing treatment was made in 1973 when Bernard and colleagues reported unusual sensitivity of the leukemia cells to single agent daunorubicin which induced complete re-

mission (CR) in 55% of patients with a longer duration of CR (median 26 months) compared to that among other subtypes of AML (median 7 months).1 However, in a landmark publication, the most important breakthrough came in 1988 when Huang and co-workers reported stunning results among patients treated with the oral vitamin derivative all-trans retinoic acid (ATRA), an agent known to induce differentiation of leukemia cells in vitro.2 Of 24 patients (8 previously treated, 16 untreated), 100% achieved CR, and remarkably, without obligatory marrow aplasia. Furthermore, cultured leukemic promyelocytes showed morphological evidence suggestive of progressive differentiation (Figure 1). Four of 6 patients maintained

Figure 1. Morphological maturation of leukemic cells of case N. 10 in vitro and in vivo. (A) Cells cultured without retinoic acid (RA), consisting of promyelocytes with characteristic cytoplasmic granules (x1,000). (B) Cells cultured with RA, showing maturation to granulocytes (x1,000). (C) Bone marrow before RA treatment. The predominance of promyelocytes (76%) indicates typical acute promyelocytic leukemia. (D) Bone marrow after five weeks of RA treatment. Promyelocyte level <2% and restoration of normal hematopoiesis without a phase of aplasia are consistent with differentiation induction.2 (Figure from Huang et al.2 Reproduced under an Elsevier user license.) Haematologica | Vol 109 - February 2024

367

LANDMARK PAPER IN HEMATOLOGY

M.S. Tallman

with ATRA alone remained in CR for 5-10 months. It was subsequently demonstrated that the mature neutrophils retained the t(15;17)(q24;q21), confirming that the mechanism of successful induction is differentiation. The introduction of ATRA represents the major turning point towards the curative treatment of APL, and also ushered in the era of effective molecularly-targeted therapy for hematologic malignancies. Given the observations of Bernard and colleagues, it was logical to combine ATRA with chemotherapy, particularly anthracyclines. Cooperative groups conducted trials of ATRA combined with either single agent anthracyclines or combination chemotherapy, yielding outstanding results. In these studies, the relapsefree survival at three years is approximately 90%. This approach established a new standard of care until a second major breakthrough came with the identification of the effectiveness of arsenic trioxide (ATO), shown to induce apoptosis with elimination of the PML-RARa fusion transcript. Lo Coco and colleagues for the GIMEMA established a new standard of care for low-risk patients combining ATRA plus ATO without chemotherapy in induction (except hydroxyurea to control the WBC), consolidation or maintenance. In this randomized trial, CR was achieved in 100% in patients treated with ATRA plus ATO and 95% in the patients treated with ATRA plus chemo-

therapy.3 The 2-year event-free survival among the patients treated with ATRA plus ATO was 97% and 86% among patients treated with ATRA plus chemotherapy. The Australasian Leukemia Lymphoma Group included highrisk patients in a phase II trial of induction with ATRA, idarubicin and ATO, ATRA plus ATO for 2 cycles of consolidation and two years of maintenance with ATRA, 6-mercaptopurine and methotrexate in previously untreated patients.4 The 2-year freedom from relapse, failurefree survival, and overall survival were 97.5%, 88.1%, and 93.2%, respectively. The last frontier in treatment may well be incorporation of an oral formulation of arsenic. In a successful phase III non-inferiority trial, Zhu and colleagues randomized newly diagnosed patients to either an oral tetra-arsenic tetra sulfide-containing formulation plus ATRA or ATO plus ATRA for induction.5 Subsequently, all patients were given 3 cycles of chemotherapy consolidation followed by maintenance. The major victory is that the majority of patients with APL can now be cured without chemotherapy since 75% of patients present with low-risk disease. Disclosures MST has received research funding from and declares advisory board participation for Orsenix.

References 1. Bernard J, Weil M, Boiron M, et al. Acute promyelocytic leukemia: results of treatment by daunorubicin. Blood. 1973;41(4):489-496. 2. Huang M-E, Ye Y-C, Chen S-R, et al. Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia. Blood. 1988;72(2):567-572. 3. Lo-Coco F, Avvisati G, Vignetti, et al. Retinoic acid and arsenic trioxide for acute promyelocytic leukemia. N Engl J Med. 2013;369(2):111-121.

4. Iland HJ, Bradstock K, Supple SG, et al. All-trans-retinoic acid, idarubicin, and IV arsenic trioxide as initial therapy in acute promyelocytic leukemia (APML4). Blood. 2012;120(8):1570-1580; quiz 1752. 5. Zhu H-H, Wu D-P, Jin J, et al. Oral tetra-arsenic tetra sulfide formula versus intravenous arsenic trioxide as first-line treatment of acute promyelocytic leukemia: a multicenter randomized controlled trial. J Clin Oncol. 2013;31(33): 4215-4221.

Haematologica | Vol 109 - February 2024

368

EDITORIAL

The splendor and the tyranny of JAK inhibition Elizabeth O. Hexner

Correspondence: E.O. Hexner

Abramson Cancer Center and Perelman School of Medicine, University of Pennsylvania,

elizabeth.hexner@pennmedicine.upenn.edu Received: Accepted: Early view:

Philadelphia, PA, USA

September 20, 2023. October 2, 2023. October 12, 2023.

If you have ever treated an ill patient with myelofibrosis with a JAK inhibitor and seen them restored, Lazarus-like, to a prior version of themselves, then you know the splendor of JAK inhibitors. Picture a patient debilitated by malaise, wasting, and a massively enlarged and symptomatic spleen; within days a cytokine-mediated fog may clear, and within weeks, measurable changes in weight and spleen size are common. This is the gratifying experience of JAK inhibitors, of which there are now four FDA-approved agents for the treatment of myelofibrosis, with momelotinib as the most recently approved in the United States.1 Each of these small molecules potently inhibits JAK2, and reduces spleen size and symptoms. They are distinguished from one another by their inhibitory activity against other kinases, and their overall safety/tolerability profile. The enthusiasm around momelotinib is related to a secondary target, ACVR1 (activin A receptor type 1). ACVR1 and its downstream pathways control iron homeostasis, including regulating hepcidin levels, which are markedly elevated in myelofibrosis and in the anemia of chronic inflammation, leading to an iron-restricted anemia. Thus, ACVR1 inhibition with momelotinib can, in some cases, improve anemia in patients with myelofibrosis or may, at a minimum, offset treatment-emergent anemia related to on target JAK2 inhibition of normal erythropoiesis.2 The effectiveness, the splendor, of JAK inhibitors is meaningful and may endure for years, though myelofibrosis inexorably progresses unless a patient undergoes allogeneic stem cell transplantation. Unlike small molecules that inhibit other drivers of myeloid diseases (e.g., ABL, FLT3, IDH2), available JAK inhibitors do not yet selectively target the mutant clone, regardless of driver, including JAK2V617F. Their benefit is largely thought to be driven by modification of the inflammatory cytokine milieu. Enter the tyranny of JAK inhibitors. The tyranny of JAK inhibitor treatment is not subtle: abrupt cessation can result in a life-threatening cytokine-driven sepsis-like syndrome, including persistent high fevers, respiratory distress, or hemodynamic instability. More commonly, even when the drug is intentionally tapered, severe pain/ rapid spleen enlargement, fevers and other symptoms may

ensue. Even at stable doses, over time, spleen responses may be lost. Cytopenias may worsen. The fog rolls back in. This rapid return of symptoms has posed challenges in clinical practice and clinical trials. The natural response to a patient progressing on one therapy is to change to another. But even in the setting of clear progression/loss of response, stopping a JAK inhibitor is fraught. In clinical trials, the conventional ‘washout’ period from a prior JAK inhibitor can render a patient so ill that they are no longer eligible for the study. Combinatorial trials of JAK inhibitors with other therapies are common now: the combinations may make biologic/translational sense, but they have the added feasibility benefit of not requiring cessation of JAK inhibitors. But how unsafe is transitioning from one JAK inhibitor to another? In this issue,3 Mesa et al. performed a posthoc analysis of the SIMPLIFY-1 trial,4 the trial randomizing JAK-inhibitor naïve patients to ruxolitinib versus momelotinib for 24 weeks followed by an open label phase where all ruxolitinib patients stopped treatment and crossed over to receive momelotinib, while momelotinib patients continued on it. This study answers an important safety question around transitioning directly and immediately from one JAK inhibitor to another. In this large study of 432 subjects, 197 subjects transitioned from ruxolitinib to momelotinib. Overall, this analysis found that there was a smooth and safe transition without “ruxolitinib discontinuation syndrome”, defined as respiratory distress, shock, worsening of cytopenias or return of spleen-related symptoms. In practice, this should offer evidence and reassurance that a direct, immediate transition from ruxolitinib or other JAK inhibitor to momelotinib will be safe. Disclosures EOH has received research funding from Blueprint Medicines and Abbvie (for clinical trials); has sat on the Tmunity Therapeutics Advisory Board; and has had a consultancy role for the American Board of Internal Medicine, Blueprint Medicines, Pharmassentia, and Abbvie.

Haematologica | 109 Mese 2024

369

E.O. Hexner

EDITORIAL

References 1. FDA Roundup: September 19, 2023. 2023. https://www.fda.gov/ news-events/press-announcements/fda-roundupseptember-19-2023. 2. Oh ST, Talpaz M, Gerds AT, et al. ACVR1/JAK1/JAK2 inhibitor momelotinib reverses transfusion dependency and suppresses hepcidin in myelofibrosis phase 2 trial. Blood Adv. 2020;4(18):4282-4291. 3. Mesa R, Verstovsek S, Platzbecker U, et al. Clinical outcomes of

patients with myelofibrosis after immediate transition to momelotinib from ruxolitinib. Haematologica. 2024:109(2):676-681. 4. Mesa RA, Kiladjian JJ, Catalano JV, et al. SIMPLIFY-1: a phase III randomized trial of momelotinib versus ruxolitinib in Janus kinase inhibitor-naive patients with myelofibrosis. J Clin Oncol. 2017;35(34):3844-3850.

Haematologica | 109 Mese 2024

370

EDITORIAL

A.C. Winters and D.A. Pollyea

Molecular measurable residual disease: staring at red herrings Amanda C. Winters1 and Daniel A. Pollyea2 1

University of Colorado Department of Pediatrics, Center for Cancer and Blood Disorders, Children's Hospital Colorado and 2University of Colorado Division of Hematology, Department of Medicine, Aurora, CO, USA

Correspondence: D.A. Pollyea daniel.pollyea@ucdenver.edu Received: Accepted: Early view:

August 4, 2023. August 9, 2023. August 17, 2023.

Red herrings, misleading or distracting clues, have been utilized in some of the most famous works of literature and films to maximum dramatic effect. The term is thought to have originated from the practice of using the pungent odor of cured fish to distract young hunting hounds in training; the more they learned to ignore the stench of the herring (which turned red in the process of being smoked), the better they were able to hone their focus on the scent of their prey. The presence of measurable residual disease (MRD) in patients achieving morphological remission after treatment is one of the most powerful predictors of acute myeloid leukemia (AML) relapse;1 it is not at all a red herring. Although conventionally performed via multiparameter flow cytometry, MRD can also be assessed through a variety of high-sensitivity molecular tools to investigate the allelic frequency of AML-associated genes.2 With respect to this so-called “molecular MRD,” given its potential as a more sensitive measure of residual disease, we must grapple with the question of when an assay’s depth is too deep, and when detectable gene mutations might not herald looming relapse but instead represent red herrings. Previous literature has provided strong evidence that clonal hematopoiesis mutations DNMT3A, TET2, and ASXL1 (“DTA” mutations, or, we might propose, “RH” mutations) may persist after therapy and are not associated with increased relapse risk.3 In the current issue of Haematologica, Murphy and colleagues describe an unbiased mathematical approach to evaluating the contribution of individual genes toward the predictive value of MRD. In their letter entitled, “Exclusion of persistent mutations in splicing factor genes and isocitrate dehydrogenase 2 improves the prognostic power of molecular measurable residual disease assessment in acute myeloid leukemia,” this group evaluated persistence of mutations in 22 AMLassociated genes in remission samples from 101 patients who received standard cytotoxic chemotherapy for newly diagnosed AML.4 In most cases two separate remission samples were evaluated with error-corrected next-gen-

eration sequencing. The authors used a conservative mutant allelic frequency (MAF) cutoff of 1% to categorize patients as MRD-positive or MRD-negative. They then systematically excluded individual genes from the MRD analysis within the cohort, yielding 2,500 permutations of MRD for which the hazard ratio for overall survival was calculated and compared. Their conclusions were that, in addition to DTA mutations, exclusion of splicing factor mutations (SRSF2, U2AF1, and SF3B1) and IDH2 enhanced the predictive value of MRD for overall survival as well as relapse-free survival and cumulative incidence of relapse. They went on to validate these findings in two historic cohorts of patients for whom next-generation sequencing MRD data were available, showing that removal of “DTASI2” mutations from MRD evaluation enhanced the prognostic value of the assay. The approach taken by the investigators is novel and does attempt to mitigate bias inherent in much of the existing molecular MRD literature. The clinical outcome is the true yardstick of a gene’s value for MRD, if molecular MRD is being utilized as a purely clinical assay without reference to its research value in imputing clonal dynamics of disease. It is interesting that the authors chose overall survival as their endpoint, rather than relapse-free survival, since residual disease is by definition a predictor of relapse, whereas the contributors to overall survival are multifactorial in the adult population. The 1% MAF cutoff also raises questions about the validity of the findings at lower thresholds such as 0.1% or 0.02%, which are more commonly used clinically as positive/negative cutoffs. It is true that the accepted thresholds for next-generation sequencing MRD have yet to be established2 and using a higher MAF burden for thresholding is more likely to capture more proximal survival events. However, it is possible that a lower MAF threshold would allow for even better discrimination between outcomes. The removal of splicing factor mutations as a class from consideration for AML MRD is supported by prior studies demonstrating their association with pre-leukemic mar-

Haematologica | 109 - February 2024

371

EDITORIAL

A.C. Winters and D.A. Pollyea

Figure 1. A fishy story. According to Murphy et al.4, inclusion of

DNMT3A, TET2, ASXL1, splicing factor genes (SRSF2, SF3B1, U2AF1), and IDH2 mutations (DTASI2) in measurable residual disease quantitation for acute myeloid leukemia may worsen the prognostic value of measurable residual disease in this context. DTASI2 mutations may be akin to red herrings that distract or mislead clinicians in their assessments of true residual disease in acute myeloid leukemia.

row disease, particularly myelodysplastic syndromes. Their persistence after both conventional chemotherapy and epigenetic agents has also been described and was not associated with inferior survival.5,6 Therefore, the present findings add further credence to their exclusion from molecular MRD assessments. Similarly, DNMT3A and TET2 are again confirmed to lessen the predictive value of molecular MRD by their exclusion, although it is interesting to note that ASXL1 was not among the genes highlighted in Figure 1A or 1B of Murphy’s publication as worthy of exclusion in the mathematical modeling, but was excluded nonetheless by convention.2,3 The exclusion of IDH2 is more controversial. While the authors show optimal hazard ratios for overall survival with “DTASI2” genes excluded, they do not directly compare these hazard ratios

to “DTAS” alone to show specifically that the exclusion of IDH2 enhances the prognostic value of molecular MRD. Furthermore, based on the heatmap in Figure 1A of the letter by Murphy et al., there is not only a cluster of IDH2 exclusion at the high hazard ratio end of the ranked permutations, but another cluster at the low hazard ratio end as well. This pattern is not seen with DNMT3A, TET2, or splicing factor mutations – or indeed any other gene in the panel. This may reflect different contributions of IDH2 to clonal evolution in individual patients. While mutations in IDH2 are known to be necessary but not sufficient for leukemic transformation in preclinical models7 and have also been described as early mutations in myelodysplastic syndromes and pre-leukemic myeloproliferative 6,8 disorders, there are numerous reports of IDH2 being used to successfully monitor MRD.9 It may be that IDH2 is a founder event in some patients, and therefore analogous to clonal hematopoiesis or splicing factor mutations in its lack of prognostic value for relapse,6,10 whereas in other instances of AML it is a later mutation, and therefore still useful for MRD monitoring. Additional studies will be necessary to reproduce the current findings in larger cohorts, paying particular attention to co-mutations and putative clonal evolution in individual patients. Despite these caveats, the authors are to be commended for their a priori approach to mutation evaluation for MRD relevance in AML patients treated with conventional induction therapies. In addition to prospective validation of these findings in a similar context, ongoing work should evaluate the utility of individual genes for molecular MRD monitoring after low-intensity therapies such as venetoclax-based regimens, as these therapies are gaining ground in particular AML populations, but very little is understood about their effect on clonal dynamics or molecular MRD. As the field gets closer to adoption of genebased MRD for clinical decision-making, we will need stringent systems in place to filter out gene mutations whose persistence smells fishy. Disclosures No conflicts of interests to disclose. Contributions ACW and DAP wrote and edited this editorial.

References 1. Short NJ, Zhou S, Fu C, et al. Association of measurable residual disease with survival outcomes in patients with acute myeloid leukemia: a systematic review and meta-analysis. JAMA Oncol. 2020;6(12):1890-1899. 2. Heuser M, Freeman SD, Ossenkoppele GJ, et al. 2021 update on MRD in acute myeloid leukemia: a consensus document from the European LeukemiaNet MRD Working Party. Blood. 2021;138(26):2753-2767. 3. Jongen-Lavrencic M, Grob T, Hanekamp D, et al. Molecular

minimal residual disease in acute myeloid leukemia. N Engl J Med. 2018;378(13):1189-1199. 4. Murphy T, Zou J, Arruda, et al. Exclusion of persistent mutations in splicing factor genes and isocitrate dehydrogenase 2 improves the prognostic power of molecular measurable residual disease assessment in acute myeloid leukemia. Haematologica. 2024;109(2):671-675. 5. Lindsley RC, Mar BG, Mazzola E, et al. Acute myeloid leukemia ontogeny is defined by distinct somatic mutations. Blood.

Haematologica | 109 - February 2024

372

EDITORIAL

A.C. Winters and D.A. Pollyea

2015;125(9):1367-1376. 6. Hasserjian RP, Steensma DP, Graubert TA, Ebert BL. Clonal hematopoiesis and measurable residual disease assessment in acute myeloid leukemia. Blood. 2020;135(20):1729-1738. 7. Kats LM, Reschke M, Taulli R, et al. Proto-oncogenic role of mutant IDH2 in leukemia initiation and maintenance. Cell Stem Cell. 2014;14(3):329-341. 8. Stengel A, Baer C, Walter W, et al. Mutational patterns and their

correlation to CHIP-related mutations and age in hematological malignancies. Blood Adv. 2021;5(21):4426-4434. 9. Ok CY, Loghavi S, Sui D, et al. Persistent IDH1/2 mutations in remission can predict relapse in patients with acute myeloid leukemia. Haematologica. 2019;104(2):305-311. 10. Ivey A, Hills RK, Simpson MA, et al. Assessment of minimal residual disease in standard-risk AML. N Engl J Med. 2016;374(5):422-433.

Haematologica | 109 - February 2024

373

EDITORIAL

Expanding treatment options by selectively targeting the cytokine storm with ruxolitinib in primary hemophagocytic lymphohistiocytosis Jan A.M. van Laar Section of Clinical Immunology, Departments of Internal Medicine and Immunology, Erasmus University Medical Center Rotterdam, Rotterdam, the Netherlands

Correspondence: J.A.M. van Laar j.vanlaar@erasmusmc.nl Received: Accepted: Early view:

August 16, 2023. August 30, 2023. September 7, 2023.

Hemophagocytic lymphohistiocytosis (HLH) is a life-threatening hyperinflammatory syndrome characterized by uncontrolled activated cells of the lymphoid and mononuclear phagocytic system (MPS), leading to excessive cytokine production and systemic tissue and organ damage. Primary HLH patients require hematopoieitic stem cell transplantation (HSCT) of which survival figures depend on prior disease control. Despite advancements in its understanding and management, HLH remains a challenging condition with a significant mortality rate. However, recent clinical data has shed light on a potential breakthrough using the Janus kinase (JAK) inhibitor, ruxolitinib. The immunopathophysiology of HLH is fascinating. Once triggered, normally self-regulating histiocytes remain activated and cause a cascade of inflammatory events eventually leading to an uncontrolled potentially fatal cytokine storm.1 The basis for understanding the immunological mechanisms of HLH was laid by the demonstration of specific mutations in the perforine genes in primary HLH. Genetic defects in the perforine functions impair the function of cytotoxic CD8⁺ lymphocytes and natural killer cells. These cells are paramount in the self-regulating feedback towards the activated MPS cells.1 Regardless of the trigger, the final common pathway in HLH involves excessive cytokine release, particularly interleukin-6 (IL-6), IL-1β, IL-2, interferon-γ (IFN-γ), granulocyte-macrophage colony-stimulating factor and tumor necrosis factor-α, which further exacerbates the immune response and eventually leads to tissue damage2 (Figure 1). Blocking the activated cells by inhibiting these cytokines theoretically might prevent these events. Ruxolitinib, a potent JAK1/2 inhibitor approved for myelofibrosis and polycythemia vera, has emerged as a potential therapeutic option for HLH by selectively inhibiting these stimulating and effector cytokines. Unlike traditional non-selective immunosuppression, ruxolitinib

specifically targets JAK-STAT signaling, preserving the immune response against infections while dampening the destructive cytokine storm.2 So far only 17 primary HLH patients (of which some may involve overlapping cases) treated with ruxolitinib as salvage therapy have been described, mainly in case reports.3 In this issue of Heamatologica Ge et al. retrospectively describe the largest cohort of primary HLH cases treated with ruxolitinib so far involving 21 children.4 Most reported 5-year survival figures in comparable cohorts are between 60% and 70%, demonstrating the highest mortality rate in the first few months.5,6 However, by adding ruxolitinib to the standard of care therapy to bridge time to HSCT Ge et al. observed a remarkable 1-year overall survival of 90.5%, higher than the about 65% in the aforementioned studies.4-6 Additionally, this study highlights the tolerability of ruxolitinib, with only limited adverse effects observed and less use of cytostatic drugs previously considered standard therapy for bridging toward HSCT. Although potential treatment bias is a well-known limitation of retrospective studies, these data might endorse a shift towards more targeted anti-inflammatory bridging therapy to HSCT in primary HLH. Despite these promising clinical data, challenges still remain. It will be of great interest to see the 5-year survival rates of this study to draw a more substantial conclusion. Furthermore, the optimal dosing, treatment duration, and the role of ruxolitinib in the management of primary HLH warrant further investigation. Additionally, its long-term safety profile in the pediatric population requires thorough evaluation. In conclusion, ruxolitinib holds tremendous potential as a novel therapeutic option for patients with primary HLH. Its targeted approach to modulate cytokine storms represents a new horizon in the treatment landscape of this overwhelming

Haematologica | 109 February 2024

374

J.A.M. van Laar

EDITORIAL

Figure 1. The cytokinestorm and its related cytokines. Activated inflammatory cells release cytokines that are causing a cytokine storm subsequently leading to irreversible tissue damage. Cells active in hemophagocytic lymphohistiocytosis (HLH): activated CD8 and CD4, monocytes, dendritic cells, neutrophils and macrophages. INF: interferon; IL: interleukin; GM-CSF: granulocyte-macrophage colony-stimulating factor; TNF: tumor necrosis factor; NK: natural killer cells.

hyper-inflammatory condition in bridging towards a more definite treatment by HSCT. As we continue to unravel the intricacies of HLH pathogenesis and treatment, further research and clinical trials are essential to substantiate the full potential of JAK inhibition, ultimately improving patient outcomes and quality of life. Future perspectives might in-

clude combined cytokine inhibition by adding IFN-γ blockers to JAK inhibitors to synergistically damp the cytokine storm as has been described recently in vivo.7 Disclosures No conflicts of interest to disclose.

References 1. Stepp SE, Dufourcq-Lagelouse R, Le Deist F, et al. Perforin gene defects in familial hemophagocytic lymphohistiocytosis. Science. 1999;286(5446):1957-1959. 2. Albeituni S, Verbist KC, Tedrick PE, et al. Mechanisms of action of ruxolitinib in murine models of hemophagocytic lymphohistiocytosis. Blood. 2019;134(2):147-159. 3. Keenan C, Nichols KE, Albeituni S. Use of the JAK inhibitor ruxolitinib in the treatment of hemophagocytic lymphohistiocytosis. Front Immunol. 2021;12:614704. 4. Ge J, Zhang Q, Ma H, et al. Ruxolitinib-based regimen in children with primary hemophagocytic lymphohistiocytosis. Haematologica. 2024;109(2):559-566.

5. Bergsten E, Horne A, Aricó M, et al. Confirmed efficacy of etoposide and dexamethasone in HLH treatment: long-term results of the cooperative HLH-2004 study. Blood. 2017;130(25):2728-2738. 6. Messina C, Zecca M, Fagioli F, et al. Outcomes of children with hemophagocytic lymphohistiocytosis given allogeneic hematopoietic stem cell transplantation in Italy. Biol Blood Marrow Transplant. 2018;24(6):1223-1231. 7. Joly JA, Vallée A, Bourdin B, et al. Combined IFN-γ and JAK inhibition to treat hemophagocytic lymphohistiocytosis in mice. J Allergy Clin Immunol. 2023;151(1):247-259.e7.

Haematologica | 109 February 2024

375

EDITORIAL

Distinct t(14;19) translocation patterns in atypical chronic lymphocytic leukemia and marginal zone lymphomas Ralf Küppers Institute of Cell Biology (Cancer Research), University of Duisburg-Essen, Medical Faculty, Essen, Germany

Correspondence: R. Küppers ralf.kueppers@uk-essen.de Received: Accepted: Early view:

August 8, 2023. August 14, 2023. August 24, 2023.

Chromosomal translocations, which typically lead to activation of proto-oncogenes, play a major role in the pathogenesis of human B-cell malignancies.1 In most instances, these translocations involve one of the immunoglobulin (Ig) loci (mostly the Ig heavy [IgH] chain locus) and occur as mistakes of one of the genetic Ig locus remodeling processes V(D)J recombination, class-switch recombination, or somatic hypermutation.1 Such translocations bring the proto-oncogene under the control of the active enhancers of the respective Ig locus, leading to constitutive and hence deregulated expression of the translocated gene. For several types of B-cell lymphomas, the presence of particular translocation events is almost a disease-defining feature. For example, practically all cases of Burkitt lymphoma carry translocations of the MYC proto-oncogene into one of the Ig loci, and mantle cell lymphoma is characterized by the translocation t(11;14) involving the CCND1 gene and the IgH locus.1 BCL3 is a member of the NF-kB family of transcription factors, and its gene was identified in 1987 as the partner of the IgH locus in a translocation t(14;19)(q32;q13) in a case of chronic lymphocytic leukemia (CLL).2 BCL3 is generally functioning as an activator of the NF-kB pathway, with survival and/or proliferation promoting effects. Only about 1% of cases of CLL carry a t(14;19),3 so this event is rare and not a CLL-defining event as the other translocation examples mentioned above. The rare occurrence of t(14;19) in CLL raises the question whether they define a particular subset of CLL, or whether CLL with BCL3 translocations have at least particular features. Moreover, translocations t(14;19)(q32;q13) have also been recurrently detected in other types of lymphomas, in particular marginal zone lymphomas (MZL). This raises the question what the common and distinct features of these lymphomas and of the translocation events are. A first comprehensive analysis of a large series of B-cell lymphomas carrying t(14;19)(q32;q13)

was performed by Martin-Subero and colleagues.4 They showed that lymphomas with this translocation can be broadly separated into two groups. One group is mainly composed of CLL with unmutated Ig variable (IgV) region genes and few additional chromosomal changes, and a second group includes mainly MZL, but also other types of B-cell lymphomas. This second group showed a higher frequency of further chromosomal alterations than the CLL, and mutated IgV genes.4 In this issue of Haematologica, Carbó-Meix and colleagues present an in depth integrative multimodal analysis of 13 B-cell lymphomas with a t(14;19), which included IgV gene sequencing, whole genome sequencing, transcriptome analysis, epigenetic DNA methylation analysis and translocation breakpoint sequencing.5 They confirm that two groups of lymphomas with t(14;19) can be distinguished, one group composed mostly of CLL with unmutated IgV genes and low chromosomal complexity, and a group of mostly nodal and splenic MZL with mutated IgV genes and more chromosomal aberrations. Through the detailed multimodal analysis, numerous further important insights into the pathobiology of t(14;19)-harboring lymphomas were obtained. Regarding CLL with t(14;19), it is shown that they also carry trisomy 12 in nearly all cases, which is generally seen in only 15-20% of CLL. Also several other CLL-typical mutations were detected. However, CLL with t(14;19) often showed an unusual morphology and immunophenotype, and their transcriptomes and DNA methylation patterns distinguished them from typical CLL.5 Hence, these cases are atypical CLL. The analysis of the translocation breakpoints revealed that all CLL with t(14;19) have a breakpoint upstream of the BCL3 gene, and that this leads to overexpression of BCL3 in CLL with the translocation in comparison to CLL without that translocation. BCL3 expression in t(14;19)-carrying CLL on the derivative chromosome 14 is likely driven by the IgH 3´ enhancers.1,5 In contrast, the

Haematologica | 109 February 2024

376

R. Küppers

EDITORIAL

Figure 1. Linking t(14;19) events to B-cell differentiation processes and the development of t(14;19)-carrying B-cell malignancies. The t(14;19) happen as mistakes during class-switch recombination (CSR). CSR mainly takes place during T-cell-dependent immune responses both in the primary focus reaction and the germinal center (GC), but also in T-independent immune responses. Atypical chronic lymphocytic leukemia (CLL) with t(14;19) harbor unmutated IgV genes, indicating that in these cases the translocation either happened in the primary focus reaction or in a T-independent immune response. In these cases, the breakpoint is upstream of BCL3 and targets this proto-oncogene, as evident by enforced BCL3 expression in the respective CLL cells. CLL is likely derived from CD5⁺ B cells, which play a major role in T-independent immune responses, but may sometimes also be driven into T-dependent immune responses.10 Lymphomas with a t(14;19) and a breakpoint downstream of BCL3 are mostly postGC marginal zone lymphomas (MZL) with mutated IgV genes. In these cases, the translocation presumably happened during the GC reaction. BCL3 is not the target gene of these events, as the lymphomas lack BCL3 expression. Arrows with dashed lines indicate the final transformation processes from the B cells having acquired a t(14;19) to the fully malignant B-cell clones. SHM: somatic hypermutation; TH: T helper; TFH: T-follicular helper.

second group of mostly MZL showed breakpoints downstream of BCL3, and there is strong indication that for those cases, BCL3 is not the relevant target gene, because these cases did not show BCL3 expression.5 It remains to be clarified which of the downstream-located gene(s) may be the relevant proto-oncogene(s) of these translocations. As translocation t(14;19) with breakpoints downstream of BCL3 are obviously not targeting BCL3, these events should not be considered as BCL3 translocations. It should be noted that principally BCL3 could also be a translocation target when positioned on the derivate chromosome 19

in the downstream breakpoint situation. This derivative chromosome carries the IgH Em intron enhancer, and there are instances where this enhancer can drive oncogene expression.6 But apparently, this does not take place in t(14;19) with breakpoints downstream of BCL3. A further very interesting and novel finding of the work by Carbó-Meix and colleagues was that all breakpoints in the IgH loci were located in one of the IgH switch regions. These are repetitive elements upstream of each IgH constant region gene (except Cd) in which DNA double strand breaks occur during class-switch recombination. Class-switching

Haematologica | 109 February 2024

377

R. Küppers

EDITORIAL

mostly takes place during the germinal center (GC) reaction, and, therefore, it is likely that the precursor cells of the MZL with t(14;19) and mutated IgV genes (as a hallmark of a GC passage), acquired the translocation as a mistake of class switch recombination in the GC (Figure 1).7 But what about the CLL with t(14;19) and unmutated IgV genes, that presumably did not pass through a GC reaction, but nevertheless have class switching-associated translocations? There are primarily two possible scenarios for this. First, class-switching can also take place in T-cell-independent immune responses, where B cells are activated through strong B-cell receptor signals, often combined with Toll-like receptor signaling.8 Second, in T-cell-dependent immune responses, a first interaction and cross-wise stimulation of antigen-activated B cells and antigen-specific T-helper cells takes place in the primary focus reaction in the T-cell area or at the border between T-cell area and B-cell follicle (Figure 1). In this initial encounter, class-switching can take place, in the absence of somatic hypermutation in the B cells. There is even recent indication from murine studies that many class-switching events in T-dependent immune responses actually take place before the activated B cells enter the GC microenvironment.9 Thus, it is principally possible that B cells with unmutated IgV genes acquire class switch-associated translocations t(14;19) in the primary focus reaction and then as premalignant cells further develop into CLL without entering a GC reaction (or enter it briefly, but exit it before initiation of somatic hypermutation) (Figure 1).

A further distinction between the two groups of lymphomas with the distinct types of t(14;19) is the cellular origin of the specific B cells that underwent malignant transformation. MZL likely derive from marginal zone B cells or conventional follicular B cells that were driven into a GC reaction and became post-GC memory B cells before final malignant transformation, whereas CD5⁺ CLL likely derives from the CD5⁺ mature B-cell subset in humans (Figure 1).7,10 In conclusion, the two types of t(14;19) with breakpoints either upstream or downstream of BCL3 are associated with distinct B-cell malignancies. Translocations with upstream breakpoints are found in a subset of atypical CLL with unmutated IgV genes and trisomy 12, and lead to upregulated BCL3 expression, whereas t(14;19) with breakpoints downstream of BCL3 are found mostly in MZL and actually do not target BCL3, but presumably one or several downstream located genes. In both instances, the translocation events occur as mistakes of the class switch recombination event, but likely in different types of immune responses. Based on these differences, it is important to distinguish the two types of t(14;19)(q32;q13) in clinical-pathological studies of lymphomas with t(14;19), and it is hence very valuable that Carbó-Meix and colleagues already established and validated a fluorescence in situ hybridization assay for this in their study.5 Disclosures No conflicts of interest to disclose.

References 1. Küppers R, Dalla-Favera R. Mechanisms of chromosomal translocations in B cell lymphomas. Oncogene. 2001;20(40):5580-5594. 2. McKeithan TW, Rowley JD, Shows TB, Diaz MO. Cloning of the chromosome translocation breakpoint junction of the t(14;19) in chronic lymphocytic leukemia. Proc Natl Acad Sci USA. 1987;84(24):9257-9260. 3. Fang H, Reichard KK, Rabe KG, et al. IGH translocations in chronic lymphocytic leukemia: clinicopathologic features and clinical outcomes Am J Hematol. 2019;94(3):338-345. 4. Martin-Subero JI, Ibbotson R, Klapper W, et al. A comprehensive genetic and histopathologic analysis identifies two subgroups of B-cell malignancies carrying a t(14;19)(q32;q13) or variant BCL3-translocation. Leukemia. 2007;21(7):1532-1544. 5. Carbó-Meix A, Guijarro F, Wang L, et al. BCL3-rearrangements in B-cell lymphoid neoplasms occur in two breakpoint clusters

associated with different diseases. Haematologica. 2024;109(2):543-558. 6. Chesi M, Nardini E, Lim RS, Smith KD, Kuehl WM, Bergsagel PL. The t(4;14) translocation in myeloma dysregulates both FGFR3 and a novel gene, MMSET, resulting in IgH/MMSET hybrid transcripts. Blood. 1998;92(9):3025-3034. 7. Seifert M, Scholtysik R, Küppers R. Origin and pathogenesis of B cell lymphomas. Meth Mol Biol. 2019;1956:1-33. 8. Chen Z, Wang JH. Signaling control of antibody isotype switching. Adv Immunol. 2019;141:105-164. 9. Roco JA, Mesin L, Binder SC, et al. Class-switch recombination occurs infrequently in germinal centers. Immunity. 2019;51(2):337-350. 10. Seifert M, Sellmann L, Bloehdorn J, et al. Cellular origin and pathophysiology of chronic lymphocytic leukemia. J Exp Med. 2012;209(12):2183-2198.

Haematologica | 109 February 2024

378

SPOTLIGHT REVIEW

Advances in next-generation sequencing and emerging technologies for hematologic malignancies Regina Kwon1 and Cecilia C. S. Yeung1,2

Correspondence: C.C.S. Yeung

Department of Laboratory Medicine and Pathology, University of Washington and 2Translational Science and Therapeutics Division, Fred Hutchinson Cancer Center, Seattle, WA, USA

cyeung@fredhutch.org Received: Accepted:

March 30, 2023. August 17, 2023.

Abstract Innovations in molecular diagnostics have often evolved through the study of hematologic malignancies. Examples include the pioneering characterization of the Philadelphia chromosome by cytogenetics in the 1970s, the implementation of polymerase chain reaction for high-sensitivity detection and monitoring of mutations and, most recently, targeted nextgeneration sequencing to drive the prognostic and therapeutic assessment of leukemia. Hematologists and hematopathologists have continued to advance in the past decade with new innovations improving the type, amount, and quality of data generated for each molecule of nucleic acid. In this review article, we touch on these new developments and discuss their implications for diagnostics in hematopoietic malignancies. We review advances in sequencing platforms and library preparation chemistry that can lead to faster turnaround times, novel sequencing techniques, the development of mobile laboratories with implications for worldwide benefits, the current status of sample types, improvements to quality and reference materials, bioinformatic pipelines, and the integration of machine learning and artificial intelligence into molecular diagnostic tools for hematologic malignancies.

Introduction Hemato-oncology has led the way for clinical implementation of molecular diagnostic testing in patients’ care, starting with chronic myeloid leukemia and the pioneering work in measurable residual disease (MRD) monitoring. The detailed genomic databases that form the basis of our understanding of drivers of disease progression and treatment response are critical to how molecular pathologists and clinical variant scientists interpret variants in clinical practice today. Despite the advances of the past two decades, relapse rates remain high (~40-50%) for patients with acute myeloid leukemia (AML).1 The current molecular standard of care for hematologic malignancies has progressed beyond diagnosis and classification to prediction of response/resistance and monitoring of MRD.2 For example, both the International Consensus Classification and the World Health Organization emphasize the importance of molecular mutations for prognostication and guidance of therapy.3,4 Still more advances are coming to molecular hematopathology through: (i) improvements in the speed and footprint of next-generation sequencing (NGS) platforms; (ii) promising novel library chemistry and sequencing tech-

niques, such as long-read and long-range sequencing; (iii) advances in bioinformatic pipelines, especially in error correction; (iv) continued efforts to bring together the numerous methods used to diagnose, manage, and monitor disease; and (v) profound advances in our conceptualization of the reference genome (Figure 1). In this review article, we touch on these new developments and discuss their implications for hematologic malignancies. Advances in turnaround time Rapid molecular diagnostics are particularly relevant for AML, in which the doubling time of malignant cells can be as brief as 1.5-4 days.5 It is generally assumed that earlier initiation of induction therapy leads to better outcomes, and most patients are treated with induction chemotherapy within 4-16 days of diagnosis. However, more recent studies have demonstrated that time to treatment has less impact on long-term survival than was previously believed,6 leaving time for comprehensive molecular genomic profiling and, therefore, a better-tailored induction regimen. However, time considerations should be balanced with the need for prognostic molecular data for informed discussions with patients. Each year brings promises of faster turnaround times for

Haematologica | 109 February 2024

379

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies

R. Kwon and C.C.S. Yeung

Figure 1. Advances in hematologic molecular testing. Improvements over the past decade have not only dramatically decreased the time from sample collection to reporting of results but have also made it possible for more people around the world to access this technology. Optimizations and novel chemistry have enabled better assay performance with improved sensitivity, while innovations in informatics and the addition of artificial intelligence tools have improved pipeline speed and accuracy. On the way from research to a clinical pathway is the use of “omics” data (spatial transcriptomics, proteomics, etc.), with the promise of improved diagnosis, prognosis, and theragnosis. Most importantly, with global adoption of next-generation sequencing into clinical practice, a better reference genome is needed to adequately represent diverse populations of patients.

NGS. As early as 2013, groups reported speedier NGS by replacing temperature-dependent polymerase chain reaction with isothermal amplification.7 One example is the research by Yao et al., who exploited this method to sequence the entire mitochondrial genome.8 LAMP-sequencing, initially developed for COVID diagnostics, uses reverse transcription-based loop-mediated amplification (LAMP) primers to amplify RNA. Ion Torrent paired isothermal chemistry, with its semiconductor-based sequencing technology, is used to create an NGS system capable of faster sequencing times compared to Illumina.9,10 An adaptation of Thermo Fisher chemistry to the Genexus instru-

ment has yielded a myeloid mutation assay that targets both RNA fusions and common DNA hotspots in AML, providing reports within 2-3 days.11 We cannot describe these methodologies in detail here; interested readers are referred to the relevant literature. Advances in sequencing platforms and library preparation Long-read sequencing, which offers average read lengths of 10,000 to 100,000 base pairs or longer, was first realized in 1980 with the passage of an unbroken fragment of nucleic acid through a nanopore enzyme, creating an energy

Haematologica | 109 February 2024

380

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies differential recorded on a semiconductor membrane.12 Although long-read sequencing by nanopore technology has been commercially available since 2015, early iterations had low accuracy (~68%).13 Steady improvements, such as the addition of a helicase enzyme to control sequencing speed, flow cell alterations, and improved informatics, have enhanced the accuracy to 99% or better, albeit in limited studies.14 Long-read data shed light on the implications of phasing; that is, whether mutations are in cis (on the same chromosome) or in trans (on homologous chromosomes). Such differences may result in either reduced or enhanced protein activity, depending on whether the gene is a tumor suppressor or an oncogene.15 Studies have shown the significance of phasing information in hematologic malignancies. For example, Intlekofer et al. noted that acquired IDH2 mutations lying in trans are an important mechanism of acquired mutations in AML.16 Optical genome mapping (OGM) platforms use probes or tags to “barcode” a genome so that the high molecular weight DNA can be rapidly scanned and then aligned to a reference. OGM can potentially improve upon current clinical cytogenetics tests in three ways: first, by decreasing turnaround times; second, by using molecular techniques, thereby eliminating the need for highly skilled cytogenetic technologists; and third, by offering higher resolution than high-resolution karyotyping or single nucleotide polymorphism arrays. Significant advancements in library preparation and analysis have made OGM platforms clinically ready - or nearly so. The BioNano OGM platform uses a short probe to label a sequence that is repeated throughout the genome. Once tagged, long DNA molecules are linearized and imaged, allowing for rapid genome scanning and alignment of chromosomal regions.17 Hi-C sequencing exploits the cross-linking action of formaldehyde on DNA segments that are in close proximity to each other in vivo, and standard NGS is then used for data generation.18 Multiple layers of whole genome and epigenetic data are produced by HiC, and analysis requires advanced software and bioinformatic pipelines and novel artificial intelligence algorithms, which could prove challenging for some laboratories to implement. Duncavage et al. describe refinements to whole-genome sequencing that could position this sequencing technique as an alternative to conventional cytogenetics in myeloid diseases. Their method uses a highly efficient 8-hour library preparation and automated data analysis pipeline to generate 50x mean genome coverage.19 Although only a few studies have explored the potential of low-pass, cell-free, DNA-based whole genome sequencing, Shao et al. demonstrated in 103 AML patients that aberrant changes in 5-hydroxymethylcytosine sre a unique, sensitive, and specific informative marker in AML for diagnosis and adverse prognosis.20

R. Kwon and C.C.S. Yeung

Although most research efforts over the past two decades have focused on genetic mutations, it has been made clear that the epigenome represents an important level of biological and regulatory function. Methylation-based biomarker assays have demonstrated the clinical significance of epigenetic variation, but implementation has been impeded by the lack of a gold-standard comparator and standardized control materials. The SEQC2 epigenomics quality-control study addresses these problems by providing three different whole-genome bisulfite sequencing protocols in addition to a comparison of the methylation signatures of several cell lines.21 Methylation sequencing holds the potential for exciting advances in hematologic malignancies, such as the ability to be used in combination with long-read sequencing to identify “dead zone” structural variants and methylation defects.22 A summary depicting new sequencing platforms is shown in Figure 2, and the current status of these technologies is detailed in Table 1. Advances in single-cell genotyping: studying clonal dynamics and clonal evolution On average, the AML cancer genome has fewer mutations than any other cancer genome. The AML testing protocol is also perhaps the most integrated of any malignancy, utilizing a broad range of tests for diagnosis, prognosis, monitoring, treatment guidance, and response prediction. However, we continue to see relapse rates of approximately 40-50%, even after transplantation.23 In contrast to bulk tumor sequencing, single-cell analysis offers the ability to track clonal diversity and observe clonal dynamics and trends. Different groups have pursued single-cell genotyping as a means of understanding the evolutionary events that allow specific leukemic clones to evade treatment. Although still in its infancy, the technique shows promise. Early work on FLT3 ITD and NPM1 mutations by Pagurian et al. showed tumor heterogeneity at the subclonal level, which could contribute to the high rates of relapse and resistance in AML.24 These single-cell studies have shed light on the complex interplay between the immune microenvironment and the heterogeneous mutant clones that drive disease progression in AML.25 The extensive clonal diversity of the AML genome was further substantiated in a study by Morita et al., who performed single-cell DNA sequencing on 123 AML patients and were able to show the evolutionary dynamics underlying therapeutic resistance in some of their patients.26 Yeaton et al. used single-cell transcriptomic profiling to demonstrate that aberrant monocytic cell inflammatory signals are associated with poor prognosis and leukemic progression.27 Despite these insights, the barriers to clinical implementation are not insubstantial. Early iterations of single-cell

Haematologica | 109 February 2024

381

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies

R. Kwon and C.C.S. Yeung

Figure 2. Advances in next-generation sequencing platforms. (A) Single-cell sequencing methods vary. Depicted here is a singlecell preparation using microfluidics, in which the cells and reagents are encapsulated with barcoded beads that tag each mRNA molecule of a single cell. (B) Long-read sequencing capitalizes on the ability to read a long string of DNA or RNA without interruptions or the need to stitch together the smaller 150-250 base pair reads typical of standard next-generation sequencing. This allows for identification of structural variations, long repeat expansions, phasing information, and pseudogenes. (C) In errorcorrected sequencing, each nucleic acid molecule is tagged with a unique barcode prior to polymerase chain reaction amplification, allowing for downstream de-multiplexing and error correction as well as identification of bias errors. (D) Methylation sequencing typically relies on a bisulfite reaction, which allows the differentiation of cytosine from methylated cytosine. The purpose is to identify hyper- or hypo-methylated gene promoter regions that may contribute to tumorigenesis. (E) The illustration shows the steps of Hi-C sequencing, in which DNA that has undergone formalin crosslinking is fragmented and the fragmented ends are ligated. Sequencing of the junctions allows for interpretation of the whole genome at a resolution that can identify small copy number alterations, indels, inversions, and translations. Optical genome mapping (OGM; not illustrated) uses light microscopy-based techniques to create a map of the genome. OGM does not sequence DNA at the resolution of nextgeneration sequencing (i.e., at the base level) but offers higher resolution than DNA microarray. NGS: next-generation sequencing; UMI: unique molecular identifier; PCR: polymerase chain reaction.

Haematologica | 109 February 2024

382

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies

R. Kwon and C.C.S. Yeung

Table 1. New and emerging sequencing technologies. A summary of new sequencing platforms, their applicability to hematologic malignancies, and their clinical availability.

Technology

Use in blood cancers

Availability of commercial kits and products

Current status

References

Improved detection of fusions Error-corrected sequencing

Long-read sequencing

More sensitive variant detection for MRD in HM (as low as ~0.001% for duplex sequencing and ~0.3% for anchored multiplex PCR sequencing) Detection of fusions

In broad clinical use

Widely available doi:10.1016/j.cancergen.2020.02. commercial and 007 laboratory-developed tests (e.g., Invitae, doi:10.1186/s12920-020-0671-8 TwinStrand Biosciences)

RUO kits from limited companies (Oxford Detection of copy number Nanopore, PacBio) variation, structural variants Not in broad clinical Limited laboratories offer doi:10.1038/s41587-021-01108-x use for HM; mainly services for LRS Detection of nuclei acid doi:10.3389/fgene.2019.00426 base modifications (e.g., research applications Some platforms require methylation) investment in new instruments Reduced turnaround time (in some cases) DNA methylation sequencing kits widely available

Methylation sequencing

Optical genome mapping

Detection of epigenetic signatures

Rapid molecular-based karyotype Higher resolution than current karyotype testing

Small number of Applicable to routine doi:10.1016/j.jmoldx.2013.05.011 targeted NGS sequencers in most methylation tests in molecular laboratories doi:10.1016/j.yexmp.2023.104855 clinical use for solid LDT have limited tumors availability in specialized molecular laboratories for specific markers (e.g., MGMT and MLH1 promoter methylation) Used in clinical trials, not yet in routine clinical practice

RNA-based single-cell gene expression Single-cell sequencing

DNA single-cell sequencing

Research applications in HM

Spatial transcriptomics

RUO instruments and research kits

doi:10.1016/j.ajhg.2021.06.001 doi:10.1016/j.jmoldx.2022.12.005

Require investment in expensive components per assay and new instruments doi:10.3390/genes12030398 RUO hematologicspecific (Mission Bio); doi:10.1038/s41587-020-0472-9 non-hematologic-specific kits from limited companies (NanoString, 10x Genomics)

MRD: measurable residual disease; HM: hematologic malignancies; PCR: polymerase chain reaction; RUO: research use only; LRS: long-read sequencing; NGS: next-generation sequencing; LDT: laboratory-developed test.

Haematologica | 109 February 2024

383

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies

R. Kwon and C.C.S. Yeung

Table 2. Comparison of reference genomes. The progression of the human reference genome and the advantages and disadvantages of each iteration.

Genome

GRCh37/hg19

GRCh38/hg38

Released

2009

2013

Organization

Adoption

Genome Reference Widespread clinical Consortium adoption

Genome Reference Consortium

Advantages

Disadvantages

Well-annotated

Missing sequence

Robust community for bioinformatics tools

No heterochromatin sequence

Limited input diversity

> 1,000 issues resolved from GRCh37

Coordinate changes

Increased alternative loci

Not all contigs can be mapped to GRCh37

Limited clinical adoption -

Increased complexity of pipeline to manage alternative loci Missing sequence Laboratories slow to validate

T2T-CHM13

Pangenomes, various

2022

Ongoing

T2T Consortium

Research use

Human Pangenome Reference Consortium

Research use

Up to 99% of genome sequenced with confidence

Source material is European

Acrocentric chromosomes mapped

Community of bioinformatics tools lags GRC assemblies

Improved accuracy of structural variation

Lacks seasoned annotation repositories

Limited reports of mapping from GRCh37/hg19

High-quality assemblies Requires new bioinformatic from diverse populations methods Partnership with T2T Con- Lacks seasoned annotasortium tion repositories

References: Schneider et al. Evaluation of GRCh38 and de novo haploid genome assemblies demonstrates the enduring quality of the reference assembly. Genome Res. 2017;27(5):849-864. Nurk et al. The complete sequence of a human genome. Science. 2022;376(6588):44-53. GRC: Genome Reference Consortium; T2T: telomere to telomere.

RNA sequencing showed high levels of cell loss during li- for somatic variant identification typically have error rates brary preparation, often with only 100 to 200 cells cap- of about 1-2%, leading to an inability to distinguish pertured per sample processed, although more recent sistent disease from sequencing artifacts. ‘Error-corrected studies have been able to capture more than 6,000 cells sequencing,’ which modifies library preparation as well as in a single marrow aspirate sample.25 In addition, as yet, data processing, has been successfully adapted from there are no standardized methods or robust and verified ultra-low variant discovery to MRD evaluation in AML with control material, the cost of sequencing is high, and the reported limits of detection (LOD95) of 0.017% variant allele fraction.30 bioinformatics complex to implement.28 The European LeukemiaNet has updated its guidelines for Advances in data analysis and bioinformatic tools AML MRD detection to include the use of NGS for this purWith each advance in technique or change in purpose pose.2 However, the European LeukemiaNet does not supcomes a corresponding need to adapt the error prevention port NGS-MRD as the sole method of MRD monitoring due and correction in the bioinformatics pipeline. For example, to higher limits of detection than those for flow cytometry the addition of neural networks for analysis of earlier gen- and possible confounding by clonal hematopoiesis and erations of more error-prone nanopore data have shown germline mutations.1,2 One area for further research is RNA the ability to improve accuracy from 68% to 99.47%.29 sequencing which, although not more sensitive for variant The potential use of NGS in MRD monitoring is a prime allele frequencies, does give better results with structural example of advances through error-correction. Pipelines variants. Haematologica | 109 February 2024

384

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies Long-read sequencing technology represents an enormous advancement for hematologic malignancies in particular, in which structural variants contribute to a significant proportion of disease.3,4 Although long-read sequencing has not yet been implemented for clinical use, greater accuracy and improved computational tools will be instrumental in the development of clinical-grade assays. Third-generation techniques have also benefited, by necessity, from bioinformatic changes. For example, a comparison of 11 variant callers found that tools popular in clinical laboratories, such as GATK, Senteion, and Mutect2, generally do not perform as well on third-generation platforms,31 and adaptations are necessary for acceptable performance. Improved computational methods at the stages of library preparation and sequencing have also helped third-generation techniques produce longer reads with fewer errors, and some investigators have found accuracy levels comparable to those of current clinical sequencing platforms.31 Ultimately, third-generation platforms have yet to accrue the community support enjoyed by traditional sequencing methods. A key opportunity for improvement continues to be the primary analysis of sequencing data, which comprises quality control, alignment, variant calling, and annotation. Thus far, no scientific society or consortium has released a comprehensive recommendation for primary analysis. Rather, there are dozens of tools and programs that could be chosen for each step in primary analysis; for example, one author estimated more than 60 read mappers in 2019.32 Some of these steps are covered by the GATK Best Practices from the Broad Institute. Even so, most clinical pipelines are custom-made, proprietary platforms built out of a few of the many open-source and commercial packages available.33 Even with a call for guidelines, it is not clear that clinical laboratories have the resources or motivation to pursue changes that may profoundly affect their workflow. In 2017, the Association for Molecular Pathology, the American Society of Clinical Oncology, and the College of American Pathologists jointly released guidelines on the interpretation and reporting of somatic variant classifications. The Association for Molecular Pathology subsequently formed the Variant Interpretation Testing Across Laboratories (VITAL) Somatic Working Group to monitor the utilization and performance of these guidelines through a series of volunteer proficiency tests, or ‘Somatic Challenges.’ A follow-up published in 2023 found that only 59% of responses thus far have agreed with the guidelines’ intended classifications.34 The lack of a standard is a natural by-product of a vigorous open-source bioinformatics community. It also speaks to the fragmentation of the field and represents a challenge for large-scale oncology studies. A change in approach may be found in nf-core, an open-source initiative that

R. Kwon and C.C.S. Yeung

has released a number of modular pipelines for various analyses, such as single-cell RNA sequencing, gene fusions, and the B- and T-cell adaptive immune receptor repertoire.33 Advances in integration and interpretation With significant advances in nearly all modalities of patient testing and interpretation over the past decade, a new grand prize has emerged in genetic testing for hematologic malignancies: the integration of data from multiple modalities. For clarity, we separate ‘integration’ into two concerns: first, the desire to combine disparate workflows and interfaces; and second, the desire to analyze data from multiple sources. Hematologic malignancies are already routinely interrogated using multiple methods: fluorescence in situ hybridization, flow cytometry, blood counts, histomorphology, polymerase chain reaction, Sanger sequencing, and chromosomal array, not to mention clinical information contained in unstructured chart notes and documents. These assays are typically performed in separate laboratories, often with separate information systems and reporting mechanisms. The hematopathologist may be asked to organize and integrate these results into a clinically useful report. However, due to different turnaround times and the separation of laboratories, such reporting is often piecemeal, and the burden is transferred to the oncologist to place data into an interpretable timeline. Currently, no automated solution exists to integrate the results of multiple workflows into a single comprehensive genetic report. The second type of integration is the ability to analyze data from multiple sources and return an interpretation. Additional data sources for hematologic malignancies include transcriptomics, proteomics, epigenomics, chromatin mapping, single-cell sequencing, and long-read sequencing. Many research studies have examined the yield of combining two or more modalities, e.g., combining multiparameter flow cytometry with NGS to reconstruct clonal evolution in patients with AML or using spatial data analysis to enable the correlation of cell subsets on glass slides with genetic and/or gene expression data. Of particular interest are efforts to associate transcriptomes with cell identity through the use of single-cell sequencing and antibody-epitope labeling. This has led to a number of promising techniques in both chemistry and computational analysis, such as genotyping of transcriptomes, CITE sequencing, and TARGET sequencing.35 Numerous computational and statistical challenges to integrated data analysis exist, including multiple-hypothesis errors and the alignment of mixed data types. It is unsurprising that advances in machine learning are central to multi-omic analysis. A group recently demonstrated the implementation of a deep learning model to derive 12

Haematologica | 109 February 2024

385

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies unique molecular subgroups of multiple myeloma using data from NGS, chromosomal array, and RNA-sequencing assays.36 Phenotype acquisition through natural-language processing provides another layer of data for integration. Researchers used this method in conjunction with multiparameter flow cytometry, cytogenetics, and NGS data to identify a new subgroup of AML. Finally, several new commercial ventures are focused on the use of multiplex results for clinical trial eligibility, including Strata Oncology and OM1. Advances in the reference genome The reimagining of the reference genome may be the most revolutionary bioinformatic advancement in the past 5 years. Table 2 shows the evolution of the reference genome. The de facto standard in use today, GRCh37/hg19, is in fact no standard when one considers the many releases, patches, and alternate allele files that are cobbled together to create the true “normal” to which we compare all patients’ sequences. The Genome Reference Consortium (GRC) consolidated these amendments and addressed many gaps and errors in its latest version, GRCh38/hg38. Despite being released nearly a decade ago, however, GRCh38 has had little uptake in the clinical realm, with a 2020 survey finding that only 7% of the 28 academic and commercial laboratories surveyed had migrated to the newer reference genome.37 The most cited reasons for not having migrated were insufficient staff and a perceived lack of benefit for the effort required. A range of strategies for improving or replacing the reference genome have been suggested, including: (i) a consensus genome comprising the population-wide majority allele at each position; (ii) population-specific consensus genomes; and (iii) the pangenome, in which the reference would no longer be a linear structure but a complex manyto-many representation of all known variants at each locus.38 The pangenome, commonly conceived as a De Bruijn graph, has been pursued vigorously from a bioinformatics perspective, requiring as it does new methods of representing, traversing, and interpreting sequences. Finally, the critical achievement of 2022 was the publication of the results of the Telomere to Telomere (T2T) Consortium’s genome, which provided clarity on the additional 8% of the human genome not previously characterized.39 The T2T project characterized regions of the genome not included in the GRC assemblies, including the short arms of

R. Kwon and C.C.S. Yeung

chromosomes 13, 14, 15, 21, and 22.39 The sequence as well as the technologies developed for the project will be used to realize the goal of a human pangenome, increasing the accuracy of molecular diagnostics for diverse populations.

Conclusion New innovations in the past decade have greatly advanced care in patients with hematologic malignancies. Cloudbased systems assisted by artificial intelligence have improved data analysis, and the development of mobile laboratories has expanded these benefits worldwide, bringing the possibility of precision-driven therapies to all patients. Advances in molecular testing have swept across developed countries and established targeted NGS as a standard of care in most hematologic malignancies. However, despite the vast amounts of new knowledge, leukemia relapse and mortality rates remain high overall. The scientific community continues to innovate to push the field of molecular diagnostics forward, seeking a deeper understanding of how patients develop resistance and what factors contribute to relapse, with the hope that these discoveries may lead to better therapies. The future will bring improved monitoring strategies with highly accurate molecular-based MRD testing, single-cell sequencing technologies to assist with clone-tracking of a patient’s disease, as well as methylation-based sequencing to help predict better and more tailored treatment regimens as a part of the routine clinical care of leukemia patients. Disclosures CCSY is a consultant for TwinStrand Biosciences, Bristol Myers Squibb, Merck, Eli Lilly, and AbbVie and has received grant funding from Sensei, Pfizer, OBI Pharma, Lonza, Signal One, and Minerva. RK has no disclosures relevant to this work. Contributions CCSY and RK contributed equally to the research for and writing, editing, and revision of the review. Acknowledgments We thank Dr. Jerald P. Radich for his intellectual input and advice and Luk Cox for his assistance with the preparation of the figures.

References 1. Döhner H, Wei AH, Appelbaum FR, et al. Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN. Blood. 2022;140(12):1345-1377.

2. Heuser M, Freeman SD, Ossenkoppele GJ, et al. 2021 update on MRD in acute myeloid leukemia: a consensus document from the European LeukemiaNet MRD Working Party. Blood. 2021;138(26):2753-2767.

Haematologica | 109 February 2024

386

SPOTLIGHT REVIEW - Advances in NGS for hematologic malignancies 3. Arber DA, Orazi A, Hasserjian RP, et al. International Consensus Classification of Myeloid Neoplasms and Acute Leukemias: integrating morphologic, clinical, and genomic data. Blood. 2022;140(11):1200-1228. 4. Khoury JD, Solary E, Abla O, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36(7):1703-1719. 5. Akinduro O, Weber TS, Ang H, et al. Proliferation dynamics of acute myeloid leukaemia and haematopoietic progenitors competing for bone marrow space. Nat Commun. 2018;9(1):519. 6. Röllig C, Kramer M, Schliemann C, et al. Does time from diagnosis to treatment affect the prognosis of patients with newly diagnosed acute myeloid leukemia? Blood. 2020;136(7):823-830. 7. Ma Z, Lee RW, Li B, et al. Isothermal amplification method for next-generation sequencing. Proc Natl Acad Sci U S A. 2013;110(35):14320-14323. 8. Yao Y, Nishimura M, Murayama K, et al. A simple method for sequencing the whole human mitochondrial genome directly from samples and its application to genetic testing. Sci Rep. 2019;9(1):17411. 9. Singh RR, Patel KP, Routbort MJ, et al. Clinical validation of a next-generation sequencing screen for mutational hotspots in 46 cancer-related genes. J Mol Diagn. 2013;15(5):607-622. 10. Rachiglio AM, De Sabato L, Roma C, et al. SARS-CoV-2 complete genome sequencing from the Italian Campania region using a highly automated next generation sequencing system. J Transl Med. 2021;19(1):246. 11. Sande CM, Wu R, Yang G, et al. Rapid and automated semiconductor-based next-generation sequencing for simultaneous detection of somatic DNA and RNA aberrations in myeloid neoplasms. J Mol Diagn. 2023;25(2):87-93. 12. Wang Y, Zhao Y, Bollas A, Wang Y, Au KF. Nanopore sequencing technology, bioinformatics and applications. Nat Biotechnol. 2021;39(11):1348-1365. 13. Ashton PM, Nair S, Dallman T, et al. MinION nanopore sequencing identifies the position and structure of a bacterial antibiotic resistance island. Nat Biotechnol. 2015;33(3):296-300. 14. Liu C, Yang X, Duffy BF, et al. High-resolution HLA typing by long reads from the R10.3 Oxford nanopore flow cells. Hum Immunol. 2021;82(4):288-295. 15. Wang Y, Zhang W, Edelmann L, Kolodner RD, Kucherlapati R, Edelmann W. Cis lethal genetic interactions attenuate and alter p53 tumorigenesis. Proc Natl Acad Sci U S A. 2010;107(12):5511-5515. 16. Intlekofer AM, Shih AH, Wang B, et al. Acquired resistance to IDH inhibition through trans or cis dimer-interface mutations. Nature. 2018;559(7712):125-129. 17. Khan WA, Toledo DM. Applications of optical genome mapping in next-generation cytogenetics and genomics. Adv Mol Pathol. 2021;427-436. 18. Lajoie BR, Dekker J, Kaplan N. The hitchhiker’s guide to Hi-C analysis: practical guidelines. Methods. 2015;72:65-75. 19. Duncavage EJ, Schroeder MC, O’Laughlin M, et al. Genome sequencing as an alternative to cytogenetic analysis in myeloid cancers. N Engl J Med. 2021;384(10):924-935. 20. Shao J, Wang S, West-Szymanski D, et al. Cell-free DNA 5hydroxymethylcytosine is an emerging marker of acute myeloid

leukemia. Sci Rep. 2022;12(1):12410. 21. Foox J, Nordlund J, Lalancette C, et al. The SEQC2 epigenomics quality control (EpiQC) study. Genome Biol. 2021;22(1):332. 22. Sanford Kobayashi E, Batalov S, Wenger AM, et al. Approaches to long-read sequencing in a clinical setting to improve diagnostic rate. Sci Rep. 2022;12(1):16945. 23. Thol F, Ganser A. Treatment of relapsed acute myeloid leukemia. Curr Treat Options Oncol. 2020;21(8):66. 24. Paguirigan AL, Smith J, Meshinchi S, Carroll M, Maley C, Radich JP. Single-cell genotyping demonstrates complex clonal diversity in acute myeloid leukemia. Sci Transl Med. 2015;7(281):281re2. 25. van Galen P, Hovestadt V, Wadsworth Ii MH, et al. Single-cell RNA-seq reveals AML hierarchies relevant to disease progression and immunity. Cell, 2019;176(6):1265-1281. 26. Morita K, Wang F, Jahn K, et al. Clonal evolution of acute myeloid leukemia revealed by high-throughput single-cell genomics. Nat Commun. 2020;11(1):5327. 27. Yeaton A, Cayanan G, Loghavi S, et al. The impact of inflammation-induced tumor plasticity during myeloid transformation. Cancer Discov. 2022;12(10):2392-2413. 28. Lähnemann D, Köster J, Szczurek E, et al. Eleven grand challenges in single-cell data science. Genome Biol. 2020;21(1):31. 29. Wick RR, Judd LM, Holt KE. Performance of neural network basecalling tools for Oxford Nanopore sequencing. Genome Biol. 2019;20(1):129. 30. Thol F, Gabdoulline R, Liebich A, et al. Measurable residual disease monitoring by NGS before allogeneic hematopoietic cell transplantation in AML. Blood. 2018;132(16):1703-1713. 31. Pei S, Liu T, Ren X, Li W, Chen C, Xie Z. Benchmarking variant callers in next-generation and third-generation sequencing analysis. Brief Bioinform. 2021;22(3):bbaa148. 32. Singer J, Irmisch A, Ruscheweyh H-J, et al. Bioinformatics for precision oncology. Brief Bioinform. 2019;20(3):778-788. 33. Ewels PA, Peltzer A, Fillinger S, et al. The nf-core framework for community-curated bioinformatics pipelines. Nat Biotechnol. 2020;38(3):276-278. 34. Li MM, Cottrell CE, Pullambhatla M, et al. Assessments of somatic variant classification using the Association for Molecular Pathology/American Society of Clinical Oncology/College of American Pathologists guidelines: a report from the Association for Molecular Pathology. J Mol Diagn. 2023;25(2):69-86. 35. O’Sullivan JM, Mead AJ, Psaila B. Single-cell methods in myeloproliferative neoplasms: old questions, new technologies. Blood. 2023;141(4):380-390. 36. Ortiz-Estévez M, Towfic F, Flynt E, et al. Integrative multi-omics identifies high risk multiple myeloma subgroup associated with significant DNA loss and dysregulated DNA repair and cell cycle pathways. BMC Med Genomics 2021;14(1):295. 37. Lansdon LA, Cadieux-Dion M, Yoo B, et al. Factors affecting migration to GRCh38 in laboratories performing clinical nextgeneration sequencing. J Mol Diagn. 2021;23(5):651-657. 38. Eizenga JM, Novak AM, Sibbesen JA, et al. Pangenome graphs. Annu Rev Genomics Hum Genet. 2020;21:139-162. 39. Nurk S, Koren S, Rhie A, et al. The complete sequence of a human genome. Science. 2022;376(6588):44-53.

Haematologica | 109 February 2024

387

R. Kwon and C.C.S. Yeung

REVIEW ARTICLE

Diffuse large B-cell lymphoma involving the central nervous system: biologic rationale for targeted therapy Mark Roschewski1 and Daniel J. Hodson2

Correspondence: M. Roschewski mark.roschewski@nih.gov

Lymphoid Malignancies Branch, Center for Cancer Research, National Cancer Institute,

D.J. Hodson djh1002@cam.ac.uk

Bethesda, MD, USA and Wellcome MRC Cambridge Stem Cell Institute, University of 2

Cambridge, Cambridge Biomedical Campus, Cambridge, UK

Received: Accepted: Early view:

May 3, 2023. September 4, 2023. September 14, 2023.

Abstract Diffuse large B-cell lymphoma (DLBCL) is an aggressive B-cell lymphoma curable even in advanced stages. DLBCL involving the central nervous system (CNS) is more difficult to cure and fewer treatment options exist. Primary CNS lymphoma (PCNSL) refers to aggressive lymphomas confined to the CNS, and are almost always DLBCL. Standard approaches for PCNSL use high-dose methotrexate-based combinations as induction therapy and younger patients often receive dose-intensive consolidation. However, dose-intensive therapies are not suitable for all patients, and older patients have fewer effective treatment options. Patients with relapsed or chemotherapy-refractory disease have a very poor prognosis. Secondary CNS lymphoma (SCNSL) describes aggressive lymphomas involving the CNS at initial presentation or relapses within the CNS after treatment for systemic DLBCL. Isolated CNS relapse is often managed as PCNSL, but patients with synchronous involvement of DLBCL in both the periphery and the CNS pose a unique clinical challenge. Insights into the molecular circuitry of DLBCL have identified distinct genetic subtypes including cases with a predilection for CNS invasion. PCNSL and subsets of SCNSL are characterized by chronically activated B-cell receptor and NFκB signaling along with genetic evidence of immune evasion which may be exploited therapeutically. Improved mechanistic understanding of targetable pathways underpinning CNS lymphomas has led to numerous clinical trials testing targeted agent combinations and immunotherapy approaches with promising early results. Biologically rational strategies may further improve the cure rate of CNS lymphomas, either by overcoming intrinsic or acquired treatment resistance and/or by being broadly applicable to patients of all ages.

Introduction Diffuse large B-cell lymphoma (DLBCL) is a spectrum of aggressive B-cell lymphomas with remarkable clinical and molecular heterogeneity. These tumors manifest in myriad clinical presentations, including involvement of the central nervous system (CNS). Importantly, DLBCL can be cured with immunochemotherapy, even in advanced stages and after relapse. Primary DLBCL of the CNS (PCNSL) is a rare subtype of DLBCL that involves the brain, eyes, leptomeninges, or spinal cord without evidence of systemic involvement. It accounts for 6% of all new CNS tumors with an annual incidence of approximately 1,500 cases in the United States.1 The incidence is rising in patients with advanced age, particularly those aged ≥75 years.2,3 Front-line treatment for PCNSL

varies across institutions, but patients deemed suitable for dose-intensive therapy receive one of several high-dose methotrexate (HD-MTX)-based regimens and consolidation with either autologous stem cell transplantation (ASCT), radiotherapy, or cytarabine-based chemotherapy.4-7 There is still no consensus on the optimal induction regimen, and treatment guidelines rely almost exclusively on phase II clinical trials.8 Furthermore, the cure rate may not be evident when surrogate markers for survival are used since late recurrences are not uncommon.9,10 Patients deemed unsuitable for HD-MTX, including those with advanced age or co-morbidities, are treated with reduced doses of chemotherapy with palliative intent.11-13 Notably, patients with chemotherapy-refractory disease have a very poor prognosis, although it is unclear if these poor clinical results are due to intrinsic chemotherapy resistance, ineffective drug

Haematologica | 109 February 2024

388

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

delivery, or a combination of these factors.14 Whilst good results are seen in patients able to tolerate dose-intensive therapy, the rising proportion of older patients means that most patients with PCNSL will die of lymphoma, and novel treatment approaches are needed. Secondary CNS lymphoma (SCNSL) is distinguished from PCNSL as either synchronous CNS and systemic involvement of lymphoma or an isolated CNS relapse. SCNSL is not a specific biologic entity, but most cases are DLBCL. The incidence of CNS involvement at diagnosis for DLBCL is only approximately 5%, but CNS progression may arise shortly after or during initial therapy;15 in a retrospective study of SCNSL, the majority (87%) occurred at the time of first relapse and 39% had synchronous systemic relapse.16 The prognosis of SCNSL is poor, so emphasis has been placed on prevention with intrathecal chemotherapy or HD-MTX prophylaxis delivered with front-line therapy in high-risk patients, but these strategies are not very effective and remain controversial.17,18 Furthermore, patients with CNS involvement are often unnecessarily excluded from clinical trials testing novel approaches, even when the agents are effective for CNS disease. Overall, there are few effective treatments for SCNSL, and this remains an unmet clinical need. We highlight advances in our understanding of the molecular pathogenesis of primary and secondary DLBCL of the CNS, including aspects of biology that are shared with genetic subtypes of DLBCL and the oncogenic pathways that are emerging as therapeutic targets. We discuss the treatment landscape of CNS lymphomas, and emphasize targeted agents with the strongest biologic rationale.

Standard approach to diffuse large B-cell lymphomas involving the central nervous system Treatment of primary diffuse large B-cell lymphoma of the central nervous system The prognosis of PCNSL is significantly worse than systemic DLBCL. Although induction therapy achieves remission in approximately 50% of patients, without consolidation, the risk of relapse remains unacceptably high.9 Furthermore, both induction therapy and consolidation are associated with significant toxicities that are not tolerated by all patients. Clinical outcomes are worse in older patients, particularly in those who do not receive dose-intensive therapy.19 The most pressing clinical needs are to improve the efficacy of induction therapy to potentially obviate the need for consolidation and to develop effective alternative strategies for patients who are not suitable candidates for dose-intensive therapy. The first therapy that improved clinical outcomes was whole brain radiotherapy (WBRT), but all patients relapsed and

M. Roschewski and D.J. Hodson

neurotoxicity was problematic.20 Thus, the original role of chemotherapy was to reduce reliance on WBRT. Indeed, when HD-MTX was added to WBRT, many patients deferred radiotherapy for an extended period of time.21 In this way, HD-MTX became the de facto cornerstone chemotherapeutic agent for PCNSL due to its reliable CNS penetration, not based on a strong biologic rationale for efficacy in DLBCL. Combination chemotherapy regimens were subsequently developed to further reduce reliance on WBRT. A randomized phase II study demonstrated that high-dose cytarabine added to HD-MTX improved the complete response (CR) rate to 46% compared to 18% with HD-MTX alone.22 Rituximab has also been used to improve the response rates of induction therapy. A multicenter phase II study comparing HD-MTX-based induction regimens found that patients treated with the HD-MTX and cytarabine backbone plus both rituximab and thiotepa (MATRix) had a CR rate of 49% compared to 30% with rituximab added to HD-MTX backbone and 23% with HD-MTX backbone alone.7 Another strategy has been to incorporate high-dose chemotherapy with ASCT as a consolidation strategy.23,24 This strategy can be effective for long-term disease control, but results vary depending on the conditioning regimen used. A retrospective registry study showed that patients who received thiotepa/busulfan/cyclophosphamide (TBC) or thiotepa/carmustine (TT-BCNU) conditioning had superior 3-year progression-free survival (PFS) compared to those who received carmustine, etoposide, cytarabine, melphalan (BEAM) conditioning.25 This improved efficacy, however, came at the cost of increased toxicity and non-relapse mortality with TBC. Indeed, a randomized phase II study in 140 patients with PCNSL aged <60 years compared thiotepa-based conditioning followed by ASCT to WBRT as a consolidation strategy and the 8-year event-free survival was 67% versus 39% (P=0.03) in favor of ASCT.5 Selected older patients without co-morbidities may be candidates for ASCT consolidation after careful consideration of the risks associated with this treatment.26 Ongoing trials are addressing the feasibility and efficacy of this approach.27 Taken together, these studies strongly suggest that HDMTX-based induction regimens with ASCT consolidation improve clinical outcomes compared to WBRT consolidation in younger patients. Furthermore, the ability of these regimens to cure PCNSL without consolidation is limited, and there are few effective alternative treatment options. Treatment of secondary diffuse large B-cell lymphoma of the central nervous system Compared to systemic DLBCL, the outcomes for SCNSL are significantly worse with an expected survival of <6 months.16,28 Patients with synchronous CNS and systemic involvement are particularly challenging since therapy needs to be both effective for systemic DLBCL and penetrate the blood-brain barrier. In a large retrospective series, the 2-year survival rate for

Haematologica | 109 February 2024

389

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

SCNSL was only 20%, but patients deemed eligible for dose-intensive therapy had a 2-year survival rate of 62%.16 Patients with isolated CNS disease may have a better prognosis; a series of 113 patients with isolated CNS relapse demonstrated that a subset of patients could survive for more than 2.5 years, but ultimately most patients died from lymphoma.29 In patients with synchronous DLBCL involving the CNS and periphery, clinical outcomes are largely related to the ability to control the CNS disease.30 A retrospective series of 80 patients with synchronous systemic DLBCL and CNS involvement demonstrated a 2-year OS of 49%. Patients treated with dose-intensive regimens had a superior 2-year PFS compared to those treated with less intensive therapy (50% vs. 31%, respectively; P=0.006). Notably, the 2-year OS of patients with relapse in the CNS was significantly worse than those with systemic relapse (13% vs. 36%, respectively; P=0.02). The potential benefit of ASCT for SCNSL is also restricted to younger patients and the supporting data mostly come from retrospective case series.31-34 A prospective multicenter phase II study of 79 patients with SCNSL tested alternating cycles of the MATRix regimen with rituximab, ifosfamide, carboplatin, etoposide (R-ICE) and responding patients received ASCT consolidation.35 In this study, only 37 (47%) patients received an ASCT. Indeed, this strategy should be considered for suitable candidates, but is largely restricted to younger patients and those without co-morbidities.

Molecular biology of primary central nervous system lymphoma and secondary central nervous system lymphoma Diffuse large B-cell lymphoma arises from the malignant transformation of B lymphocytes that are either engaged in, or exiting, the germinal center (GC) reaction. GC are transient microanatomic structures that form in secondary lymphoid tissue in response to antigenic stimulation, where GC B cells undergo somatic hypermutation of their immunoglobulin genes associated with intense competition for T cell prior to differentiation into plasma or memory B cells. As such, systemic DLBCL typically develops within lymphoid tissue, and tumor cells remain reliant on the lymph node microenvironment. This presents a biological challenge for CNS lymphomas, since GC are not found within the CNS, and the usual supportive microenvironment of the lymph node is absent. Whilst secondary central nervous system lymphoma (SCNSL) represents spread from cells that initiated as systemic DLBCL, the case of PCNSL is even more intriguing since these cases lack evidence of lymphoma outside of the CNS, and relapses are almost universally restricted to the CNS. The question of how and where these tumors initiate, and how they survive or

M. Roschewski and D.J. Hodson

even depend upon the microenvironment of the CNS are key to understanding how biology might ultimately direct the therapy for CNS lymphomas. Two simplistic ‘pull or push’ hypotheses can be formed: firstly, the pull of some survival factor exclusive to the CNS, to which tumor cells are addicted, or secondly, the push from outside the CNS of some factor which tumor cells are unable to tolerate. Research over the last two decades, reinforced by recent genomic studies, suggest that both play a role in terms of neural-specific antigen within the CNS and host antitumor immunity outside the CNS. These studies highlight vulnerabilities that might be exploited for the biologically-directed therapy of CNS lymphomas. We will mainly consider CNS lymphomas that arise within immunocompetent patients, since these are a biologically homogeneous group, but also highlight important contrasts with the biology of CNS lymphomas in the immune deficient patient. We will focus on aspects of the biology that have the greatest therapeutic relevance. Primary central nervous system lymphoma shares molecular biology with specific diffuse large B-cell lymphoma subtypes Diffuse large B-cell lymphoma can be classified by gene expression profiling into subtypes that resemble either GC B cells (GCB DLBCL) or in vitro activated B cells (ABC DLBCL), and, more recently, genetic subtypes have emerged.36-39 Early work using immunohistochemistry showed that the majority of PCNSL cases could be classified as ABC DLBCL,40 a finding borne out by transcriptional analysis using Nanostring.41 Early studies also revealed that PCNSL was enriched for mutations associated with ABC DLBCL. These included mutation of MYD88, CD79B, TBL1XR1, PIM1, and deletions of the CDKN2A and HLA loci.42-46 Since the advent of next-generation sequencing technologies, more than 250 PCNSL exomes or whole genomes have now been published.47-52 Four main conclusions,51 summarized from across all these studies, capture the unique biological aspects of PCNSL: 1) an extremely high frequency of MYD88 (67-86%) and CD79B (61%) mutations, implicating corrupted toll-like receptor (TLR) and B-cell receptor (BCR) signaling as a mechanism of enhanced NFκB activity; 2) frequent mutations that impede onward differentiation and lock tumor cells into a GC-like proliferative state, including mutation of TBL1XR1, PRDM1 and translocation of BCL6; 3) the near universal loss of negative regulator of cell cycle, CDKN2A, with only rare mutation of TP53; and 4) the very high frequency of mutations that lead to immune evasion, including deletion of HLA loci, or mutation of B2M or CD58 (Figure 1). Intriguingly, this genomic profile is shared with other forms of extranodal lymphoma such as primary DLBCL of the testis, breast, and skin.39,53 Moreover, this genetic profile of PCNSL aligns precisely to a specific subtype of systemic DLBCL, variably termed C5, MCD or MYD88, identified

Haematologica | 109 February 2024

390

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

from DLBCL genetic clustering studies.36-39 The C5/MCD/ MYD88 subtype is characterized by BCR/TLR activation and immune evasion, and is strongly enriched for cases with extranodal involvement. Where genetic classifiers such as LymphGen have been applied to cases of PCNSL, almost all classified cases are found to fall within the MCD subtype;39 this suggests an overlap between PCNSL and systemic MCD DLBCL in terms of biology, but also therapeutically exploitable vulnerabilities. Addiction to chronic active B-cell receptor signaling The strikingly high frequency of MYD88 and CD79 mutations points strongly to oncogenic NFκB signaling as a critical driver of PCNSL. These mutations affect specific hotspots, resulting in a leucine to proline change at amino acid 265 in MYD88 (L265P) and the exchange of tyrosine 196 in CD79B. The function of these hotspot mutations

M. Roschewski and D.J. Hodson

is already well established from the study of ABC DLBCL. MYD88 is an adapter protein that is activated downstream of the toll-like receptor TLR9. It supports the assembly of a signaling complex that also includes IRAK 1 and IRAK4, thereby promoting activation of NFκB, and further enhanced by the L265P mutation.54 CD79B forms a heterodimer with CD79A to create the signaling component of the B-cell antigen receptor. Tyrosine 196 sits within the immunoreceptor tyrosine-based activation motif (ITAM) and its mutation interrupts a negative feedback pathway, contributing to a hyperactive signaling state termed chronic active BCR signaling.55 For reasons that remain elusive, this state of chronic active BCR signaling appears to be almost always associated with the IgM isotype. Despite strong expression of the class switching enzyme activation-induced cytidine deaminase (AID), PCNSL show blocked class switch recombination while retaining expression of surface IgM.56

Figure 1. Four biological cornerstones of PCNSL biology. 1) Chronic active BCR signaling driven by neural-specific autoantigen, activating mutations of MYD88 and CD79B, and formation of the MYD88-TLR9-BCR (My-T-BCR) complex. 2) Cell cycle activation secondary to deletion of the CDKN2A locus. 3) Escape from host immunity, resulting from loss of HLA expression, mutation of CD58 or enhanced expression of immune checkpoint ligands. 4) Transcription factor mutations (BCL6, PRDM1 and TBL1XR1) that block exit from, or promote re-entry into, the germinal center phenotype of proliferation and ongoing hypermutation. NK cell: natural killer cell.

Haematologica | 109 February 2024

391

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

MYD88, TLR9 and the IgM BCR have been found to associate into a multiprotein, oncogenic supercomplex known as the My-T-BCR, which drives activation of NFκB in DLBCL. Consistent with the frequent double mutation of MYD88 and CD79B, the oncogenic My-T-BCR complex was strongly detected in biopsies of PCNSL, suggesting it is a critical driver of oncogenic NFκB activity in PCNSL.57 Importantly, the formation of this complex depends upon an active BCR signal and can be blocked by pharmacological inhibitors of BCR signaling such as ibrutinib, highlighting a potential therapeutic vulnerability in PCNSL. In ABC DLBCL, CD79B ITAM mutations are frequently associated with a BCR signal that is initiated by engagement of autoantigen. Mouse models show how combined CD79B and MYD88 mutation allows autoreactive B cells to escape deletion.58 Together with the strict tissue trophism seen in PCNSL, this has opened questions as to whether PCNSL might receive chronic BCR stimulation from a neural-specific autoantigen. In support of this, PCNSL shows strong bias towards usage of the immunoglobulin variable gene segment VH4-34, which encodes polyreactive or autoreactive immunoglobulin.59-61 Moreover, PCNSL immunoglobulin genes show heavy somatic hypermutation that is biased towards codon-altering but non-destructive mutations, a pattern suggesting antigen-driven selection.61 Production of recombinant antibody based upon PCNSL immunoglobulin V gene sequences confirm that these antibodies are polyreactive and bind CNS proteins. Moreover, comparison of tumor sequences with the predicted naïve BCR showed that somatic hypermutation further increased self-reactivity, suggesting that immunoglobulin variants are selected according to their interaction with neural autoantigen.62 Protein microarrays have identified putative interacting neural antigens. Whilst PCNSL-derived recombinant antibodies failed to recognize common autoantigens, they showed reactivity to a large range of CNS-specific antigens.63 Overall, these studies reveal the reliance of PCNSL on chronic activated signaling through the BCR. This may provide a partial explanation for the unique CNS trophism of PCNSL, but it also reveals an important tumor vulnerability that might be exploited by pharmacological inhibitors of the BCR pathway. Heightened sensitivity to antitumor immunity Despite their location in an immune-privileged site, PCNSL cells remain under strong selective pressure from the immune system. Although the need to escape immune recognition is shared with systemic DLBCL, one of the most striking features of PCNSL is the almost universal presence of genetic alterations that promote immune evasion.64 This prominent immune vulnerability may be secondary to the high expression of the DNA mutator enzyme AID in PCNSL, which leads to ongoing aberrant somatic hypermutation and an increasing neoantigen burden.64,65

M. Roschewski and D.J. Hodson

Multiple studies show that most PCNSL tumors have lost expression of both MHC class I and class II.66-68 This results from deletions within the HLA locus on chromosome 6p, truncating mutations of individual HLA genes and in B2M.52,53,65,69,70 Inactivating mutations of TAP1 and TAP2 are additional mechanisms that limit immunogenic peptide display on MHC.39,65 Loss of MHC renders tumor cells invisible to cytotoxic T cells that typically infiltrate the PCNSL tumor microenvironment.67 However, MHC I loss may render tumor cells susceptible to destruction by natural killer (NK) cells. CD58 is a cell surface antigen and ligand for the receptor CD2, expressed on T cells and NK cells. CD58-CD2 ligation is essential for tumor lysis by T or NK cells, and CD58 is commonly inactivated in DLBCL that has lost MHC I expression.71 Indeed, CD58 is mutated in over a third of PCNSL cases.65 Mutations of CD70 and CD80 represent other mechanisms used by PCNSL to avoid cytotoxic T-cell killing.48,64,65 An alternative immune evasion strategy is the expression of immune suppressive ligands such as PDL1 and PDL2, inducing an exhaustion phenotype in T cells. Copy number gains of 9p24.1, which contains the PDL1 and PDL2 loci, and structural variants of both loci leading to increased surface expression of PDL1 and PDL2, were reported in PCNSL.53 However, the impact of immune checkpoint ligands in PCNSL remains controversial as increased surface expression on tumor cells was not identified in other studies.50,68 Regardless of this mechanism, single cell transcriptomic analysis has revealed clear signatures of T-cell exhaustion within the PCNSL microenvironment.72 Overall, it appears that almost every case of PCNSL is associated with genetic evidence of immune evasion, highlighting a critical vulnerability. Whilst this raises enticing therapeutic possibilities, greater understanding is still required to define precisely which of these genetic alterations might render cells more, or perhaps less, sensitive to immunotherapeutic approaches. The enigmatic origins of primary central nervous system lymphoma A fascinating enigma is the precise origin of PCNSL. DLBCL arises from a B cell engaged in or exiting the GC. In ABC DLBCL, a simplistic model of malignant transformation includes NFκB activation to drive proliferation and survival, combined differentiation block, precluding cells from exiting the GC phenotype. Superficially, this fits with the genetics of PCNSL, including chronic BCR activation leading to oncogenic NFκB activation, and blocked plasma cell differentiation as a result of frequent BCL6 translocation, or genetic loss of PRDM1.53,65,73 Moreover, high levels of somatic hypermutation in the immunoglobulin genes support this post-GC origin of PCNSL.62 However, there is scant evidence for GC formation within the CNS, suggesting that this step may occur in peripheral lymphoid tissue, or alternatively reflects AID induction through a GC-indepen-

Haematologica | 109 February 2024

392

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

dent process. A slightly different model has been proposed prompted by the high frequency of mutations in TBL1XR1. Mouse models show how mutant TBL1XR1 drives B cells towards a memory B (MB) fate.74 TBL1XR1 mutant MB cells are unable to differentiate to plasma cells, but, instead, undergo repeated cycles of re-entry into the GC reaction upon subsequent antigen exposure. Mouse models show how MB differentiation is also promoted by MYD88 mutation. Of particular relevance to PCNSL, these mice develop increased self-reactive MB cells that can be stimulated to reactivate and proliferate with minimal T-cell co-stimulation.75 Therefore, the genetics of PCNSL suggest a model where mutant, polyreactive MB engage neural-specific autoantigen outside a formal GC environment, which drives cellular activation, proliferation, up-regulated AID expression, and subsequent somatic hypermutation, leading to enhanced autoantigen recognition and malignant transformation in the CNS.76 The point at which B cells enter the CNS during this process of malignant transformation remains unclear. However, it is notable that MYD88 mutant, non-tumor cells were detected in peripheral blood in PCNSL patients, suggesting this is an early initiating mutation that occurs outside the CNS.48 The distinction is of significance as the early transformed cell may represent a common precursor cell (CPC) that seeds relapse and may be the source of late relapses. Comparison to secondary central nervous system lymphoma and immune deficient central nervous system lymphomas Whilst PCNSL in the immunocompetent patient represents a comparatively homogeneous entity as described above, SCNSL and PCNSL arising in the context of immune deficiency show important biological differences. Gandhi et al. studied the genetics of PCNSL in immune deficient patients, most associated with Epstein-Barr virus (EBV) infection.70 In contrast to the genetic profile of immune competent patients, almost all cases lacked mutation of MYD88 and CD79B. Moreover, most cases retained surface HLA expression. There was evidence of an induced immune tolerant microenvironment with increased macrophage infiltration and increased expression of PDL1 and PDL2 on the surface of microenvironmental cells. These findings suggest that EBV may itself drive NFκB activation and immune evasion without the need for the genetic alterations seen in immune competent PCNSL. SCNSL also shows important distinctions. Whilst PCNSL is almost always classified as the ABC subtype of DLBCL, SCNSL covers a more diverse spectrum of biology comprising an equal division of ABC and GCB subtypes.77 Systemic DLBCL of the MCD genetic subtype is associated with extranodal involvement, including sites such as kidney and adrenal, considered to be of high risk for CNS recurrence. MCD DLBCL with secondary CNS involvement may share considerable overlap with PCNSL. However, GCB DLBCL with CNS involvement represents a

M. Roschewski and D.J. Hodson

different biological entity. For example, whilst MYC and BCL2 proteins are highly expressed, translocation of these loci is rarely seen in PCNSL. In contrast, translocation of MYC and/or BCL2 is enriched in GCB DLBCL involving the CNS. Indeed, MYC translocated lymphoma, including high-grade B-cell lymphomas with double-hit and Burkitt lymphoma, have an increased risk of CNS involvement. The important implication of all these contrasts is that biologically directed, therapeutic vulnerabilities of CNS lymphomas may differ depending on whether tumors arise as true PCNSL or, alternatively, in the context of immune deficiency or secondary to system disease. Multiple studies have collectively made tremendous progress in dissecting the biology of CNS lymphomas. They paint a picture of PCNSL as a homogeneous biological entity, which, like systemic MCD DLBCL, is characterized by chronically activated BCR and NFκB signaling combined with a pronounced requirement for immune evasion. This phenotype presents a specific set of therapeutic vulnerabilities, many of which are already being tested in clinical trials. However, there are important biological differences between true PCNSL and other forms of CNS lymphoma that may require a different treatment approach. Rational targeted agents for central nervous system lymphomas The critical question then becomes whether the improved mechanistic understanding of the therapeutic vulnerabilities of CNS lymphomas can be translated to improve clinical outcomes. To this end, rational targeted agents and combinations have been tested in prospective trials (Tables 1 and 2). Notably, targeted agents have been given on indefinite treatment schedules which may increase the toxicity over time. To reduce treatment duration and improve efficacy, many ongoing clinical trials are testing novel combinations with fixed duration treatment schedules (Table 3). Clinical studies in systemic DLBCL first demonstrated that the Bruton’s tyrosine kinase (BTK) inhibitor ibrutinib inhibits chronic active BCR signaling and has selective activity in ABC DLBCL tumors that harbor both MYD88L265P and CD79B mutations.78 Given that this genetic profile is common in PCNSL and subsets of SCNSL, BTK is a highly rational therapeutic target in CNS lymphomas, and multiple inhibitors of BTK are being tested.79-81 Another class of rational targeted agents are the immunomodulatory drugs, lenalidomide and pomalidomide, which exert direct cytotoxic effects, recruitment of NK cells, upregulation of CD80 and CD40, impairment of inflammatory cytokines, and effects on the tumor microenvironment.82 Lenalidomide down-regulates the transcription factor interferon regulatory factor 4 (IRF4) and augments interferon β production to kill ABC DLBCL cell lines in vitro.83 Furthermore, pre-clinical models have demonstrated synergy between lenalidomide and ibrutinib to inhibit NFκB signaling in ABC DLBCL cell lines.84 A third rational strategy involves various forms of immu-

Haematologica | 109 February 2024

393

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

M. Roschewski and D.J. Hodson

Table 1. Prospective studies of novel targeted agents as monotherapy in central nervous system lymphomas. Novel mechanism

Novel agent

Design

Ibrutinib

Phase I

Ibrutinib

Phase II

Tirabrutinib

Phase II

Immunomodulatory

Lenalidomide

Phase I

Inhibitor of mTOR

Temsirolimus

Phase II

Inhibitor of BTK

Study population

Response rate, %

PCNSL, relapsed (N=13) SCNSL, relapsed (N=7) PCNSL, relapsed (N=52) PCNSL, relapsed (N=44) PCNSL, relapsed (N=7) SCNSL, relapsed (N=7) PCNSL, relapsed (N=37)

ORR: 77 CR: 38 ORR: 71 CR/CRu: 57 ORR: 52 CR/CRu: 19 ORR: 64 CR/CRu: 34 ORR: 86 CR: 14 ORR: 57 CR/CRu: 29 ORR: 54 CR/CRu: 22

Response duration

Reference

Median PFS 4.6 mths

Grommes et al.87

Median PFS 7.4 mths Median PFS 4.8 mths

Soussain et al.88

Median PFS 2.9 mths

Narita et al.79

PFS not reported

Rubenstein et al.89

Median PFS 2.1 mths

Korfel et al.86

CNS: central nervous system; PCNSL: primary CNS lymphoma; SCNSL: secondary CNS lymphoma; BTK: Bruton’s tyrosine kinase; mTOR: mammalian target of rapamycin; PI3K: phosphoinositide 3-kinase; N: number; mths: months; ORR: overall response rate; CR: complete response; CRu: unconfirmed CR; PFS: progression-free survival.

Table 2. Prospective studies of novel targeted agents as combination therapy in central nervous system lymphomas. Novel mechanism

Inhibitor of BTK

Immunomodulatory

Targeted agent

Design

Study population

Response rate, %

Ibrutinib + TEDD-R

Phase I

PCNSL (N=18)

ORR: 94 CR/CRu: 86 ORR: 89 CR/CRu: 67 ORR: 67 CR/CRu: 33 ORR: 32 CR/CRu: 29 ORR: 48 CR/CRu: 32

Ibrutinib + HDMTX-R Lenalidomide + rituximab Pomalidomide + dexamethasone

Phase I Phase II Phase I

PCNSL (N=9) SCNSL (N=6) PCNSL (N=50) PCNSL (N=25)

Response duration

Reference

2-year PFS 66%

Lionakis et al.91

Median PFS not reached Median PFS not reported

Grommes et al.93

Median PFS 7.5 mths

Soussain et al.94

Median PFS 5.3 mths

Tun et al.95

N: number; BTK: Bruton’s tyrosine kinase; TEDD-R: temozolomide, etoposide, doxil, dexamethasone, rituximab; PCNSL: primary CNS lymphoma; SCNSL: secondary CNS lymphoma; ORR: overall response rate; CR: complete response; CRu: unconfirmed complete response; PFS: progression-free survival; HD-MTX-R: high-dose methotrexate and rituximab; mths: months.

notherapy, including immune checkpoint inhibitors and chimeric antigen receptor (CAR) T cells. Given the increased surface expression of PDL1 and PDL2 in PCNSL, targeting the PD-1 receptor has a biologic basis. CAR T-cell therapy is also rational for CNS lymphomas since this approach has efficacy in chemotherapy-refractory systemic DLBCL, but it remains unknown as to whether the underlying genetic basis promoting immune evasion in CNS lymphomas will limit the efficacy of these approaches. Targeted agents as monotherapy for central nervous system lymphomas A fundamental limitation of monotherapy for CNS lymphomas is rapid onset of drug resistance, as observed in systemic DLBCL.85 Indeed, despite remarkably high initial response rates, the responses after BTK monotherapy are often incomplete and of short duration.86 In a phase I study, escalating doses of ibrutinib monotherapy were given to 20 patients with relapsed or refractory PCNSL

(N=13) and SCNSL (N=7).87 The clinical activity of ibrutinib was excellent and 10 (77%) PCNSL patients responded, including 5 (38%) who achieved a CR. Yet the durability of response was short and the median PFS was 4.6 months. The clinical activity of ibrutinib for SCNSL was also tested and 5 (71%) patients responded, including 4 (57%) CR, but the median PFS was 7.4 months. A multicenter phase II study tested ibrutinib in 52 patients with relapsed or refractory PCNSL and 27 (52%) patients responded, including 10 (19%) CR, but the median PFS was 3.3 months.88 Taken together, these data show that ibrutinib is highly active in PCNSL, and probably in subsets of SCNSL, but resistance to monotherapy develops quickly. In a phase I study, 14 patients with relapsed or refractory CNS lymphoma received lenalidomide at escalating doses and 6 (86%) patients with PCNSL responded, including one (14%) CR.89 In SCNSL, 4 (57%) responses were observed, including 2 (29%) CR. The duration of responses was variable, but included 4 responses lasting >12 months.

Haematologica | 109 February 2024

394

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

M. Roschewski and D.J. Hodson

Table 3. Ongoing clinical trials testing novel targeted agents or combinations in central nervous system lymphomas. Novel mechanism Inhibitor of BTK

Inhibitor of BTK + chemotherapy

Inhibitor of BTK + Immunomodulator + chemotherapy Inhibitor of BTK + Inhibitor of PI3K Inhibitor of BTK + Immunomodulator Inhibitor of BTK + Immunomodulator + Inhibitor of BCL2 Inhibitor of PD-1 Inhibitor of BTK + Inhibitor of PD-1 Immunomodulator + Inhibitor of PD-1

Targeted agent combination

Design

Zanubrutinib

Phase II

Ibrutinib + TEDD-R

Phase I

Ibrutinib + TEDD-R

Phase II

1) Ibrutinib + MER

Study population PCNSL and SCNSL, relapsed PCNSL, relapsed and untreated SCNSL, relapsed and untreated

Target enrollment, N

Trial reference

NCT05117814

NCT02203526

NCT03964090

Randomized phase II

PCNSL, relapsed

120

NCT04129710

Phase II

PCNSL, untreated

Not reported

Ma et al.81

Phase II

PCNSL, untreated

112

NCT04947319

Zanubrutinib + lenalidomide + MTR

Phase II

PCNSL, untreated

Song et al.80

Ibrutinib + copanlisib

Phase I

PCNSL, relapsed

NCT03581942

Ibrutinib + lenalidomide + rituximab

Phase Ib

PCNSL and SCNSL, relapsed

NCT03703167

Venetoclax, ibrutinib, prednisone, obinutuzumab, revlimid

Phase I

PCNSL and SCNSL, relapsed and untreated

NCT05211336

Nivolumab

Phase II

NCT02857426

Pembrolizumab

Phase II

NCT03255018

Ibrutinib + nivolumab

Phase II

NCT03770416

Acalabrutinib + durvalumab

Phase I

PCNSL, relapsed PCNSL and SCNSL, relapsed PCNSL and SCNSL, relapsed PCNSL and SCNSL, relapsed

NCT04462328

Pomalidomide + nivolumab

Phase I

PCNSL, relapsed

2) Lenalidomide + MER 3) MER Orelabruitnib + MR Tirabrutinib + MTR or R-MVP

NCT03798314

N: number; BTK: Bruton’s tyrosine kinase; TEDD-R: temozolomide, etoposide, doxil, dexamethasone, rituximab; PCNSL: primary central nervous system lymphoma; SCNSL: secondary CNS lymphoma; MER: high-dose methotrexate, etoposide, rituximab; MR: high-dose methotrexate, rituximab; MTR: high-dose methotrexate, temozolomide, rituximab; R-MVP: high-dose methotrexate, rituximab, procarbazine, vincristine; PI3K: phosphoinositide 3-kinase; PD-1: programmed death ligand.

A retrospective series of 4 patients with relapsed and refractory CNS lymphoma treated with the immune checkpoint inhibitor nivolumab demonstrated clinical responses in all patients.90 Two ongoing phase II studies of PD-1 inhibitors as monotherapy are ongoing and have yet to publish clinical results. Combination targeted therapy for central nervous system lymphomas In order to overcome the drug resistance that rapidly occurs after monotherapy, prospective clinical trials

have tested combination regimens for CNS lymphoma. In a phase I study, patients with relapsed or refractory PCNSL (N=13) and untreated PCNSL (N=5) were treated with escalating doses of ibrutinib monotherapy along with temozolomide, etoposide, doxil, dexamethasone, and rituximab (TEDDi-R), and most patients achieved a CR, including patients refractory to HD-MTX-based regimens.91 A follow-up report on this small series suggests that remissions were often durable.92 A major limitation of this regimen was the risk of developing aspergillosis infections when no antifungal prophylaxis was given. This regimen

Haematologica | 109 February 2024

395

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

has been revised to include anti-fungal prophylaxis and is being studied in both PCNSL and SCNSL (clinicaltrials. gov identifiers NCT02203526 and NCT03964090). Ibrutinib has also been safely added to HD-MTX and rituximab in CNS lymphomas.93 In a phase Ib study, 15 patients with relapsed PCNSL (N=9), relapsed SCNSL (N=3), and untreated SCNSL (N=3) received HD-MTX, rituximab, and ibrutinib. No unexpected toxicities were observed and no cases of aspergillosis were reported. In PCNSL, 8 (89%) patients responded, including 6 (67%) CR. The durability of response to the regimen was unclear as 4 of the responding patients received ASCT consolidation. In 4 cases without consolidation, 3 remained in remission >10 months. In SCNSL, 4 (67%) patients responded and 2 (33%) achieved a CR. Two responding patients progressed after a remission of <6 months, one received an ASCT, and one had an ongoing remission of 6 months. Numerous combination regimens using ibrutinib and second-generation BTK inhibitors, including zanubrutinib, acalabrutinib, tirabrutinib and orelabrutinib, are now being tested in ongoing clinical trials. Lenalidomide and pomalidomide have both been tested as combination therapy for relapsed or refractory PCNSL.94,95 In a multicenter phase II study, 50 patients with relapsed or refractory PCNSL received lenalidomide and rituximab for 8 cycles and responding patients continued lenalidomide monotherapy for another 12 cycles.94 The overall response rate was 32%, including 13 (29%) CR, but the median PFS was only 7.8 months. In a phase I study, escalating doses of pomalidomide were tested in combination with dexamethasone for 2 cycles followed by pomalidomide monotherapy until disease progression.95 In 25 evaluable patients, the overall response rate was 48% including 8 (32%) CR. The median PFS for this study was 5.3 months. Taken together, these results support immunomodulatory agents as clinically active and biologically rational targeted agents for CNS lymphomas that are likely to be most effective as part of combination therapy. CAR T-cell therapies targeting CD19 can overcome chemotherapy resistance in systemic DLBCL, and have been tested in small studies for CNS lymphoma.96,97 One concern about CAR T-cell therapy for CNS lymphomas is related to a potential increased incidence in immune cell-associated neurotoxicity syndrome (ICANS). In a pilot study of 12 patients with PCNSL, tisagenlecleucel resulted in a response in 7 (58%) patients, including 6 (50%) CR. Notably, only one patient had grade 3 ICANS. In another pilot study of axicabtagene ciloleucel in 6 patients with PCNSL and 3 patients with SCNSL, a CR was achieved by 6 (86%) patients with 3 months of follow-up and only one case of grade 3 or higher ICANS. These results suggest that effector CAR T cells cross the blood brain barrier and can induce remissions in CNS lymphoma. Indeed, a meta-analysis of 128 patients with CNS lymphoma suggested that the safety and efficacy of anti-CD19 CAR T-cell therapy is similar to that observed

M. Roschewski and D.J. Hodson

in systemic DLBCL.98 These strategies have only very short follow-up and many questions remain regarding persistence of CAR T cells in the CSF, mechanisms of treatment failure, and if biologic differences between PCNSL and SCNSL will impact treatment efficacy.

Fundamental questions and future directions Primary and secondary DLBCL involving the CNS pose unique clinical challenges regarding delivery of therapy into tumor tissue, treatment-related toxicities, and the diversity of the underlying biology. Recent insights regarding the molecular biology of PCNSL and DLBCL subsets with CNS predilection, along with an expanding list of effective targeted agents, have introduced important fundamental questions that can only be answered with prospective clinical trials. What are the safest and most effective targeted agents for CNS lymphoma, and which combinations are tolerable in patients of all ages? Will predictive biomarkers based on the underlying biology be identified, and will they be reliable enough to select therapy? Will technological advances such as next-generation sequencing of cell-free DNA in the plasma and/or CSF help in clinical decision-making?99 Well-designed clinical trials with strong translational molecular endpoints will be essential to address these questions and ultimately improve the cure rate of CNS lymphoma. Disclosures No conflicts of interest to disclose. Contributions MR and DJH contributed equally to this manuscript. Acknowledgments The authors would like to thank patients and their families who participate in clinical trials testing novel agents in CNS lymphoma. DJH was supported by a Fellowship from CRUK (RCCFEL\100072). Funding Research in the Hodson group is funded in part by the Wellcome Trust who support the Wellcome-MRC Cambridge Stem Cell Institute (203151/Z/16/Z), the CRUK Cambridge Major Centre (C49940/A25117), and the NIHR Cambridge Biomedical Research Centre (BRC-1215-20014). The views expressed are those of the authors and not necessarily those of the NIHR or the UK Department of Health and Social Care. Data-sharing statement Relevant primary source scientific publications are cited in the References section.

Haematologica | 109 February 2024

396

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma

M. Roschewski and D.J. Hodson

References 1. Ostrom QT, Gittleman H, Truitt G, et al. CBTRUS Statistical Report: primary brain and other central nervous system tumors diagnosed in the United States in 2011-2015. Neuro Oncol. 2018;20(Suppl 4):iv1-iv86. 2. Villano JL, Koshy M, Shaikh H, et al. Age, gender, and racial differences in incidence and survival in primary CNS lymphoma. Br J Cancer. 2011;105(9):1414-1418. 3. Shiels MS, Pfeiffer RM, Besson C, et al. Trends in primary central nervous system lymphoma incidence and survival in the U.S. Br J Haematol. 2016;174(3):417-424. 4. Rubenstein JL, Hsi ED, Johnson JL, et al. Intensive chemotherapy and immunotherapy in patients with newly diagnosed primary CNS lymphoma: CALGB 50202 (Alliance 50202). J Clin Oncol. 2013;31(25):3061-3068. 5. Houillier C, Dureau S, Taillandier L, et al. Radiotherapy or autologous stem-cell transplantation for primary CNS lymphoma in patients age 60 years and younger: long-term results of the randomized phase II PRECIS study. J Clin Oncol. 2022;40(32):3692-3698. 6. Omuro A, Correa DD, DeAngelis LM, et al. R-MPV followed by high-dose chemotherapy with TBC and autologous stem-cell transplant for newly diagnosed primary CNS lymphoma. Blood. 2015;125(9):1403-1410. 7. Ferreri AJ, Cwynarski K, Pulczynski E, et al. Chemoimmunotherapy with methotrexate, cytarabine, thiotepa, and rituximab (MATRix regimen) in patients with primary CNS lymphoma: results of the first randomisation of the International Extranodal Lymphoma Study Group-32 (IELSG32) phase 2 trial. Lancet Haematol. 2016;3(5):e217-227. 8. Fox CP, Phillips EH, Smith J, et al. Guidelines for the diagnosis and management of primary central nervous system diffuse large B-cell lymphoma. Br J Haematol. 2019;184(3):348-363. 9. Ambady P, Holdhoff M, Bonekamp D, et al. Late relapses in primary CNS lymphoma after complete remissions with highdose methotrexate monotherapy. CNS Oncol. 2015;4(6):393-398. 10. Biccler JL, Savage KJ, Brown PDN, et al. Risk of death, relapse or progression, and loss of life expectancy at different progression-free survival milestones in primary central nervous system lymphoma. Leuk Lymphoma. 2019;60(10):2516-2523. 11. Pulczynski EJ, Kuittinen O, Erlanson M, et al. Successful change of treatment strategy in elderly patients with primary central nervous system lymphoma by de-escalating induction and introducing temozolomide maintenance: results from a phase II study by the Nordic Lymphoma Group. Haematologica. 2015;100(4):534-540. 12. Fritsch K, Kasenda B, Schorb E, et al. High-dose methotrexatebased immuno-chemotherapy for elderly primary CNS lymphoma patients (PRIMAIN study). Leukemia. 2017;31(4):846-852. 13. Omuro A, Chinot O, Taillandier L, et al. Methotrexate and temozolomide versus methotrexate, procarbazine, vincristine, and cytarabine for primary CNS lymphoma in an elderly population: an intergroup ANOCEF-GOELAMS randomised phase 2 trial. Lancet Haematol. 2015;2(6):e251-259. 14. Langner-Lemercier S, Houillier C, Soussain C, et al. Primary CNS lymphoma at first relapse/progression: characteristics, management, and outcome of 256 patients from the French LOC network. Neuro Oncol. 2016;18(9):1297-1303. 15. Bernstein SH, Unger JM, Leblanc M, et al. Natural history of

CNS relapse in patients with aggressive non-Hodgkin's lymphoma: a 20-year follow-up analysis of SWOG 8516 - the Southwest Oncology Group. J Clin Oncol. 2009;27(1):114-119. 16. El-Galaly TC, Cheah CY, Bendtsen MD, et al. Treatment strategies, outcomes and prognostic factors in 291 patients with secondary CNS involvement by diffuse large B-cell lymphoma. Eur J Cancer. 2018;93:57-68. 17. Wilson MR, Eyre TA, Kirkwood AA, et al. Timing of high-dose methotrexate CNS prophylaxis in DLBCL: a multicenter international analysis of 1384 patients. Blood. 2022;139(16):2499-2511. 18. Bobillo S, Khwaja J, Ferreri AJM, et al. Prevention and management of secondary central nervous system lymphoma. Haematologica. 2023;108(3):673-689. 19. David KA, Sundaram S, Kim SH, et al. Older patients with primary central nervous system lymphoma: survival and prognostication across 20 U.S. cancer centers. Am J Hematol. 2023;98(6):900-912. 20. Berry MP, Simpson WJ. Radiation therapy in the management of primary malignant lymphoma of the brain. Int J Radiat Oncol Biol Phys. 1981;7(1):55-59. 21. Cher L, Glass J, Harsh GR, Hochberg FH. Therapy of primary CNS lymphoma with methotrexate-based chemotherapy and deferred radiotherapy: preliminary results. Neurology. 1996;46(6):1757-1759. 22. Ferreri AJ, Reni M, Foppoli M, et al. High-dose cytarabine plus high-dose methotrexate versus high-dose methotrexate alone in patients with primary CNS lymphoma: a randomised phase 2 trial. Lancet. 2009;374(9700):1512-1520. 23. Colombat P, Lemevel A, Bertrand P, et al. High-dose chemotherapy with autologous stem cell transplantation as first-line therapy for primary CNS lymphoma in patients younger than 60 years: a multicenter phase II study of the GOELAMS group. Bone Marrow Transplant. 2006;38(6):417-420. 24. Abrey LE, Moskowitz CH, Mason WP, et al. Intensive methotrexate and cytarabine followed by high-dose chemotherapy with autologous stem-cell rescue in patients with newly diagnosed primary CNS lymphoma: an intent-totreat analysis. J Clin Oncol. 2003;21(22):4151-4156. 25. Scordo M, Wang TP, Ahn KW, et al. Outcomes associated with thiotepa-based conditioning in patients with primary central nervous system lymphoma after autologous hematopoietic cell transplant. JAMA Oncol. 2021;7(7):993-1003. 26. Schorb E, Fox CP, Fritsch K, et al. High-dose thiotepa-based chemotherapy with autologous stem cell support in elderly patients with primary central nervous system lymphoma: a European retrospective study. Bone Marrow Transplant. 2017;52(8):1113-1119. 27. Schorb E, Isbell L, Kerkhoff A, et al. High-dose chemotherapy and autologous stem cell transplant in elderly and fit primary CNS lymphoma patients - a multicenter study by the Cooperative PCNSL Study Group (MARTA study). Blood. 2022;140(Suppl 1):1773-1774. 28. Hollender A, Kvaloy S, Lote K, et al. Prognostic factors in 140 adult patients with non-Hodgkin's lymphoma with systemic central nervous system (CNS) involvement. A single centre analysis. Eur J Cancer. 2000;36(14):1762-1768. 29. Doolittle ND, Abrey LE, Shenkier TN, et al. Brain parenchyma involvement as isolated central nervous system relapse of systemic non-Hodgkin lymphoma: an International Primary CNS

Haematologica | 109 February 2024

397

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma Lymphoma Collaborative Group report. Blood. 2008;111(3):1085-1093. 30. Wight JC, Yue M, Keane C, et al. Outcomes of synchronous systemic and central nervous system (CNS) involvement of diffuse large B-cell lymphoma are dictated by the CNS disease: a collaborative study of the Australasian Lymphoma Alliance. Br J Haematol. 2019;187(2):174-184. 31. Khwaja J, Kirkwood AA, Isbell LK, et al. International multicenter retrospective analysis of thiotepa-based autologous stem cell transplantation for secondary central nervous system lymphoma. Haematologica. 2023;108(3):882-888. 32. Korfel A, Elter T, Thiel E, et al. Phase II study of central nervous system (CNS)-directed chemotherapy including high-dose chemotherapy with autologous stem cell transplantation for CNS relapse of aggressive lymphomas. Haematologica. 2013;98(3):364-370. 33. Ferreri AJ, Donadoni G, Cabras MG, et al. High doses of antimetabolites followed by high-dose sequential chemoimmunotherapy and autologous stem-cell transplantation in patients with systemic B-cell lymphoma and secondary CNS involvement: final results of a multicenter phase II trial. J Clin Oncol. 2015;33(33):3903-3910. 34. Doorduijn JK, van Imhoff GW, van der Holt B, et al. Treatment of secondary central nervous system lymphoma with intrathecal rituximab, high-dose methotrexate, and R-DHAP followed by autologous stem cell transplantation: results of the HOVON 80 phase 2 study. Hematol Oncol. 2017;35(4):497-503. 35. Ferreri AJM, Doorduijn JK, Re A, et al. MATRix-RICE therapy and autologous haematopoietic stem-cell transplantation in diffuse large B-cell lymphoma with secondary CNS involvement (MARIETTA): an international, single-arm, phase 2 trial. Lancet Haematol. 2021;8(2):e110-e121. 36. Chapuy B, Stewart C, Dunford AJ, et al. Molecular subtypes of diffuse large B cell lymphoma are associated with distinct pathogenic mechanisms and outcomes. Nat Med. 2018;24(5):679-690. 37. Schmitz R, Wright GW, Huang DW, et al. Genetics and pathogenesis of diffuse large B-cell lymphoma. N Engl J Med. 2018;378(15):1396-1407. 38. Lacy SE, Barrans SL, Beer PA, et al. Targeted sequencing in DLBCL, molecular subtypes, and outcomes: a Haematological Malignancy Research Network report. Blood. 2020;135(20):1759-1771. 39. Wright GW, Huang DW, Phelan JD, et al. A probabilistic classification tool for genetic subtypes of diffuse large B cell lymphoma with therapeutic implications. Cancer Cell. 2020;37(4):551-568.e14. 40. Camilleri-Broet S, Criniere E, Broet P, et al. A uniform activated B-cell-like immunophenotype might explain the poor prognosis of primary central nervous system lymphomas: analysis of 83 cases. Blood. 2006;107(1):190-196. 41. Bödör C, Alpár D, Marosvári D, et al. Molecular subtypes and genomic profile of primary central nervous system lymphoma. J Neuropathol Exp Neurol. 2020;79(2):176-183. 42. Montesinos-Rongen M, Godlewska E, Brunn A, et al. Activating L265P mutations of the MYD88 gene are common in primary central nervous system lymphoma. Acta Neuropathol. 2011;122(6):791-792. 43. Gonzalez-Aguilar A, Idbaih A, Boisselier B, et al. Recurrent mutations of MYD88 and TBL1XR1 in primary central nervous system lymphomas. Clin Cancer Res. 2012;18(19):5203-5211. 44. Montesinos-Rongen M, Schäfer E, Siebert R, et al. Genes regulating the B cell receptor pathway are recurrently mutated

in primary central nervous system lymphoma. Acta Neuropathol. 2012;124(6):905-906. 45. Nakamura T, Tateishi K, Niwa T, et al. Recurrent mutations of CD79B and MYD88 are the hallmark of primary central nervous system lymphomas. Neuropathol Appl Neurobiol. 2016;42(3):279-290. 46. Yamada S, Ishida Y, Matsuno A, et al. Primary diffuse large B-cell lymphomas of central nervous system exhibit remarkably high prevalence of oncogenic MYD88 and CD79B mutations. Leuk Lymphoma. 2015;56(7):2141-2145. 47. Braggio E, Van Wier S, Ojha J, et al. Genome-wide analysis uncovers novel recurrent alterations in primary central nervous system lymphomas. Clin Cancer Res. 2015;21(17):3986-3994. 48. Fukumura K, Kawazu M, Kojima S, et al. Genomic characterization of primary central nervous system lymphoma. Acta Neuropathol. 2016;131(6):865-875. 49. Chapuy B, Roemer MG, Stewart C, et al. Targetable genetic features of primary testicular and primary central nervous system lymphomas. Blood. 2016;127(7):869-881. 50. Nayyar N, White MD, Gill CM, et al. MYD88 L265P mutation and CDKN2A loss are early mutational events in primary central nervous system diffuse large B-cell lymphomas. Blood Adv. 2019;3(3):375-383. 51. Radke J, Ishaque N, Koll R, et al. The genomic and transcriptional landscape of primary central nervous system lymphoma. Nat Commun. 2022;13(1):2558. 52. Hernandez-Verdin I, Kirasic E, Wienand K, et al. Molecular and clinical diversity in primary central nervous system lymphoma. Ann Oncol. 2023;34(2):186-199. 53. Chapuy B, Roemer MGM, Stewart C, et al. Targetable genetic features of primary testicular and primary central nervous system lymphomas. Blood. 2016;127(7):869-881. 54. Ngo VN, Young RM, Schmitz R, et al. Oncogenically active MYD88 mutations in human lymphoma. Nature. 2011;470(7332):115-119. 55. Davis RE, Ngo VN, Lenz G, et al. Chronic active B-cell-receptor signalling in diffuse large B-cell lymphoma. Nature. 2010;463(7277):88-92. 56. Montesinos-Rongen M, Schmitz R, Courts C, et al. Absence of immunoglobulin class switch in primary lymphomas of the central nervous system. Am J Pathol. 2005;166(6):1773-1779. 57. Phelan JD, Young RM, Webster DE, et al. A multiprotein supercomplex controlling oncogenic signalling in lymphoma. Nature. 2018;560(7718):387-391. 58. Wang JQ, Jeelall YS, Humburg P, et al. Synergistic cooperation and crosstalk between MYD88(L265P) and mutations that dysregulate CD79B and surface IgM. J Exp Med. 2017;214(9):2759-2776. 59. Montesinos-Rongen M, Küppers R, Schlüter D, et al. Primary central nervous system lymphomas are derived from germinalcenter B cells and show a preferential usage of the V4–34 gene segment. Am J Pathol. 1999;155(6):2077-2086. 60. Thompsett AR, Ellison DW, Stevenson FK, et al. VH gene sequences from primary central nervous system lymphomas indicate derivation from highly mutated germinal center B cells with ongoing mutational activity. Blood. 1999;94(5):1738-1746. 61. Montesinos-Rongen M, Purschke F, Küppers R, et al. Immunoglobulin repertoire of primary lymphomas of the central nervous system. J Neuropathol Exp Neurol. 2014;73(12):1116-1125. 62. Montesinos-Rongen M, Terrao M, May C, et al. The process of somatic hypermutation increases polyreactivity for central nervous system antigens in primary central nervous system

Haematologica | 109 February 2024

398

M. Roschewski and D.J. Hodson

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma lymphoma. Haematologica. 2021;106(3):708-717. 63. Montesinos-Rongen M, Purschke FG, Brunn A, et al. Primary central nervous system (CNS) lymphoma B cell receptors recognize CNS proteins. J Immunol. 2015;195(3):1312-1319. 64. Takahara T, Nakamura S, Tsuzuki T, Satou A. The immunology of DLBCL. Cancers (Basel). 2023;15(3):835. 65. Radke J, Ishaque N, Koll R, et al. The genomic and transcriptional landscape of primary central nervous system lymphoma. Nat Commun. 2022;13(1):2558. 66. Riemersma SA, Jordanova ES, Schop RFJ, et al. Extensive genetic alterations of the HLA region, including homozygous deletions of HLA class II genes in B-cell lymphomas arising in immune-privileged sites. Blood. 2000;96(10):3569-3577. 67. Riemersma SA, Oudejans JJ, Vonk MJ, et al. High numbers of tumour-infiltrating activated cytotoxic T lymphocytes, and frequent loss of HLA class I and II expression, are features of aggressive B cell lymphomas of the brain and testis. J Pathol. 2005;206(3):328-336. 68. Minderman M, Amir A, Kraan W, et al. Immune evasion in primary testicular and central nervous system lymphomas: HLA loss rather than 9p24.1/PD-L1/PD-L2 alterations. Blood. 2021;138(13):1194-1197. 69. Schwindt H, Vater I, Kreuz M, et al. Chromosomal imbalances and partial uniparental disomies in primary central nervous system lymphoma. Leukemia. 2009;23(10):1875-1884. 70. Gandhi MK, Hoang T, Law SC, et al. EBV-associated primary CNS lymphoma occurring after immunosuppression is a distinct immunobiological entity. Blood. 2021;137(11):1468-1477. 71. Challa-Malladi M, Lieu Yen K, Califano O, et al. Combined genetic inactivation of β2-microglobulin and CD58 reveals frequent escape from immune recognition in diffuse large B cell lymphoma. Cancer Cell. 2011;20(6):728-740. 72. Heming M, Haessner S, Wolbert J, et al. Intratumor heterogeneity and T cell exhaustion in primary CNS lymphoma. Genome Med. 2022;14(1):109. 73. Braggio E, Van Wier S, Ojha J, et al. Genome-wide analysis uncovers novel recurrent alterations in primary central nervous system lymphomas. Clin Cancer Res. 2015;21(17):3986-3994. 74. Venturutti L, Teater M, Zhai A, et al. TBL1XR1 mutations drive extranodal lymphoma by inducing a pro-tumorigenic memory fate. Cell. 2020;182(2):297-316.e27. 75. Venturutti L, Rivas MA, Pelzer BW, et al. An aged/autoimmune B-cell program defines the early transformation of extranodal lymphomas. Cancer Discov. 2023;13(1):216-243. 76. Venturutti L, Melnick AM. The dangers of deja vu: memory B cells as the cells of origin of ABC-DLBCLs. Blood. 2020;136(20):2263-2274. 77. Bodor C, Alpar D, Marosvari D, et al. Molecular subtypes and genomic profile of primary central nervous system lymphoma. J Neuropathol Exp Neurol. 2020;79(2):176-183. 78. Wilson WH, Young RM, Schmitz R, et al. Targeting B cell receptor signaling with ibrutinib in diffuse large B cell lymphoma. Nat Med. 2015;21(8):922-926. 79. Narita Y, Nagane M, Mishima K, et al. Phase I/II study of tirabrutinib, a second-generation Bruton's tyrosine kinase inhibitor, in relapsed/refractory primary central nervous system lymphoma. Neuro Oncol. 2021;23(1):122-133. 80. Song J, Liu H, Jiao Z, et al. Zanubrutinib, lenalidomide, rituximab, temozolomide and methotrexate (RLZT±MTX) as first-line treatment for newly diagnosed PCNSL: a prospective, open-lable, multicenter clinical trial. Blood. 2022;140(Suppl 1):3739-3740. 81. Ma J, He Q, Lu K, Wang L. A multicenter, single-arm, prospective

phase II study of orelabrutinib combined with high-dose methotrexate and rituximab sequential autologous hematopoietic stem cell transplantation in newly-dignosed primary central nervous system lymphoma. Blood. 2022;140(Suppl 1):3744-3745. 82. Tageja N. Lenalidomide - current understanding of mechanistic properties. Anticancer Agents Med Chem. 2011;11(3):315-326. 83. Zhang LH, Kosek J, Wang M, et al. Lenalidomide efficacy in activated B-cell-like subtype diffuse large B-cell lymphoma is dependent upon IRF4 and cereblon expression. Br J Haematol. 2013;160(4):487-502. 84. Yang Y, Shaffer AL, Emre NC, et al. Exploiting synthetic lethality for the therapy of ABC diffuse large B cell lymphoma. Cancer Cell. 2012;21(6):723-737. 85. Shaffer AL, Phelan JD, Wang JQ, et al. Overcoming acquired epigenetic resistance to BTK inhibitors. Blood Cancer Discov. 2021;2(6):630-647. 86. Korfel A, Schlegel U, Herrlinger U, et al. Phase II trial of temsirolimus for relapsed/refractory primary CNS lymphoma. J Clin Oncol. 2016;34(15):1757-1763. 87. Grommes C, Pastore A, Palaskas N, et al. Ibrutinib unmasks critical role of Bruton tyrosine kinase in primary CNS lymphoma. Cancer Discov. 2017;7(9):1018-1029. 88. Soussain C, Choquet S, Blonski M, et al. Ibrutinib monotherapy for relapse or refractory primary CNS lymphoma and primary vitreoretinal lymphoma: final analysis of the phase II 'proof-ofconcept' iLOC study by the Lymphoma study association (LYSA) and the French oculo-cerebral lymphoma (LOC) network. Eur J Cancer. 2019;117:121-130. 89. Rubenstein JL, Geng H, Fraser EJ, et al. Phase 1 investigation of lenalidomide/rituximab plus outcomes of lenalidomide maintenance in relapsed CNS lymphoma. Blood Adv. 2018;2(13):1595-1607. 90. Nayak L, Iwamoto FM, LaCasce A, et al. PD-1 blockade with nivolumab in relapsed/refractory primary central nervous system and testicular lymphoma. Blood. 2017;129(23):3071-3073. 91. Lionakis MS, Dunleavy K, Roschewski M, et al. Inhibition of B cell receptor signaling by ibrutinib in primary CNS lymphoma. Cancer Cell. 2017;31(6):833-843.e5. 92. Roschewski M, Lionakis MS, Melani C, et al. Dose-adjusted Teddi-R induces durable complete remissions in relapsed and refractory primary CNS lymphoma. Blood. 2018;132(Suppl 1):4195. 93. Grommes C, Tang SS, Wolfe J, et al. Phase 1b trial of an ibrutinib-based combination therapy in recurrent/refractory CNS lymphoma. Blood. 2019;133(5):436-445. 94. Ghesquieres H, Chevrier M, Laadhari M, et al. Lenalidomide in combination with intravenous rituximab (REVRI) in relapsed/ refractory primary CNS lymphoma or primary intraocular lymphoma: a multicenter prospective 'proof of concept' phase II study of the French Oculo-Cerebral lymphoma (LOC) Network and the Lymphoma Study Association (LYSA)†. Ann Oncol. 2019;30(4):621-628. 95. Tun HW, Johnston PB, DeAngelis LM, et al. Phase 1 study of pomalidomide and dexamethasone for relapsed/refractory primary CNS or vitreoretinal lymphoma. Blood. 2018;132(21):2240-2248. 96. Frigault MJ, Dietrich J, Gallagher K, et al. Safety and efficacy of tisagenlecleucel in primary CNS lymphoma: a phase 1/2 clinical trial. Blood. 2022;139(15):2306-2315. 97. Jacobson CA, Falvey C, Bouvier R, et al. A pilot study of axicabtagene ciloleucel (axi-cel) for the treatment of relapsed/ refractory primary and secondary central nervous system

Haematologica | 109 February 2024

399

M. Roschewski and D.J. Hodson

REVIEW ARTICLE - Biology and targeted therapy for CNS lymphoma lymphoma (CNSL). Blood. 2022;140(Suppl 1):1060-1061. 98. Cook MR, Dorris CS, Makambi KH, et al. Toxicity and efficacy of CAR T-cell therapy in PCNSL and SCNSL: a meta-analysis of 128 patients. Blood Adv. 2023;7(1):32-39.

99. Mutter JA, Alig SK, Esfahani MS, et al. Circulating tumor DNA profiling for detection, risk stratification, and classification of brain lymphomas. J Clin Oncol. 2023;41(9):1684-1694.

Haematologica | 109 February 2024

400

M. Roschewski and D.J. Hodson

ARTICLE - Acute Myeloid Leukemia

Quantification of measurable residual disease using duplex sequencing in adults with acute myeloid leukemia Laura W. Dillon,1* Jake Higgins,2* Hassan Nasif,3 Megan Othus,3 Lan Beppu,4 Thomas H. Smith,2 Elizabeth Schmidt,2 Charles C. Valentine III,2 Jesse J. Salk,2 Brent L. Wood,5 Harry P.

Correspondence: C.S. Hourigan

Erba, Jerald P. Radich

hourigan@nih.gov

4,7#

and Christopher S. Hourigan

1,8,#

Laboratory of Myeloid Malignancies, Hematology Branch, National Heart, Lung, and Blood

Institute, National Institutes of Health, Bethesda, MD; 2TwinStrand Biosciences, Seattle, WA;

Received: Accepted: Early view:

May 15, 2023. July 28, 2023. August 3, 2023.

Public Health Sciences Division, Fred Hutchinson Cancer Center, Seattle, WA; 4Clinical

https://doi.org/10.3324/haematol.2023.283520

Research Division, Fred Hutchinson Cancer Center, Seattle, WA; 5Department of Pathology

and Laboratory Medicine, Children’s Hospital Los Angeles, Los Angeles, CA; Duke University 6

School of Medicine, Durham, NC; 7Division of Medical Oncology, Department of Medicine, University of Washington, Seattle, WA and 8Myeloid Malignancies Program, National Institutes of Health, Bethesda, MD, USA LWD and JH contributed equally as first authors.

JPR and CSH contributed equally as senior authors.

Abstract The presence of measurable residual disease (MRD) is strongly associated with treatment outcomes in acute myeloid leukemia (AML). Despite the correlation with clinical outcomes, MRD assessment has yet to be standardized or routinely incorporated into clinical trials and discrepancies have been observed between different techniques for MRD assessment. In 62 patients with AML, aged 18-60 years, in first complete remission after intensive induction therapy on the randomized phase III SWOG-S0106 clinical trial (clinicaltrials gov. Identifier: NCT00085709), MRD detection by centralized, high-quality multiparametric flow cytometry was compared with a 29-gene panel utilizing duplex sequencing (DS), an ultrasensitive next-generation sequencing method that generates double-stranded consensus sequences to reduce false positive errors. MRD as defined by DS was observed in 22 (35%) patients and was strongly associated with higher rates of relapse (68% vs. 13%; hazard ratio [HR] =8.8; 95% confidence interval [CI]: 3.2-24.5; P<0.001) and decreased survival (32% vs. 82%; HR=5.6; 95% CI: 2.3-13.8; P<0.001) at 5 years. DS MRD strongly outperformed multiparametric flow cytometry MRD, which was observed in ten (16%) patients and marginally associated with higher rates of relapse (50% vs. 30%; HR=2.4; 95% CI: 0.9-6.7; P=0.087) and decreased survival (40% vs. 68%; HR=2.5; 95% CI: 1.0-6.3; P=0.059) at 5 years. Furthermore, the prognostic significance of DS MRD status at the time of remission for subsequent relapse was similar on both randomized arms of the trial. These findings suggest that next-generation sequencing-based AML MRD testing is a powerful tool that could be developed for use in patient management and for early anti-leukemic treatment assessment in clinical trials.

Introduction Acute myeloid leukemia (AML) is a rare blood cancer diagnosed in approximately 20,000 Americans annually. While most patients treated with chemotherapy will achieve an initial complete remission (CR), less than one-third are expected to survive after 5 years.1,2 Measurable residual disease (MRD) is the presence of leukemia below the threshold set for remission by traditional clinical criteria but detectable with higher sensitivity approaches.3 The presence of MRD is strongly associated with treatment outcomes.4,5 However, despite being well

established as correlated with the antileukemic effect of treatment interventions,6-11 clinical implementation has been limited. While no standard technique is currently used for AML MRD testing,12 multiple methodologies exist including detection of aberrant cell surface protein expression by multiparametric flow cytometry (MFC) or detection of genetic alterations by molecular assays.13 MFC has been widely used for AML MRD detection, but there are concerns that inter-laboratory variability and a lack of standardization could limit applicability of the technique on a broader scale.14,15 MFC and next-generation sequencing (NGS) have been found to provide discordant

Haematologica | 109 February 2024

401

ARTICLE - Duplex sequencing for measurable residual disease in AML

MRD results,16,17 potentially capturing different residual cell populations. Furthermore, while MRD detection of certain highly prevalent genetic variants, including FLT3 internal tandem duplications (FLT3-ITD) and NPM1 insertions, by NGS has been shown to be strongly correlated with adverse clinical outcomes,10,11,18 decentralized flow cytometry on the same patients was not prognostic.11 There remains a need to compare AML MRD assessment using both centralized, high-quality MFC and ultra-sensitive NGS for detection of a broad range of variants in the same patients. The SWOG Cancer Research Network S0106 study was an open-label randomized phase III clinical trial of adults aged 18-60 years with previously untreated de novo nonacute promyelocytic leukemia (non-APL) AML comparing standard induction therapy with daunorubicin (60 mg/m2 intravenously [IV] day [d]1,2,3) and cytarabine (100 mg/m2/d continuous infusion d1-7) (“DA”) against the combination of daunorubicin (45 mg/m2 d1,2,3), cytarabine (100 mg/m2/d continuous infusion d1-7), and gemtuzumab ozogamicin (6 mg/m2 d4) (“DA+GO”). Rates of cytomorphological CR (69% and 70%), 5-year relapse-free survival (RFS, 43% and 42%), and 5-year overall survival (OS, 46% and 50%) have previously been reported as not different between DA and DA+GO arms respectively.19 Utilizing samples and clinical data from patients treated on the S0106 trial, we explored the utility of MRD to predict treatment outcomes by both MFC and NGS. Bone marrow (BM) specimens obtained prior to treatment and at time of CR underwent centralized, prospective assessment of MRD using a three-tube, ten-color MFC assay.20 Banked samples from a total of 67 patients were available at diagnosis and CR after first induction, and 62 patients with trackable variants identified using a 29-gene NGS panel at diagnosis underwent retrospective genomic analysis with error-corrected duplex sequencing (DS) for MRD at time of CR.

Methods Patients Archival BM aspirates or peripheral blood (PB) from 67 patients enrolled on the SWOG trial S0106 (clinicaltrials gov. Identifier: NCT00085709) were available for this study. A total of 62 patients were selected for MRD analysis if they (i) achieved a first morphological CR with protocol induction therapy, (ii) had both diagnosis and remission samples after first induction, (iii) had central flow cytometry results on their remission BM, and (iv) had a variant detected at diagnosis for tracking in remission. Samples described in this manuscript were collected at time of first morphologic CR, but if CR samples were not available the first sample collected after achieving CR was used. BM (n=56) and PB (n=6) remission samples were collected a median of 34 days (range, 25-162) post-randomization

L.W. Dillon et al.

and a median of 0 days (range, -6 to 121) from clinically defined remission. The Institutional Review Board of the Fred Hutchinson Cancer Center gave ethical approval for this work, and patients were treated according to the Declaration of Helsinki. Duplex sequencing Retrospective targeted DNA sequencing of 29 genes recurrently mutated in adult AML was performed on genomic DNA (gDNA) collected from paired diagnostic and remission BM or PB samples utilizing the TwinStrand Duplex SequencingTM AML-29 Panel (Online Supplementary Table S1). Non-error corrected sequencing was performed on diagnostic samples (500 ng gDNA) and error-corrected DS was performed on remission samples (1μg gDNA). DS was performed essentially as described21 and further detailed in the Online Supplementary Appendix. Bioinformatics Alignment, duplex consensus sequence generation, and variant calling were performed as described.21 For each patient, potential germline variants were identified and excluded from the analysis if the variant allele fraction (VAF) was ≥35% at both diagnosis and remission, or ≥40% at either time point and a gnomAD allele frequency ≥0.05. Somatic variants present at diagnosis were classified as potentially deleterious if computationally predicted as such and with a VAF ≥ 5% (≥1% for FLT3-ITD/NPM1 insertions). Somatic variants in remission followed the same classification rules for deleterious impact and required an alternative depth of ≥2 (≥1 for FLT3-ITD/NPM1 insertions detected at diagnosis). All remaining variants were manually curated for pathogenicity. MRD by NGS was defined using conditions previously identified as prognostic.7,11 Multiparametric flow cytometry BM samples collected at diagnosis and remission were analyzed for MRD using a three-tube, ten-color MFC assay with a sensitivity of 0.1% in most cases; data and details of which were reported previously.20 Statistics Morphologic complete remission was defined per contemporary consensus criteria definitions and required count recovery with absolute neutrophil count >1,000 and platelets >100,000. Time-to-event outcomes analyzed were OS (event=death), RFS (event=relapse or death) and time to relapse (TTR; event=relapse, death in remission a competing event). All outcomes were measured from date of morphologic remission to date of event, with patients without event censored at date of last contact. Associations between residual disease and outcomes were assessed using Cox regression models (cause-specific model for TTR); model discrimination was assessed using C-statistics.

Haematologica | 109 February 2024

402

ARTICLE - Duplex sequencing for measurable residual disease in AML

L.W. Dillon et al.

Results

1-9) that could be tracked in remission.

Patient characteristics The median age of the 62 patients in this study was 48 years (range, 18-60) (Table 1). Thirty-two patients were randomized to DA and 30 patients to DA+GO. At 5 years, the rate of non-relapse mortality (NRM) was 10%, relapse was 33%, RFS was 57%, and OS was 64% for the entire cohort (Figure 1). Overall patient demographics and clinical outcomes of the 62 patients analyzed in this study align with the full S0106 clinical trial cohort (Online Supplementary Table S2). Targeted sequencing analysis of diagnostic samples at an average raw sequencing read depth of 279x utilizing a 29 gene panel identified a total of 172 potentially deleterious variants across the 62 patients. Variants had a median VAF of 34% (range, 1.4-91.5) and were detected in 23 genes, with FLT3 being the most frequently mutated (Online Supplementary Table S3; Online Supplementary Figure S1). Patients had a median of two variants detected at diagnosis (range

Technical performance of duplex sequencing Technical performance of the 29-gene DS assay was assessed on contrived mutation mixes versus healthy donor DNA. A single nucleotide variant mix containing 15 variants, an insertion-deletion mix containing four variants, and four separate serial dilutions of a FLT3-ITD/NPM1 mutant mix were analyzed, with predicted VAF ranging from 1.0x10-2 to 3.9x10-6. Data combined from four replicate libraries per mix generated 135,065-142,707x mean duplex consensus molecular depth (from the 1.5 μg DNA input libraries), with max depths 186,645-196,896x. All expected variants were detected in the mutation mixes and the observed VAF were significantly correlated with the predicted VAF (r2>0.99; Online Supplementary Figure S2). When the 21 spike-in mutation positions were assessed in the pure healthy donor DNA, a total of four mutant allele counts were detected out of a total duplex molecular depth of 2,993,429x at the 21 spike-in sites, for a combined mutation frequency of 1.3x10-6. The highest single background VAF at a spike-in site in the pure healthy donor DNA was 1.3x10-5.

Table 1. Patient clinical characteristics. Covariate

Patient cohort

Patients, N

Randomized arm, N (%) DA DA+GO Age in years, median (range)

Detection of residual variants in remission DS of remission samples utilizing the same 29-gene panel at a median error-corrected duplex molecular depth of 27,996x (range, 11,958x-35,131x) identified 82 diagnostic variants remaining in remission, with a median VAF of 0.059% (range, 0.005-41.8) (Online Supplementary Table S3). Variants were detected in 18 genes, with DNMT3A being the most frequently mutated, followed by NPM1 and FLT3 (Online Supplementary Figure S3). Forty-three patients (69%) had at least one diagnostic variant detectable in remission, with a median of two residual variants per positive patient (range, 1-5). Residual diagnostic variants in remission had a median 2.60 (range, 0.06-3.96) log10 reduction in VAF. Not surprisingly, mutations in DNMT3A and TET2, genes commonly associated with clonal hematopoiesis, showed the least change in VAF between diagnosis and remission: median 1.23 (range, 0.06-3.31) and median 1.32 (range 1.23-2.29) log10 reduction, respectively. Mutations in FLT3 showed the greatest change in VAF, median 3.12 (range, 1.5-3.8) log10 reduction.

32 (52) 30 (48) 48 (18-60)

Sex, N (%) Female Male

28 (45) 34 (55)

Performance status, N (%) 0-1 2-3

58 (84) 11 (16

Cytogenetic risk, N (%) Favorable Intermediate Adverse Missing

13 (23) 30 (54) 13 (23) 6

WBC x103/uL, median (range)

18.0 (0.2-214)

Platelets x103/uL, median (range)

48.5 (10-449)

Hemoglobin g/dL, median (range)

9.4 (3.5-13.6)

Race, N (%) Asian Black Native American/Alaskan Pacific Islander White Unknown

1 (2) 4 (6) 1 (2) 0 54 (87) 2 (3)

Specimen for sequencing, N (%) Bone marrow Peripheral blood

56 (90) 6 (10)

WBC: white blood cell count; DA: daunorubicin and cytarabine; GO: gemtuzumab ozogamicin.

Measurable residual disease as defined by flow cytometry MFC analysis of BM collected at the time of remission using a three-tube, ten-color assay identified MRD in ten (16%) patients (Figure 2) and the median MRD level was 0.25% (range, 0.002-6.2%) (Online Supplementary Table S4). Patients who were MFC MRD-positive had increased rates of relapse (50% vs. 30% at year 5; hazard ratio [HR] =2.4; 95% confidence interval [CI]: 0.9-6.7; P=0.087) and decreased rates of RFS (40% vs. 61% at year 5; HR=2.2; 95% CI: 0.9-5.4; P=0.095) and OS (40% vs. 68% at year 5;

Haematologica | 109 February 2024

403

ARTICLE - Duplex sequencing for measurable residual disease in AML

L.W. Dillon et al.

Figure 1. Clinical outcomes of S0106 acute myeloid leukemia patients analyzed for measurable residual disease. Rates of (A) non-relapse related mortality (NRM), (B) relapse, (C) relapse-free survival (RFS), and (D) overall survival (OS) are shown for the 62-patient cohort from the S0106 clinical trial analyzed for measurable residual disease (MRD) by duplex sequencing and multiparametric flow cytometry. CR: complete remission; No.: number.

Figure 2. Mutational spectrum, measurable residual disease status, and clinical outcomes of patients in complete remission. The heatmap displays variants detected at diagnosis and the presence (divided into variant allele fraction [VAF] ≥ or <5%) or absence at the time of complete remission (CR) by duplex sequencing (DS), DS measurable residual disease (MRD) status, multiparametric flow cytometry (MFC) MRD status, and clinical outcome at 5 years (relapse, no relapse, or non-relapse mortality [NRM]). The presence of a mutation within a gene is denoted in the heatmap, with the color corresponding to the highest VAF within each gene per patient. Variants identified in remission that were not identified at diagnosis are also marked (*). pos: positive. Haematologica | 109 February 2024

404

ARTICLE - Duplex sequencing for measurable residual disease in AML

HR=2.5; 95% CI: 1.0-6.3; P=0.059) compared to patients that were MFC MRD-negative (Figure 3A; Table 2). While not statistically significant in this subset of S0106 patients,

L.W. Dillon et al.

these results, including the magnitude of the HR, are in line with the significant findings previously published for the larger cohort of S0106 patients analyzed by MFC.20

Figure 3. Impact of measurable residual disease status on clinical outcomes. Rates of non-relapse mortality (NRM), relapse, relapse-free survival, and overall survival are shown based on measurable residual disease (MRD) status as determine by (A) multiparametric flow cytometry (MFC) and (B) duplex sequencing (DS). pos: positive; neg: negative; No.: number; CR: complete remission. Haematologica | 109 February 2024

405

L.W. Dillon et al.

ARTICLE - Duplex sequencing for measurable residual disease in AML

Measurable residual disease as defined by detection of residual diagnostic variants by duplex sequencing We defined DS test positivity utilizing criteria previously demonstrated to be prognostic for AML MRD by NGS,7,11 which included non-DTA (DNMT3A, TET2, ASXL1) time-of-diagnosis mutations with a VAF ≥0.1% and/or an FLT3-ITD/ NPM1 VAF ≥0.01% (Figure 2). Using this definition, 22 patients (35%) were DS MRD-positive. Compared to MFC, DS MRD provided a superior prediction of clinical outcomes, such that patients who were DS MRD-positive had significantly increased rates of relapse (68% vs. 13% at year 5; HR=8.8; 95% CI: 3.2-24.5; P<0.001) and decreased rates of RFS (23% vs. 77% at year 5; HR=5.4; 95% CI: 2.4-12.3; P<0.001) and OS (32% vs. 82% at year 5; HR=5.6; 95% CI: 2.3-13.8; P<0.001) compared to patients that were DS MRD-negative (Figure 3B; Table 2). Additional criteria for defining MRD by DS were also explored, including investigating the presence of any residual diagnostic variant and filtering based on VAF, gene, and VAF log10 reduction relative to diagnosis (Online Supplementary Figure S4; Online Supplementary Table S5). While a naïve definition of AML MRD as the detection of any residual diagnostic variant in remission was not associated with statistically significant difference in rates of relapse or survival (Online Supplementary Figure S4; Online Supplementary Table S5), the addition of VAF cutoffs, removal of mutations in genes associated with clonal hematopoiesis (DTA), and limiting calls to variants with no more than a log10 reduction of 2 between diagnosis and remission all resulted in statistically significant increased rates of relapse and decreased OS and RFS compared to patients testing negative, but none outperformed the criteria previously established as prognostic. Measurable residual disease as defined by de novo detection of deleterious variants by duplex sequencing We also explored the value of detecting AML-associated variants in remission that were not detected at the time of diagnosis. Utilizing the same variant filtering as defined above but agnostic to variant status at diagnosis, we identified 12 additional variants across nine patients with a median VAF of 0.24% (range, 0.08-15.1) (Online Supplementary Table S3). This resulted in three additional patients being defined as DS MRD-positive, for a total of 25 (40%) patients (Figure 2). Use of this NGS MRD definition agnostic to di-

agnostic variants provided a similar prediction of clinical outcomes to that of the initial prognostic criteria, such that patients who were DS MRD-positive had significantly increased rates of relapse (64% vs. 11% at year 5; HR= 8.7; 95% CI: 2.9-26.1; P<0.001) and decreased rates of RFS (28% vs. 78% at year 5; HR=4.8; 95% CI: 2.1-11.1; P<0.001) and OS (36% vs. 83% at year 5; HR=5.4; 95% CI: 2.1-13.8; P<0.001) compared to patients that were DS MRD-negative (Online Supplementary Figure S4F; Online Supplementary Table S5). Comparison of measurable residual disease detection by multiparametric flow cytometry versus duplex sequencing Next, we examined the differences between MFC and DS MRD calls. Of the 62 patients analyzed, five (8%) were called positive and 35 (56%) were called negative for MRD by both MFC and DS (Figure 4A). Five of the ten (50%) patients called MRD-positive by MFC were called negative by DS and 17 of the 22 (77%) patients called MRD-positive by DS were called negative by MFC. Comparing clinical outcomes of the discordant cases revealed that 59% of patients called MFC MRD-negative/DS MRD-positive relapsed, while only 20% of patients called MFC MRD-positive/DS MRD-negative relapsed (Figure 4B). While patients defined as MRD-positive by both MFC and DS had the highest rate of relapse (80% at year 5), there was no significant difference in rates of relapse between DS MRD-positive/MFC-positive and DS MRD-positive/MFC-negative (80% vs. 65% at year 5; cause-specific P=0.59) or DS MRD-negative/MFC-positive and DS MRD-negative/MFCnegative (20% vs. 12% at year 5; cause-specific P=0.57), indicating DS MRD was the main driver of outcomes prediction (Figure 4C; Online Supplementary Figure S5A). Looking closer at the disease burden in the discordant cases that experienced relapse, we found that the median VAF of variants identified in the DS MRD-positive/MFC-positive patients was 25 times higher than those identified in the DS MRD-positive/MFC-negative patients (1% vs. 0.04%). Additionally, five of the ten (50%) DS MRD-positive/MFC-negative patients that experienced relapse had a residual variant in NPM1 detected, compared to none in the DS MRD-positive/MFC-positive patients. Furthermore, the addition of MFC to the DS MRD definition did not significantly improve outcome predictions. While patients who were MFC- and/or DS MRD-positive had sig-

Table 2. Univariate cox regression model for associations between measurable residual disease definitions and clinical outcomes. MRD definition MFC DS MFC+DS

Relapse

Relapse-free survival

Overall survival

HR (95% CI)

2.4 (0.9-6.7) 8.8 (3.2-24.5) 7.8 (2.6-23.5)

0.087 <0.001 <0.001

2.2 (0.9-5.4) 5.4 (2.4-12.3) 5.2 (2.2-12.5)

0.095 <0.001 <0.001

2.5 (1-6.3) 5.6 (2.3-13.8) 6.2 (2.3-16.9)

0.059 <0.001 <0.001

MFC: multiparametric flow cytometry; DS: duplex sequencing; HR: hazard ratio; CI: confidence interval

Haematologica | 109 February 2024

406

ARTICLE - Duplex sequencing for measurable residual disease in AML

L.W. Dillon et al. Figure 4. Analysis of discordant measurable residual disease results by duplex sequencing and flow cytometry. (A) Number and percentage of patients called measurable residual disease (MRD)-positive (pos) versus MRD-negative (neg) by duplex sequencing (DS) versus multiparametric flow cytometry (MFC). (B) Clinical outcomes (non-relapse mortality [NRM], relapse, or no relapse) of MFC MRD versus DS MRD discordant cases. (C) Rates of relapse for patients grouped by MRD status as defined by MFC MRD and DS MRD. CR: complete remission; No.: number.

nificantly increased rates of relapse (59% vs. 12% at year 5; HR=7.8; 95% CI: 2.6-23.5; P<0.001) and decreased rates of RFS (30% vs. 79% at year 5; HR=5.2; 95% CI: 2.2-12.5; P<0.001) and OS (37% vs. 85% at year 5; HR=6.2; 95% CI: 2.3-16.9; P<0.001) compared to patient who were MFC- and DS MRD-negative, this did not significantly differ from DS MRD alone (Table 2; Online Supplementary Figure S5B). We assessed the ability of covariates to predict relapse in individual patients. Baseline clinical characteristics (including age, performance status, and cytogenetics) yielded a C-statistic of 0.66. Inclusion of MFC MRD status did not improve the discrimination of the model with a C-statistic of 0.67, while inclusion of DS MRD status did improve the model discrimination yielding a C-statistic of 0.77. Impact of duplex sequencing measurable residual disease status and treatment regimen on clinical outcomes Finally, we examined the impact of DS MRD status and patient randomization to DA versus DA+GO on clinical outcomes. In concordance with results from the full S0106 cohort,19 the subset of 62 patients in this study showed no difference in rates of relapse between patients treated with DA versus DA+GO (35% vs. 31% at year 5; P=0.62) (Figure 5A). Adding information on DS MRD status showed that patients who were DS MRD-positive had significantly

higher rates of relapse compared to patients that were DS MRD-negative regardless of the treatment regimen (DA: 60% vs. 12% at year 5; P=0.017, DA+GO: 86% vs. 13% at year 5; P<0.001) (Figure 5B). No difference was seen in rates of relapse between patients treated with DA versus DA+GO based on DS MRD status (DS MRD-positive: 60% vs. 86% at year 5; P=0.2; DS MRD-negative, 12% vs. 13%; P=0.98).

Discussion MRD has been well established as a method for quantifying the antileukemic effect of interventional therapies, but implementation in the clinic has thus far been limited for AML. MFC has been widely used for AML MRD detection, but concerns exist over inter-laboratory variability which could hinder widespread applicability of this technique. NGS for AML MRD detection could be more amenable to decentralized clinical testing and has been shown to outperform decentralized flow cytometry in the context of FLT3-ITD- and NPM1-mutated AML.11 Utilizing a subset of 62 AML patients treated on the S0106 phase III randomized trial of DA versus DA+GO induction chemotherapy, we compared the performance of MRD detection by high quality, centralized MFC and ultra-sensitive DS across a broad 29 gene panel to predict clinical response in first

Haematologica | 109 February 2024

407

ARTICLE - Duplex sequencing for measurable residual disease in AML

L.W. Dillon et al.

Figure 5. Impact of treatment randomization and duplex sequencing measurable residual disease status on relapse. Rates of relapse for patients as defined by (A) treatment randomization to daunorubicin and cytarabine (DA) versus daunorubicin, cytarabine, and gemtuzumab ozogamicin (DA+GO) and (B) treatment randomization (DA or DA+GO) and duplex sequencing (DS) measurable residual disease (MRD) status. pos.: positive; neg.: negative; No.: number; CR: complete remission.

remission and found the latter to broadly have superior outcome-predicting performance. Application of NGS for AML MRD detection has varied across the literature, and questions remain regarding the impact of assay sensitivity, gene targets, variant status at diagnosis, and the applicability across patients with diverse baseline genetics.22,23 In the 67 patients screened in this study, we found that 62 (93%) had at least one variant present at diagnosis in the gene panel examined that could be tracked by NGS. The mutations identified spanned 23 genes, representing a broad set of AML MRD targets. The highly sensitive DS assay detected residual mutations at some level in most patients, rendering the naïve designation of MRD positivity clinically uninformative. However, application of previously established, clinically relevant variant filtering conditions, including VAF thresholds well above the assay limit of detection and removal of less informative genes (DNMT3A, TET2, ASXL1) associated with clonal hematopoiesis,24,25 was highly predictive of adverse clinical outcomes. These results highlight the importance of establishing informed guidelines for interpreting the presence of molecular MRD in the clinical setting. Additionally, we found that utilizing these filtering criteria remains highly predictive when agnostic to diagnostic variants. Therefore, the DS assay may have utility even when a diagnostic sample is not available. Future studies are needed to assess clinically relevant VAF thresholds at later treatment time points where residual disease may be present at a lower level. In comparison to MFC, DS was significantly better at stratifying patients at risk of adverse clinical outcomes. Additional prognostic value was not seen when combining MRD detection by DS and MFC. Of the DS MRD-positive patients that relapsed, the median VAF of patients with MRD also detected by MFC was 25-times higher than MFC-negative patients and all relapses occurred within the first year,

indicating that this subset of patients had a higher disease burden at the time of clinical remission. Of the DS MRD-positive patients that relapsed but were MFC-negative, 50% (n=5) had residual NPM1 mutations, in contrast to none in the double-positive group. NPM1-mutated AML characteristically has absent/low CD34 expression with heterogeneity seen in the observed leukemia-associated immunophenotypes,26,27 making it uniquely challenging to track by MFC. This combined with increased assay sensitivity could explain most of the discrepant results and improved prognostic power of DS. The S0106 phase III clinical trial found that randomization of AML patients to DA versus DA+GO induction chemotherapy provided no significant difference in clinical outcomes. One potential value of MRD testing is to provide a surrogate endpoint to predict long-term patient response, allowing for faster drug development/approval and to identify patients in need of additional therapy versus those who do not. Recent trials have found that addition of GO to standard induction chemotherapy leads to deeper molecular responses and lower relapse rates in patients with NPM1 mutated AML.28,29 In this cohort we found that DS MRD was able to predict clinical relapse, with no significant difference for the prognostic implications of MRD status seen in patients who received DA versus DA+GO. Follow-up studies are needed to confirm the applicability of this technology as a definitive surrogate biomarker in clinical trial settings. Limitations of this study include (i) the small sample size, (ii) the retrospective nature of the DS MRD analysis, (iii) the comparison to an early generation MFC assay, and (iv) the age of the S0106 study potentially limiting comparability to contemporary AML standard of care. The findings of this study need to be confirmed in a larger cohort using prospective analysis by both DS and a more modern MFC MRD assay. While both flow cytometry and

Haematologica | 109 February 2024

408

ARTICLE - Duplex sequencing for measurable residual disease in AML

molecular methods such as NGS can be used for AML MRD detection in both centralized and local settings, the level of expertise required for interpretation and the test performance characteristics differ.30 Evidence-based recommendations support the use of molecular testing for a stable AML MRD target, in preference to flow cytometry, in situations where a validated test is available.12,31 There are now multiple ongoing efforts to define appropriate targets, test requirements, interpretation, and clinical implications of AML MRD molecular testing.31,32 Methods to suppress the false-positive error rates, such as DS, enable low-level variant discovery and could potentially expand the range of suitable targets for NGS-based AML MRD. In this study only five of the 67 (7%) patients screened were excluded due to lack of a mutation detected at diagnosis available for tracking by DS in the panel used. However, these patients did have cytogenetic abnormalities present. Whole exome or genome sequencing at diagnosis could inform individualized MRD panels, targeting a combination of recurrently mutated genes, novel variants, and structural alterations. Future work needs to be done exploring the use of patient personalized MRD targets to expand applicability to all patients. In conclusion, we provide evidence that in a group of genetically diverse de novo adult AML patients randomized to DA versus DA+GO induction chemotherapy that ultra-sensitive detection of residual variants by DNA sequencing in the BM or PB at the time of first CR can outperform centralized, high-quality MFC in identifying patients at high risk of adverse clinical outcomes and predicting patient clinical response to treatment. Disclosures The National Heart, Lung, and Blood Institute receives research funding for the laboratory of CSH from the Foundation of the NIH AML MRD Biomarkers Consortium. MO consults for Merck and Biosight and serves on the data safety monitoring committee for Celgene, Glycomimetic, and Grifols. JH, THS, CCV, ES and JJS are employees and stockholders of TwinStrand Biosciences. BLW consults for

L.W. Dillon et al.

Amgen and Kite Pharma. HPE discloses grants and other support from AbbVie, Agios Pharmaceuticals, ALX Oncology, Amgen, Daiichi Sankyo, FORMA Therapeutics, Forty Seven, Gilead Sciences, GlycoMimetics, ImmunoGen, Jazz Pharmaceuticals, MacroGenics, Novartis and PTC Therapeutics; has received research funding from AbbVie, Agios Pharmaceuticals, Bristol Myers Squibb, Celgene, Incyte Corporation, Jazz Pharmaceuticals and Novartis; is part of the speakers bureau of AbbVie; is on the independent review committee of AbbVie, Agios Pharmaceuticals, Astellas, Bristol Myers Squibb, Celgene, Daiichi Sankyo, Genentech, GlycoMimetics, Incyte Corporation, Jazz Pharmaceuticals and Kura Oncology. All other authors have no conflicts of interest to disclose. Contributions JPR conceived and designed the study. JH and ES performed and analyzed laboratory experiments. LB managed clinical samples. THS and CCV performed bioinformatic analysis. HN and MO performed statistical analysis. BLW performed flow cytometry analysis. HPE chaired the clinical trial. LWD performed genetic variant interpretation. LWD and CSH directed integrative analysis and wrote the original version of the manuscript. All authors contributed to reviewing and editing the manuscript and are accountable for the final version. Funding This work was supported in part by the Intramural Research Program of the National Heart, Lung, and Blood Institute; National Cancer Institute of the National Institutes of Health under award no. R44CA233381 (to JS); and National Cancer Institute CA175008, 180888, 180819, 233338, and 233381 (to JPR). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Data-sharing statement Raw FASTQ files are available in the NCBI Small Reads Archive (accession: PRJNA945188).

References 1. Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin. 2022;72(1):7-33. 2. Dohner H, Wei AH, Appelbaum FR, et al. Diagnosis and management of AML in adults: 2022 recommendations from an international expert panel on behalf of the ELN. Blood. 2022;140(12):1345-1377. 3. Hourigan CS, Karp JE. Minimal residual disease in acute myeloid leukaemia. Nat Rev Clin Oncol. 2013;10(8):460-471. 4. Short NJ, Fu C, Berry DA, et al. Association of hematologic response and assay sensitivity on the prognostic impact of measurable residual disease in acute myeloid leukemia: a systematic review and meta-analysis. Leukemia.

2022;36(12):2817-2826. 5. Short NJ, Zhou S, Fu C, et al. Association of measurable residual disease with survival outcomes in patients with acute myeloid leukemia: a systematic review and meta-analysis. JAMA Oncol. 2020;6(12):1890-1899. 6. Paras G, Morsink LM, Othus M, et al. Conditioning intensity and peritransplant flow cytometric MRD dynamics in adult AML. Blood. 2022;139(11):1694-1706. 7. Hourigan CS, Dillon LW, Gui G, et al. Impact of conditioning intensity of allogeneic transplantation for acute myeloid leukemia with genomic evidence of residual disease. J Clin Oncol. 2020;38(12):1273-1283.

Haematologica | 109 February 2024

409

ARTICLE - Duplex sequencing for measurable residual disease in AML 8. Dillon R, Hills R, Freeman S, et al. Molecular MRD status and outcome after transplantation in NPM1-mutated AML. Blood. 2020;135(9):680-688. 9. Thol F, Gabdoulline R, Liebich A, et al. Measurable residual disease monitoring by NGS before allogeneic hematopoietic cell transplantation in AML. Blood. 2018;132(16):1703-1713. 10. Loo S, Dillon R, Ivey A, et al. Pretransplant FLT3-ITD MRD assessed by high-sensitivity PCR-NGS determines posttransplant clinical outcome. Blood. 2022;140(22):2407-2411. 11. Dillon LW, Gui G, Page KM, et al. DNA sequencing to detect residual disease in adults with acute myeloid leukemia prior to hematopoietic cell transplant. JAMA. 2023;329(9):745-755. 12. Heuser M, Freeman SD, Ossenkoppele GJ, et al. 2021 Update on MRD in acute myeloid leukemia: a consensus document from the European LeukemiaNet MRD Working Party. Blood. 2021;138(26):2753-2767. 13. Blachly JS, Walter RB, Hourigan CS. The present and future of measurable residual disease testing in acute myeloid leukemia. Haematologica. 2022;107(12):2810-2822. 14. Paiva B, Vidriales MB, Sempere A, et al. Impact of measurable residual disease by decentralized flow cytometry: a PETHEMA real-world study in 1076 patients with acute myeloid leukemia. Leukemia. 2021;35(8):2358-2370. 15. Tettero JM, Freeman S, Buecklein V, et al. Technical aspects of flow cytometry-based measurable residual disease quantification in acute myeloid leukemia: experience of the European LeukemiaNet MRD working party. Hemasphere. 2022;6(1):e676. 16. Jongen-Lavrencic M, Grob T, Hanekamp D, et al. Molecular minimal residual disease in acute myeloid leukemia. N Engl J Med. 2018;378(13):1189-1199. 17. Patkar N, Kakirde C, Shaikh AF, et al. Clinical impact of panelbased error-corrected next generation sequencing versus flow cytometry to detect measurable residual disease (MRD) in acute myeloid leukemia (AML). Leukemia. 2021;35(5):1392-1404. 18. Grob T, Sanders MA, Vonk CM, et al. Prognostic value of FLT3internal tandem duplication residual disease in acute myeloid leukemia. J Clin Oncol. 2023;41(4):756-765. 19. Petersdorf SH, Kopecky KJ, Slovak M, et al. A phase 3 study of gemtuzumab ozogamicin during induction and postconsolidation therapy in younger patients with acute myeloid leukemia. Blood. 2013;121(24):4854-4860. 20. Othus M, Wood BL, Stirewalt DL, et al. Effect of measurable (‘minimal’) residual disease (MRD) information on prediction of relapse and survival in adult acute myeloid leukemia. Leukemia.

2016;30(10):2080-2083. 21. Valentine CC, 3rd, Young RR, Fielden MR, et al. Direct quantification of in vivo mutagenesis and carcinogenesis using duplex sequencing. Proc Natl Acad Sci U S A. 2020;117(52):33414-33425. 22. Ghannam J, Dillon LW, Hourigan CS. Next-generation sequencing for measurable residual disease detection in acute myeloid leukaemia. Br J Haematol. 2020;188(1):77-85. 23. Walter RB, Ofran Y, Wierzbowska A, et al. Measurable residual disease as a biomarker in acute myeloid leukemia: theoretical and practical considerations. Leukemia. 2021;35(6):1529-1538. 24. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488-2498. 25. Genovese G, Kahler AK, Handsaker RE, et al. Clonal hematopoiesis and blood-cancer risk inferred from blood DNA sequence. N Engl J Med. 2014;371(26):2477-2487. 26. Alcalay M, Tiacci E, Bergomas R, et al. Acute myeloid leukemia bearing cytoplasmic nucleophosmin (NPMc+ AML) shows a distinct gene expression profile characterized by up-regulation of genes involved in stem-cell maintenance. Blood. 2005;106(3):899-902. 27. Verhaak RG, Goudswaard CS, van Putten W, et al. Mutations in nucleophosmin (NPM1) in acute myeloid leukemia (AML): association with other gene abnormalities and previously established gene expression signatures and their favorable prognostic significance. Blood. 2005;106(12):3747-3754. 28. Kapp-Schwoerer S, Weber D, Corbacioglu A, et al. Impact of gemtuzumab ozogamicin on MRD and relapse risk in patients with NPM1-mutated AML: results from the AMLSG 09-09 trial. Blood. 2020;136(26):3041-3050. 29. Lambert J, Lambert J, Nibourel O, et al. MRD assessed by WT1 and NPM1 transcript levels identifies distinct outcomes in AML patients and is influenced by gemtuzumab ozogamicin. Oncotarget. 2014;5(15):6280-6288. 30. Freeman SD, Hourigan CS. MRD evaluation of AML in clinical practice: are we there yet? Hematology Am Soc Hematol Educ Program. 2019;2019(1):557-569. 31. Hourigan CS. Achieving MRD negativity in AML: how important is this and how do we get there? Hematology Am Soc Hematol Educ Program. 2022;2022(1):9-14. 32. Wong ZC, Dillon LW, Hourigan CS. Measurable residual disease in patients undergoing allogeneic transplant for acute myeloid leukemia. Best Pract Res Clin Haematol. 2023;36(2):101468.

Haematologica | 109 February 2024

410

L.W. Dillon et al.

ARTICLE - Acute Myeloid Leukemia

miR-30e-5p regulates leukemia stem cell self-renewal through the Cyb561/ROS signaling pathway Yanwen Ge,1 Mei Hong,1 Yu Zhang,1 Jiachen Wang,1 Lei Li,2 Hongkai Zhu,3 Yue Sheng,3 Wen-Shu Wu4 and Zhonghui Zhang1,5

Correspondence: Z. Zhang zhonghui@shu.edu.cn

School of Life Sciences, Shanghai University, Shanghai, China; 2Department of Pediatrics,

W-S. Wu wuwenshu@uic.edu

Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China; 3Department of Hematology, The Second Xiangya Hospital, Central South University, Changsha, China; 4Division of Hematology/Oncology, Department of Medicine and University of Illinois Cancer Center, the University of Illinois at Chicago, Chicago, IL, USA and Shaoxing Institute of Technology, Shanghai University, Shaoxing, China

Received: Accepted: Early view:

January 27, 2023. August 8, 2023. August 17, 2023.

Abstract Leukemia stem cells (LSC) represent a crucial and rare subset of cells present in acute myeloid leukemia (AML); they play a pivotal role in the initiation, maintenance, and relapse of this disease. Targeting LSC holds great promise for preventing AML relapse and improving long-term outcomes. However the precise molecular mechanisms governing LSC self-renewal are still poorly understood. Here, we present compelling evidence that the expression of miR-30e-5p, a potential tumor-suppressive microRNA, is significantly lower in AML samples than in healthy bone marrow samples. Forced expression of miR30e effectively inhibits leukemogenesis, impairs LSC self-renewal, and delays leukemia progression. Mechanistically, Cyb561 acts as a direct target of miR-30e-5p in LSC, and its deficiency restricts the self-renewal of LSC by activating reactive oxygen series signaling and markedly prolongs recipients’ survival. Moreover, genetic or pharmacological overexpression of miR-30e-5p or knockdown of Cyb561 suppresses the growth of human AML cells. In conclusion, our findings establish the crucial role of the miR-30e-5p/Cyb561/ROS axis in finely regulating LSC self-renewal, highlighting Cyb561 as a potential therapeutic target for LSC-directed therapies.

Introduction Acute myeloid leukemia (AML) is a highly aggressive and often fatal blood malignancy primarily caused by genetic mutations in hematopoietic stem cells (HSC) or committed progenitor compartments.1,2 Despite ongoing efforts, the standard therapy for most types of AML has not shown significant improvements, with a 5-year overall survival rate of approximately 24% in the USA.3 Leukemia stem cells (LSC), an indispensable subset of AML cells with a limitless capacity of self-renewal and differentiation block, play a key role in the initiation, maintenance, and propagation of AML.4 Targeting LSC has emerged as a promising strategy to address AML relapse and enhance long-term treatment outcomes. MicroRNA (miRNA, miR), typically consisting of 18-24 nucleotides, which function by binding to and cleaving mRNA or inhibiting their translation. Accumulating evidence supports the critical role of miRNA in regulating LSC self-renewal and

differentiation.5-7 Among them, miR-30e-5p is implicated in many important biological regulation processes, such as cancer development and metastasis, epithelial-mesenchymal transition, fatty acid metabolism, and osteogenesis.8-12 Previous studies have documented lower levels of expression of miR-30e-5p in patients with chronic myeloid leukemia. Additionally, enforced expression of miR-30e-5p has demonstrated inhibitory effects on proliferation and induction of apoptosis in K562 cells.13 Another study highlighted the involvement of circPVT1, which regulates the miR-30e/DLL4 pathway, in suppressing the proliferation of T-cell acute lymphoblastic leukemia cells.14 However, the precise roles of miR-30e-5p in LSC initiation and maintenance and its therapeutic potential in AML remain largely unexplored. In this study, we aimed to investigate the roles of miR-30e5p in LSC initiation and maintenance in KMT2A::MLLT3-induced leukemia, as well as explore the potential of this microRNA as a therapeutic target in human AML cells. We

Haematologica | 109 February 2024

411

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

demonstrated that miR-30e-5p is expressed at lower levels in human AML bone marrow (BM) samples compared to healthy samples. Forced expression of miR-30e delays the onset of KMT2A::MLLT3-driven leukemia by reducing cycling LSC and promoting LSC apoptosis. Using RNA-sequencing analysis, we found that endogenous Cyb561 is significantly suppressed in the LSC overexpressing miR-30e compared with LSC expressing empty retroviral miRVector. By quantitative PCR (qPCR), reporter assays, and functional assays, we identified Cyb561 as a direct downstream target of miR-30e-5p in leukemogenesis. Consistently, knockdown of Cyb561 resulted in reduced colony formation capacity in vitro, decreased LSC frequency, and delayed progression of mouse AML in vivo. Moreover, the miR-30e-5p/Cyb561 signaling pathway was found to enhance intracellular reactive oxygen species (ROS) levels, while concurrently decreasing glutathione and ascorbate levels. The ROS scavenger N-acetyl-L-cysteine (NAC) impeded the roles of miR-30e-5p and Cyb561 in leukemia progression by regulation of cell cycle and apoptosis of LSC. In addition, both genetic and pharmacological upregulation of miR-30e-5p hindered the growth of human AML cells. Consistent with these findings, we observed a significant upregulation of CYB561 expression in AML patients, and high-level expression of CYB561 correlated with shorter overall survival in AML patients. Inhibition of CYB561 had suppressive effects on human AML cell growth in vitro. Together, our study provides compelling evidence supporting the essential role of the miR-30e-5p/Cyb561/ ROS signaling pathway in the initiation and maintenance of LSC in KMT2A::MLLT3-induced leukemia. Furthermore, we highlight the potential of Cyb561 as a therapeutic target for LSC in AML.

murine fibroblast growth factor-1, and 100 ng/mL human angiopoietin-like protein 3. All recombinant proteins were purchased from PeproTech (Rock Hill, NJ, USA) or Genscript (Nanjing, China). All cell culture products were obtained from Jet Biofil (Guangzhou, China) unless otherwise specified in the text and figure legends. Plasmids KMT2A::MLLT3 (previously called MLL-AF9) was cloned into the retroviral vector pMIGR1 containing green fluorescent protein (GFP). The entire loci of mouse and human miR30e, containing 100-bp upstream and downstream native flank sequences, were amplified by PCR from genomic DNA; then cloned into pMXs-miR-Puro (miRVector) retroviral vector. To generate short hairpin (sh)RNA-expressing plasmids targeting mouse and human CYB561, we cloned shRNA control and CYB561 shRNA into Age I and EcoR I sites of Tet-pLKO-puro vector (Addgene, USA). To construct the luciferase reporters for identifying true targets of miR-30e-5p, the sequences of target 3’ untranslated regions (3’UTR) were amplified from genomic DNA of C57/ B6J BM cells by PCR using specific primers and cloned into the pGL3-control vector (Promega). To generate the mutant of Cyb561 3’UTR, point mutations in the miR-30e5p binding site were introduced by PCR and cloned into the pGL3-control vector. The reporter plasmids were validated by DNA sequencing. For the ectopic expression of Cyb561, the coding sequence of Cyb561 was amplified from mouse BM cDNA and then cloned into a retroviral vector pMIBSD containing the selective antibiotic blasticidin. All the primers for plasmid cloning are listed in Online Supplementary Table S1.

Results

Methods Mice C57BL/6J mice were purchased from VITALSTAR (Beijing, China). All the animal studies conducted in this research were approved by the Animal Care and Use Committee at Shanghai University. Cell cultures 293T cells were cultured in DMEM (high glucose) containing 10% fetal bovine serum. THP-1, MONOMAC-6, NOMO-1, MV4-11, and NB-4 cells were cultured in RPMI1640 plus 10% fetal bovine serum. Human MA9.3 leukemic cells were cultured in IMDM plus 20% fetal bovine serum with defined cytokines.15 Enriched fresh hematopoietic stem and progenitor cells (HSPC) were cultured in vitro in StemSpan (STEMCELL Technologies, Vancouver, Canada) supplemented with 10 μg/mL heparin (Sigma, St Louis, MO, USA), 10 ng/mL murine stem cell factor, 20 ng/mL murine thrombopoietin, 20 ng/mL human insulin-like growth factor-II, 10 ng/mL

miR-30e-5p expression is downregulated in patients with acute myeloid leukemia Previous clinical reports showed that miR-30e-5p is upregulated in patients with B-cell acute lymphoblastic leukemia and childhood acute lymphoblastic leukemia but downregulated in patients with chronic myeloid leukemia.13,16,17 To assess the potential clinical relevance of miR-30e-5p expression levels in AML samples and healthy samples, we carried out qPCR analysis of the expression of mature miR-30e-5p and miR-30e-3p. Our findings revealed a significant downregulation of miR-30e-5p in AML BM samples compared to healthy BM samples (Figure 1A, Online Supplementary Table S2). No significant difference was observed in the expression of miR-30e-3p between AML samples and healthy samples (Online Supplementary Figure S1A). To assess the expression levels of miR-30e-5p in different subtypes of AML, defined according to the French-American-British (FAB) classification, and determine its prognostic value for overall survival, we searched clinical databases

Haematologica | 109 February 2024

412

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

online (http://ualcan.path.uab.edu/index.html) and found that the expression of miR-30e-5p was lower in human AML-M3, -M4, and -M5 subtypes than in AML-M2 (Online Supplementary Figure S1B). The overall survival of AML patients was not significantly related to the expression level of miR-30e-5p (Online Supplementary Figure S1C). To provide more conclusive evidence for the decreased expression of miR-30e-5p in LSC, we examined a mouse AML model driven by the KMT2A::MLLT3 fusion oncogene. Our results demonstrated a reduction in miR-30e-5p expression in LSC-enriched granulocyte-monocyte progenitors (referred to as L-GMP, Lin–GFP+c-Kit+CD34+CD16/32+) compared to HSPC and granulocyte-monocyte progenitors (GMP) (Figure 1B). Furthermore, we observed the downregulation of endogenous miR-30e-5p in HSPC expressing various oncogenes (e.g., HOXA9, MESI1, and NUP98::HOXA9) (Figure 1C). These findings in the AML mouse model are consistent with the results obtained from human AML samples, suggesting a potentially significant role for miR-30e-5p in AML. Overexpression of miR-30e delays the development of KMT2A::MLLT3-driven leukemia To determine the role of miR-30e in AML cells, we enriched Lin–Sca-1+ HSPC by a single intraperitoneal dose of 5-fluorouracil. These cells were subsequently co-transduced with retroviral particles containing KMT2A::MLLT3/miRVector and KMT2A::MLLT3/miR-30e. Overexpression of miR-30e in HSPC resulted in a 5.36-fold increase in miR-30e-5p expression compared to the level in miRVector-HSPC (Online Supplementary Figure S2A). Colony-forming/replating assays showed that forced expression of miR-30e suppressed KMT2A::MLLT3-induced immortalization of mouse HSPC (Figure 2A, Online Supplementary Figure S2B). To explore the roles of miR-30e in AML in vivo, equal numbers of KMT2A::MLLT3/miRVector- and KMT2A::MLLT3/miR-30e-transduced HSPC were transplanted into irradiated recipients. Both sets of HSPC developed AML in recipient mice with full penetrance; however, the onset was significantly delayed in miR-30e-overexpressing AML compared to miRVector-AML (median survival, 78 days vs. 52 days, respectively; P<0.01) (Figure 2B, C). The expression level of miR-30e-5p was elevated 3.6-fold in miR-30e-overexpressing AML cells (Online Supplementary Figure S2C). Consistent with these findings, other parameters of AML severity, such as peripheral blood white cell count and spleen weight, were also reduced (Online Supplementary Figure S2D-F). In line with the primary BM transplantation results, the secondary recipients of miRVector developed AML significantly faster than the miR-30e-overexpressing secondary recipients (median survival, 41 days vs. 53 days, respectively; P<0.01) (Online Supplementary Figure S2G). The frequency of LSC is thought to be associated with patients’ prognosis as well as leukemia progression in murine models. To further investigate the impact of miR-30e on LSC, we analyzed L-GMP (i.e., a LSC-enriched population)

frequency in the secondary recipients overexpressing the miRVector or miR-30e. The results showed that forced expression of miR-30e resulted in a lower LSC frequency (Figure 2D, Online Supplementary Figure S3), indicating a potential role for miR-30e in modulating LSC self-renewal. Additionally, cell cycle analysis demonstrated a decreased proportion of miR-30e-overexpressing LSC in the S/G2/M phases, accompanied by a concomitant increase in the G1 phase, compared to miRVector LSC (Figure 2E, Online Supplementary Figure S3). Furthermore, miR-30e-overexpressing LSC also showed heightened levels of apoptosis compared to miRVector-overexpressing LSC (Figure 2F). To directly evaluate the effect of miR-30e overexpression on the frequency of LSC, we conducted a limiting dilution

Figure 1. miR-30e-5p is repressed in both human and mouse acute myeloid leukemia cells. (A) Quantitative polymerase chain reaction (qPCR) analysis of the expression of miR-30e-5p between bone marrow cells of healthy controls and acute myeloid leukemia (AML) patients. Results are normalized to U6 expression and expressed relative to miR-30e-5p expression in the healthy group (healthy patients, N=6; AML patients, N=29). (B) qPCR analysis of the expression of endogenous miR-30e-5p in mouse HSC, LSK, GMP, and L-GMP. The indicated cells were sorted by flow cytometry. Results are normalized to U6 expression and expressed relative to miR-30e-5p expression in the L-GMP group (N=3). (C) qPCR analysis of the expression of endogenous miR-30e-5p in mouse hematopoietic stem and progenitor cells transduced with the retroviral vector pMIGR1 only or retrovirus expressing KMT2A::MLLT3, HOXA9, MESI1, NUP98::HOXA9. Results are normalized to U6 expression and expressed relative to miR-30e-5p expression in the retroviral empty vector group (N=3). All data are represented as mean ± standard deviation. Two-tailed Student t tests were used to assess statistical significance (*P<0.05; **P<0.01). BM: bone marrow; HSC: hematopoietic stem cells (Lin–Sca-1+c-Kit+CD150+CD48–); LSK cells (Lin–Sca-1+c-Kit+); GMP: granulocyte-monocyte progenitors (Lin–c-Kit+CD34+CD16/32+); L-GMP: leukemia stem cell-enriched granulocyte-monocyte progenitors (GFP+Linlow c-Kit+CD34+CD16/32+) (N=3).

Haematologica | 109 February 2024

413

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

F F

Figure 2. Forced expression of miR-30e impairs leukemia stem cell self-renewal and delays the onset of KMT2A::MLLT3 leukemia. (A) Colony-forming assay of miRVector- or miR-30e-overexpressing acute myeloid leukemia (AML) cells (N=3). (B) Percentage green fluorescent protein positive cells in the peripheral blood at week 5 after primary bone marrow transplant with 5x104 AML cells (N=6). (C) Survival analysis of primary recipient mice. The post-transplant median survival was 52 versus 78 days for primary recipients of miRVector- or miR-30e-overexpressing AML cells, respectively (P<0.01, Mantel-Cox test, N=7). (D) Frequency of leukemia stem cell-enriched granulocyte-monocyte progenitors (L-GMP) in the bone marrow (BM) and spleen from primary recipients injected with 5x104 miRVector- or miR-30e-overexpressing AML cells (N=5). (E) Cell cycle phase distribution of L-GMP cells in BM from primary recipients of 5x104 with miRVector- or miR-30e-overexpressing AML cells at week 5 after transplantation (N=4). (F) Percentage of apoptotic L-GMP cells in the BM from primary recipients (N=5). Data are representative of two or three independent experiments. Excluding survival analysis, all data are represented as mean ± standard deviation. Two-tailed Student t tests were used to assess statistical significance (N.S.: not significant; *P<0.05; **P<0.01). miRVector: control; OE-miR-30e: cells with miR-30e overexpression; GFP: green fluorescent protein; PBMC: peripheral blood mononuclear cells; SP: spleen.

assay. As expected, the estimated LSC frequency in the miR-30e-overexpressed group was 17-fold lower than that in the miRVector-group (1/529.9 vs. 1/31.1) (Online Supplementary Figure 2H, Online Supplememtary Table S3). Considering that the survival disparity between miRVector- and miR-30e-overexpressing AML recipients might be partly attributed to differences in homing efficiency, we determined the homing capability of secondary transplanted miRVector- and miR-30e-overexpressing AML cells. The results showed that homing efficiency was similar for both miRVector- and miR-30e-overexpressing AML cells at 16 h after BM transplantation (Online Supplementary Figure S2I). Taken together, these data demonstrated that miR30e impaired cell cycle and enhanced apoptosis in LSC, thereby delaying AML progression in vivo. Since the miR-30e locus contains two mature miRNA (i.e., miR-30e-5p and miR-30e-3p), we next determined which mature miRNA affects AML colony formation. To this end, we transfected corresponding mature miRNA mimics into KMT2A::MLLT3 AML cells. The results revealed that overexpression of miR-30e-5p significantly suppressed AML colony formation. In contrast, forced expression of miR30e-3p did not have a noticeable effect on colony formation (Online Supplementary Figure S2J). These findings suggest

that miR-30e-5p, rather than miR-30e-3p, plays a pivotal regulatory role in KMT2A::MLLT3-driven leukemogenesis. Cyb561 is a direct target of miR-30e-5p in KMT2A::MLLT3-driven leukemia stem cells To gain insights into the underlying mechanisms by which miR-30e regulates the self-renewal of LSC, we sorted miRVector- and miR-30e-overexpressing LSC from two groups of recipients with leukemia driven by KMT2A:: MLLT3. We assessed the gene expression profiles of these cells using RNA-sequencing analysis (Online Supplementary Figure S4A). We found that 689 genes were upregulated, and 531 genes were downregulated by >2.0-fold (P<0.01) in miR-30e-overexpressing LSC compared with LSC expressing miRVector (Online Supplementary Table S4). Through gene set enrichment analysis (GSEA), we found that the downregulated genes in miR-30e-overexpressing LSC were associated with electron transport chain, oxidative phosphorylation, and glutathione metabolism (Figure 3A, Online Supplementary Figure S4B, Online Supplementary Table S5). Further analysis of Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways revealed that these differentially expressed genes could be classified into five functional groups, with more than 150 genes involved in

Haematologica | 109 February 2024

414

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

signal transduction, including the MAPK pathway (Online Supplementary Figure S5, Online Supplementary Table S6). Considering that miRNA typically inhibit the translation of their mRNA by binding with the 3’UTR of their direct targets, we focused on the downregulated genes. Using the miRbase online database (www.mirbase.org) and qPCR, we validated seven genes as the most likely direct targets of miR-30e-5p, namely Dock7, Six1, Hoxa11, Cyb561, Six4, Prickle1, and Ikzf2 (Figure 3B, Online Supplementary Table

S7). Among these seven miR-30e-5p-regulated genes, some (Ikzf2, Six1, and Hoxa11) have been shown to play important roles in LSC and AML.18-20 To confirm the true target of miR-30e-5p in LSC, we cloned the DNA fragment covering these genes’ 3’UTR into pGL3-control luciferase vector. Subsequently, we performed a luciferase reporter assay to validate this bioinformatic prediction. Our data revealed that miR-30e-5p inhibited the luciferase activity of the wild-type Cyb561 3’UTR construct by 45%, whereas

Figure 3. RNA-sequencing analysis identifies potential targets of miR-30e-5p in L-GMP cells from KMT2A::MLLT3-driven leukemia. (A) Enrichment plot of downregulated gene sets in miR-30e-overexpressing leukemia stem cells (LSC), as determined by gene set enrichment analysis. RNA-sequencing data for 27,359 transcripts were used for the analysis. (B) Quantitative polymerase chain reaction analysis of candidate target genes in LSC-enriched granulocyte-monocyte progenitors (L-GMP) sorted from recipients transplanted with miRVector- or miR-30e-overexpressing acute myeloid leukemia cells. Results are normalized to Hprt expression and expressed relative to the expression of target genes in miRVector L-GMP (N=3). (C) Luciferase reporter assay to identify the true target of miR-30e-5p (N=3). (D) Cyb561 3’UTR luciferase reporter assay to identify the binding site of miR-30e-5p (N=3). (E) Overexpression of Cyb561 impaired the suppressive function of miR-30e on LSC (N=3). (F) Forced expression of Cyb561 impeded the prolonged survival of miR-30e on KMT2A::MLLT3-driven leukemia (N=7). Data are representative of two or three independent experiments. All data are represented as mean ± standard deviation. Two-tailed Student t tests were used to assess statistical significance (N.S.: not significant; *P<0.05; **P<0.01). miRVector: control; OE-miR-30e: cells with miR-30e overexpression; NOM: nominal; WT: wild-type; mut: mutated. Haematologica | 109 February 2024

415

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

the mutation in the binding site fully restored luciferase activity (Figure 3C, D, Online Supplementary Figure S6A). Additionally, the forced expression of miR-30e decreased the protein level of Cyb561 in mouse AML cells (Online Supplementary Figure S6B). These findings demonstrate that Cyb561 is a direct downstream target of miR-30e-5p in LSC. Consequently, a key question arises: does miR-30e5p regulate LSC self-renewal through Cyb561? To address this question, we generated retroviral particles expressing Cyb561 and transduced the miRVector- and miR-30e-overexpressing AML cells. The expression level of Cyb561 was upregulated 9.86-fold in miRVector/pMIBSD-Cyb561 AML cells compared with miRVector/pMIBSD AML cells (Online Supplementary Figure S7A). Overexpression of Cyb561 rescued the colony-forming capacity of forced-expressed miR-30e LSC in vitro (Figure 3E). Furthermore, functional assays on LSC revealed that the overexpression of Cyb561 attenuated the effects of miR-30e on leukemia progression by regulating the cell cycle and apoptosis of LSC (Figure 3F, Online Supplementary Figure S7B-D). The limiting dilution assay showed that overexpression of Cyb561 resulted in an approximately 3.80-fold increase in LSC frequency (1/124.7) in AML mice overexpressing miR-30e compared to AML mice transplanted with overexpressing-miR-30e/ pMIBSD-AML cells (1/474.3) (Online Supplementary Table S8). Together, these findings demonstrate that Cyb561 is a direct functional target of miR-30e-5p in LSC. Cyb561 deficiency delays the development of KMT2A::MLLT3-driven acute myeloid leukemia by miR-30e-5p/ROS signaling pathway Since miR-30e-5p suppressed the expression of endogenous Cyb561, we aimed to determine whether the knockdown of endogenous Cyb561 could impair AML development. To address this issue, we knocked down Cyb561 in KMT2A::MLLT3-driven LSC by inducible lentivirus-expressing shRNA. Our data showed that Cyb561 shRNA #1, #2 and #3 reduced the expression of Cyb561 in mouse AML cells by 57%, 80% and 31%, respectively (Online Supplementary Figure S8A). Subsequently, we conducted cell growth and colony-forming/replating assays using KMT2A::MLLT3-driven LSC transduced with scrambled shRNA (control) and two Cyb561 shRNA (#1 and #2). Knockdown of Cyb561 significantly inhibited the cell growth and reduced colony numbers compared to the control group (Online Supplementary Figure S8B, C). We then transplanted an equal number of shRNA-transduced LSC into irradiated recipients. Three to 5 weeks after BM transplantation, we analyzed AML cells in peripheral blood, spleen weight, LSC frequency, cell cycle, and apoptosis of LSC. The results showed that inhibition of Cyb561 decreased the frequency of GFP+ AML cells in peripheral blood, spleen weight, and LSC frequency in BM compared with the control group (Figure 4A, Online Supplementary Figure S8D, E). Suppression of Cyb561 expression also attenuated the cell cycle and promoted apoptosis of

LSC (Figure 4B, C, Online Supplementary Figure S8F). The limiting dilution assay revealed that knockdown of endogenous Cyb561 resulted in an 11.8-fold reduction in LSC frequency (1/335.3) in AML mice transplanted with Cyb561 shRNA #2-AML cells compared to AML mice transplanted with shRNA control (1/28.4) (Online Supplementary Table S9). It has been reported that CYB561 is a type of transmembrane protein consisting of six transmembrane helices and two b-type hemes on each side of the membrane, and it plays a role in the ascorbate recycling process as dehydroascorbate (DHA) reductase.21 Ascorbate oxidation/ reduction is closely associated with ROS homeostasis.22 This suggests a possible link between CYB561 and intracellular ROS, which may be directly or indirectly involved in the cell cycle and apoptosis. Therefore, we investigated whether the knockdown of Cyb561 affects ROS production. Mean fluorescence intensity analysis revealed elevated intracellular ROS levels in Cyb561-knockdown LSC (Online Supplementary Figure S8G). Next, we examined the effect of miR-30e on the level of ROS in LSC from the recipients. As expected, overexpression of miR-30e increased intracellular ROS levels in LSC. However, overexpression of Cyb561 reduced intracellular ROS levels in ectopic miR-30e-expressing LSC (Online Supplementary Figure S7E, F). To further investigate whether miR-30e/Cyb561 impaired LSC by regulating intracellular ROS, we used the ROS scavenger NAC and Tiron to decrease ROS levels in the AML cells. Both NAC and Tiron rescued the colony-forming capacity of LSC overexpressing miR-30e and LSC with Cyb561 knockdown (Online Supplementary Figures S9A and S10A). LSC functional assays, showed that NAC treatment impaired the roles of miR-30e and Cyb561 in LSC and leukemia progression (Figure 4D, E, Online Supplementary Figures S9B-D and S10B-D). Given that CYB561 acts as a reductase in the ascorbate-DHA recycling process and there is an enriched gene set (i.e., glutathione metabolism) in LSC according to GSEA, we tested whether the miR-30e/Cyb561/ROS axis affects the levels of endogenous glutathione, ascorbate and DHA in AML cells. The results revealed a decrease in the levels of glutathione and ascorbate accompanied by an increase in DHA levels in AML cells overexpressing miR-30e and in AML cells with Cyb561 knockdown (Online Supplementary Figure S11). Furthermore, we examined the functions of miR-30e and Cyb561 in normal HSC, given the intriguing findings in LSC. As shown in Online Supplementary Figure S12, neither overexpression of miR-30e nor knockdown of Cyb561 influenced the repopulation capacity of normal HSC. Collectively, our data indicate that miR-30e-5p functions as a negative regulator of the Cyb561-ROS signaling pathway (Figure 4F). Overexpression of Homo sapiens-miR-30e and knockdown of CYB561 impairs human acute myeloid leukemias Although the results from the murine models provided robust evidence for the critical role of miR-30e-5p and

Haematologica | 109 February 2024

416

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

Cyb561 in the initiation and maintenance of murine LSC induced by KMT2A::MLLT3, it remained unclear whether miR-30e-5p or Cyb561 also affects human leukemia. To address this question, we generated retroviral particles expressing Homo sapiens (hsa)-miR-30e and transduced them into six different human leukemic cells, namely THP1, MONOMAC-6, NOMO-1, Kasumi-1, NB4, and human MA9.3 cells. Cell growth/proliferation assays demonstrated that forced expression of hsa-miR-30e inhibited the growth of all human leukemic cells (Figure 5A-D, Online Supplementary Figure S13A-C). Furthermore, overexpression of miR-30e induced extensive apoptosis in human AML cells without affecting their differentiation (Online Supplementary Figure S13D-F). In vitro colony-forming assays revealed reduced colony numbera in the hsa-miR-30e-overexpressing groups in comparison to the miRVector-expressing groups (Figure 5E). Moreover, ectopic expression of hsamiR-30e suppressed the expression level of CYB561 in hu-

man AML cells (Figure 5F). To further investigate whether miR-30e exerted similar effects on the functions of human LSC, we analyzed the cell cycle and apoptosis of human LSC (CD34+CD38–) in human MA9.3 cells, which closely resemble AML cell models containing the KMT2A::MLLT3 fusion found in patients’ AML samples.15 As depicted in Online Supplementary Figure S13G-H, overexpression of miR-30e increased the G1 phase while decreasing the S/ G2/M phases of LSC and promoted the apoptosis of LSC. To assess the relevance of CYB561 in human AML cells, we first examined the clinical significance of CYB561 expression levels in AML patients and the normal human population. The results showed that CYB561 expression was upregulated in BM mononuclear cells of FAB subtypes AML-M1, M2, M3, M4, M5 compared to normal BM monocytes (Figure 6A). Additionally, CYB561 expression was increased in LSC from AML patients compared to normal GMP (Figure 6B). Notably, AML patients with higher levels of

Figure 4. Knockdown of Cyb561 diminishes self-renewal of KMT2A::MLLT3-driven leukemia stem cells by enhancing intracellular levels of reactive oxygen species. (A) Frequency of leukemia stem cell-enriched granulocyte-monocyte progenitors (L-GMP) in the bone marrow (BM) from primary recipients injected with 1x105 acute myeloid leukemia (AML) cells transduced with control short hairpin (sh) RNA (shRNA CTR) or Cyb561 shRNA #2, at week 4 after transplantation (N=5). (B) Cell cycle phase distribution of L-GMP cells in BM from primary recipients injected with shRNA CTR or Cyb561 shRNA #2 AML cells, at week 5 after transplantation (N=4). (C) Percentage of apoptotic L-GMP cells in the BM from primary recipients transplanted with shRNA CTR or Cyb561 shRNA #2 AML cells (N=4). (D) Frequency of L-GMP in the BM from recipients receiving AML cells transduced with shRNA CTR or Cyb561 shRNA #2 following treatment with saline or N-acetyl-L-cysteine (NAC) (N=5). (E) Survival analysis of recipient mice receiving AML carrying shRNA CTR or Cyb561 shRNA #2 after treatment with saline or NAC. The median survival of the four groups of recipients was 32, 30, 55, and 35 days after transplantation (P<0.01, Mantel-Cox test, N=6). (F) Diagrammatic model of the miR-30e-5p/Cyb561/ROS signaling pathway. Data are representative of two or three independent experiments. All data are represented as mean ± standard deviation. Two-tailed Student t tests were used to assess statistical significance (N.S.: not significant; *P<0.05; **P<0.01). AsA: ascorbate; GSSG: oxidated glutathione; GSH: reduced glutathione; DHA: dehydroascorbate; ROS: reactive oxygen species. Haematologica | 109 February 2024

417

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

Figure 5. Overexpression of miR-30e impairs human acute myeloid leukemia cell growth. (A) Quantitative polymerase chain reaction (qPCR) analysis of miR-30e-5p expression in THP-1 cells. Results are normalized to U6 expression and expressed relative to miR-30e-5p expression in the miRVector group (N=3). (B-D) Ectopic expression of miR-30e suppresses the growth of THP-1 (B), MONOMAC-6 (C), and Hu MA9.3 (D) cells (N=3). (E) Colony-forming assay of human leukemia cells by overexpressed-miR-30e (N=3). (F) qPCR analysis of the expression level of CYB561 in THP-1 cells transduced with miRVector or miR-30e. Results are normalized to GADPH expression and expressed relative to CYB561 in the miRVector group (N=3). Data are representative of two or three independent experiments. All data are represented as mean ± standard deviation. Two-tailed Student t tests were used to assess statistical significance (*P<0.05; **P<0.01). miRVector: control; OE-hsa-miR-30e: cells with Homo sapiens miR-303 overexpression.

CYB561 expression displayed a significantly shorter overall survival time, as observed through gene expression profiling interactive analysis (GEPIA: http://gepia.cancer-pku. cn/) (Figure 6C). Next, we examined functional potentials of CYB561 in human AML cells by doxycycline-inducible shRNA. As shown in Figure 7A, CYB561 shRNA #1 and #2 effectively reduced the expression of CYB561 in human leukemic cells by 56% and 45%, respectively. The knockdown of CYB561 suppressed the growth of all human leukemic cells and promoted apoptosis in vitro, while having no impact on the differentiation of human AML cells (Figure 7B-E, Online Supplementary Figure S14A-F). Moreover, inhibition of CYB561 increased the G1 phase while decreasing the S/G2/M phases of human LSC and promoted the apoptosis of human LSC (Online Supplementary Figure S14G, H). To explore the therapeutic potential of targeting miR-30e5p and CYB561 in AML, we designed mimics of hsa-miR30e-5p and siRNA corresponding to the CYB561 coding region. These molecules were delivered into various human leukemic cells. We observed inhibition of human leukemic

cell growth upon treatment with these miRNA mimics and siRNA across all AML cell lines (Online Supplementary Figures S15 and S16). Taken together, these findings provide compelling evidence for the promising therapeutic potential of miR-30e-5p and CYB561 in human AML.

Discussion LSC, a rare population of AML cells, play unique roles in the initiation, maintenance, and propagation of AML. Similar to normal HSC, LSC reside in the hypoxic BM niche and their biological functions are closely related to the intracellular ROS level and oxidative stress status.23 It has been observed that excessively elevated ROS levels can impair the self-renewal capacity of LSC.24 Therefore, targeting ROS and ROS-associated regulators in LSC represents a promising therapeutic strategy for improving long-term outcomes in AML. In this study, we have revealed that miR-30e-5p negatively regulates LSC self-renewal mainly through its direct target Cyb561, a transmembrane protein

Haematologica | 109 February 2024

418

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

Figure 6. CYB561 is upregulated in acute myeloid leukemia patients. (A) The difference in expression of CYB561 between normal bone marrow cells from healthy controls and from patients with different subtypes of acute myeloid leukemia (AML) according to the French-American-British classification. The data were obtained from a public microarray database (access numbers: GSE11504, GSE19429, GSE10358) (normal patients, N=42; M0 patients, N=12; M1 patients, N=35; M2 patients, N=36; M3 patients, N=31; M4 patients, N=29; M5 patients, N=9). (B) The expression of CYB561 between normal granulocyte-monocyte progenitors (GMP) and leukemia stem cell-enriched granulocyte-monocyte progenitors (L-GMP) from patients with AML (normal patients, N=11; AML patients, N=20). (C) Correlation analysis between CYB561 expression level and overall survival in AML patients according to GEPIA. BM: bone marrow; FAB: French-American-British; TPM: transcripts per million; HR: hazards ratio.

that functions as a reductase for regenerating ascorbate and providing reduced iron for transmembrane transport.21 The functional roles of Cyb561 in leukemogenesis have not been previously reported. However, in our investigations, we observed that miR-30e-5p/Cyb561 modulates LSC self-renewal by activating intracellular ROS. Intriguingly, both forced expression of miR-30e-5p and inhibition of Cyb561 have shown the ability to delay leukemia progression, suggesting that both miR-30e-5p and Cyb561 possess potential therapeutic effects on LSC in AML. miR-30e-5p is expressed at a high level in normal HSC and plays a significant role in various crucial biological processes. Previous studies have demonstrated its potential as a tumor suppressor in different types of cancer, such as bladder cancer, nasopharyngeal carcinoma, and liver cancer, and induces apoptosis of chronic myeloid leukemia cells.9,13,25,26 The level of expression of miR-30e-5p is elevated in patients with acute lymphoblastic leukemia and reduced in patients with chronic myeloid leukemia.13,16 Despite these notable findings regarding miR-30e-5p in cancer development and leukemia, to the best of our knowledge, there is no existing evidence demonstrating that overexpression of miR-30e-5p not only delays the onset of AML and targets LSC by regulating the cell cycle and apoptosis in vivo, but also acts as a negative regulator of the transmembrane reductase CYB561. CYB561 is a di-heme transmembrane protein that plays

a crucial role in ascorbate recycling and iron homeostasis and is associated with cellular senescence.21,27 It is expressed highly in metabolically active human tissues, such as the brain, kidney, and heart.28 Despite its clear physiological significance, the functions of Cyb561 in animal cells, particularly in LSC, remain unexplored. Previous studies have indicated that CYB561 mutations in patients result in a novel orthostatic hypotension syndrome, and that CYB561 could serve as a potential prognostic biomarker for breast cancer.28,29 In our current study, we found that miR-30e-5p downregulates the expression of Cyb561 in LSC. Intriguingly, the knockdown of Cyb561 in KMT2A::MLLT3-driven LSC prolongs the survival of recipient mice with a reduced frequency of LSC. Furthermore, we observed a negative correlation between overall survival and CYB561 expression level in AML patients. These findings suggest that Cyb561 acts as a potential suppressor of AML initiation and development and could serve as a biomarker for leukemia prognosis. Through GSEA of downregulated gene sets in miR-30e-overexpressed LSC, we identified significant enrichment of gene sets associated with the electron transport chain, oxidative phosphorylation, and glutathione metabolism. These gene sets are highly relevant to intracellular ROS production and balance. Glutathione, as the most abundant endogenous antioxidant in mammalian cells, plays a crucial role in maintaining a “non-toxic” level of ROS.

Haematologica | 109 February 2024

419

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC

Figure 7. Knockdown of CYB561 suppresses the growth of human acute myeloid leukemia cells. (A) Quantitative polymerase chain reaction analysis of inducible CYB561 knockdown in NB4 cells. NB4 cells were transduced with lenitiviruses containing a scrambled short hairpin (sh) RNA control (CTR), or inducible CYB561 shRNA. After puromycin selection, 1 μg/ mL of doxycycline was added to the medium for 72 h. Results are normalized to HPRT expression and expressed relative to the shRNA control group (N=3). (B-E) Knockdown of CYB561 suppresses the growth of THP-1 (B), NOMO-1 (C), MONOMAC-6 (D), and Hu MA9.3 (E) cells (N=3). Data are representative of two or three independent experiments. All data are represented as mean ± standard deviation. Two-tailed Student t tests were used to assess statistical significance (**P<0.01).

Previous studies have demonstrated a significant reduction in glutathione levels in primary human CD34+ AML cells compared to CD34+ normal BM cells. Consistent with these findings, our study revealed a decrease in endogenous glutathione levels in miR-30e-overexpressing LSC, which was attributed to elevated levels of intracellular ROS, similar to the results observed in Cyb561-knockdown LSC. CYB561, functioning as a reductase, is involved in the reduction process from DHA to ascorbate, and its deficiency can disrupt this process, leading to intracellular ROS accumulation. Our data demonstrated that both overexpression of miR-30e-5p and knockdown of Cyb561 resulted in reduced intracellular ascorbate levels, accompanied by elevated DHA levels in LSC. Impaired ascorbate recycling due to endogenous Cyb561 knockdown consequently led to decreased glutathione levels and increased ROS levels in LSC. Integrating these findings with our results, we propose that miR-30e-5p regulates LSC self-renewal through the Cyb561/ROS signaling pathway, involving intracellular ascorbate/glutathione metabolism. In summary, our study unveils a previously unknown miR30e-5p/Cyb561 axis that finely modulates an intracellular ROS signaling pathway in LSC. Based on our findings, pharmacological targeting of LSC through manipulation of miR-30e-5p or Cyb561, in combination with chemothera-

py, holds great potential as a highly effective strategy to enhance therapeutic regimens. Disclosures No conflicts of interest to disclose. Contributions ZZ and WSW were responsible for the design of the experiments, data analysis, and manuscript preparation. YG, MH, YZ, JW, and LL performed experiments. HZ and YS contributed to the bioinformatics analysis. Funding This research was supported in part by the Natural Science Foundation of Shanghai (grant n. 18ZR1414900), the Program for Professor of Special Appointment (Eastern Scholar) at Shanghai Institutions of Higher Learning, Special Project of Science and Technology Plan of Shaoxing Science and Technology Bureau (grant n. 2020B33004), and the National Natural Science Foundation of China (grant n. 82070106). Data-sharing statement Data that support the findings of this study are available from the corresponding author upon reasonable request.

References 1. Krivtsov AV, Twomey D, Feng Z, et al. Transformation from committed progenitor to leukaemia stem cell initiated by MLL-AF9. Nature. 2006;442(7104):818-822. 2. Wiseman DH, Greystoke BF, Somervaille TCP. The variety of

leukemic stem cells in myeloid malignancy. Oncogene. 2014;33(24):3091-3098. 3. Shallis RM, Wang R, Davidoff A, Ma X, Zeidan AM. Epidemiology of acute myeloid leukemia: recent progress and enduring

Haematologica | 109 February 2024

420

Y. Ge et al.

ARTICLE - miR-30e-5p effectively targets the LSC challenges. Blood Rev. 2019;36:70-87. 4. Thomas D, Majeti R. Biology and relevance of human acute myeloid leukemia stem cells. Blood. 2017;129(12):1577-1585. 5. Zhang L, Nguyen LXT, Chen YC, et al. Targeting miR-126 in inv(16) acute myeloid leukemia inhibits leukemia development and leukemia stem cell maintenance. Nat Commun. 2021;12(1):6154. 6. Gonzales-Aloy E, Connerty P, Salik B, et al. miR-101 suppresses the development of MLL-rearranged acute myeloid leukemia. Haematologica. 2019;104(7):e296-e299. 7. Khalaj M, Woolthuis CM, Hu W, et al. miR-99 regulates normal and malignant hematopoietic stem cell self-renewal. J Exp Med. 2017;214(8):2453-2470. 8. Ma YX, Zhang H, Li XH, Liu YH. MiR-30e-5p inhibits proliferation and metastasis of nasopharyngeal carcinoma cells by targeting USP22. Eur Rev Med Pharmacol Sci. 2018;22(19):6342-6349. 9. Zhang Z, Qin H, Jiang B, et al. miR-30e-5p suppresses cell proliferation and migration in bladder cancer through regulating metadherin. J Cell Biochem. 2019;120(9):15924-15932. 10. Liang Z, Tang S, He R, Luo W, Qin S, Jiang H. The effect and mechanism of miR-30e-5p targeting SNAI1 to regulate epithelial-mesenchymal transition on pancreatic cancer. Bioengineered. 2022;13(4):8013-8028. 11. Chen Z, Qiu H, Ma L, et al. miR-30e-5p and miR-15a synergistically regulate fatty acid metabolism in goat mammary epithelial cells via LRP6 and YAP1. Int J Mol Sci. 2016;17(11):1909. 12. Ding W, Li J, Singh J, et al. miR-30e targets IGF2-regulated osteogenesis in bone marrow-derived mesenchymal stem cells, aortic smooth muscle cells, and ApoE-/- mice. Cardiovasc Res. 2015;106(1):131-142. 13. Hershkovitz-Rokah O, Modai S, Pasmanik-Chor M, et al. MiR-30e induces apoptosis and sensitizes K562 cells to imatinib treatment via regulation of the BCR-ABL protein. Cancer Lett. 2015;356(2 Pt B):597-605. 14. Jia Y, Gu W. Up-regulation of circPVT1 in T cell acute lymphoblastic leukemia promoted cell proliferation via miR30e/DLL4 induced activating NOTCH signaling. Pathol Res Pract. 2021;224:153536. 15. Wei J, Wunderlich M, Fox C, et al. Microenvironment determines lineage fate in a human model of MLL-AF9 leukemia. Cancer Cell. 2008;13(6):483-495. 16. Luna-Aguirre CM, de la Luz Martinez-Fierro M, Mar-Aguilar F, et al. Circulating microRNA expression profile in B-cell acute lymphoblastic leukemia. Cancer Biomark. 2015;15(3):299-310. 17. Schotte D, Chau JC, Sylvester G, et al. Identification of new microRNA genes and aberrant microRNA profiles in childhood

acute lymphoblastic leukemia. Leukemia. 2009;23(2):313-322. 18. Park SM, Cho H, Thornton AM, et al. IKZF2 drives leukemia stem cell self-renewal and inhibits myeloid differentiation. Cell Stem Cell. 2019;24(1):153-165. 19. Chu Y, Chen Y, Li M, et al. Six1 regulates leukemia stem cell maintenance in acute myeloid leukemia. Cancer Sci. 2019;110(7):2200-2210. 20. Fu JF, Shih LY, Yen TH. HOXA11 plays critical roles in disease progression and response to cytarabine in AML. Oncol Rep. 2021;46(1):150. 21. Asard H, Barbaro R, Trost P, Bérczi A. Cytochromes b561: ascorbate-mediated trans-membrane electron transport. Antioxid Redox Signal. 2013;19(9):1026-1035. 22. Shen J, Griffiths PT, Campbell SJ, Utinger B, Kalberer M, Paulson SE. Ascorbate oxidation by iron, copper and reactive oxygen species: review, model development, and derivation of key rate constants. Sci Rep. 2021;11(1):7417. 23. Chen Y, Liang Y, Luo X, Hu Q. Oxidative resistance of leukemic stem cells and oxidative damage to hematopoietic stem cells under pro-oxidative therapy. Cell Death Dis. 2020;11(4):291. 24. Herault O, Hope KJ, Deneault E, et al. A role for GPx3 in activity of normal and leukemia stem cells. J Exp Med. 2012;209(5):895-901. 25. Ma YX, Zhang H, Li XH, Liu YH. MiR-30e-5p inhibits proliferation and metastasis of nasopharyngeal carcinoma cells by target-ing USP22. Eur Rev Med Pharmacol Sci. 2018;22(19):6342-6349. 26. Mao J, Hu X, Pang P, Zhou B, Li D, Shan H. miR-30e acts as a tumor suppressor in hepatocellular carcinoma partly via JAK1/ STAT3 pathway. Oncol Rep. 2017;38(1):393-401. 27. Kang MK, Kameta A, Shin KH, Baluda MA, Kim HR, Park NH. Senescence-associated genes in normal human oral keratinocytes. Exp Cell Res. 2003;287(2):272-281. 28. van den Berg MP, Almomani R, Biaggioni I, et al. Mutations in CYB561 causing a novel orthostatic hypotension syndrome. Circ Res. 2018;122(6):846-854. 29. Yang X, Zhao Y, Shao Q, Jiang G. Cytochrome b561 serves as a potential prognostic biomarker and target for breast cancer. Int J Gen Med. 2021;14:10447-10464. 30. Szarka A, Tomasskovics B, Bánhegyi G. The ascorbateglutathione-α-tocopherol triad in abiotic stress response. Int J Mol Sci. 2012;13(4):4458-4483. 31. Pei S, Minhajuddin M, Callahan KP, et al. Targeting aberrant glutathione metabolism to eradicate human acute myelogenous leukemia cells. J Biol Chem. 2013;288(47):33542-33558.

Haematologica | 109 February 2024

421

ARTICLE - Bone Marrow Failure

Spontaneous remission and loss of monosomy 7: a window of opportunity for young children with SAMD9L syndrome Miriam Erlacher,1,2* Felicia Andresen,1°* Martina Sukova,3 Jan Stary,3 Barbara de Moerloose,4 Jutte van der Werff Ten Bosch,5 Michael Dworzak,6,7 Markus G. Seidel,8 Sophia Polychronopoulou,9 Rita Beier,10 Christian P. Kratz,10 Michaela Nathrath,11,12 Michael C. Frühwald,13 Gudrun Göhring,14 Anke K. Bergmann,14 Christina Mayerhofer,1 Dirk Lebrecht,1 Senthilkumar Ramamoorthy,1,15 Ayami Yoshimi,1 Brigitte Strahm, Marcin W. Wlodarski 1

1,16

and Charlotte M. Niemeyer

1,2

Department of Pediatrics and Adolescent Medicine, Division of Pediatric Hematology and

Oncology, Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany; German Cancer Consortium (DKTK) and German Cancer Research Center (DKFZ), Partner Site

Correspondence: M. Erlacher Miriam.erlacher@uniklinik-freiburg.de Received: Accepted: Early view:

May 25, 2023. August 7, 2023. August 17, 2023.

Freiburg, Germany; 3Department of Pediatric Hematology and Oncology, Second Faculty of Medicine, Charles University and University Hospital Motol, Prague, Czech Republic; 4Department of Pediatric Hematology-Oncology and Stem Cell Transplantation, Ghent University Hospital, Ghent, Belgium; 5Department of Pediatric Hematology-Oncology, University Hospital Brussel, Brussels, Belgium; 6St. Anna Children’s Hospital, Medical University of Vienna, Department of Pediatrics and Adolescent Medicine, Vienna, Austria; 7St. Anna Children’s Cancer Research Institute, Vienna, Austria; 8Division of Pediatric Hematology-Oncology, Department of Pediatrics and Adolescent Medicine, Medical University of Graz, Graz, Austria; 9Department of Pediatric Hematology-Oncology (T.A.O.), Aghia Sophia Children’s Hospital, Athens, Greece; 10Department of Pediatric Hematology and Oncology, Hannover Medical School, Hannover, Germany; 11Department of Pediatric Hematology and Oncology, Klinikum Kassel, Kassel, Germany; 12Department of Pediatrics and Children’s Cancer Research Center, Klinikum rechts der Isar, Technical University of Munich, School of Medicine, Munich, Germany; 13Pediatrics and Adolescent Medicine, University Medical Center Augsburg, Augsburg, Germany; 14Department of Human Genetics, Hannover Medical School, Hannover, Germany; 15Institute of Medical Bioinformatics and Systems Medicine, Medical Center, Faculty of Medicine, University of Freiburg, Freiburg, Germany and Department of Hematology, St. Jude Children’s Research Hospital, Memphis, TN, USA

ME and FA contributed equally as first authors.

Current address: Division of Hematology/Oncology, Boston Children’s Hospital, Harvard Medical

School, Boston, MA, USA.

Abstract Monosomy 7 is the most common cytogenetic abnormality in pediatric myelodysplastic syndrome (MDS) and associated with a high risk of disease progression. However, in young children, spontaneous loss of monosomy 7 with concomitant hematologic recovery has been described, especially in the presence of germline mutations in SAMD9 and SAMD9L genes. Here, we report on our experience of close surveillance instead of upfront hematopoietic stem cell transplantation (HSCT) in seven patients diagnosed with SAMD9L syndrome and monosomy 7 at a median age of 0.6 years (range, 0.4-2.9). Within 14 months from diagnosis, three children experienced spontaneous hematological remission accompanied by a decrease in monosomy 7 clone size. Subclones with somatic SAMD9L mutations in cis were identified in five patients, three of whom attained hematological remission. Two patients acquired RUNX1 and EZH2 mutations during the observation period, of whom one progressed to myelodysplastic syndrome with excess of blasts (MDS-EB). Four patients underwent allogeneic HSCT at a median time of 26 months (range, 14-40) from diagnosis for MDSEB, necrotizing granulomatous lymphadenitis, persistent monosomy 7, and severe neutropenia. At last follow-up, six patients were alive, while one passed away due to transplant-related causes. These data confirm previous observations that monosomy 7 can be transient in young children with SAMD9L syndrome. However, they also indicate that delaying HSCT poses a substantial risk of severe infection and disease progression. Finally, surveillance of patients with SAMD9L syndrome and monosomy 7 is critical to define the evolving genetic landscape and to determine the appropriate timing of HSCT (clinicaltrials gov. Identifier: NCT00662090).

Haematologica | 109 February 2024

422

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome

Introduction Loss of chromosome 7 and partial deletion of its long arm (i.e., monosomy 7 and del(7q)) are frequent non-random cytogenetic aberrations in pediatric patients with myeloid malignancies, including myelodysplastic syndrome (MDS).1-3 Several tumor suppressors and regulators of myeloid differentiation have been identified on chromosome 7. Among them, deletion of EZH2, MLL3/KMT2C, SAMD9L, and CUX1 has been shown to promote malignant transformation in mouse models.4-8 The presence of monosomy 7 is generally associated with a high risk of disease progression and acquisition of oncogenic mutations, and timely allogeneic hematopoietic stem cell transplantation (HSCT) is warranted.1,9-11 However, Scheurlen et al. first reported in 1994 on a 14-month-old boy with MDS and monosomy 7 who achieved spontaneous hematologic recovery within 2 years.12 Subsequently, further case reports and series of transient monosomy 7 in infants and young children with MDS have been published demonstrating spontaneous hematologic recovery upon loss of monosomy 7.13-17 Chromosome 7 aberrations have been associated with several germline conditions predisposing to hematopoietic neoplasia, such as GATA2 deficiency syndrome and Fanconi anemia.18-20 In GATA2 deficiency syndrome, 40% of pediatric patients with cytopenia carry monosomy 7, del(7q), or an unbalanced translocation der(1;7), with the highest prevalence in adolescence.18,19 Recently, a growing body of research has highlighted the association of monosomy 7 and del(7q) with germline mutations in sterile α-motif domain-containing protein 9 (SAMD9) and SAMD9-like (SAMD9L).19,21,22 Gain of function (GOF) mutations in SAMD9 and its paralogue SAMD9L were first reported in 2016 in children with MIRAGE (myelodysplasia, infection, restriction of growth, adrenal hypoplasia, genital phenotype, enteropathy) syndrome and ataxia pancytopenia syndrome (ATXPC), respectively.19,23,24 Our recent analysis of pediatric patients with the histopathological phenotype of refractory cytopenia of childhood (RCC) and MDS enrolled in the registry of the European Working Group of MDS in Childhood (EWOG-MDS) revealed that patients with germline SAMD9 and SAMD9L mutations presented with a broad and overlapping phenotypic spectrum. The majority of cases presented with RCC, whereas MDS with excess of blasts (MDS-EB) was diagnosed in only 7% of patients. Furthermore, chromosome 7 aberrations were observed in 55% of all SAMD9and SAMD9L-mutated cases.19 SAMD9 and SAMD9L are located adjacent to each other on chromosome 7. Germline SAMD9/9L GOF mutations result in reduced proliferation and survival of hematopoietic stem and progenitor cells (HSPC).15,19,25,26 Acquisition of monosomy 7 and del(7q) can be considered somatic rescue events because they result in a non-random loss of the mutant allele, conferring an increased competitive

M. Erlacher et al.

fitness of HSPC. Yet, concomitant loss of tumor suppressors located on chromosome 7 renders monosomy 7 and del(7q) a maladaptive somatic mechanism with potential for leukemic transformation. Other somatic rescue events frequently detected in bone marrow (BM) cells of affected patients include somatic loss of function (LOF) mutations of SAMD9/9L in cis and uniparental isodisomy 7q (UPD7q) with loss of the mutated allele.19,21,25,27 Due to the strong selection pressure, multiple hematopoietic clones with clone-defining somatic mutations can simultaneously exist within the marrow. Moreover, in line with the distinct plasticity of hematopoiesis in young children, the polyclonal composition of BM cells may change over time.19 Consequently, monosomy 7 may spontaneously disappear in patients with SAMD9/9L germline disorders, followed by hematologic recovery.15,16,25,27 This phenomenon, primarily observed in infants, represents a form of “natural gene therapy” that might spare patients from therapy like HSCT. However, little is known about the frequency, natural history of loss of monosomy 7, and factors influencing outcome of affected patients.

Methods Seven patients less than 5 years of age diagnosed with a SAMD9L germline disorder with monosomy 7 and normal blast percentage in BM were followed closely to allow for spontaneous recovery. BM examinations every 3 to 4 months included histopathology, cytogenetics, and SAMD9L sequencing. Six of the seven patients were enrolled in the prospective study EWOG-MDS 2006 (clinicaltrials gov. Identifier: NCT00662090); for patient 5 (P5), parental consent was obtained for the respective data analyses. University of Freiburg Institutional Ethics Committee approved the research (ethics vote no. 247/05). Patient 2 (P2, D1300) has already been described by Sahoo et al.19 The diagnosis of RCC and MDS-EB was established according to the International Consensus Classification (ICC) of hematologic neoplasms.28 Cytogenetics included classical karyotyping and interphase fluorescence in situ hybridization (FISH). Targeted next-generation sequencing (NGS) of SAMD9L allowed determination of the variant allele frequency (VAF) and detection of newly acquired somatic SAMD9L mutations in cis. In order to screen for somatic oncogenic mutations and corrective UPD7q, a custom-made NGS panel including genes frequently mutated in myeloid neoplasia (in the following called “myeloid NGS panel”, see the Online Supplementary Appendix S1) and single nucleotide polymorphism (SNP) array were performed; data from the BM sample prior to HSCT or at last follow-up is reported. The American College of Medical Genetics and Association for Molecular Pathology (ACMG-AMP) 2015 guidelines were employed for the interpretation of variants identified in the myeloid NGS panel.29

Haematologica | 109 February 2024

423

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome

Results Clinical presentation with pancytopenia, immunodeficiency, and multiple non-hematological phenotypes We report on seven patients with SAMD9L germline disorder who presented with cytopenia and monosomy 7 at a median age of 0.6 years (range, 0.4-2.9). At diagnosis, all patients had moderate to severe neutropenia and three children (P1, P2, P3) required platelet transfusions. Six patients had normocytic (n=5) or macrocytic (n=1) anemia, one patient had a normal hemoglobin concentration with macrocytosis of red cells (P5). Histopathology was compatible with RCC in all seven patients. Immunodeficiency with a variable clinical presentation was present in six patients. The most consistent findings of immunological impairment were hypogammaglobulinemia found in five patients (P1, P2, P5, P6, P7) and B- and natural killer (NK)-cell deficiency present in four (P2, P5, P6, P7). Patient characteristics are summarized in Table 1 and in the Online Supplementary Appendix; complete genetic data are listed in the Online Supplementary Table S1. Six of the seven patients had a non-hematological phenotype compatible with SAMD9L germline disorder (Table 1). Three patients were born small for gestational age (P1, P2, P7). Patient 1 (P1) was also diagnosed with cerebellar atrophy and global developmental delay. Patient 2 (P2) was a hypotrophic preterm neonate born at 36 gestational weeks with bilateral cleft lip and palate. Patient 3 (P3) presented with macrocephaly. Patient 5 (P5) was a triplet born preterm at 30 gestational weeks who developed hydrocephalus requiring a ventriculoperitoneal shunt in the first year of life. Patient 6 (P6) presented with mild macrocephaly and short thumbs. Patient 7 (P7) had enlarged eye bulbs and failure to thrive. Spontaneous hematopoietic recovery and clinical course With a median follow-up of 43 months (range, 40-55), three of the seven patients (P3, P4, P6) experienced spontaneous regression of monosomy 7 (Figure 1; Online Supplementary Figure S1). In patient 4 (P4) and patient 6 (P6), monosomy 7 was no longer detectable 6 and 16 months from diagnosis, respectively. P4 presented at 7 months of age with pancytopenia, requirement for platelet transfusions, and a normal karyotype. Three months after the initial presentation, a small monosomy 7 clone of 7% was noted, which was no longer detectable 3 months later. Concurrently, the absolute neutrophil count (ANC) slowly increased to 1.23x109/L at 3 months and 3.97x109/L at 6 months from diagnosis, and hemoglobin concentration and platelet count reached normal values at six and 14 months, respectively. Concomitantly, BM cellularity increased to an age-adjusted normal value. P6 presented with neutropenia, macrocytosis, and moderate thrombocytopenia at the age of 2.8 years. BM analysis was compatible with hypocellular RCC

M. Erlacher et al.

with a monosomy 7 clone in 63% of BM interphases. The ANC had slowly increased 7 months after diagnosis and reached close to normal values (ANC 1.47x109/L) 21 months from diagnosis. By that time, the monosomy 7 clone was no longer detectable and hemoglobin and platelet count were normalized at 16 and 32 months from diagnosis, respectively. P3 presented with pancytopenia at the age of 4 months. BM histopathology was compatible with hypocellular RCC with monosomy 7 in 37% of BM interphases. Nine months later, hemoglobin concentration and platelet count had normalized, but isolated neutropenia (ANC 0.38x109/L) persisted. Thirteen months after diagnosis, trephine biopsy showed a normocellular BM according to age and the complete blood count (CBC) gradually improved. Further invasive procedures were denied, the child was alive with a normalized CBC 4.3 years after diagnosis (age 4.5 years). All four patients that did not experience spontaneous remission (P1, P2, P5, P7) received an allogeneic HSCT with a median interval from diagnosis to HSCT of 26 months (range, 14-40) (Online Supplementary Figure S1). Indications for HSCT were severe neutropenia and/or bacterial infection (P1, P7), persistence of monosomy 7 (P5), and disease progression to MDS-EB (P2). Three of the four transplanted patients (P1, P5, P7) were alive with stable engraftment and complete donor chimerism at last follow-up. P2 who succumbed to transplant-related causes had disease progression prior to HSCT. At the age of 15 months (i.e., 10 months after diagnosis), the patient developed severe neutropenia prompting a 12-day treatment course with granulocyte colony-stimulating factor (G-CSF). Subsequently, the BM blast percentage and cellularity increased consistent with progression to MDS-EB and persisted after G-CSF withdrawal. At the time, two RUNX1 variants (see below) were detected in the BM. After a myeloablative conditioning regimen and an allogeneic HSCT from a matched unrelated donor, the patient developed acute respiratory distress syndrome and veno-occlusive disease and died in hematologic remission. Somatic events influencing disease outcome The three patients with spontaneous remission (P3, P4, P6) had somatic SAMD9L LOF mutations in cis known to disrupt the germline allele. Figure 1 depicts the courses of the VAF of SAMD9L, monosomy 7, and ANC. P3 acquired a somatic SAMD9L variant (SAMD9L c.1765C>T, p.R589*) 13 months from diagnosis (VAF 12%), molecular analyses at later time points could not be obtained. P4 acquired two somatic SAMD9L variants with a VAF of 27% (SAMD9L c.683G>A, p.C228Y) and 4% (SAMD9L c.2699>G, p.Y900C) 6 months from diagnosis concomitantly with loss of the previously diagnosed monosomy 7. Both SAMD9L clones remained stable over time, a third somatic SAMD9L variant (SAMD9L c.3562C>T, p.R1188*, VAF 4%) was detected 41 months after diagnosis. P6 harbored a somatic SAMD9L variant (SAMD9L c.4224dupA, p.Q1409Tfs*49) with a VAF of 6% at diagnosis. The somatic SAMD9L variant slowly in-

Haematologica | 109 February 2024

424

M. Erlacher et al.

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome

creased to 35%, while the monosomy 7 clone decreased in size over time and subsequently disappeared at 16 months from diagnosis. Importantly, we also detected somatic SAMD9L mutations in cis in two patients (P2, P7) who did not experience hematological recovery and/or a decrease in monosomy 7 clone size (Online Supplementary Table S1; Online Supplementary Figure S1). P2 acquired a somatic SAMD9L variant 1 month after diagnosis (SAMD9L c.768dup, p.K257*, VAF 6%). Within 12 months, the VAF of this variant increased

to 23%, however, the disease progressed to MDS-EB. P7 had a large monosomy 7 clone at diagnosis and acquired a somatic SAMD9L variant (SAMD9L c.2385C>A, p.Y795*) with a VAF of 6% 6 months later. The patient received an allograft for persistent severe neutropenia 10 months later. The myeloid NGS panel analysis demonstrated that two patients (P2, P5) had acquired oncogenic mutations during the course of their disease (Online Supplementary Table S1; Online Supplementary Figure S1). P2 acquired two pathogenic RUNX1 mutations 12 months from diagnosis

Table 1. Phenotype and clinical course of seven patients with SAMD9L germline mutations and monosomy 7. CBC at dx Age Patient N (WBC, Sex at dx (UPN) ANC, PLT, in mths Hb, MCV)

P1 (D1297)

P2 (D1300)

P3 (GR012)

P4 (B063)

P5 (KM)

P6 (CZ132)

P7 (A146)

Non-hematological phenotype

Somatic SAMD9L mutation at last FU (N)

Immunological phenotype

6.40x109/L, SGA, cerebellar atrophy, 0.70x109/L, global developmental delay 100x109/L #, 9.5 g/dL, Hypogammaglobulinemia 79 fL 3.90x109/L, 0.39x109/L, 3x109/L, 6.7 g/dL, 75 fL 3.14x109/L, 0.35x109/L, 12x109/L, 8 g/dL, 80 fL 7.10x109/L, 0.71x109/L, 107x109/L #, 8 g/dL, 78 fL 4.50x109/L, 0.50x109/L, 170x109/L, 12.3 g/dL, 92 fL 4.10x109/L, 0.81x109/L, 46x109/L, 11 g/dL, 94 fL 2.54x109/L, 0.25x109/L, 79x109/L, 7.7 g/dL, 89 fL

Preterm infant (36 GW), SGA, cleft lip and palate

Monosomy 7 Age at HSCT Somatic clone size in mths cancer gene at dx mutation at HSCT Loss of last FU (N) indication monosomy 7 73%

46 No

Necrotizing granulomatous lymphadenitis

80%

19 RUNX1 (2)

Yes (1)

Hypogammaglobulinemia, B/NK cell deficiency

Status at last FU

Alive

Dead MDS-EB

Macrocephaly Yes (1) None

None

7% Yes (3)

None

Yes

Preterm triplet (30 GW), hydrocephalus

52% No

Hypogammaglobulinemia, B/NK cell deficiency

Mild macrocephaly, short thumbs

Alive

RUNX1 (1), EZH2 (2)

53 Alive Persistent -7

63% Yes (1)

Hypogammaglobulinemia, B cell deficiency

Alive

Yes

SGA, failure to thrive, big eye bulbs

Yes (1)

Hypogammaglobulinemia, B/NK cell deficiency

66 % No

32 Severe neutropenia

Alive

Post-transfusion. SAMD9L: sterile α-motif domain-containing protein 9-like; N: number; UPN: unique patient number; dx: diagnosis; mths: months; CBC: complete blood count; WBC: white blood cells; ANC: absolute neutrophil count; PLT: platelets; Hb: hemoglobin; MCV: mean corpuscular volume; FU: follow-up; HSCT: hematopoietic stem cell transplantation; SGA: small for gestational age; NA: not applicable; GW: weeks of gestation; MDS-EB: myelodysplastic syndrome with excess blasts. #

Haematologica | 109 February 2024

425

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome

M. Erlacher et al. Figure1. Course of SAMD9L variant allele frequency, monosomy 7 (analyzed by fluorescence in situ hybridization), and absolute neutrophil count of patient 3, 4, and 6. ANC: absolute neutrophil count; BM: bone marrow; SAMD9L: sterile α-motif domain-containing protein 9-like.

Haematologica | 109 February 2024

426

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome

(RUNX1 c.317G>A, p.W106*; RUNX1 c.496_508+2dup), which was accompanied by disease progression to MDS-EB. In P5, a pathogenic variant in RUNX1 (RUNX1 c.593A>G, p.Asp198Gly) and a variant of unknown significance in EZH2 (EZH2 c.1672+3_1672+4del) were detected 30 months after diagnosis. At this time point, the monosomy 7 clone size had increased again to 12% following an initial decrease from 52% to 4% and the patient had already been scheduled for HSCT for persistent monosomy 7.

Discussion Somatic genetic rescue (SGR) events that (at least in part) abrogate an underlying deleterious germline defect and thereby confer a selective advantage have been described in several inherited bone marrow failure and leukemia predisposition syndromes.30-36 Prime examples are SAMD9/9L germline disorders with multiple SGR events typically resulting in a polyclonal BM (Figure 2).19 This case series describes the natural history of transient monosomy 7 in seven children with SAMD9L germline disorders below the age of 5 years at diagnosis. Spontaneous hematologic recovery was observed in three children; complete loss of monosomy 7 was demonstrated in two children and can be assumed in the third patient who was withdrawn from follow-up BM studies but normalized his CBC. In all three patients, improvement of blood counts was accompanied by the emergence and/or expansion of clones harboring somatic SAMD9L mutations in cis as well as a gradual decrease in monosomy 7 clone size. However, somatic SAMD9L mutations in cis were also detected in two patients who did not experience hematologic recovery.

M. Erlacher et al.

This finding underlines that the acquisition of adaptive SGR clones such as somatic SAMD9L mutations in cis does not reliably predict loss of monosomy 7 or favorable clinical outcome.19 Interestingly, we did not detect somatic genetic reversion by UPD7q with non-random loss of the SAMD9L germline variant in either of the two patients with spontaneous remission and BM material available for SNP array. Whether somatic SAMD9L mutations in cis and UPD7q are functionally equivalent with respect to longterm sustainability of normal hematopoietic function is currently unknown. It is well established that the somatic mutational landscape in children with MDS differs from that observed in adults.19,21,37-39 The nature of somatic variants noted in pediatric MDS is largely dependent on the underlying genetic predisposition,19,34,36,40,41 the morphological subtype (MDS-EB vs. RCC), and the karyotype (monosomy 7 vs. other karyotypes).15,21 Our recently published data demonstrated that SAMD9/9L-mutated patients most frequently harbored mutations in SETBP1, ASXL1, RUNX1, EZH2, and the RAS pathway genes, thereby resembling the somatic mutational pattern of GATA2 deficiency and MDS with monosomy 7 without underlying GATA2 or SAMD9/9L germline mutations.19 In two of the patients described here, we observed the emergence of somatic mutations in RUNX1 and EZH2 during BM surveillance. Somatic variants in RUNX1 play a critical role in leukemic transformation in a number of predisposition syndromes, specifically in severe congenital neutropenia SCN and Fanconi anemia: Skokowa et al. detected somatic RUNX1 mutations in 64% of pediatric patients with MDS/ acute myeloid leukemia following SCN,40 while Sebert et al. described RUNX1 alterations in 34% of Fanconi anemia patients with clonal hematopoiesis.36

Figure 2. Somatic rescue events in SAMD9/9L germline disorders. UPD: uniparental isodisomy; LOF: loss-of-function; HSPC: hematopoietic stem and progenitor cells; SAMD9L: sterile α-motif domain-containing protein 9-like. Created with BioRender. com. Haematologica | 109 February 2024

427

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome

Meticulous description of the sequence of acquisition of individual somatic aberrations during leukemogenesis is a prerequisite for unraveling their impact on outcome of patients with pediatric MDS. Schwartz et al. reported that the presence of somatic SETBP1 mutations is associated with a poor outcome in patients with MDS-EB.21 SETBP1 mutations are often noted in the context of monosomy 7.15 In our previous report, 16 of 59 patients (27%) with SAMD9/9L germline disorders and normal blast count carried somatic oncogenic mutations, and twelve of the 16 patients had monosomy 7. The presence of monosomy 7 had a negative impact on patient outcome; among patients with SAMD9/9L syndrome and normal blast count, 5-year overall survival was 93% in patients with normal karyotype and 73% in the presence of monosomy 7.19 To the best of our knowledge, somatic genetic rescue with complete hematological recovery in SAMD9L syndrome and monosomy 7 has only been observed in patients less than 5 years of age.15,16,19,22,42 This age dependency may indicate that BM plasticity substantially decreases during the first years of life. At the same time, it provides a window of opportunity for BM surveillance that might spare some of these affected children from allogeneic HSCT. Reoccurrence of monosomy 7 has not been described, the longest reported follow-up is 20 years.15 Nevertheless, monosomy 7 remains a risk factor for potential disease progression. Although BM sampling is an invasive procedure requiring analgesic sedation or anesthesia in small children, frequent BM examinations (e.g., q3-4 months) are advised to detect emerging oncogenic mutations early during malignant transformation. Of note, two of the seven patients included in the surveillance strategy reported were transplanted for infection and severe neutropenia. When initiating a surveillance strategy, treating physicians need to be aware that these young children with severe to moderate neutropenia, B- and NK-cell deficiency, and/or hypogammaglobulinemia are at increased risk for life-threatening infections. This case series has two main limitations. First, the small number of patients renders it difficult to draw definitive conclusions about patient management. Furthermore, since our myeloid NGS panel has a detection limit of 5%, small somatic clones might have been missed in some patients. Future studies of larger patient cohorts using more sensitive sequencing methods such as duplex unique molecular identifiers sequencing43 or single-cell sequencing approaches could address these open questions. In summary, surveillance instead of upfront HSCT can give some patients less than 5 years of age with SAMD9L syndrome and monosomy 7 a chance to experience spontaneous

M. Erlacher et al.

hematological remission. However, stringent indications for HSCT are recommended to render this expectant approach a safe procedure in this young patient population. Although the mechanism of loss of monosomy 7 is not fully elucidated yet, clinical experience indicates that expansion of clones with somatic rescue mutations in cis is sufficient to allow for hematopoietic regeneration in some patients. Such “natural gene therapy” provides a promising template for novel gene therapy approaches. Introducing stop mutations, which are the most frequent somatic rescue events in SAMD9L disease, is technically well feasible rendering SAMD9/9L germline syndromes most amenable to gene therapy. Disclosures No conflicts of interest to disclose. Contributions ME and FA designed the research, analyzed, interpreted clinical data and wrote the manuscript. CM, CMN and MWW contributed to the manuscript conception. ME, BS, AY, CMN, MS, JS, BdW, JvdWTB, MD, MGS, SP, RB, CK, MN and MCF were involved in patient care, testing and data presentation. All authors contributed to the manuscript and approved its final version. Acknowledgments We thank A-R. Kaya, M. Teller, C. Jaeger, and S. Zolles, for excellent laboratory assistance; P. Noellke for assistance with the statistical analysis; A. Breier, W. Truckenmueller, and A. Gebert for data management (all University of Freiburg). We acknowledge the contribution of the Hilda Biobank at the Department of Pediatrics and Adolescent Medicine, Freiburg, Germany. We are also deeply grateful for continuous effort of the National Reference Pathologists, National Reference Cytogeneticists, physicians, nurses and other staff of pediatric hematology/oncology units and transplant centers in all participating countries within the EWOG-MDS consortium (www.ewog-mds-saa.org). Funding This work was generated within the European Reference Network for Pediatric Cancer (PAEDCAN). It was supported by the German Federal Ministry of Education and Research (BMBF) 01GM1911A “MyPred - Network for young individuals with syndromes predisposing to myeloid malignancies” to BS, CMN, GG, ME, AY and MWW. Data-sharing statement All data relevant to the study are included in the article or uploaded in the Online Supplementary Appendix.

Haematologica | 109 February 2024

428

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome

M. Erlacher et al.

References 1. Hasle H, Arico M, Basso G, et al. Myelodysplastic syndrome, juvenile myelomonocytic leukemia, and acute myeloid leukemia associated with complete or partial monosomy 7. European Working Group on MDS in Childhood (EWOG-MDS). Leukemia. 1999;13(3):376-385. 2. Stieglitz E, Taylor-Weiner AN, Chang TY, et al. The genomic landscape of juvenile myelomonocytic leukemia. Nat Genet. 2015;47(11):1326-1333. 3. Gohring G, Michalova K, Beverloo HB, et al. Complex karyotype newly defined: the strongest prognostic factor in advanced childhood myelodysplastic syndrome. Blood. 2010;116(19):3766-3769. 4. Inaba T, Honda H, Matsui H. The enigma of monosomy 7. Blood. 2018;131(26):2891-2898. 5. An N, Khan S, Imgruet MK, et al. Gene dosage effect of CUX1 in a murine model disrupts HSC homeostasis and controls the severity and mortality of MDS. Blood. 2018;131(24):2682-2697. 6. Chen C, Liu Y, Rappaport AR, et al. MLL3 is a haploinsufficient 7q tumor suppressor in acute myeloid leukemia. Cancer Cell. 2014;25(5):652-665. 7. Mochizuki-Kashio M, Aoyama K, Sashida G, et al. Ezh2 loss in hematopoietic stem cells predisposes mice to develop heterogeneous malignancies in an Ezh1-dependent manner. Blood. 2015;126(10):1172-1183. 8. Nagamachi A, Matsui H, Asou H, et al. Haploinsufficiency of SAMD9L, an endosome fusion facilitator, causes myeloid malignancies in mice mimicking human diseases with monosomy 7. Cancer Cell. 2013;24(3):305-317. 9. Kardos G, Baumann I, Passmore SJ, et al. Refractory anemia in childhood: a retrospective analysis of 67 patients with particular reference to monosomy 7. Blood. 2003;102(6):1997-2003. 10. Hasle H, Alonzo TA, Auvrignon A, et al. Monosomy 7 and deletion 7q in children and adolescents with acute myeloid leukemia: an international retrospective study. Blood. 2007;109(11):4641-4647. 11. Pastor V, Hirabayashi S, Karow A, et al. Mutational landscape in children with myelodysplastic syndromes is distinct from adults: specific somatic drivers and novel germline variants. Leukemia. 2017;31(3):759-762. 12. Scheurlen W, Borkhardt A, Ritterbach J, Huppertz HI. Spontaneous hematological remission in a boy with myelodysplastic syndrome and monosomy 7. Leukemia. 1994;8(8):1435-1438. 13. Mantadakis E, Shannon KM, Singer DA, et al. Transient monosomy 7: a case series in children and review of the literature. Cancer. 1999;85(12):2655-2661. 14. Parker TM, Klaassen RJ, Johnston DL. Spontaneous remission of myelodysplastic syndrome with monosomy 7 in a young boy. Cancer Genet Cytogenet. 2008;182(2):122-125. 15. Pastor VB, Sahoo SS, Boklan J, et al. Constitutional SAMD9L mutations cause familial myelodysplastic syndrome and transient monosomy 7. Haematologica. 2018;103(3):427-437. 16. Wong JC, Bryant V, Lamprecht T, et al. Germline SAMD9 and SAMD9L mutations are associated with extensive genetic evolution and diverse hematologic outcomes. JCI Insight. 2018;3(14):e121086. 17. Tsang MMC, Ha SY, Chan N, So CC, Cheuk DKL, Chan GCF. Spontaneous resolution of refractory cytopenia of childhood with monosomy 7 in an infant without an identifiable genetic cause. Pediatr Blood Cancer. 2022;69(10):e29654. 18. Kozyra EJ, Gohring G, Hickstein DD, et al. Association of unbalanced translocation der(1;7) with germline GATA2

mutations. Blood. 2021;138(23):2441-2445. 19. Sahoo SS, Pastor VB, Goodings C, et al. Clinical evolution, genetic landscape and trajectories of clonal hematopoiesis in SAMD9/SAMD9L syndromes. Nat Med. 2021;27(10):1806-1817. 20. Dutzmann CM, Spix C, Popp I, et al. Cancer in children with Fanconi anemia and ataxia-telangiectasia - a nationwide register-based cohort study in Germany. J Clin Oncol. 2022;40(1):32-39. 21. Schwartz JR, Ma J, Lamprecht T, et al. The genomic landscape of pediatric myelodysplastic syndromes. Nat Commun. 2017;8(1):1557. 22. Bluteau O, Sebert M, Leblanc T, et al. A landscape of germ line mutations in a cohort of inherited bone marrow failure patients. Blood. 2018;131(7):717-732. 23. Narumi S, Amano N, Ishii T, et al. SAMD9 mutations cause a novel multisystem disorder, MIRAGE syndrome, and are associated with loss of chromosome 7. Nat Genet. 2016;48(7):792-797. 24. Chen DH, Below JE, Shimamura A, et al. Ataxia-pancytopenia syndrome is caused by missense mutations in SAMD9L. Am J Hum Genet. 2016;98(6):1146-1158. 25. Tesi B, Davidsson J, Voss M, et al. Gain-of-function SAMD9L mutations cause a syndrome of cytopenia, immunodeficiency, MDS, and neurological symptoms. Blood. 2017;129(16):2266-2279. 26. Thomas ME, 3rd, Abdelhamed S, Hiltenbrand R, et al. Pediatric MDS and bone marrow failure-associated germline mutations in SAMD9 and SAMD9L impair multiple pathways in primary hematopoietic cells. Leukemia. 2021;35(11):3232-3244. 27. Buonocore F, Kuhnen P, Suntharalingham JP, et al. Somatic mutations and progressive monosomy modify SAMD9-related phenotypes in humans. J Clin Invest. 2017;127(5):1700-1713. 28. Rudelius M, Weinberg OK, Niemeyer CM, Shimamura A, Calvo KR. The International Consensus Classification (ICC) of hematologic neoplasms with germline predisposition, pediatric myelodysplastic syndrome, and juvenile myelomonocytic leukemia. Virchows Arch. 2023;482(1):113-130. 29. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-424. 30. Ramirez MJ, Pujol R, Trujillo-Quintero JP, et al. Natural gene therapy by reverse mosaicism leads to improved hematology in Fanconi anemia patients. Am J Hematol. 2021;96(8):989-999. 31. Jongmans MC, Verwiel ET, Heijdra Y, et al. Revertant somatic mosaicism by mitotic recombination in dyskeratosis congenita. Am J Hum Genet. 2012;90(3):426-433. 32. Sharma R, Sahoo SS, Honda M, et al. Gain-of-function mutations in RPA1 cause a syndrome with short telomeres and somatic genetic rescue. Blood. 2022;139(7):1039-1051. 33. Perdigones N, Perin JC, Schiano I, et al. Clonal hematopoiesis in patients with dyskeratosis congenita. Am J Hematol. 2016;91(12):1227-1233. 34. Kennedy AL, Myers KC, Bowman J, et al. Distinct genetic pathways define pre-malignant versus compensatory clonal hematopoiesis in Shwachman-Diamond syndrome. Nat Commun. 2021;12(1):1334. 35. Revy P, Kannengiesser C, Fischer A. Somatic genetic rescue in Mendelian haematopoietic diseases. Nat Rev Genet. 2019;20(10):582-598.

Haematologica | 109 February 2024

429

ARTICLE - Transient monosomy 7 in young children with SAMD9L syndrome 36. Sebert M, Gachet S, Leblanc T, et al. Clonal hematopoiesis driven by chromosome 1q/MDM4 trisomy defines a canonical route toward leukemia in Fanconi anemia. Cell Stem Cell. 2023;30(2):153-170. 37. Coutinho DF, Boroni M, Batalha ABW, et al. Somatic genomic variants in refractory cytopenia of childhood. J Pediatr Hematol Oncol. 2021;6(3):123-126. 38. Noy-Lotan S, Krasnov T, Dgany O, et al. Incorporation of somatic panels for the detection of haematopoietic transformation in children and young adults with leukaemia predisposition syndromes and with acquired cytopenias. Br J Haematol. 2021;193(3):570-580. 39. Tan S, Kermasson L, Hilcenko C, et al. Somatic genetic rescue of a germline ribosome assembly defect. Nat Commun. 2021;12(1):5044.

40. Skokowa J, Steinemann D, Katsman-Kuipers JE, et al. Cooperativity of RUNX1 and CSF3R mutations in severe congenital neutropenia: a unique pathway in myeloid leukemogenesis. Blood. 2014;123(14):2229-2237. 41. West RR, Calvo KR, Embree LJ, et al. ASXL1 and STAG2 are common mutations in GATA2 deficiency patients with bone marrow disease and myelodysplastic syndrome. Blood Adv. 2022;6(3):793-807. 42. Csillag B, Ilencikova D, Meissl M, et al. Somatic mosaic monosomy 7 and UPD7q in a child with MIRAGE syndrome caused by a novel SAMD9 mutation. Pediatr Blood Cancer. 2019;66(4):e27589. 43. Schmitt MW, Kennedy SR, Salk JJ, Fox EJ, Hiatt JB, Loeb LA. Detection of ultra-rare mutations by next-generation sequencing. Proc Natl Acad Sci U S A. 2012;109(36):14508-14513.

Haematologica | 109 February 2024

430

M. Erlacher et al.

ARTICLE - Cell Therapy & Immunotherapy

Allogeneic hematopoietic stem cell transplantation in patients aged 60-79 years in Germany (1998-2018): a registry study Jan Frederic Weller,1 Claudia Lengerke,1 Jürgen Finke,2 Johannes Schetelig,3 Uwe Platzbecker,4 Hermann Einsele,5 Thomas Schroeder,6 Christoph Faul,1 Matthias Stelljes,7 Peter Dreger,8 Igor W. Blau,9 Gerald Wulf,10 Johanna Tischer,11 Christoph Scheid,12 Ahmet Elmaagacli,13 Helga Neidlinger,14 Sarah Flossdorf,14,15 Martin Bornhäuser,3 Wolfgang Bethge,1 Katharina Fleischhauer,14,16 Nicolaus Kröger,14,17 Liesbeth C. de Wreede18,19 and Maximilian Christopeit;1 for Deutsches Register für Stammzelltransplantationen DRST/German Registry for Stem Cell

Correspondence: M. Christopeit m.christopeit@uke.de Received: Accepted: Early view:

March 20, 2023. August 11, 2023. August 31, 2023.

Transplantation and the Cooperative Group for Stem Cell Transplantation of the German

https://doi.org/10.3324/haematol.2023.283175

Working Group for Transplantation and Cellular Therapy (DAG-HSZT)

Department of Internal Medicine II, Hematology, Oncology, Clinical Immunology and

Rheumatology, University Hospital Tübingen, Tübingen, Germany; 2University Medical Center Freiburg, Department of Hematology, Oncology and Stem Cell Transplantation, Faculty of Medicine, University of Freiburg, Freiburg, Germany; 3Department of Internal Medicine I, University Hospital Carl Gustav Carus, TU Dresden, Dresden, Germany; 4Medical Clinic and Policlinic I, Hematology and Cellular Therapy, Leipzig University Hospital, Leipzig, Germany; Department of Internal Medicine II, University Hospital of Würzburg, Würzburg, Germany;

Department of Hematology and Stem Cell Transplantation, West German Cancer Center

Essen, University Hospital Essen, Essen, Germany; 7Department of Medicine A, University Hospital of Münster, Münster, Germany; 8Department of Medicine V, University of Heidelberg, Heidelberg, Germany; 9Medical Clinic, Charité University Medicine Berlin, Berlin, Germany; Hematology and Medical Oncology, University Medicine Göttingen, Göttingen, Germany;

Internal Medicine III, Hematology/Oncology/Stem Cell Transplantation, Ludwig-Maximilians-

University, Munich, Germany; 12Faculty of Medicine and Cologne University Hospital, Center for Integrated Oncology Aachen-Bonn-Cologne-Düsseldorf (CIO ABCD), University of Cologne, Cologne, Germany; 13Department of Hematology/Oncology and Stem Cell Transplantation, Asklepios Klinik St. Georg, Hamburg, Germany; 14German Registry for Stem Cell Transplantation, DRST, Ulm, Germany; 15Institute for Medical Informatics, Biometry and Epidemiology (IMIBE), University of Duisburg-Essen, Essen, Germany; 16Institute for Experimental Cellular Therapy, University Hospital Essen, Essen, Germany; 17Department of Stem Cell Transplantation, University Medical Center Eppendorf, Hamburg, Germany; Department of Biomedical Data Sciences, Leiden University Medical Center, Leiden, The

Netherlands and 19DKMS Clinical Trials Unit, Dresden, Germany

Abstract Incidences of diseases treated with transplantation frequently peak at higher age. The contribution of age to total risk of transplantation has not been estimated amidst an aging society. We compare outcomes of 1,547 patients aged 70-79 years and 9,422 patients aged 60-69 years transplanted 1998-2018 for myeloid, lymphoid and further neoplasia in Germany. To quantify the contribution of population mortality to survival, we derive excess mortality based on a sex-, year- and agematched German population in a multistate model that incorporates relapse and graft-versus-host-disease (GvHD). Overall survival, relapse-free survival (RFS) and GvHD-free-relapse-free survival (GRFS) is inferior in patients aged 70-79 years, compared to patients aged 60-69 years, with 36% (95% Confidence Interval [CI]: 34-39%) versus 43% (41-44%), 32% (3035%) versus 36% (35-37%) and 23% (21-26%) versus 27% (26-28%) three years post-transplant (P<0.001). Cumulative incidences of relapse at three years are 27% (25-30%) for patients aged 70-79 versus 29% (29-30%) (60-69 years) (P=0.71), yet the difference in non-relapse mortality (NRM) (40% [38-43%] vs. 35% [34-36%] in patients aged 70-79 vs. 60-69 years) (P<0.001) translates into survival differences. Median OS of patients surviving >1 year relapse-free is 6.7 (median, 95% CI: 4.5-9.4, 70-79 years) versus 9 (8.4-10.1, 60-69 years) years since landmark. Three years after RFS of one year, excess NRM is Haematologica | 109 February 2024

431

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

14% (95% CI: 12-18%) in patients aged 70-79 versus 12% [11-13%] in patients aged 60-69, while population NRM is 7% (6-7%) versus 3% (3-3%). Mortality for reasons other than relapse, GvHD, or age is as high as 27% (24-29%) and 22% (22-23%) four years after transplantation. In conclusion, survival amongst older patients is adequate after allogeneic stem cell transplantation.

Introduction Cancer is intrinsically tied to age.1,2 Age has long been viewed as one prominent risk factor for survival after cancer.3 Incidences of most diseases treated with allogeneic hematopoietic stem cell transplantation peak beyond the 6th decade of life. In contrast, transplantation had long been restricted to younger individuals since older persons were considered too frail to transplant. With the ongoing demographic changes, this resulted in an offer-demand shortage.4 Consecutively, huge efforts were made to facilitate transplantation in older persons. While in the year 2000 patients undergoing transplantation were rarely ≥60 years, nowadays this intervention is offered to patients up to 80 years of age.5-7 Transplantation in patients ≥70 years results in 2-year-overall survival (OS) of around 40%, with non-relapse mortality (NRM) exceeding 30%.6-12 However, mortality can be considered as a combination of population mortality and excess mortality. Population mortality reflects the baseline mortality that also occurs in the ageand sex-matched general population. Excess mortality is a disease- or treatment-related component.13 Analyses investigating procedures in older patients need to give special consideration to the age-dependent contribution of population mortality to mortality after a particular procedure.8,11,14 Recently, relative survival has been integrated into multistate modeling.13-18 Here, we apply this modeling to the largest set of real-world data reported so far. Data in this analysis span two decades (1998-2018) of treatment in Germany and include 9,422 patients aged 60-69 years and 1,547 patients aged 70-79 years. In this way, we analyze population mortality and excess mortality with and without GvHD and relapse. To our knowledge, we here present the largest analysis of this kind. We report OS, relapse-free survival (RFS), graft-versus-host-disease-free-relapse-free survival (GRFS), non-relapse-mortality (NRM), relapse, and the risk factors thereof for patients in their 7th and 8th decades of life. We conclude that transplantation can efficiently treat diseases in both age groups. Excess mortality is particularly high during the first year after transplantation.

Methods Study population The patient cohort consisted of patients transplanted between 1998-2018 in 55 German centers and reported to the German registry for stem cell transplantation (Deutsches

Register für Stammzelltransplantation). Inclusion criteria were: 1) transplantation of bone marrow or peripheral blood stem cells from a mismatched or matched related or unrelated allogeneic donor for any malignant disease between 1998-2018; 2) age at transplantation ≥60 years; 3) consent to reporting data to DRST. Exclusion criteria were: i) cord blood transplantation; ii) partial/ full duplicates using the reported co-variates UPN, center and patient-ID; iii) missing OS data; and, iv) beyond first allogeneic transplantation. Of 11,031 patients, 62 (0.56%) were excluded. Information was extracted from the DRST database on May 1, 2021. Informed consent had been retrieved prior to reporting. The study was approved by the data access committees of the DRST and by the institutional review board of the Medical Faculty/ University Hospital Tübingen (N. 291/2021BO2) (Online Supplementary Figure S1). Definitions Myeloablative5 and reduced intensity conditioning (RIC) as well as total conditioning intensity (TCI) were defined as published (Online Supplementary Table S1).19-21 OS was defined as time to event (death) after transplantation with patients censored if still alive at last follow-up. Similarly, RFS was defined as the time to either relapse or confirmed progression or death, whichever came first. GRFS was defined as the time to the first of RFS-events or severe GvHD (acute [aGvHD]: grade III or IV; chronic [cGvHD] extensive). Missing data for relapse, aGvHD and cGvHD were treated as “no relapse”, “no aGvHD”, and “no cGvHD”, and were censored at last follow-up regarding these events (Online Supplementary Appendix Chapter 3). Cumulative incidences of relapse / progression and NRM were calculated regarding the two as competing risks. Cumulative incidences of severe aGvHD and cGvHD were analyzed with death as a competing risk. Median follow-up was calculated using the reverse Kaplan-Meier method.22 Multistate models (see below) allowed information to be obtained about patients dying after relapse, after GvHD, or after both, as well as patients alive with these complications at different timepoints during follow-up. Statistical analysis We performed univariable and multivariable analysis for OS, RFS and GRFS using Kaplan-Meier estimations, logrank testing and Cox regression with period-dependent effects. In an environment regarding relapse and death as competing risks, cumulative incidence of relapse and NRM were estimated.23 Likewise, this was repeated for severe GvHD incidence and death without GvHD. To identify risks beyond

Haematologica | 109 February 2024

432

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

the first year we chose a landmarking approach to examine OS for the patients alive at one year after transplantation (OS landmark [LM] population), alive and relapse-free at that timepoint (RFS LM population), and alive without previous relapse or GvHD at that timepoint (GRFS population). We split all mortality into population (the mortality that the patients would have faced in the absence of their disease and treatment) and excess mortality (all additional mortality associated with the disease, and with previous and current treatment) using techniques from relative survival. To apply this method, we had to assume that the population mortality risk of the transplanted patients is equal to that of the German population matched by sex, age, and year of transplant. While this approach still ignores other covariates that are considered important to create appropriate population hazards (socioeconomic status, comorbidities) (see Discussion), this already offers a promising approach to obtain increasingly precise estimates.13 Data for the general population were derived from population tables as available from the Human Mortality Database (accessed Nov 21, 2021).24 We created time-inhomogeneous Markov Multistate models for relapse / progression, GvHD, and death that integrate population mortality to discover differences in excess mortality between age groups, and before and after post-transplantation events. These models allow an in-depth evaluation of complex outcomes. In addition to Kaplan-Meier analysis, this provided information about the probability of reaching different states of death (death after relapse, death after GvHD, non-relapse-death) and about different states of being alive (alive with or after relapse or GvHD). We split all death states into a population and excess part.13 The use of the suffix “.e” denotes excess, while “.p” denotes population contribution. Relevant comorbidities are defined as pre-existing medical conditions that have previously been associated with increased risk of mortality after transplantation. These comorbidities include conditions such as pulmonary, cardiovascular, hepatic, and renal disease, as well as diabetes, malignancies, and infections, and were decided by the treating physician. Excess mortality could be analyzed in an additive model (relsurv package, rsadd), (Online Supplementary Appendix Chapter 1). Because of their reported impact on transplantation outcomes, for all multivariable regression models, we selected the covariates sex, diagnosis, period of transplantation, HLA-match, related / unrelated donors, donor age, stem cell source, conditioning intensity, Karnofsky performance status, known comorbidities prior to transplantation, relationship of cytomegalovirus (CMV) status between donor and patient, and remission status at transplantation. Where applicable, models allowing the effect of baseline covariates to differ between the first year and later years were applied.25 Statistical analysis was carried out in R (v.4.0.5, 2021-03-31, Shake and Throw) using the packages dplyr, survival, survminer, prodlim, mstate, mice,

and relsurv.13,15-17,25-27 All P values are two-sided. P<0.05 is considered significant. To handle missing predictor values in this registry analysis, we chose multiple imputation by chained equations and result-pooling by Rubin’s rules as the main strategy (Online Supplementary Appendix Chapter 3).27 Finally, we included only patients from centers that completed ≥50% of follow-up into outcomes and risk factor analyses to avoid too heavy a reliance on the assumption of non-informative censoring for assessing long-term outcomes (50 centers, N=8,560 aged 60-69, N=1,427 aged 70-79) (Online Supplementary Appendix). When we compared these results to centers with a follow-up of >80%, no difference in outcomes with respect to age as the main covariate of interest were observed. Age was tested via Schoenfeld Residuals in different follow-up approaches (Online Supplementary Figure 7.1, 7.2). Prior to analysis, the study was approved by the Institutional Review Board of the Medical Faculty of Tübingen University (N. 291/2021BO2).

Results Patients’ characteristics We present epidemiological data from 10,969 patients (9,422 aged 60-69 years and 1,547 aged 70-79 years) from 55 German transplantation centers (Table 1). In both age groups, acute leukemia, myelodysplastic syndromes (MDS) and myeloproliferative neoplasms (MPN) are the most frequent indications for transplantation. The percentage of unrelated donors increased to >50% after 2002 and is higher in older patients (60-69 years: 75%; 70-79 years: 86%; P<0.001). Donors of older patients are significantly younger than donors of younger patients (37 years for patients aged 60-69 years vs. 35 years for patients aged 70-79 years; P<0.001). Bone marrow as a graft source has decreased from around 10% to just below 5% in recent years. Higher age is associated with less myeloablative conditioning (MAC) (60-69 years: 37%; 70-79 years: 24%; P<0.001). TCI is significantly lower in older patients. In both age groups, approximately 20% of patient-donor-pairs are CMV IgG seropositive patient and seronegative donor (Table 1). The number of annual transplantations in patients aged 60 years or older rose from a minimum of 13 (1998) to a maximum of 1,140 (2018). The number of patients aged 70 years or older increased from 1 (1999) to 201 (2018) (Figure 1). Changes over time We compared the two decades analyzed in the cohort described. Changes in patients aged 60-69 years and >70 years are significant in parts: median age increased from 63.69 (1998-2008: 61.79-65.81) to 64.37 (2009-2018: 62.24-66.90) in patients aged 60-69 years, and from 71.25 (70.50-73.11) to 72.09 (70.91-73.64) in patients aged 70-79 years (Wilcoxon rank

Haematologica | 109 February 2024

433

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

sum P=0.001). Again, comparing the two periods studied, 1998-2008 and 2009-2018, we observe a decrease in allogeneic stem cell transplantation to treat chronic lymphocytic leukemia (CLL) and chronic myeloid leukemia (CML), likely due to the introduction of novel therapies. We observe an increase in the use of a combination of calcineurin inhibi-

tors (CNI) and mycophenolic acid (MPA), while TCI is lower in patients aged >70 years. Patients with a Karnofsky score <80 are less frequently transplanted while patients with a score of 90-100 undergo transplant more frequently. Furthermore, we find an increase in the number of patients with a complete response (1st or other CR) status before

Table 1. Demographic and clinical characteristics of the patients at transplantation. DRST data Age group 60-69, N=9,422

Characteristic N Patient sex Female Male

9,420

Patient age in years

9,422

Diagnosis Acute leukemia CLL CML Lymphoma MDS/MPN Others Plasma cell disorders

9,422

Donor age in years

4,310

Stem cell source Bone marrow PBSC

9,409

GvHD-prophylaxis CNI+MPA CsA+MTX Other

8,655

TCI-category High Int Low

7,309

Karnofsky Performance Index <80 80 90-100

7,970

Relevant comorbidities

4,897

CMV status relationship Both negative Both positive Negative patient, positive donor Positive patient, negative donor

8,483

Disease status at HSCT - only CR

8,954

1998-2008 N=2,2171

2009-2018 N=7,2051

881 (40) 1,336 (60)

2,839 (39) 4,364 (61)

63.69 (61.79-65.81)

64.37 (62.24-66.90)

1,197 (54) 115 (5.2) 99 (4.5) 141 (6.4) 431 (19) 77 (3.5) 157 (7.1)

3,789 (53) 240 (3.3) 76 (1.1) 646 (9.0) 1,867 (26) 213 (3.0) 374 (5.2)

42 (32-58)

36 (27-49)

119 (5.4) 2,095 (95)

383 (5.3) 6,812 (95)

701 (46) 389 (25) 442 (29)

4,337 (61) 1,878 (26) 908 (13)

41 (3.3) 654 (53) 532 (43)

126 (2.1) 3,489 (57) 2,467 (41)

127 (11) 334 (28) 727 (61)

535 (7.9) 1,814 (27) 4,433 (65)

301 (62)

2,804 (64)

375 (26) 549 (38) 156 (11) 350 (24)

1,862 (26) 3,042 (43) 581 (8.2) 1,568 (22)

644 (31)

2,534 (37)

Age group 70-79, N=1,547 P2

0.8

1,545

<0.001 1,547 <0.001 1,547

<0.001 790 >0.9

1,541

<0.001 1,517

0.002 1,307

0.002 1,415

0.6

980

<0.001 1,494

<0.001 1,489

1998-2008 N=1121

2009-2018 N=1,4351

37 (33) 75 (67)

508 (35) 925 (65)

71.25 (70.50-73.11)

72.09 (70.91-73.64)

80 (71) 0 (0) 2 (1.8) 3 (2.7) 25 (22) 2 (1.8) 0 (0)

934 (65) 37 (2.6) 10 (0.7) 60 (4.2) 343 (24) 37 (2.6) 14 (1.0)

40 (31-65)

34 (27-44)

0 (0) 112 (100)

74 (5.2) 1,355 (95)

46 (49) 10 (11) 37 (40)

948 (67) 266 (19) 210 (15)

1 (1.2) 56 (67) 26 (31)

9 (0.7) 614 (50) 601 (49)

13 (17) 27 (35) 38 (49)

122 (9.1) 410 (31) 805 (60)

35 (78)

645 (69)

25 (27) 28 (30) 12 (13) 28 (30)

385 (27) 615 (44) 104 (7.4) 297 (21)

32 (29)

528 (38)

P2 0.6

0.001 0.3

0.001 0.014

<0.001

0.005

0.041

0.2 0.015

0.040

N (%); median (IQR); 2Pearson’s X2 test; Wilcoxon rank sum test; Fisher’s exact test. Values per decade of transplant. IQR: interquartile range; DRST: German registry for stem cell transplantation; PBSC: peripheral blood stem cell; CsA: cyclosporine A; MPA: mycophenolic acid; CMV: cytomegalovirus; GvHD: graft-versus-host disease; CNI: calcineurin inhibitors; MTX: methotrexate; TCI: total conditioning intensity; CLL: chronic lymphocytic leukemia; CML: chronic myeloid leukemia; MDS/MPN: myelodysplastic syndromes/myeloproliferative neoplasms; CR: complete remission. Full table in the Online Supplementary Appendix. 1

Haematologica | 109 February 2024

434

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

SCT (Table 1, Online Supplementary Table S2). Completed follow-up Follow-up estimated by reverse Kaplan-Meier is 3.9 [95%

CI: 3.8-4] years in the whole cohort.22 After exclusion of 5 centers for a completeness of follow-up <50%, data from 8,560 patients (aged 60-69 years, 91%) and 1,427 patients (aged 70-79 years, 92%) (total 9,987 patients) were included

Figure 1. Epidemiology of reported population. (A) Frequency of age groups among all patients. (B) Frequency of disease among all patients. (C) Frequency of age groups in patients with acute leukemia. (D) Frequency of age groups in patients with myelodysplasia. (E) Frequency of age groups in patients with chronic leukemia. (F) Frequency of age groups in patients with lymphoma. (G) Frequency of age groups in patients with other diseases than those described above (bone marrow failure, solid tumors, hemoglobinopathies, etc.). HSCT: hematopoietic stem cell transplantation. Haematologica | 109 February 2024

435

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

in further analysis. In this analysis cohort, follow-up is 4.2 [95%CI 4.1-4.4] years (Online Supplementary Chapter 2). Survival estimates and cumulative incidences Overall survival, RFS, and GRFS of the entire cohort are 1.6 (median, 95% CI: 1.5-1.7), 1 (95% CI: 0.9-1) and 0.6 (95% CI: 0.5-0.6) years (Figure 2A). Median OS is 1.7 (95% CI: 1.6-1.9) years in patients aged 60-69 years versus 1.1 (95%CI 1-1.3; logrank P<0.001) years in patients aged 70-79 years. Median RFS is 1.0 (95% CI: 0.9-1.1) years versus 0.8 (95% CI: 0.7-0.9 years; P<0.001). Median GRFS is 0.6 (95% CI: 0.5-0.6) years versus 0.5 (95% CI: 0.4-0.5 years; P<0.001). (See Table 2 for detailed survival probabilities.) Overall survival improved in patients treated more recently exclusively in the group 60-69 (median OS, 1998-2008 1.15 [95% CI: 1.03-1.38] years, and 2009-2018 1.94 [95% CI: 1.76-2.13] years; logrank P<0.001). We observe no significant difference in the group 70-79 (median OS for 1998-2008: 1.18 [95% CI: 0.82-3.39] years; for 2009-2018: 1.08 [95% CI: 0.94-1.31] years; logrank P=0.44) (Figure 2E). To examine long-term survivorship, we performed landmark analyses for the 1-year OS, RFS, and GRFS LM populations. Outcome after LM is 7.4 (median, 95% CI: 6.8-8.2) versus 4.8 (95% CI: 4.1-7.1) years regarding the OS-LM population, 9 (95% CI: 8.4-10.1) versus 6.7 (95% CI: 4.5-9.4) years regarding the RFS-LM population, and 10.1 (95% CI: 9.1-11.1) versus 6.7 (95% CI: 4.8-NA) years regarding the GRFS-LM population, when patients aged 60-69 years are compared to those aged 70-79 years, respectively (Figure 2B). Multivariable analysis for 1-year and subsequent survival We performed multivariable analyses in the entire cohort (60-79 years old) including age, diagnosis, remission status, sex, comorbidities, conditioning intensity, use of TBI, Karnofsky performance status, CMV status, donor type (family or unrelated, HLA-match, age), stem cell source (bone marrow or peripheral blood), treatment period (1998-2008, 2009-2018) (Table 3, Online Supplementary Table S5). To estimate different risk factor effects early and later after transplantation, alongside multivariable analysis addressing OS in total, we also performed an analysis split into first and subsequent years of follow-up. Probability of OS is lower in the age group 70-79 years versus 60-69 years, with similar effect sizes during follow-up (overall: Hazard Ratio [HR] 1.19 [95% CI: 1.1-1.28], P<0.001; first year after transplantation: HR 1.18 [1.08-1.29], P<0.001; subsequent years of follow-up: HR 1.22 [1.07-1.4], P=0.004). Transplantation for CLL (HR 0.58 [0.49-0.68], P<0.001), CML (HR 0.73 [0.59-0.9], P=0.003) and MDS/ MPN (HR 0.76 [0.71-0.81], P<0.001) is associated with longer OS. Transplantation in complete remission is protective both in the first year after transplantation and thereafter (HR 0.71 [0.66-0.75], P<0.001). Transplantation from unrelated versus family donors adversely influences OS during

the first year of follow-up but not thereafter, and donor age below median age of all donors (HR 0.86 [0.79-0.93], P<0.001) positively affects OS. Male patient sex correlates with reduced OS in total (HR 1.08 [1.03-1.14], P=0.004) with insignificant differences during the first year but significant impact during subsequent years of follow-up (HR 1.16 [1.05-1.27], P=0.003). Additional adverse factors for OS are the presence of relevant comorbidities as defined by the treating physician (HR 1.16 [1.07-1.25], P<0.001), a low Karnofsky performance status prior to transplantation (<80 vs. 90-100, HR 1.84 [1.68-2.02], P<0.001), and a mismatched HLA-donor (HR 1.33 [1.19-1.49], P<0.001). Low or intermediate TCI do not impact OS significantly, versus high intensity (HR 0.91 [0.75-1.1], P=0.32, HR 0.93 [0.77-1.12], P=0.44). In multivariable analysis, transplantation before 2009 shows poorer survival when compared to transplantation between 2009-2018 (HR 0.9 [0.84-0.97], P=0.009) only during the first year. Table 3 and Online Supplementary Table S5 show the entire regression analysis. Multistate model calculation of survival: quantification of excess mortality Relapse-free survival environment - Recently, the integration of relative survival into multistate modeling offered a new method to estimate excess and population survival with and without intermediate events in an age-dependent fashion (Online Supplementary Figure S2.1).13 We built such an RFS multistate model to quantify the dependence of excess mortality on age (Figure 3, Online Supplementary Figures S5.1-S5.4, S6). The model was set for all patients as well as only for patients with RFS of at least one year (60-69 years: 49.91% [95% CI: 49.32-50.51]; 70-79 years: 45.59% [95% CI: 43.26-48.05%], LM). Among patients aged 70-79 years, 35.44% [95% CI: 33.6137.36%] and 38.14% [95% CI: 35.57-40.91%] died from excess NRM (NRM.e) two and four years after transplantation. Among patients aged 60-69 years, this excess mortality is lower: 31.28% (95% CI: 30.16-32.44%) and 33.81% (95%CI 32.51-35.16%). The proportion of population deaths without relapse (population mortality, NRM.p) steadily increases from 1.42% (95% CI: 1.39-1.45%) to 2.48% (95% CI: 2.412.55%) (60-69 years) and 2.58% (95% CI: 2.47-2.7%) to 4.54% (95% CI: 4.33-4.76%) (70-79 years). One and three years after 1-year LM, NRM.p is 1.18% (95% CI: 1.15-1.21%) and 3.3% (95% CI: 3.21-3.4%) (60-69 years) versus 2.28% (95% CI: 2.18-2.39%) and 6.57% (95% CI: 6.35-6.8%) (70-79 years) (Figure 3A, B). Graft-versus-host-relapse-free survival environment The referenced RFS-multistate model allows the addition of further possible states.13 These, such as living with severe GvHD, can more precisely specify the estimation of non-disease-non-treatment-related death as non-relapse-non-GvHD-mortality (Online Supplementary Figure S2.2).

Haematologica | 109 February 2024

436

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

Figure 2. Survival analysis. (A) Kaplan-Meier estimations of overall survival (OS) and relapse-free survival (RFS) by age group. (B) Kaplan-Meier estimations of OS after having reached 1-year failure-free survival of different definitions: death (OS), death + relapse (RFS), death + relapse + severe graft-versus-host disease (GvHD) (graft-versus-host-relapse-free-survival; GRFS). All patients with respective events before 1-year landmark (LM) were excluded. (C) Kaplan-Meier estimations of age-group-specific OS during 1998-2008 and 20092018. (D) Competing risk analysis of relapse and non-relapse mortality (NRM). (E) Competing risk analysis of severe GvHD and non-GvHD-mortality (NGM). (F) Competing risk analysis of relapse and NRM with regards to low versus intermediate versus high TCI. alloSCT: allogeneic stem cell transplantation; TCI: total conditioning intensity. Haematologica | 109 February 2024

437

Relapse-free survival P<0.001+

Graft-versus-hostrelapse-free survival P<0.001+ NRM P<0.001*

Relapse P=0.71*

Sev. cGvHD P=0.58*

Sev. aGvHD P=0.74*

41.0 (38.4-43.9)

36.4 (33.7-39.3)

32.4 (29.7-35.3)

57.8 (56.7-58.9)

48.1 (47-49.3)

1 year, % (95% CI)

438

2 years, % (95% CI)

3 years, % 42.5 (95% CI) (41.4-43.7)

4 years, % 39.0 (95% CI) (37.8-40.2)

Haematologica | 109 February 2024

32.7% (31.6-33.9)

35.8 (34.7-37.0)

40.7 (39.6-41.8)

49.9 (48.8-51)

1 (0.9-1.1)

28.3 (25.7-31.1)

32.4 (29.8-35.2)

36.8 (34.2-39.5)

45.6 (42.9-48.4)

0.8 (0.7-0.9)

24.2 (23.2-25.3)

26.9 (25.9-27.9)

30.9 [29.8-31.9]

39.4 (38.3-40.4)

0.6 (0.5-0.6)

20.1 (17.9-22.7)

23.3 (21-25.9)

26.7 (24.3-29.3)

34.1 (31.6-36.8)

0.5 (0.4-0.5) -

36.3 (35.2-37.4)

34.8 (33.8-35.9)

26.6 (25.6-27.6)

21.4 (20.5-22.3)

42.7 31.0 (39.8-45.5) (29.9-32.1)

40.4 29.3 (37.7-43.1) (28.3-30.4)

32.7 38.0 (31.7-33.7) (35.4-40.7)

28.7 33.1 (27.7-29.7) (30.6-35.7)

16.2 (14.2-18.3)

14.0 (12.1-16.0)

17.6 (15.5-19.9)

16.0 16.8 (15.1-16.8) (14.8-19.0) 29.1 16.7 (26.5-31.7) (15.9-17.6)

27.2 (24.7-29.7)

25.2 15.0 (22.9-27.6) (14.2-15.8)

21.3 12.7 (19.1-23.5) (12.0-13.5)

NRM: non-relapse mortality; Sev. cGvHD: severe chronic GvHD and Sev. aGvHD: severe acute GvHD. +Logrank. *Gray’s test; CI: confidence interval.

51.9 (49.2-54.7)

Day 100, % (95% CI)

1.1 (1-1.3)

1.7 (1.6-1.9)

Median in years (95% CI)

11.8 (11.1-12.5)

11.3 (9.7-13.0)

Age group Age group Age group Age group Age group Age group Age group Age group Age group Age group Age group Age group Age group Age group 60-69 70-79 60-69 70-79 60-69 70-79 60-69 70-79 60-69 70-79 60-69 70-79 60-69 70-79

Overall survival P<0.001+

Table 2. Detailed outcome data.

ARTICLE - Transplantation in patients aged 60-79 years J.F. Weller et al.

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

To refine the analysis of NRM as non-relapse-non-GvHD-mortality (NRNGM), multistate analysis was repeated in a GRFS-environment (1-year-GRFS, 60-69 years: 39.37% [95% CI: 38.41-40.34%]; 70-79 years: 34.08% [95% CI: 32.27-36%]). At two and four years, NRNGM.e is 21% (95% CI: 20.3621.66%], 22.22% (95% CI: 21.61-22.84%] and NRNGM.p 1.17% (95% CI: 1.13-1.2%), 1.97% (95% CI: 1.9-2.04%) for patients aged 60-69 years versus 25.02% (95% CI: 23.13-27.07%), 26.58% (95% CI: 24.42-28.93%) and 2.06% (95% CI: 1.982.15%), 3.48% (95% CI: 3.28-3.69%) for patients aged 7079 years, respectively. In patients GvHD- and relapse-free after one year, 4.84% (95% CI: 4.25-5.51%) and 7.93% (95% CI: 7.26-8.66%) (60-69 years) versus 6.08% (95% CI: 5.396.86%) and 10.64% (95% CI: 7.27-15.59%) (70-79 years) still suffer NRNGM.e at one and three years after LM. NRNGM.p increases to 1.15% (95% CI: 1.14-1.17%] and 3.2% (95% CI:

3.13-3.27%] (60-69) versus 2.23% (95% CI: 2.12-2.34%] and 6.38% (95% CI: 5.95-6.85%] (70-79 years) (Figure 3C, D). Full data and state proportions can be found in the Online Supplementary Appendix and Online Supplementary Table S3.1.-S3.4.

Risk factors for excess mortality The integration of relative survival into regression models can enable the impact of risk factors on excess mortality to be estimated. (See Online Supplementary Chapter 1 for additive modeling.) Age influences excess mortality during the entire, and especially the first year, of follow-up (HR 1.14 [1.05-1.24], P=0.001, 70-79 years vs. 60-69 years, first year: HR 1.16 [1.061.27], P=0.001). However, significance is lost after 1-year of OS (HR 1.09 [0.91-1.3], P=0.35) despite similar effect size.

Figure 3. Multistate modeling of relapse-free survival and graft-versus-host-relapse-free-survival. (A) Multistate progress of follow-up after one year without failure (death / relapse) in patients aged 70-79 years. (B) Multistate progress of follow-up after one year without failure (death / relapse) in patients aged 60-69 years. (C) Multistate progress of follow-up after one year without failure (death / relapse / severe graft-versus-host disease [GvHD]) in patients aged 70-79 years. (D) Multistate progress of follow-up after one year without failure (death / relapse / severe GvHD) in patients aged 60-69 years. NRM: non-relapse-mortality; DAR: death after relapse; NRNGM: non-relapse-non-GvHD-mortality; DaGvHD+R: death after GvHD and relapse; DaGvHD: death after GvHD; R: Relapse; p: due to population mortality (NRM and NRNGM in bold); e: due to excess mortality. Haematologica | 109 February 2024

439

Haematologica | 109 February 2024

440

1 1.33 (1.23-1.44), P<0.001 2.04 (1.82-2.3), P<0.001 -

1 1.3 (1.21-1.39), P<0.001 1.95 (1.77-2.15), P<0.001 -

1 1.17 (1.04-1.32), P=0.008 1.53 (1.27-1.85), P<0.001 P<0.001

1 1.31 (1.22-1.42), P<0.001 1.96 (1.75-2.18), P<0.001 P<0.001

1 0.66 (0.61-0.72), P<0.001 P<0.001

1 1.27 (1.19-1.35), P<0.001 1.84 (1.68-2.02), P<0.001 P<0.001

1 0.95 (0.89-1.01), P=0.09

1 0.71 (0.66-0.75), P<0.001 P<0.001

Karnofsky Perf. Index, ref. 90-100 80 <80 P value after Cox Anova

Period, ref. 1998-2008 2009-2018

Remission state at HSCT, ref. No CR Complete remission

1 0.9 (0.84-0.97), P=0.009

P<0.001

1 0.82 (0.73-0.92), P<0.001

1 1.04 (0.93-1.16), P=0.51

1 0.67 (0.63-0.72), P<0.001

1 0.92 (0.85-0.99), P=0.021

1 0.65 (0.6-0.7), P<0.001

1 0.9 (0.83-0.97), P=0.008

1 0.76 (0.66-0.88), P<0.001

1 1 (0.86-1.15), P=0.96

1 1.2 (1.01-1.42), P=0.035 1.56 (1.24-1.97), P<0.001 -

1 1.05 (0.89-1.24), P=0.54

1 1.11 (0.98-1.25), P=0.1

1 0.62 (0.44-0.87), P=0.006 0.54 (0.31-0.94), P=0.028 0.79 (0.61-1.04), P=0.09 0.76 (0.65-0.9), P=0.001 1.07 (0.84-1.37), P=0.59 1.18 (0.87-1.61), P=0.3 -

1 1.09 (0.91-1.3), P=0.35

First year of FU* HR (95% CI)

*See Online Supplementary Appendix for model: added total conditioning intensity (TCI), type of allograft (peripheral blood [PB] vs. bone marrow [BM]), donor age, cytomegalovirus (CMV) status relation, donor type, total body irradiation (TBI) part of conditioning and follow-up (FU) hazards; Online Supplementary Table S5). Values obtained from a multivariable Cox regression model are depicted overall and separately for the first year and subsequent years of (FU). Likewise, a Cox model was fitted for excess hazard only. An overall P value of a likelihood ratio test was added. CLL: chronic lymphocytic leukemia; CML: chronic myeloid leukemia; MDS/MPN: myelodysplastic syndromes/myeloproliferative neoplasms; CR: complete remission; HSCT: hematopoietic stem cell transplantation; HR: hazard ratio; CI: confidence interval; NA: not available; ref.: reference.

Likelihood Ratio Test

1 1.2 (1.1-1.31), P<0.001

1 1.16 (1.07-1.26), P<0.001

1 1.12 (0.99-1.26), P=0.07

1 1.18 (1.07-1.29), P<0.001

1 1.16 (1.07-1.25), P<0.001

1 1.04 (0.97-1.11), P=0.23

Comorbidity ref. No Yes

1 1.06 (1-1.12), P=0.06

1 1.08 (1.03-1.14), P=0.004

Sex ref. female Male

1 1.16 (1.05-1.27), P=0.003

1 0.5 (0.41-0.62), P<0.001 0.77 (0.59-1), P=0.05 1.19 (1.06-1.34), P=0.004 0.73 (0.67-0.8), P<0.001 0.8 (0.69-0.94), P=0.006 1.03 (0.86-1.22), P=0.77 -

1 0.53 (0.44-0.63), P<0.001 0.72 (0.56-0.91), P=0.006 1.1 (0.99-1.23), P=0.09 0.74 (0.68-0.8), P<0.001 0.87 (0.76-0.99), P=0.036 1.06 (0.91-1.23), P=0.47 -

1 0.68 (0.53-0.88), P=0.003 0.65 (0.45-0.94), P=0.021 0.83 (0.68-1.02), P=0.08 0.78 (0.69-0.89), P<0.001 1.29 (1.07-1.55), P=0.007 1.15 (0.89-1.49), P=0.28 P<0.001

1 0.53 (0.43-0.65), P<0.001 0.78 (0.6-1), P=0.048 1.19 (1.06-1.33), P=0.004 0.75 (0.69-0.81), P<0.001 0.81 (0.7-0.94), P=0.007 1.02 (0.86-1.22), P=0.78 P<0.001

1 0.58 (0.49-0.68), P<0.001 0.73 (0.59-0.9), P=0.003 1.08 (0.97-1.19), P=0.16 0.76 (0.71-0.81), P<0.001 0.97 (0.86-1.09), P=0.56 1.06 (0.92-1.22), P=0.42 P<0.001

Disease, ref. acute leukemia CLL CML Lymphoma MDS/MPN Plasma cell disorders Others P value after Cox Anova 1 1.05 (0.98-1.12), P=0.17

1 1.16 (1.06-1.27), P=0.001

First year of FU* HR (95% CI)

1 1.14 (1.05-1.24), P=0.001

Total FU* HR (95% CI)

1 1.22 (1.07-1.4), P=0.004

Years 1+ of FU* HR (95% CI)

1 1.18 (1.08-1.29), P<0.001

First year of FU* HR (95% CI)

Additive proportional hazards model regarding excess mortality+

1 1.19 (1.1-1.28), P<0.001

Total FU* HR (95% CI)

Cox regression regarding overall survival

Age group ref. 60-69 70-79

Co-variate

Table 3. Regression analyses.

ARTICLE - Transplantation in patients aged 60-79 years J.F. Weller et al.

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

Interestingly, the negative impact of male sex is lost when only excess mortality (HR 1.06 [1-1.12], P=0.06) is considered. Table 3 shows the regression analysis.

Discussion This study on 1,547 patients aged 70-79 years and 9,422 patients aged 60-69 years, transplanted during two decades (1998-2018), is the largest real-world data analysis investigating the influence of age on survival after transplantation for myeloid neoplasia, but also chronic leukemia, lymphoma, MDS/ MPN, plasma cell disorders, and other types of neoplasia. Notably, no patient over the age of 79 years is documented in the database. This real-world dataset is a picture of the heterogeneous landscape (e.g., disease, conditioning, GvHD-prophylaxis) of stem cell transplantation. We preserve this heterogeneity on purpose.6 The design of the study with a long inclusion period enabled us to capture time trends in numbers of transplanted older patients and outcomes. Absolute hazards for excess mortality are known to increase with age.14,28 Here we aim to dissect mortality in older cohorts into the naturally occurring, age-dynamic, population-based mortality and excess (disease- and treatment-related) mortality in order to better define the impact of age as risk factor for transplantation. In general, our study confirms older age as a risk factor for poor survival in patients receiving transplantation. OS of patients aged 60-69 years exceeds that of their older peers (70-79 years) by 7.5 months (Table 2).6-8,28 Death rates in older patients (70-79 years) are markedly increased during the first year of follow-up. Importantly, overall survivors after 1-year are expected to survive a median of 7.4 years (60-69 years) and 4.8 years (70-79 years), respectively, consistent with other studies.6,7 The contribution of population mortality to NRM has been shown in long-term survivors of transplantation for MDS.14 Furthermore, over the age of 55, the risk of death roughly doubles in every age decade within the general population, which would give a hazard ratio of 2.0 for age decade (Online Supplementary Figure S3).29 To separate the contribution of age, GvHD and relapse to post-transplantation mortality, we performed multistate analysis in an RFS- and GRFS-environment for all patients (Online Supplementary Figure S5.1-S5.4) and exclusively for 1-year event-free survivors (Figure 3). Two main messages stand out clearly. First, NRM and NRNGM seem to gradually outweigh “death after GvHD” and “death after relapse” in older patients. Strikingly, up to 25% of older patients will have died for reasons other than GvHD, relapse or age up to four years after transplantation (Online Supplementary Table S3.3, Online Supplementary Figure S4.1, S4.2). However, in one-year survivors followed up to four years after transplantation (i.e., LM plus three years), population mortality contributes roughly from a fifth

to a third to NRM: NRM.p divided by NRM.p+.e for 70-79 years equals 31.4%, for 60-69 years 21.6%. The quotient of NRNGM.p and NRNGM.p+.e for the group 70-79 years is 37.3%, and for 60-69 years 29.2%. Thus, the contribution of population mortality is more pronounced with higher age. Hence, differences in excess survival are less pronounced (Table 3), especially at longer follow-up. Thus, we add the most accurate estimation of the increased hazard for excess mortality for higher age and this is revealed to be, indeed, moderate (HR 1.2-1.3 per decade).8,14,28 As a first conclusion, whereas excess mortality drives mortality the year after transplantation, the contribution of population mortality increases with distance to transplantation but excess mortality remains dominant. Defining and addressing the remaining causes of death should be a matter of further research. It remains speculative as to why OS of patients aged 60-69 years improves over time whereas it does not change for patients aged 70-79 years. Patient selection is one possible explanation. Furthermore, our comprehensive analysis confirms previous results on the impact of donor characteristics (related vs. unrelated, young donor age30-34), disease (CLL, CML, MDS, MPN), disease state at transplant (complete remission or not) on outcome. One reason for the dominance of the male sex within the dataset is probably the tendency towards curative treatment options and against palliative regimen for men.35 Regression analysis suggests that worse outcome in male patients is largely due to higher population mortality. There is no significant difference in excess mortality between male and female patients during all follow-up periods. Sex differences, both as to whether frequencies and outcome are concerned, must remain a matter of future research. Limitations of our analysis are the degree of missing data any registry analysis must face. It remains debatable if an adequate follow-up might have differed to any great extent. We are sure that a maximum error of 5% must be anticipated for registry analysis; however, this should be tackled within the registries. With regards to our regression analysis, it needs to be mentioned that MICE would not generally accept missing not-at-random values. This is an issue of the model assumption that is described in the literature.36 In addition, our data span two decades, contributing to the heterogeneity in this work. To account for medical progress during that time, we considered outcomes from patients transplanted between 2006-2018 in a separate analysis (Online Supplementary Tables S5, S6, S7, Online Supplementary Figure S8.1, 8.2). It is important to be aware that relative survival analysis models rely on the assumption that the patients’ state of health before transplantation does not differ from that of the general population. This assumption might not be met in the context of a treatment-based registry; it is likely that patients who have severe comorbidities or a disadvantaged socioeconomic

Haematologica | 109 February 2024

441

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years

background will not be offered transplantation. This would imply that the background mortality is somewhat smaller in our patient cohort than in the general population and the excess mortality thus larger than estimated. This would not change our main messages though. Next to an improved completeness of follow-up, future registry documentation might boost coverage of frailty and comorbidity indices. We, therefore, further encourage the community to focus on raw data for variables like the HCT-CI comborbidity index and the Disease Risk Index. In conclusion, transplantation is particularly challenging for older adults due to comorbidities, impaired functional status, and cognitive impairment. These factors may negatively affect recovery after transplantation and increase the risk of complications. Therefore, thorough screening and careful planning are critical for older adults considering transplantation. Practical clinical implications on whom to choose and how to prepare for transplantation cannot be concluded from this registry analysis and have been reviewed elsewhere.37-40 Although it is evident that older patients generally have worse outcomes, the difference between patients aged 70 to 79 years compared with patients aged 60 to 69 years is not so high as to justify exclusion from transplantation on the basis of age alone.

Disclosures HN is an employee of the German Registry for Stem Cell Transplantation (DRST). SF is, in part, financed through the German Registry for Stem Cell Transplantation (DRST). In an honorary capacity, NK, KF, PD, JS and MS are on the board of the German Registry for Stem Cell Transplantation (DRST). The other authors have no conflicts of interest to disclose. Contributions JFW and MC are responsible for the conception of the analysis, data retrieval, statistical analysis and interpretation, and writing of the manuscript. LdW is responsible for the discussion of results, statistical analysis, and writing of the manuscript. CL contributed to the discussion of results. JF, JS, UP, HE, TS, CF, MS, PD, IWB, GW, JT, CS, AE, MB, WB, KF and NK are responsible for the contribution of patient data and discussion of results. HN and SF curated primary data. All authors read the final version of the manuscript and agreed to its content. Data-sharing statement The code used for the analyses is available upon request from the corresponding author. Data of the individual participants will not be shared.

References 1. Li J, Smith A, Crouch S, Oliver S, Roman E. Estimating the prevalence of hematological malignancies and precursor conditions using data from Haematological Malignancy Research Network (HMRN). Cancer Causes Control. 2016;27(8):1019-1026. 2. White MC, Holman DM, Boehm JE, Peipins LA, Grossman M, Henley SJ. Age and cancer risk: a potentially modifiable relationship. Am J Prev Med. 2014;46(3 Suppl 1):S7-15. 3. Bassily MN, Wilson R, Pompei F, Burmistrov D. Cancer survival as a function of age at diagnosis: a study of the Surveillance, Epidemiology and End Results database. Cancer Epidemiol. 2010;34(6):667-681. 4. Ferrucci L, Giallauria F, Guralnik JM. Epidemiology of aging. Radiol Clin North Am. 2008;46(4):643-652. 5. Camara Rdl, Alonso A, Steegmann JL, et al. Allogeneic hematopoietic stem cell transplantation in patients 50 years of age and older. Haematologica. 2002;87(9):965-972. 6. Muffly L, Pasquini MC, Martens M, et al. Increasing use of allogeneic hematopoietic cell transplantation in patients aged 70 years and older in the United States. Blood. 2017;130(9):1156-1164. 7. Duarte RF, Sánchez-Ortega I. HSCT in elderly patients. In: Carreras E, Dufour C, Mohty M, Kröger N, editors. The EBMT Handbook: Hematopoietic Stem Cell Transplantation and Cellular Therapies. Cham (CH): Springer; 2019. p. 499-503. 8. Robin M, De Wreede LC, Wolschke C, et al. Long-term outcome after allogeneic hematopoietic cell transplantation for myelofibrosis. Haematologica. 2019;104(9):1782. 9. Kyriakou C, Boumendil A, Finel H, et al. The impact of advanced patient age on mortality after allogeneic hematopoietic cell transplantation for non-Hodgkin lymphoma: a retrospective study by the European Society for Blood and Marrow Transplantation Lymphoma Working Party. Biol Blood Marrow

Transplant. 2019;25(1):86-93. 10. Heidenreich S, Ziagkos D, de Wreede LC, et al. Allogeneic stem cell transplantation for patients age ≥ 70 years with myelodysplastic syndrome: a retrospective study of the MDS Subcommittee of the Chronic Malignancies Working Party of the EBMT. Biol Blood Marrow Transplant. 2017;23(1):44-52. 11. van Gelder M, de Wreede LC, Bornhauser M, et al. Long-term survival of patients with CLL after allogeneic transplantation: a report from the European Society for Blood and Marrow Transplantation. Bone Marrow Transplant. 2017;52(3):372-380. 12. Ringden O, Boumendil A, Labopin M, et al. Outcome of allogeneic hematopoietic stem cell transplantation in patients age >69 years with acute myelogenous leukemia: on behalf of the Acute Leukemia Working Party of the European Society for Blood and Marrow Transplantation. Biol Blood Marrow Transplant. 2019;25(10):1975-1983. 13. Manevski D, Putter H, Pohar Perme M, Bonneville EF, Schetelig J, de Wreede LC. Integrating relative survival in multi-state models - a non-parametric approach. Stat Methods Med Res. 2022;31(6):997-1012. 14. Schetelig J, de Wreede LC, van Gelder M, et al. Late treatmentrelated mortality versus competing causes of death after allogeneic transplantation for myelodysplastic syndromes and secondary acute myeloid leukemia. Leukemia. 2019;33(3):686-695. 15. de Wreede LC, Fiocco M, Putter H. The mstate package for estimation and prediction in non- and semi-parametric multistate and competing risks models. Comput Methods Programs Biomed. 2010;99(3):261-274. 16. Pohar M, Stare J. Relative survival analysis in R. Comput Methods Programs Biomed. 2006;81(3):272-278. 17. Pohar Perme M, Estève J, Rachet B. Analysing population-based

Haematologica | 109 February 2024

442

J.F. Weller et al.

ARTICLE - Transplantation in patients aged 60-79 years cancer survival - settling the controversies. BMC Cancer. 2016;16(1):933. 18. Robin M, de Wreede LC, Padron E, et al. Role of allogeneic transplantation in chronic myelomonocytic leukemia: an international collaborative analysis. Blood. 2022;140(12):1408-1418. 19. Spyridonidis A, Labopin M, Savani BN, et al. Redefining and measuring transplant conditioning intensity in current era: a study in acute myeloid leukemia patients. Bone Marrow Transplant. 2020;55(6):1114-1125. 20. Spyridonidis A, Labopin M, Savani BN, et al. Correction: Redefining and measuring transplant conditioning intensity in current era: a study in acute myeloid leukemia patients. Bone Marrow Transplant. 2020;55(6):1213. 21. Bacigalupo A, Ballen K, Rizzo D, et al. Defining the intensity of conditioning regimens: working definitions. Biol Blood Marrow Transplant. 2009;15(12):1628-1633. 22. Schemper M, Smith TL. A note on quantifying follow-up in studies of failure time. Control Clin Trials. 1996;17(4):343-346. 23. Gray B, Gray MB. Package ‘cmprsk’. Subdistribution analysis of competing risks R package version. Ann Stat. 2014;2:2-7. 24. Wilmoth JR, Andreev K, Jdanov D, et al. Methods protocol for the human mortality database. University of California, Berkeley, and Max Planck Institute for Demographic Research, Rostock. http://mortalityorg Accessed October 26, 2021. 25. Therneau TM, Lumley T. Package ‘survival’. R Top Doc. 2015;128(10):28-33. 26. Therneau T, Crowson C, Atkinson E. Multi-state models and competing risks. CRAN-R. https://cran Accessed October 26, 2021. 27. Van Buuren S, Groothuis-Oudshoorn K. Mice: multivariate imputation by chained equations in R. J Stat Softw. 2011;45(3):1-67. 28. McClune BL, Weisdorf DJ, Pedersen TL, et al. Effect of age on outcome of reduced-intensity hematopoietic cell transplantation for older patients with acute myeloid leukemia in first complete remission or with myelodysplastic syndrome. J Clin Oncol. 2010;28(11):1878-1887. 29. Partridge L. Evolutionary biology and age-related mortality. In: Wachter KW, Finch CE, Population NRCCo, editors. Between Zeus and the Salmon: The Biodemography of Longevity: National Academies Press (US); 1997. 30. Mehta J, Gordon LI, Tallman MS, et al. Does younger donor age

affect the outcome of reduced-intensity allogeneic hematopoietic stem cell transplantation for hematologic malignancies beneficially? Bone Marrow Transplant. 2006;38(2):95-100. 31. DeZern AE, Franklin C, Tsai HL, et al. Relationship of donor age and relationship to outcomes of haploidentical transplantation with posttransplant cyclophosphamide. Blood Adv. 2021;5(5):1360-1368. 32. Wang Y, Wu DP, Liu QF, et al. Donor and recipient age, gender and ABO incompatibility regardless of donor source: validated criteria for donor selection for haematopoietic transplants. Leukemia. 2018;32(2):492-498. 33. Kollman C, Spellman SR, Zhang MJ, et al. The effect of donor characteristics on survival after unrelated donor transplantation for hematologic malignancy. Blood. 2016;127(2):260-267. 34. Kollman C, Howe CW, Anasetti C, et al. Donor characteristics as risk factors in recipients after transplantation of bone marrow from unrelated donors: the effect of donor age. Blood. 2001;98(7):2043-2051. 35. Saeed F, Hoerger M, Norton SA, Guancial E, Epstein RM, Duberstein PR. Preference for palliative care in cancer patients: are men and women alike? J Pain Symptom Manage. 2018;56(1):1-6. 36. Hammon A. Multiple imputation of ordinal missing not at random data. AStA Adv Stat Anal. 2022 Dec 8. doi: 10.1007/ s10182-022-00461-9 [preprint, not peer-reviewed] 37. Rosko AE, Cordoba R, Abel G, Artz A, Loh KP, Klepin HD. Advances in management for older adults with hematologic malignancies. J Clin Oncol. 2021;39(19):2102-2114. 38. Lin RJ, Artz AS. Allogeneic hematopoietic cell transplantation for older patients. Hematology Am Soc Hematol Educ Program. 2021;2021(1):254-263. 39. Mishra A, Preussler JM, Bhatt VR, et al. Breaking the age barrier: physicians’ perceptions of candidacy for allogeneic hematopoietic cell transplantation in older adults. Transplant Cell Ther. 2021;27(7):617. 40. Olin RL, Fretham C, Pasquini MC, et al. Geriatric assessment in older alloHCT recipients: association of functional and cognitive impairment with outcomes. Blood Adv. 2020;4(12):2810-2820.

Haematologica | 109 February 2024

443

ARTICLE - Cell Therapy & Immunotherapy

Regulatory T-cell dysfunctions are associated with increase in tumor necrosis factor α in autoimmune hemolytic anemia and participate in Th17 polarization Marion Ciudad,1,2 Sethi Ouandji,1,2 Baptiste Lamarthée,2 Claudie Cladière,1,2 Thibault Ghesquière,1,2 Martin Nivet,1,2 Marine Thébault,1,2 Romain Boidot,3 Agnès Soudry-Faure,4 Sandy Chevrier,3 Corentin Richard,3 Thibault Maillet,5 François Maurier,6 Hélène Greigert,1,2 Coraline Genet,2 André Ramon,2 Malika Trad,2 Valérie Predan,1 Philippe Saas,2 Maxime Samson,1,2 Bernard Bonnotte 1

1,2

and Sylvain Audia

1,2

Department of Internal Medicine and Clinical Immunology, Referral Center for Adult

Autoimmune Cytopenia (CeReCAI) - Dijon University Hospital, Dijon; Université de 2

Bourgogne, INSERM, UMR1098, RIGHT, Dijon; 3Unit of Molecular Biology, Georges-François

Correspondence: S. Audia sylvain.audia@u-bourgogne.fr Received: Accepted: Early view:

February 3, 2023. July 25, 2023. August 3, 2023.

Leclerc Cancer Center, Dijon; 4Department of Clinical Research and Innovation (DRCI), Clinical Research Unit-Methodological Support Network (USMR), Dijon Bourgogne University Hospital, Dijon; 5Department of Internal Medicine - Centre Hospitalier de Mâcon, Groupe Hospitalier Bourgogne Méridionale, Macon and 6Department of Internal Medicine, Groupe Hospitalier UNEOS, Metz, France

Abstract Warm autoimmune hemolytic anemia (wAIHA) is a rare acquired autoimmune disease mediated by antibodies targeting red blood cells. The involvement of CD4 T-helper cells has been scarcely explored, with most findings extrapolated from animal models. Here, we performed quantification of both effector T lymphocytes (Teff) and regulatory T cells (Treg), associated with functional and transcriptomic analyses of Treg in human wAIHA. We observed a shift of Teff toward a Th17 polarization concordant with an increase in serum interleukin-17 concentration that correlates with red blood cell destruction parameters, namely lactate dehydrogenase and bilirubin levels. A decrease in circulating Treg, notably effector Treg, associated with a functional deficiency, as represented by their decrease capability to inhibit Teff proliferation, were also observed. Treg deficiency was associated with a reduced expression of Foxp3, the master transcription factor known to maintain the Treg phenotype stability and suppressive functions. Transcriptomic profiling of Treg revealed activation of the tumor necrosis facto (TNF)-α pathway, which was linked to increased serum TNF-α concentrations that were twice as high as in controls. Treg transcriptomic profiling also suggested that post-translational mechanisms possibly accounted for Foxp3 downregulation and Treg dysfunctions. Since TNF-α participates in the rupture of immune tolerance during wAIHA, its inhibition could be of interest. To this end, the effects of fostamatinib, a SYK inhibitor, were investigated in vitro, and we showed that besides the inhibition of erythrocyte phagocytosis by monocytes, fostamatinib is also able to dampen TNF-α production, thus appearing as a promising multitargeting therapy in wAIHA (clinicaltrials gov. Identifier: NCT02158195).

Introduction Warm autoimmune hemolytic anemia (wAIHA) is a rare acquired autoimmune disease due to antibodies targeting red blood cell (RBC) antigens.1-3 Anti-RBC antibodies are mostly immunoglobulin (Ig)G, underlying a class-switch recombination, a mechanism involving cooperation between B and T cells. Until now, the involvement of helper CD4 T cells (Th) has been scarcely studied in human wAIHA, and most of the conclusions have been derived

from animal models. In mice, a Th1 polarization has been first observed,4 with an improvement of the disease by Th2 cytokines such as interleukin (IL)-4.5 On the opposite, human wAIHA was first thought to be associated with a Th2 skewing as suggested by an increased production of IL-4 while interferon (IFN)-γ was reduced.6 However, when CD4 T cells from patients were stimulated with specific RBC antigens, IFN-γ secretion was enhanced whereas IL-4 was not detected, arguing for a Th1 polarization.7 This was secondly amended by the fact that IL-17 and Th17 were

Haematologica | 109 February 2024

444

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

increased and associated with wAIHA severity.8,9 A defect in regulatory T cells (Treg), together with a pro-inflammatory T-cell response, are commonly observed in autoimmune diseases.10 During immune thrombocytopenia (ITP), the most prevalent autoimmune cytopenia, such abnormalities have been observed,11-13 with a negative correlation between Th17 and Treg frequencies.13 To date, the results remain controversial in wAIHA. In animals, whereas the role of Treg has been shown in a xenogeneic model in which Treg depletion enhanced the occurrence of wAIHA while their adoptive transfer prevented its establishment,14 it was not confirmed in a non-xenogeneic model.15 In humans, only one study reported on a decrease in circulating Treg, but without assessing their function.16 With the aim to better understand the pathophysiology of human wAIHA, and thus opening the way to new therapies, a better comprehensive appraisal of its global T-cell immune response is required. Here we provide the first concurrent evaluation of circulating effector T lymphocytes (Teff) and Treg, associated with Treg function and transcriptomic analyses.

Methods Patients Healthy controls (HC) and patients with newly diagnosed wAIHA were proposed to participate. The diagnosis of wAIHA was defined as hemoglobin <11 g/dL, with a low haptoglobin level and a positive direct antiglobulin test (DAT) for IgG +/- complement (C3d). The diagnosis of primary wAIHA was retained after exclusion of lymphoproliferative malignancies, other autoimmune diseases, primary immunodeficiency and infections, as recommended.17,18 All patients were included before receiving immunomodulating drugs, especially steroids. Study approval All patients gave a written informed consent in accordance with the Declaration of Helsinki before participating to this prospective study (clinicaltrials gov. Identifier: NCT02158195). The research was approved by the Institutional Review Board and the Independent Ethics Committee (Comité de Protection des Personnes, CPP Est 1). Cell isolation, culture and storage A more detailed methods is provided in the Online Supplementary Appendix. Whole blood from patients or HC was collected for peripheral blood mononuclear cells (PBMC) isolation to perform immunophenotyping, cell culture and storage. T-cell proliferation suppression assay Teff (CD4+CD25-) and Treg (CD4+CD25hi) were purified by magnetic cell isolation. Labeled Teff were activated with

anti-CD2/CD3/CD28 microbeads and cultured for 4 days with or without Treg (Teff/Treg ratio=2/1). Proliferation was measured by flow cytometry based on CellTrace dilution. Assessment of monocyte function during hemolysis Monocytes were isolated based on CD14 expression and cultured for 24 hours either alone or stimulated with heme or RBC coated with either anti-glycophorin A or in the presence of isotype control. RBC collected from one patient with relapsing wAIHA were also used. The effect of R406, the active metabolite of fostamatinib was also investigated. Flow cytometry was used to quantify TNF-α production and RBC phagocytosis. RNA sequencing, gene ontology terms, Treg signature and index computations Fluorescent-activated cell sorting (FACS) of Treg (FVS780-CD3+CD4+CD25hiCD127-) from four primary wAIHA and four sex- and age-matched HC were used for RNA sequencing (RNAseq). RNAseq was processed in one batch. Only protein-coding transcripts and genes were included in the downstream analysis. Genes with a P value below 0.05 and normalized read count greater than 25 were considered as significantly differentially expressed. RNAseq data are available in the Gene Expression Omnibus database (accession number: GSE195791). Treg transcriptomic identity was assessed by referring to the Treg signature genes.19 Tumor necrosis factor (TNF) index was calculated for each subject by averaging the normalized expression of all genes belonging to TNF pathway (normalized expression vs. mean expression of all HC) similarly to published data on Treg signature.19,20 Cytokine assays IFN-γ, IL-4, IL-17 and TNF-α were quantified in culture supernatants using multiplex immunoassay. High sensitivity enzyme-linked immunosorbant assay was performed for serum IL-17A. Statistics Statistical analyses and graphs were performed with Prism v9.3.0 (GraphPad Software), R Studio (1.4.1717) and Heatmaper. Data are reported as the median with interquartile range (IQR) and were compared using Mann-Whitney test for independent data, Wilcoxon signed-rank test for paired conditions, and Spearman’s rank correlation test for correlation analyses, unless otherwise specified. P value <0.05 was considered significant.

Results Patients Twenty-two patients with wAIHA were enrolled at diagnosis, before initiation of any immunomodulatory drugs

Haematologica | 109 February 2024

445

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA Table 1. Baseline characteristics of warm autoimmune hemolytic anemia patients and healthy controls. HC N=30

wAIHA N=22

67 (58-77)

65 (53-76)

0.95

Sex ratio, female/male

18/12

15/7

0.63

Primary AIHA, N (%)

17 (77)

14. (13.2- 15.2)

7.4 (6.4-9.5)

<0.0001

Reticulocytes, x109/L, median (IQR)

189 (139-324)

Leukocytes, x109/L, median (IQR)

5.8 (4.9-7.0)

5.8 (4.6- 8.5)

0.97

Lymphocytes, x109/L, median (IQR)

1.6 (1.2-2.1)

1.3 (0.7-2.5)

0.31

Platelets, x109/L, median (IQR)

224 (204-296)

191 (162-314)

0.47

Haptoglobin, g/L, median (IQR)

0 (0.0-0.2)

Direct antiglobulin test, N (%) IgG alone IgG and C3d

22 (100) 15 (68) 7 (32)

Characteristics Age in years, median (IQR)

Hemoglobin, g/dL, median (IQR)

wAIHA: warm autoimmune hemolytic anemia; HC: healthy controls; IQR: interquartile range; NA: not applicable.

(Table 1). Their median age was 65 (IQR, 53-76) years with a female/male ratio of 2.1. Secondary wAIHA was diagnosed in five patients (22,7%), associated with lupus (n=1), ITP (n=1) or B-cell lymphoproliferation (B-cell monoclonal lymphocytosis, n=2; indolent marginal zone B cell, n=1). As none of the measured parameters were different between primary and secondary AIHA, the data set was kept as a single group (Online Supplementary Table S3). Patients were compared to 30 HC, with a median age of 67 years (IQR, 58-77) (P=0.9) and a female/male ratio of 1.5 (P=0.6). Hemoglobin was lower in wAIHA patients (7.4 vs. 14.2 g/dL; P<0.0001) while total leukocyte and lymphocyte counts were similar. The flowchart of the study is depicted in Figure 1A. Quantitative and functional alterations of circulating Treg in warm autoimmune hemolytic anemia The frequency of circulating Treg (CD3+CD4+CD25hiFoxp3+) was lower in wAIHA (3.2% vs. 4.5% of CD4 T cells; P=0.01; Figure 1B). Moreover, the quantification of Treg subtypes21 showed that CD4+CD45RA-Foxp3hi effector Treg (Fr.II/eTreg), known to have the strongest suppressive capabilities, were twice less frequent in wAIHA (1.1% vs. 2.1%; P=0.0008; Figure 1C), while the proportions of naive Treg (CD45RA+Foxp3lo, Fr.I/nTreg) were similar. The frequency of CD25hiFoxp3+Helios+ cells among total CD4 T cells was also lower in wAIHA patients (2.4 vs. 3.1%; P=0.007; Figure 1D). As lymphocyte count and CD4 T-cell proportions were similar between patients and controls, the decrease in Treg subsets was also observed when considering cell absolute numbers (Online Supplementary Table S4). Associated with this quantitative reduction, the suppressive function of Treg was also altered in wAIHA, as shown by a decreased ability to inhibit Teff proliferation (51% vs. 73%; P=0.003; Figure 1E).

Warm autoimmune hemolytic anemia is associated with a Th17 polarization Effector T-cell subpopulations were similar between HC and patients (Online Supplementary Figure S1; Online Supplementary Table S4). We then focused on Teff polarization determined by their cytokine production. Measurement of cytokines in culture supernatants of Teff activated for 4 days showed a higher production of IL-17 in wAIHA (25.5 pg/mL vs. 7.5 pg/mL; P=0.02; Figure 2A), whereas IFN-γ concentration was similar and IL-4 was not detected. Supporting this Th17 polarization, we observed an increase in serum IL-17 concentrations (0.41 pg/mL vs. 0.68 pg/mL; P=0.02; Figure 2B) and an imbalance in the Th17/Treg ratio (0.22 vs. 0.11; P=0.02; Figure 2C). Interestingly, the serum concentration of IL-17 positively correlated with markers of RBC destruction, namely lactate dehydrogenase (LDH) and bilirubin (R=0.58, P=0.03 and R=0.68, P=0.02, respectively; Figure 2D), but not with reticulocyte count, a marker of RBC production, nor with hemoglobin level, resulting from the production and destruction RBC. Overall, these results reflected an imbalance between proand anti-inflammatory immune responses in wAIHA, with a quantitative and functional alteration of circulating Treg, associated with Th17 polarization, the latter being positively correlated with disease activity, notably markers of RBC destruction such as LDH and bilirubin. Transcriptomic profiling showed Treg activation during warm autoimmune hemolytic anemia In order to further characterize the mechanisms involved in Treg dysfunctions, a transcriptomic analysis was performed on sorted CD3+CD4+CD25hiCD127lo Treg obtained from four representative patients (Online Supplementary Table S2) and four HC. The isolated cells were first confirmed to be

Haematologica | 109 February 2024

446

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Continued on following page.

Haematologica | 109 February 2024

447

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Figure 1. Quantitative and functional alterations of circulating Treg in warm autoimmune hemolytic anemia. (A) Flowchart of the overall experimental approach. (B) Dot plot showing the gating strategy for the quantification of regulatory T cells (Treg), defined as CD4+CD25hiFoxp3+ lymphocytes, by flow cytometry (left panel). Scatter dot plot (right panel) showing Treg frequencies among circulating CD4 T cells in healthy controls (HC) (N=24) and warm autoimmune hemolytic anemia (wAIHA) patients (N=20). (C) Dot plot (left panel) showing the gating strategy for the determination of naive Treg (CD45RA+Foxp3low: Fr. I/nTreg) and effector Treg (CD45RA-Foxp3hi: Fr. II/eTreg) by fluorescense-activated cell sorting (FACS) (HC N=20; wAIHA N=19). Scatter dot plots (right panel) showing the frequency of Treg subsets among circulating CD4 T cells. (D) Scatter dot plot showing the frequency of Treg expressing Helios among circulating CD4 T cells. (E) Effector T-cell proliferation assay. Representative dot plots (left panel) showing the proliferation of effector T cells (Teff) stimulated or not with anti-CD2/CD3/CD28 microbeads, in presence or not of Treg (Teff/Treg ratio: 2/1), assessed by CellTrace dilution by flow cytometry after 4 days of culture, for one representative HC and one representative wAIHA patient. Proliferation index (PI) is mentioned. Histogram with scatter dot plot (right panel) showing Treg function as determined by the inhibition of Teff proliferation for 9 HC and 9 wAIHA patients. The PI of stimulated Teff alone is used as reference. P values derived from Mann-Whitney test. Median with 1st and 3rd quartiles are depicted on graphs. PBMC: peripheral blood mononucelar cells; FACS: fluorescence-activated cell sorting; RNAseq: RNA sequencing; NS: not significant.

Treg as shown by the specific upregulation of 194 genes and downregulation of 192 genes (Figure 3A) consistent with Treg signature.19 The transcriptomic profile of Treg from patients was distinct from the one of HC, as revealed by the differential expression of 455 genes, some of which being implicated in transcription or translation processes or reflecting the activation of the T-cell receptor (TCR) and TNF pathways (Figure 3B). The activation of the transcription and translational processes (Figure 3B) were revealed by increased transcription of RNA splicing proteins (DHX38, U2AF1, PNN), transcription elongation or termination factors (SUPT6H, XRN2), ribosomal proteins (RPL, RPS) and translation initiation factor (EIF3B). The engagement of the TCR pathway in wAIHA Treg (Figure 3C) was supported by the overexpression of 16 genes such as CD247 (TCR ζ-chain), ZAP70 (ζ-chain associated protein 70), CD4, and Janus kinase 3 (JAK3). Upon activation, multiple mechanisms are involved in the suppressive functions of Treg.22,23 Although our functional analysis showed an impairment of Treg function, a slight but significant increase in the transcripts of CTLA-4 (CTLA4), CD25 (IL2RA), TIGIT (TIGIT), and GARP (LRRC32), the protein binding latent TGF-β at cell surface, was observed, while others were expressed at similar rates. The expression of GPA33 a marker that allows the identification of a subset of naive Treg of thymic origin, expressing Foxp3 and Helios, with immunosuppressive functions and not secreting pro-inflammatory cytokines such as IL-1724 was not different between patients and controls (Figure 3D). The RORC (ROR-γt), CCR6 and IL17A transcripts were compared between patients and controls and found similar (Figure 3E), arguing against the hypothesis of a loss of inhibitory functions of Treg due to their differentiation into IL-17 producing cells in a Th17 environment.23,25 α signaling pathway is engaged in Treg and is TNF-α correlated with the decreased Foxp3 expression Transcriptomic analyses of Treg showed a strong engagement of TNF-α signaling pathway in wAIHA, as supported by the overexpression of ten genes such as MADD (MAP kinase-activating death domain), CASP8 (caspase 8), TAB2 and TRAF1 (P<0.0001; Figure 4A). In order to better under-

stand these results, TNF-α was measured in sera and found to be almost twice as high in wAIHA patients (5.7 pg/mL vs. 3.0 pg/mL; P<0.0001; Figure 4B). The expression of the transcripts of the two TNF receptors were investigated in Treg. TNFR2 transcripts were more abundant than the ones of TNFR1, both in HC and patients (Table 2), and were higher in patients (average expression: 138 vs. 103; P=0.02; Table 2). The activation of TNFR2 can promote the inhibitory functions of Treg through NF-κB pathway, with the transcription of NFLB2 and RELB.26 These transcripts were increased by 1.4-fold in wAIHA Treg (P=0.04 and P=0.07, respectively; Table 2). However, repressive effects have also been reported after TNFR2 ligation,27 involving the NF-κB28 pathway or molecules such as DBC1 (deleted breast cancer 1)29 and PP1 (protein phosphatase 1).30 Their transcripts were also increased in wAIHA patients (Table 2), thus arguing for an engagement of both activating and inhibiting pathways in Treg during wAIHA. In order to assess the effect of TNF-α stimulation on Treg, the level of TNF engagement was quantified using a TNF index, as previously described for Treg signature.19,20 This TNF index inversely correlated with the frequency of circulating Treg (R=-0.89; P=0.01; Figure 4C), notably with the one of eTreg (R=-0.79; P=0.05; Figure 4C), thus supporting an inhibitory effect of TNF-α on Treg. Considering the major role of Foxp3 in the maintenance of Treg phenotype and functions31 and that TNF-α could participate to its downregulation,32 the expression of Foxp3 protein was measured and found to be lower in wAIHA patient Treg (mean fluorescence intensity [MFI]: 662 vs. 1,149; P=0.012; Figure 4D). Consistently, the TNF index inversely correlated with Foxp3 protein level (R=0.89; P=0.01; Figure 4E), although FOXP3 transcripts were increased (Figure 4F). Of note, neither Treg frequencies nor TNF-α concentrations correlated with disease activity markers such as hemoglobin level, reticulocyte count, bilirubin and LDH concentrations (Online Supplementary Figure S2). As Treg functions are strengthened by the interaction of Foxp3 with Helios,33,34 its expression was also determined and found to be decreased in wAIHA (MFI: 291 vs. 1,346; P=0.002). In contrast to Foxp3, the reduction of Helios protein expression was not correlated with the TNF index (Online Supplementary Figure S3).

Haematologica | 109 February 2024

448

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Figure 2. Warm autoimmune hemolytic anemia is associated with a Th17 polarization. (A) Scatter dot plots of interferon (IFN)-γ, interleukin (IL)-4 and IL-17 concentrations measured in supernatants of 4-day culture of CD4 T cells stimulated by anti-CD2/CD3/ CD28 microbeads (healthy controls [HC] N=9; warm autoimmune hemolytic anemia [wAIHA] N=9; upper panel). Line plots with histograms showing the concentration of IL-17 measured in culture supernatants of effector T cells (Teff) activated or not with microbeads, with or without regulatory T cells (Treg) (Teff/Treg ratio:2/1; lower panel). (B) Scatter dot plots of serum IL-17A concentrations measured in 20 wAIHA patients and 26 HC. (C) Scatter dot plots showing the balance between pro- and anti-inflammatory immune responses as represented by Th1/Treg, Th2/Treg and Th17/Treg ratios (HC N=23; wAIHA N=20). P values derived from Mann-Whitney test, Wilcoxon signed-rank as appropriate. (D) Correlation between serum IL-17 concentration and red blood cell (RBC) destruction markers (lactate dehydrogenase [LDH] and bilirubin), hemoglobin and reticulocyte count. Spearman’s rank correlation coefficient (R) and P value are depicted. Line represents linear regression. ND: not detected; NS: not significant. Haematologica | 109 February 2024

449

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Post-translational mechanisms may downregulate Foxp3 expression in Treg The maintenance of Foxp3 protein expression is crucial to ensure a stable pool of functional Treg and depends on multiple mechanisms such as transcriptional, translational and post-translational regulation.35 The fact that FOXP3 transcripts were increased during wAIHA while Foxp3 protein was diminished suggested translational or post-translational regulatory dysfunctions rather than transcriptional alterations. We thus investigated post-translational mechanisms known to regulate Foxp3 protein expression or function. We observed a higher transcription of genes coding for Pim-2, a protein involved in Foxp3 phosphorylation, for HDAC7, responsible for Foxp3 deacetylation, but also an increase in DBC1 and caspase 8, previously reported to degrade Foxp3 in an inflammatory environment (Online Supplementary Table S5).35 Suggesting a role for TNF-α in these post-translational regulatory mechanisms of Foxp3, TNF index positively correlated with these different transcripts, while negatively associated with Foxp3 protein expression and circulating Treg frequency (Online Supplementary Figure S4).

In vitro, the Syk inhibitor fostamatinib decreased both the phagocytosis of red blood cells and the production of TNF-α α by monocytes TNF-α has been targeted in clinical practice for years, notably in rheumatoid arthritis, in which anti-TNF monoclonal antibodies were shown to restore Treg functions.36 Considering that during wAIHA monocytes and macrophages that are activated after phagocytosis of RBC represent an important source of TNF-α,37,38 what we confirmed in vitro (Online Supplementary Figure S5), and that the spleen tyrosine kinase SYK participates to the downstream signaling of the Fc portion of immunoglobulin G receptor (FcγR), we investigated the effects of fostamatinib, on monocyte functions in vitro, as it showed promising results in wAIHA.39 As SYK also participates to the production of TNF-α by monocytes,40,41 its inhibition could have a dual interest in wAIHA. As expected, R406, the active metabolite of fostamatinib, dramatically decreased phagocytosis of RBC coated with anti-GPA antibody (mean of 58.8% vs. 5.6% of monocytes; P=0.03; Figure 5A). Interestingly, the production of TNF-α induced by heme and either by anti-GPA-coated RBC or RBC obtained from a patient with active wAIHA were profoundly

Continued on following page.

Haematologica | 109 February 2024

450

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Figure 3. Transcriptomic profiling of Treg showing T-cell receptor activation and the engagement of transcriptional and translational processes. (A) Volcano plot (fold change vs. P value) displaying the transcriptomes of regulatory T cells (Treg) from patients with warm autoimmune hemolytic anemia (wAIHA) compared to healthy controls (HC). Genes that are part of the Treg signature are highlighted in red and blue when their expression is increased or decreased in wAIHA, respectively. P values derived from X2 test. (B) Heat map showing Treg genes that are differentially expressed between HC (N=4) and wAIHA patients (N=4). The relative gene expression (ζ-score) is represented by color gradient (decreased expression in blue and increased in red). Four of the key terms predicted by gene ontology analysis are annotated along with the main genes implicated. (C) T-cell receptor (TCR) signaling pathway evaluated by gene ontology analysis. P value is shown (upper panel) with the heat map of differentially expressed genes (lower panel). (D) Radar chart showing the average expression of transcripts of Treg effector molecules (HC N=4; wAIHA N=4). (E) Heat map showing the relative expression of transcripts used for the identification of ROR-γ+ Treg between HC (N=4) and wAIHA patients (N=4). The relative gene expression (ζ-score) is depicted by color gradient of blue (downregulated) and red (upregulated). *P<0.05.

Table 2. Expression of genes coding for TNFR pathway effectors involved in Treg function. Effectors

Genes

TNFR1

Average expression HC

wAIHA

Fold change

TNFRSF1A

0.78

0.117

TNFR2

TNFRSF1B

103

138

1.34

0.023

NF-kB

NFKB2

1.34

0.041

RelB

RELB

1.47

0.074

DBC1

CCAR2

1.33

0.030

PP1

PPP1CA

1.44

0.032

TNFR: tumor necrosis factor receptor; Treg: regulatory T cell; wAIHA: warm autoimmune hemolytic anemia; HC: healthy controls. Haematologica | 109 February 2024

451

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Figure 4. TNF-α α signaling engagement correlates with Treg dysfunctions. (A) Tumor necrosis factor (TNF)-mediated signaling pathway was evaluated by gene ontology analysis. P value is shown (upper panel) with heat map (lower panel) showing the relative expression of genes (ζ-score) between healthy controls (HC) (N=4) and warm autoimmune hemolytic anemia (wAIHA) patients (N=4), depicted by color gradient of blue (downregulated) and red (upregulated). (B) Scatter dot plots of TNF-α concentration measured in sera from HC (N=26) and wAIHA patients (N=20). P values derived from Mann-Whitney test. (C) The activation of the TNF pathway is represented as TNF index, calculated by averaging the normalized expression of TNF-associated genes differentially expressed in regulatory T cells (Treg). Correlation of the TNF index with Treg frequency (upper panel) and Fr.II/effector Treg (eTreg) frequency (lower panel) among circulating CD4 T cells from HC (grey diamonds, N=3) and wAIHA patients (red triangles, N=4). P and R values derived from Spearman correlation analysis. (D) Nuclear Foxp3 protein expression assessed by the mean fluorescence intensity (MFI) measured by flow cytometry in circulating Treg from HC (N=24) and wAIHA patients (N=15). Representative histogram of nuclear expression (left panel). Scatter dot plots of Foxp3 MFI among circulating Treg (right panel). (E) Correlation between TNF index and Foxp3 MFI in circulating Treg (HC N=3; wAIHA N=4). P and R values derived from Spearman correlation analysis. (F) Box-and-whiskers plots of normalized expression of FOXP3 transcripts in Treg (HC N=4; wAIHA N=4). P values derived from Mann-Whitney test. NS: not significant. Haematologica | 109 February 2024

452

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

inhibited (mean of 60.3 vs. 3.1-fold change compared to unstimulated monocytes; P=0.03; 90 vs. 19.7-fold change; P=0.005; and 35.6 vs. 3.3-fold changes; P=0.01 respectively; Figure 5B).

Discussion Conversely to ITP, the most frequent autoimmune cytopenia, in which a decrease in Treg has been clearly demonstrated,11,12 the literature is scarce in human wAIHA. We here confirmed that the frequency of circulating Treg in human wAIHA was reduced,16 and specified that eTreg, the subset with the most intense inhibitory activity,21 was mostly affected as represented by its two-times lower frequency. We also provide the first functional evaluation of Treg in wAIHA and observed altered functions characterized by a reduced inhibition of Teff proliferation and by the promotion of IL-17 secretion. A dysfunction of Treg has been implicated in the occurrence and the maintenance of multiple autoimmune diseases,10,13 although the underlying mechanisms are still puzzling. In wAIHA, we observed that Treg dysfunctions were associated with a decreased expression of the major transcription factor Foxp3, that is critical for Treg survival and the maintenance of their inhibitory functions,31,42 but also to prevent their conversion toward Teff.43 The lower level of the transcription factor Helios, which associates with Foxp3 and enhances Treg functions, might also participate in these Treg dysfunctions. In order to clarify the processes involved in the diminution of Foxp3 during wAIHA, we used transcriptomic analysis to assess the mechanisms that have been described so far as key regulators of Foxp3,35 and observed higher transcription of genes involved in Foxp3 phosphorylation or deacetylation, such as Pim2 and HDAC7. The transcripts of DBC1 and caspase 8, previously reported to degrade Foxp3 in an inflammatory environment were also increased.29 In order to formally demonstrate their involvement in the downregulation of Foxp3 during wAIHA, a specific assessment of Foxp3 phosphorylation and acetylation would be of interest. Until recently, TNF-α was not known to be increased in wAIHA and had even been found diminished.44 However, when measured in the active phase of the disease and prior to the initiation of any treatment, as done in our study and in a recent report,45 the concentration of TNF-α is twice as high than in controls. In vitro, we observed that in conditions mimicking wAIHA, monocytes were the main cells producing TNF-α. As previous studies showed that TNF-α could alter Treg functions28,36,46,47 and downmodulate Foxp3 expression,32 we investigated the potential link between TNF-α and Foxp3. Indeed, the more the TNF-α signaling pathway was engaged in Treg, the lower the number of circulating Treg was and the more the level of Foxp3 protein expression was reduced, highly suggesting a pivotal role of TNF-α in Treg dysregulation in wAIHA. However, these

results appeared controversial, as it is increasingly recognized that TNF-α, by its binding to TNFR2, has a positive effect on Treg, as represented by the increased expression of CD25 and Foxp3, their expansion and the fostering of their inhibitory properties.27,48 However, all these conclusions were drawn from studies using Treg from healthy subjects. Moreover, the presence of other cytokines such as IL-2 were required to allow TFN-α to increase Foxp3 expression, the generation of Treg and to foster their inhibiting functions.48 Conversely, a study performed on Treg obtained under inflammatory conditions from synovial fluid during juvenile arthritis showed an increase in TNFR2 expression and a decrease in Treg immunosuppressive functions, similarly21 to our results. In the same way, in the presence of TNF-α, a reduction of suppressive functions from Treg obtained from rheumatoid arthritis patients was also observed.46 On the other hand, TNF blockade restores Treg suppressive functions28 in rheumatoid arthritis36,46 and enables the expansion of Treg in ITP.47 In wAIHA, although we observed an overexpression of TNFRII transcripts and an activation of the NF-κB pathway, as reflected by the increase in RELB and NFKB2 transcripts, in accordance with previous reports,26,28 our functional assay clearly demonstrated a Treg dysfunction. Thus, while TNF-α promotes the functions of Treg from healthy donors in vitro,48 during wAIHA, there is both a functional and quantitative deficit of Treg. Moreover, this deficit is associated with a decrease in Foxp3 expression which could result from post-translational regulation mechanisms that correlate with TNF-α pathway engagement. Notably, our results are supported by a previous study showing that the regulation by DBC1, a protein binding Foxp3 and leading to its degradation in a caspase 8-dependent mechanism can be initiated in an inflammatory environment, i.e., in the presence of IL-6 or TNF-α.29 In a mouse model of AIHA, Treg deficiency was not sufficient for disease initiation.15 In humans, it is impossible to determine whether Treg dysfunction precedes the onset of the disease and promotes it, or whether this deficit merely sustains the disease. However, our data suggest that during wAIHA, the Treg deficit at least perpetuates the autoimmune response, as suggested by the trend to an increased production of IL-17 when Teff are cocultured with Treg. Moreover, hemolysis by itself could indirectly maintain Treg deficiency, as supported by the negative correlation between Foxp3 expression or Treg frequency and TNF-index, while the production of TNF-α by monocytes is increased in vitro in the presence of either IgG-opsonized RBC or heme. As the release of heme from RBC can activate Toll-like receptor (TLR)-4 and induces TNF-α secretion by macrophages through a mechanism involving the adaptor protein SYK,40,41 we investigated the effect of the SYK inhibitor fostamatinib, on monocytes in conditions mimicking wAIHA. The primary mechanism of action of fostamatinib is the inhibition of FcγR signaling pathway and thus phagocytosis of autoantibody-recognized RBC,39 what we confirmed in vitro. We

Haematologica | 109 February 2024

453

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Continued on following page.

Haematologica | 109 February 2024

454

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Figure 5. Fostamatinib reduced red blood cell phagocytosis and TNF-α α production by monocytes in vitro. Monocytes from 7 healthy controls were cultured 24 hours alone or in conditions mimicking hemolysis, i.e., in the presence of heme or red blood cells (RBC) labeled with CellTrace and either coated with immunoglobulin (Ig)G anti-glycophorin A (RBC/monocytes ratio: 5/1) or in the presence of isotype control (RBC-Ig). RBC obtained from a patient with active warm autoimmune hemolytic anemia (wAIHA) were also used (wAIHA RBC). R406 (2 µM), the active metabolite of fostamatinib was added in the different conditions. In order to exclude RBC attached to monocyte membrane during phagocytosis assay, RBC lysis was performed before flow cytometry. (A) Histograms showing the mean percentage of RBC phagocytosis by monocytes with standard of the mean. (B) Histograms representing the mean fold changes of tumor necrosis factor (TNF)-α production in hemolysis conditions, with or without R406 (lower panel, N=7). The production of TNF-α by unstimulated monocytes was used as reference. P values derived from paired t test. AGPA: anti-glycophorin A antibody.

also observed a profound diminution of TNF-α production by monocytes in the presence of fostamatinib. Whether neutrophils, that are stimulated by TNF-α and are involved in RBC phagocytosis during wAIHA,38 also participate in the increase in TNF-α remains to be determined. However, previous publications have reported that neutrophils produce only little amount of TNF-α in humans.49,50 Finally, fostamatinib, with a response rate observed in nearly half of wAIHA patients in a phase I/II clinical trial, may play a broader action than previously thought, by acting on multiple pathways of wAIHA pathogenesis. These multiple effects could confer to fostamatinib both a short-term effect by reducing RBC phagocytosis and possibly the production of anti-RBC antibodies,39,51 but also a long-term effect, by restoring the immune tolerance mediated by Treg, which could allow its discontinuation overtime. The interest of our study also relies on the concomitant evaluation of the anti-inflammatory and the pro-inflammatory responses, as an imbalance is frequent in autoimmune diseases. T-cell polarization in wAIHA was not formally established in humans and was successively related to Th2 polarization,6 then Th17 and finally Th17.8,9 Our results firmly support the latter as evidenced by the increase in IL-17 concentration in serum and the imbalance of the Th17/Treg ratio, while Th1/Treg and Th2/Treg ratios were unaffected. However, although blood samples were taken before any treatment was started, serum IL-17 concentrations or its production by Teff was highly variable from one patient to another, resulting in overlapping results with controls. Thus, the involvement of pathophysiological mechanisms differing from one patient to another, as is the case in ITP notably,52 cannot be ruled out. However, the positive correlation between IL-17 concentration and hemolysis parameters such as LDH and bilirubin strongly supports a link between RBC destruction and Th17 polarization. There was also a trend to an increase in IL-17 secretion in co-culture of Teff and Treg from wAIHA conversely to healthy donors, suggesting that in addition to the loss of their inhibitory functions, Treg could promote Th17 polarization. Taken altogether, our results show that in wAIHA, Treg harbor both quantitative and functional dysfunctions that cannot counteract the pro-inflammatory Th17 polarization of Teff, with a production of IL-17 that correlates with the intensity of RBC destruction. These Treg defects are associated with a downmodulation of Foxp3 expression

that could be driven by post-translational mechanisms such as deacetylation and phosphorylation of Foxp3, and also involved DBC1 and caspase 8. The activation of these mechanisms positively correlates with the engagement of the TNF-α pathway in Treg, whose serum concentration is increased. Targeting TNF-α could be a novel approach complementary to current treatments to restore immune tolerance in wAIHA. Along this line, fostamatinib could be a promising treatment as it reduces both the production of TNF-α and the phagocytosis of RBC by monocytes. Disclosures SA served on advisory committee boards for Novartis and SOBI; received consultancy fees from Amgen, Argenx and Novartis; received lecture fee/congress support from Amgen, Grifols and Novartis; and received a research grant from Novartis. BB served on advisory committee boards for Novartis; received consultancy fees from Amgen and Novartis; received lecture fees/congress support from Amgen and Novartis. MS received consultancy fees from Abvvie, Boehringer Ingelheim, Novartis, Roche Chugai and Vifor. All other authors have no conflicts of interest to disclose. Contributions SA, BB, MC and ASF designed the study. SA, PS and BB funded the research. SA, BB, TM, VP, MS and FM recruited patients and provided clinical information. SA and MC designed the experiments. MC, SO, CC, TG, MN, MT, HG, CG, AR and MT performed experiments. RB, SC, and CR performed RNA sequencing. MC, BL and SA analyzed data. MC, BL and SA drew the figures. SA and MC wrote the manuscript. SA, MC, MS, BL, PS and BB edited the paper. Acknowledgments We thank Y. Duffourd and Dr. H. Begue for their valuable advice for RNA-sequencing analysis. We thank Dr. A. Legrand, N. Pernet and S. Monier, from the flow cytometry core facility (INSERM UMR1231, Université de Bourgogne Franche-Comté), for their technical support for flow cytometry and cell sorting. Funding This investigation was supported by a research grant (2012A001154-39) from the GIRCI Est (Groupement Interrégional pour la Recherche Clinique et l’Innovation), by the French Ministry of Health and Solidarity (Referral Centers for Rare

Haematologica | 109 February 2024

455

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA

Diseases), and by the French National Institute of Health and Medical Research (INSERM). The flow cytometry core facility is supported by the Burgundy Regional Council.

Data-sharing statement RNA-sequencing data are available in the Gene Expression Omnibus database (accession number GSE195791).

References 1. Michel M. Warm autoimmune hemolytic anemia: advances in pathophysiology and treatment. Presse Med. 2014;43(4 Pt 2):e97-e104. 2. Barcellini W, Zaninoni A, Giannotta JA, Fattizzo B. New insights in autoimmune hemolytic anemia: from pathogenesis to therapy stage 1. J Clin Med. 2020;9(12):3859. 3. Mahévas M, Michel M, Vingert B, et al. Emergence of long-lived autoreactive plasma cells in the spleen of primary warm autoimmune hemolytic anemia patients treated with rituximab. J Autoimmun. 2015;62:22-30. 4. Shen CR, Mazza G, Perry FE, et al. T-helper 1 dominated responses to erythrocyte Band 3 in NZB mice. Immunology. 1996;89(2):195-199. 5. Youssef AR, Shen CR, Lin CL, Barker RN, Elson CJ. IL-4 and IL-10 modulate autoimmune haemolytic anaemia in NZB mice. Clin Exp Immunol. 2005;139(1):84-89. 6. Fagiolo E, Toriani-Terenzi C. Th1 and Th2 cytokine modulation by IL-10/IL-12 imbalance in autoimmune haemolytic anaemia (AIHA). Autoimmunity. 2002;35(1):39-44. 7. Hall AM, Ward FJ, Vickers MA, Stott LM, Urbaniak SJ, Barker RN. Interleukin-10-mediated regulatory T-cell responses to epitopes on a human red blood cell autoantigen. Blood. 2002;100(13):4529-4536. 8. Hall AM, Zamzami OM, Whibley N, et al. Production of the effector cytokine interleukin-17, rather than interferon-gamma, is more strongly associated with autoimmune hemolytic anemia. Haematologica. 2012;97(10):1494-1500. 9. Xu L, Zhang T, Liu Z, Li Q, Xu Z, Ren T. Critical role of Th17 cells in development of autoimmune hemolytic anemia. Exp Hematol. 2012;40(12):994-1004. 10. Grant CR, Liberal R, Mieli-Vergani G, Vergani D, Longhi MS. Regulatory T-cells in autoimmune diseases: challenges, controversies and - yet - unanswered questions. Autoimmun Rev. 2015;14(2):105-116. 11. Stasi R, Cooper N, Del Poeta G, et al. Analysis of regulatory T-cell changes in patients with idiopathic thrombocytopenic purpura receiving B cell-depleting therapy with rituximab. Blood. 2008;112(4):1147-1150. 12. Yu J, Heck S, Patel V, et al. Defective circulating CD25 regulatory T cells in patients with chronic immune thrombocytopenic purpura. Blood. 2008;112(4):1325-1328. 13. Zufferey A, Kapur R, Semple JW. Pathogenesis and therapeutic mechanisms in immune thrombocytopenia (ITP). J Clin Med. 2017;6(2):16. 14. Mqadmi A, Zheng X, Yazdanbakhsh K. CD4+CD25+ regulatory T cells control induction of autoimmune hemolytic anemia. Blood. 2005;105(9):3746-3748. 15. Richards AL, Kapp LM, Wang X, Howie HL, Hudson KE. Regulatory T cells are dispensable for tolerance to RBC antigens. Front Immunol. 2016;7:348. 16. Ahmad E, Elgohary T, Ibrahim H. Naturally occurring regulatory T cells and interleukins 10 and 12 in the pathogenesis of idiopathic warm autoimmune hemolytic anemia. J Investig Allergol Clin Immunol. 2011;21(4):297-304.

17. Barcellini W, Fattizzo B. The changing landscape of autoimmune hemolytic anemia. Front Immunol. 2020;11:946. 18. Jager U, Barcellini W, Broome CM, et al. Diagnosis and treatment of autoimmune hemolytic anemia in adults: recommendations from the First International Consensus Meeting. Blood Rev. 2019;41:100648. 19. Ferraro A, D’Alise AM, Raj T, et al. Interindividual variation in human T regulatory cells. Proc Natl Acad Sci U S A. 2014;111(12):E1111-1120. 20. Galvan-Pena S, Leon J, Chowdhary K, et al. Profound Treg perturbations correlate with COVID-19 severity. Proc Natl Acad Sci U S A. 2021;118(37):e2111315118. 21. Miyara M, Yoshioka Y, Kitoh A, et al. Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor. Immunity. 2009;30(6):899-911. 22. Grover P, Goel PN, Greene MI. Regulatory T cells: regulation of identity and function. Front Immunol. 2021;12:750542. 23. Sakaguchi S, Mikami N, Wing JB, Tanaka A, Ichiyama K, Ohkura N. Regulatory T cells and human disease. Annu Rev Immunol. 2020;38:541-566. 24. Opstelten R, de Kivit S, Slot MC, et al. GPA33: a marker to identify stable human regulatory T cells. J Immunol. 2020;204(12):3139-3148. 25. Komatsu N, Okamoto K, Sawa S, et al. Pathogenic conversion of Foxp3+ T cells into TH17 cells in autoimmune arthritis. Nat Med. 2014;20(1):62-68. 26. Wang J, Ferreira R, Lu W, et al. TNFR2 ligation in human T regulatory cells enhances IL2-induced cell proliferation through the non-canonical NF-κB pathway. Sci Rep. 2018;8(1):12079. 27. Salomon BL. Insights into the biology and therapeutic implications of TNF and regulatory T cells. Nat Rev Rheumatol. 2021;17(8):487-504. 28. Nagar M, Jacob-Hirsch J, Vernitsky H, et al. TNF activates a NF-kappaB-regulated cellular program in human CD45RAregulatory T cells that modulates their suppressive function. J Immunol. 2010;184(7):3570-3581. 29. Gao Y, Tang J, Chen W, et al. Inflammation negatively regulates FOXP3 and regulatory T-cell function via DBC1. Proc Natl Acad Sci U S A. 2015;112(25):E3246-3254. 30. Nie H, Zheng Y, Li R, et al. Phosphorylation of FOXP3 controls regulatory T cell function and is inhibited by TNF-α in rheumatoid arthritis. Nat Med. 2013;19(3):322-328. 31. Williams LM, Rudensky AY. Maintenance of the Foxp3dependent developmental program in mature regulatory T cells requires continued expression of Foxp3. Nat Immunol. 2007;8(3):277-284. 32. Valencia X, Stephens G, Goldbach-Mansky R, Wilson M, Shevach EM, Lipsky PE. TNF downmodulates the function of human CD4+CD25hi T-regulatory cells. Blood. 2006;108(1):253-261. 33. Seng A, Krausz KL, Pei D, et al. Coexpression of FOXP3 and a Helios isoform enhances the effectiveness of human engineered regulatory T cells. Blood Adv. 2020;4(7):1325-1339. 34. Takatori H, Kawashima H, Matsuki A, et al. Helios enhances Treg cell function in cooperation with FoxP3. Arthritis Rheumatol.

Haematologica | 109 February 2024

456

M. Ciudad et al.

ARTICLE - T-cell dysregulation in warm AIHA 2015;67(6):1491-1502. 35. Colamatteo A, Carbone F, Bruzzaniti S, et al. Molecular mechanisms controlling Foxp3 expression in health and autoimmunity: from epigenetic to post-translational regulation. Front Immunol. 2019;10:3136. 36. Ehrenstein MR, Evans JG, Singh A, et al. Compromised function of regulatory T cells in rheumatoid arthritis and reversal by anti-TNFalpha therapy. J Exp Med. 2004;200(3):277-285. 37. Gallagher MT, Branch DR, Mison A, Petz LD. Evaluation of reticuloendothelial function in autoimmune hemolytic anemia using an in vitro assay of monocyte-macrophage interaction with erythrocytes. Exp Hematol. 1983;11(1):82-89. 38. Meinderts SM, Oldenborg PA, Beuger BM, et al. Human and murine splenic neutrophils are potent phagocytes of IgGopsonized red blood cells. Blood Adv. 2017;1(14):875-886. 39. Kuter DJ, Rogers KA, Boxer MA, et al. Fostamatinib for the treatment of warm antibody autoimmune hemolytic anemia: phase 2, multicenter, open-label study. Am J Hematol. 2022;97(6):691-699. 40. Fortes GB, Alves LS, de Oliveira R, et al. Heme induces programmed necrosis on macrophages through autocrine TNF and ROS production. Blood. 2012;119(10):2368-2375. 41. Prestes EB, Alves LS, Rodrigues DAS, et al. Mitochondrial reactive oxygen species participate in signaling triggered by heme in macrophages and upon hemolysis. J Immunol. 2020;205(10):2795-2805. 42. Zhou X, Bailey-Bucktrout S, Jeker LT, Bluestone JA. Plasticity of CD4(+) FoxP3(+) T cells. Curr Opin Immunol. 2009;21(3):281-285. 43. Bailey-Bucktrout SL, Martinez-Llordella M, Zhou X, et al. Selfantigen-driven activation induces instability of regulatory T cells during an inflammatory autoimmune response. Immunity. 2013;39(5):949-962.

44. Barcellini W, Zaja F, Zaninoni A, et al. Low-dose rituximab in adult patients with idiopathic autoimmune hemolytic anemia: clinical efficacy and biologic studies. Blood. 2012;119(16):3691-3697. 45. Branch DR, Leger RM, Sakac D, et al. Chemokines IP-10/CXCL10 and IL-8/CXCL8 are potential novel biomarkers of warm autoimmune hemolytic anemia. Blood Adv. 2022;7(10):2166-2170. 46. Zanin-Zhorov A, Ding Y, Kumari S, et al. Protein kinase C-theta mediates negative feedback on regulatory T cell function. Science. 2010;328(5976):372-376. 47. Zhong H, Bussel J, Yazdanbakhsh K. In vitro TNF blockade enhances ex vivo expansion of regulatory T cells in patients with immune thrombocytopenia. Br J Haematol. 2015;168(2):274-283. 48. Zaragoza B, Chen X, Oppenheim JJ, et al. Suppressive activity of human regulatory T cells is maintained in the presence of TNF. Nat Med. 2016;22(1):16-17. 49. Cassatella MA, Meda L, Bonora S, Ceska M, Constantin G. Interleukin 10 (IL-10) inhibits the release of proinflammatory cytokines from human polymorphonuclear leukocytes. Evidence for an autocrine role of tumor necrosis factor and IL-1 beta in mediating the production of IL-8 triggered by lipopolysaccharide. J Exp Med. 1993;178(6):2207-2211. 50. Tecchio C, Micheletti A, Cassatella MA. Neutrophil-derived cytokines: facts beyond expression. Front Immunol. 2014;5:508. 51. Roders N, Herr F, Ambroise G, et al. SYK inhibition induces apoptosis in germinal center-like B cells by modulating the antiapoptotic protein myeloid cell leukemia-1, affecting B-cell activation and antibody production. Front Immunol. 2018;9:787. 52. Audia S, Mahevas M, Nivet M, Ouandji S, Ciudad M, Bonnotte B. Immune thrombocytopenia: recent advances in pathogenesis and treatments. Hemasphere. 2021;5(6):e574.

Haematologica | 109 February 2024

457

ARTICLE - Cell Therapy & Immunotherapy

Ruxolitinib-based regimen in children with primary hemophagocytic lymphohistiocytosis Jian Ge,1-3* Qing Zhang,2-4* Honghao Ma,1-3 Dong Wang,1-3 Yunze Zhao,1-3 Ting Zhu,2-4 Wenqian Wang,1-3 Chenxin Zhou,1-3 Ang Wei,1-3 Hongyun Lian,1-3 Maoquan Qin,1 Jun Yang,1 Zhigang Li,2-4 Tianyou Wang1-3 and Rui Zhang1-3 Hematology Oncology Center, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health; 2Beijing Key Laboratory of Pediatric Hematology Oncology, National Key Discipline of Pediatrics, Capital Medical University; 3Key Laboratory of Major Diseases in Children, Ministry of Education and 4Hematologic Disease Laboratory, Beijing Pediatric Research Institute, Beijing Children's Hospital, Capital Medical University, National Center for Children's Health, Beijing, China 1

Correspondence: R. Zhang ruizh1973@126.com Received: Accepted: Early view:

May 3, 2023. July 7, 2023. July 20, 2023.

JG and QZ contributed equally as first authors.

Abstract Primary hemophagocytic lymphohistiocytosis (pHLH) is a rare immune disorder and hematopoietic stem cell transplantation (HSCT) is the only potentially curative treatment. Given the high pre-HSCT mortality of pHLH patients reported in the HLH-2004 study (17%), more regimens to effectively control the disease and form a bridge with HSCT are needed. We conducted a retrospective study of pHLH children treated by ruxolitinib (RUX)-based regimen. Generally, patients received RUX until HSCT or unacceptable toxic side-effect. Methylprednisolone and etoposide were added sequentially when the disease was suboptimally controlled. The primary end point was 1-year overall survival. Twenty-one pHLH patients (12 previously treated and 9 previously untreated) were included with a median follow-up of 1.4 years. At last follow-up, 17 (81.0%) patients were alive with a 1-year overall survival of 90.5% (95% confidence interval: 84.1-96.9). Within the first 8 weeks, all patients had an objective response, of which 19 (90.5%) achieved complete response (CR) and two (9.5%) achieved partial response (PR) as a best response. Seventeen (81.0%) patients received HSCT, of which 13 (76.5%) had CR, three (17.6%) had PR and one (5.9%) had disease reactivation at the time of HSCT. Fifteen (88.2) patients were alive postHSCT. Notably, eight (38.1%) patients received zero doses of etoposide, suggesting the potential of RUX-based regimen to reduce chemotherapy intensity. Patients tolerated RUX-based regimen well and the most frequently observed adverse events were hematologic adverse events. Overall, RUX-based regimen was effective and safe and could be used as a bridge to HSCT for pHLH children.

Introduction Hemophagocytic lymphohistiocytosis (HLH) is a rare hematologic disorder characterized by pathologic immune activation and extreme inflammation.1,2 Primary HLH (pHLH) is a genetic disorder caused by mutation of genes involved in cytotoxicity machinery of natural killer and CD8+ T cells. pHLH can be diagnosed at any age but typically manifests during childhood. If left untreated, uncontrolled inflammation may result in severe organ dysfunction and death.3,4 As allogeneic hematopoietic stem cell transplantation (HSCT) is the only possible curative therapy for pHLH, it is critical to rapidly control inflammatory response to allow for early HSCT.5,6 Over the past 20 years, HLH-1994 and HLH-2004 regimen are widely used as a bridge to

transplantation and has significantly improved the survival of patient with pHLH. Despite these advancements, preHSCT mortality still remains high (17% according to the HLH-2004 study) due to frequent disease recurrence and toxic effects of chemotherapy.7,8 In addition, as an integral part of regimen, etoposide may have non-negligible longterm side effects including secondary tumor risk.9 These dilemmas make pHLH clinical management challenging and encourage physicians to search for new drugs to treat the disease. With increasing understanding of immunopathology of pHLH, two studies investigated efficacy of targeted therapy agents including alemtuzumab and emapalumab. Encouraging results of the studies demonstrate over 90% of patients treated by alemtuzumab as the first-line treatment survived to HSCT and 65% of patients who received emapalumab had a response.5,10

Haematologica | 109 February 2024

458

ARTICLE - RUX-based regimen in children with pHLH

J. Ge et al.

Ruxolitinib (RUX), an oral selective JAK1 and 2 inhibitor, has been considered promising targeted drug for HLH because it can inhibit signaling of key proinflammatory cytokines including interferon-γ (IFN-γ) involved in the disease.11-13 Our previous studies have demonstrated RUX is effective and safe in children with HLH and RUX monotherapy could control the disease rapidly. However, only four patients with pathogenic gene mutations associated with pHLH were enrolled in the two clinical trials.14,15 Individual case reports have indicated the benefits of using RUX as a bridge to HSCT for patients with pHLH that escaped etoposide-based regimen.16,17 In addition, several case studies have showed the efficacy of RUX in treating refractory HLH, as well as its potential as a first-line treatment for secondary HLH.18-20 However, the specific treatment plans in these studies were not uniform and there is still a lack of robust data on efficacy and safety of RUX in pHLH patients. In order to improve treatment outcomes of pHLH patients and reduce chemotherapy dose intensity, a RUX-based regimen was used in our center. In this study, we reported the findings of our retrospective study evaluating the efficacy and safety of RUX-based regimen in children with pHLH.

Methods Study design and patients We conducted a retrospective study of children with pHLH receiving RUX-based therapy in Beijing Children’s Hospital from January 2020 to October 2022. This study was approved by the Ethics Committee of Beijing Children’s Hospital and the written informed consent from the patients or their parents was obtained. Inclusion criteria included: i) previously treated and untreated children with a diagnosis of pHLH; ii) had active disease before RUX treatment. Patients were excluded for the following reasons: i) did not cooperate with treatment; ii) data were not available. Ruxolitinib-based treatment and response assessment In order to improve the treatment outcomes for patients with pHLH and reduce chemotherapy dosage intensity, RUX-based regimen was used as a bridge to HSCT. Patients received oral RUX phosphate tablets treatment within 48 hours after hospital admission, and the dose was 2.5 mg, 5 mg or 10 mg twice a day depending on the body weight (≤10 kg, ≤20 kg or >20 kg, respectively). Generally, RUX treatment was continued until HSCT unless the occurrence of intolerable adverse events or condition in critically ill (e.g., multiorgan system failure). Additional drugs including methylprednisolone (initially 2 mg/kg/d, tapered off in 8 weeks) and etoposide (100 mg/m2/dose,

once/week, continued until achieving complete response [CR]) were added in order. Methylprednisolone was added during RUX monotherapy if any of the following appeared: HLH relapse, marked worsening or no remission of disease symptoms and HLH-related indicators until day 3, or RUX withdrawal due to intolerance. During RUX plus methylprednisolone treatment, etoposide was added when any of the following appeared: HLH relapse, marked worsening or no remission until day 3 after methylprednisolone was added. One patient with pHLH triggered by Epstein-Barr virus (EBV) infection received low-dose liposomal doxorubicin (25 mg/m2) and pegaspargase (2,000 U/m2) as an auxiliary treatment. For patients with central nervous system (CNS) involvement, intrathecal treatment with corticosteroids and methotrexate was performed. When an acceptable donor was available, HSCT would be performed as early as possible. The preconditioning regimen we used was a fludarabine-based myeloablative regimen, including fludarabine, busulfan, cyclophosphamide, etoposide, and rabbit anti-thymocyte globulin. Treatment response was evaluated twice a week until patients achieved CR. Furthermore, we performed assessments at the designated time points, including on the third day of administering RUX, glucocorticoids, or etoposide, and whenever necessary as determined by the treating physician. The response evaluation criteria are provided in the Online Supplementary Table S1, which was mainly based on the criteria previously described with some modifications according to our clinical experience.21 Outcomes The primary outcome of this study was the 1-year overall survival. Secondary outcome included the best response within the first 8 weeks, disease status before HSCT, duration from RUX treatment to HSCT, death before and after HSCT, dose intensity of etoposide chemotherapy, and safety. Overall response (OR) rate included the proportion of patients with a CR and a partial response (PR). Response to treatment was evaluated as previously described, including CR, PR and no response (NR). Adverse events (AE) were assessed according to National Cancer Institute Common Terminology Criteria for Adverse Events (version 4.0). Statistical analysis Descriptive data are presented as the means ± standard deviation for variables with a normal distribution and the medians (minimum - maximum) for variables without a normal distribution. For categorical variables, number and percentage are presented. Characteristics were compared using two-tailed Student's t test or Wilcoxon rank sum test for continuous variables and χ2 test or Fisher exact test for categorical variables. Kaplan-Meier curves were used to analyze survival, and the log-rank test were used

Haematologica | 109 February 2024

459

ARTICLE - RUX-based regimen in children with pHLH

J. Ge et al.

to compare the differences in survival among patient subgroups. All statistical analyses were conducted using SPSS version 25.0 software (IBM, https://www.ibm.com/analytics/spss-statistics-software) and R version 3.6.3 (R Foundation for Statistical Computing, https://www.r-project.org), and a two-sided P value of 0.05 was used to determine the statistical significance.

Results Patients’ characteristics Between January 2020 and October 2022, a total of 27 children with pHLH were treated with RUX-based regimen at Beijing Children’s Hospital and six of them were excluded for various reasons (Online Supplementary Figure S1). Among the 21 patients included in the analysis, nine (42.9%) were previously untreated and 12 (57.1%) were previously treated at other hospitals.

Patient characteristics are summarized in Table 1. The median age was 3.10 years (range, 0.13-15.03 years). Ten (47.6%), six (28.6%), four (19.0%) and one (4.8%) patient had a genetic mutation in UNC13D, PRF1, XIAP and ITK, respectively. HLH in seven (33.3%) patients was triggered by EBV infection, and 11 patients (52.4%) had CNS involvement presenting abnormalities in one or more CNS symptoms, cerebrospinal fluid and radiological findings before RUX treatment. Characteristics between previously untreated patients and previously treated patients were consistent. Among 12 previously treated patients, 12 (100.0%) received glucocorticoids, eight (66.7%) received etoposide, two (16.7%) received cyclosporine A and one (8.3%) received liposomal doxorubicin before RUX treatment. The median duration of previous treatment was 55 days (range, 19-350 days). Seven (58.3%) had a CR and five (41.7%) had a PR as the best response to previous treatment. Before RUX treatment, all the previously treated patients had active

Table 1. Clinical patient characteristics. Total N=21

Previously untreated N=9

Previously treated N=12

3.10 (0.13-15.03)

3.60 (0.13-7.78)

2.30 (0.15-15.03)

0.098

Sex, N (%) Female Male

8 (38.1) 13 (61.9)

4 (44.4) 5 (55.6)

4 (33.3) 8 (66.7)

0.604

Gene, N (%) UNC13D PRF1 XIAP ITK

10 (47.6) 6 (28.6) 4 (19.0) 1 (4.8)

4 (44.4) 2 (22.2) 2 (22.2) 1 (11.1)

6 (50.0) 4 (33.3) 2 (16.7) 0 (0.0)

Trigger, N (%) EBV infection Unknown

7 (33.3) 14 (66.7)

3 (33.3) 6 (66.7)

4 (33.3) 8 (66.7)

1.000

Fever (>38.5℃), N (%)

7 (33.3)

3 (33.3)

4 (33.3)

1.000

Splenomegaly, N (%)

13 (61.9)

6 (66.7)

7 (58.3)

0.697

Neutrophils, ×109/L (range)

0.81 (0.36-3.40)

0.79 (0.51-1.82)

0.808

Platelets, ×109/L (range)

83 (21-569)

68 (21-152)

85 (55-569)

0.111

Hemoglobin, g/L (range)

83 (62-127)

83 (69-105)

86 (62-127)

0.298

Fibrinogen, g/L (range)

1.65 (0.68-3.05)

1.67 (0.68-2.44)

1.65 (1.20-3.05)

0.614

Triglycerides, mmol/L (range)

2.54 (0.51-11.42)

3.03 (1.74-3.55)

1.77 (0.51-11.42)

0.193

AST, U/L (range)

58.4 (16.9-1,435.7)

54.5 (16.9-1,435.7)

63.4 (33.7-452.9)

0.345

ALT, U/L (range)

47.6 (10.7-680.6)

28.2 (10.7-680.6)

59.0 (19.7-480.2)

0.102

IFN-γ, increase (fold)

12.3 (1.65-246.2)

13.9 (2.6-246.2)

8.1 (1.6-120.3)

0.382

sCD25, increase (fold)

3.0 (0.5-16.6)

4.4 (0.6-16.6)

1.7 (0.5-12.2)

0.169

Ferritin, increase (fold)

3.4 (0.1-39.8)

3.5 (0.1-22.1)

3.3 (0.2-39.8)

0.776

CNS involvement, N (%)

11 (52.4)

5 (55.6)

6 (50.0)

0.801

Patient subgroup Median age in years (range)

0.643

The baseline values of IFN-γ, ferritin and soluble CD25 were described as “increase (fold)”, which was calculated based on the upper limits of reference range (8 pg/mL, 500 μg/L and 6,400 pg/mL). ALT: alanine aminotransferase; LDH: lactate dehydrogenase; CNS: central nervous system; IFN-γ: interferon-γ; sCD25: soluble CD25. Laboratory clinical reference range: AST ≤40 U/L; ALT ≤40U/L; IFN-γ ≤8 pg/mL; ferritin ≤500 μg/L; soluble CD25 ≤6,400 pg/mL. Haematologica | 109 February 2024

460

ARTICLE - RUX-based regimen in children with pHLH

J. Ge et al.

disease. The other details for previous HLH treatment were described in the Online Supplementary Table S2. Efficacy Throughout the treatment, one (4.8%) patient received RUX monotherapy, seven (33.3%) patients received RUX plus methylprednisolone, thirteen (61.9%) patients received RUX plus methylprednisolone plus etoposide. Detailed treatment information for each patient is shown in the Online Supplementary Table S3. Clinical outcome of the patients is summarized in Table 2. Within the first 8 weeks, all patients had an objective response, of which 19 (90.5%) patients achieved CR and two (9.5%) patients achieved PR as a best response. Eight (42.1%) patients relapsed after achieving CR, and the triggers for relapse in four (50.0%) patients were infections and in four (50.0%) patients were unknown factors. Seventeen (81.0%) patients received HSCT, and the duration from RUX-based therapy to HSCT was 106.6±50.8 days. Among 17 patients received HSCT, 13 (76.5%) patients had a CR, three (17.6%) patients had a PR, one (5.9%) patient had HLH reactivation at the time of HSCT. Seventeen (100%) patients got CR after HSCT. Fifteen (88.2%) patients were alive post-HSCT, while two (11.8%) patients died more than 1 year after HSCT. Two patients with XIAP gene mutation did not undergo HSCT at the discretion of their physicians and they had sustained control of disease over 1 year after treatment was stopped. Four (19.0%) patients died, two (50.0%) of whom died before HSCT due to persistent HLH activation and two (50.0%) of whom died after HSCT due to severe graft-versus-host disease

(GvHD). Of note, both patients died before HSCT had CNS symptoms as the initial presentation of disease reactivation. In addition, we would like to highlight that serum levels of interleukin-6 and IFN-γ decreased rapidly and significantly during treatment (Online Supplementary Figure S2), providing further evidence of the effectiveness of the RUXbased regimen. Survival All patients were followed up to date of death or October 1, 2022 (time of data cutoff), with a mean follow-up of 1.4±0.7 years. Survival to HSCT and overall survival are shown in Figure 1. At the last follow up, 17 (81.0%) patients were alive with a 1-year cumulative probability of survival of 90.5% (95% confidence interval [CI]: 84.1- 96.9). For previously treated patients, ten (83.3%) patients were alive with an estimated 1-year survival of 91.7% (95% CI: 82.799.7). For previously untreated patients, seven (77.8%) patients were alive with an estimated 1-year survival of 88.9% (95% CI: 78.4-99.4) (Online Supplementary Figure S3). Etoposide dose intensity and cumulative glucocorticoid dose Dose intensity of chemotherapy in the duration of waiting for HSCT was a subject of intense scrutiny. Given etoposide was the most predominant chemotherapy drug for our patients, we calculated the dose of etoposide of every patient and compared it to the dose according to HLH2004 regimen under the same duration of waiting for HSCT. Three (33.3%) patients in the previously untreated

Table 2. Clinical outcome. Outcome

Total N=21

Previously untreated N=9

Previously treated N=12

Duration of follow-up in years, mean ± SD

1.4 ± 0.7

1.1 ± 0.6

1.6 ± 0.8

0.095

Achieve CR within the first 8 weeks of therapy, N (%) Relapse, N (%)

19 (90.5) 8 (42.1)

8 (88.9) 3 (37.5)

11 (91.7) 5 (45.5)

0.368 0.914

HSCT, N (%)

17 (81.0)

7 (77.8)

10 (83.3)

0.811

106.6 ± 50.8

102.1 ± 33.3

109.7 ± 61.9

0.883

Response at the time of HSCT,* N (%) CR PR Active

13 (76.5) 3 (17.6) 1 (5.9)

6 (85.7) 1 (14.3) 0 (0.0)

7 (70.0) 2 (20.0) 1 (10.0)

Death, N (%) Death before HSCT due to HLH

4 (19.0) 2 (9.5)

2 (22.2) 1 (11.1)

2 (16.7) 1 (8.3)

0.748 0.830

Death after HSCT, N (%)

2 (9.5)

1 (14.3)

1 (10.0)

0.787

Duration in days from RUX therapy to HSCT, mean ± SD*

0.635

*17 patients who underwent HSCT were included for analysis. HSCT: hematopoietic stem cell transplantation; CR: complete response; PR: partial response; SD: standard deviation; HLH: hemophagocytic lymphohistiocytosis. Haematologica | 109 February 2024

461

ARTICLE - RUX-based regimen in children with pHLH

J. Ge et al.

group and five (41.7%) patients in previously treated group received zero doses of etoposide. Totally, eight (38.1%) patients received zero doses of etoposide during the during the whole treatment. Under the same waiting time for HSCT, patients treated by RUX-based regimen might receive less doses of etoposide than that of the patients treated by HLH-2004 regimen (Figure 2), suggesting that RUX-based regimen had the potential to reduce chemotherapy intensity. Likewise, we calculated the cumulative dose of glucocorticoids of every patient. One (4.8%) patient received 0 mg of glucocorticoids during treatment. Under the same wait-

ing time for HSCT, patients treated by RUX-based regimen might receive lower cumulative dose of glucocorticoids than patients treated according to the HLH-2004 regimen (Online Supplementary Figure S4), suggesting that RUXbased regimen allows for a reduction in cumulative glucocorticoid dosing in pHLH patients.

Safety AE were summarized in Table 3. Overall, patients tolerated RUX-based treatment well and most AE were grade 1/2. Grade 3/4 AE that observed most frequently were hematologic AE, including anemia (23.8%), thrombocytopenia

Figure 1. Kaplan-Meier estimates of survival. (A) Survival until hematopoietic stem cell transplantation (HSCT). Two patients with XIAP genetic mutation didn’t receive HSCT were excluded from analyses. (B) Overall survival. RUX: ruxolitinib.

Figure 2. Patients treated by RUX-based regimen received reduced intensity of etoposide chemotherapy. (A) Previously treated patients. (B) Previously untreated patients (at the last follow up, 2 patients with XIAP genetic mutation didn't receive hematopoietic stem cell transplantation [HSCT] and their disease was controlled well over more than 1 year after the discontinuation of ruxolitinib [RUX]). After discontinuation of RUX, they didn't receive chemotherapy, so we only calculated the expected etoposide doses of HLH-2004 regimen in the RUX treatment duration). HLH: hemophagocytic lymphohistiocytosis. Haematologica | 109 February 2024

462

ARTICLE - RUX-based regimen in children with pHLH

J. Ge et al.

(23.8%), neutropenia (33.3%) and myelosuppression (33.3%). One patient had grade 3 pancreatic damage and was treated with somatostatin without RUX discontinuation. After about 3 weeks of treatment, her pancreatitis was resolved. One patient was diagnosed with pulmonary tuberculosis and received antituberculosis therapy with RUX discontinuation. After 6 months of treatment, his tuberculosis was resolved. Among eight patients who did not receive etoposide chemotherapy, apart from grade 3 anemia observed in two (25.0) patients, no other grade 3/4 AE were observed. Of note, AE might be caused by RUX, chemotherapy drugs, HLH activation and co-existing conditions of them.

Discussion PHLH is a rare and life-threating disorder characterized by hyperinflammation and immune dysregulation. Currently, the primary goal of therapy for pHLH patients is stably controlling the disease in order to perform HSCT, the only curative therapy.5 In this study, we presented the efficacy and safety of RUX-based regimen in children with pHLH. To the best of our knowledge, this is the first cohort study demonstrating the clinical benefits of treating pHLH patients with RUX-based regimen. At the last follow-up, 17 (81.0%) patients were alive with a 1-year overall survival of 90.5% (95% CI: 84.1-96.9) and 17 (81.0%) patients received HSCT. Interestingly, our results indicated patients treated by RUX-based therapy received relative lower intensity of etoposide, suggesting the potential of RUXbased regimen to help patients reduce chemotherapy

intensity. CNS involvement is a portion of systemic immune response in HLH.22 During the period of HLH treatment, frequent reactivation within CNS could occur independent of or concomitant with systemic relapses, which may be associated with the high risk of mortality.23 Up to now, there have been no clinical trials focusing specifically on CNS involvement in HLH patients. Currently, intrathecal treatment with corticosteroids and methotrexate is the standard care for CNS symptoms and may have beneficial effects.24 Results of an animal experiments indicated that RUX could penetrate the blood brain barrier of mice and RUX therapy could reduce CNS involvement in the Rab27a-/- mice, but these findings haven't been confirmed in human patients.13,25 In this study, all patients with CNS involvement received intrathecal therapy and RUX treatment. However, two patients with CNS-HLH died due to disease reactivation with somnolence and coma as the first symptoms before undergoing HSCT. Two patients achieved PR but not CR as a best response because CNS involvement could not be completely remitted. Four patients with CNS involvement relapsed after achieving CR, of which three patients had CNS symptoms when the disease relapsed. These observations suggest RUX is probably not an ideal drug for CNS involvement and more effective treatments are needed. There is also evidence for the importance of HSCT in CNS involvement. Results from a retrospective study of 18 patients in a single center indicated immediate HSCT may be beneficial even if there is active disease.26,27 Given the long waiting time for acceptable donors, physicians have to consider other treatments for CNS-HLH. Thus, the most plausible intervention for CNS involvement in pHLH pa-

Table 3. Possible adverse event.* Outcome

Any grade

Grade ≥3

Hematologic AE, N (%) Anemia Thrombocytopenia Neutropenia Myelosuppression

15 (71.4) 13 (61.9) 15 (71.4) 14 (66.7)

5 (23.8) 5 (23.8) 7 (33.3) 7 (33.3)

Non-hematologic abnormalities, N (%) Constipation Pancreatic damage Rash Diarrhea Liver damage Sweating Gastritis Secondary infection Heart damage Kidney damage Gastrointestinal hemorrhage

6 (28.6) 7 (33.3) 4 (19.0) 4 (19.0) 5 (23.8) 1 (4.8) 1 (4.8) 3 (14.3) 3 (14.3) 3 (14.3) 3 (14.3)

0 1 (4.8) 0 0 0 0 0 1 (4.8) 0 0 1 (4.8)

A part of patients had more than 1 adverse event (AE). *AE might be caused by drugs, hemophagocytic lymphohistiocytosis activation and co-existing conditions of them. Haematologica | 109 February 2024

463

ARTICLE - RUX-based regimen in children with pHLH

J. Ge et al.

tients warrants further exploration. During HLH treatment, the high dosage of chemotherapy drugs remains an important concern. Etoposide-based HLH-1994 and HLH-2004 regimens are the most frequently used chemotherapy regimens to treat HLH and the treatment-related morbidity and potential mortality has been observed.28 Possible AE from the therapy, especially etoposide, include secondary infections, hepatic dysfunction, myelotoxicity and secondary malignancies. Moreover, the toxic effects will be increased with reintensification of etoposide among patients with HLH flares.5,9 Therefore, alternative regimens with less toxicity are urgently needed. According to the recommendations provided by HLH steering committee of the histiocyte society for the use of etoposide-based therapy for the treatment of HLH, treatment may have to be individualized depending on the clinical context, and the drug doses and/or dosing intervals can be altered.29 Remarkably, our results suggest that RUX-based regimen have good efficacy and the potential to reduce chemotherapy intensity. pHLH is known to be characterized by frequent reactivations.24 However, for most patients in this study, the disease was controlled rapidly by treated with RUX-based regimen and parts of them had well-controlled disease persistently without HLH reactivation after achieving CR. Moreover, results of our previous studies demonstrated RUX had a quick effect on HLH because all responding patients achieved the first response to RUX monotherapy within 3 days.15 Therefore, RUX-based regimen or RUX-contained regimen are worth considering to minimize the toxicity of chemotherapy when making treatment plans for HLH patients. Management and treatment for pHLH patients with XIAP gene mutation may be different. In a study from Europe, 54 HLH patients with X-linked inhibitor of apoptosis deficiency did not undergo HSCT and 49 of them survived at a median time of 4 years after diagnosis.30 In our study, two patients with pathogenic genetic mutations of XIAP did not receive HSCT and have stopped the drugs for more than 1 year. They survived without disease activation until the last follow-up. For these patients, the most beneficial treatment decisions should be made according to the clinical features. There are several limitations in this study. First, this is a retrospective study and there could be sources of bias including confounding bias, elective bias and observational bias. Therefore, our results of this study need to be further validated by well-designed, prospective study. Second, this study included patients who had received prior treatment that had either been ineffective in achieving the desired outcome or had caused intolerable side effects. However, there is still no evidence of worse prognosis for

these patients and their clinical characteristics and outcome were not significantly different with initially treated patients. Third, all patients included in this study were of Chinese origin, and the most highly mutated gene was UNC13D, which is different from studies conducted by the Histiocyte Society.7,8 Thus, results from this study may not be applicable to other pHLH populations. Fourth, it is important to acknowledge that the median follow-up duration in this study was relatively short, spanning only 1.4 years, and therefore, the long-term outcomes of survival post-HSCT remain unclear. Nevertheless, previous research indicates pHLH patients who had CR at HSCT may have a more favorable long-term overall survival than those with PR.31 Notably, 76.5% of patients in this study had CR at the time of HSCT, suggesting that the use of RUX-based regimen as a pre-HSCT therapy may hold great promise for pHLH patients. In summary, our study demonstrates that for children with pHLH, RUX-based regimen was effective and safe and could be used as a bridge to HSCT. Disclosures No conflicts of interest to disclose. Contributions RZ, JG and QZ conceptualized and designed the study and drafted the initial manuscript. ZGL and TYW conceptualized and designed the study and reviewed and revised the manuscript. AW, TZ, WQW and CXZ collected the data, carried out the initial analyses, and reviewed and revised the manuscript. HHM, DW, YZZ, HYL, MQQ and JY conceptualized and designed the study, coordinated and supervised the data collection, and critically reviewed the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work. Acknowledgments The authors thank the patients and their families for contributing to this study. Funding This work was supported by Beijing Municipal Science & Technology Commission (no. Z221100007422054), the Beijing research Ward Project (no. BCRW202101), the Special Fund of the Pediatric Medical Coordinated Development Center of Beijing Municipal Administration (no. XTZD20180202). Data-sharing statement Original data and protocols are available to other investigators upon request by contacting the corresponding author.

Haematologica | 109 February 2024

464

ARTICLE - RUX-based regimen in children with pHLH

J. Ge et al.

References 1. Henter JI, Elinder G, Soder O, Ost A. Incidence in Sweden and clinical features of familial hemophagocytic lymphohistiocytosis. Acta Paediatr Scand. 1991;80(4):428-435. 2. Canna SW, Marsh RA. Pediatric hemophagocytic lymphohistiocytosis. Blood. 2020;135(16):1332-1343. 3. Arico M, Janka G, Fischer A, et al. Hemophagocytic lymphohistiocytosis. Report of 122 children from the International Registry. FHL Study Group of the Histiocyte Society. Leukemia. 1996;10(2):197-203. 4. Henter JI, Elinder G, Soder O, Hansson M, Andersson B, Andersson U. Hypercytokinemia in familial hemophagocytic lymphohistiocytosis. Blood. 1991;78(11):2918-2922. 5. Locatelli F, Jordan MB, Allen C, et al. Emapalumab in children with primary hemophagocytic lymphohistiocytosis. N Engl J Med. 2020;382(19):1811-1822. 6. Marsh RA, Haddad E. How I treat primary haemophagocytic lymphohistiocytosis. Br J Haematol. 2018;182(2):185-199. 7. Trottestam H, Horne A, Arico M, et al. Chemoimmunotherapy for hemophagocytic lymphohistiocytosis: long-term results of the HLH-94 treatment protocol. Blood. 2011;118(17):4577-4584. 8. Bergsten E, Horne A, Arico M, et al. Confirmed efficacy of etoposide and dexamethasone in HLH treatment: long-term results of the cooperative HLH-2004 study. Blood. 2017;130(25):2728-2738. 9. Johnson TS, Terrell CE, Millen SH, Katz JD, Hildeman DA, Jordan MB. Etoposide selectively ablates activated T cells to control the immunoregulatory disorder hemophagocytic lymphohistiocytosis. J Immunol. 2014;192(1):84-91. 10. Fischer A, Blanche S, Neven B, et al. Alemtuzumab as first line treatment in children with familial lymphohistiocytosis. Blood. 2019;134(Suppl 1):S80. 11. Das R, Guan P, Sprague L, et al. Janus kinase inhibition lessens inflammation and ameliorates disease in murine models of hemophagocytic lymphohistiocytosis. Blood. 2016;127(13):1666-1675. 12. Albeituni S, Verbist KC, Tedrick PE, et al. Mechanisms of action of ruxolitinib in murine models of hemophagocytic lymphohistiocytosis. Blood. 2019;134(2):147-159. 13. Maschalidi S, Sepulveda FE, Garrigue A, Fischer A, de Saint Basile G. Therapeutic effect of JAK1/2 blockade on the manifestations of hemophagocytic lymphohistiocytosis in mice. Blood. 2016;128(1):60-71. 14. Zhang Q, Wei A, Ma HH, et al. A pilot study of ruxolitinib as a front-line therapy for 12 children with secondary hemophagocytic lymphohistiocytosis. Haematologica. 2021;106(7):1892-1901. 15. Zhang Q, Zhao YZ, Ma HH, et al. A study of ruxolitinib responsebased stratified treatment for pediatric hemophagocytic lymphohistiocytosis. Blood. 2022;139(24):3493-3504. 16. Marois L, Touzot F, Haddad E, et al. Successful management of familial hemophagocytic lymphohistiocytosis by the JAK 1/2 inhibitor ruxolitinib. Pediatr Blood Cancer. 2021;68(6):e28954. 17. Ramanan KM, Uppuluri R, Ravichandran N, et al. Successful remission induction in refractory familial hemophagocytic lymphohistiocytosis with ruxolitinib as a bridge to hematopoietic stem cell transplantation. Pediatr Blood Cancer.

2020;67(3):e28071. 18. Zandvakili I, Conboy CB, Ayed AO, Cathcart-Rake EJ, Tefferi A. Ruxolitinib as first-line treatment in secondary hemophagocytic lymphohistiocytosis: a second experience. Am J Hematol. 2018;93(5):E123-E125. 19. Slostad J, Hoversten P, Haddox CL, Cisak K, Paludo J, Tefferi A. Ruxolitinib as first-line treatment in secondary hemophagocytic lymphohistiocytosis: a single patient experience. Am J Hematol. 2018;93(2):E47-E49. 20. Broglie L, Pommert L, Rao S, et al. Ruxolitinib for treatment of refractory hemophagocytic lymphohistiocytosis. Blood Adv. 2017;1(19):1533-1536. 21. Zhao Y, Li Z, Zhang L, et al. L-DEP regimen salvage therapy for paediatric patients with refractory Epstein-Barr virusassociated haemophagocytic lymphohistiocytosis. Br J Haematol. 2020;191(3):453-459. 22. Horne A, Wickstrom R, Jordan MB, et al. How to treat involvement of the central nervous system in hemophagocytic lymphohistiocytosis? Curr Treat Options Neurol. 2017;19(1):3. 23. Horne A, Trottestam H, Arico M, et al. Frequency and spectrum of central nervous system involvement in 193 children with haemophagocytic lymphohistiocytosis. Br J Haematol. 2008;140(3):327-335. 24. Henter JI, Horne A, Arico M, et al. HLH-2004: Diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer. 2007;48(2):124-131. 25. Haile WB, Gavegnano C, Tao S, Jiang Y, Schinazi RF, Tyor WR. The Janus kinase inhibitor ruxolitinib reduces HIV replication in human macrophages and ameliorates HIV encephalitis in a murine model. Neurobiol Dis. 2016;92(Pt B):137-143. 26. Deiva K, Mahlaoui N, Beaudonnet F, et al. CNS involvement at the onset of primary hemophagocytic lymphohistiocytosis. Neurology. 2012;78(15):1150-1156. 27. Sparber-Sauer M, Honig M, Schulz AS, et al. Patients with early relapse of primary hemophagocytic syndromes or with persistent CNS involvement may benefit from immediate hematopoietic stem cell transplantation. Bone Marrow Transplant. 2009;44(6):333-338. 28. Jordan MB, Allen CE, Weitzman S, Filipovich AH, McClain KL. How I treat hemophagocytic lymphohistiocytosis. Blood. 2011;118(15):4041-4052. 29. Ehl S, Astigarraga I, von Bahr Greenwood T, et al. Recommendations for the use of etoposide-based therapy and bone marrow transplantation for the treatment of HLH: consensus statements by the HLH Steering Committee of the Histiocyte Society. J Allergy Clin Immunol Pract. 2018;6(5):1508-1517. 30. Yang L, Booth C, Speckmann C, et al. Phenotype, genotype, treatment, and survival outcomes in patients with X-linked inhibitor of apoptosis deficiency. J Allergy Clin Immunol. 2022;150(2):456-466. 31. Bergsten E, Horne A, Hed Myrberg I, et al. Stem cell transplantation for children with hemophagocytic lymphohistiocytosis: results from the HLH-2004 study. Blood Adv. 2020;4(15):3754-3766.

Haematologica | 109 February 2024

465

ARTICLE - Chronic Lymphocytic Leukemia

IGH 3’RR recombination uncovers a non-germinal center imprint and c-MYC-dependent IGH rearrangement in unmutated chronic lymphocytic leukemia Israa Al Jamal,1,2 Milène Parquet,1+ Kenza Guiyedi,1+ Said Aoufouchi,3,4+ Morwenna Le Guillou,3 David Rizzo,1,5 Justine Pollet,1 Marine Dupont,1,5 Mélanie Boulin,1,5 Nathalie Faumont,1 Hend Boutouil,1 Fabrice Jardin,6 Philippe Ruminy,6 Chahrazed El Hamel,7 Justine Lerat,8 Samar Al Hamaoui,2 Nehman Makdissy,2 Jean Feuillard,1,5 Nathalie Gachard1,5 and Sophie Peron1 1

Centre National de la Recherche Scientifique (CNRS), Unité Mixte de Recherche (UMR),

7276/INSERM U1262, Université de Limoges, Limoges, France; 2Faculty of Sciences, GSBT Genomic Surveillance and Biotherapy Team, Mont Michel Campus, Lebanese University, Tripoli, Lebanon; 3CNRS UMR9019, Gustave Roussy, B-cell and Genome Plasticity Team,

Correspondence: S. Peron sophie.peron@unilim.fr Received: Accepted: Early view:

February 13, 2023. July 20, 2023. July 27, 2023.

Villejuif, France; 4Université Paris-Saclay, Orsay, France; 5Laboratoire d’Hématologie Biologique, Centre Hospitalier Universitaire de Limoges, Limoges, France; 6Department of Henri-Becquerel Hematology Center and Normandie, INSERM U1245, Université de Rouen, Rouen, France; 7Collection Biologique Hôpital de la Mère et de l’Enfant (CB-HME), Department of Pediatrics, Limoges University Hospital, Limoges, France and 8Department of Otorinolaryngology, Limoges University Hospital, Limoges, France MP, KG and SA contributed equally.

Abstract Chronic lymphocytic leukemia (CLL) is an incurable indolent non-Hodgkin lymphoma characterized by tumor B cells that weakly express a B-cell receptor. The mutational status of the variable region (IGHV) within the immunoglobulin heavy chain (IGH) locus is an important prognosis indicator and raises the question of the CLL cell of origin. Mutated IGHV gene CLL are genetically imprinted by activation-induced cytidine deaminase (AID). AID is also required for IGH rearrangements: class switch recombination and recombination between switch Mu (Sμ) and the 3’ regulatory region (3’RR) (Sμ-3’RRrec). The great majority of CLL B cells being unswitched led us to examine IGH rearrangement blockade in CLL. Our results separated CLL into two groups on the basis of Sμ-3’RRrec counts per sample: Sμ-3’RRrecHigh cases (mostly unmutated CLL) and Sμ-3’RRrecLow cases (mostly mutated CLL), but not based on the class switch recombination junction counts. Sμ-3’RRrec appeared to be ongoing in Sμ-3’RRrecHigh CLL cells and comparison of Sμ-3’RRrec junction structural features pointed to different B-cell origins for both groups. In accordance with IGHV mutational status and PIM1 mutation rate, Sμ-3’RRrecHigh CLL harbor a non-germinal center experienced B-cell imprint while Sμ-3’RRrecLow CLL are from AID-experienced B cells from a secondary lymphoid organ. In addition to the proposals already made concerning the CLL cell of origin, our study highlights that analysis of IGH recombinatory activity can identify CLL cases from different origins. Finally, on-going Sμ-3’RRrec in Sμ-3’RRrecHigh cells appeared to presumably be the consequence of high c-MYC expression, as c-MYC overexpression potentiated IGH rearrangements and Sμ-3’RRrec, even in the absence of AID for the latter.

Introduction Being one of the most frequent B-cell cancers of the elderly, chronic lymphocytic leukemia (CLL) is characterized by lymphocytosis exceeding ≥5.0×109/L, and is composed of small circulating monomorphic round CD19+ CD23+ CD5+ B cells, as well as bone marrow and secondary lymphoid organ infiltration in most cases.1 CLL evolution is highly

variable, with overall survival ranging from a few years to decades and is still incurable despite the development of new therapeutics such as Bruton tyrosine kinase or Bcl2 inhibitors. Binet and Rai classifications are still the most reliable staging systems to predict CLL course and are the keystones of clinical decision for treatment.2,3 However, patients show marked karyotypic and genetic heterogeneity which also

Haematologica | 109 February 2024

466

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

influences overall survival rate and prediction of therapeutic response. Major prognosis factors are chromosomal abnormalities such del17p, del11q, trisomy 12, isolated del13q or complex karyotypes which highlight genomic instability in CLL pathogenesis.4 Among poor prognosis genetic abnormalities are those involving the Notch pathway (Notch1 mutations), NF-κB activation (BIRC3 or MYD88 mutations), splicing (SF3B1), the DNA lesion sensor ATM or the TP53 anti-oncogene.5 Underscoring the role of the B-cell receptor (BCR) in this B-cell cancer, another important prognosis indicator is the mutational status of the variable region (IGHV) within the immunoglobulin heavy chain (IGH) locus, which separates CLL patients into two groups: those with an unmutated variable region and those with a mutated IGHV rearranged gene (umCLL and mCLL, respectively). The former bestows a poor prognosis while patients with a very long survival rate are found in the latter group.6,7 Enforcing the role of the BCR is the fact that 30% of CLL patients express a socalled “stereotyped receptor”, which suggests the role of common antigenic determinants in the promotion of B-cell transformation.8 The fact that mCLL are genetically imprinted by activation-induced cytidine deaminase (AID)-dependent IGHV somatic hypermutation (SHM) raises the question of a CLL group with a post-germinal center (GC) B-cell counterpart. Recent methylome analyses suggest proximities between mCLL and GC experienced memory B cells on the one hand and umCLL and naïve B cells on the other hand.9–11 However, CLL cells exhibit a unique CD5+, CD23+, CD27+, CD43+ with low levels of surface immunoglobuln (Ig)M and IgD immunophenotype, which is different from that of any normal B cell.12,13 Gene expression profiles revealed that both CLL groups share a characteristic gene expression signature that is close to that of antigen-experienced B cells.12,14 Reconstruction of B-cell differentiation trajectories indeed suggested that precursors of both umCLL and mCLL have reached the antigen-experienced memory B-cell stage.13 It has also been proposed that CLL may originate from an extra-follicular B-cell response since maturation of these cells is antigen driven and can be either mutated or not.12 Therefore, more than 20 years after the papers of Hamblin and Damle,6,7 this question of the CLL cell of origin (COO) has not been clearly answered and umCLL could differ from mCLL mainly by expressing BCR-related mitogenic markers.14 The fact that most CLL cells express an IgM raises the question of class switch recombination (CSR) blockade in this B-cell cancer. Low levels of CSR can be observed in a small fraction of CLL tumor cells and seem to correlate with AID expression in these intraclonal switched CLL cells.15–17 AID has been repeatedly detected in CLL B cells independently of the IGHV mutational status.15 AID likely contributes to CLL evolution and seems to generate intraclonal diversity targeting the IGH locus and non-Ig

I. Al Jamal et al.

off-targets.18 Physiologically, AID induces CSR in activated proliferating B cells.19,20 CSR results from IGH intrachromosomal recombination between the switch μ (Sμ) region and another switch region (so-called Sx) located upstream of one of the constant genes.21 CSR requires double strand DNA breaks (DSB). Converting a cytosine into uracil, AID creates a U:G mismatch that is targeted by the base excision repair (BER) pathway to excise the uracil base using uracil-DNA-glycosylase (UNG). This results in an abasic site which is processed and generates a single strand DNA break (SSB). Several SSB on both DNA strands then produce DSB which are repaired through the ubitquitous DSB repair (DSBR) response. In B cells, during CSR, DSB repair occurs by the joint action of the non-homologous end joining (NHEJ) and alternative end-joining (Alt-EJ) pathways.22,23 With XRCC4 for the ligation step, NHEJ depends on 53BP1/ Rif124,25 which recruits the Shieldin complex. Both 53BP1/ Rif1 and Shieldin complexes are essential for protection against the resection of DNA ends.26–28 Less understood and described, Alt-EJ involves PARP1, POLQ and/or LigIII. These DSBR pathways differently shape the structure of the repair junction; each single junction is thus unique. Being highly conserved across mammalian species (human, mouse, dog, rabbit etc.), the 3’ regulatory regions (3’RR) of the IGH locus are key regulatory regions for CSR. 3’RR exhibits a singular structure with three DNaseI hypersensitive (HS) sites (HS3, HS1-2, and HS4) harboring strict specific B-lineage transcriptional enhancer activity related to a “quasi-palindrome” organization where inverted repeated sequences flank the HS1-2 sequence which is the symmetry center element.29 The 3’RR can also be recombined with the Sμ region in an AID-dependent manner (Sμ-3’RRrec).30 Sμ-3’RRrec has been shown to occur in vitro and in vivo in activated murine and human B cells.30–32 It has been repeatedly detected in secondary lymphoid organs and peripheral blood mononuclear cells (PBMC) of both mice and humans. Similar to CSR in activated mature B cells, Sμ-3’RRrec occurs between transcribed recombination donor and acceptor DNA segments. However, we observed that the structure of Sμ-3’RRrec junctions in murine B cells was different from that of CSR.31 Indeed, Sμ-3’RRrec junctions are reminiscent of the usage of NHEJ and/or Alt-EJ, with regard to the repair signature at Sμ-3’RRrec joints and the recruitment of Alt-EJ but not NHEJ components at the 3’RR locus in mice.31 In contrast with CSR, Alt-EJ seems, therefore, to surface as the major contributor. Therefore, some arguments support the fact that Sμ-3’RRrec is a different IGH recombination from that of CSR. When Sμ-3’RRrec hits the IGH locus, it results in the excision of the whole cluster of constant IGH genes. This should kill BCR expression if occurring on the functional IGH allele. Since in vivo loss of BCR induces B-cell death,33 Sμ-3’RRrec was initially called “locus suicide recombination”. However, to date, the search for such B cells lacking BCR due to Sμ-3’RRrec has failed. Moreover, high throughput sequencing studies on human

Haematologica | 109 February 2024

467

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

BCR-positive circulating B cells revealed that amplicons covering Sμ-3’RRrec junctions certainly came from the non-functional IGH allele.32 Thus, at present, the function of Sμ-3’RRrec in B-cell physiology remains unclear. In this study, we raised the question of IGH switch blockade in CLL. For that purpose, we analyzed both CSR and Sμ-3’RRrec junctions as reflections of putative switch activity. Our results revealed that CLL patients could be separated into two groups with different prognoses on the basis of Sμ-3’RRrec counts but not CSR junction counts. Cases with increased Sμ-3’RRrec were indeed related to umCLL. Comparison between both groups revealed that Sμ-3’RRrec was likely to be ongoing in tumor cells from CLL patients with increased Sμ-3’RRrec counts. Structural features of Sμ-3’RRrec junctions revealed an imprint that pointed to a different B-cell origin for both groups. Moreover, Sμ-3’RRrec appeared to be potentiated by MYC overexpression even in the absence of AID.

Methods Human materials and ethics The project was conducted according to the guidelines of the Declaration of Helsinki. CLL peripheral blood mononucelar cells (PBMC) were obtained from CRBioLim, CHU Dupuytren, Limoges Hospital (authorization no.: DC-2008-604, AC-2016-2758, and AC-2019-3418). Tonsils were obtained from children scheduled for elective tonsillectomy from CRBioLim (authorization no.: DC-2008-604, AC-2018-3157). PBMC from healthy volunteers (HV) were collected through the research project approved by CPP Sud MéditerranéeI (authorization no.: 2021-A00778-33). Human class switch recombination and Sμ μ-3’RRrec junction counts Human CSR and Sμ-3’RRrec junctions were amplified as described32 and used to prepare next-generation sequencing (NGS) libraries (Ion Xpress™ Plus Fragment Library Kit, Life Technologies, Thermofisher, 447269) sequenced with an Ion Proton or S5 chip (Life Technologies). FastQ were analyzed using CSReport.34 CSR and Sμ-3’RRrec junction diversities were estimated through the Shannon Diversity Index (see the Online Supplementary Appendix). The Jurkat cell line and naïve B cells sorted from PB of healthy donors (n=2) served as negative controls for Sμ-3’RRrec junction detection (no Sμ-3’RRrec junctions were detected). μ-3’RRrec CH12F3 class switch recombination and Sμ junction counts The CH12F3 cells were transfected or not by MYC expression vector (Plasmid#74164, Addgene) and cultured in RPMI1640 with Ultra Glutamine, 10% fetal calf serum (FCS) (Lonza), sodium pyruvate (Lonza), penicillin/streptomycin (Lonza), non-essential amino acids (Lonza) and β2-mercaptoetha-

I. Al Jamal et al.

nol. Cells were stimulated for CSR toward IgA for 72 hours with murine interleukin (IL)-4 (5 ng/mL; PeproTech), human TGF-β1 (1 ng/mL; R&D Systems), and murine anti-CD40 antibody (Ab) (1 μg/mL; eBioscience). CSR and Sμ-3’RRrec junctions were amplified by nested polymerase chain reaction (PCR) using specific primers (Online Supplementary Table S3) as described.34 IGHV sequence analysis V, D, and J rearranged genes were amplified using the Biomed-2 strategy with FR1 and FR2 primers and sequence analyses were performed as described.35 The IGHV intra-clonal diversity analysis workflow is described in the Online Supplementary Appendix. Diffuse large B-cell lymphoma IGHV sequences are from Rizzo et al.36 Flow cytometry analysis Immunophenotyping was done on a Navios-flow cytometer (Beckman Coulter) with the protocol for routine CLL diagnosis using: CD5-APC (Beckman Coulter PN a60790, clone BL1a), CD19-ECD (Beckman Coulter A07770, clone J3-119) and anti-human κ light chain/anti-human λ light chains/RPE (Dako, FR481 X0935). Results were analyzed with Kaluza software version 2.1 (Beckman Coulter). RNA extraction, cDNA synthesis and quantitative real time polymerase chain reaction Total RNA was isolated (TRIzol™ Reagent, 15596018) and reverse-transcribed (Advantage RT-for-PCR kit Applied Biosystems™, Thermofisher 4368814/10400745). Quantitative real time PCR (qRT-PCR) were performed with the SYBRGreen PCR mix (SensiFast hi ROX Syber Green BIO820025) and primers (Online Supplementary Table S3) or with Taqman PCR mix (SensiFast Probe Hi-Rox kit BIO820025) and MYC probe (4331182 Hs00905030_m1, Thermofisher). Normal centroblasts and naïve B cells were sorted from tonsils as described.37 Relative telomere length assay DNA (25 ng) extracted from PBMC was used in triplicate to assess relative telomere assay (RTL) by qPCR as described previously.38 Mutation analysis of PIM1 We amplified PIM1 exon 4 (Online Supplementary Table S3), containing a CLL AID-targeted nucleotide,18 using Phusion High Fidelity Taq (Thermo Scientific, F-530XL). Products were used to build NGS libraries. The analysis workflow is described in the Online Supplementary Appendix. Statistical analysis Graphs, histograms, curves, and standard statistical analyses were designed using GraphPad Prism 6x software. Fisher tests were done with R (version 4.3.0) using the RStudio interface (RStudio 2023.03.0 Build 386). Kaplan

Haematologica | 109 February 2024

468

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

Meyer survival curves and Cox univariate and multivariates analyses were done using the R Survival package (URL: https://github.com/therneau/survival).

Results Chronic lymphocytic leukemia patients can be separated into two groups according to Sμ μ-3’RR recombination We analyzed counts of Sμ-3’RR recombination (Sμ-3’RRrec) in DNA samples collected at CLL diagnosis from 47 patients. Blood tumor infiltration was over 90% circulating lymphocytes in 42 of 47 (89%) cases and over 98% in 37 of 42 cases (79%) (Online Supplementary Figure S1A). Comparison of junction counts was performed with results obtained from DNA from PBMC of nine HV. As negative controls, we used the Jurkat cell line and naïve B cells sorted from PB of healthy donors (n=2), in which no Sμ-3’RRrec junctions could be detected (data not shown). Even at low levels, Sμ-3’RRrec was found at comparable levels in both HV and CLL (Online Supplementary Figure S1B), and was undetectable in only three of 47 (6.3%) CLL patients. We separated CLL patients into two groups, using as a threshold value the mean of Sμ-3’RRrec counts in HV (Online Supplementary Table S1), called Sμ-3’RRrecHigh (12/47 patients =26%), and Sμ-3’RRrecLow (35/47 patients =74%) (Figure 1A). Analysis of CSR and Sμ-3’RR recombinations was based on a nested-PCR approach. Somatic hypermutation (SHM) could theoretically introduce mutations in primer binding DNA, particularly in CLL.39 But here, low levels of Sμ (mutation rate average +/- standard error of the mean for Sμ-3’RRrecLow, Sμ-3’RRrecHigh and HV PBMC respectively are: 1.828+/-0.306, 1.475+/-0.302 and 0.870+/-0.090) and 3’RR2 (mutation rate average +/- standard error of the mean

I. Al Jamal et al.

for Sμ-3’RRrecLow, Sμ-3’RRrecHigh and HV PBMC respectively: 0.048+/-0.018, 0.062+/-0.020 and 0.056+/-0.033) mutation frequency. In DNA segments from Sμ-3’RRrec junctions, absence of significant differences between samples ruled out significant bias in amplification of the Sμ-3’RRrec junctions and comparison between Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL. Sμ-3’RRrec counts were not dependent on CLL B-cell richness as shown in the Online Supplementary Figure S1C. Moreover, the percentages of CLL B cells were similar in Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL samples (Online Supplementary Figure S1D). As expected in this IgM+ B-cell cancer, CSR junction levels were much lower in CLL than in HV samples. CSR counts were similarly low in both the Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL groups (Figure 1B). We did not find any significant association between increased Sμ-3’RRrec and CSR counts as shown in the Online Supplementary Table S2. Moreover, the correlation between CSR and Sμ-3’RRrec counts was poor (correlation coefficient r=0.2; data not shown). Thus, Sμ-3’RRrecHigh CLL specifically exhibited increased Sμ-3’RRrec counts when compared to CSR. μ-3’RRrec counts had Most patients with Increased Sμ unmutated chronic lymphocytic leukemia In order to further study Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL, we analyzed the IGHV mutational status, an important prognosis indicator of poor outcome. Even if being cautious for small numbers, no significant IGHV gene repertoire bias was found between CLL groups (Online Supplementary Figure S2). IGHV clonal rearrangements of Sμ-3’RRrecHigh CLL cases exhibited stronger homology to IGHV reference sequences (Figure 2A). With the threshold of 98% homology, nine of 12 (75%) Sμ-3’RRrecHigh CLL were not or only weakly mutated (mean IGHV mutation rate =98.1%). One additional

Figure 1. Sμ μ-3’RRrec is detectable in chronic lymphocytic leukemia patients and Sμ μ-3’RRrec counts are significantly increased μ-3’RRrecHigh samples. (A) Sμ-3’RR recombination (Sμ-3’RRrec) junction counts analyzed by next-generation sequencing and in Sμ CSReport in healthy volunteer peripheral blood mononuclear cells (PBMC) (N=9, 239 Sμ-3’RRrec junctions) and chronic lymphocytic leukemia (CLL) patients, divided into 2 groups based on the mean of junction counts obtained in healthy PBMC: Sμ-3’RRrecLow CLL (≤27 junctions per sample, N=35, 357 junctions) and Sμ-3’RRrecHigh CLL (>27 junctions per sample, N=12, 703 junctions). (B) Class switch recombination (CSR) junction counts were at comparable levels in both CLL groups (Sμ-3’RRrecLow: N=35, 22,247 junctions; Sμ-3’RRrecHigh: N=11, 10,739 junctions) and were lower than in HV PBMC (N=11, 27,528 junctions). Graphs represent the mean ± standard error of the mean. Statistical analyses were performed using unpaired t test. 3’RR: 3’ regulatory region; NS: not significant; **P<0.01; ***P<0.001. Haematologica | 109 February 2024

469

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

Sμ-3’RRrecHigh CLL had 97% IGHV sequence homology with the reference. Sμ-3’RRrecLow cases were mCLL for which 20 of 35 (57%) cases had a mean IGHV mutation rate =4.8%; P=0.043. Consistently, we found that Sμ-3’RRrecHigh patients exhibited low rates of AID off-target PIM1 mutations (Figure 2B). As CD19 transcription and expression at the cell surface are specific for the B-cell compartment and were similar between Sμ-3’RRrecLow and Sμ-3’RRrecHigh CLL (Online Supplementary Figure S3A, B), transcript expression levels were normalized to those of CD19. When compared to centroblasts and naïve B cells sorted from benign inflammatory tonsils, AID transcript levels were comparable in both Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL, being as low

I. Al Jamal et al.

as in naïve B cells, regardless of the mutated or unmutated IGHV status (Figure 2C). In agreement with the strong predominance of umCLL in this group, Sμ-3’RRrecHigh CLL were associated with decreased treatment-free survival (TFS) (≈14 months compared to ≈71 months; P<0.001; Figure 2D). In comparison, separating patients into CSRLow and CSRHigh groups did not result in significant differential TFS even if survival curves were separated (Online Supplementary Figure S1E). For this series, TFS also strongly depended on the Binet stage and the IGHV mutation status, and marginally depended on lymphocytosis and cytogenetics (Online Supplementary Figure S5). In order to search for independent variables, a

Figure 2: Enrichment in unmutated chronic lymphocytic leukemia and poor prognosis of chronic lymphocytic leukemia patients μ-3’RRrec counts. (A) Low Sμ-3’RR recombination (Sμ-3’RRrecLow) chronic lymphocytic leukemia (CLL) cases with increased Sμ (N=34) had lower percentages of sequence identity with the reference sequence compared to the high homology of the variable region of the immunoglobulin heavy chain variable region (IGHV) segments in Sμ-3’RRrecHigh CLL (N=12). For each CLL group, the somatic hypermutation rate mean is indicated above the graph. (B) Sequence analysis of PIM1, activation induced-cytidine deaminase (AID) off-target gene. The mutation rate of PIM1 was significantly increased in Sµ-3’RRrecLow patients (N=8) compared to healthy peripheral blood mononuclear cells (PBMC) (N=8) and Sμ-3’RRrecHigh CLL (N=7). (C) AID transcripts, relative to CD19 transcripts, were lower in Sμ-3’RRrecLow CLL (N=7) and Sμ-3’RRrecHigh CLL (N=6) compared to normal B centroblasts (N=4) used as positive controls and comparable to AID transcript levels in sorted naïve B cells (N=4) used as negative controls. Purple dots correspond to mutated IGHV CLL samples. (D) Cumulative survival (Cum survival) time (years) without treatment (treatment-free survival [TFS]) for patients indicated shorter TFS in Sμ-3’RRrecHigh CLL (N=12) than Sμ-3’RRrecLow CLL (N=34). Graphs represent the mean ± standard error of the mean. Statistical analyses were performed using unpaired t test (A, B, C) or Χ2 test (D). 3’RR: 3’ regulatory region; NS: not significant; *P<0.05; **P<0.01; ***P<0.001. Haematologica | 109 February 2024

470

I. Al Jamal et al.

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

Cox univariate analysis was first done for Sμ-3’RRrec status, Binet stage, lymphocytosis, age, IGHV mutation status and cytogenetics (Table 1). A first Cox multivariate model was constructed with variables with a P value <0.2, id est Sμ-3’RRrec status, Binet stage, IGHV mutation status, and cytogenetics (Table 1). In this model, IGHV mutation status and Binet stage were the two independent variables. However, a Cox model including only IGHV mutation and Sμ-3’RRrec status suggested that the confounding variable was the IGHV mutation. Indeed, a second model replacing IGHV mutation status by lymphocytosis pointed on Sμ-3’RRrec status as the sole independent variable (Table 1). Revealing strong overlaps between IGHV mutation and Sμ-3’RRrec status in terms of TFS, these analyses reflect the enrichment Sμ-3’RRrecHigh group in umCLL, which are very well known to have a poor prognosis.6,7

In contrast to poorly-diversified IGHV clonal rearrangements, Sμ μ-3’RRrecHigh chronic lymphocytic leukemia exhibited increases in both Sμ μ-3’RRrec and class switch recombination diversities with increased IGH locus accessibility Sμ-3’RRrec and CSR result from random IGH recombination involving two DNA DSB, one in the Sμ donor region and one in the 3’RR or Sx acceptor region respectively. Since CLL is IgM+, Sμ-3’RR rearrangements have to occur on the non-productive IGH allele; this raises the question of Sμ-3’RRrec clonality. Because CSR can also take place on the non-productive allele, both Sμ-3’RRrec and CSR junction diversities were evaluated using the Shannon index which measures the number of different junctions in sequencing libraries.39 HV were used here as controls of “polyclonal” junctions. Diversities of both Sμ-3’RRrec and

Table 1. Univariate and multivariate analysis of treatment-free survival including biological parameters in the chronic lymphocytic leukemia cases from the study.

Multivariate analysis model 1 (Sµ-3’RRrec status, IGHV mutation status, Binet stage, Cytogenetics)

Univariate analysis

Multivariate analysis model 2 (Sµ-3’RRrec status, lymphocytosis, Binet stage, cytogenetics)

LCI

UCI

LCI

UCI

LCI

UCI

Sµ-3’RRrecLow

Sµ-3’RRrecHigh

5.07

2.08

12.36

0.0004

5.07

2.08

12.36

0.0004

muCLL

umCLL

3.233

1.37

7.631

0.0079

3.554

1.426

8.853

0.0068

3.199

1.368

7.48

0.0073

3.197

1.362

7.506

0.0076

<30 G/L

>30 G/L

1.687

0.7452

3.819

0.2100

1.926

0.8306

4.466

0.1270

<70

>70

1.054

0.4599

2.417

0.9010

Sµ-3’RRrec status

IGHV mutation status

Binet stage A B or C Lymphocytosis

Cytogenetics Normal karyotype or isolated del(13)q Other karyotypes Age in years

Cox multivariate model 1 included variables with a P value below 0.2, Sμ-3’RRrec status, Binet stage, IGHV mutation status (chronic lymphocytic leukemia with mutated IGHV gene [muCLL] or with unmutated IGHV gene [umCLL]), lymphocytosis and cytogenetics. Cox multivariate model 2 included Sμ-3’RRrec status, Binet stage, lymphocytosis and cytogenetics. Statistical analyses were performed usung the survival R packages (see materials and methods). 3’RR: 3’ regulatory region; CLL: chronic lymphocytic leukemia; HR: hazard ratio; LCI: lower confidence interval; UCI: upper confidence interval; P: P value.

Haematologica | 109 February 2024

471

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

two groups (data not shown), coding Cμ transcripts were also increased in Sμ-3’RRrecHigh CLL (Figure 4C). Altogether, our results suggest abnormal intratumoral IGH remodeling activity in the Sμ-3’RRrecHigh CLL group.

CSR junctions were strongly decreased in Sμ-3’RRrecLow CLL samples, a result that should be expected in this clonal IgM+ B-cell cancer. Strikingly, the Shannon diversity index was significantly higher in Sμ-3’RRrecHigh CLL samples than in Sμ-3’RRrecLow CLL (Figure 3A, B). Both the absolute numbers of B cells (Online Supplementary Figure S4) and diversity indexes of CLL IGHV clonal rearrangements (Figure 3C) were similar in both the Sμ-3’RRrecHigh and Sμ-3’RRrecLow groups. CLL IGHV diversity was much lower than DLBCL, known to harbor intra-tumoral subclones with divergent IGHV after SHM and taken here as positive controls of intra-tumor diversity. Therefore, increased Sμ-3’RRrec and CSR junction diversities were not likely to be influenced by B-cell richness but would rather reflect diversification of heavy chain rearrangements in Sμ-3’RRrecHigh CLL. Because increased diversities of Sμ-3’RRrec and CSR junctions in Sμ-3’RRrecHigh CLL are evocative of an on-going process, we evaluated whether IGH locus DNA was accessible to recombination machinery. In order to assess locus aperture, we analyzed the expression of non-coding and coding transcripts from the constant part of the IGH locus (Figure 4A). We found higher levels of Sμ, Sγ1, Sγ3, HS1.2 and HS4 sterile transcripts in Sμ-3’RRrecHigh than in Sμ-3’RRrecLow CLL (Figure 4B), meaning that the IGH locus was accessible to the recombination machinery in these patients. While levels of surface Ig were comparable between the

I. Al Jamal et al.

In contrast with those of class switch recombination, structural features of Sμ μ-3’RRrec junctions harbor an activated B-cell imprint and discriminate both Sμ3’RRrecLow and Sμ μ-3’RRrecHigh chronic lymphocytic leukemia In addition to being unique, joint structure is differently shaped according to the DSBR machinery. In B cells, DSBR occurs mainly through NHEJ and, to a lesser extent, through Alt-EJ pathways for IGH recombination.22,23 The joint structure of each single Sμ-3’RRrec and CSR junction can be determined by alignment to reference sequences. We performed structural analyses of Sμ-3’RRrec and CSR junctions in CLL samples, HV PBMC and benign inflammatory tonsil cells. Here, while circulating B lymphocytes from HV, included because they were exempt of any known disease, were predominantly resting, tonsils were analyzed since they are very well known benign inflammatory lymphoid tissues with highly active B-cell responses and numerous GC, which are the main site of post-medullary Ig gene recombination. CSR joint structures were comparable between HV PBMC and tonsils and were similar to

Figure 3. Intratumoral IGH remodeling activity in the Sμ μ-3’RRrecHigh chronic lymphocytic leukemia group. The Shannon diversity index was used to estimate class switch recombination (CSR), Sμ-3’RR recombination (Sμ-3’RRrec) and intra-clonal immunoglobulin heavy chain variable region (IGHV) diversities. (A) Higher Sμ-3’RRrec junction diversity was observed in Sμ-3’RRrecHigh chronic lymphocytic leukemia (CLL) samples (N=11) compared to Sμ-3’RRrecLow (N=35) CLL and healthy peripheral blood mononuclear cells (PBMC) (N=9). (B) CSR junction diversity was increased in Sμ-3’RRrecHigh CLL samples (N=12) compared to Sμ-3’RRrecLow (N=35) CLL and comparable to those of healthy PBMC (N=11). (C) CLL IGHV diversities were lower than those observed in diffuse large B-cell lymphomas (DLBCL) harboring intra-tumoral subclones with divergent IGHV after somatic hypermutation (SHM) used as positive controls (N=10). Graphs represent the mean ± standard error of the mean. Statistical analyses were performed using unpaired t test. IGH: immunoglobulin heavy chain; 3’RR: 3’ regulatory region; NS: not significant; *P<0.05; **P<0.01; ***P<0.001.

Haematologica | 109 February 2024

472

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

I. Al Jamal et al.

Figure 4. Increased IGH locus accessibility in Sμ μ-3’RRrecHigh chronic lymphocytic leukemia. Quantification of immunoglobulin heavy chain (IGH) locus non-coding transcripts (Sµ, Sγ1, Sγ3, HS1.2 and HS4) relative to those of CD19 (A) and coding transcripts (Cµ and surface immunglobulin M [sIgM]) (B) in Sµ-3’RRrecLow (N=4-7) and Sµ-3’RRrecHigh (N=3-5) chronic lymphocytic leukemia (CLL). Sµ-3’RRrecHigh exhibited high levels of IGH locus transcription in both productive and non-productive transcripts. Graphs represent the mean ± standard error of the mean. Statistical analyses were performed using unpaired t test. 3’RR: 3’ regulatory region; NS: not significant; PBMC: peripheral blood mononuclear cells; *P<0.05; **<0.01. Haematologica | 109 February 2024

473

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

those of CLL regardless of Sμ-3’RRrec status (Figure 5A). In contrast, the structure of Sμ-3’RRrec joints differed between HV PBMC and tonsils. The latter exhibited more Sμ-3’RRrec junctions with small microhomologies (1-2 bp) and blunt junctions while long insertions (≥4 bp) were predominant in the former (Figure 5B). Therefore, Sμ-3’RRrec

I. Al Jamal et al.

joint structures seem to be differently imprinted according to tissue origin and/or B-cell activation. This was not the case for CSR, a strong indication that these two IGH recombination events, even if mechanistically close, are not linked. Strikingly, the structures of Sμ-3’RRrec junctions differed between Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL,

Figure 5. Sμ μ-3’RRrec junction structural features are related to different lymphoid tissue imprints and discriminate between μ-3’RRrecLow and Sμ μ-3’RRrecHigh chronic lymphocytic leukemia. Structures at the Sμ-3’RR recombination (Sμ-3’RRrec) and class Sμ switch recombination (CSR) junctions were determined using CSReport by alignment to reference sequences for chronic lymphocytic leukemia (CLL) samples, healthy volunteer (HV) peripheral blood mononuclear cells (PBMC) and benign inflammatory tonsil cells. Structural features account for length in base pairs (bp) of nucleotide insertions at the joint, of short homology (microhomology) between acceptor and donor sequences and absence of insertions and homology (blunt) (A) Sμ-Sγ1, Sµ-Sγ2, Sμ-Sγ3, Sμ-Sγ4 CSR joint structures were comparable between HV PBMC, benign inflammatory tonsils and CLL. (B) The Sμ-3’RRrec junction structure differed between CLL: Sμ-3’RRrecHigh samples were comparable to HV PBMC and predominantly exhibited junctions with long insertions (≥4 bp) while Sμ-3’RRrecLow CLL and benign inflammatory tonsils were enriched in Sμ-3’RRrec junctions with small microhomologies (1-2 bp) and blunt junctions. Quantification of transcripts coding for actors implicated in double strand break (DSB) repair by quantitative real-time polymerase chain reaction relative to CD19 transcripts (C) by non-homologous end joining (NHEJ) (53BP1, RIF1, Rev7 and LIGIV) and (D) by alternative end-joining (Alt-EJ) (PARP-1, POLθ and LIGIII). Statistical analyses were performed using the Χ2 test (A, B) or unpaired t test (C, D). 3’RR: 3’ regulatory region; NS: not significant; ***P<0.001; ****P<0.0001. Haematologica | 109 February 2024

474

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

the former being close to those of HV PMBC and the latter more similar to benign inflammatory tonsils. Decreased small microhomologies (1-2 bp) and blunt junctions argue against NHEJ involvement in DSBR of Sμ-3’RRrec junctions from both HV PBMC and Sμ-3’RRrecHigh CLL.22,23,40 We, therefore, quantified transcripts coding for actors implicated in the protection of DSB DNA ends and favoring NHEJ (53BP1, Rif1 and Rev7, a Shieldin complex component), NHEJ actor (LIGIV) and Alt-EJ components (PARP-1, POLθ and LIGIII). We did not detect any significant difference in the tested transcripts of NHEJ proteins (Figure 5C) or Alt-EJ actors (Figure 5D) between Sμ-3’RRrecHigh and Sμ-3’RRrecLow. As already suggested by the fact that CSR junction structures were similar between Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL, this suggests that Sμ-3’RRrec junction structural differences observed for both Sμ-3’RRrec CLL groups were not due to imbalances in NHEJ/Alt-EJ actors. Altogether, these results indicate that Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL have different Sμ-3’RRrec imprints, the former being close to recirculating resting B cells and the latter reflecting activated B cells in a benign inflammatory secondary lymphoid organ. Overexpression of c-MYC potentiated Sμ μ-3’RR recombination even in the absence of AID Because IGH locus accessibility is also linked to B-cell activation and proliferation,21 we evaluated the past history of CLL B-cell proliferation by measuring relative telomere length. While homogeneous in HV (mean of 2.14), telomere lengths were very heterogeneous in Sμ-3’RRrecLow patients; 13 of 32 (40%) had long or very long telomeres, indicating that cells underwent few proliferation cycles. In contrast, all but one Sμ-3’RRrecHigh patient homogeneously exhibited telomeres that were shorter than HV, reflecting increased proliferation cycle numbers (Figure 6A). Shorter telomeres in Sμ-3’RRHigh CLL were associated with increased MYC expression (Figure 6B). We thus raised the question of the impact of c-MYC on IGH recombination. For this purpose, we used the murine B-cell lymphoma CH12F3 cell line and its AID knockout (KO) counterpart stably transfected or not with a MYC overexpression vector and stimulated in vitro to undergo CSR and Sμ-3’RRrec. As shown in Figure 6C, the levels of both CSR and Sμ-3’RRrec were increased when MYC was overexpressed in the AID context (AID+MYCtg). In the absence of AID (AIDKO), CSR was undetectable either in absence or presence of MYC overexpression. Some Sμ-3’RRrec junctions were detectable in the absence of both AID and the MYC overexpressing vector. Induction of MYC overexpression resulted in increased numbers of Sμ-3’Rrrec events (Figure 6D). Contamination could be ruled out because sequences of these Sμ-3’RRrec junctions were unique. Some of these junctions contained sequence fragments of Sε and Sγ2 regions between Sμ and 3’RR, which suggests that these Sμ-3’RRrec events were sequentially preceded

I. Al Jamal et al.

by a CSR event. These results show that MYC potentiated both CSR and Sµ-3’RRrec when AID was expressed. Even at low frequency, Sµ-3’RRrec was possible in the absence of AID but in the presence of MYC overexpression.

Discussion In this study, we observed that Sμ-3’RRrec was detectable in CLL patients. Moreover, we showed that the Sμ-3’RRrec rate was increased at levels even higher than in polyclonal HV in one group of CLL cases, the Sμ-3’RRrecHigh group. In the Sμ-3’RRrecHigh group, Sμ-3’RRrec appeared to be on-going, as supported by increased diversity of the Sμ-3’RRrec junctions. Despite low CSR junction counts, CSR diversity was also significantly higher in Sμ-3’RRrecHigh CLL samples than in Sμ-3’RRrecLow CLL meaning that CSR was on-going in these cases. On-going IGH CSR can be observed in a restricted subpopulation of tumoral CLL cells.15,17,41,42 In our study, increased Sμ-3’RRrec and CSR diversities in Sμ-3’RRrecHigh CLLs reflected IGH remodeling in the tumoral B-cell clone. Since CLL cases of this series were all IgM+, Sμ-3’RRrec had occurred on the non-productive IGH allele while CSR could occur on both alleles. CSR mainly occurs on the functional allele.43,44 Here, the numbers of CSR junctions were strongly decreased in CLL without any differences between Sμ-3’RRrecHigh and Sμ-3’RRrecLow CLL. This indicates that despite increased IGH remodeling, Sμ-3’RRrecHigh class-switched CLL cells could have been counterselected. Indeed, as reviewed recently, IgM BCR is a key component of CLL pathogenesis development and evolution, not only for its antigenic recognition properties but also likely through its structure and its signaling capacities.45 Even if the CSR junction structure appeared identical among samples, Sμ-3’RRrecLow and Sμ-3’RRrecHigh CLL samples exhibited different Sμ-3’RRrec structural profiles. In Sμ-3’RRrecHigh CLL, Sμ-3’RRrec junctions were close to those of circulating B cells, while Sμ-3’RRrec junctions from Sμ-3’RRrecLow CLL were similar to those of tonsils B cells. According to Sμ-3’RRrec junction imprints, our results are full agreement with the hypothesis that CLL can be subdivided according to two different COO. Since physiologically most circulating B cells are IgM+IgD+ pre-GC cells and since tonsils contain numerous active GC, we suggest that the imprint of Sμ-3’RRrec junctions would be from circulating nonGC experienced B cells for Sμ-3’RRrecHigh cases (mostly umCLL) while Sμ-3’RRrecLow cases (mostly mCLL) would have Sμ-3’RRrec junction imprints of AID-experienced B cells issued from secondary lymphoid organs. This correlates with IGHV mutational status and PIM1 mutation rate that were decreased in Sμ-3’RRrecHigh CLL and increased in Sμ-3’RRrecLow CLL. Consistent with increased IGH recombination activity, the IGH locus was strongly transcribed in Sμ-3’RRrecHigh CLL and

Haematologica | 109 February 2024

475

I. Al Jamal et al.

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

Figure 6. Increased IGH remodeling in Sμ μ-3’RRrecHigh chronic lymphocytic leukemia cells is potentiated by high levels of c-MYC expression. (A) Relative telomere length (RTL) measured by specific quantitative real time polymerase chain reaction relative to the human β globin gene (Sμ-3’RRrecLow, N=33; Sµ-3’RRrecHigh, N=12; healthy volunteer [HV] peripheral blood mononuclear cells [PBMC], N=6). Telomere length was significantly shorter in the Sµ-3’RRrecHigh group compared to the Sµ-3’RRrecLow group and HV PBMC. (B) c-MYC expression was higher in Sµ-3’RRrecHigh (N=6) compared to Sµ-3’RRrecLow chronic lymphocytic leukemia (CLL) samples (N=7). (C) Detection of class switch recombination (CSR) and Sµ-3’RR recombination (Sµ-3’RRrec) junctions in activated CH12F3 clones overexpressing or not MYC in the presence of activation induced-cytidine deaminase (AID) (CSR: AID+, N=1 and AID+MYCtg, N=3; Sµ-3’RRrec: AID+, N=1 and AID+ MYCtg, N=3) suggested that c-MYC tends to increase CSR and Sµ-3’RRrec counts. This was also observed for Sµ-3’RRrec in the absence of AID (D), as Sµ-3’RRrec junctions, even if rare, were detectable in CH12F3 AIDKO clones (N=2) and appeared to increase with MYC overexpression (AIDKO MYCtg, N=2). No CSR junctions were detected in the absence of AID (AIDKO, N=1), even with MYC overexpression (AIDKOMYCtg, N=3). Graphs represent the mean ± standard error of the mean. Statistical analyses were performed using unpaired t test. KO: knockout; tg: transgenic; 3’RR: 3’ regulatory region; NS: not significant; *P<0.05; **P<0.01; ***P<0.001.

was thus targetable by the IGH recombination machinery. Characteristic of activated and proliferating cells, shorter telomeres in Sμ-3’RRrechigh CLL indicate increased numbers of past mitoses. Upon B-cell stimulation, the MYC gene frequently relocates to the transcription factory occupied by the IGH locus.46 Consistently, we observed increased c-MYC expression levels in Sμ-3’RRrecHigh CLL. This result led us to evaluate the impact of c-MYC on CSR and Sμ-3’RRrec. We found that c-MYC overexpression potentiated both CSR and Sμ-3’RRrec recombination in the presence of AID. Moreover, c-MYC induced increases in Sμ-3’RRrec counts in the absence of AID. It has been recently shown that some residual CSR can occur in absence of the AID.47 Due

to specific constitutively occuring IGH loop conformation between 3’RR, Eμ and Sμ in mature B cells,48 DNA segments from both Sμ and 3’RR are likely to be in close proximity and to be recombined together when DSB-targeted even in the absence of AID. Regardless of the AID status of CLL cells, this c-MYC effect contributes to genetic instability. Indeed, CLL is known to harbor DNA repair alterations and to accumulate DSB across the genome.49,50 Moreover, both NHEJ and Alt-EJ are good candidate processes for chromosomal material exchange due to their capacity to ligate DNA ends from independent molecular origins. IGHV mutation and Sμ-3’RRrec status exhibited strong overlaps in terms of TFS, which reflects that fact that

Haematologica | 109 February 2024

476

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL

Sμ-3’RRrecHigh group was enriched in umCLL. Here, TFS very likely reflects the different natural histories of the disease, which should be underpinned by biological differences. Separating CLL patients in the CLL Sμ-3’RRrecLow group that would originate from GC-experienced B cells or in the CLL Sμ-3’RRrecHigh group with probable non-GC experienced B cells and close to umCLL, our study characterizes for the first time active IGH recombination potentiated by c-Myc overexpression in Sμ-3’RRrecHigh CLL, even in the absence of AID. Disclosures No conflicts of interest to disclose. Contributions IAJ performed experiments and participated in writing of the original draft. MP, KG, SA, MLG, MD, MB and HB participated in the experiments. DR, JP, MD, NF, FJ, PR and NG participated in data curation. CEH and JL provided tonsils from patients undergoing tonsillectomies performed in Limoges Dupuytren Hospital. SA, SAH and NM participated in writing the original draft. JF, NG and SP led the conceptualization, data curation, funding acquisition and manuscript writing.

I. Al Jamal et al.

and Dr. Vallejo and the INSERM CIC 1435, Dupuytren Hospital, Limoges, France, for providing human samples. We thank Dr. Jeanne Cook Moreau (UMR CNRS 7276 INSERM 1276) for careful English editing. Funding This work was supported by grants to SP from la Ligue Contre le Cancer (Comité de la Haute-Vienne, CD87), the Fondation ARC pour la Recherche sur le Cancer (ARCPJA2022060005138), the INCa-Cancéropôle GSO (Emergence program no. 2019-E11) and grants to CRIBL laboratory from the Institut CARNOT CALYM. IAJ is supported by Fondation pour la Recherche Medicale and Lebanese associations (AZM and Saade, LASeR). MP is supported by Région Nouvelle Aquitaine and Université de Limoges. KG is supported by Région Nouvelle Aquitaine and Délégation INSERM Nouvelle Aquitaine. JP is supported by the Institut CARNOT CALYM. SP is a National Institute of Health and Medical Research (INSERM) investigator. Data-sharing statement Sequencing data produced in this study have been deposited in the National Center for Biotechnology Information’s BioProject (PRJNA830327).

Acknowledgments The authors would like to thank Dr. Villéger at the CRBioLim

References 1. Swerdlow SH, Campo E, Pileri SA, et al. The 2016 revision of the World Health Organization classification of lymphoid neoplasms. Blood. 2016;127(20):2375-2390. 2. Rai KR, Sawitsky A, Cronkite EP, Chanana AD, Levy RN, Pasternack BS. Clinical staging of chronic lymphocytic leukemia. Blood. 1975;46(2):219-234. 3. Binet JL, Auquier A, Dighiero G, et al. A new prognostic classification of chronic lymphocytic leukemia derived from a multivariate survival analysis. Cancer. 1981;48(1):198-206. 4. Baliakas P, Jeromin S, Iskas M, et al. Cytogenetic complexity in chronic lymphocytic leukemia: definitions, associations, and clinical impact. Blood. 2019;133(11):1205-1216. 5. Rossi D, Gaidano G. The clinical implications of gene mutations in chronic lymphocytic leukaemia. Br J Cancer. 2016;114(8):849-854. 6. Damle RN, Wasil T, Fais F, et al. Ig V gene mutation status and CD38 expression as novel prognostic indicators in chronic lymphocytic leukemia. Blood. 1999;94(6):1840-1847. 7. Hamblin TJ, Davis Z, Gardiner A, Oscier DG, Stevenson FK. Unmutated Ig V(H) genes are associated with a more aggressive form of chronic lymphocytic leukemia. Blood. 1999;94(6):1848-1854. 8. Stamatopoulos K, Agathangelos A, Rosenquist R, Ghia P. Antigen receptor stereotypy in chronic lymphocytic leukemia. Leukemia. 2017;31(2):282-291. 9. Kulis M, Heath S, Bibikova M, et al. Epigenomic analysis detects widespread gene-body DNA hypomethylation in chronic lymphocytic leukemia. Nat Genet. 2012;44(11):1236-1242.

10. Kulis M, Merkel A, Heath S, et al. Whole-genome fingerprint of the DNA methylome during human B cell differentiation. Nat Genet. 2015;47(7):746-756. 11. Wierzbinska JA, Toth R, Ishaque N, et al. Methylome-based cell-of-origin modeling (Methyl-COOM) identifies aberrant expression of immune regulatory molecules in CLL. Genome Med. 2020;12(1):29. 12. Klein U, Tu Y, Stolovitzky GA, et al. Gene expression profiling of B cell chronic lymphocytic leukemia reveals a homogeneous phenotype related to memory B cells. J Exp Med. 2001;194(11):1625-1638. 13. Ng A, Chiorazzi N. Potential relevance of B-cell maturation pathways in defining the cell(s) of origin for chronic lymphocytic leukemia. Hematol Oncol Clin North Am. 2021;35(4):665-685. 14. Rosenwald A, Alizadeh AA, Widhopf G, et al. Relation of gene expression phenotype to immunoglobulin mutation genotype in B cell chronic lymphocytic leukemia. J Exp Med. 2001;194(11):1639-1647. 15. Palacios F, Moreno P, Morande P, et al. High expression of AID and active class switch recombination might account for a more aggressive disease in unmutated CLL patients: link with an activated microenvironment in CLL disease. Blood. 2010;115(22):4488-4496. 16. Patten PEM, Chu CC, Albesiano E, et al. IGHV-unmutated and IGHV-mutated chronic lymphocytic leukemia cells produce activation-induced deaminase protein with a full range of biologic functions. Blood. 2012;120(24):4802-4811.

Haematologica | 109 February 2024

477

ARTICLE - c-MYC and Sµ-3’RRrec denote different COO in CLL 17. Oppezzo P, Vuillier F, Vasconcelos Y, et al. Chronic lymphocytic leukemia B cells expressing AID display dissociation between class switch recombination and somatic hypermutation. Blood. 2003;101(10):4029-4032. 18. Morande PE, Yan X-J, Sepulveda J, et al. AID overexpression leads to aggressive murine CLL and nonimmunoglobulin mutations that mirror human neoplasms. Blood. 2021;138(3):246-258. 19. Revy P, Muto T, Levy Y, et al. Activation-induced cytidine deaminase (AID) deficiency causes the autosomal recessive form of the Hyper-IgM syndrome (HIGM2). Cell. 2000;102(5):565-575. 20. Muramatsu M, Kinoshita K, Fagarasan S, Yamada S, Shinkai Y, Honjo T. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell. 2000;102(5):553-563. 21. Stavnezer J, Guikema JEJ, Schrader CE. Mechanism and regulation of class switch recombination. Annu Rev Immunol. 2008;26261-26292. 22. Soulas-Sprauel P, Le Guyader G, Rivera-Munoz P, et al. Role for DNA repair factor XRCC4 in immunoglobulin class switch recombination. J Exp Med. 2007;204(7):1717-1727. 23. Yan CT, Boboila C, Souza EK, et al. IgH class switching and translocations use a robust non-classical end-joining pathway. Nature. 2007;449(7161):478-482. 24. Manis JP, Morales JC, Xia Z, Kutok JL, Alt FW, Carpenter PB. 53BP1 links DNA damage-response pathways to immunoglobulin heavy chain class-switch recombination. Nat Immunol. 2004;5(5):481-487. 25. Ward IM, Reina-San-Martin B, Olaru A, et al. 53BP1 is required for class switch recombination. J Cell Biol. 2004;165(4):459-464. 26. Chapman JR, Barral P, Vannier J-B, et al. RIF1 is essential for 53BP1-dependent nonhomologous end joining and suppression of DNA double-strand break resection. Mol Cell. 2013;49(5):858-871. 27. Di Virgilio M, Callen E, Yamane A, et al. Rif1 prevents resection of DNA breaks and promotes immunoglobulin class switching. Science. 2013;339(6120):711-715. 28. Noordermeer SM, Adam S, Setiaputra D, et al. The shieldin complex mediates 53BP1-dependent DNA repair. Nature. 2018;560(7716):117-121. 29. Pinaud E, Marquet M, Fiancette R, et al. The IgH locus 3’ regulatory region: pulling the strings from behind. Adv Immunol. 2011;11027-11070. 30. Péron S, Laffleur B, Denis-Lagache N, et al. AID-driven deletion causes immunoglobulin heavy chain locus suicide recombination in B cells. Science. 2012;336(6083):931-934. 31. Boutouil H, Boyer F, Cook-Moreau J, Cogné M, Péron S. IgH locus suicide recombination does not depend on NHEJ in contrast to CSR in B cells. Cell Mol Immunol. 2019;16(2):201-202. 32. Dalloul I, Boyer F, Dalloul Z, et al. Locus suicide recombination actively occurs on the functionally rearranged IgH allele in B-cells from inflamed human lymphoid tissues. PLoS Genet. 2019;15(6):e1007721. 33. Lam KP, Kühn R, Rajewsky K. In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell. 1997;90(6):1073-1083. 34. Boyer F, Boutouil H, Dalloul I, et al. CSReport: a new

computational tool designed for automatic analysis of class switch recombination junctions sequenced by high-throughput sequencing. J Immunol. 2017;198(10):4148-4155. 35. Gachard N, Parrens M, Soubeyran I, et al. IGHV gene features and MYD88 L265P mutation separate the three marginal zone lymphoma entities and Waldenström macroglobulinemia/ lymphoplasmacytic lymphomas. Leukemia. 2013;27(1):183-189. 36. Rizzo D, Viailly P-J, Mareschal S, et al. Oncogenic events rather than antigen selection pressure may be the main driving forces for relapse in diffuse large B-cell lymphomas. Am J Hematol. 2017;92(1):68-76. 37. Wohlford EM, Baresel PC, Wilmore JR, Mortelliti AJ, Coleman CB, Rochford R. Changes in tonsil B cell phenotypes and EBV receptor expression in children under 5-years-old. Cytometry B Clin Cytom. 2018;94(2):291-301. 38. Joglekar MV, Satoor SN, Wong WKM, Cheng F, Ma RCW, Hardikar AA. An optimised step-by-step protocol for measuring relative telomere length. Methods Protoc. 2020;3(2):E27. 39. Rosati E, Dowds CM, Liaskou E, Henriksen EKK, Karlsen TH, Franke A. Overview of methodologies for T-cell receptor repertoire analysis. BMC Biotechnol. 2017;17(1):61. 40. Boboila C, Yan C, Wesemann DR, et al. Alternative end-joining catalyzes class switch recombination in the absence of both Ku70 and DNA ligase 4. J Exp Med. 2010;207(2):417-427. 41. Oppezzo P, Magnac C, Bianchi S, et al. Do CLL B cells correspond to naive or memory B-lymphocytes? Evidence for an active Ig switch unrelated to phenotype expression and Ig mutational pattern in B-CLL cells. Leukemia. 2002;16(12):2438-2446. 42. Oppezzo P, Navarrete M, Chiorazzi N. AID in chronic lymphocytic leukemia: induction and action during disease progression. Front Oncol. 2021;11:634383. 43. Perlot T, Alt FW, Bassing CH, Suh H, Pinaud E. Elucidation of IgH intronic enhancer functions via germ-line deletion. Proc Natl Acad Sci U S A. 2005;102(40):14362-14367. 44. Sakai E, Bottaro A, Alt FW. The Ig heavy chain intronic enhancer core region is necessary and sufficient to promote efficient class switch recombination. Int Immunol. 1999;11(10):1709-1713. 45. Young RM, Phelan JD, Wilson WH, Staudt LM. Pathogenic B cell receptor signaling in lymphoid malignancies: new insights to improve treatment. Immunol Rev. 2019;291(1):190-213. 46. Osborne CS, Chakalova L, Mitchell JA, et al. Myc dynamically and preferentially relocates to a transcription factory occupied by Igh. PLoS Biol. 2007;5(8):e192. 47. Dalloul I, Laffleur B, Dalloul Z, et al. UnAIDed Class switching in activated B-cells reveals intrinsic features of a self-cleaving IgH locus. Front Immunol. 2021;12:7374. 48. Wuerffel R, Wang L, Grigera F, et al. S-S synapsis during class switch recombination Is promoted by distantly located transcriptional elements and activation-induced deaminase. Immunity. 2007;27(5):711-722. 49. Deriano L, Guipaud O, Merle-Béral H, et al. Human chronic lymphocytic leukemia B cells can escape DNA damage-induced apoptosis through the nonhomologous end-joining DNA repair pathway. Blood. 2005;105(12):4776-4783. 50. Popp HD, Flach J, Brendel S, et al. Accumulation of DNA damage and alteration of the DNA damage response in monoclonal B-cell lymphocytosis and chronic lymphocytic leukemia. Leuk Lymphoma. 2019;60(3):795-804.

Haematologica | 109 February 2024

478

I. Al Jamal et al.

ARTICLE - Chronic Lymphocytic Leukemia

Global miRNA profiling reveals key molecules that contribute to different chronic lymphocytic leukemia incidences in Asian and Western populations Panpan Liu,1,2* Kefeng Wang,3,4* Jianan Li,1,5* Marcia A. Ogasawara,6 Zhongjun Xia,1,5 William G. Wierda,7 Michael J. Keating,7 Yiqing Li,3,8# and Peng Huang1#

Correspondence: P. Liu liupp@sysucc.org.cn

State Key Laboratory of Oncology in South China, Collaborative Innovation Center for

Y. Li liyiqing@mail.sysu.edu.cn

Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou, China; 2Department of Medical Oncology, Sun Yat-sen University Cancer Center, Guangzhou, China; 3Guangdong Provincial Key Laboratory of Malignant Tumor Epigenetics and Gene Regulation, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China; 4Department of Thoracic Surgery, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China; Department of Hematologic Oncology, Sun Yat-sen University Cancer Center, Guangzhou,

P. Huang huangpeng@sysucc.org.cn Received: Accepted: Early view:

March 21, 2023. August 14, 2023. August 31, 2023.

China; 6Department of Translational Molecular Pathology, the University of Texas MD Anderson Cancer Center, Houston, TX, USA; 7Department of Leukemia, the University of Texas MD Anderson Cancer Center, Houston, Texas, USA and 8Department of Hematology,

Sun Yat-Sen Memorial Hospital, Sun Yat-Sen University, Guangzhou, China PPL, KFW and JNL contributed equally as first authors.

YQL and PH contributed equally as senior authors.

Abstract It has been known for decades that the incidence of chronic lymphocytic leukemia (CLL) is significantly lower in Asia than in Western countries, but the reason responsible for this difference still remains a major knowledge gap. Using GeneChip® miRNA array to analyze the global microRNA expression in B lymphocytes from Asian and Western CLL patients and healthy individuals, we have identified microRNA with CLL-promoting or suppressive functions that are differentially expressed in Asian and Western individuals. In particular, miR-4485 is upregulated in CLL patients of both ethnic groups, and its expression is significantly lower in Asian healthy individuals. Genetic silencing of miR-4485 in CLL cells suppresses leukemia cell growth, whereas ectopic expression of miR-4485 promotes cell proliferation. Mechanistically, miR-4485 exerts its CLL-promoting activity by inhibiting the expression of TGR5 and activating the ERK1/2 pathway. In contrast, miR-138, miR-181a, miR181c, miR-181d, and miR-363 with tumor-suppressive function are highly expressed in Asian healthy individuals. Our study suggests that differential expression of several important microRNA with pro- or anti-CLL functions in Asian and Western B lymphocytes likely contributes to the difference in CLL incidence between the two ethnic groups, and that miR-4485 and its downstream molecule TGR5 could be potential therapeutic targets.

Introduction Chronic lymphocytic leukemia (CLL) is a malignancy characterized by the accumulation of mature but dysfunctional B lymphocytes with abnormal expression of CD5.1-3 CLL is the most common form of adult leukemia in the United States and western European countries, but is rare in the Asian population.4-10 Although the exact reasons for such a substantial difference in disease incidences remain unclear, it appears that genetic rather than environmental factors are the most likely mechanistic explanation. Genetic con-

tribution to CLL incidence was previously suggested by observational studies showing that individuals from Japan who settled in Hawaii did not exhibit any increase in CLL incidence.11,12 Further evidence supporting the strong genetic impact on CLL incidence was provided by a study which analyzed the Los Angeles County Population-based Cancer Registry and showed that Asians (Chinese, Japanese, Filipinos, and Koreans) had a significantly lower incidence of CLL compared to non-Hispanic whites in the same geographic region.13 Interestingly, the same study also revealed that the birthplace or socioeconomic status did not account for the

Haematologica | 109 February 2024

479

ARTICLE - MicroRNA affects CLL incidence in different races

difference in CLL incidence, and that CLL incidence among the offspring of the Asian immigrants remained significantly lower compared to the local non-Hispanic whites. These observations together suggest that genetic factors are mainly responsible for this difference in CLL incidences, but the key molecular players remain to be identified. MicroRNA (miRNA) are non-coding, single-stranded RNA (usually 20-23 nucleotides) with diverse biological functions. They can modulate the expression of genes at the post-transcriptional level and are involved in cancer, apoptosis, and cell metabolism.14-16 In CLL cells, miRNA has been shown to interact with BCL2 (B-cell lymphoma-2) and TP53 (tumor protein p53) and might play a role in CLL pathogenesis.17,18 In a study of 56 patients with CLL/SLL (small lymphocytic lymphoma), an overexpression of two miRNA, miR-21 and miR-155, was seen in most samples analyzed.19 Loss of miR15a and miR-16-a, associated with a deletion of chromosome 13q, could lead to leukemogenesis.20,21 A study found that the expression of miR-15a, miR-16-1, miR-181a, miR-181b, and miR-29b in Chinese CLL patients was lower than in healthy donors.22 Also, the expression of miR-29b and miR-181a/b was significantly correlated to the immunoglobulin heavychain variable region (IGHV) mutational status,18,23 which is a strong prognostic indicator for CLL. Another study found that low expression of miR-223, miR-29c, miR-29b and miR-181, along with unmutated IGHV and high expression of ZAP-70 (Ζ chain-associated protein kinase 70) was associated with CLL progression.18 Thus, it is evident that miRNA play an important role in CLL development, but it remains unclear if there are differentially expressed miRNA that could account for the difference in CLL incidence in Asia and western countries. Interestingly, analysis of miRNA expression in lymphoblastoid cell lines derived from Northern/Western European and from Nigeria Yoruba ethnic groups revealed a significant difference in miRNA expression between these two ethnic groups with an interesting correlation with drug sensitivity phenotypes.24 Based on the observations that miRNA could significantly affect CLL development and disease progression, and that population differences in miRNA expression have been detected in different ethnic groups, we hypothesized that the B lymphocytes from Asian and Western individuals might have differential expression of certain key miRNA which play an important role in B-cell survival and proliferation, thus contributing to the difference in CLL incidence in these two ethnic groups. This study was designed to test this possibility by using B lymphocytes isolated from Asian and Western individuals for miRNA analysis and functional study.

Methods Primary B lymphocytes and cell lines In the present study, CLL peripheral blood samples of Asian and Western CLL patients were obtained from Sun Yat-

P. Liu et al.

sen University Cancer Center (Guangzhou, China) and the University of Texas MD Anderson Cancer Center (Texas, USA), respectively. All patients were diagnosed according to the National Cancer Institute Criteria.25 Informed consents under research protocols approved by the Institutional Review Board (IRB) of Sun Yat-sen University Cancer Center and MD Anderson Cancer Center were obtained from all patients before the collection of blood samples. CLL peripheral blood mononuclear cells (PBMC) samples with leukemia cells ≥70% were selected for this study. CLL cells were isolated from blood samples as previously described.26 Normal PBMC were isolated from the buffy coats of blood samples from normal healthy individuals who donated blood at the regional blood banks. These samples were obtained in an anonymous fashion without personal information. After Ficoll-PaqueTM PLUS (GE Healthcare Bio-Sciences, Uppsala, Sweden) gradient centrifugation, normal CD19+ B cells were purified using CD19 MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany). Direct purification of CD19+ B cells from PB was accomplished using EasySep™ Direct Human B-Cell Isolation Kit (Stem Cell Technologies, Cat #19674). The purity of B lymphocytes was analyzed by flow cytometry to confirm that more than 95% of the purified cells were CD19+/CD3- (Online Supplementary Figure S1A). Normal blood samples were analyzed by sensitive flow cytometry to confirm their negativity for monoclonal B lymphocytosis (Online Supplementary Figure S1B-D). Two previously published CLL cell lines (MEC1 and MEC2)27 were cultured in RPMI 1640 with glutamine (Invitrogen, Carlsbad, USA) supplemented with 10% fetal bovine serum. HEK293T cells were cultured in DMEM (Invitrogen, Carlsbad, USA) supplemented with 10% fetal bovine serum. Microarray assays and data analysis Total RNA was extracted from CD19+ cells of CLL and healthy donors using TRIzol reagent (Invitrogen, USA) according to the manufacturer’s instructions. RNA was subjected to analysis of microRNA expression using the Affymetrix GeneChip® miRNA 3.0 Array. The detail procedures are provided in the Online Supplementary Appendix. The differences between samples were analyzed using the SAM (significance analysis of microarray) R package. The criteria for identifying differential expression between miRNA were a q value <0.05 and more than 2-fold change (FC >2.0 or <0.5). The microarray datasets of miRNA expression of this study have been deposited in the Gene Expression Omnibus database (GEO, GSE216258). Comparison analyses were performed using samples from four different groups including two CLL groups (Asian and Western) and two normal control groups (healthy donors from healthy Asian or Western individuals). Details of data comparisons and statistical analyses are described in the respective figure legends and in the Online Supplementary Appendix.

Haematologica | 109 February 2024

480

ARTICLE - MicroRNA affects CLL incidence in different races

Vectors and oligonucleotides transfection Lentiviral miArrestTM miRNA inhibitor vector for miRNA-4485 (Cat. # HmiR-AN2125-AM03, Genecopoeia, Guangzhou, China) and miExpressTM miRNA expression precursor for miR-4485 (Cat. # HmiR1168-MR03, Genecopoeia) were transduced into proper host cells. The transduced cells were selected with puromycin (Invivogen, San Diego, USA) or hygromycin (Invitrogen, Carlsbad, CA, USA) for 2-3 weeks to obtain cells with stable overexpression or low expression of miR-4485. MicrON miR-4485 mimic (dsRNA oligo, Cat. # miR10019019) and micrOFF miR-4485-3p inhibitor (single-stranded oligonucleotide, Cat. # miR2160621085914) were obtained from RiboBio Co., Ltd (Guangzhou, China). The miRNA mimic and inhibiting oligos were transfected into cells using a riboFect™ CP Transfection Kit as described in the Online Supplementary Appendix. Detailed description of experimental procedures, data analysis, and statistical test is provided in the Online Supplementary Appendix and detailed information on antibodies used in the western blot is provided in Online Supplementary Table S5.

Results Differential microRNA expression in primary chronic lymphocytic leukemia cells versus normal B-lymphocytes in Asian and Western individuals The study design to reveal the potential differences in miRNA expression in primary CLL cells versus normal B-lymphocytes of Asian and Western individuals is illustrated in Figure 1A. In order to compare the microRNA expression profiles in the B lymphocytes and identify the miRNA that were differentially expressed in these two ethnic groups, total RNA was purified from B lymphocytes isolated from the blood samples of the following four groups of individuals: healthy Asians (n=9), healthy Western individuals (n=6), Asian CLL patients (n=22 including six patient samples in the initial GeneChip miRNA array analysis and 16 additional samples in the validation study), and Western CLL patients (n=9). The demographic and clinicopathological information of the CLL patients is summarized in the Online Supplementary Table S1. The global miRNA expression of each individual sample was profiled using the Affymetrix GeneChip® miRNA 3.0 Array. Using the t-distributed stochastic neighbor embedding (t-SNE) and unsupervised hierarchical clustering heatmap visualization, we showed that the four groups exhibited clearly distinguishable miRNA expression profiles (Figure 1B, C; Online Supplementary Figure S2A-C). Uniform manifold approximation and projection (UMAP) analysis also showed that each group had its own UMPA distribution pattern (Online Supplementary Figure S2D). Multivariate analyses including unsupervised principal component anlysis and supervised orthogonal partial least-squares discrimination analysis (OPLS-DA) revealed clear differ-

P. Liu et al.

ences in miRNA expression in term of separation trend and clustering among the four groups (Figure 1D; Online Supplementary Figure S3A, B, D, E, G, H, J, K). We also used a permutation test with 200 random runs permutations to evaluate the reliability of our analyses. As shown in Figure 1D (right panel) and the Online Supplementary Figure 3C, F, I, L, the intercept of the Q2 regression line with the Y-axis was less than 0, indicating the OPLS-DA model was reliable and not overfitted. Figure 1E shows a significant difference in miRNA expression in the primary leukemia cells from six Asian CLL patients and normal B lymphocytes from nine Asian healthy donors, using a FC>2 and a q value <0.05 as the criteria for statistical significance. The detail miRNA expression data and q values are shown in Online Supplementary Table S2. Similarly, the common miRNA expressed in the primary leukemia cells from nine Western CLL patients were also significantly different from the common miRNA expressed in the normal B lymphocytes from the healthy Western donors (Figure 1F; Online Supplementary Table S3). These data together showed that the differentially expressed miRNA between CLL and normal lymphocytes of the Asian samples were substantially different from that of the Western samples. Some of these differentially expressed miRNA such as miR-1295 and miR-4485 appeared in both ethnic groups (indicated by the red arrows in Figure 1E, F). These miRNA, either common in two ethnic groups or unique to a specific group, were further analyzed to identify candidate miRNA that might potentially contribute to the different CLL incidences between the two ethnic groups. Identification of microRNA that potentially contribute to low chronic lymphocytic leukemia incidence in Asians In order to identify the miRNA that might contribute to different CLL incidences in Asian and Western groups, we used Venn diagram and heatmap to visualize the differential miRNA expression profiles in normal lymphocytes and CLL cells of each ethnic group, using miRNA expression in normal B lymphocytes from healthy Western donors as the control for comparison against the other three groups. Because CLL incidence is high in the Western population, we reasoned that if a miRNA functioned to promote CLL development, it would be highly expressed in the CLL samples of both ethnic groups and would also be relatively lower in the normal lymphocytes of the Asian group compared to that of the Western group (low pro-CLL factor in Asians). Conversely, if a miRNA functioned as a CLL suppressor, its expression would be low in the CLL samples of both ethnic groups and also relatively low in the normal lymphocytes of the Western group compared to that of the Asian group. Using this analytical logic and a 2-FC (FC >2.0 or <0.5) in miRNA expression (increase or decrease) with a q value <0.05 as the cutoff criteria, we identified only a single microRNA (miR-4485) that met the criteria as a potential molecule that could promote CLL development but with relatively

Haematologica | 109 February 2024

481

P. Liu et al.

ARTICLE - MicroRNA affects CLL incidence in different races

Figure 1. GeneChip® microRNA array analysis of microRNA expression profiles in primary chronic lymphocytic leukemia cells and normal B lymphocytes from Asian and Western individuals. (A) Schematic illustration of the study design, experimental approaches, and data analyses. (B) T-distribution stochastic neighbor embedding (t-SNE) plot showing sample clustering according to microRNA (miRNA) expression profiles. (C) Unsupervised hierarchical clustering of miRNA in different groups of samples. Each row indicates a miRNA, each column indicates a sample. The miRNA-clustering tree is on the left. (D) Score plots of the principal component analysis (PCA) model were set up using data from Asian CLL (A-C) (green plots) and Asian Normal (A-N) (blue plots), Western Normal (W-N) (red plots), Western CLL (W-C) (yellow plots), R2X=0.585; the score plots of the orthogonal partial least-squared discrimination analysis (OPLS-DA) models discriminated the indicated four groups, and was confirmed by permutation test (PT). R2X=0.854, R2Y=0.976, Q2=0.852. (E) Heatmap revealing the differentially expressed miRNA in chronic lymphocytic leukemia (CLL) cells versus normal B lymphocytes from Asian individuals, using 2-fold differential expression with a q value <0.05 as the cutoff values. (F) Heatmap showing the differentially expressed miRNA in CLL cells versus normal B lymphocytes from Western individuals. The miRNA with at least 2-fold differentially expressed levels with a q value <0.05 are shown. Haematologica | 109 February 2024

482

P. Liu et al.

ARTICLE - MicroRNA affects CLL incidence in different races

Table 1. Quantitation of changes in microRNA expression in chronic lymphocytic leukemia cells and normal B lymphocytes from Asian and Western individuals. MicroRNA name

FC (q value) Asian-CLL Asian-N

FC (q value) Western-CLL Western-N

FC (q value) Asian-N Western-N

FC (q value) Asian-CLL Western-CLL

miR-4485

9.49 (<0.001)

2.34 (0.008)

0.01 (<0.001)

0.05 (<0.001)

miR-181d

0.24 (0.007)

0.43 (0.030)

6.85 (<0.001)

3.86 (<0.001)

miR-181c

0.13 (0.007)

0.10 (<0.001)

2.98 (<0.001)

3.89 (0.005)

miR-363

0.22 (0.007)

0.16 (<0.001)

2.67 (<0.001)

3.66 (0.003)

miR-138

0.06 (<0.001)

2.52 (0.006)

2.00 (0.040)

miR-181a

0.11 (0.001)

0.08 (<0.001)

2.07 (0.006)

1.48 (0.072)

Asian-CLL: microRNA expression in chronic lymphocytic leukemia (CLL) cells from Asian patients; Asian-N: microRNA expression in normal B lymphocytes from healthy Asians; Western-CLL: microRNA expression in CLL cells from Western patients; Western-N: microRNA expression in normal B lymphocytes from healthy Westeners. FC: fold change. The values in the table show the ratios of miR expression between the indicated ethnic samples, and the numbers in paratheses indicate the respective q values.

low expression in Asians (Figure 2A). Among the 84 miRNA with low expression in the normal B lymphocytes of Asian origin, only miR-4485 was highly expressed in CLL cells of both ethnic groups. Of note, similarly, although miR-1295 was upregulated in CLL cells of both ethnic groups as indicated in Figure 1E, F by the red arrows, its expression in Asian normal B lymphocytes was similar to that of the Western samples (Figure 2E, left panel; Online Supplementary Tables S2 and S3), suggesting that miR-1295 was a potential pro-oncogenic molecule but could not account for the low CLL incidence in Asians. The Venn diagram also revealed five miRNA (miR-181d, miR-181c, miR-363, miR-138 and miR-181a) that met the criteria as potential CLL suppressors with low expression in CLL cells of both ethnic groups and higher expression in the normal B lymphocytes from healthy Asian individuals compared with the normal B lymphocytes from healthy Western individuals (Figure 2B). The expression profiles of these six miRNA in each group were further analyzed using volcano plots and visualized by heatmaps (Figure 2C-E). The results revealed that miR-4485 was upregulated and miR-181d, miR-181c, miR-363, miR-138, miR-181a were downregulated in all CLL samples of both ethnic groups (Figure 2C, D). Importantly, miR-4485 was significantly downregulated, while miR-181d, miR-181c, miR-363, miR-138, and miR-181a were highly elevated in the normal B lymphocytes from Asian healthy individuals (Figure 2E). The quantitative data for these comparisons and the respective q values are summarized in Table 1, which shows the ratios of miR expression in CLL cells/normal B lymphocytes in the two ethnic groups and the ratios of miR expression in Asian/ Western normal B lymphocytes and CLL patients. Of special note, the expression of miR-4485 was 9.5-fold and 2.3-fold higher in Asian and Western CLL patients compared to the respective normal B lymphocytes of the same ethnic groups, whereas its expression in the normal lymphocytes from Asian healthy individuals was 83-fold lower than that in

the normal lymphocytes from healthy Western individuals. High expression of miR-4485 promotes chronic lymphocytic leukemia cell proliferation Quantitative reverse-transcription polymerase chain reaction (qRT-PCR) was first used to confirm the low expression of miR-4485 in Asian normal B lymphocytes and its upregulation in CLL cells, using samples from nine healthy Asian individuals and 12 Asian CLL patients. As shown in Figure 3A, miR-4485 expression was low in primary normal B lymphocytes and significantly increased in the CLL samples (P<0.05; Figure 3A, left panel). The expressions of miR-138, miR-181c, miR-363, miR-181a and miR-181d which was found to be upregulated in normal lymphocytes in miRNA array analysis was also validated using qRT-PCR analysis. The results also showed that the levels of miR-138, miR-181c, miR-363, miR-181a and miR-181d were significantly higher in the Asian normal B lymphocytes than in the primary CLL cells from Asian patients (Figure 3A, right 5 panels), similar to that observed in miRNA array analysis and consistent with their presumed tumor suppressor function. Consistently, analysis of two publicly available datasets GSE108901 and GSE66186, which contained miRNA expression data mainly from Western CLL patients and Western normal individuals, also showed higher expression of pro-CLL miR-4485 and lower expression of the CLL-suppressive miRNA in CLL patients compared with healthy individuals (Online Supplementary Figure S4). Since the five miRNA identified in Figure 2B (miR-181d, miR181c, miR-363, miR-138, and miR-181a) with putative anti-CLL function had been reported previously,23,28-34 whereas miR-4485 was the only microRNA identified as a putative CLL-promoting molecule with high expression in CLL cells in both ethnic groups and a downregulation in Asian normal B lymphocytes, we thus focused our efforts on investigating the role of miR-4485. In order to investigate the functional impact of miR-4485 on CLL cells, we first used a lentiviral

Haematologica | 109 February 2024

483

ARTICLE - MicroRNA affects CLL incidence in different races

P. Liu et al.

Figure 2. Identification of microRNA differentially expressed in chronic lymphocytic leukemia cells and normal B lymphocytes from Asian and Western individuals. (A) Venn diagram revealing the differentially expressed microRNA (miRNA) that exhibited at least 2-fold difference (ratio >2 or <0.5) in miRNA expression with a q value <0.05 for the indicated comparison groups: Asian chronic lymphocytic leukemia (CLL) (A-C) versus Asian Normal (Norm) (A-N); Western-CLL (W-C) versus Western-Norm (W-N); Asian-Norm (A-N) versus Western-Norm (W-N). The red color represents the elevated miRNA expression with a ratio >2.0 for the indicated groups, while the green color shows lower expression with a ratio <0.5 for the indicated groups. The red number “1” in the center of the Continued on following page.

Haematologica | 109 February 2024

484

ARTICLE - MicroRNA affects CLL incidence in different races

P. Liu et al.

diagram indicates the only miRNA (miR-4485) that was highly expressed in all CLL samples of both Asian and Western groups but was significantly lower in Asian normal B lymphocytes compared with Western normal B lymphocytes. (B) Venn diagram showing the differentially expressed miRNA that exhibited at least 2-fold difference (ratio >2.0 or <0.5) in miRNA expression with a q value <0.05 for the comparison groups. The color codes are the same as in (A). The blue number “5” in the center of the diagram indicates 5 miRNA that were downregulated in CLL cells of both Asian and Western groups but highly expressed in Asian normal B lymphocytes compared with Western normal B lymphocytes. (C-E) Left panels: volcano plots showing the differentially expressed miRNA in Asian-CLL compared to Asian-Norm (C), Western-CLL compared to Western-Norm (D), and Western-Norm compared to Asian-Norm (E). Differentially expressed miRNA were identified that meet fold change (FC) >2 (log2 FC>1) and P<0.05 (-log10 P>1.30) calculated using two-tailed Student’s t test. Significant upregulated and downregulated miRNA are respectively indicated with red and blue points that exhibit at least 2-fold change (x axis; log2 FC>1) and statistical significance (y axis; -log10 P>1.30). The top selected miRNA are marked with annotation on the plot; right panels: heatmaps illustrating the expression of the single microRNA (miR-4485) identified in (A) and the five miRNA (miR-181d, miR-181c, miR-363, miR-138, and miR-181a) identified in (B) in Asian CLL cells and Asian normal lymphocytes (C), in Western CLL cells and Western normal lymphocytes (D), and in Asian and Western normal B lymphocytes (E). N: normal B lymphocytes; C: CLL; A: Asian; W: Western.

vector (pEZX-AM03) expressing miR-4485 antagonist to transfect two CLL cell lines, MEC1 and MEC2 (both with a high level of miR-4485), and then tested the effect of miR-4485 inhibition on cell proliferation at 24 hour (hrs), 48 hrs, and 72 hrs. There was a significant decrease in cell proliferation after MEC1 or MEC2 cells were transfected to express miR-4485 antagonist, whereas transfection using the control vector did not affect cell proliferat6ion (Figure 3B). The effect of miR-4485 on CLL cell proliferation was further tested using miR-4485 mimic or inhibiting oligos in MEC1 and MEC2 cells. As shown in Figure 3C, cell proliferation was significantly enhanced by miR-4485 mimic compared to the oligo control, whereas the miR-4485 inhibitor significantly reduced cell proliferation in both cell lines. We also used HEK293T cell line with a low level of endogenous miR-4485 for infection with a lentivirus containing miR-4485 overexpression vector (pEZX-MR03) or with the control vector. The increased expression of miR-4485 in HEK293T cells led to a moderate but significant increase in cell proliferation (P<0.01; Figure 3D, E). Consistently, miR-4485 mimic also significantly increased HEK293T cell proliferation, whereas miR-4485 inhibitor markedly reduced cell growth (Figure 3F). Suppression of TGR5 and activation of ERK1/2 pathway by miR-4485 In order to investigate the potential mechanisms by which miR-4485 promoted cell proliferation that might contribute to CLL development, we first performed target prediction analysis using multiple web tools (TargetScan, miRwalk, miRPathDB, miRDB) to generate a list of potential target genes by each web tool, and used Venn diagram to reveal the common candidate genes shared by all four sets of analyses (Figure 4A). We found six genes (ASB7, CTDSPL2, FBXO32, JOSD1, MED1, TGR5) that were likely regulated by miR-4485. qRT-PCR was then used to measure the expression of these six genes in CD19+ normal B lymphocytes purified from healthy Asian donors (n=7) in comparison with primary CLL cell isolated from Asian patients (n=11). We found that only TGR5 gene expression was significantly suppressed in CLL cells compared with normal lymphocytes (Figure 4B),

whereas the expression levels of the other five genes were similar in normal lymphocytes and leukemia cells (Online Supplementary Figure S5A-E). We also compared the expression of these six genes using a dataset from Gene Expression Omnibus (GEO, GSE66117), which contained RNA expression data from 45 CLL patients and five normal individuals. The results consistently showed that TGR5 expression was lower in CLL cells than in normal lymphocytes, whereas the expression of the other five genes was similar in CLL and normal lymphocytes (Figure 4C; Online Supplementary Figure S5F-J; Online Supplementary Table S4). Based on these observations, we focused our study on the potential regulation of TGR5 by miR-4485, using a luciferase reporter assay in which HEK293T cells were co-transfected with a miR-4485 mimic and a Dual-Glo® luciferase assay system containing either the wild-type 3’UTR (untranslated region) of TGR5 with a miR-4485 binding sequence or with a mutated miR-4485 binding sequence (Figure 4D). The results showed that miR-4485 expression significantly suppressed the luciferase activity in cells containing the wild-type miR4485 binding sequence, whereas such inhibition was not observed in cells transfected with the mutated miR-4485 binding sequence (Figure 4D). In order to further test the ability of miR-4485 to downregulate TGR5 expression, we transfected HEK293T cells with a lentiviral-mediated miR-4485 overexpression vector, and measured TGR5 mRNA and protein expression by qRTPCR and western blot analyses, respectively. The results showed that overexpression of miR-4485 caused a significant decrease in TGR5 mRNA and protein (Figure 5A, B). Conversely, transfection of HEK293T cells with a lentiviral vector expressing a miR-4485 antagonist (miR-4485 inhibitory sequence) led to a significant increase in TGR5 mRNA and protein expression (Figure 5C). In order to further test if miR-4485 could regulate TGR5 expression in CLL cells, MEC1 and MEC2 cells were transfected with miR-4485 mimic, miR-4485 inhibitor, or their respective control oligos, and their effect on TGR5 expression was measured. The results showed that miR-4485 mimic caused a decrease in TGR5 protein, whereas miR-4485 inhibitor enhanced TGR5 protein expression (Figure 5D; Online Supplementary Figure

Haematologica | 109 February 2024

485

P. Liu et al.

ARTICLE - MicroRNA affects CLL incidence in different races

S6A). The upregulation of TGR5 mRNA and protein was also consistently observed in primary CLL cells transfected with miR-4485 inhibitor (Figure 5E). Since the roles of miR-4485 and its target gene TGR5 in CLL development largely remained unknown, we then ex-

amined the potential effect of miR-4485 on the expression of molecules such as STAT3, Bcl-2, Bcl-xL, MEK1/2 , ERK1/2 and c-Jun which are important for CLL cell survival and proliferation. Among all the molecules tested, overexpression of miR-4485 in HEK293T cells caused a decrease in TGR5 pro-

Figure 3. Quantitative analysis of key microRNA expression and the impact of miR-4485 on cell proliferation. (A) Expression of miR-4485, miR-138, miR-181c, miR-181d, miR-363, miR-181a in primary chronic lymphocytic leukemia (CLL) cells from patients and normal B lymphocytes from Asian individuals. The levels of microRNA (miRNA) were quantified by quantitative reverse-transcription polymerase chain reaction (qRT-PCR) and presented as mean ± standard deviation (SD) (Norm, N=9; CLL, N=12 for analysis of miR-4485, miR-138 and miR-181c; Norm, N=6; CLL, N=10 for miR-181d, miR-363 and miR-181a). (B) Effect of miR-4485 inhibition on CLL cell growth. MEC1 cells and MEC2 cells were transfected with a miArrest™ miRNA inhibitor vector (pEZX-AM03-based) containing miR-4485 antagonist sequence or the control vector as indicated, and cell numbers were counted at 24 hours (hrs), 48 hrs, and 72 hrs. (C) MEC1 and MEC2 cells were transiently transfected with miR-4485 mimic or inhibitor oligos and their corresponding negative control oligos; 48 hrs after transfection, cell numbers were counted every 24 hrs. (D, E) HEK293T cells were infected with lentiviral containing miR-4485 expression vector (pEZX-MR03-based) or the control vector, and the level of miR4485 expression was measured by qRT-PCR (D) and cell numbers were quantified after 24 hrs, 48 hrs, and 72 hrs (E). (F) HEK293T cells were transiently transfected with miR-4485 mimic or inhibitor oligonucleotides and their corresponding negative controls, 48 hrs after transfection, cell numbers were counted every 24 hrs. Two-way ANOVA analysis was used to determine the significance of the proliferation differences between 2 groups. The mean value ± SD of 3 experiments is shown. **P<0.01; ***P<0.001. Haematologica | 109 February 2024

486

P. Liu et al.

ARTICLE - MicroRNA affects CLL incidence in different races

tein expression and an increase in ERK1/2 phosphorylation without an increase in total ERK1/2 protein. The expression of phosphorylated STAT3, total STAT3 protein, Bcl-2, and BclxL remained unchanged (Figure 5B). Conversely, inhibition of miR-4485 consistently led to an upregulation of TGR5 mRNA and protein (Figure 5C), a decrease of phosphorylated ERK1/2 and downstream c-Jun and p-c-Jun proteins in MEC-1 and MEC-2 cells, while transfection of miR-4485 mimic inhibited TGR5 expression and activated ERK1/2 and its downstream p-c-Jun and c-Jun (Figure 5D; Online Supplementary Figure S6A). Inhibition of miR-4485 in primary CLL cells also decreased phosphorylation of ERK1/2 and c-Jun (Figure 5E). In

order to evaluate the relationship between TGR5 expression and ERK1/2 activation, we first tested the effect of ERK1/2 specific inhibitor SCH772984 on TGR5 expression in MEC1 and MEC2 cell lines and in primary CLL cells, and found that inhibition of ERK1/2 led to an increased expression of TGR5 (Figure 6A, Online Supplementary Figure S6B), suggesting a negative regulation of TGR5 by ERK1/2. Importantly, TGR5 overexpression in MEC1 and MEC2 cells or activation of TGR5 with a TGR5 agonist treatment in primary CLL cells caused a decrease in ERK1/2 phosphorylation (Figure 6B, Online Supplementary Figure S6C, D), suggesting an inhibitory effect of TGR5 on ERK1/2 activation. This effect was

Figure 4. Identification of TGR5 as the target molecule of miR-4485. (A) Venn diagram revealing the putative target molecules of miR-4485 predicted by mirWalk, TargetScan, miRPathDB, and miRDB. Six genes (TGR5, ASB7, CTDSPL2, FBXO32, JOSD1, MED1) were commonly identified by all 4 webtools as potential targets for miR-4485. (B) Expression level of TGR5 mRNA, quantified by quantitative reverse-transcription polymerase chain reaction (qRT-PCR), in primary chronic lymphocytic leukemia (CLL) cells from Asian patients (N=11) and normal B lymphocytes from Asian healthy donors (N=7, mean ± standard deviation [SD]); *P<0.05. (C) Comparison of TGR5 gene expression in primary leukemia cells from CLL patient and normal B lymphocytes from healthy donors, using the GSE66117 dataset; data are presented as mean ± SD; *P<0.05. (D) Key nucleotide sequences of the luciferase reporters: the wild-type sequence of TGR5 3’ untranslated region (3’UTR) containing potential miR-4485 binding site is shown on the upper line; the complementary sequence of miR-4485 is shown in red, with corresponding miR-4485 sequence shown in the middle line; the mutated 3’UTR sequence containing 7 altered nucleotides (red color) are shown in the third line. The bar graph in the lower panel shows the results of luciferase assay using HEK293 cells co-transfected with miR-4485 mimic and the luciferase reporter containing either wild-type TGR5 3’UTR or the mutated TGR5 3’UTR; **P<0.01. Haematologica | 109 February 2024

487

P. Liu et al.

ARTICLE - MicroRNA affects CLL incidence in different races

unlikely mediated by the canonical MEK/ERK1/2 pathway, since TGR5 did not cause any significant change in MEK or p-MEK (Online Supplementary Figure S6C, D). At the cellular level, TGR5 overexpression in MEC1 and MEC2 cells led to a significant decrease in cell proliferation (Online Supplementary Figure S6E, F), consistent with its inhibitory effect on ERK1/2 activation. In order to further evaluate the potential role of TGR5 in mediating the regulation of miR-4485 on ERK1/2 and cell proliferation, we first used MEC2 cell line to generate

stable TGR5-knockdown cells that exhibited an increase in ERK1/2 phosphorylation and cell proliferation (Online Supplementary Figure S7A, B), and then transfected the cells with miR-4485 inhibitory oligos to test its impact on ERK1/2 and cell proliferation. In MEC2 cells transfected with control small hairpin RNA (shRNA), inhibition of miR-4485 by inhibitory oligos consistently led to upregulation of TGR5 expression, suppression of ERK1/2 phosphorylation (Figure 6C), and a significant decrease in cell proliferation (Online Supplementary Figure S7C). In contrast, miR-4485 inhib-

Figure 5. miR-4485 suppresses TGR5 expression and promotes phosphorylation of ERK1/2. (A) HEK293T cells were transfected with miR-4485 overexpression vectors, and 2 stable clones were selected for analysis of TGR5 expression by quantitative reverse-transcription polymerase chain reaction (qRT-PCR) (N=3, mean ± standard deviation [SD]). (B) HEK293T cells were transfected with miR-4485 overexpression vectors, and 2 stable clones were selected for analysis of protein expression by western blotting, using the specific antibodies for the molecules of interest as indicated. (C) HEK293T cells were transfected with a miArrest™ microRNA (miRNA) inhibitor vector (pEZX-AM03) containing an miR-4485 antagonist sequence. The expression of TGR5 mRNA and protein were measured by qRT-PCR and western blotting, respectively (N=3, mean ± SD). (D) MEC1 cells were transiently transfected with miR-4485 mimic or inhibitor, and their corresponding negative controls, and the expression levels of TGR5, phosphorylated MEK1/2 (Ser217/221), MEK1/2, phosphorylated ERK1/2 (Thr202/Tyr204), ERK1/2, phosphorylated-c-Jun (S63), and c-Jun was measured by western blot analysis. (E) Primary chronic lymphocytic leukemia (CLL) cells were transiently transfected with miR-4485 inhibitor or its corresponding negative control, the expression of TGR5 mRNA and protein were measured by qRT-PCR and western blotting, respectively. The expression of ERK1/2, phosphorylated ERK1/2, phosphorylated-c-Jun and c-Jun was also measured by western blot analysis. miRi Ctrl: miRNA inhibitor control; miR-4485i: miR-4485 inhibitor; miRm Ctrl: miRNA mimic control; miR-4485m: miR-4485 mimic. ***P<0.001. Haematologica | 109 February 2024

488

ARTICLE - MicroRNA affects CLL incidence in different races

itor could no longer induce TGR5 expression and did not suppress ERK1/2 phosphorylation in MEC2 cells with TGR5 knockdown by shRNA (Figure 6C), and its inhibitory effect on cell proliferation was also largely attenuated in MEC2 cell with TGR5 knockdown (Online Supplementary Figure S7D, E). These data together suggest that TGR5 might play an important role in mediating the effect of miR-4485 on ERK1/2 activation and cell growth, as illustrated in Figure 6D.

Discussion It has been known for decades that the incidence of CLL is significantly lower in Asians compared to that in the Western population.4-10 The fact that CLL disease frequency remains low in Asians despite living in the western countries for multiple generations suggests the existence of strong genetic factors that affect CLL development. However, the exact identities of such genetic factors remain largely unknown, although certain genes with general pro-CLL

P. Liu et al.

functions have been identified. For example, unmutated IGHV, p53 mutation, Bcl-2 overexpression, 17p deletion [Del (17p)], 11q deletion [Del (11q)], Trisomy 12, 13q deletion [Del (13q)], and low expression of miR-15/16 have been implicated to play certain roles in CLL development,35-38 but these genetic events did not show differential occurrence between Asian and Western populations and thus could not account for the different CLL incidences in the two ethnic groups. Thus, further investigations are required to reveal the possible genetic differences between Asian and Western populations that could explain the population difference in CLL leukemogenesis, which would not only gain further insights into CLL biology, but would also provide a basis for developing new strategies for effective CLL treatment. Based on the significant role of miRNA in affecting CLL development,39,40 the objectives of the current study were to identify candidate miRNA that might contribute to the different incidences of CLL in Asian and Western populations, and to investigate the relevant molecular mechanisms. Using the GeneChip® miRNA array analysis, we compared

Figure 6. TGR5 plays a key role in mediating the effect of miR-4485 on ERK1/2 activation. (A) MEC1 cells and primary chronic lymphocytic leukemia (CLL) cells isolated from CLL patients were treated with or without SCH772984 (an ERK inhibitor) at 1-5 µM for 48 hours (hrs), and the indicated molecules were determined by western blot. (B) MEC1 cells were transduced with empty control vector or TGR5 expression vectors (TGR5-OE), and primary CLL cells were incubated with TGR5 agonist at indicated concentrations for 48 hrs. The cells were then collected for western blot analysis of indicated proteins. (C) MEC2 cells stably transduced with empty vector or TRG5 knockdown vectors (shTGR5-#2, shTGR5-#3) were transiently transfected with miR-4485 inhibitor (miR4485i) or its negative inhibitor control, and cells were collected for western blot analysis of indicated proteins. (D) Schematic model illustrating possible mechanisms by which miR-4485 promotes CLL development through inhibiting TGR5 expression and thus decreasing its suppressive function, leading to an increase of ERK1/2 phosphorylation and enhancing CLL cell proliferation. The activation of ERK1/2 could also suppress TGR5 expression and thus further reduce its inhibitory effect on ERK1/2, leading to further activation ERK1/2. As such, TGR5 and ERK1/2 form a double-negative loop, which could amplify the pro-CLL effect of miR-4485. miR4485i: miR-4485 inhibitor.

Haematologica | 109 February 2024

489

ARTICLE - MicroRNA affects CLL incidence in different races

the global miRNA expression profiles in primary CLL cells and normal lymphocytes of the two ethnic groups, aiming to identify (i) pro-CLL miRNA with high expression in all CLL cells and lowest expression in Asian normal B lymphocytes (lower than Western normal B lymphocytes), and (ii) anti-CLL miRNA with low expression in all CLL cells and highest expression in Asian normal B lymphocytes (higher than Western normal B lymphocytes). Using this logical approach, we identified a single miRNA (miR-4485) in the first category and five miRNA (miR-181d, miR-181c, miR363, miR-138, miR-181a) in the second category. It is likely that the low expression of pro-CLL miR-4485 and the high expression of the five anti-CLL miRNA in Asian normal B lymphocytes collaboratively or collectively contribute to the low CLL incidence in Asian population. Of note, although the tumor suppressive functions of miR-181d, miR-181c, miR-363, miR-138, and miR-181a have been reported previously,23,28-34 the novelty of our study was the finding that these anti-CLL miRNA were expressed significantly higher in Asian normal B lymphocytes than in Western normal B lymphocytes, and this could decrease the chance of CLL development in the Asian population. In contrast, the expression of miR-4485 with pro-CLL function was higher in Western normal B lymphocytes than in the Asian normal B lymphocytes, and might likely contribute to the higher CLL incidence in the Western population. The observations that miR-4485 expression was significantly higher in primary CLL cells of both Asian and Western patients compared to that in the normal B lymphocytes from healthy individuals of the respective ethnic groups suggest that this miRNA might potentially have a role in promoting CLL development. Indeed, we showed that inhibition of miR-4485 in two CLL cell lines resulted in a significant retardation of cell proliferation, whereas ectopic expression of miR-4485 in HEK293T cells led to an increase in cell proliferation, indicating the cancer promoting function of this miRNA. Importantly, we found that while the expression of miR-4485 was generally low in normal B lymphocytes, its level was particularly lower in Asian normal B lymphocytes compared to Western normal lymphocytes. Together, these novel findings suggest that the differential expression of miR4485 in the normal B lymphocytes of Asian versus Western groups likely contributed to the different CLL incidences in these two populations, although the exact mechanisms for its pro-CLL function remain to be further explored. In human cells, miR-4485 is located on chromosome 11p15.4, and its function has not been well-characterized. Jima et al. used deep sequencing to analyze miRNA expression in normal B lymphocytes from the tonsils of healthy individuals in comparison with that in malignant B cells from lymphoma patients, and found higher expression of miR-4485 in malignant B cells.41 A study by Sripada et al. showed that miR-4485 was translocated to mitochondria where it regulated the processing of mitochondrial 16S rRNA and affected mitochondrial functions and cellular metabolism in breast

P. Liu et al.

cancer cells.42 Interestingly, the same study suggested that miR-4485 might function as a tumor suppressor in breast carcinoma since its expression was low in breast cancer tissues and ectopic expression of miR-4485 in breast cancer cells led to a decrease in tumor growth.42 Thus, miR4485 appeared to have tumor promoting or suppressor functions depending on cell type. In our study, we found that miR-4485 was highly expressed in primary leukemia cells from CLL patients and could promote cell proliferation in two CLL cell lines, suggesting that it might have a pro-oncogenic function in B lymphocytes, consistent with the observation in B-cell lymphoma.41 The exact mechanism by which miR-4485 promotes CLL would provide an important insight into the regulation of CLL development. Our study showed that TGR5, also known as G-protein coupled bile acid receptor 1 (GPBAR1), seems to be a direct target molecule of miR-4485. This conclusion is supported by multiple evidence: (i) luciferase assay showed that miR-4485 could suppress the luciferase reporter activity in a vector containing 3’-UTR of TGR5, (ii) ectopic expression of miR-4485 downregulated the expression of TGR5, (iii) transfection of CLL cells with miR-4485 inhibitor led to the upregulation of TGR5 expression in two CLL cell lines, and (iv) inhibition of miR-4485 in primary CLL cells also resulted in an elevated expression of TGR5. Several studies have shown that TGR5 is highly expressed in normal tissues, has the ability to regulate glucose and lipid metabolism and maintain energy homeostasis, inhibits inflammatory reactions, and might have a tumor-suppressing function.43-45 Activation of TGR5 has been shown to suppress cell proliferation and migration through dephosphorylation of STAT3 in gastric cancer.46 The tumor-suppressive effect of TGR5 was also observed in hepatocellular carcinoma, again through inhibition of STAT3 signaling pathway.47 In our study, although we consistently found that miR-4485 could significantly downregulate TGR5 expression in CLL cells and HEK cells, the decrease in TGR5 did not cause any detectable changes in STAT3 protein level or its phosphorylation. Instead, we observed a significant increase in phosphorylation of the ERK1/2, suggesting that activation of the ERK1/2 signaling pathway might be a potential mechanism by which miR-4485 promote CLL development. TGR5 seems to play a key role in mediating the effect of miR-4485 on activation of ERK1/2 and promotion of cell proliferation, as a knockdown of TGR5 abolished such effect of miR-4485. As illustrated in Figure 6D, miR-4485 seems to exert its pro-CLL function through a double-negative mechanism with a double-negative amplification loop in the following fashion. miR-4485 negatively regulates the expression of TGR5 and thus reduces its negative impact on ERK1/2 phosphorylation, leading to activation of ERK1/2 and cell proliferation. This double negative effect could be further amplified by a double-negative regulatory loop between ERK1/2 and TGR5. Since inhibition of ERK1/2 could enhance TGR5 expression (Figure 6A), the activation of ERK1/2 would then suppress TGR5

Haematologica | 109 February 2024

490

P. Liu et al.

ARTICLE - MicroRNA affects CLL incidence in different races

expression and thus reduce its inhibitory effect on ERK1/2, leading to further activation ERK1/2. This double-negative loop could potentially provide a novel mechanism to amplify the pro-CLL effect of miR-4485. Our study suggests that differential expression of miRNA in Asin and Western individuals may contribute to the different CLL incidents in the two ethnic populations, and provide novel insights into the regulatory mechanism of the miR4485/TGR5/ERK1/2 axis in the context of CLL development. However, it should be noted that our study has several limitations including small sample size, a lack of longitudinal samples at various stages of CLL development, and potential batch effects on miRNA assays due to samples acquisition and analysis in two different geological locations, although every effort was made to keep the experimental conditions identical and to minimize the risk of batch effect. Because of limited number of samples, we have not performed analysis of several same samples in both countries to evaluate the magnitude of potential batch effect. Thus, the differential expression of miR-4485 and several other miRNA should be further validated in larger sample sizes in future study. A longitudinal follow-up with analysis of blood samples at different stages during CLL development has the advantage of identifying the critical time when abnormal expression of key miRNA and activation of the pro-CLL pathway may occur. This may potentially identify the miRNA markers related to disease initiation and progression and the potential underlying mechanisms and therapeutic targets. Also, it would be interesting to include samples from Western CLL patients who live in Asia and samples from Asian CLL patients who live in the Western country as crucial controls for ideal comparison. Thus, it would be important to conduct a prospective study with larger sample sizes and all necessary control groups, and with sufficient follow-up time to determine whether high miR-4485 is associated with increased risk of developing CLL. In summary, through comparative analyses of the global miRNA expression profiles in primary CLL cells and normal

B lymphocytes from Asian and Western individuals, we have identified miR-4485 with CLL-promoting function and miR138, miR-181a, miR-181c, miR-181d, and miR-363 with tumor suppressor functions that were differentially expressed in Asian and Western individuals and likely contributed to the different CLL incidences in the two ethnic groups. The proCLL activity of miR-4485 is likely mediated by inhibition of TGR5 expression and thus reducing its tumor-suppressive effect on ERK1/2 phosphorylation, leading to ERK1/2 activation and leukemia cell proliferation. Disclosures No conflicts of interest to disclose. Contributions PPL, KFW and JNL performed experiments. PPL, KFW, JNL and PH designed the study, analyzed and interpreted the data. PPL and PH wrote the paper. KFW, MAO, WGW and PH aided in design of the microarray, processed its data. ZJX, MJK and YQL provided clinical samples and critical research materials. PPL, PH and YQL supervised the study. Acknowledgements The authors are most grateful to the Flow Cytometry Core Facility of Sun Yat-sen University Cancer Center for their technical support. Funding This work was supported by grants from the National Natural Science Foundation of China (grant no. 81972595), Natural Science Foundation of Guangdong Province (2023A1515010123) and a grant from the CLL Global Research Foundation. Data-sharing statement MicroRNA assay data has been made deposited in the Gene Expression Omnibus (GSE216258). All other datasets analyzed during the current study are available from the corresponding author on reasonable request.

References Blood. 2010;116(22):4430-4435. 6. Mak V, Ip D, Mang O, et al. Preservation of lower incidence of chronic lymphocytic leukemia in Chinese residents in British Columbia: a 26-year survey from 1983 to 2008. Leuk Lymphoma. 2014;55(4):824-827. 7. Miranda-Filho A, Pineros M, Ferlay J, Soerjomataram I, Monnereau A, Bray F. Epidemiological patterns of leukaemia in 184 countries: a population-based study. Lancet Haematol. 2018;5(1):e14-e24. 8. Ko BS, Chen LJ, Huang HH, Chen HM, Hsiao FY. Epidemiology, treatment patterns and survival of chronic lymphocytic leukaemia/small lymphocytic lymphoma (CLL/SLL) in Taiwan, 2006-2015. Int J Clin Pract. 2021;75(8):e14258. 9. Yang S, Varghese AM, Sood N, et al. Ethnic and geographic

1. Hallek M, Shanafelt TD, Eichhorst B. Chronic lymphocytic leukaemia. Lancet 2018;391(10129):1524-1537. 2. Kikushige Y. Pathogenesis of chronic lymphocytic leukemia and the development of novel therapeutic strategies. J Clin Exp Hematop. 2020;60(4):146-158. 3. Hallek M, Al-Sawaf O. Chronic lymphocytic leukemia: 2022 update on diagnostic and therapeutic procedures. Am J Hematol. 2021;96(12):1679-1705. 4. Yamamoto JF, Goodman MT. Patterns of leukemia incidence in the United States by subtype and demographic characteristics, 1997-2002. Cancer Causes Control. 2008;19(4):379-390. 5. Wu SJ, Huang SY, Lin CT, Lin YJ, Chang CJ, Tien HF. The incidence of chronic lymphocytic leukemia in Taiwan, 19862005: a distinct increasing trend with birth-cohort effect.

Haematologica | 109 February 2024

491

ARTICLE - MicroRNA affects CLL incidence in different races diversity of chronic lymphocytic leukaemia. Leukemia. 2021;35(2):433-439. 10. Gale RP. Chronic lymphocytic leukemia in China. Chin Med J (Engl). 2022;135(8):883-886. 11. Haenszel W, Kurihara M. Studies of Japanese migrants. I. Mortality from cancer and other diseases among Japanese in the United States. J Natl Cancer Inst. 1968;40(1):43-68. 12. Yanagihara ET, Blaisdell RK, Hayashi T, Lukes RJ. Malignant lymphoma in Hawaii-Japanese: a retrospective morphologic survey. Hematol Oncol. 1989;7(3):219-232. 13. Gale RP, Cozen W, Goodman MT, Wang FF, Bernstein L. Decreased chronic lymphocytic leukemia incidence in Asians in Los Angeles County. Leuk Res. 2000;24(8):665-669. 14. Agbu P, Carthew RW. MicroRNA-mediated regulation of glucose and lipid metabolism. Nat Rev Mol Cell Biol. 2021;22(6):425-438. 15. Lynam-Lennon N, Maher SG, Reynolds JV. The roles of microRNA in cancer and apoptosis. Biol Rev Camb Philos Soc. 2009;84(1):55-71. 16. Di Leva G, Garofalo M, Croce CM. MicroRNAs in cancer. Annu Rev Pathol. 2014;9:287-314. 17. Calin GA, Croce CM. Chronic lymphocytic leukemia: interplay between noncoding RNAs and protein-coding genes. Blood. 2009;114(23):4761-4770. 18. Visone R, Rassenti LZ, Veronese A, et al. Karyotype-specific microRNA signature in chronic lymphocytic leukemia. Blood. 2009;114(18):3872-3879. 19. Fulci V, Chiaretti S, Goldoni M, et al. Quantitative technologies establish a novel microRNA profile of chronic lymphocytic leukemia. Blood. 2007;109(11):4944-4951. 20. Klein U, Lia M, Crespo M, et al. The DLEU2/miR-15a/16-1 cluster controls B cell proliferation and its deletion leads to chronic lymphocytic leukemia. Cancer Cell. 2010;17(1):28-40. 21. Calin GA, Dumitru CD, Shimizu M, et al. Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci U S A. 2002;99(24):15524-15529. 22. Zhu DX, Miao KR, Fang C, et al. Aberrant microRNA expression in Chinese patients with chronic lymphocytic leukemia. Leuk Res. 2011;35(6):730-734. 23. Zhu DX, Zhu W, Fang C, et al. miR-181a/b significantly enhances drug sensitivity in chronic lymphocytic leukemia cells via targeting multiple anti-apoptosis genes. Carcinogenesis. 2012;33(7):1294-1301. 24. Huang RS, Gamazon ER, Ziliak D, et al. Population differences in microRNA expression and biological implications. RNA Biol. 2011;8(4):692-701. 25. Hallek M, Cheson BD, Catovsky D, et al. Guidelines for the diagnosis and treatment of chronic lymphocytic leukemia: a report from the International Workshop on Chronic Lymphocytic Leukemia updating the National Cancer Institute-Working Group 1996 guidelines. Blood. 2008;111(12):5446-5456. 26. Huang P, Sandoval A, Van Den Neste E, Keating MJ, Plunkett W. Inhibition of RNA transcription: a biochemical mechanism of action against chronic lymphocytic leukemia cells by fludarabine. Leukemia. 2000;14(8):1405-1413. 27. Stacchini A, Aragno M, Vallario A, et al. MEC1 and MEC2: two new cell lines derived from B-chronic lymphocytic leukaemia in prolymphocytoid transformation. Leuk Res. 1999;23(2):127-136. 28. Pekarsky Y, Santanam U, Cimmino A, et al. Tcl1 expression in chronic lymphocytic leukemia is regulated by miR-29 and miR-181. Cancer Res. 2006;66(24):11590-11593.

29. Berg V, Rusch M, Vartak N, et al. miRs-138 and -424 control palmitoylation-dependent CD95-mediated cell death by targeting acyl protein thioesterases 1 and 2 in CLL. Blood. 2015;125(19):2948-2957. 30. Jiang K, Xie LF, Xiao TZ, Qiu MY, Wang WL. MiR-181d inhibits cell proliferation and metastasis through PI3K/AKT pathway in gastric cancer. Eur Rev Med Pharmacol Sci. 2019;23(20):8861-8869. 31. Zhang W, Zhang J, Hoadley K, et al. miR-181d: a predictive glioblastoma biomarker that downregulates MGMT expression. Neuro Oncol. 2012;14(6):712-719. 32. Wang W, Chen R, Droll S, et al. miR-181c regulates MCL1 and cell survival in GATA2 deficient cells. J Leukoc Biol. 2022;111(4):805-816. 33. Feng WT, Yao R, Xu LJ, et al. Effect of miR-363 on the proliferation, invasion and apoptosis of laryngeal cancer by targeting Mcl-1. Eur Rev Med Pharmacol Sci. 2018;22(14):4564-4572. 34. Pallasch CP, Patz M, Park YJ, et al. miRNA deregulation by epigenetic silencing disrupts suppression of the oncogene PLAG1 in chronic lymphocytic leukemia. Blood. 2009;114(15):3255-3264. 35. Rossi S, Shimizu M, Barbarotto E, et al. microRNA fingerprinting of CLL patients with chromosome 17p deletion identify a miR-21 score that stratifies early survival. Blood. 2010;116(6):945-952. 36. Otake Y, Soundararajan S, Sengupta TK, et al. Overexpression of nucleolin in chronic lymphocytic leukemia cells induces stabilization of bcl2 mRNA. Blood. 2007;109(7):3069-3075. 37. Zenz T, Eichhorst B, Busch R, et al. TP53 mutation and survival in chronic lymphocytic leukemia. J Clin Oncol. 2010;28(29):4473-4479. 38. Kipps TJ, Stevenson FK, Wu CJ, et al. Chronic lymphocytic leukaemia. Nat Rev Dis Primers. 2017;3:16096. 39. Katsaraki K, Karousi P, Artemaki PI, et al. MicroRNAs: tiny regulators of gene expression with pivotal roles in normal B-cell development and B-cell chronic lymphocytic leukemia. Cancers. 2021;13(4):593. 40. D’Abundo L, Callegari E, Bresin A, et al. Anti-leukemic activity of microRNA-26a in a chronic lymphocytic leukemia mouse model. Oncogene. 2017;36(47):6617-6626. 41. Jima DD, Zhang J, Jacobs C, et al. Deep sequencing of the small RNA transcriptome of normal and malignant human B cells identifies hundreds of novel microRNAs. Blood. 2010;116(23):e118-e127. 42. Sripada L, Singh K, Lipatova AV, et al. hsa-miR-4485 regulates mitochondrial functions and inhibits the tumorigenicity of breast cancer cells. J Mol Med (Berl). 2017;95(6):641-651. 43. Houten SM, Watanabe M, Auwerx J. Endocrine functions of bile acids. EMBO J. 2006;25(7):1419-1425. 44. Zhao L, Zhang H, Liu X, et al. TGR5 deficiency activates antitumor immunity in non-small cell lung cancer via restraining M2 macrophage polarization. Acta Pharm Sin B. 2022;12(2):787-800. 45. Qi Y, Duan G, Wei D, Zhao C, Ma Y. The bile acid membrane receptor TGR5 in cancer: friend or foe? Molecules. 2022;27(16):5292. 46. Guo C, Su J, Li Z, et al. The G-protein-coupled bile acid receptor Gpbar1 (TGR5) suppresses gastric cancer cell proliferation and migration through antagonizing STAT3 signaling pathway. Oncotarget. 2015;6(33):34402-34413. 47. Chen WD, Yu D, Forman BM, Huang W, Wang YD. Deficiency of G-protein-coupled bile acid receptor Gpbar1 (TGR5) enhances chemically induced liver carcinogenesis. Hepatology. 2013;57(2):656-666.

Haematologica | 109 February 2024

492

P. Liu et al.

ARTICLE - Chronic Lymphocytic Leukemia

BCL3 rearrangements in B-cell lymphoid neoplasms occur in two breakpoint clusters associated with different diseases Anna Carbó-Meix,1* Francesca Guijarro,1,2* Luojun Wang,2* Marta Grau,1* Romina Royo,3 Gerard Frigola,1,2 Heribert Playa-Albinyana,1,4 Marco M. Bühler,5 Guillem Clot,1 Martí DuranFerrer,1,4 Junyan Lu,6 Isabel Granada,7 Maria-Joao Baptista,7 José-Tomás Navarro,7 Blanca Espinet,8 Anna Puiggros,8 Gustavo Tapia,9 Laura Bandiera,10 Gabriella De Canal,10 Emanuela Bonoldi,10 Fina Climent,11 Inmaculada Ribera-Cortada,12 Mariana Fernández-Caballero,7 Esmeralda de la Banda,13 Janilson do Nascimento,14 Alberto Pineda,15 Dolors Vela,16 María Rozman,2 Marta Aymerich,1,2 Charlotte Syrykh,17 Pierre Brousset,17,18,19 Miguel Perera,20 Lucrecia Yáñez,21 Jesús Xavier Ortin,22 Esperanza Tuset,23 Thorsten Zenz,24 James R. Cook,25 Steven H. Swerdlow,26 José I. Martín-Subero,1,4,27,28 Dolors Colomer,1,2,4,27 Estella Matutes,2 Sílvia Beà,1,2,4,27 Dolors Costa,1,2,4 Ferran Nadeu1,4# and Elías Campo1,2,4,27# 1

Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Spain; Hematopathology Section, Pathology Laboratory, Hospital Clínic de Barcelona, Barcelona, Spain; 3Barcelona Supercomputing Center (BSC), Barcelona, Spain; 4Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), Madrid, Spain; 5Department of Pathology and Molecular Pathology, University Hospital Zürich, Zürich, Switzerland; 6 European Molecular Biology Laboratory, Heidelberg, Germany; 7Department of Hematology, Institut Català d’Oncologia, Hospital Germans Trias i Pujol, Josep Carreras Research Institute, Universitat Autònoma de Barcelona, Badalona, Spain; 8Molecular Cytogenetics Laboratory, Department of Pathology, Hospital del Mar, Barcelona, Spain and Translational Research on Hematological Neoplasms Group (GRETNHE) - Institut Hospital del Mar d’Investigacions Mèdiques (IMIM), Barcelona, Spain; 9Department of Pathology, Hospital Germans Trias i Pujol, Badalona, Spain; 10Dipartimento Ematologia, Oncologia e Medicina Molecolare, Anatomia Istologia Patologica e Citogenetica, Niguarda Cancer Center, Milano, Italy; 11Department of Pathology, Hospital Universitari de Bellvitge, Institut d’Investigació Biomèdica de Bellvitge (IDIBELL), L'Hospitalet De Llobregat, Spain; 12 Department of Pathology, Hospital Nostra Senyora de Meritxell, Escaldes-Engordany, Principat d'Andorra; 13Hematology Laboratory, Hospital Universitari Bellvitge, Institut d’Investigació Biomèdica de Bellvitge (IDIBELL), L'Hospitalet De Llobregat, Spain; 14 Department of Hematology, Hospital Joan XXIII, Institut Català d'Oncologia, Tarragona, Spain; 15Department of Hematology, Fundació Hospital de l'Esperit Sant, Badalona, Spain; 16 Hematologia Clínica, Hospital General de Granollers, Granollers, Spain; 17Department of Pathology, Toulouse University Hospital Center, Cancer Institute University of ToulouseOncopole, Toulouse, France; 18INSERM UMR1037 Cancer Research Center of Toulouse (CRCT), ERL 5294 National Center for Scientific Research (CNRS), University of Toulouse III Paul-Sabatier, Toulouse, France; 19Institut Carnot Lymphome CALYM, Laboratoire d'Excellence 'TOUCAN', Toulouse, France; 20Hematology Department, Hospital Dr Negrín, Las Palmas de Gran Canaria, Spain; 21Hematology Department, Hospital Universitario Marqués de Valdecilla-Instituto de Investigación Valdecilla (IDIVAL), Santander, Spain; 22 Hematology Department, Hospital Verge de la Cinta, Tortosa, Spain; 23Hematology Department, Institut Català d'Oncologia, Hospital Dr. Josep Trueta, Institut d'Investigació Biomèdica de Girona (IDIBGI), Girona, Spain; 24Department of Medical Oncology and Hematology, University Hospital and University of Zürich, Zurich, Switzerland; 25Pathology and Laboratory Medicine Institute, Cleveland Clinic, Cleveland, OH, USA; 26Department of Pathology, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA; 27Department of Fonaments Clinics, Universitat de Barcelona, Barcelona, Spain and 28Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain 2

AC-M, FG, LW and MG contributed equally as first authors. FN and EC contributed equally as senior authors.

Haematologica | 109 - February 2024

493

Correspondence: E. Campo ecampo@clinic.cat Received: Accepted: Early view:

March 23, 2022 July 31, 2023. August 10, 2023.

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

Abstract The t(14;19)(q32;q13) often juxtaposes BCL3 with immunoglobulin heavy chain (IGH) resulting in overexpression of the gene. In contrast to other oncogenic translocations, BCL3 rearrangement (BCL3-R) has been associated with a broad spectrum of lymphoid neoplasms. Here we report an integrative whole-genome sequence, transcriptomic, and DNA methylation analysis of 13 lymphoid neoplasms with BCL3-R. The resolution of the breakpoints at single base-pair revealed that they occur in two clusters at 5’ (n=9) and 3’ (n=4) regions of BCL3 associated with two different biological and clinical entities. Both breakpoints were mediated by aberrant class switch recombination of the IGH locus. However, the 5’ breakpoints (upstream) juxtaposed BCL3 next to an IGH enhancer leading to overexpression of the gene whereas the 3’ breakpoints (downstream) positioned BCL3 outside the influence of the IGH and were not associated with its expression. Upstream BCL3-R tumors had unmutated IGHV, trisomy 12, and mutated genes frequently seen in chronic lymphocytic leukemia (CLL) but had an atypical CLL morphology, immunophenotype, DNA methylome, and expression profile that differ from conventional CLL. In contrast, downstream BCL3-R neoplasms were atypical splenic or nodal marginal zone lymphomas (MZL) with mutated IGHV, complex karyotypes and mutated genes typical of MZL. Two of the latter four tumors transformed to a large B-cell lymphoma. We designed a novel fluorescence in situ hybridization assay that recognizes the two different breakpoints and validated these findings in 17 independent tumors. Overall, upstream or downstream breakpoints of BCL3-R are mainly associated with two subtypes of lymphoid neoplasms with different (epi)genomic, expression, and clinicopathological features resembling atypical CLL and MZL, respectively.

mors diagnosed as B-cell non-Hodgkin lymphomas, some of them with evidence of transformation.6,7 Whether this diThe t(14;19)(q32;q13) is a balanced translocation found in versity of entities associated with BCL3-R corresponds to less than 1% of lymphoid neoplasms that often leads to the a real biological promiscuity is not clear. Some reports injuxtaposition of BCL3 (B-cell leukemia/lymphoma 3) with cluded tumors with the t(14;19) by cytogenetics without the regulatory elements of the immunoglobulin heavy chain specific analysis of BCL3-R. Since this translocation may (IGH) gene, resulting in the overexpression of the gene.1 rearrange genes other than BCL3, it is possible that some BCL3 encodes an IκB-like nuclear protein that regulates of the series reported may have included tumors that did NF-κB activity apparently as a molecular adaptor between not involve BCL3. Furthermore, some studies included tuNF-κB transcription factors and nuclear co-activator and mors for which the pathological features were not thorco-repressor complexes.2 Although the function of BCL3 in oughly reviewed.6,7 B cells is not fully understood, this gene seems to be in- The purposes of this study were to characterize the gevolved in regulation of cell proliferation, differentiation, and nomic configuration of BCL3-R in B-cell neoplasms and to survival.3,4 In transgenic mice, Bcl3 overexpression pro- understand the clinical and biological significance of this moted accumulation of mature B cells but it was not suf- alteration using an integrative (epi)genomic and transcripficient to drive malignant transformation.5 tomic analysis in a cohort of patients with available clinical Chromosomal translocations activating oncogenes in lym- and pathological characteristics. phoid neoplasms are usually associated with relatively specific tumor subtypes. However, the t(14;19) and BCL3 rearrangement (BCL3-R) have been identified in a broad Methods spectrum of different tumor subtypes.6,7 Most patients have been diagnosed with chronic lymphocytic leukemia (CLL), Patients and samples atypical CLL, or transformed CLL. These tumors frequently We searched the cytogenetic files of lymphoid B-cell neohave an unmutated IGHV (U-IGHV) and trisomy 12. However, plasms with t(14;19) or BCL3-R in three institutions from they also have atypical features for CLL, including cytology 2008 to 2019. Fluorescence in situ hybridization (FISH) with and immunophenotype not characteristic of CLL, frequent dual-color break-apart probes for IGH and BCL3 genes (XL IGHV stereotype #8, and aggressive behavior in some IGH BA and XL BCL3 BA, Metasystems) was performed in series.6-9 Some authors have suggested that B-cell neo- patients with available material. Patients with t(14;19), but plasms carrying the t(14;19) could represent an entity dif- lacking confirmation of BCL3-R, were excluded. Overall, 13 ferent from CLL.6 In addition to these tumors resembling B-cell neoplasms carrying BCL3-R, with available material CLL, the t(14;19) and BCL3-R have been also identified in for genomic studies, were identified (Table 1; Online Supdiffuse large B-cell lymphomas (DLBCL), marginal zone plementary Figure S1; Online Supplementary Table S1). These lymphomas (MZL), splenic small B-cell lymphomas, and tu- cases represent 0.28% of all small B-cell lymphomas

Introduction

Haematologica | 109 - February 2024

494

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms studied genetically. The initial diagnoses were atypical CLL (aCLL) (n=5), SLL/CLL (n=3), splenic marginal zone lymphoma (SMZL) (n=3), lymphoplasmacytic lymphoma (n=1), and unclassifiable low-grade B-cell leukemic neoplasm (n=1). Tumor DNA was obtained from cryopreserved blood cells or frozen tumor tissue in all patients, germline DNA from non-neoplastic blood cells or saliva (n=10), and RNA from peripheral blood purified cells (n=5) or frozen tissue (n=2). Informed consent was obtained from all patients, and the study was approved by the Ethics Committee of the Hospital Clínic of Barcelona. Genomic studies Whole-genome sequencing (WGS) of the 13 tumors and 10 paired normal DNA samples was performed using the TruSeq DNA PCR Free or TruSeq DNA nano library preparation. Raw reads were mapped to the human reference genome (GRCh37) using the BWA-MEM algorithm.10 Immunoglobulin gene rearrangements were extracted using IgCaller (version 1.1)11 and annotated using IMGT/V-QUEST.12 Genomic alterations were identified using a multi-caller bioinformatics approach (Online Supplementary Appendix).13 Driver mutations were studied considering a list of 247 recurrently mutated genes in B-cell neoplasms (Online Supplementary Appendix; Online Supplementary Table S2). Total RNA sequencing (RNA-seq) was performed in seven tumors with BCL3-R and nine CLL without BCL3-R. Raw data were analyzed as previously described (Online Supplementary Appendix)13 using the DESeq2 package.14 mRNA-seq data from our International Cancer Genome Consortium CLL cohort were used for comparison.15 DNA methylation profiles of ten BCL3-R tumors were generated using EPIC methylation arrays. Similar data from 85 CLL were obtained for comparison from two previous publications: cohort 1 (C1) included 12 CLL from our institution,13 and cohort 2 (C2) 73 CLL from University Hospital Heidelberg.16 Data analyses were performed using minfi and limma packages.17,18 The AME tool from the MEME suite19 was used for enrichment analysis of known motifs (2022 JASPAR database; Online Supplementary Appendix).20 WGS, RNA-seq, and DNA methylation data are deposited in the European Genome-phenome Archive. Immunohistochemistry BCL3 protein expression was studied by immunohistochemistry (IHC) in tumors with formalin-fixed paraffin-embedded tissue. Tissue sections (3 mm) were stained using a Leica Bond-MAX stainer (Leica Biosystems) and the antiBCL3 primary antibody (23959-1-AP; Proteintech) (Online Supplementary Appendix). Custom BCL3 fluorescence in situ hybridization Custom BCL3 break-apart FISH probes to detect 5’ and 3’ BCL3 breakpoints were designed using three differentially labeled BAC clones: RP11-927F16 (spectrum orange), CTD-

A. Carbó-Meix et al. 2608C5 (spectrum aqua), and RP11-423N20 (spectrum green) from the Children´s Hospital Oakland Research Institute library obtained from the Molecular Cytogenetics Platform of IMIM (Barcelona, Spain) and Life Technologies. BAC extraction and labeling, slide preparation, and hybridization were performed according to standard procedures.21

Results Genomic characterization of the BCL3 rearrangement We first characterized the breakpoints of the BCL3 rearrangement at base-pair resolution using WGS data from 13 tumors (Online Supplementary Table S3). BCL3 was rearranged with the IGH region as a clonal event in all but one tumor (3646), in which the number of reads suggested a subclonal distribution. All tumors had breakpoints on chromosome 14 within class switch recombination (CSR) regions of the IGH locus (Figure 1A). Breakpoints occurred in IGHA2 (n=1), IGHG2 (n=3), IGHA1 (n=4), IGHG1 (n=3), and IGHG3 (n=3). Breakpoints on chromosome 19 (chr19) were found upstream of the 5’ untranslated region (UTR) of BCL3 in eight of 13 (61.5%) tumors (Figure 1A). These breakpoints occurred within a window of 13 kb, and the translocation juxtaposed BCL3 downstream of the CSR (Figure 1B). Notably, all eight tumors had U-IGHV, six had 100%, and two had 99.6% IGHV identity with the germline (Figure 1A; Online Supplementary Table S4). One additional tumor (3698) with mutated IGHV (M-IGHV) (94.4% identity) had a breakpoint further upstream of BCL3 truncating CEACAM16, although the result of the translocation also placed BCL3 downstream of the CSR (Figures 1A, C). The four remaining tumors had breakpoints downstream of BCL3, two within CBLC, one in BCAM, and one after NECTIN2 (Figure 1A). In these four translocations, BCL3 was not located after the CSR of IGH; therefore, it does not seem to be the target of the translocation (Figure 1C). All tumors carrying the breakpoint downstream of BCL3 had M-IGHV with <98% germline identity. In order to determine the influence of the chr19 breakpoint on BCL3 expression, we studied 12 tumors, seven by RNA-seq (6 with the breakpoint upstream and 1 downstream) and seven by IHC (3 upstream and 4 downstream). Two tumors were studied using both approaches (Online Supplementary Table S5). Eight tumors carrying the upstream BCL3-R, including one further upstream (3698), overexpressed BCL3 in comparison to CLL without this rearrangement (Figures 2A, B). No protein expression was detected by IHC in ten additional nodal CLL with unmutated IGHV (U-CLL) with trisomy 12 and without BCL3-R. In contrast, the four tumors downstream BCL3-R did not express BCL3 (Figures 2A, B). The only downstream BCL3R tumor with RNA available (3676) showed overexpression

Haematologica | 109 - February 2024

495

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

Table 1. Clinical and pathological features of 13 patients with BCL3 rearrangement.

Clinical data at diagnosis Median age in years (range) Male, N (%) ALC, x109/L (range) Lymphadenopathy, N (%) Splenomegaly, N (%) B symptoms, N (%) Clinical data at follow-up Need for treatment, N (%) Median time in years to first treatment Large B-cell lymphoma transformation, N (%) Median survival time in years Genetics, N (%) Trisomy 12 Complex karyotype Unmutated IGHV status Deletion 11q TP53 mutation Deletion 17p Phenotype Flow cytometry, N (%) Typical for CLL* Bright sIg Bright B-cell markers CD5+ CD43+ CD23+ CD200+ Immunohistochemistry, N/N CD5+ CD23+ LEF1+ BCL3+

Total N=13

Upstream BCL3-R N=9

Downstream BCL3-R N=4

69 (50-81) 7 (54) 8.1 (0.9-161.8) 3 (25) 4 (33) 3 (25)

69 (50-78) 5 (56) 9 (2.9-161.8) 3 (38) 1 (13) 1 (13)

64 (53-81) 2 (50) 1 (0.9-2.9) 0 3 (75) 2 (50)

1 1 0.01 0.49 0.07 0.24

9 (69) 2.6 2 (15) 10.5

5 (56) 4.6 0 11.1

4 (100) 0.9 2 (50) 10.5

0.23 0.7 0.08 0.5

7 (54) 8 (61) 8 (61) 2 (15) 0 0

7 (78) 6 (67) 8 (89) 1 (11) 0 0

0 2 (67) 0 1 (25) 0 0

0.04 1 0.007 1

3 (23) 8 (61) 9 (69) 9 (69) 7 (58) 6 (50) 9 (75)

3 (33) 4 (44) 5 (56) 9 (100) 6 (67) 6 (67) 8 (89)

0 4 (100) 4 (100) 0 1 (33) 0 1 (33)

3/7 4/7 0/7 3/6

3/4 2/4 strong, 2/4 weak 0/4 3/3

0/3 0/3 0/3 0/3

Quantitative parameters are expressed as median (range). *Typical chronic lymphocytic leukemia (CLL) immunophenotype CD19+ with dim expression of CD20, CD22, CD79b, CD5+, CD23+, CD43+, CD200+, and FMC7-. BCL3-R: BCL3 rearrangement; IGHV: variable region of the immunoglobulin heavy chain gene; ALC: absolute lymphocyte count; sIg: surface immunoglobulins.

of NECTIN2, which was negative in all upstream BCL3-R tumors (Figures 2A). Overall, the location of the chr19 breakpoint distinguishes two main subgroups: i) tumors with upstream BCL3-R breakpoints, which overexpress BCL3 and are enriched in U-IGHV, and ii) tumors with downstream BCL3-R breakpoints, which do not overexpress BCL3 and carry M-IGHV.

SBS5, and SBS8 in all cases, SBS18 in seven cases, and SBS9 in three tumors with M-IGHV (Online Supplementary Figure S2; Online Supplementary Table S7). In addition, we searched for activation-induced deaminase (AID) motifs in the mutations occurring in IGH locus between the constant gene and class switch regions. We found that 17 of 25 (68%) mutations occurred in AID motifs (Online Supplementary Table S8). Genomic landscape The driver mutations in the upstream BCL3-R subgroup Tumors with upstream BCL3-R, excluding the three patients were very heterogeneous, with only MED12 and FAT4 relacking normal DNA, had significantly fewer somatic muta- currently mutated in two tumors each. Other mutated tions (mean 2,511; range, 1,825-3,165; n=7) than tumors with genes have also been frequently described in CLL (ATM, downstream BCL3-R (mean 6,271.7; range, 4,535-9,125; n=3) NOTCH1, POT1, KHL6). In the four downstream BCL3-R tu(P<0.05; Figure 3A, B; Online Supplementary Table S6). Mu- mors, two carried mutations in KMT2D, NOTCH2, and KLF2, tational signature analysis identified the presence of SBS1, frequently seen in MZL, whereas the remaining two tuHaematologica | 109 - February 2024

496

Figure 1. Characterization of the breakpoints derived from the t(14;19) at base-pair resolution. (A) Representation of the breakpoints on chromosome 19 (chr19) and chr14 for each patient (red vertical line). Unmutated variable region of the immunoglobulin heavy chain gene (U-IGHV) and mutated IGHV (M-IGHV) tumors are represented in gray and black labels, respectively. Tumors are classified based on the breakpoint on chromosome 19 (chr19): further upstream BCL3 (pale blue), upstream BCL3 (blue), and downstream BCL3 (orange). (B) Schema of the most recurrent translocation pattern observed in the upstream BCL3 subgroup with IGH and its corresponding class switch recombination (CSR) located upstream BCL3, suggesting a constitutive upregulation of BCL3. (C) Depiction of 5 patients, 1 with the translocation further upstream BCL3 and 4 with the translocation downstream BCL3. In the further upstream tumor (3698), the t(14;19) truncates CEACAM16 and, similar to upstream BCL3 translocations, IGH is located 5’ of BCL3 suggesting a constitutive upregulation of this gene. In the downstream tumors, the t(14;19) affects 3 different genes (CBLC, BCAM, NECTIN2) located downstream of BCL3. The resulting derivatives of the t(14;19) suggest that BCL3 is not placed under the regulation of the enhancers of the IGH and, therefore, its expression remains unchanged.

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

Haematologica | 109 - February 2024

497

A. Carbó-Meix et al.

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

Figure 2. BCL3 expression in upstream and downstream tumors with BCL3 rearrangement. (A) RNA-sequencing data shows that BCL3 is upregulated in the upstream BCL3 tumors, except tumor 3646 carrying a subclonal t(14;19), compared to unmutated chronic lymphocyitc leukemia (U-CLL). Contrarily, the downstream BCL3 tumor (3676) upregulated NECTIN2 while showed lower BCL3 expression than any of the upstream and CLL tumors. (B) Immunohistochemistry images (400x) displaying a positive BCL3 expression in the further upstream tumor and in a representative upstream tumor, but not in a representative downstream tumor.

mors had recurrent mutations in TBL1XR1, detected in aggressive lymphomas, but also in some MZL (Figure 3A). In terms of CNA, upstream BCL3-R tumors had a significantly lower genomic complexity (mean 2.9; range, 1-9; n=9) than downstream BCL3-R tumors (mean 11.7; range, 5-19; n=3) (P<0.05; Figure 3B; Online Supplementary Table S9). All but one upstream BCL3-R tumor carried trisomy 12, but this aberration was not observed in any of the downstream BCL3-R tumors (Figures 3B, C; Online Supplementary Figure S3). In line with the copy number alterations (CNA), the number of structural variations (SV) was lower in upstream BCL3-R tumors than in downstream BCL3-R tumors (mean 4.8 SV; range, 2-10 [n=6] vs. 18 SV; range, 8-28 [n=3], respectively) (Figures 3B; Online Supplementary Figure S4; Online Supplementary Table S10). Gene expression profiling In order to determine the gene expression profile of the BCL3-R tumors, we compared the RNA-seq data of seven BCL3-R tumors (6 with upstream and 1 downstream breakpoint) with nine CLL (4 U-CLL and 5 mutated IGHV [M-CLL]). An unsupervised principal component analysis (PCA) suggested that upstream BCL3-R tumors displayed a distinct gene expression profile with some similarities with both M-CLL and U-CLL, whereas the downstream BCL3-R tumor did not cluster with any of the other tumors (Figure 4A). Then, we conducted a differential expression analysis (DEA) between upstream BCL3-R tumors, all UIGHV with trisomy 12, four CLL with U-IGHV, and one with trisomy 12 (excluding tumor 3646 with subclonal BCL3-R). This analysis identified 1,298 differentially expressed genes

(DEG): 578 upregulated and 720 downregulated in the upstream BCL3-R subgroup (q<0.05; Figure 4B; Online Supplementary Table S11). These genes showed similar expression levels in U- and M-CLL (Figure 4B; Online Supplementary Table S12). Significant expression differences were found in genes previously described as characteristically down- or upregulated in CLL compared with other B-cell neoplasms.22-24 Among them, upstream BCL3-R tumors had significant overexpression of EBF1, usually not expressed in CLL, and, in contrast, downregulation of LEF1, FMOD, ADTRP, CLNK, IGSF3, and TCF4, frequently overexpressed in CLL (Figure 4C). To rule out a potential confounding effect of trisomy 12, we performed a DEA between 16 U-CLL with trisomy 12 and 49 U-CLL without trisomy 12 using data from our ICGC CLL cohort.15 These analyses identified 1,527 DEG (q<0.05, absolute (log2 fold change [FC])>0.1; Online Supplementary Table S13). Among them, only 129 (9.9%) were shared by the upstream BCL3-R tumors, suggesting that most DEG observed in BCL3-R tumors were not related to trisomy 12 (Figure 4D). Interestingly, most CLL-specific genes modulated in the upstream BCL3-R tumors appeared to be independent of trisomy 12 in U-CLL (Figure 4D; Online Supplementary Figure S5A). Gene set enrichment analyses of upstream BCL3-R tumors and U-CLL with trisomy 12 showed that, while both subgroups of tumors shared some genes related to trisomy 12, most other pathways identified were expressed at lower levels in BCL3-R tumors, such as B-cell receptor (BCR) signaling or TNFα signaling via NF-κB (Online Supplementary Figure S5B; Online Supplementary Tables S14,

Haematologica | 109 - February 2024

498

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

Figure 3. Mutations and structural alterations in B-cell neoplasms with t(14;19) and BCL3 rearrangement identified by wholeexome sequencing. (A) Oncoprint representation of driver gene mutations frequently observed in chronic lymphocytic leukemia (CLL) (red) or in other B-cell lymphomas (blue-green). Total number of mutations are not reported in samples 3649, 3696, and 624 due to the lack of germline DNA (Online Supplementary Appendix; Online Supplementary Table S1). (B) Comparison of the number of mutations, copy number alterations (CNA) and structural variations (SV) between upstream BCL3 rearrangement (BCL3-R) and downstream BCL3-R tumors. (C) Copy number profile of BCL3-R tumors. Tumors are shown in rows and chromosomes in columns. The variable region of the immunoglobulin heavy chain gene (IGHV) mutational status, breakpoint location on chromosome 19, and number of CNA are shown on the right. Sample 3649 had an estimated tumor cell content of 20% that allowed the detection of driver somatic mutations and the BCL3-R but was not sufficient for a proper analysis of CNA (Online Supplementary Appendix; Online Supplementary Table S1). MZL: marginal zone lymphoma; CNN-LOH: copy number neutral loss of heterozygosity; Num: number; mut: mutational; NA: not available; M-IGHV: mutated IGHV; U-IGHV: unmutated IGHV. Haematologica | 109 - February 2024

499

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

Figure 4. Gene expression profile of upstream tumors with BCL3 rearrangement. (A) Principal component analysis of RNA sequencing data of 6 upstream BCL3 rearrangement (BCL3-R) tumors, 1 downstream BCL3-R tumor, and 9 chronic lymphocytic leukemia (CLL) (1st component is shown against 2nd, 3rd, 4th and 5th components). (B) Heatmap of the differential gene expression analysis between 5 upstream BCL3-R tumors and 4 unmutated CLL (U-CLL), also compared to 1 tumor with downstream BCL3R tumor and CLL without t(14;19). Tumor 3646 was excluded from the analysis due to its subclonal BCL3-R. Hallmark CLL genes differentially expressed between BCL3-R tumors and CLL are flagged. (C) Expression of CLL hallmark genes in the upstream BCL3-R tumors compared to U-CLL. Q-values are from the differential gene expression analysis. (D) Venn diagram showing the overlap of the differentially expressed genes among upstream BCL3-R versus U-CLL and U-CLL with versus without trisomy 12. Hallmark CLL genes are highlighted. IGHV: variable region of the immunoglobulin heavy chain gene; M-IGHV: mutated IGHV; UIGHV: unmutated IGHV; NA: not available; M-CLL: mutated CLL; w/o: without.

S15). The lower BCR-signaling capacity of BCL3-R tumors was confirmed by measuring Ca2+ mobilization upon BCR stimulation with IgM (Online Supplementary Figure S6; Online Supplementary Appendix). These findings suggest that, although upstream BCL3-R tumors share a subset of commonly expressed genes in CLL carrying trisomy 12, they also have a remarkably distinct profile. DNA methylation We analyzed the DNA methylation profile of eight upstream BCL3-R tumors, one of which was subclonal, and two downstream BCL3-R, and compared them with that of 85 CLL classified as naive-like CLL (n-CLL) (n=33), intermediate CLL (i-CLL) (n=7), or memory-like (m-CLL) (n=45),25,26 and seven normal B-cell subsets (2 naive, 1 ger-

minal center, 3 memory, and 1 plasma cell). We first performed PCA using 764159 CpG (Figure 5A). Principal component 1 (PC1) reflected the variability related to the proliferative history of the cells captured by the epiCMIT score,26 whereas PC2 grouped samples based on the cell of origin, in which upstream BCL3-R clustered with n-CLL (Figure 5A). Upstream BCL3-R tumors had a higher proliferative history than n-CLL. This observation was confirmed by comparing epiCMIT scores between BCL3-R and n-CLL in the C1 (P=0.0043) and C2 (P=0.00016) CLL cohorts (Figure 5B). In order to gain further insight into the differences between upstream BCL3-R and CLL, we performed differential methylation analysis between both subgroups of tumors adjusted for trisomy 12, IGHV status, epitype, and

Haematologica | 109 - February 2024

500

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms cohort (Figure 5C; Online Supplementary Figure S7A). This analysis showed 795 differentially methylated CpG (DMCpG), with 80 hyper- and 715 hypomethylated in BCL3R tumors (q<0.05; log FC=0.25; Online Supplementary Table S16). A subset of 21 hypomethylated CpG in BCL3-R tumors was modulated during B-cell differentiation, being

A. Carbó-Meix et al.

hypomethylated in germinal center-experienced normal B cells and M-CLL. Unmethylated CpG were enriched in heterochromatin and gene bodies, whereas hypermethylated CpG were enriched in enhancer-promoter regions (Online Supplementary Figure S7B). Among the DMCpG, 69 mapped to 37 DEG, with 45 of 64 (70%) hypomethylated CpG lo-

Figure 5. DNA methylation profile of upstream tumors with BCL3 rearrangement. (A) Principal component analysis (PCA) of DNA methylation data of 10 B-cell neoplasms with BCL3 rearrangement (BCL3-R), 85 chronic lymphocytic leukemia (CLL), and 7 normal B-cells subsets (1st and 2nd components are shown). The shape corresponds to the tumor types while the color represents the proliferative history (epiCMIT score). (B) Comparison of the epiCMIT score between upstream BCL3-R tumors and naive-like CLL (n-CLL) from cohorts C1 and C2, respectively. The upstream BCL3-R subgroup of tumors does not include the tumor 3646 carrying a subclonal t(14;19). (C) Heatmap of the differentially methylated CpG between 7 upstream BCL3-R tumors and 85 CLL. The chromatin state of each CpG is shown on the right. Differentially methylated CpG mapping to differentially expressed CLL genes of interest are labeled. (D) TP63 expression in the upstream BCL3-R subgroup compared to CLL. NBC: naive B cell; GC: germinal center B cell; MBC memory B cell; PC: plasma cell; n-CLL: naive-like CLL; i-CLL: intermediate CLL; m-CLL: memory-like CLL; IGHV: variable region of the immunoglobulin heavy chain gene; NA: not available. Haematologica | 109 - February 2024

501

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms cated in the gene body (5’UTR/first exon/body/3’UTR, n=38) or promoter region (TSS1500/TSS200, n=7) of upregulated genes and four of five (80%) hypermethylated CpG mapped to the gene body (n=2) or promoter (n=2) of downregulated genes (Figure 5C; Online Supplementary Table S16). These genes include EBF1, CREBBP, and genes associated with NOTCH1 pathway (EPS15L1, ZMIZ1),27,28 cell proliferation (BHLHE40, TP63),29-31 cell motility and migration (CORO1C, GAB1, GRAMD1B, ITGB2),32-36 and poor outcomes in CLL or other lymphoid neoplasms (IMMP2L, OSBPL10, TP63).31,37,38 Notably, TP63, previously shown to be a pro-survival factor in CLL subset #8,31 was overexpressed in BCL3-R cases (Figure 5D). A subsequent transcription factor (TF) binding analysis in the hypomethylated CpG revealed a significant enrichment in the binding sites of B-cell-related TF such as BCL11B, RUNX3, IRF, JUN/FOS, and FOX families (Online Supplementary Table S17). Pathology and clinical characteristics Given the marked genomic differences between upstream and downstream BCL3-R tumors, we reanalyzed their pathological and clinical features separately (Table 1; Figure 6; Online Supplementary Tables S18, S19). Upstream BCL3 rearrangement tumors Tumor cells in peripheral blood were small medium-sized with condensed non-clumped chromatin and broader pale cytoplasm than expected in typical CLL/SLL. Typical clumped chromatin was observed in only one tumor. Five tumors had cells with indented nuclei and seven tumors had prominent nucleoli (Figure 6A). Lymph node biopsies showed diffuse infiltration by small-to medium-sized cells in all tumors. In two tumors the cells had irregular nuclei and prominent nucleoli. Variable numbers of dispersed large cells were observed in all tumors, but clear proliferation centers were observed in only two. Flow cytometry showed expression of mature B-cell markers with CD5 and CD200 positivity in all tumors, but a typical CLL immunophenotype (CD19+, CD79b+, CD5+, CD23+, CD43+, CD200+ with dim expression of CD20, CD22, and FMC7-) was only found in three of nine tumors (Table 1; Online Supplementary Table S19). The other tumors expressed bright B-cell antigens/surface IgG and/or were dim/negative for CD23 and CD43. In the tissue sections, the four tumors studied were LEF1 negative and no or very scant follicular dendritic networks were observed (Figure 6B).

A. Carbó-Meix et al. cal small cells that partially preserved the architecture, with open sinusoids and occasional residual germinal centers. Tumor cells expanded the perifollicular areas and colonized germinal centers. One patient showed marked monotypic plasmacytosis. The two spleens showed expansion of the white pulp and partial infiltration of the red pulp by small-to medium-sized lymphoid proliferation with occasional larger cells, consistent with SMZL (Figure 6C). The four tumors expressed strong B-cell markers and were CD5 and CD23 negative. CD200 and CD43 were positive in one of three of the tumors, and two of three expressed IgD. Follicular dendritic cells highlighted the presence of residual germinal centers in all tumors (Figure 6C). Clinical characteristics The main clinical difference between the two subgroups was the higher lymphocyte count in the upstream BCL3R subgroup (P=0.01) and splenomegaly in three of the four patients with downstream BCL3-R (P=0.07) (Table 1). Two of the latter patients transformed to a large B-cell lymphoma 5 and 11 years after diagnosis. Transformations were not observed in the upstream BCL3-R subgroup with a similar median follow-up time as the downstream tumors (6.3 years vs. 5.3 years, respectively; P=0.4). All downstream BCL3-R patients required therapy, in contrast to only five of nine patients from the upstream BCL3-R subgroup, although no significant differences were found in the median time to first treatment. There were six deaths in the whole cohort, four of which were disease-related, two in the upstream BCL3-R subgroup, and two in the downstream BCL3-R subgroup, without differences in median survival time. Patients with upstream BCL3-R tumors had a similar overall survival as patients with U-CLL and trisomy 12 in our ICGC CLL cohort (Online Supplementary Figure S8).

Fluorescence in situ hybridization validation of BCL3 rearrangement breakpoints and expanded cohort As the current commercially available FISH probes do not distinguish between the 5’ and 3’ rearrangements of BCL3R identified in this study, we designed a new three-color FISH assay that could identify the new BCL3 5’ and 3’ breakpoints (Figure 7A). We tested the new assay in nine of our 13 tumors and confirmed the breakpoints concordantly with the WGS in all tumors, seven upstream and two downstream (Figure 7B; Online Supplementary Figure S9A; Downstream BCL3-R tumors Online Supplementary Table S1). Tumor cells in peripheral blood were larger than those in We used the new FISH assay in 17 additional B-cell neothe previous group, and three of four tumors had villi or plasms with t(14;19) or BCL3-R (Online Supplementary clasmatosis (Figure 6A). The patient without villous lym- Table S20, S21; Online Supplementary Figure S9B). We phocytes had multiple chromosomal alterations that were identified an upstream breakpoint in 13 tumors and downnot specific to any lymphoid neoplasm. The two lymph stream in four. In line with our previous observations, 11 nodes examined in these patients had infiltration by atypi- tumors with upstream breakpoints were diagnosed as Haematologica | 109 - February 2024

502

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

Figure 6. Images of tumors with representative upstream and downstream BCL3 rearrangement. (A) Cells in peripheral blood smears from representative tumors. Both tumors show features such as nuclear irregularities and lobulation, non-clumped chromatin, central nucleoli, ample cytoplasm, or villi, which are atypical for conventional chronic lymphocytic leukemia (CLL). 1000x oil immersion, light microscope and camera, Leishman stain. (B, C) Histology (hematoxilin & eosin staining) and immunohistochemistry images were obtained from scanned slides (Ventana DP200 scanner, Roche Diagnostics). The upstream BCL3 rearrangement (BCL3-R) tumor had a diffuse growth pattern, resembling chronic lymphocyitc leukemia (CLL), but without proliferation centers (100x). At high power (600x), the cells were small, with scarce cytoplasm, distinct irregular nuclei, and central nucleoli. Larger scattered cells were observed. The immunophenotype is atypical for a CLL tumor (CD5-, CD23+ weak, and CD43+ weak), and the cells are LEF1 negative. CD5 was negative in the lymph nodes by immunohistochemistry but positive in the peripheral blood according to flow cytometry. The downstream BCL3-R tumor has a perifollicular growth pattern (100x), leaving residual germinal centers (400x), with a residual follicular dendritic network on CD21 and germinal center cells on BCL6, resembling marginal zone lymphoma. This tumor has a non-specific B-cell phenotype and plasma cell differentiation with κ light-chain restriction. MZL: marginal zone lymphoma. Haematologica | 109 - February 2024

503

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

Figure 7. Custom fluorescence in situ hybridization assay to map the breakpoints of the BCL3 rearrangement and images of a representative tumor from the validation cohort. (A) Schematic representation of the custom design of BCL3 break-apart fluorescence in situ hybridization (FISH) probe. BCL3 gene and BCL3 FISH probe are annotated based on GRCh37/hg19 assembly. (B) Interphase nucleus of tumor 3783 (left panel) and 3676 (right panel). Tumor 3783 shows a positive signal constellation indicating a break upstream of BCL3 since the BAC-clone RP11-927F16 is split from CTD2608C5 and RP11-423N20. Tumor 3676 displays a positive signal constellation suggesting a break downstream of BCL3 with the BAC-clone RP11-423N20 split from CTD2608C5 and RP11-927F16. (C) Histology (hematoxilin & eosin staining) and immunohistochemistry images of tumor 1 from the validation cohort. Low power magnification (50x) of lymph node shows clear proliferation centers. CD20 shows diffuse positivity (100x). CD23 is only partially and faintly expressed in proliferation centers (100x). CD3 highlights few admixed T cells (100x). CD5 shows few admixed T cells (strong staining intensity) and low expression in tumor cells in the lymph node (100x). LEF1 shows expression in T cells and few cells in proliferation centers but mainly negative in tumor cells (100x). BCL3-R: BCL3 rearrangement. Haematologica | 109 - February 2024

504

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms aCLL (n=8) or CLL (n=3) and two as leukemic non-nodal MCL. The aCLL had bright B-cell markers and LEF1 was negative in the six tumors studied. Trisomy 12 was present in eight of eleven and six of seven had U-IGHV. Lymph nodes examined in four cases were consistent with CLL, including prominent proliferation centers in two patients (Figure 7C; Online Supplementary Figure S10A). The two MCL were leukemic non-nodal, with CCND1 rearrangement and overexpression, and SOX11 negative (Online Supplementary Figure S10B). Three of the four patients with downstream BCL3-R were SMZL, one of them with atypical features previously published,6 splenomegaly and leukemic disease. Two cases carried del(7)(q32) and one case studied mutations frequent in SMZL (TNFAIP3, NOTCH1, KMT2D) (Online Supplementary Table S21).39

Discussion In this study, we characterized the breakpoints of t(14;19) at base-pair resolution in 13 patients with B-cell neoplasms in whom the BCL3 rearrangement had been detected by FISH. These tumors showed marked molecular, pathological, and clinical differences according to the location of the breakpoint in the 5’ or 3’ BCL3 region, suggesting that they correspond to different entities. Specifically, tumors upstream BCL3-R showed BCL3 overexpression, unmutated IGHV, low genomic complexity, trisomy 12, gene mutations and mutational signatures typically observed in CLL. In contrast, tumors with downstream BCL3-R did not upregulate BCL3 and carried MIGHV, high genomic complexity, and mutations typically observed in MZL. Intriguingly, all the breakpoints in the IGHV were mediated by aberrant CSR, but eight of the nine tumors with the 5’ BCL3 breakpoints had U-IGHV and six of them had 100% identity with the germline, consistent with the fact that CSR occurs before germinal cell commitment and initiation of somatic mutations in the immunoglobulin genes.40,41 The pathological features of both subgroups were atypical for CLL or MZL, raising difficulties in their precise taxonomic classification. Upstream BCL3-R tumors have characteristics supporting their relationship with CLL including the presence of nodal proliferation centers in some tumors, trisomy 12 in virtually all tumors, and mutations in genes seen in CLL and uncommon in other lymphoid neoplasms. However, the cytological and phenotypic features of most tumors are not completely typical of CLL with bright expression of B-cell antigens and surface Ig, weak or negative CD23 and the expression profile of a subset of genes different from that seen in UCLL with trisomy 12 such as negative/low expression of LEF1 and upregulation of EBF1 among others. In addition, the BCR signaling response was lower than that in U-CLL

A. Carbó-Meix et al. with trisomy 12. These findings were confirmed in the validation cohort and suggest that lymphoid neoplasms with upstream BCL3-R may correspond to a distinct atypical subset of CLL. Downstream BCL3-R tumors had features of MZL with the presence of villous lymphocytes and genetic alterations frequently seen in these tumors (KLF2, NOTCH2, TBL1XR1). However, they also had some atypical characteristics, such as the exclusive leukemic presentation for 5 and 11 years in two patients and large cell transformation in two of them, an event only seen in 10-15% of SMZL cases.42 Three of four tumors with downstream BCL3-R in the validation series were also SMZL, two of them with del(7)(q32).6 The candidate gene of the downstream BCL3-R is unclear. We could only study one of these cases using RNA-seq, which overexpressed NECTIN2. This gene, also known as PVRL2 or CD112, is a member of immunoglobulin-like cell adhesion molecules and a ligand for natural killer cells. Although its potential oncogenic role is unknown, translocations of this gene with IG and T-cell recepetor have been detected in occasional DLBCL and peripheral T-cell lymphomas, respectively.43,44 Further studies are required to determine whether tumors with downstream BCL3-R are a homogeneous group within the marginal zone spectrum. The biological and clinical differences between tumors with 5’ and 3’ BCL3-R observed in our study may explain the heterogeneity described in the literature. Most of the published tumors resemble our atypical CLL subgroup with an increased frequency of trisomy 12, U-IGHV, and atypical morphology and immunophenotype, although some tumors have also been described as having typical CLL features.7-9,45 The other subgroup is more heterogeneous with frequent M-IGHV and also MZL characteristics, although with occasional atypical features. Some of the tumors had large B-cell morphology similar to our transformed 3’ BCL3-R tumors.6,7,46 The possible prognostic impact of BCL3-R in lymphoid neoplasms in the literature is also controversial. Some studies have indicated that CLL or aCLL with BCL3-R have an adverse prognosis8,47-49 but this was not confirmed by others.50 Our patients with upstream BCL3-R had a similar time to the first treatment and overall survival as U-CLL with trisomy 12. Our new BCL3-R FISH assay identified two breakpoints in 11 of 12 (92%) initial tumors studied and in all 17 independent lymphoid neoplasms, 13 with a 5’ breakpoint and four with a 3’ breakpoint. Interestingly, 11 tumors with upstream BCL3-R had pathological and genetic features similar to those of aCLL/CLL with U-IGHV, trisomy 12, and negative LEF1 expression. The tumors with the 3’ breakpoint were three SMZL, with some atypical features.6 These results confirm the value of this new FISH assay in identifying different BCL3 breakpoints and diseases. The finding of a 5’ BCL3-R in two nnMCL suggests that, similar

Haematologica | 109 - February 2024

505

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms to other translocations in lymphoid neoplasms, BCL3-R is not specific to a single entity and needs to be interpreted in the appropriate context. In conclusion, identification of breakpoints upstream or downstream of BCL3 revealed two different subgroups of lymphoid neoplasms. Tumors with a 5’ breakpoint may correspond to a distinct subset of aCLL/CLL with distinct (epi)genomic, transcriptomic, and clinicopathological features, whereas 3’-rearranged tumors appear to be in the MZL spectrum. We developed a novel FISH assay that recognizes these two BCL3 breakpoints and is therefore useful in clinical practice to identify the two subgroups of patients.

A. Carbó-Meix et al. FN analyzed and interpreted the data, supervised the bioinformatic analyses, wrote the manuscript, and contributed to the design of the study. EC reviewed and supervised the pathology, analyzed and interpreted the data, wrote the manuscript, and designed the study.

Acknowledgments The authors thank the Hematopathology Collection registered at the Biobank of Hospital Clí nic - Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), the Biobank HUB-ICO-IDIBELL (PT17/0015/0024), integrated in the Spanish Biobank Network and funded by Instituto de Salud Carlos III (PT17/0015/0024), and Xarxa de Bancs de Tumors de Catalunya sponsored by Pla Director d’OncoloDisclosures gia de Catalunya (XBTC), and the Molecular Cytogenetics MJB is currently an employee of Swedish Orphan Biovit- Platform of IMIM, Hospital del Mar (Barcelona) for providing rum. FN received honoraria from Janssen, AbbVie, and BAC clones. This work was partially developed at the SOPHiA GENETICS for speaking in educational activities. Center Esther Koplowitz (CEK, Barcelona, Spain). EC has been a consultant for Takeda, NanoString, AbbVie and Illumina; has received honoraria from Janssen, EUSA Funding Pharma, Takeda and Roche for speaking at educational This study was supported by the “la Caixa" Foundation activities and research funding from AstraZeneca and is (CLLEvolution - LCF/PR/HR17/52150017 [HR17-00221LCF] an inventor on two patents filed by the National Institutes and CLLSYSTEMS - LCF/PR/HR22/52420015 [HR22-00172] of Health, National Cancer Institute: “Methods for selecting Health Research 2017 and 2022 Programs, to EC), the Euroand treating lymphoma types,” licensed to NanoString pean Research Council (to EC and JIM-S) under the EuroTechnologies, and “Evaluation of mantle cell lymphoma pean Union’s Horizon 2020 research and innovation and methods related thereof”, not related to this project. program (810287, BCLLatlas, to EC), Ministry of Science and FN and EC licensed the use of the protected IgCaller al- Innovation (MCIN) /AEI/10.13039/501100011033/ and Eurogorithm for Diagnóstica Longwood. The remaining authors pean Regional Development Fund “Una manera de hacer have no conflicts of interest to disclose. Europa” (PID2021-123054OB-I00 to EC) and the Generalitat de Catalunya Suport Grups de Recerca AGAUR (2021-SGRContributions 01172 to EC and 2021-SGR-01293 to SB). HP-A is a recipient AC-M analyzed and interpreted the WGS, RNA-seq, and of a pre-doctoral fellowship from the Spanish Ministry of DNA methylation data and wrote the manuscript. FG col- Science, Innovation and Universities (FPU19/03110). MD-F lected the samples and clinical data, reviewed the histol- acknowledges the research support from the AECC Scienogy, and contributed to manuscript preparation. LW tific Foundation. FN acknowledges research support from reviewed the pathology and contributed to the manuscript the American Association for Cancer Research (2021 AACRpreparation. MG performed custom FISH experiments and Amgen Fellowship in Clinical/Translational Cancer Recontributed to the manuscript preparation. RR designed search, 21-40-11-NADE), European Hematology Association and performed the bioinformatics pipelines for WGS and (EHA Junior Research Grant 2021, RG-202012-00245), and RNA-seq data analyses and contributed to the manuscript Lady Tata Memorial Trust (International Award for Research preparation. GF performed the immunohistochemistry ex- in Leukemia 2021-2022, LADY_TATA_21_3223). EC is an Acaperiments and contributed to manuscript preparation. HP demia Researcher of the “Institució Catalana de Recerca i performed the calcium flux analyses and contributed to Estudis Avançats” (ICREA) of the Generalitat de Catalunya. manuscript preparation. MB contributed to the cases, reviewed the pathology, and prepared the manuscript. GC, Data-sharing statement MD-F, JL, IG, M-JB, JTN, BE, APu, GT, LB, GDC, EB, FC, IR- Whole genome sequencing, RNA-sequencing, and DNA C, MF-C, EDB, JDN, AP, DV, MR, MA, CS, PB, MP, LY, JXO, ES, methylation data are available from the European GeTZ, JRC, SHS, JIM-S, DC, EM, SB and DC provided samples nome–phenome Archive (http://www.ebi.ac.uk/ega/) under and/or data, performed experiments, and interpreted data. accession no. EGAS00001007465.

Haematologica | 109 - February 2024

506

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms

A. Carbó-Meix et al.

References 1. Michaux L, Mecucci C, Stul M, et al. BCL3 rearrangement and t(14;19)(q32;q13) in lymphoproliferative disorders. Genes Chromosomes Cancer. 1996;15(1):38-47. 2. Palmer S, Chen YH. Bcl-3, a multifaceted modulator of NFkappaB-mediated gene transcription. Immunol Res. 2008;42(1–3):210–218. 3. Liu H, Zeng L, Yang Y, Guo C, Wang H. Bcl-3: a double-edged sword in immune cells and inflammation. Front Immunol. 2022;13:847699. 4. Zhang X, Paun A, Claudio E, Wang H, Siebenlist U. The tumor promoter and NF-κB modulator Bcl-3 regulates splenic B cell development. J Immunol. 2013;191(12):5984-5992. 5. Ong ST, Hackbarth ML, Degenstein LC, Baunoch DA, Anastasi J, McKeithan TW. Lymphadenopathy, splenomegaly, and altered immunoglobulin production in BCL3 transgenic mice. Oncogene. 1998;16(18):2333-2343. 6. Soma LA, Gollin SM, Remstein ED, et al. Splenic small B-cell lymphoma with IGH/BCL3 translocation. Hum Pathol. 2006;37(2):218-230. 7. Martín-Subero JI, Ibbotson R, Klapper W, et al. A comprehensive genetic and histopathologic analysis identifies two subgroups of B-cell malignancies carrying a t(14;19)(q32;q13) or variant BCL3translocation. Leukemia. 2007;21(7):1532-1544. 8. Kelly RJ, Wright D, Patil K, et al. t(14;19)(q32;q13) incidence and significance in B-cell lymphoproliferative disorders. Br J Haematol. 2008;141(4):561-563. 9. Huh YO, Schweighofer CD, Ketterling RP, et al. Chronic lymphocytic leukemia with t(14;19)(q32;q13) is characterized by atypical morphologic and immunophenotypic features and distinctive genetic features. Am J Clin Pathol. 2011;135(5):686-696. 10. Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009;25(14):1754-1760. 11. Nadeu F, Mas-de-les-Valls R, Navarro A, et al. IgCaller for reconstructing immunoglobulin gene rearrangements and oncogenic translocations from whole-genome sequencing in lymphoid neoplasms. Nat Commun. 2020;11(1):3390. 12. Brochet X, Lefranc M-P, Giudicelli V. IMGT/V-QUEST: the highly customized and integrated system for IG and TR standardized V-J and V-D-J sequence analysis. Nucleic Acids Res. 2008;36(Web Server issue):W503-508. 13. Nadeu F, Royo R, Massoni-Badosa R, et al. Detection of early seeding of Richter transformation in chronic lymphocytic leukemia. Nat Med. 2022;28(8):1662-1671. 14. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. 15. Puente XS, Beà S, Valdés-Mas R, et al. Non-coding recurrent mutations in chronic lymphocytic leukaemia. Nature. 2015;526(7574):519-524. 16. Lu J, Cannizzaro E, Meier-Abt F, et al. Multi-omics reveals clinically relevant proliferative drive associated with mTORMYC-OXPHOS activity in chronic lymphocytic leukemia. Nat Cancer. 2021;2(8):853-864. 17. Ritchie ME, Phipson B, Wu D, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43(7):e47. 18. Aryee MJ, Jaffe AE, Corrada-Bravo H, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics.

2014;30(10):1363-1369. 19. Bailey TL, Johnson J, Grant CE, Noble WS. The MEME suite. Nucleic Acids Res. 2015;43(W1):W39-W49. 20. Castro-Mondragon JA, Riudavets-Puig R, Rauluseviciute I, et al. JASPAR 2022: the 9th release of the open-access database of transcription factor binding profiles. Nucleic Acids Res. 2022;50(D1):D165-D173. 21. Ventura RA, Martin-Subero JI, Jones M, et al. FISH analysis for the detection of lymphoma-associated chromosomal abnormalities in routine paraffin-embedded tissue. J Mol Diagn. 2006;8(2):141-151. 22. Navarro A, Clot G, Martínez-Trillos A, et al. Improved classification of leukemic B-cell lymphoproliferative disorders using a transcriptional and genetic classifier. Haematologica. 2017;102(9):e360-e363. 23. Seifert M, Sellmann L, Bloehdorn J, et al. Cellular origin and pathophysiology of chronic lymphocytic leukemia. J Exp Med. 2012;209(12):2183-2198. 24. Gutierrez A, Tschumper RC, Wu X, et al. LEF-1 is a prosurvival factor in chronic lymphocytic leukemia and is expressed in the preleukemic state of monoclonal B-cell lymphocytosis. Blood. 2010;116(16):2975-2983. 25. Kulis M, Heath S, Bibikova M, et al. Epigenomic analysis detects widespread gene-body DNA hypomethylation in chronic lymphocytic leukemia. Nat Genet. 2012;44(11):1236-1242. 26. Duran-Ferrer M, Clot G, Nadeu F, et al. The proliferative history shapes the DNA methylome of B-cell tumors and predicts clinical outcome. Nat Cancer 2020;1(11):1066–1081. 27. Griffen TL, Dammer EB, Dill CD, et al. Multivariate transcriptome analysis identifies networks and key drivers of chronic lymphocytic leukemia relapse risk and patient survival. BMC Med Genomics. 2021;14(1):171. 28. Pinnell N, Yan R, Cho HJ, et al. The PIAS-like coactivator Zmiz1 is a direct and selective cofactor of Notch1 in T cell development and leukemia. Immunity. 2015;43(5):870-883. 29. Cook ME, Jarjour NN, Lin C-C, Edelson BT. Transcription factor Bhlhe40 in immunity and autoimmunity. Trends Immunol. 2020;41(11):1023-1036. 30. Rauschmeier R, Reinhardt A, Gustafsson C, et al. Bhlhe40 function in activated B and TFH cells restrains the GC reaction and prevents lymphomagenesis. J Exp Med. 2022;219(2):e20211406. 31. Papakonstantinou N, Ntoufa S, Tsagiopoulou M, et al. Integrated epigenomic and transcriptomic analysis reveals TP63 as a novel player in clinically aggressive chronic lymphocytic leukemia. Int J Cancer. 2019;144(11):2695-2706. 32. Roadcap DW, Clemen CS, Bear JE. The role of mammalian coronins in development and disease. Subcell Biochem. 2008;48:124-135. 33. Seda V, Vojackova E, Ondrisova L, et al. FoxO1-GAB1 axis regulates homing capacity and tonic AKT activity in chronic lymphocytic leukemia. Blood. 2021;138(9):758-772. 34. Khanna P, Lee JS, Sereemaspun A, Lee H, Baeg GH. GRAMD1B regulates cell migration in breast cancer cells through JAK/STAT and Akt signaling. Sci Rep. 2018;8(1):9511. 35. Hutterer E, Asslaber D, Caldana C, et al. CD18 (ITGB2) expression in chronic lymphocytic leukaemia is regulated by DNA methylation-dependent and -independent mechanisms. Br J Haematol. 2015;169(2):286-289. 36. Goldin LR, McMaster ML, Rotunno M, et al. Whole exome sequencing in families with CLL detects a variant in Integrin β 2

Haematologica | 109 - February 2024

507

ARTICLE - BCL3 breakpoints identify two B-cell neoplasms associated with disease susceptibility. Blood. 2016;128(18):2261-2263. 37. Dobashi A, Togashi Y, Tanaka N, et al. TP53 and OSBPL10 alterations in diffuse large B-cell lymphoma: prognostic markers identified via exome analysis of cases with extreme prognosis. Oncotarget. 2018;9(28):19555-19568. 38. Schweighofer CD, Coombes KR, Majewski T, et al. Genomic variation by whole-genome SNP mapping arrays predicts timeto-event outcome in patients with chronic lymphocytic leukemia: a comparison of CLL and HapMap genotypes. J Mol Diagn. 2013;15(2):196-209. 39. Grau M, López C, Navarro A, et al. Unraveling the genetics of transformed splenic marginal zone lymphoma. Blood Adv. 2023;7(14):3695-3709. 40. Oppezzo P, Vuillier F, Vasconcelos Y, et al. Chronic lymphocytic leukemia B cells expressing AID display dissociation between class switch recombination and somatic hypermutation. Blood. 2003;101(10):4029-4032. 41. Roco JA, Mesin L, Binder SC, et al. Class-switch recombination occurs infrequently in germinal centers. Immunity. 2019;51(2):337-350. 42. Bastidas‐Mora G, Beà S, Navarro A, et al. Clinico‐biological features and outcome of patients with splenic marginal zone lymphoma with histological transformation. Br J Haematol. 2022;196(1):146-155. 43. Otto C, Scholtysik R, Schmitz R, et al. Novel IGH and MYC translocation partners in diffuse large B-cell lymphomas. Genes

A. Carbó-Meix et al. Chromosomes Cancer. 2016;55(12):932-943. 44. Almire C, Bertrand P, Ruminy P, et al. PVRL2 is translocated to the TRA@ locus in t(14;19)(q11;q13)-positive peripheral T-cell lymphomas. Genes Chromosomes Cancer. 2007;46(11):1011-1018. 45. Chapiro E, Radford-Weiss I, Bastard C, et al. The most frequent t(14;19)(q32;q13)-positive B-cell malignancy corresponds to an aggressive subgroup of atypical chronic lymphocytic leukemia. Leukemia. 2008;22(11):2123-2127. 46. Salido M, Baró C, Oscier D, et al. Cytogenetic aberrations and their prognostic value in a series of 330 splenic marginal zone B-cell lymphomas: a multicenter study of the Splenic B-Cell Lymphoma Group. Blood. 2010;116(9):1479-1488. 47. Busschots AM, Mecucci C, Stul M, et al. Translocation (14;19)(q32;q13.1) in a young patient who developed a large cell lymphoma after an initial diagnosis of CLL. Leuk Lymphoma. 1991;5(4):281-286. 48. Michaux L, Dierlamm J, Wlodarska I, Bours V, Van Den Berghe H, Hagemeijer A. t(14;19)/BCL3 rearrangements in lymphoproliferative disorders: a review of 23 cases. Cancer Genet Cytogenet. 1997;94(1):36-43. 49. Fang H, Reichard KK, Rabe KG, et al. IGH translocations in chronic lymphocytic leukemia: Clinicopathologic features and clinical outcomes. Am J Hematol. 2019;94(3):338-345. 50. Rossi D, Deambrogi C, Monti S, et al. BCL3 translocation in CLL with typical phenotype: assessment of frequency, association with cytogenetic subgroups, and prognostic significance. Br J Haematol. 2010;150(6):702-704.

Haematologica | 109 - February 2024

508

ARTICLE - Hematopoiesis

Prognostic relevance of clonal hematopoiesis in myeloid neoplastic transformation in patients with follicular lymphoma treated with radioimmunotherapy Zhuoer Xie,1,2 Terra Lasho,1 Arushi Khurana,1 Alejandro Ferrer,1 Christy Finke,1 Abhishek A. Mangaonkar,1 Stephen Ansell,1 Jenna Fernandez,1 Mithun Vinod Shah,1 Aref Al-Kali,1 Naseema Gangat,1 Jithma Abeykoon,1 Thomas E. Witzig1 and Mrinal M. Patnaik1 Mayo Clinic, Department of Internal Medicine, Hematology Division, Rochester, MN and

Malignant Hematology Department, H. Lee Moffitt Cancer Center & Research Institute,

Tampa, FL, USA

Correspondence: M. Patnaik Patnaik.Mrinal@mayo.edu Received: Accepted: Early view:

June 13, 2023. August 16, 2023. August 31, 2023.

Abstract While novel radioisotope therapies continue to advance cancer care, reports of therapy-related myeloid neoplasms (t-MN) have generated concern. The prevalence and role of clonal hematopoiesis (CH) in this process remain to be defined. We hypothesized that: (i) CH is prevalent in relapsed follicular lymphoma and is associated with t-MN transformation, and (ii) radiation in the form of radioimmunotherapy (RIT) plays a role in clonal progression. In this retrospective cohort study, we evaluated the prevalence and prognostic impact of CH on clinical outcomes in 58 heavily pre-treated follicular lymphoma patients who received RIT. Patients had been given a median of four lines of therapy before RIT. The prevalence of CH prior to RIT was 46%, while it was 67% (P=0.15) during the course of RIT and subsequent therapies in the paired samples. Fourteen (24%) patients developed t-MN. Patients with t-MN had a higher variant allele fraction (38% vs. 15%; P=0.02) and clonal complexity (P=0.03) than those without. The spectrum of CH differed from that in age-related CH, with a high prevalence of DNA damage repair and response pathway mutations, absence of spliceosome mutations, and a paucity of signaling mutations. While there were no clear clinical associations between RIT and t-MN, or overall survival, patients with t-MN had a higher mutant clonal burden, along with extensive chromosomal abnormalities (median survival, afer t-MN diagnosis, 0.9 months). The baseline prevalence of CH was high, with an increase in prevalence on exposure to RIT and subsequent therapies. The high rates of t-MN with marked clonal complexities and extensive chromosomal damage underscore the importance of better identifying and studying genotoxic stressors accentuated by therapeutic modalities.

Introduction Clonal hematopoiesis (CH) is defined by the acquisition and subsequent expansion of somatic DNA variants, including somatic mutations and copy number alterations in hematopoietic stem and progenitor cells (HSPC).1-4 When CH mutations occur in leukemia-associated genes, with a variant allele fraction (VAF) ≥2% in individuals without a diagnosed hematologic disorder, the condition is termed CH of indeterminant potential. While CH is ubiquitous with aging, context-specific development of CH is heterogeneous and dependent on clonal selection pressures. CH mutations have differential rates of fitness and stability and expand based on clonal selection pressures.5 Retrospective series have demonstrated the role of CH in therapy-related myeloid neoplasms (t-MN) and have

documented associations with inferior overall survival (OS) in the setting of prior cytotoxic therapies.6 CH of indeterminate potential is considered the first step in a multi-hit model for the development of t-MN.7 DNA-damage-inducing therapies such as chemotherapy or radiation used in primary cancers can lead to collateral alterations in HSPC, resulting in clonal populations with enhanced fitness and propagation potential,6,8,9 particularly when they involve DNA damage response and repair (DDR) genes, such as TP53 and PPM1D. 6,8,10-12 Although the impact of CH on t-MN has been extensively investigated in settings of chemoradiation therapy,6,8,9 autologous stem cell transplantation (ASCT),13-15 and chimeric antigen receptor (CAR) T-cell therapy,16,17 the prevalence and impact of CH in the context of systemic radioisotope therapy

Haematologica | 109 February 2024

509

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

remains to be further clarified.18 The need for these data is particularly relevant, given the increasing use of radioisotopes.19-22 A few examples include lutetium dotatate for neuroendocrine tumors and metastatic prostate cancer and radioimmunotherapy (RIT) 90Y ibritumomab tiuxetan (90YIT, Acrotech Biopharma) for relapsed. low-grade and follicular non-Hodgkin lymphoma (FL). It has been reported that the incidence of t-MN is between 5-10% after chemotherapy with ASCT and 2-20% in the context of radioisotope therapy combined with cytotoxic chemotherapy in non-Hodgkin lymphoma. 23-29 Given the evidence of a higher risk of t-MN in patients with CH of indeterminate potential, we hypothesized that exposure to β radiation would enhance the prevalence and growth of CH in HSPC, resulting in a higher prevalence of t-MN, with CH negatively impacting OS. The group of patients that we studied was unique because all patients had normal bone marrow morphology for t-MN and normal cytogenetics prior to RIT and had a very mature follow-up duration, providing valuable data on the role of clonal progression in the

context of cancer-directed therapies.

Methods Cohort of patients After institutional review board approval, we identified 58 patients with relapsed FL treated with RIT who had cryopreserved peripheral blood DNA for CH assessment at the Mayo Clinic. Of note, all samples were banked in the relapsed or refractory disease phase. Among these, 24 (group 1) had paired samples before and after RIT exposure; 22 (group 2) had one or more samples after RIT exposure, and 13 (group 3) had samples only before RIT exposure. Given our hypothesis that RIT could cause clonal expansion secondary to its genotoxic effects, we compared clonal VAF and complexity in group 1 (n=24 patients with paired samples); we also identified 13 patients in group 2 who had two serial cryopreserved samples after RIT exposure (group 2a) and five patients in group 3 who had two serial samples

Figure 1. Swimmer plot for group 1 patients (patients with paired samples, with radioimmunotherapy administered between samples). PT_9, 30, 31, and 32 received a total of 6, 6, 3, and 5 different therapies, respectively. RIT: radioisotope therapy; PT: patient; NA: not available; CH: clonal hematopoiesis; TMN: treatment-related myeloid malignancy. Haematologica | 109 February 2024

510

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

Figure 2. Swimmer plot for group 2 patients (patients with one or more samples taken after radioimmunotherapy). Symbols and abbreviations as in Figure 1.

Figure 3. Swimmer plot for group 3 patients (patients with samples taken only prior to radioimmunotherapy). Symbols and abbreviations as in Figure 1.

prior to RIT exposure (group 3a), and we compared clonal evolution in subgroups 2a and 3a (Figures 1-3). Clinical data, including prior therapy regimens, were retrospectively abstracted from clinical records. Outcomes of interest included the spectrum and diversity of CH, clonal dynamics, the development of t-MN, and OS. Detection of clonal hematopoiesis DNA was extracted from peripheral blood mononuclear cells and subjected to a customized, targeted next-gen-

eration sequencing assay, as previously described30 (Online Supplementary Material, Online Supplementary Table S1). A VAF ≥2% was considered as CH. Cryopreserved bone marrow DNA, collected at the time of t-MN diagnosis, was also available for eight patients and was subjected to sequencing with the same panel. Based on the recent description of the involvement of genes in lymphoproliferative disorders, we considered ARID1A, ARID1B, CARD11, CD79B, CREBBP, EP300, EZH2 (gain-of-function variants only), HIST1H1C, HIST1H1D, KMT2D, NOTCH1, STAT6, and TNFRSF14 mutations

Haematologica | 109 February 2024

511

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

as lymphoid CH (L-CH) and the rest as myeloid CH (M-CH), including loss-of-function EZH2 mutations.31,32 We classified our observations based on the following categories: L-CH, M-CH, both L-CH and M-CH (LM-CH), DDR mutations and mutations in DNMT3A, TET2, and ASXL1 (the so-called DTA genes). For patients with multiple pathogenic variants, we used the maximum VAF for VAF comparisons. Mutation patterns were analyzed using ProteinPaint.33 Statistical analysis We compared clinical characteristics, mutation patterns, VAF, and outcomes of patients with and without t-MN. Continuous variables are presented as a median with interquartile range (IQR) or mean with standard deviation, and categorical variables as frequency (percentage). Differences in the distribution of nonparametric continuous variables between categories were compared using the Wilcoxon matched pairs signed rank test for paired samples. Categorical variables were compared using the χ2 or Fisher exact test. OS was measured from the date of RIT exposure to the date of death from any cause; data were censored at the time patients were last known to be alive. The univariable logistic regression model was used to evaluate potential risk factors for outcomes. The median point estimate and 95% confidence interval (95% CI) for follow-up time, t-MN, and OS were estimated using the Kaplan-Meier method. All P values were two-sided tests. All statistical calculations

were carried out using R version 4.0.1. Considering the hypothesis-generating character of the study, no multiple testing correction was implemented, and the reported P values should be interpreted as exploratory.

Results Patients’ characteristics and clinical outcomes Fifty-eight patients with relapsed FL were included. Their median age was 48.5 years (range, 28-79) and 23 (40%) were females. The median follow-up duration was 17.8 years (IQR, 15.7-21.9). At the last follow-up, 14 (24%) patients had developed t-MN, and 23 (40%) had died, including eight deaths from t-MN. Eleven (19%) patients had diffuse large B-cell lymphoma transformation and in this group, three (17.5%) developed t-MN. The median time from FL diagnosis to t-MN was 14.4 years (IQR, 11.3, 18.6), with the median latency from RIT to t-MN diagnosis being 8.8 years (IQR, 6.7-12.6) and the median OS after t-MN diagnosis being 0.9 years (IQR, 0.24-2.2). All 58 patients had significant exposures to chemo-immunotherapy or involved field radiation therapy with a median of four prior regimens (Table 1). There were no significant differences in the total number of therapies received between patients with and without CH (median 5 vs. 4; P=0.5). Twenty-eight patients (48%) had received purine/pyrimidine analogs, 23 (57%) DNA topoisomerase

Figure 4. Mutation prevalence and co-mutation status. The light blue color indicates the number of patients with the specific mutation, and the light pink color indicates the number of patients with other mutations. Haematologica | 109 February 2024

512

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

inhibitors, 52 (90%) alkylating agents, and 21 (36%) ionizing radiation therapy. In the 14 patients who developed t-MN, all had been previously exposed to chemotherapy, including 11 (79%) with exposure to alkylating agents and eight (57%) with exposure to topoisomerase inhibitors. In addition, eight (57%) had undergone prior conventional radiation, including six (43%) exposed to alkylating agents/topoisomerase inhibitors and radiation therapy. Prevalence and mutational spectrum of clonal hematopoiesis Immediately prior to RIT administration, all 58 patients had a bone marrow biopsy in which no morphological atypia was found, and all patients had a normal karyotype. Despite these normal findings, the prevalence of CH at any timepoint was 60% (35/58), with 97 somatic variants identified. The most frequent mutations were DNMT3A (25%), followed by PPM1D (23%), KMT2D (8%), and TP53 (8%) (Figure 4, Online Supplementary Figure S1). Among patients with CH, 12 (34%) had one mutation, and 23 (66%) had two or more mutations. The median VAF was 19% (IQR, 4-39%). The co-mutation status and VAF are shown in Figure 4

and Online Supplementary Figure S2. In the entire cohort, 21 (60%) patients had M-CH only, two (6%) had L-CH only, and 12 (34%) had LM-CH. The pathogenic variants for each patient are listed in Online Supplementary Table S2. The oncoplot for the entire cohort is shown in Figure 5. The common mutation patterns were missense, followed by nonsense and frameshift mutations. Clonal expansion after radioimmunotherapy In patients with paired samples (n=24; group 1), the median time interval between the samples was 1 year. The prevalence of CH before and after RIT exposure was 46% versus 67% (P=0.15) and the prevalence of M-CH was 42% versus 61% (P=0.19). There were 17 and 33 variants identified in the pre- and post-exposure samples, respectively, with DNMT3A being the most frequent (47% and 30%), followed by PPM1D (12% and 21%). Five patients without CH prior to RIT had M-CH in the post-RIT samples. (Online Supplementary Figure S2: PT_36, PT_42, and PT_44 with DNMT3A mutations and PT_38 with mutations in DNMT3A and PPM1D, and PT_45 with a MED12 mutation). The median VAF was not significantly different between the paired

Figure 5. Co-mutation plot showing mutations present in 35 patients with somatic mutation(s). Each column represents a single patient. The top row denotes the translational effect and mutations per megabase. The bar graph on the left designates the prevalence of mutations (count for once regardless of the prevalence/patient). The mutation subtypes are represented by colors, red indicates nonsense, light-blue indicates missense; yellow indicates frameshift; gray indicates splice effect; purple indicates duplication, pink indicates deletion. The bar at the bottom designates the proportion of the mutation subtypes for each patient. Proportions are from 0 (bottom) to 25%, 50%, 75%, and 100% (top). MB: megabase. Haematologica | 109 February 2024

513

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

samples (19.1% vs. 14.3%; P=0.27) (Figure 6). The median annual change in VAF for DNMT3A (n=15) was 2% (range, -8% to 14%), and that for DDR mutations (n=8) was 4.5% (range, 2% to 19%). Four patients developed t-MN in this group, three with existing CH before the diagnosis of t-MN and one without (Figure 1, Online Supplementary Table S3). Online Supplementary Figures S3 and S4 demonstrate the changes in clone VAF and complexity in this group. In patients with two serial samples from subgroups 2a and 3a (Figures 2 and 3), there were no differences in the prevalence of CH or M-CH between the samples at the two timepoints in either group. In subgroup 2a the prevalence of CH at the two timepoints was 69% versus 75% (P=0.75) and the prevalence of M-CH was 62% versus 67% (P=0.79). In subgroup 3a, at both timepoints the prevalence of CH was 60% and that of M-CH was 40%. Furthermore, there were no significant difference in VAF between the two timepoint samples in subgroups 2a or 3a (VAF in subgroup 2a, 15.5% vs. 14.8%, P=0.73; VAF in subgroup 3a, 29% vs. 23.8%, P>0.99) (Figure 6). These results may be limited by the smaller sample size and shorter time interval between sampling points (time intervals were 1, 0.48, and 0.04 years for group 1, subgroup 2a, and subgroup 3a, respectively). These time intervals were statistically not different (P=0.32) and hence we were not able to draw additional conclusions on the impact of time differences on CH prevalence and CH-VAF changes. CREBBP and KMT2D are FL-associated genes and were annotated as L-CH in our cohort: these clones decreased with lymphoma-directed therapy and could in fact represent circulating tumor cells that were inadvertently included in the peripeheral blood mononuclear cell fraction. In addition, all CH mutations became undetectable in one patient who underwent allogeneic stem cell transplantation (Online Supplementary Figure S3; PT_43). Clinical association with therapy-related myeloid neoplasms Fourteen (24%) patients developed t-MN (7 with acute

myeloid leukemia, 7 with myelodysplastic syndromes). Their median age was 51.5 years (range, 33-79) and nine (64%) were males. There were no significant differences in baseline age (P=0.44), gender (P=0.72), FL stage (P=0.54), grade (P=0.89), Follicular Lymphoma International Prognostic Index (FLIPI) score (P=0.71), or the number of prior therapies (median, 4 in both groups) (P=0.66), including prior ASCT (P=0.18), between the t-MN and non-t-MN groups. Among the 97 identified somatic variants, 27 (28%) were in the t-MN group and 70 (72%) in the non-t-MN group. The most frequent variants seen in the t-MN group included PPM1D (n=7, 28%), DNMT3A (n=3, 11%), and TP53 (n=2, 8%), whereas the most frequent variants seen in the non-t-MN group included DNMT3A (n=16, 23%), PPM1D (n=15, 21%), KMT2D (n=7, 10%), and TP53 (n=6, 9%). The prevalence of CH was not significantly different between the t-MN (64%) and non-t-MN (59%) groups (P=0.97). However, the t-MN group had a higher CH complexity (M-CH other than DDR and DTA mutations) (P=0.03) and a higher VAF compared to those without (38% vs. 15%; P=0.02). (Table 1). The presence of CH prior to RIT or after therapy was not associated with t-MN development (P=0.6 and P=0.73, respectively). Detailed clinical, molecular, and cytogenetic information on patients with t-MN are provided in Table 2. Thirteen of the 14 evaluable patients with t-MN had an abnormal karyotype. These abnormal karyotypes comprised seven (50%) monosomal karyotypes, one (7%) complex karyotype, four chromosome 17 abnormalities, eight (57%) chromosome 5 abnormalities, and 11 (79%) chromosome 7 abnormalities. Twenty-one variants were found in the eight sequenced t-MN bone marrow samples, with 19 new pathogenic variants emerging that were not present in the pre-t-MN assessments (Table 2). Among these eight patients, four (50%) had no CH in the prior samples. Common mutations gained at t-MN transformation were TP53 (n=6, 32%), with four patients having biallelic TP53 inactivation. PPM1D and SETBP1

Figure 6. Changes in variant allele fraction in patients with paired samples in group 1, with two paired serial samples in subgroup 2a; and with two paired serial samples in subgroup 3a. VAF: variant allele fraction; RIT: radioimmunotherapy. Haematologica | 109 February 2024

514

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

Table 1. Clinical and demographic comparisons between patients with or without therapy-related myeloid neoplasms. Characteristic N Age in years, median (IQR) Sex, N (%) Female Male

FNHL grade, N (%) 1 2 3 Missing information

FNHL stage, N (%) I II III IV Missing information

t-MN

Non-t-MN

51.5 (44.3-63) 48 (41- 61) 5 (36) 9 (64)

18 (41) 26 (59)

7 (50) 5 (36) 0 (0) 4 (29)

22 (50) 17 (39) 1 (2) 2 (4)

1 (7) 2 (14) 0 (0) 9 (64) 1 (7)

4 (9) 4 (9) 8 (18) 26 (59) 2 (4)

P 0.46 0.97 0.89

0.53

FLIPI score, N (%) 0 1 2 3 4 Missing information

1 (7) 3 (21) 5 (36) 4 (29) 0 (0) 1 (7)

4 (9) 6 (14) 24 (55) 6 (14) 1 (2) 3 (7)

CH, N of pts (%)

9 (64)

26 (59)

0.97

DTA mutations, N of pts (%) DNMT3A TET2 AXSL1

4 (15) 3 (11) 0 (0) 1 (4)

18 (26) 16 (23) 1 (1) 1 (1)

0.14 0.19 0.48

DDR mutations, N of pts (%) TP53 PPM1D

9 (33) 2 (7) 7 (26)

21 (30) 6 (9) 15 (21)

0.54

0.75 0.85 0.63

Characteristic

t-MN

Non-t-MN

Other myeloid CH, N of pts (%) ASXL2 BRCC3 CBL CDKN2A IDH2 LTB MED12 NF1 PHIP POU2AF1 PTEN SBDS SMC3 STAG2 STAT3 TET1

9 (33) 0 (0) 1 (4) 1 (4) 0 (0) 1 (4) 1 (4) 1 (4) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (4) 1 (4) 1 (4) 0 (0)

10 (14) 1 (1) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 1 (1) 1 (1) 2 (3) 1 (1) 1 (1) 1 (1) 0 (0) 0 (0) 0 (0) 1 (1)

0.03 0.48 0.48 0.48 -

Lymphoid CH, N of pts (%) ARID1A ARID1B CAR11 CD79B CREBBP EP300 EZH2 HIST1HIC KMT2D STAT6 TNFRSF14

0 (0) 1 (4) 0 (0) 0 (0) 1 (4) 0 (0) 1 (4) 0 (0) 1 (4) 1 (4) 0 (0)

21 (30) 1 (1) 0 (0) 1 (1) 1 (1) 4 (6) 1 (1) 2 (3) 2 (3) 7 (10) 1 (1) 1 (1)

0.38 0.68 0.83 0.31 0.48 -

N of therapies, median (IQR)

4 (3-6)

4 (2-6)

0.9

Death, N (%)

8 (48)

15 (34)

0.22

t-MN: therapy-related myeloid neoplasms; IQR: interquartile range; FNHL: follicular, non-Hodgkin lymphoma; FLIPI: Follicular Lymphoma International Prognostic Index; CH: clonal hematopoiesis; pts: patients; DTA: DNMT3A, TET2, and ASXL1; DDR: DNA damage response and repair.

were the next most frequent mutations seen (n=3, 16%). Within the limitations of the small sample size, we did not see an impact of CH on t-MN development; nine (26%) of 35 patients with CH developed t-MN, compared to five (22%) of 23 without CH (P=0.97). We were also limited in identifying potential risk factors for t-MN development, including CH subtypes, e.g., TP53 (P=0.53) or PPM1D (P=0.4), and the impact of prior therapies (Online Supplementary Figure S5). Risk factors for overall survival Patients with t-MN had a trend towards a shorter OS compared to those without (median OS: 5.16 years vs. 17.8; P=0.1). Age ≥50 years was identified as a risk factor for shorter OS (15-year survival probability: 42% vs. 82%; P=0.001) (Online Supplementary Figure S6). Other variables, including CH (Online Supplementary Figure S7) and its subtype, mutation numbers, and nature and the number of prior cytotoxic therapies, had no impact on OS. The relevance of these

findings is limited by the small sample size and the retrospective nature of the study that was not powered to detect a difference.

Discussion In this retrospective study, we focused on a unique cohort of patients with low-grade relapsed FL, which represents a very particular clinical situation as patients have long survivals, punctuated by multiple treatment modalities.34 This provides a great scenario to study CH in the context of evolving selection pressures. 90Y ibritumomab tiuxetan for FL was the first RIT approved by the US Food and Drug Administration35 in 2002 and, although not currently in use, provides a valuable database with a long follow-up that can serve as a model to inform risk for patients now receiving newer RIT for solid tumors and potentially the re-emergence of RIT for lymphoma. Using paired pre- and

Haematologica | 109 February 2024

515

516

Haematologica | 109 February 2024

NGS: negative in 2019 and 2020

TP53 (89)

No CH at baseline

PPM1D (2) TP53 (5)

ASXL1 (3) NOTCH1 (48) PPM1D (2) PPM1D (3) SMC3 (2)

46, XX, -7[3]/46, idem,+r[17] 42-45, XX, add(5)(q11.2), -7,add(14)(q32),ins(15;?)(q13;?)[3], -18, -21,-21[3],+0-2mar[cp16]/46, XX[4]5q del t-MDS 46,XX,del(1)(p32p36.1), del(5)(q22q35), -7,+8[16]/46,XX[4] progressed to t-AML 46,XX,del(1)(p32p36.1),del(5) (q22q35),-7,+8[20]

t-MDS

t-MDS → AML

t-MDS (MDS-EB-2) → t-AML

t-MDS

No CH at baseline

40-41,XY,add(5)(q11.2),-7,add(8) (p11.2),add(12)(q13),-16, -17,-18,-20,-22, add(22)(q11.2),+04mar[cp15]/78-81,idemx2[2]/46,XY[3]

Continued on following page.

TP53 (37) TP53 (3.6) PPM1D (3.1)

46, XY,del(7)(q22)[10]/46,XY[10]

Pure erythroid leukemia t-AML

NA (CG was normal 2 years ago before t-MN diagnosis)

t-MDS

EZH2 (30) IDH2 (37) PPM1D (32) STAG2 (36)

45,XY,add(5)(q11.2),add(7)(q22), add(10)(q24),-12, add(13)(p11.2),-17, -18,+2mar[19]/46, XY[1]

t-MDS (EB-2)

CBL (18) CREBBP (32) KMT2D (36) LTB (38) STAT6 (36)

del(5q) and monosomy 7

t-AML

PPM1D (45)

PPM1D (19) TP53 (9)

40-43,X,-Y, 3,add(3)(q21), add(5) (q13),del(5)(q13q33), -7,-10, -12, del(12)(q13q24.1), der(15)t(1;15) (q21;q22),-16,inv(16)(p13.3q22),17,add(17)(p11.2),add(18)(p11.2),19,+22,mar[cp20]

DNMT3A, (46) SETBP1 (45) U2AF1 (15)

No CH at baseline

Cytogenetics

Pathogenic variants (VAF %) at t-MN diagnosis

Baseline CH mutations and VAF (%)

Purine/ Age in years TopoN of prior pyrimidine Radiation MDS/AML at FL Sex isomerase Alkylators therapies nucleoside therapy subtype diagnosis inhibitor analog

Pt #

Table 2. Clinical, cytogenetics, and somatic mutations in patients who developed therapy-related myeloid neoplasms.

ARTICLE - CHIP in patients with follicular lymphoma Z. Xie et al.

517

Haematologica | 109 February 2024

46,XY,del(20)(q11.2q13.3) [17]/46,sl,t(3;17)(p21;q25)[3] Complex, details unknown, treated elsewhere

t-MDS (MDS-EB-2) t-AML with monocytic differentiation

No CH at baseline

PPM1D (4)

PPM1D (3.1)

BCOR (76) DNMT3A (40) RUNX1 (8) SF3B1 (41)

SETBP1 (17.8) SETBP1 (7.8)

46,XY,add(2)(p11.2), 3,add(3) (p21),del(5)(q22q35), der(7)t(3;7)(p13;p22),add(8)(q13), der(8)t(8;10)(p21;q11.2),der(10)t(9;10) (q13;q11.2),+11, der(15;17)(q10;q10), DNMT3A (3) +mar[3]/49 53,XY,add(2)(p11.2),add(3) DNMT3A (48) (p13),t(3;7)(p13;p22), del(5)(q22q35), +6,+add(8)(q13)x2,der(8)t(8;10) (p21;q11.2)x2der(10)t(9;10) (q13;q11.2),+11,+13, +19,+0-1r[cp17] 46, XX,-7,+r[20]

DNMT3A (4) TP53 (65.9) STAG2 (10)

45-46,XX,del(1)(p32p36.1),del(5) (q13q33),-7, der(9;18)(p10;q10),add(15)(q22),+r, +0-1mar[cp13]/46,XX[7]

t-AML

STAT3 (1.1) TP53 (60) PPM1D (3.5) TET2 (1.6) ARIDIB (9) MED12 (33) STAT3 (10)

Cytogenetics

Pathogenic variants (VAF %) at t-MN diagnosis

Baseline CH mutations and VAF (%)

Pt#: patient’s number; FL: follicular lymphoma; MDS: myelodysplastic syndrome; AML: acute myeloid leukemia; CH: clonal hematopoiesis; VAF: variant allele fraction; t-MN: therapy-related myeloid neoplasms; F: female; M: male; Y: yes; N: no; t-MDS: therapy-related MDS; EB-2: excess blasts-2; t-AML: therapy-related AML; NA: not available/applicable; NGS: next-generation sequencing; CG: cytogenetics.

Purine/ Age in years TopoN of prior pyrimidine Radiation MDS/AML at FL Sex isomerase Alkylators therapies nucleoside therapy subtype diagnosis inhibitor analog

Pt #

ARTICLE - CHIP in patients with follicular lymphoma Z. Xie et al.

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

post-RIT samples, we found that 42% of patients with relapsed FL had CH despite normal conventional cytogenetic studies at the time of RIT, with an eventual 67% prevalence rate for CH after RIT. In the context of a median follow-up duration of 17.8 years, 24% of patients had t-MN with high clonal burdens and with extensive chromosomal damage. The median latency of t-MN development was 8.8 years after RIT exposure, which highlights the difficulty of fully assessing bone marrow toxicity in the short term for any new agent being tested for the treatment of indolent non-Hodgkin lymphomas. The survival after t-MN diagnosis of less than a year also underscores the dismal outlook and a key unmet need for better treatment for patients with t-MN. The spectrum of CH seen in this group was different from that in classical age-related CH, with a high prevalence of DDR pathway mutations, absence of spliceosome mutations, and a paucity of signaling mutations, including at t-MN diagnosis. We observed clonal evolution in the paired samples before and after RIT without a clear inference of causality between exposure to RIT and clonal expansion. Furthermore, CH was not associated with an increased risk of t-MN or inferior OS. This is likely due to the small sample size and the high prevalence of CH in both the t-MN and non-t-MN groups. In fact, while new CH mutations were encountered at t-MN diagnosis, including somatic TP53 mutations, all patients had extensive chromosomal damage, highlighting the genomic instability seen in this heavily treated patient population. There are several unique findings in our study. First, we found a significantly higher incidence of t-MN in heavily treated FL patients who received RIT, higher than stated in previous reports.23-28,36 Two reports on RIT suggested that the cumulative incidence of t-MN was 2.5% at 5 years37 and 10% at 10 years, with a median latency of 6.6 years.38 In our study, the incidence of t-MN was 24% after 17.8 years of follow-up, with the median latency being 8.8 years. The latency is also strikingly longer than that which can be seen after exposure to alkylating agents (5-7 years) or topoisomerase II inhibitors (1-3 years),39 suggesting mechanistic differences that might contribute to t-MN development. Secondly, in patients with non-Hodgkin lymphoma undergoing ASCT or CAR T-cell therapy, the prevalence of CH was estimated to be 25-50%, with the most common mutations being DNMT3A, PPM1D, TET2, and TP53.13-17,40 In our cohort, the prevalence of CH was higher, at 60%. One explanation is that our CH panel was larger (>200 genes) and was able to detect mutations with greater sensitivity (VAF of >0.5%) compared to other studies.16,41 Consistent with other studies, the most frequent mutations were DNMT3A (25%), PPM1D (23%), KMT2D (8%), and TP53 (8%), which reflect the effect of cytotoxic therapy and the FL genetic landscape.31,42-44 PPM1D mutations were highly enriched in this cohort, suggesting that these mutations represent convergent mechanisms of clonal fitness in the constraints of oncogenic exposures.

Third, in paired samples taken 1 year apart before and after exposure to RIT, we observed clonal evolutionary changes as reflected by an increase, albeit not statistically significant, in the prevalence of CH. In addition, five (21%) patients without CH prior to RIT had CH after RIT. There were no significant changes in mutational VAF, a phenomenon not seen in the other two groups either. These findings may be limited by the retrospective nature of cohort assignment, smaller sample size, and the impact of natural clonal growth rates, which are hard to approximate for individual patients.5 We found that all evaluable patients with t-MN had extensive cytogenetic abnormalities at the time of t-MN, with 79% demonstrating chromosome 5 and/or 7 abnormalities and 36% having chromosome 17 abnormalities, all of which are associated with poor outcomes.45 Our data suggest that radioisotope therapy may play a role in chromosomal damage/alterations; however, it is difficult to parse out the role of RIT on t-MN development as all patients in our study had multiple cumulative exposure to chemoradiation therapy. Our study highlights the somatic genomic landscape in FL patients treated with cytotoxic chemotherapy, ionizing radiation, and radioisotope therapy, demonstrating the impact of these treatment modalities on CH, chromosomal damage, and clonal evolution. The strikingly high prevalence of CH and t-MN highlights the oncogenic potential of such therapies, while the exact mechanisms of oncogenesis and the impact of CH remain to be elucidated. This study has important implications as different types of RIT in cancer are rapidly gaining approval for a variety of malignancies. It has been reported that the incidence of t-MN in recipients of peptide receptor radionuclide therapy ranges between 1.8-5.4%, although the latency of t-MN development in such patients was shorter, suggesting different mechanisms of t-MN development.46,47 Recently, radioligand therapy with 177Lu-PSMA-617 was approved for prostate-specific membrane antigen-positive, metastatic, castration-resistant prostate cancer22 and CAR-T therapy with purine nucleoside analog lymphodepletion has been approved for relapsed and refractory FL.48 Lastly, the widening long-term use of poly ADP ribose polymerase inhibitors in patients with breast, ovarian, and prostate cancers has brought increasing awareness to the problem of t-MN and acute myeloid leukemia.49 It will, therefore, be important to monitor the impact of CH and clonal evolution closely in such patients. The advantages of our study include the baseline bone marrow studies with conventional analysis, the extensive molecular analysis, and the long follow-up duration, all of which confirm the need for long-term monitoring and the relevance to solid tumor oncology in which these agents are gaining popularity. The limitations of our study include the relatively small sample size, retrospective study design, and heterogeneity regarding prior therapies. In summary, we report the prevalence of CH and t-MN in FL patients who relapsed after conventional therapy and then received radiation therapy with RIT. With a

Haematologica | 109 February 2024

518

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma

long follow-up, we found a high prevalence of CH and t-MN in this cohort. The spectrum of CH was unique and different from that in age-related CH, with no clear causality between RIT exposure and clonal expansion or t-MN development. All patients with t-MN had high clonal burdens and demonstrated extensive chromosomal damage, with very poor survival outcomes. Our study provides the rationale for future prospective studies with paired samples before and after RIT interventions, evaluating the impact of CH by specific CH subtype and by clonal complexity, on t-MN occurrence and non-relapse mortality, given the rapidly growing indications for this type of treatment. Disclosures MMP has received research funding from Stem Line Pharmaceuticals and Kura Oncology and has served on an advisory board for CTI Pharmaceuticals. TEW and the Mayo

Clinic received research support from IDEC, Spectrum and Acrotech Biopharma for clinical trials of 90-yttrium ibritumomab tiuxetan. ZX has no conflicts of interest to disclose. Contributions ZX and MMP designed the study using samples from the radioimmunotherapy database (TEW). ZX, MMP and TEW wrote, reviewed, and edited the manuscript. All authors approved the final version. Funding We acknowledge support from the Mayo Clinic SPORE (CA97274), R21 to TEW CA87912, and the Predolin Foundation Biobank. Data-sharing statement Please contact the author for correspondence to discuss data sharing.

References 1. Saiki R, Momozawa Y, Nannya Y, et al. Combined landscape of single-nucleotide variants and copy number alterations in clonal hematopoiesis. Nat Med. 2021;27(7):1239-1249. 2. Jacobs KB, Yeager M, Zhou W, et al. Detectable clonal mosaicism and its relationship to aging and cancer. Nat Genet. 2012;44(6):651-658. 3. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488-2498. 4. Genovese G, Kähler AK, Handsaker RE, Lindberg J, Rose SA, Bakhoum SF. Clonal hematopoiesis and blood-cancer risk inferred from blood DNA sequence. N Engl J Med. 2014;371(26):2477-2487. 5. Fabre MA, de Almeida JG, Fiorillo E, et al. The longitudinal dynamics and natural history of clonal haematopoiesis. Nature. 2022;606(7913):335-342. 6. Coombs CC, Zehir A, Devlin SM, et al. Therapy-related clonal hematopoiesis in patients with non-hematologic cancers is common and associated with adverse clinical outcomes. Cell Stem Cell. 2017;21(3):374-382. 7. Voso MT, Falconi G, Fabiani E. What’s new in the pathogenesis and treatment of therapy-related myeloid neoplasms. Blood. 2021;138(9):749-757. 8. Bolton KL, Ptashkin RN, Gao T, et al. Cancer therapy shapes the fitness landscape of clonal hematopoiesis. Nat Genet. 2020;52(11):1219-1226. 9. Gillis NK, Ball M, Zhang Q, et al. Clonal haemopoiesis and therapy-related myeloid malignancies in elderly patients: a proof-of-concept, case-control study. Lancet Oncol. 2017;18(1):112-121. 10. Wong TN, Ramsingh G, Young AL, et al. Role of TP53 mutations in the origin and evolution of therapy-related acute myeloid leukaemia. Nature. 2015;518(7540):552-555. 11. Hsu JI, Dayaram T, Tovy A, et al. PPM1D mutations drive clonal hematopoiesis in response to cytotoxic chemotherapy. Cell Stem Cell. 2018;23(5):700-713. 12. Takahashi K, Wang F, Kantarjian H, et al. Preleukaemic clonal

haemopoiesis and risk of therapy-related myeloid neoplasms: a case-control study. Lancet Oncol. 2017;18(1):100-111. 13. Gibson CJ, Lindsley RC, Tchekmedyian V, et al. Clonal hematopoiesis associated with adverse outcomes after autologous stem-cell transplantation for lymphoma. J Clin Oncol. 2017;35(14):1598-1605. 14. Husby S, Favero F, Nielsen C, et al. Clinical impact of clonal hematopoiesis in patients with lymphoma undergoing ASCT: a national population-based cohort study. Leukemia. 2020;34(12):3256-3268. 15. Wong TN, Miller CA, Jotte MRM, et al. Cellular stressors contribute to the expansion of hematopoietic clones of varying leukemic potential. Nat Commun. 2018;9(1):455. 16. Miller PG, Sperling AS, Brea EJ, et al. Clonal hematopoiesis in patients receiving chimeric antigen receptor T-cell therapy. Blood Adv. 2021;5(15):2982-2986. 17. Saini NY, Swoboda DM, Greenbaum U, et al. Clonal hematopoiesis is associated with increased risk of severe neurotoxicity in axicabtagene ciloleucel therapy of large B-cell lymphoma. Blood Cancer Discov. 2022;3(5):385-393. 18. Singh A, Mencia-Trinchant N, Griffiths EA, et al. Mutant PPM1Dand TP53-driven hematopoiesis populates the hematopoietic compartment in response to peptide receptor radionuclide therapy. JCO Precis Oncol. 2022;6:e2100309. 19. Puvvada SD, Guillén-Rodríguez JM, Yan J, et al. Yttrium-90ibritumomab tiuxetan (Zevalin®) radioimmunotherapy after cytoreduction with ESHAP chemotherapy in patients with relapsed follicular non-Hodgkin lymphoma: final results of a phase II study. Oncology. 2018;94(5):274-280. 20. Alcindor T, Witzig TE. Radioimmunotherapy with yttrium-90 ibritumomab tiuxetan for patients with relapsed CD20+ B-cell non-Hodgkin’s lymphoma. Curr Treat Options Oncol. 2002;3(4):275-282. 21. Strosberg J, El-Haddad G, Wolin E, et al. Phase 3 trial of 177 Lu-dotatate for midgut neuroendocrine tumors. N Engl J Med. 2017;376(2):125-135. 22. Sartor O, de Bono J, Chi KN, et al. Lutetium-177–PSMA-617 for

Haematologica | 109 February 2024

519

Z. Xie et al.

ARTICLE - CHIP in patients with follicular lymphoma metastatic castration-resistant prostate cancer. N Engl J Med. 2021;385(12):1091-1103. 23. Jacobs SA, Swerdlow SH, Kant J, et al. Phase II trial of shortcourse CHOP-R followed by 90Y-ibritumomab tiuxetan and extended rituximab in previously untreated follicular lymphoma. Clin Cancer Res. 2008;14(21):7088-7094. 24. Scholz CW, Pinto A, Linkesch W, et al. 90Yttrium-ibritumomabtiuxetan as first-line treatment for follicular lymphoma: 30 months of follow-up data from an international multicenter phase II clinical trial. J Clin Oncol. 2013;31(3):308-313. 25. Morschhauser F, Radford J, van Hoof A, et al. 90Yttriumibritumomab tiuxetan consolidation of first remission in advanced-stage follicular non-Hodgkin lymphoma: updated results after a median follow-up of 7.3 years from the international, randomized, phase III first-line indolent trial. J Clin Oncol. 2013;31(16):1977-1983. 26. Illidge TM, Mayes S, Pettengell R, et al. Fractionated 90Y-ibritumomab tiuxetan radioimmunotherapy as an initial therapy of follicular lymphoma: an international phase II study in patients requiring treatment according to GELF/BNLI criteria. J Clin Oncol. 2014;32(3):212-218. 27. Kaminski MS, Tuck M, Estes J, et al. Tositumomab and iodine I-131 tositumomab for previously untreated, advanced-stage, follicular lymphoma: median 10 year follow-up results. Blood. 2009;114(22):3759-3759. 28. Press OW, Unger JM, Rimsza LM, et al. Phase III randomized intergroup trial of CHOP plus rituximab compared with CHOP chemotherapy plus 131iodine-tositumomab for previously untreated follicular non-Hodgkin lymphoma: SWOG S0016. J Clin Oncol. 2013;31(3):314-320. 29. Bennett JM, Kaminski MS, Leonard JP, et al. Assessment of treatment-related myelodysplastic syndromes and acute myeloid leukemia in patients with non-Hodgkin lymphoma treated with tositumomab and iodine I131 tositumomab. Blood. 2005;105(12):4576-4582. 30. Pritzl SL, Gurney M, Badar T, et al. Clinical and molecular spectrum and prognostic outcomes of U2AF1 mutant clonal hematopoiesis - a prospective Mayo clinic cohort study. Leuk Res. 2023;125:107007. 31. Kumar E, Okosun J. Follicular lymphoma genomics. Hemasphere. 2019;3(Suppl):79-81. 32. Niroula A, Sekar A, Murakami MA, et al. Distinction of lymphoid and myeloid clonal hematopoiesis. Nat Med. 2021;27(11):1921-1927. 33. Zhou X, Edmonson MN, Wilkinson MR, et al. Exploring genomic alteration in pediatric cancer using ProteinPaint. Nat Genet. 2016;48(1):4-6. 34. Maurer MJ, Bachy E, Ghesquières H, et al. Early event status informs subsequent outcome in newly diagnosed follicular lymphoma. Am J Hematol. 2016;91(11):1096-1101. 35. Grillo-López AJ. Zevalin: the first radioimmunotherapy approved for the treatment of lymphoma. Expert Rev Anticancer Ther. 2002;2(5):485-493. 36. Radivoyevitch T, Sachs RK, Gale RP, et al. Defining AML and MDS second cancer risk dynamics after diagnoses of first cancers

treated or not with radiation. Leukemia. 2015;30(2):285-294. 37. Czuczman MS, Emmanouilides C, Darif M, et al. Treatmentrelated myelodysplastic syndrome and acute myelogenous leukemia in patients treated with ibritumomab tiuxetan radioimmunotherapy. J Clin Oncol. 2007;25(27):4285-4292. 38. Reiss J, Link B, Ruan J, et al. Long-term follow up of rates of secondary malignancy and late relapse of two trials using radioimmunotherapy consolidation following induction chemotherapy for previously untreated indolent lymphoma. Leuk Lymphoma. 2015;56(10):2870-2875. 39. McNerney ME, Godley LA, le Beau MM. Therapy-related myeloid neoplasms: when genetics and environment collide. Nat Rev Cancer. 2017;17(9):513. 40. Teipel R, Kroschinsky F, Kramer M, et al. Prevalence and variation of CHIP in patients with aggressive lymphomas undergoing CD19-directed CAR T-cell treatment. Blood Adv. 2022;6(6):1941-1946. 41. Gibson CJ, Lindsley RC, Tchekmedyian V, et al. Clonal hematopoiesis associated with adverse outcomes after autologous stem-cell transplantation for lymphoma. J Clin Oncol. 2017;35(14):1598-1605. 42. Teipel R, Kroschinsky F, Kramer M, et al. Prevalence and variation of CHIP in patients with aggressive lymphomas undergoing CD19-directed CAR T-cell treatment. Blood Adv. 2022;6(6):1941-1946. 43. Mouhieddine TH, Sperling AS, Redd R, et al. Clonal hematopoiesis is associated with adverse outcomes in multiple myeloma patients undergoing transplant. Nat Commun. 2020;11(1):2996. 44. Ortmann CA, Dorsheimer L, Abou-El-Ardat K, et al. Functional dominance of CHIP-mutated hematopoietic stem cells in patients undergoing autologous transplantation. Cell Rep. 2019;27(7):2022-2028. 45. Döhner H, Wei A, Niederwieser D, et al. Diagnosis and management of AML in adults: 2022 ELN recommendations from an international expert panel. Blood. 2022;140(12):1345-1377. 46. Chantadisai M, Kulkarni HR, Baum RP. Therapy-related myeloid neoplasm after peptide receptor radionuclide therapy (PRRT) in 1631 patients from our 20 years of experiences: prognostic parameters and overall survival. Eur J Nucl Med Mol Imaging. 2021;48(5):1390-1398. 47. Kesavan M, Claringbold PG, Turner JH. Hematological toxicity of combined 177Lu-octreotate radiopeptide chemotherapy of gastroenteropancreatic neuroendocrine tumors in long-term follow-up. Neuroendocrinology. 2014;99(2):108-117. 48. FDA grants accelerated approval to axicabtagene ciloleucel for relapsed or refractory follicular lymphoma. https//:www.fda.gov Accessed December 17, 2022. 49. Morice PM, Leary A, Dolladille C, et al. Myelodysplastic syndrome and acute myeloid leukaemia in patients treated with PARP inhibitors: a safety meta-analysis of randomised controlled trials and a retrospective study of the WHO pharmacovigilance database. Lancet Haematol. 2021;8(2):e122-e134.

Haematologica | 109 February 2024

520

ARTICLE - Myeloproliferative Disorders

Phenotypic profiling of CD34+ cells by advanced flow cytometry improves diagnosis of juvenile myelomonocytic leukemia Cristina Bugarin,1 Laura Antolini,2 Chiara Buracchi,1,2 Sergio Matarraz,3 Tiziana Angela Coliva,4 Vincent H. van der Velden,5 Tomasz Szczepanski,6 Elaine Sobral da Costa,7 Alita van der Sluijs,8 Michaela Novakova,9 Ester Mejstrikova,9 Stefan Nierkens,10 Fabiana Vieira de Mello,7 Paula Fernandez,11 Carmen Aanei,12 Łukasz Sędek,6 Luisa Strocchio,13 Riccardo Masetti,14 Laura Sainati,15 Jan Philippé,16 Maria Grazia Valsecchi,2 Franco Locatelli,13 Jacques J.M. van Dongen,3,8 Andrea Biondi,1,17 Alberto Orfao3# and Giuseppe Gaipa1# on behalf of the EuroFlow Consortium

Correspondence: A. Biondi abiondi.unimib@gmail.com M.G. Valsecchi grazia.valsecchi@unimib.it Received: Accepted: Early view:

February 6, 2023. July 26, 2023. August 3, 2023.

Centro Tettamanti, Fondazione IRCCS San Gerardo dei Tintori, Monza (MB), Italy; 2Center of

https://doi.org/10.3324/haematol.2023.282805

Biostatistics for Clinical Epidemiology, Dipartimento di Medicina e Chirurgia, Università degli

Studi Milano-Bicocca, Monza (MB), Italy; 3Cancer Research Center (IBMCC-CSIC), Department of Medicine and Cytometry Service (NUCLEUS), University of Salamanca, CIBERONC and Institute of Biomedical Research of Salamanca (IBSAL), Salamanca, Spain; Department of Pediatrics, Fondazione IRCCS San Gerardo dei Tintori, Monza (MB), Italy;

Department of Immunology, Erasmus MC, University Medical Center Rotterdam, Rotterdam,

the Netherlands; 6Department of Pediatric Hematology and Oncology, Medical University of Silesia (SUM), Zabrze, Poland; 7Department of Pediatrics, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil; 8Department of Immunohematology and Blood Transfusion (IHB), Leiden University Medical Center (LUMC), Leiden, the Netherlands; 9CLIP-Department of Pediatric Hematology and Oncology, Second Faculty of Medicine, Charles University and University Hospital Motol, Prague, Czech Republic; 10Princess Máxima Center for Pediatric Oncology, Utrecht, the Netherlands; 11Institute for Laboratory Medicine, Kantonsspital Aarau AG, Aarau, Switzerland; 12Hematology Laboratory CHU de Saint-Etienne, Saint-Etienne, France; 13Department of Pediatric Hematology and Oncology IRCCS Ospedale Pediatrico Bambino Gesu’, Sapienza University of Rome, Italy; 14Pediatric Oncology and Hematology Unit ‘Lalla Seràgnoli’, IRCCS Azienda Ospedaliero-Universitaria di Bologna, Bologna, Italy; Dipartimento di Salute della Donna e del Bambino, Clinica di Oncoematologia Pediatrica,

Azienda Ospedale Università di Padova, Padua, Italy; 16Department of Laboratory Medicine, Ghent University Hospital, Ghent, Belgium and 17Dipartimento di Medicina e Chirurgia, Università degli Studi Milano-Bicocca, Monza (MB), Italy AO and GG contributed equally as senior authors.

Abstract Diagnostic criteria for juvenile myelomonocytic leukemia (JMML) are currently well defined, however in some patients diagnosis still remains a challenge. Flow cytometry is a well established tool for diagnosis and follow-up of hematological malignancies, nevertheless it is not routinely used for JMML diagnosis. Herewith, we characterized the CD34+ hematopoietic precursor cells collected from 31 children with JMML using a combination of standardized EuroFlow antibody panels to assess the ability to discriminate JMML cells from normal/reactive bone marrow cell as controls (n=29) or from cells of children with other hematological diseases mimicking JMML (n=9). CD34+ precursors in JMML showed markedly reduced B-cell and erythroid-committed precursors compared to controls, whereas monocytic and CD7+ lymphoid precursors were significantly expanded. Moreover, aberrant immunophenotypes were consistently present in CD34+ precursors in JMML, while they were virtually absent in controls. Multivariate logistic regression analysis showed that combined assessment of the number of CD34+CD7+ lymphoid precursors and CD34+ aberrant precursors or erythroid precursors had a great potential in discriminating JMMLs versus controls. Importantly our scoring model allowed highly efficient discrimination of truly JMML Haematologica | 109 February 2024

521

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

versus patients with JMML-like diseases. In conclusion, we show for the first time that CD34+ precursors from JMML patients display a unique immunophenotypic profile which might contribute to a fast and accurate diagnosis of JMML worldwide by applying an easy to standardize single eight-color antibody combination.

Introduction Juvenile myelomonocytic leukemia (JMML) is a rare, unique myeloproliferative/myelodysplastic neoplasia of early childhood driven by canonical Ras-pathway mutations in PTPN11, N-RAS, K-RAS, NF1, or CBL and characterized by in vitro hypersensitivity of hematopoietic progenitor to granulocyte-macrophage colony-stimulating factor (GM-CSF).1,2 JMML has an incidence of 1.2 cases per million children per year and, if untreated, patients can progress toward an uncontrolled disease.3 Recently the World Health Organization (WHO) classification placed JMML in the group of myelodysplastic/myeloproliferative disorders.4,5 Clinical signs of JMML are well defined.6 Peripheral blood (PB) smears typically shows leukocytosis with monocytosis and low blast cell counts (median 2% myeloblasts),7 and presence of immature hematopoietic precursor cells.8 Bone marrow (BM) examination reveals hypercellularity with myelomonocytic cell proliferation, reduction of megakaryocytes and moderate increase of blasts (<20% myeloblasts). Thus, the combination of young age, hepato-splenomegaly, appearance of myeloid and erythroid precursors in the PB, and/or elevated levels of fetal hemoglobin should alert the clinicians to suspect JMML and initiate specific tests. These generally include the molecular analysis of driver mutations in the PTPN11, K-RAS, N-RAS, and CBL genes, and the search for features of neurofibromatosis type 1 (NF1) including family history,9 in addition to the contemporary exclusion of the BCR-ABL transcript. Monosomy 7 is the most frequent cytogenetic aberration found in ~25% of JMML patients.8,10,11 A conventional hallmark of clonogenic JMML cells is their in vitro hypersensitivity to GM-CSF, although this test is laborious and poorly standardized. Flow cytometry-based assay of STAT5 hyperphosphorylation after stimulation with GM-CSF was proposed to aid in distinguishing JMML from other disease conditions.12,13 In addition to the above diagnostic parameters the DNA methylation analysis has been also validated as prognostic biomarker potentially aiming at tailoring treatment strategies for JMML.14-16 Despite progress made in the diagnosis of JMML, it can still be challenging due to the existence of several diseases presenting with clinical features mimicking JMML, such as human herpesvirus infections, leukocyte-adhesion molecule deficiency, infantile malignant osteopetrosis, hemophagocytic lymphohistiocytosis, and Wiskott-Aldrich syndrome.17-19 Diagnostic criteria for JMML have been recently revised and summarized in several comprehensive reviews.20-23 However, in contrast to other hematological malignancies, immunophenotyping is not part of the diagnostic work-up. Despite this, Oliveira et al. have recently shown a significant decrease of T lympho-

cytes and the presence of several phenotypic abnormalities within CD34+ cells of JMML patients including the aberrant expression of CD7 in the majority of CD34+CD117+CD13+ cells, associated with a decrease or complete lack of CD19+CD10+ B-cell precursors,24 similarly to what has been described in adults with myelodysplastic syndrome (MDS) and/or myeloproliferative neoplasm (MPN).25,26 Since 2012, the EuroFlow consortium has developed fully standardized approaches for immunophenotyping of hematological malignancies, including validated antibody panels, standardized sample preparation protocols, gating strategies and innovative (smart) data analysis and software tools embedded with artificial intelligence, to prospectively classify individual patients against a predefined reference database of normal/reactive and disease associated groups of subjects.27 However, these approaches have never been applied for the diagnosis of JMML. Herewith we applied the EuroFlow eight-color acute leukemia orientation tube (ALOT), and AML/MDS/MPN antibody panel for the phenotypic characterization of major BM myeloid lineages (neutrophil, monocytic, and erythroid cells), and other minor BM cell subsets.28 In this study we used such antibody panels for in depth characterization of the immunophenotypic profile of CD34+ hematopoietic precursor cells (HPC) of children with confirmed JMML and comparison with either normal/non-malignant cells or cells from children with JMML-like diseases. Our ultimate goal was to identify an immunophenotypic profile that could help in fast and objective differential diagnosis of JMML versus patients with normal/reactive BM or patients with JMML-like diseases.

Methods Pediatric samples Samples for flow cytometric immunophenotyping were collected in eight EuroFlow centers (Brazil, Czech Republic, France, Italy, the Netherlands, Poland, Spain and Switzerland) from 31 children newly diagnosed as JMML according to the WHO 2016 criteria.4 From each child either BM (n=22) or PB (n=9) was collected. In eight of 31 JMML patients paired BM/ PB samples were available, obtaining a total of 39 samples (22 BM and 17 PB) for the analyses. The immunophenotype of JMML BM/PB cells was compared with that of BM cells from 29 children without hematological malignancies (referred in the following as control group) or with that of BM/PB cells from nine children presenting with a suspected diagnosis of JMML which was not subsequently

Haematologica | 109 February 2024

522

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

confirmed (referred in the following as non-confirmed JMML patients). For validation purposes, an additional cohort consisting of four control subjects and three JMML patients was collected. Informed consent was obtained from each patient and/or his/her legal guardian. Flow cytometry immunophenotypic studies Flow cytometry immunophenotyping was performed as described in Online Supplementary Appendix with both the eight-color EuroFlow ALOT and the AML/MDS/MPN antibody panels (Online Supplementary Table S1), following the EuroFlow standard operating procedures for sample preparation, instrument set-up and calibration.27,29 The gating strategies used to identify each cell population are described in Online Supplementary Figure S1A-F. Statistical analysis Descriptive statistics for each CD34+ HPC subset was first performed separately for JMML cases, controls and non-confirmed JMML using median, interquartile range (IQR) (p25, p75) displayed in box plot graphics. For the comparison between groups of subjects, the non-parametric Wilcoxon rank sum test was used and the degree of discrimination was investigated using receiver operating characteristic (ROC) curves. The cross-validated area under the curve (AUC) was used to rank the discriminatory potential of each phenotypic parameter. A logistic regression model, with a binary response variable indicating the diagnosis of JMML (i.e., presence vs. absence), was then estimated including combinations of phenotypic parameters among those that showed the greatest discrimination potential in the ROC curve analyses. The ROC analysis through the AUC enabled us to assess the diagnostic potential in separating JMML cases from controls by the probability that, for a random pair of JMML cases and controls, the model risk score was greater in the JMML case than in the control case. The number of phenotypic parameters included in the model was guided from considerations of significance on couples, triplets etc. The number of parameters in the final models was the maximum with significant contributions that will be lost by adding further parameters. The choice of the final model was then reassessed by its potential in terms of ROC analysis. This model was used to obtain a risk score for confirming the diagnosis of JMML, where the greater the score was obtained the higher the predicted probability of JMML was observed.

Results Patients and controls features We investigated a total of 31 patients with a median age at diagnosis of 1 year (IQR, 1 month to 4 years) fulfilling clinical and hematological characteristics of JMML, as summarized in Table 1. Molecular screening performed in 30 of 31 patients (in 1 case data was not available) showed mutations in the

RAS signaling pathway and/or clinical findings consistent with NF1 with a distribution in line with previous studies.30 Specifically, 13 patients (45%) carried somatic PTPN11 mutations, eight (26%) showed RAS somatic mutations (4 N-RAS and 4 K-RAS). In three cases (10%) a clinical diagnosis of NF1 was made, and six patients (21%) carried CBL germline mutations. In all studied patients the absence of Philadelphia chromosome (BCR/ABL rearrangement) was assessed. The control group consisted of 29 pediatric BM samples including two healthy hematopoietic stem cell donors and 27 BM samples obtained from children non-suspected for JMML undergoing diagnostic BM aspiration for suspicious of idiopathic thrombocytopenic purpura, cytopenia after viral infection, arthralgias and transient neutropenia. Morphological examination of the BM aspirates confirmed the absence of BM involvement at the time of immunologic investigation in all cases. The median age of the children was 4 years (IQR, 1 month to 12 years). Nine patients with clinical features mimicking JMML (median age 6 months; IQR, 0-2 years) for whom diagnosis of JMML was finally ruled out (non-confirmed JMML) were also analyzed (Table 2). Finally, we used additional samples as a validation cohort consisting of four control subjects (3 BM samples from healthy hematopoietic stem cell donors, with age of 3, 6 and 8 years, and 1 BM from a 6-year-old child without hematological disease, who underwent clinical observation for arthralgias) and three JMML patients (1 BM sample and 2 paired BM/PB samples) whose characteristics are reported in Online Supplementary Table S2. Analysis of CD34+ hematopoietic precursor cells Overall, the median frequency of CD34+ HPC, referred to total nucleated cells, was significantly higher in BM of JMML patients (n=22) than in control BM samples (n=29) being 3.0% (IQR, 2.4-4.9%) versus 1.8% (IQR, 1.0-2.4%) (P=0.0038; Figure 1). The distribution of CD34+ HPC in PB and BM of JMML patients was 2.0% (IQR, 0.7-3.0%) versus 3.0% (IQR, 2.4-4.9%) (P=0.1155), respectively (Online Supplementary Figure S2). For this reason, we decided to pool all the JMML samples regardless of the collection source (BM or PB) obtaining a series of 39 samples (22 BM + 17 PB) that have been compared to CD34+ HPC from control BM samples. Both B-cell and erythroid precursors subsets of CD34+ HPC, were strongly reduced in JMML versus control: 2.4% (IQR, 0.6-5.7%) versus 60.1% (IQR, 41.8-66.7%), (P<0.0001) and 0.2% (IQR, 0.1-0.6%) versus 1.9% (IQR, 1.7-3.6%) (P<0.0001). By contrast, the median percentage of monocytic precursors and CD7+ precursors were both significantly expanded within the CD34+ HPC compartment in JMML versus control group: 10.4% (IQR, 3.4-18.0%) versus 3.2% (IQR, 1.2-6.1%) (P=0.0004) and 28.7% (IQR, 8.3-76.0%) versus 2.9% (IQR, 1.5-4.4%) (P<0.0001), respectively. In turn, the proportion of cells showing early commitment to the neutrophil lineage was similar in JMML versus control samples: 32.5% (IQR, 13.0-43.1%) versus 23.9% (IQR, 20.3-32.3%) (P=0.9). Interestingly, immunophenotyp-

Haematologica | 109 February 2024

523

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML Table 1. Clinical and laboratory findings of 31 juvenile myelomonocytic leukemia patients at diagnosis. Patient code JMML 1 JMML 2 JMML 3 JMML 4 JMML 5 JMML 6 JMML 7 JMML 8 JMML 9 JMML 10 JMML 11 JMML 12 JMML 13 JMML 14 JMML 15 JMML 16 JMML 17 JMML 18 JMML 19 JMML 20 JMML 21 JMML 22 JMML 23 JMML 24 JMML 25 JMML 26 JMML 27 JMML 28 JMML 29 JMML 30 JMML 31

Age in years

Sex

Splenomegaly

WBC x109/L

Monocytes x109/L

% Blasts (by morphology)

Genetic subgroup§

Karyotype

2 3 <1 2 1 <1 1 3 <1 1 1 1 3 1 <1 <1 <1 <1 2 1 3 1 <1 <1 <1 2 3 <1 4 <1 2

F F F M F F M F M M M F M M M M M M M F M F M M F F F F M M F

P P P P P P A P P NK P NK P P P P NK P P P P P P P NK P P P P P P

50.0 52.4 26.4 47.7 23.6 8.2 12.2 20.5 12.7 22.8 20.0 65.0 52.0 24.7 19.5 149.0 28.0 65.0 46.2 13.8 6.7 50.0 14.0 10.1 28.4 10.8 23.5 56.8 24.0 26.9 34.2

5.5 3.7 5.2 7.2 14.5 1.5 2.6 1.2 1.7 1.6 3.0 >1.0 8.5 2.4 3.8 26.8 7.8 13.7 8.8 4.7 2.0 20.0 4.9 1.3 7.1 1.2 9.8 21.6 5.5 8.6 1.9

5.0 1.0 2.7 4.0 5.0 4.0 0.0 15.0 0.0 NK 0.0 NK NK 10.0 1.0 3.0 3.0 8.0 2.0 5.0 5.5 4.8 2.2 8.0 1.4 10.0 4.8 NK. 0.0 3.2 0.0

PTPN11 K-RAS PTPN11 PTPN11 PTPN11 CBL K-RAS PTPN11 CBL PTPN11 NF1 CBL NF1 N-RAS PTPN11 N-RAS PTPN11 PTPN11 CBL K-RAS PTPN11 K-RAS NK PTPN11 CBL PTPN11 PTPN11 N-RAS CBL N-RAS NF1

NK 46,XX 47,XX +21 45,XY,-7 46,XX NK 45,XY,-7 46,XX 46,XY NK NK NK NK 46,XY 46, XY 46, XY 46,XY 46,XY NK 45,XX,-7 45,XY,-7 46,XX 45,XY,-7 46, XY 46,XX 47,XX,+8 45,XX,-7 NK NK 46, XY 46, XX

PTPN11 or K-RAS or N-RAS or RAS are intended as somatic mutations (germline status was excluded based on buccal swab testing), CBL is intended as germline mutation ± loss of heterozygosity (LOH), NF1 is intended as clinical diagnosis of neurofibromatosis type 1. JMML: juvenile myelomonocytic leukemia; WBC: white blood cells; F: female; M: male; P: present; A: absent; NK: not known. §

ically aberrant CD34+ HPC subsets such as CD7+/cyMPO+/ cyCD3- and cyCD79a+/CD7+/cyCD3- were detected in most of JMML samples (3.9%; IQR, 1.3-8.4%), while they were virtually absent in the control group (0.05%; IQR, 0.0-0.3%) (P<0.0001) (Figure 2A-F). In order to rule out any age-specific effects on the comparison between JMML and non-malignant control subjects we analyzed this latter series by dividing the patients into those with an age of less than 4 years (within the age range of JMML patients) and those with an age greater than or equal to 4 years. As shown in Online Supplementary Figure S3 no significant differences in the phenotypic profile of CD34+ HPC were observed. In addition, to verify the homogeneity of our control series, we compared the phenotypic profile of the healthy hematopoietic stem cell donors (n=5) with that of the 27 children

with non-malignant hematopoietic abnormalities, and we did not find any remarkable differences (Online Supplementary Figure S4). Besides of CD34+ HPC we also dissected seven major compartments of more mature (i.e., CD34-) hematopoietic cells in BM, including: B and T lymphocytes, natural killer cells, monocytes, neutrophils, eosinophils, and erythroid cells. Statistical differences are reported in Online Supplementary Figure S5. Assessment of the discriminatory potential of the phenotypic profile and distribution of hematopoietic cells in juvenile myelomonocytic leukemia patients versus controls Once identified those immunophenotypic parameters with significantly different expression in JMML versus controls, we studied their potential in discriminating between JMML and

Haematologica | 109 February 2024

524

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML Table 2. Clinical and laboratory findings of nine non-confirmed juvenile myelomonocytic leukemia patients. Patient code

Age in years

Sex

Splenomegaly

WBC x109/L

Monocytes x109/L

Final diagnosis

Non-confirmed JMML 1

30.1

1.5

LAD II

Non-confirmed JMML 2

20.4

1.5

CMV infection

Non-confirmed JMML 3

7.3

0.8

EBV infection

Non-confirmed JMML 4

26.8

2.6

Transient myeloproliferative reaction with karyotype: 46, XY

Non-confirmed JMML 5

30.5

2.3

Chronic eosinophilic leukemia [inv(9), t(8;9), PMC1-JAK2 fusion gene]

Non-confirmed JMML 6

Non-confirmed JMML 7

30.0

NK (monocytosis)

Aphthous fever, JMML and Noonan syndrome were excluded*

Non-confirmed JMML 8

7.9

0.98

AML NOS with megakaryoblastic maturation

Non-confirmed JMML 9

20.6

1.1

CMV congenital infection and Noonan syndrome (pathogenic mutation c.846 C-G in PTPN11 in heterozygotic status)

NK NK (leukocytosis) (monocytosis)

AML M7

*This patient underwent observation because of Noonan syndrome facies not subsequently confirmed. JMML: juvenile myelomonocytic leukemia; WBC: white blood cells; M: male; F: female; NK: not known; A: absent; P: present; CMV: Cytomegalovirus; EBV: Epstein-Barr virus; LAD II: leukocyte adhesion deficiency type II; AML: acute myeloid leukemia; NOS: not otherwise specified.

control subjects. To this aim we used a ROC curve analysis approach based on the relative distribution of immature CD34+ HPC compartments. Then, the cross-validated AUC was used to rank the discriminatory potential of each single CD34+ phenotypic parameter along the range of possible values. Based on the rank of accuracy identified for each cell population the most discriminatory phenotypic parameters, measured as percentage of cells on total CD34+ HPC, were: CD7+ precursors (AUC=0.944), aberrant precursors (AUC =0.943), erythroid precursors (AUC=0.936), and B-lymphoid precursors (AUC=0.930) (Online Supplementary Figure S6). Logistic regression model for the immunophenotypic diagnosis of juvenile myelomonocytic leukemia In order to better discriminate JMML cases from controls we developed a logistic regression model using the diagnosis of JMML (yes or no) as a dependent variable, and those immunophenotypic parameters detected within the CD34+ HPC having (individually) the highest discriminatory potential (i.e., AUC), as regressors. To this aim, we combined the highest discriminatory parameter (i.e., the proportion of CD7+ precursors within CD34+ HPC: AUC=0.944) with either the percentage of CD34+ aberrant precursors (AUC=0.943), or the number of CD34+ erythroid precursors (AUC= 0.936), to obtain two different combinations of informative immunophenotypic parameters. This model was

then used to obtain a risk score for being diagnosed as JMML according to the following algorithms: score model 1= (0.619 * %CD7 precursors) + (1.444 * %aberrant precursors); and score model 2 = (0.497 * %CD7 precursors) - (1.573 * %erythroid precursors). The obtained scores (Online Supplementary Table S3) are a scale transformation of the predicted probability of being a JMML, thus the greater the score, the greater is

Figure 1. Percentage of CD34+ hematopoietic precursor cells in juvenile myelomonocytic leukemia and control samples. Box plot graphics show a significantly increased percentage of CD34+ hematopoietic precursor cells in juvenile myelomonocytic leukemia (JMML) bone marrow (BM) as compared to control (CTR) BM samples (P=0.0038).

Haematologica | 109 February 2024

525

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

the predicted probability. This probability will depend on the prevalence of JMML cases in the target population (Online Supplementary Table S4). The discriminatory potential (score value) of each score model was then assessed by the AUC of the linear predictor, both score models displaying high discriminatory potential between JMML and control subjects: AUC=0.973, P<0.0001, and AUC=0.982, P<0.0001, respectively (Figure 3A, B). By applying each risk score to the additional validation series of controls and JMML samples (Online Supplementary Table S5) we confirmed highly significant differences in the discriminatory potential with both models (P=0.0079 and P=0.0008; Figure 3C). Distribution of CD34+ hematopoietic precursor cells subsets in juvenile myelomonocytic leukemia versus nonconfirmed juvenile myelomonocytic leukemia patients The median percentage of CD34+ HPC in pooled BM/PB samples of non-confirmed JMML samples was 1.5% (IQR, 0.8-2.6%) with non-significant difference with that of JMML (P=0.121) even when comparing BM and PB separately (Online Supplementary Figure S7). In order to confirm the diagnostic value of the developed immunophenotypic scores, we subsequently compared the distribution of different subsets of CD34+ HPC

in JMML patients with those of non-confirmed JMML patients. Overall, JMML samples showed a significantly higher proportion (median and range) of both CD7+ precursors with 28.7% (IQR, 8.3-76.0%) versus 3.7% (IQR, 2.2-4.8%) (P=0.0006) and aberrant precursors with 3.9% (IQR, 1.3-8.4%) versus 0.2% (IQR, 0.0-1.0%) (P=0.0047); by contrast B-cell precursors were decreased with 2.4% (IQR, 0.5-6.2%) versus 44.6% (IQR, 12.668.5%) (P<0.0001) (Figure 4A-F). We then applied both score model 1 and 2 to non-confirmed JMML samples to assess their risk score (Online Supplementary Table S6), and we obtained a highly efficient discrimination between JMML and non-confirmed JMML as shown in Figure 5A, B. The calculated discriminatory values (AUC of JMML vs. non-confirmed JMML) were 0.954 and 0.903 for score model 1 and 2, respectively (data not shown). Prevalence of the juvenile myelomonocytic leukemia L-associated immunophenotypic signature according to the underlying genetic mutation We, therefore, wanted to evaluate the JMML-associated immunophenotypic profile more in-depth by analyzing each CD34+ phenotypic parameter within each JMML genetic subgroup in comparison with control subjects. As reported in Figure

Figure 2. Distribution of different immunophenotypic subsets in CD34+ hematopoietic precursor cells in juvenile myelomonocytic leukemia and control samples. Box plot graphics show severely reduced B-cell and erythroid precursors (P<0.0001) (A, F) and significantly increased CD7+ lymphoid, CD34+ aberrant precursors, (both P<0.0001) (B, C), as well as increased monocytic precursors (P=0.0004) in juvenile myelomonocytic leukemia (JMML) bone marrow/peripheral blood as compared to control (CTR) bone marrow precursor cells (E). Non-significant (NS) differences were found in the distribution of neutrophil precursors (P≥0.05) (D). Percentages of each immunophenotypic subset are referred to 100% of CD34+ cells. Haematologica | 109 February 2024

526

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

Figure 3. Discriminatory potential of the CD34+ immunophenotypic profile in juvenile myelomonocytic leukemia patients versus controls. The risk score for being diagnosed as juvenile myelomonocytic leukemia (JMML) is shown in the left graphs (A, B) for score model 1 (A) and score model 2 (B), respectively. The distribution of risk scores in JMML and control samples are compared in each graph showing highly significant differences. Area under the ROC curve (AUC) to assess the discriminatory potential of each score model is reported in the right graphs of both panels. (C) Represents the additional validation series of controls and JMML samples that confirms significant discriminatory potential with both models (P=0.0079 and P=0.0008).

6, CD34+ HPC such as B-cell precursors, CD7+, and aberrant precursors, maintained a significantly different expression as compared to controls even when analyzed within the PTPN11, NF1 and RAS genetic subgroups. In contrast such parameters in patients carrying germline CBL gene mutations were non significantly different than controls. Despite this, JMML patients with the CBL mutation could still be discriminated from controls due to a significant increase in the percentage of monocytic precursors, equal to 16.9% (IQR, 8.6-26.1%) compared to 3.2% (IQR, 1.2-6.1%) (P<0.001), and a decrease in the percentage of CD34+ erythroid precursor cells being 0.5% (IQR, 0.1-1.1%) versus 1.9% (IQR, 1.7-3.6%) (P<0.01). When we applied the score model 1 or the score model 2 specifically to CBL-mutated JMML patients we observed a non-significant discriminatory power versus control with either model (Online Supplementary Figure S8A, B). Finally, we displayed the distribution of each single sample

along the score values according to its genetic alteration (Online Supplementary Figure S8C, D). Indeed, by applying the score model 1, among the six samples resulted with lowest score values (below the discriminatory value of 6.178), three were CBL, two were RAS and one was PTPN11. Whereas by applying the score model 2, we found one CBL and one RAS below the value of 1.455.

Discussion Although JMML is rare, the criteria proposed for diagnosing this malignant disease have been progressively refined over time.11 Nevertheless, diagnosis of JMML can still be challenging especially when specific laboratory tests (e.g., molecular assays and next-generation sequencing-based genetic diagnostics) are not available. Moreover, a con-

Haematologica | 109 February 2024

527

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

Figure 4. Distribution of different immunophenotypic subsets in CD34+ hematopoietic precursor cells in juvenile myelomonocytic leukemia and non-confirmed juvenile myelomonocytic leukemia. Box plot graphics show significantly differences in both bone marrow/peripheral blood samples for immunophenotypic parameters such as CD34+ B-cell precursors (P<0.0001), CD7+ precursors (P=0.0006) and aberrant precursors (P=0.0047) (A-C). Non-significant (NS) differences were found in the distribution of neutrophil, monocytic and erythroid precursors (P≥0.05) (D-F).

sistent plethora of other confounding malignancies, either hematological or non-hematological, mimicking JMML, can further complicate the diagnosis of this disease and delay clinical decision-making.6,11,23,30 Taking advantage of a fully standardized flow cytometric platform27 herewith we investigated in depth the immunophenotypic profile of immature CD34+ HPC of JMML patients compared to both control children and patients with other diseases mimicking JMML. Our ultimate goal was to identify immunophenotypic alterations that could be used in a score system to discriminate between JMML and non-confirmed JMML cases. Interestingly, an in-depth analysis of JMML CD34+ HPC revealed a markedly reduced percentages of both B-cell precursors and erythroid precursors, in parallel to a marked expansion of monocytic precursors and CD7+ lymphoid precursors. In addition, JMML CD34+ HPC were characterized by the systematic presence of aberrant immunophenotypes, being virtually absent in the BM of control subjects. Based on these findings we used ROC curve analysis to estimate the discriminatory potential of each individual immunophenotypic parameter. By this approach we then designed two different highly predictive score models for diagnosing JMML (score model 1 and score model 2). By applying each

risk score model to an additional, albeit limited, validation series of controls and JMML samples we confirmed a high discriminatory potential with both of them. Indeed, we are aware that our control series is mainly constituted by non-healthy donor children, however we did not find any significant differences between the immunophenotypic profile of the truly healthy donors and that of children who underwent BM puncture for diagnostic purposes. Further application of these score models in a cohort of patients with suspected diagnosis of JMML, non-confirmed after completion of the diagnostic investigations, demonstrated a greater predictive value of score model 1 as compared to score model 2. Indeed, model 1 may be a more robust tool in routine diagnostics since in some patients erythroid precursors (used in score model 2) were present at very low frequencies, which may hamper their unequivocal identification particularly in hypocellular BM specimens. Moreover, score model 1 would require cell staining only with the ALOT single-tube antibody combination whereas score model 2 would require the addition of the AML tube 3 staining. Of note, despite the decreased number of erythroid precursors observed within CD34+ HPC in BM, this did not translate into a decreased (relative) production of more

Haematologica | 109 February 2024

528

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

Figure 5. Discriminatory potential of the CD34+ immunophenotypic profile in juvenile myelomonocytic leukemia patients versus non-confirmed juvenile myelomonocytic leukemia patients. (A) Discriminatory potential between juvenile myelomonocytic leukemia (JMML) and non-confirmed JMML patients for the score model 1 resulted highly significant (P<0.0001). Dashed line at 6.178 represents the maximum score obtained in control subjects. Scores for all non-confirmed JMML patients fell below this cutoff value. (B) Discriminatory potential between JMML and non-confirmed JMML patients for the score model 2. The ability to discriminate JMML from the non-confirmed JMML was feasible. However, 3 of 9 of non-confirmed JMML samples resulted slightly above the maximum score of control subjects setting at 1.455; specifically non-confirmed JMML # 4, #8, #9 reported in Online Supplementary Table S6. Control samples are also reported in each graph. No differences were found between non-confirmed JMML and controls when applying both score models.

Figure 6. Distribution of different immunophenotypic subsets in CD34+ hematopoietic precursor cells in juvenile myelomonocytic leukemia patients according to gene mutations. All gene mutations were associated with a specific immunophenotypic parameter of CD34+ cells (indicated in the y axis). CBL-mutated patients showed immunophenotypic features of CD34+ hematopoietic precursor cells (HPC) more similar to those of control (CTR) cases without remarkable differences in B cell, CD7+, aberrant, and neutrophil precursors (A-D), but with significant differences in monocytic and erythroid cell precursors (P<0.01) (E, F). All other genetic subgroups maintained highly relevant differences compared to CTR in most of the immunophenotypic parameters including those used for score models (i.e., CD7+, aberrant and erythroid precursors). Haematologica | 109 February 2024

529

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

mature CD34- nucleated red blood cells, which showed median values within the normal ranges. Whether, such apparent discrepancy is due to a greater expansion/proliferation potential of the fewer CD34+ erythroid precursors, the relative decrease of maturing neutrophil precursors and/or to the relative overall increased number of CD34+ HPC in JMML versus normal/reactive BM, deserves further investigations. It is important to note that the antibody combinations proposed here are all highly validated and have proven to be easy to standardize, with highly reproducible results in different laboratories.27,29,31 It is known that for most patients with JMML, early allogeneic HSCT is the mandatory therapy of choice,11,32–34 however, this indication may vary according to several risk factors to be assessed early after diagnosis, including the type of mutation and the gene methylation status.6,35,36 Indeed, patients carrying NF1 or PTPN11 gene mutations are at fatal risk without rapid HSCT. In turn, children carrying K-RAS and N-RAS mutations are associated with variable clinical risk, this also considering entities such Ras-associated autoimmune leukoproliferative disorder (RALD),37 whereas patients with germline CBL mutations undergo a watch-and-wait approach.22,23,38 In this context, a goal of our study was to determine whether different gene mutations occurring in JMML were associated with specific immunophenotypic profiles of CD34+ HPC. Interestingly we observed a step-wise decline of score levels seen from highest levels in NF1, then PTPN11 and RAS to lowest in CBL cases, which were almost close to normal, similarly to their well established clinical risk profile. In this regard, the sensitivity of our scoring models may be underpowered mainly by the CBL and RAS cases. These findings suggest that the here proposed immunophenotypic score could also reflect a clinical-biological significance with potential prognostic value. These aspects are currently being investigated and they may also contribute to a better understanding of the genetic heterogeneity of JMML patients.24,26,39 However, given the limited sample size of our series and the possibility that RAS cases with lower scores could be RALD and not JMML, the power to assess phenotypic-genotypic associations should be considered with caution. The novelty of our study relies on the design of a diagnostic algorithm based on immunophenotypic scoring as a new tool to be integrated in the laboratory diagnostic work-up of JMML. It allows rapid identification of children with this disease, even in those patients with confounding clinical signs, just within a few hours from sample collection while molecular tests take longer. Further, this assay can be easily implemented worldwide being eight-color flow cytometers available in virtually every pediatric oncology center. Importantly, it can also be successfully applied in PB samples with great advantage in clinical practice, especially for infant patients.

In summary, the novel flow cytometric assay proposed here can contribute to a faster and accurate diagnosis of JMML allowing a prompt start of both treatment with demethylating agents40 and of the search for locating a suitable HSCT donor. Disclosures JJMvD, AO, TS, and VHvdV each report being one of the inventors on the EuroFlow-owned patent PCT/NL2010/050332 (methods, reagents and kits for flow cytometric immunophenotyping of normal, reactive and malignant leukocytes). The Infinicyt software is based on intellectual property of AO and ESdC, licensed to Cytognos SL (Salamanca, Spain) which pays royalties to the University of Salamanca (Salamanca, Spain), and the scientific input of other EuroFlow members. JJMvD and AO are chairmen of the EuroFlow scientific foundation, which receives royalties from licensed patents (Cytognos - Salamanca, ES and BD Biosciences - San José, CA) that are collectively owned by the participants of the EuroFlow Foundation. These royalties are exclusively used for continuation of EuroFlow collaboration and sustainability of the EuroFlow consortium. VHvdV reports a Laboratory Services Agreement with BD Biosciences. In addition, JJMvD and AO report a Laboratory Services Agreement with BD Biosciences, an Educational Services Agreement with BD Biosciences and a Scientific Advisor Agreement with Cytognos SL and BD Biosciences; all related fees and honoraria of these agreements go to Leiden University Medical Center and University of Salamanca, respectively. AB is on the speakers bureau of Amgen and Novartis. FL has served in an advisory role for Amgen, Bellicum Pharmaceuticals, Neovii, Novarti, Sanofì and Vertex; and has served on a speakers’ bureau for Amgen, Bellicum Pharmaceuticals, Bluebird bio, Gilead, Jazz Pharmaceuticals, Medac, Miltenyi Biotec, Novartis, Neovii, SOBI and Takeda. All other authors have no conflicts of interest to disclose. Contributions CrB performed research, analyzed and interpreted the data, and wrote the paper. LA and MGV performed statistical analyses and wrote the paper. ChB collected samples and analyzed flow cytometric data. AO and GG designed the research, lead the project, and wrote the paper. FL, JJMvD and AB supervised the project. SM, TC, VHvdV, TS, AJvdS, ESdC, MN, EM, SN, FVDM, CA, PF, LS, LuS, RM and LaS collected samples and patient data. All authors reviewed and approved the final version of the manuscript. Funding This project was supported by Fondazione Alessandro Maria Zancan “GrandeAle ONLUS”, Fondazione M. Tettamanti De Marchi. It was also partially funded by Fondazione Regionale per la Ricerca Biomedica (FRRB, Regione Lombardia), project no. CP2_10/2018 (Plagencell); AIRC IG 2017 ref. id 20564 (to AB), AIRC 5x1000 ref. id 21147 (to AB and FL), AIRC Accelerator

Haematologica | 109 February 2024

530

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML

Award 2018 id 22791 (to AB and FL); Ministero Università e Ricerca, PRIN2017, project no. 2017WC8499 (to FL and AB). SM was supported by Acción Estratégica en Salud (AES) (grant no. PI21_01115) and the grant of CIBERONC of the Instituto de Salud Carlos III, Ministerio de Ciencia e Innovación, Madrid, Spain and FONDOS FEDER (no. CB16/12/00400). MN and EM

were supported by the Ministry of Health of the Czech Republic, grant no. NU20J-07-00028. Data-sharing statement The data supporting the findings of this study are available upon request to the corresponding author.

References 1. Emanuel P, Bates L, Castleberry R, Gualtieri R, Zuckerman K. Selective hypersensitivity to granulocyte-macrophage colonystimulating factor by juvenile chronic myeloid leukemia hematopoietic progenitors. Blood. 1991;77(5):925-929. 2. Freedman MH, Cohen A, Grunberger T, et al. Central role of tumour necrosis factor, GM-CSF, and interleukin 1 in the pathogenesis of juvenile chronic myelogenous leukaemia. Br J Haematol. 1992;80(1):40-48. 3. Hasle H, Kerndrup G, Jacobsen BB. Childhood myelodysplastic syndrome in Denmark: Incidence and predisposing conditions. Leukemia. 1995;9(9):1569-1572. 4. Baumann I, Bennett JM, Niemeyer CM. Juvenile myelomonocytic leukemia. In: Organization WH, editor. WHO Classification of tumours of haematopoietic and lymphoid tissues. IARC: Lyon; 2017. p. 89-92. 5. Khoury JD, Solary E, Abla O, et al. The 5th edition of the World Health Organization classification of haematolymphoid tumours: myeloid and histiocytic/dendritic neoplasms. Leukemia. 2022;36(7):1703-1719. 6. Niemeyer CM, Flotho C. Juvenile myelomonocytic leukemia: who’s the driver at the wheel? Blood. 2019;133(10):1060-1070. 7. Castro-malaspina H, Schaison G, Passe S, et al. Subacute and chronic myelomonocytic leukemia in children (juvenile CML). Clinical and hematologic observations, and identification of prognostic factors. Cancer. 1984;54(4):675-686. 8. Niemeyer CM, Aricó M, Basso G, et al. Chronic myelomonocytic leukemia in childhood: a retrospective analysis of 110 cases. Blood. 1997;89(10):3534-3543. 9. Greenmyer JR, Kohorst M. Pediatric neoplasms presenting with monocytosis. Curr Hematol Malig Rep. 2021;16(3):235-246. 10. Murakami N, Okuno Y, Yoshida K, et al. Integrated molecular profiling of juvenile myelomonocytic leukemia. Blood. 2018;131(14):1576-1586. 11. Locatelli F, Niemeyer CM. How I treat juvenile myelomonocytic leukemia. Blood. 2015;125(7):1083–1090. 12. Hasegawa D, Bugarin C, Giordan M, et al. Validation of flow cytometric phospho-STAT5 as a diagnostic tool for juvenile myelomonocytic leukemia. Blood Cancer J. 2013;3(11):e160. 13. Kotecha N, Flores NJ, Irish JM, et al. Single-cell profiling identifies aberrant STAT5 activation in myeloid malignancies with specific clinical and biologic correlates. Cancer Cell. 2008;14(4):335-343. 14. Olk-Batz C, Poetsch AR, Nöllke P, et al. Aberrant DNA methylation characterizes juvenile myelomonocytic leukemia with poor outcome. Blood. 2011;117(18):4871-4880. 15. Schonung M, Meyer J, Nollke P, et al. International consensus definition of DNA methylation subgroups in juvenile myelomonocytic leukemia. Clin Cancer Res. 2021;27(1):158-168. 16. Poetsch AR, Lipka DB, Witte T, et al. RASA4 undergoes DNA hypermethylation in resistant juvenile myelomonocytic leukemia. Epigenetics 2014;9(9):1252-1260.

17. Pinkel D, Arico M, Biondi A, et al. Differentiating juvenile myelomonocytic leukemia from infectious disease. Blood. 1998;91(1):365-367. 18. Karow A, Baumann I, Niemeyer CM. Morphologic differential diagnosis of juvenile myelomonocytic leukemia-pitfalls apart from viral infection. J Pediatr Hematol Oncol. 2009;31(5):380. 19. Yoshimi A, Kamachi Y, Imai K, et al. Wiskott-Aldrich syndrome presenting with a clinical picture mimicking juvenile myelomonocytic leukaemia. Pediatr Blood Cancer. 2013;60(5):836-841. 20. Loh ML. Recent advances in the pathogenesis and treatment of juvenile myelomonocytic leukaemia. Br J Haematol. 2011;152(6):677-687. 21. Chan RJ, Cooper T, Kratz CP, Weiss B, Loh ML. Juvenile myelomonocytic leukemia: a report from the 2nd International JMML Symposium. Leuk Res. 2009;33(3):355-362. 22. Locatelli F, Algeri M, Merli P, Strocchio L. Novel approaches to diagnosis and treatment of juvenile myelomonocytic leukemia. Exp Rev Hematol. 2018;11(2):129-143. 23. Gupta AK, Meena JP, Chopra A, Tanwar P, Seth R. Juvenile myelomonocytic leukemia-A comprehensive review and recent advances in management. Am J Blood Res. 2021;11(1):1-21. 24. Oliveira AF, Tansini A, Toledo T, et al. Immunophenotypic changes in juvenile myelomonocytic leukaemia after treatment with hypomethylating agent: do they help to evaluate depth of response? Br J Haematol. 2022;197(3):339-348. 25. Oliveira AF, Tansini A, Vidal DO, Lopes LF, Metze K, LorandMetze I. Characteristics of the phenotypic abnormalities of bone marrow cells in childhood myelodysplastic syndromes and juvenile myelomonocytic leukemia. Pediatr Blood Cancer. 2017;64(4):e26285. 26. Frisanco Oliveira A, Tansini A, Toledo TR, et al. Immunophenotypic characteristics of juvenile myelomonocytic leukaemia and their relation with the molecular subgroups of the disease. Br J Haematol. 2021;192(1):129-136. 27. Van Dongen JJM, Lhermitte L, Böttcher S, et al. EuroFlow antibody panels for standardized n-dimensional flow cytometric immunophenotyping of normal, reactive and malignant leukocytes. Leukemia. 2012;26(9):1908-1975. 28. Matarraz S, Almeida J, Flores-Montero J, et al. Introduction to the diagnosis and classification of monocytic-lineage leukemias by flow cytometry. Cytometry B Clin Cytom. 2017;92(3):218-227. 29. Kalina T, Flores-Montero J, Van Der Velden VHJ, et al. EuroFlow standardization of flow cytometer instrument settings and immunophenotyping protocols. Leukemia. 2012;26(9):1986-2010. 30. Niemeyer CM. JMML genomics and decisions. Hematology Am Soc Hematol Educ Program. 2018;2018(1):307-312. 31. Lhermitte L, Barreau S, Morf D, et al. Automated identification of leukocyte subsets improves standardization of databaseguided expert-supervised diagnostic orientation in acute leukemia: a EuroFlow study. Mod Pathol. 2021;34(1):59-69.

Haematologica | 109 February 2024

531

C. Bugarin et al.

ARTICLE - Phenotypic profiling of CD34+ cells in JMML 32. Locatelli F, Nöllke P, Zecca M, et al. Hematopoietic stem cell transplantation (HSCT) in children with juvenile myelomonocytic leukemia (JMML): results of the EWOG-MDS/ EBMT trial. Blood. 2005;105(1):410-419. 33. Yabe M, Ohtsuka Y, Watanabe K, et al. Transplantation for juvenile myelomonocytic leukemia: a retrospective study of 30 children treated with a regimen of busulfan, fludarabine, and melphalan. Int J Hematol. 2015;101(2):184-190. 34. Dvorak CC, Satwani P, Stieglitz E, et al. Disease burden and conditioning regimens in ASCT1221, a randomized phase II trial in children with juvenile myelomonocytic leukemia: a Children’s Oncology Group study. Pediatr Blood Cancer. 2018;65(7):e27034. 35. Mayerhofer C, Niemeyer CM, Flotho C. Current treatment of juvenile myelomonocytic leukemia. J Clin Med. 2021;10(14):3084. 36. Lipka DB, Witte T, Toth R, et al. RAS-pathway mutation patterns define epigenetic subclasses in juvenile myelomonocytic

leukemia. Nat Commun. 2017;8(1):2126. 37. Calvo KR, Price S, Braylan RC, et al. JMML and RALD (Rasassociated autoimmune leukoproliferative disorder): common genetic etiology yet clinically distinct entities. Blood. 2015;125(18):2753-2758. 38. Hecht A, Meyer JA, Behnert A, et al. Molecular and phenotypic diversity of CBL-mutated juvenile myelomonocytic leukemia. Haematologica. 2022;107(1):178-186. 39. Mariani RA, Jennings L, Zhang S, Bhat R, Gong S. Morphologic and immunophenotypic differences in juvenile myelomonocytic leukemias with CBL and other canonical RAS-pathway gene mutations: a single institutional experience. J Pediatr Hematol Oncol. 2021;43(6):e819-e825. 40. Niemeyer CM, Flotho C, Lipka DB, et al. Response to upfront azacitidine in juvenile myelomonocytic leukemia in the AZAJMML-001 trial. Blood Adv. 2021;5(14):2901-2908.

Haematologica | 109 February 2024

532

ARTICLE - Non-Hodgkin Lymphoma

Results from a phase I trial of pembrolizumab plus vorinostat in relapsed/refractory B-cell non-Hodgkin lymphoma James Godfrey,1 Matthew Mei,1 Lu Chen,2 Joo Y. Song,3 Victoria Bedell,3 L. Elizabeth Budde,1 Saro Armenian,4 Sandrine Puverel,1 Liana Nikolaenko,1 Robert Chen,1 Shari Daniels,1 Neena

Correspondence: A. F. Herrera

Kennedy,1 Lacolle Peters,1 Steven T. Rosen,1 Stephen J. Forman,1 Leslie L. Popplewell,1 Larry W.

aherrera@coh.org

Kwak1 and Alex F. Herrera1 1

Received: Accepted: Early view:

Information Sciences; 3Department of Pathology and 4Department of Pediatrics, City of Hope

https://doi.org/10.3324/haematol.2023.283002

Department of Hematology & Hematopoietic Cell Transplantation; 2Department of

Medical Center, City of Hope, Duarte, CA, USA

March 3, 2023. July 7, 2023. July 20, 2023.

Abstract Outcomes after programmed death-1 (PD-1) blockade in B-cell lymphomas are disappointing with few durable responses. Histone deacetylase inhibitors exhibit favorable immunomodulatory effects and demonstrate synergistic anti-tumor immune responses with anti-PD-1 therapy in preclinical models. We, therefore, developed a phase I study to evaluate the safety and preliminary efficacy of pembrolizumab with vorinostat in relapsed/refractory B-cell lymphomas. Patients were treated in a dose-escalation cohort using a Rolling 6 design followed by an expansion cohort at the recommended phase II dose (R2PD). Fifty-two patients were enrolled (32 Hodgkin and 20 non-Hodgkin lymphoma [NHL]). Here, we report safety data from the dose escalation cohort, and the toxicity and efficacy within NHL patients. Vorinostat was administered twice daily on days 1-5 and 8-12 (dose-level [DL]1: 100 mg; DL2: 200 mg) and pembrolizumab (200 mg) was administered on day 1 of each 3-week cycle. Of six patients treated at DL1, one had a dose-limiting toxicity (DLT) (Stevens-Johnson syndrome [SJS]), and one of six had a DLT at DL2 (thromboembolism); therefore, DL2 was the RP2D. The patient developing SJS was treated with corticosteroids, infliximab, and cyclosporine but ultimately died of invasive fungal infection from the extensive immunosuppression used to treat the SJS. The most common adverse events were hypertension, diarrhea, and cytopenias. Of 20 NHL patients, nine had follicular lymphoma (FL) and 11 had diffuse large B-cell lymphoma (DLBCL). Five DLBCL patients had primary mediastinal B-cell lymphoma (PMBL). The complete and overall response rates (CR and ORR) were 11% and 22% for FL and 45% and 55% for all DLBCL. Amongst DLBCL, the CR and ORR was 80% and 80% for PMBL and 17% and 33% for non-PMBL. In conclusion, pembrolizumab with vorinostat was tolerable and produced responses in relapsed/refractory B-cell NHL, with particularly notable efficacy in PMBL (clinicaltrials gov. Identifier: NCT03150329).

Introduction B-cell non-Hodgkin lymphoma (NHL) such as follicular lymphoma (FL) and diffuse large B-cell lymphoma (DLBCL) are the most common lymphomas diagnosed in the United States.1 Currently, most FL and a large proportion of DLBCL patients will develop relapsed/refractory (r/r) disease, which has been historically associated with poor outcomes in the chemotherapy era.2,3 More recently, however, the treatment paradigm of r/r DLBCL and FL has shifted away from conventional chemotherapy and towards the use of highly effective novel immunotherapies. This includes the recent development of CD19-directed chimeric antigen receptor

(CAR) T cells,4 CD20 bispecific antibodies (BsAb),5 antibody drug conjugates (e.g., polatuzumab vedotin, loncastuximab tesirine),6,7 and other antibody-based therapies (e.g., tafasitamab).8 These treatments are each associated with high response rates and have greatly improved outcomes in patients with r/r B-cell NHL. However, despite these therapeutic advances, the reality remains that the majority of patients with r/r B-cell NHL will still not achieve a durable remission with immunotherapy or more traditional approaches like stem cell transplantation.2,4,5 Treatment options are also limited for patients who progress after CAR T-cell therapy or stem cell transplantation, and the median overall survival (OS) is only 5.2 months for patients

Haematologica | 109 February 2024

533

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL

with aggressive B-cell lymphomas relapsing after CAR T-cell therapy.9 The development of safe and effective therapies for r/r B-cell NHL, therefore, remains an area of significant unmet need The clinical efficacy of PD-1 blockade has transformed the treatment landscape of a number of cancer types.10–13 The outcomes of anti-PD-1 monotherapy in r/r B-cell NHL, however, have been disappointing. For instance, the response rates to single-agent PD-1 blockade in r/r FL and DLBCL are only 4% and 10%, respectively.14,15 The one notable exception are B-cell NHL that harbor programmed death-1 (PD-L1) gene alterations, including gray zone lymphomas, primary mediastinal B-cell lymphoma (PMBL), and certain subtypes of DLBCL.16–18 In r/r PMBL for instance, up to 50% of patients respond to anti-PD-1 monotherapy. However, while complete responses (CR) appear durable, only a minority of PMBL patients achieve CR and the majority will eventually progress following anti-PD-1 therapy.17 This presents a major challenge in a population of patients commonly defined by young age and where cure is the goal of treatment. Given these challenges, we sought to investigate whether a PD-1-based combination regimen could result in improved outcomes. Histone deacetylase inhibitors (HDACi) such as vorinostat are epigenetic-modifying cancer treatments that are Food and Drug Administration-approved for the treatment of certain NHL.19,20 HDACi also exhibit favorable immunomodulatory effects that improve anti-tumor immune responses generated in the setting of PD-1 blockade in various preclinical tumor models.21–24 HDACi, for example, enhance tumor antigen presentation, increase recruitment of T cells into the tumor environment, and promote the function of tumor-reactive T cells, which results in significantly improved responses to PD-1 blockade therapy in preclinical models.21–24 HDACi also increase PD-L1 expression on malignant cells from various tumor types, which may be an important determinant associated with PD-1 response in lymphoma and other malignancies.23 HDACi and other epigenetic modifying therapies (DNMT3A inhibitors) have now been studied in combination with PD-1 blockade for a number of cancers and there have been early clinical signs of potential synergy.21,25,26 In classic Hodgkin lymphoma (cHL), for example, the combination of PD-1 blockade with DNMT3Ai was associated with a higher CR rate and progression-free survival (PFS) compared to anti-PD-1 monotherapy.25,26 We, therefore, hypothesized that adding the immunomodulatory pan-HDACi, vorinostat, to the anti-PD1 antibody, pembrolizumab, in patients with r/r FL, DLBCL, PMBL, and cHL would be safe and boost the anti-tumor activity of PD-1 blockade in B-cell lymphomas. Here we report the results of our phase I clinical trial evaluating the safety and preliminary efficacy of pembrolizumab with vorinostat in r/r B-cell lymphomas. In this manuscript, we report the safety data from the dose escalation cohort and the toxicity data and efficacy results from the NHL cohort of patients.

Methods Patients This was a single-center phase I dose-escalation trial with a planned expansion cohort. Eligible patients were 18 years old or older with r/r FL, DLBCL, PMBL, or cHL who had relapsed or progressed after at least one prior line of therapy and were transplant-ineligible or who refused transplant. Initially, prior anti-PD-1 exposure was allowed if patients had evidence of a prior objective response to PD-1 blockade. On May 14, 2019, the protocol was amended to allow patients to enroll regardless of response to prior anti-PD-1 therapy to facilitate study enrollment and assess efficacy in patients with cHL who progressed on prior PD-1 blockade. Additional inclusion criteria were as follows: Eastern Cooperative Oncology Group performance status of 0-1, total bilirubin ≤1.5x upper limit of normal (ULN) or direct bilirubin ≤ULN for subjects with total bilirubin levels >1.5x ULN, AST/ALT ≤2.5x ULN, and PT/INR ≤1.5x ULN and PTT (aPTT) ≤1.5x ULN unless the patient was receiving anticoagulant therapy in which case PT or PTT had to be within therapeutic range of the intended use of anticoagulants. Patients with known Gilbert’s disease were allowed to have a total bilirubin of up to ≤3x ULN and AST/ALT up to ≤3x ULN, and patients with lymphomatous involvement of the liver were allowed to be enrolled as long as AST/ALT ≤5x ULN. Hematologic parameters required were as follows: absolute neutrophil count (ANC) ≥1,000/μL, platelet count (Plt) ≥75,000/μL, and hemoglobin ≥8 g/dL without use of an erythropoiesis-stimulating agent within 7 days of assessment; patients with known bone marrow involvement by lymphoma were not required to meet these parameters. Patients had to be willing to provide tissue from a fresh core or excisional biopsy prior to starting study therapy or from archival tissue of a biopsy that was performed after the most recent systemic therapy. Other exclusion criteria included a diagnosis of immunodeficiency or any immunosuppressive therapy including systemic corticosteroids within 7 days prior to the first dose of trial treatment, prior allogeneic stem cell transplantation within 5 years or active graft-versus-host-disease, prior autoHCT within 60 days, active autoimmune disease requiring systemic treatment (replacement therapy such as thyroxine or insulin excepted), history of non-infectious pneumonitis requiring steroids or current pneumonitis, or a QT interval corrected for heart rate (QTc) >470 ms using the Fridericia formula. Patients with known active HIV, hepatitis B, or hepatitis C infection were ineligible. All patients provided informed consent for participation in the clinical trial. The study was approved by the Institutional Review Board and conducted in accordance with the principles of the Declaration of Helsinki. Study treatment Patients were treated in a dose-escalation cohort with two dose levels (DL) using a Rolling 6 design and then onto

Haematologica | 109 February 2024

534

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL Table 1. Clinical characteristics. Clinical characteristics

All DLBCL, N=11

PMBCL, N=5

Age in years, median (range) Male, N (%)

51 (21-79) 5 (45)

34 (21-51) 2 (40)

0 (0) 6 (55) 4 (36) 1 (9)

0 (0) 2 (40) 2 (40) 1 (20)

Ethnicity, N (%) Hispanic White Asian Black

Histology, N (%) DLBCL PMBCL Transformed DLBCL

4 (36) 5 (45) 2 (18)

Cell-of-origin by Hans criteria, N (%) Non-GCB Unknown Stage 3-4, N (%) Extranodal involvement, N (%) Primary refractory, N (%) Refractory to most recent therapy, N (%) IPI Score, N (%) Low risk (0-1) Low-intermediate risk (2) Double-expressor, N (%) Yes No Unknown

Double-hit, N (%) Yes No Unknown Prior CAR T-cell therapy, N (%)

9 (82) 2 (18) 4 (36) 5 (45) 7 (64) 10 (91)

4 (80) 1 (20) 3 (60) 2 (40) 4 (80) 5 (100)

5 (45) 6 (55)

2 (40) 3 (60)

4 (36) 5 (45) 2 (18)

0 (0) 4 (80) 1 (20)

1 (9) 8 (73) 2 (18) 3 (21)

1 (20) 3 (60) 1 (20) 1 (20)

Clinical characteristics

FL, N=9

Age in years, median (range) Male, N (%) Ethnicity, N (%) Hispanic White Asian

60 (28-78) 8 (89) 2 (22) 7 (78) 2 (22)

Grade, N (%) 2 3A

6 (67) 3 (33)

Stage 3-4, N (%) Extranodal involvement, N (%) Refractory to most recent therapy, N (%) Rituximab-resistant, N (%)

6 (67) 3 (33) 5 (56) 9 (100) 9 (100)

Double-refractory*, N (%)

FLIPI score category, N (%) Good (0-1) Intermediate (2) Poor (3-5)

4 (44) 2 (22) 3 (33)

Best response

FL, N=9

DLBCL, N=11

Non-PMBL DLBCL, N=6

PMBL, N=5

Cycles, median (range) Complete response, N (%) Partial response, N (%) Stable disease, N (%) Progressive disease, N (%)

4 (2-16) 1 (11) 1 (11) 6 (67) 1 (11)

5 (1-35) 5 (45) 1 (9) 2 (18) 3 (27)

3 (1-32) 1 (17) 1 (17) 2 (33) 2 (33)

32 (1-35) 4 (80) 0 (0) 0 (0) 1 (20)

* Double-refractory indicates refractory to rituximab and an alkylating chemotherapy. DLBCL: diffuse large B-cell lymphoma; PMBCL: primary mediastinal B-cell lymphoma; GCB: germinal center B cell; CAR: chimeric antigen rector; IPI: International Prognostic Index; FL: follicular lymphoma; FLIPI: Follicular Lymphoma International Prognostic Index.

Haematologica | 109 February 2024

535

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL

an expansion cohort with treatment at the recommended phase II dose (RP2D). In DL1, vorinostat was administered orally at 100 mg twice daily on days 1-5 and 8-12 and in DL2, vorinostat was administered at 200 mg twice daily on days 1-5 and 8-12. Pembrolizumab dose was 200 mg intravenously (IV) on day 1; the cycle length was 21 days. Treatment could continue for a maximum of 2 years. Patients with disease progression could continue on therapy at the discretion of the principal investigator provided that there were no signs or symptoms of progressive disease, no decline in Eastern Cooperative Oncology Group performance status, and absence of progressive tumor at critical anatomical sites requiring urgent medical intervention. Study assessment and endpoints Safety was monitored continuously with toxicities assessed using the Common Terminology Criteria for Adverse Events (CTCAE) v4.0. Dose-limiting toxicity (DLT) was assessed in the first two cycles. DLT was defined as any of the following at least possibly related to study treatment: grade 4 neutropenia lasting >7 days, grade 4 thrombocytopenia lasting >7 days or requiring platelet transfusion, grade 3 or 4 thrombocytopenia associated with grade 2 or higher bleeding, grade 4 anemia not associated with lymphoma, grade ≥3 pneumonitis that does not resolve to grade ≤1 within 3 days after initiation of supportive care measures, any clinically relevant grade 3 or grade 4 non-hematologic AE with certain exceptions, and any grade 5 AE. Positron emission tomography/computed tomography (PET/CT) scans were performed at baseline and then every four cycles until disease progression or off-study therapy; a diagnostic quality CT with IV contrast was acceptable if a CR was previously confirmed by PET/CT. Responses were assessed by investigators according to the 2014 Lugano Classification.27 The primary endpoints were safety, tolerability, and determination of the RP2D. Secondary endpoints included the CR rate, duration of response (DOR), OS, and PFS. For DOR and PFS, failures included disease relapse/progression or death due to any cause. DOR/PFS were censored at the last follow-up or start of other non-protocol therapy, whichever occurred earlier. Trial design and statistical considerations Dose escalation followed a Rolling 6 design.28 Maximum tolerated dose (MTD) was defined as the highest dose level with at most one out of six participants with DLT; RP2D would be at or lower than MTD. Participants evaluable for DLT during dose escalation needed to complete treatment through DLT period (2 cycles), receive the planned pembrolizumab dose and miss <30% vorinostat doses during DLT period, except due to DLT. DOR, PFS and OS were estimated based on Kaplan-Meier product limit method with Greenwood variance estimator and along with confidence interval (CI) estimated based on log-log transformation.

Correlative analyses PD-L1 protein expression was assessed by immunohistochemistry (IHC) (SP263 clone) using an H-score system, and PD-L1 gene alterations were identified using fluorescent in situ hybridization as previously described.18

Results Patients Twenty patients with FL, DLBCL, or PMBL were enrolled and received study therapy. Baseline characteristics are listed in Table 1. Of the 20 patients enrolled, nine had a diagnosis of FL and 11 had DLBCL. Of the 11 DLBCL patients, five had PMBL. Thirteen (65%) patients were male and the median age was 59 years (range, 21-79). The median number of prior lines of therapy were three (range, 1-7 prior lines). A large proportion of patients exhibited high-risk features, including 15 (75%) who were refractory to their most recent line of therapy, three (15%) who progressed after prior CAR T-cell therapy, and four (20%) who progressed after prior BsAb therapy. Treatment disposition and safety All 20 patients received at least one dose of treatment and were evaluable for safety endpoints. A median number of four cycles of therapy were administered (range, 1-35 cycles), and all subjects have discontinued treatment. Reasons for treatment discontinuation included: disease progression/insufficient response/stable disease (n=12, 60%), toxicity (n=3, 15%), completion of study treatment (n=2, 10%), discontinuation of study treatment while in CR (n=1, 5%), proceeding to consolidative stem cell transplant (n=1, 5%), and patient preference (n=1, 5%). On the entire study (including cHL patients not included in this report), two subjects experienced a DLT during the dose escalation portion of the study (6 each treated at DL1 and DL2), both had FL. One patient experienced a grade 3 Stevens-Johnson syndrome (SJS) at DL1, and one patient experienced a DLT of grade 3 pulmonary embolism at DL2. Therefore, DL2 (200 mg vorinostat twice daily and 200 mg pembrolizumab) was established as the RP2D for the dose expansion cohort. The most common adverse events (AE) among the 20 NHL patients treated on the study (DL1 or DL2) are listed in Table 2. These included hypertension (70%), diarrhea (65%), nausea (65%), fatigue (60%), leukopenia (60%), and anemia (55%). Grade ≥3 AE included neutropenia (15%), lymphopenia (10%), mucositis (5%), hyperkalemia (5%), hypertension (5%), pulmonary embolism (5%), anemia (5%), and leukopenia (5%). Immune-related AE among NHL patients included the patient with SJS. Three other patients experienced grade 1-2 immune-related hypo- and/ or hyper-thyroidism. There were three patients with a delay of pembrolizumab, one patient discontinued vorinostat for grade 2 nausea and anorexia, and one patient had a vori-

Haematologica | 109 February 2024

536

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL

nostat dose reduction from 200 mg twice daily to 100 mg twice daily due to abdominal cramping. The patient with extensive mucocutaneous SJS was treated with high-dose corticosteroids without resolution. Infliximab and cyclosporine were also used, but the patient died of invasive fungal infection due to the extensive immunosuppression used to treat the ongoing SJS.

Table 2. Adverse events with an attribution of possibly or higher.

Efficacy and correlatives Among all 20 patients, the ORR and CR were 8 and 20 (40%) and 6 and 20 (30%), respectively. Individual response characteristics by disease subtype are shown in Table 1 and Figure 1. The disease-specific CR and ORR were 1 and 9 and 2 and 9 (11% and 22%) for FL; 5 and 11 and 6 and 11 (45% and 55%) for all DLBCL; and 4 and 5 and 4 and 5 (80% and 80%) for PMBL. Disease control rates (CR, PR, and SD) were 89% for FL, 73% for all DLBCL, and 80% for PMBL. Among the three patients who received prior CAR T-cell therapy, there were one CR and two PD; among the four patients with prior BsAb therapy, there were one CR and three SD. Of the two patients achieving CR after prior CAR T-cell or BsAb therapy, one had PMBL with a best response of stable disease to prior Liso-cel CAR T. The patient achieved a CR to pembrolizumab plus vorinostat and then proceeded to a consolidative allogeneic stem cell transplant after five cycles on study and remains in ongoing remission after approximately 4 years. The second CR patient had FL and progressed after prior CD20 BsAb. This patient remained in CR on study before progression after 12 cycles and off-treatment after 16 cycles. The patient subsequently had a brief response to CAR T-cell therapy and was then lost to follow up. Seven subjects have died as of the data cutoff with a median follow-up of 3.8 years (range, 2.9-5.1) among the survivors. The most common cause of death was disease progression (n=5, 2 FL and 3 DLBCL). One FL patient died from pulmonary fungal infection approximately 3 months after coming off study treatment, and another FL patient died approximately 10 months after off-study treatment and cause of death was unavailable. Four subjects started other therapy without disease progression and therefore were censored for PFS/DOR, three of them were SD on treatment and the fourth was a CR patient who went to transplant. Median PFS and DOR were 8.0 (95% CI 2.5-25.0) and 22.3 (95% CI: 0.6-not available) months for the total NHL population (Figure 2). Disease-specific median PFS were 4.0 months (95% CI: 2.1-12.5) for FL, 8.2 months (95% CI: 0.7-not available) for all DLBCL, and not reached amongst the PMBL patients. DOR were 0.6 and 2.8 months for the 2 FL responders, and median DOR was not reached for all patients with DLBCL, including PMBL. Median OS was not reached for the population overall nor for any subset. The 2-year PFS was 29.0% (95% CI: 9.7-52.0) overall, 0% for FL, 49.1% (95% CI: 16.7-75.3) for all DLBCL, and 80%

All grades (>10%)

N (%)

Hypertension

14 (70)

Diarrhea

13 (65)

Nausea

13 (65)

Fatigue

12 (60)

White blood cell decreased

12 (60)

Anemia

11 (55)

Abdominal pain

10 (50)

Neutrophil count decreased

10 (50)

Vomiting

8 (40)

Hyponatremia

8 (40)

Platelet count decreased

7 (35)

Anorexia

7 (35)

Dyspepsia

6 (30)

Myalgia

6 (30)

Constipation

5 (25)

Creatinine increased

5 (25)

Hypermagnesemia

5 (25)

Hypophosphatemia

5 (25)

Lymphocyte count decreased

5 (25)

Weight loss

5 (25)

Hypocalcemia

4 (20)

Headache

4 (20)

Dehydration

4 (20)

Rash maculopapular

4 (20)

Hyperkalemia

3 (15)

Dizziness

3 (15)

Hypoalbuminemia

3 (15)

Fever

3 (15)

Sinus tachycardia

3 (15)

Chills

3 (15)

Sore throat

3 (15)

Malaise

3 (15)

Grade 3+

N (%)

Neutrophil count decreased

3 (15)

Lymphocyte count decreased

2 (10)

Mucositis oral

1 (5)

Hyperkalemia

1 (5)

Hypertension

1 (5)

Thromboembolic event

1 (5)

Anemia

1 (5)

White blood cell decreased

1 (5)

Haematologica | 109 February 2024

537

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL

Figure 1. Response characteristics. Swimmer plot demonstrating response characteristics to pembrolizumab with vorinostat. Of note, 1 follicular lymphoma patient discontinued treatment after 1 month of therapy due to toxicity, but remained on study and achieved a partial response (PR). CR: complete response, PD: progressive disease, DLBCL: diffuse large B-cell lymphoma, FL: follicular lymphoma; PMBCL: primary mediastinal B-cell lymphoma.

(95% CI: 20.4-96.9) for PMBL. The 2-year OS was 65.0% (95% CI: 40.3-81.5) overall, 55.6% (95% CI: 20.4-80.5) for FL, 72.7% (95% CI: 37.1-90.3) for all DLBCL, and 100% for PMBL. We performed PD-L1 immunohistochemistry and PD-L1 fluorescence in situ hybridization on archived tumor samples and assessed association between PD-L1 expression, 9p24.1 alterations, and response to treatment. Eight patients had samples available for correlative testing (5 DLBCL and 3 PMBL). All three PMBL patients had PD-L1 gene amplifications, and two of these cases demonstrated intense PD-L1 protein expression on tumor cells (Online Supplementary Table S1; Online Supplementary Figure S1). All three of these patients achieved a CR to pembrolizumab plus vorinostat. Among the five non-PMBL DLBCL cases, there were no PD-L1 amplifications or PD-L1 copy gains. PD-L1 protein expression was modest or low in all cases, and was not associated with response. Interestingly, the DLBCL patient achieving a 2-year remission was Epstein-Barr virus-positive as assessed by EBER in situ hybridization staining.

Discussion In this phase I study of pembrolizumab and vorinostat in patients with r/r B-cell NHL, we observed objective responses in highly refractory patients, with particularly notable preliminary activity in PMBL. The combination of vorinostat and pembrolizumab exhibited only modest clinical activity in FL or in non-PMBL DLBCL (ORR: 22% and 33%, respectively). The combination was however well-tolerated in most patients. The RP2D was determined to be 200 mg twice daily of vorinostat on days 1-5 and 8-12 in combination with 200 mg of pembrolizumab on day 1 of each 3-week cycle. There was a low incidence of grade ≥3 AE with this dosing schedule, and the overall safety profile was consistent with the known single-agent toxicities of each drug. Side effects were primarily related to hypertension, hematological, and gastrointestinal toxicities, and were largely manageable with a low rate of treatment discontinuation. In our study, we noted a high ORR and CR rate to pembrolizumab with vorinostat in PMBL. The 80% CR rate we observed in PMBL patients compares favorably to the

Haematologica | 109 February 2024

538

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL

Figure 2. Survival outcomes. (A) Progression-free survival (PFS), (B) overall survival (OS) and (C) duration of response (DOR) to pembrolizumab with vorinostat. (D) PFS, (E) OS and (F) DOR stratified by follicular lymphoma (FL) and diffuse large B-cell lymphoma (DLBCL). Haematologica | 109 February 2024

539

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL

reported data of pembrolizumab monotherapy where CR rates were noted to only be 19% in a combined phase I/II study of 74 PMBL patients.17 Achieving a CR to PD-1-based treatment in PMBL appears to be an especially important milestone because all PMBL patients achieving a CR on the aforementioned phase I/II study of pembrolizumab remain in remission.17 Similarly durable responses were observed in our study as well as a study of nivolumab plus brentuximab vedotin.29 While there were only five PMBL patients enrolled on our study and our findings would require validation in a larger cohort, the high CR rate is nonetheless an encouraging early signal of activity in PMBL based on our similar findings of encouraging activity in heavily treated cHL, a disease that shares biological features with PMBL.30–32 In particular, PMBL and cHL are both JAK/STAT-driven tumors that frequently exhibit intense PD-L1 expression as the result of gene amplifications and activating translocations occurring within the CD274 (PD-L1) locus.33,34 Thus, adding vorinostat to PD-1 blockade may be a particularly useful method by which to improve the efficacy of anti-PD-1 monotherapy in strongly PD-L1+ lymphomas such as cHL, PMBL, and certain subsets of DLBCL.18,33–35 Interestingly, other epigenetic modifying therapies have also demonstrated robust clinical responses when combined with PD-1 blockade in cHL. In particular, combining the DNMT3A inhibitor, decitabine, with PD-1 blockade is associated with high CR and ORR in PD-1 refractory cHL,26,36 and the combination significantly improves CR rates and PFS compared to anti-PD-1 monotherapy in anti-PD-1 naïve patients.25 Combining DNMT3A inhibitors with PD-1 blockade may, therefore, represent a promising treatment strategy in PMBL and other strongly PD-L1+ NHL given the biological similarities between cHL and PMBL noted above. Moreover, these data suggest that epigenetic modifying therapies should be explored further in combination with PD-1 blockade in PD-L1+ lymphomas, either in novel combinations (e.g., HDACi + DNTM3Ai + anti-PD-1) or during earlier lines of therapy (e.g., elderly/frail patients or as a bridge to consolidative cellular therapy or stem cell transplantation). Alternatively, pembrolizumab plus vorinostat could also be evaluated in PD-L1+ lymphomas relapsing after CAR T-cell therapy, as this represents a growing patient population in need of improved treatment options as their expected median survival is only 5.2 months.9 Lastly, future correlative analyses should investigate the mechanisms by which DNMT3Ai and HDACi enhance the efficacy of anti-PD-1 therapies in lymphoma and whether those mechanisms are potentially complementary in action. While the clinical outcomes of pembrolizumab plus vorinostat are promising in PMBL, the combination exhibited only modest activity in FL and non-PMBL DLBCL. Collectively, the CR and ORR of pembrolizumab with vorinostat were only 13% and 26% in these NHL subtypes, respectively. These data suggest only a marginal incremental benefit

compared to historical data of anti-PD-1 monotherapy where ORR are 4-10%.14,15 Together, our results contribute to a growing number of negative studies evaluating PD1-based combinations in DLBCL and FL. These include studies evaluating PD-1 blockade in combination with other immune checkpoint inhibitors,37 BTK inhibitors,38 as well as with CAR T-cell therapy.39 Thus, future efforts investigating anti-PD-1 therapies in these diseases should have strong scientific merit or restrict inclusion to PD-L1+ cases that may have increased sensitivity to PD-1-based therapy. Moreover, the results of these studies provide an important lesson for the field on clinical trial development, as numerous negative PD-1 combination trials have now been developed in FL and DLBCL based on the results of early anti-PD-1 trials that included only small numbers of patients.40 Thus, while accelerating drug development is critical to advancing the field and improving treatment options for our patients, we should ideally wait for more mature efficacy results from larger datasets before allocating significant resources to study a novel therapy. In summary, the combination of pembrolizumab with vorinostat is safe in r/r B-cell NHL and elicits objective responses in patients with highly refractory disease. Particularly high rates of durable CR in PMBL support further investigation of this treatment regimen in this specific patient population, both in the PD-1 naïve and PD-1 refractory settings. Disclosures AFH reports research funding from BMS, Merck, Genentech, Inc/F. Hoffman-La Roche Ltd, Gilead Sciences, Seattle Genetics, AstraZeneca, and ADC Therapeutics; consultancy for BMS, Merck, Genentech, Inc/F. Hoffmann-La Roche Ltd, Kite Pharma/Gilead, Seattle Genetics, Karyopharm, Takeda, Tubulis, AstraZeneca, Pfizer, Caribou Biosciences, Adicet Bio, Abbvie, Allogene Therapeutics, Genmab, ADC Therapeutics, and Regeneron. MM reports research funding from TG Therapeutics, Epizyme, and BMS; consultancy with Morphosys and GlaxoSmithKline. JG reports research funding from Verastem, Merck, and Forty Seven, Inc. LN is a current employee of Kite Pharma. Contributions AFH, SP and RC conceived the study design. JG, AFH, MM and LC analyzed the data. JG, AFH and LC wrote the manuscript. AFH, LC and JG collected, assembled and interpreted the data. MM, EB, SA, SP, LN, RC, SD, NK, LP, STR, SJF, LLP, LWK, JS and VB collected and assembled the data, interpreted the data, and revised the manuscript. Funding AFH was supported by the Emmet and Toni Stephenson Leukemia, the Lymphoma Society Scholar Award, the Lymphoma Research Foundation Larry and Denise Mason Clinical Investigator Career Development Award. Research reported

Haematologica | 109 February 2024

540

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL

in this article included work performed in the Biostatistics and Mathematical Oncology Core supported by the National Cancer Institute of the National Institutes of Health under award number P30CA033572. The research was also supported by the City of Hope Cytogenetics Department. This study was also supported in part by a research grant from Investigator-Initiated Studies Program of Merck Sharp &

Dohme LLC. The opinions expressed in this paper are those of the authors and do not necessarily represent those of Merck Sharp & Dohme LLC. Data-sharing statement Data used to support the findings of this study are available from the corresponding author upon request.

References 1. Teras LR, DeSantis CE, Cerhan JR, Morton LM, Jemal A, Flowers CR. 2016 US lymphoid malignancy statistics by World Health Organization subtypes. CA Cancer J Clin. 2016;66(6):443-459. 2. Gisselbrecht C, Glass B, Mounier N, et al. Salvage regimens with autologous transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin Oncol. 2010;28(27):4184-4190. 3. Casulo C, Byrtek M, Dawson KL, et al. Early relapse of follicular lymphoma after rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone defines patients at high risk for death: an analysis from the National LymphoCare Study. J Clin Oncol. 2015;33(23):2516-2522. 4. Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene ciloleucel CAR T-cell therapy in Refractory Large B-Cell Lymphoma. N Engl J Med. 2017;377(26):2531-2544. 5. Budde LE, Assouline S, Sehn LH, et al. Single-agent mosunetuzumab shows durable complete responses in patients with relapsed or refractory B-cell lymphomas: phase I dose-escalation study. J Clin Oncol. 2022;40(5):481-491. 6. Tilly H, Morschhauser F, Sehn LH, et al. Polatuzumab vedotin in previously untreated diffuse large B-cell lymphoma. N Engl J Med. 2022;386(4):351-363. 7. Caimi PF, Ai W, Alderuccio JP, et al. Loncastuximab tesirine in relapsed or refractory diffuse large B-cell lymphoma (LOTIS-2): a multicentre, open-label, single-arm, phase 2 trial. Lancet Oncol. 2021;22(6):790-800. 8. Salles G, Duell J, González Barca E, et al. Tafasitamab plus lenalidomide in relapsed or refractory diffuse large B-cell lymphoma (L-MIND): a multicentre, prospective, single-arm, phase 2 study. Lancet Oncol. 2020;21(7):978-988. 9. Di Blasi R, Le Gouill S, Bachy E, et al. Outcomes of patients with aggressive B-Cell lymphoma after failure of anti-CD19 CAR T-cell therapy: a DESCAR-T analysis. Blood. 2022;140(24):2584-2593. 10. Herbst RS, Baas P, Kim D-W, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial. Lancet. 2016;387(10027):1540-1550. 11. Seiwert TY, Burtness B, Mehra R, et al. Safety and clinical activity of pembrolizumab for treatment of recurrent or metastatic squamous cell carcinoma of the head and neck (KEYNOTE-012): an open-label, multicentre, phase 1b trial. Lancet Oncol. 2016;17(7):956-965. 12. Robert C, Long G v., Brady B, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med. 2015;372(4):320-330. 13. Le DT, Uram JN, Wang H, et al. PD-1 Blockade in tumors with mismatch-repair deficiency. N Engl J Med. 2015;372(26):2509-2520.

14. Armand P, Janssens A, Gritti G, et al. Efficacy and safety results from CheckMate 140, a phase 2 study of nivolumab for relapsed/ refractory follicular lymphoma. Blood. 2021;137(5):637-645. 15. Ansell SM, Minnema MC, Johnson P, et al. Nivolumab for relapsed/refractory diffuse large B-cell lymphoma in patients ineligible for or having failed autologous transplantation: a single-arm, phase II study. J Clin Oncol. 2019;37(6):481-489. 16. Melani C, Major A, Schowinsky J, et al. PD-1 blockade in mediastinal gray-zone lymphoma. N Engl J Med. 2017;377(1):89-91. 17. Armand P, Rodig S, Melnichenko V, et al. Pembrolizumab in relapsed or refractory primary mediastinal large b-cell lymphoma. J Clin Oncol. 2019;37(34):3291-3299. 18. Godfrey J, Tumuluru S, Venkataraman G, et al. PD-L1 gene alterations identify a subset of diffuse large B cell lymphoma harboring a T cell inflamed phenotype. Blood. 2019;133(21):2279-2290. 19. Olsen EA, Kim YH, Kuzel TM, et al. Phase IIB multicenter trial of vorinostat in patients with persistent, progressive, or treatment refractory cutaneous t-cell lymphoma. J Clin Oncol. 2007;25(21):3109-3115. 20. Coiffier B, Pro B, Prince HM, et al. Results from a pivotal, openlabel, phase II study of romidepsin in relapsed or refractory peripheral T-cell lymphoma after prior systemic therapy. J Clin Oncol. 2012;30(6):631-636. 21. Wen T, Sun G, Jiang W, et al. Histone deacetylases inhibitor chidamide synergizes with humanized PD1 antibody to enhance T-cell chemokine expression and augment Ifn-γ response in NK-T cell lymphoma. EBioMedicine. 2023;87:104420. 22. Wang X, Waschke BC, Woolaver RA, et al. Histone deacetylase inhibition sensitizes PD1 blockade-resistant B-cell lymphomas. Cancer Immunol Res. 2019;7(8):1318-1331. 23. Woods DM, Sodré AL, Villagra A, Sarnaik A, Sotomayor EM, Weber J. HDAC inhibition upregulates PD-1 ligands in melanoma and augments immunotherapy with PD-1 blockade. Cancer Immunol Res. 2015;3(12):1375-1385. 24. Akimova T, Beier UH, Liu Y, Wang L, Hancock WW. Histone/ protein deacetylases and T-cell immune responses. Blood. 2012;119(11):2443-2451. 25. Liu Y, Wang C, Li X, et al. Improved clinical outcome in a randomized phase II study of anti-PD-1 camrelizumab plus decitabine in relapsed/refractory Hodgkin lymphoma. J Immunother Cancer. 2021;9(4):e002347. 26. Nie J, Wang C, Liu Y, et al. Addition of low-dose decitabine to anti–PD-1 antibody camrelizumab in relapsed/refractory classical Hodgkin lymphoma. J Clin Oncol. 2019;37(17):1479-1489. 27. Cheson BD, Fisher RI, Barrington SF, et al. Recommendations

Haematologica | 109 February 2024

541

J. Godfrey et al.

ARTICLE - Pembro plus Vorinostat in NHL for initial evaluation, staging, and response assessment of hodgkin and non-hodgkin lymphoma: the Lugano Classification. J Clin Oncol. 2014;32(27):3059-3067. 28. Skolnik JM, Barrett JS, Jayaraman B, Patel D, Adamson PC. Shortening the timeline of pediatric phase I trials: The rolling six design. J Clin Oncol. 2008;26(2):190-195. 29. Zinzani PL, Santoro A, Gritti G, et al. Nivolumab combined with brentuximab vedotin for relapsed/refractory primary mediastinal large B-cell lymphoma: efficacy and safety from the phase II checkmate 436 study. J Clin Oncol. 2019;37(33):3081-3089. 30. Rosenwald A, Wright G, Leroy K, et al. Molecular diagnosis of primary mediastinal B cell lymphoma identifies a clinically favorable subgroup of diffuse large B cell lymphoma related to Hodgkin lymphoma. J Exp Med. 2003;198(6):851-862. 31. Savage KJ, Monti S, Kutok JL, et al. The molecular signature of mediastinal large B-cell lymphoma differs from that of other diffuse large B-cell lymphomas and shares features with classical Hodgkin lymphoma. Blood. 2003;102(12):3871-3879. 32. Herrera AF, Chen L, Budde LE, et al. Pembrolizumab plus vorinostat induces responses in patients with Hodgkin lymphoma who are refractory to prior PD-1 blockade. Blood. 2021;138(Suppl 1):S234. 33. Green MR, Monti S, Rodig SJ, et al. Integrative analysis reveals selective 9p24.1 amplification, increased PD-1 ligand expression, and further induction via JAK2 in nodular sclerosing Hodgkin lymphoma and primary mediastinal large B-cell lymphoma.

Blood. 2010;116(17):3268-3277. 34. Roemer MG, Advani RH, Ligon AH, et al. PD-L1 and PD-L2 genetic alterations define classical Hodgkin lymphoma and predict outcome. J Clin Oncol. 2016;34(23):2690-2697. 35. Kline J, Godfrey J, Ansell SM. The immune landscape and response to immune checkpoint blockade therapy in lymphoma. Blood. 2020;135(8):523-533. 36. Wang C, Liu Y, Dong L, et al. Efficacy of decitabine plus antiPD-1 camrelizumab in patients with Hodgkin lymphoma who progressed or relapsed after PD-1 blockade monotherapy. Clin Cancer Res. 2021;27(10):2782-2791. 37. Armand P, Lesokhin A, Borrello I, et al. A phase 1b study of dual PD-1 and CTLA-4 or KIR blockade in patients with relapsed/ refractory lymphoid malignancies. Leukemia. 2021;35(3):777-786. 38. Younes A, Brody J, Carpio C, et al. Safety and activity of ibrutinib in combination with nivolumab in patients with relapsed nonHodgkin lymphoma or chronic lymphocytic leukaemia: a phase 1/2a study. Lancet Haematol. 2019;6(2):e67-e78. 39. Jacobson CA, Locke FL, Miklos DB, et al. End of phase 1 results from Zuma-6: axicabtagene ciloleucel (axi-cel) in combination with atezolizumab for the treatment of patients with refractory diffuse large B cell lymphoma. Blood. 2018;132(Supplement 1):4192. 40. Lesokhin AM, Ansell SM, Armand P, et al. Nivolumab in patients with relapsed or refractory hematologic malignancy: preliminary results of a phase ib study. J Clin Oncol. 2016;34(23):2698-2704.

Haematologica | 109 February 2024

542

ARTICLE - Non-Hodgkin Lymphoma

Feasibility and outcomes after dose reduction of immunochemotherapy in young adults with Burkitt lymphoma and leukemia: results of the BURKIMAB14 trial Josep-Maria Ribera,1 Mireia Morgades,1 Olga García-Calduch,1 Maialen Sirvent,2 Buenaventura Buendía,3 Marta Cervera,4 Hugo Luzardo,5 Jesús-María Hernández-Rivas,6 Marta Sitges,7 Irene García-Cadenas,8 Pau Abrisqueta,9 Pau Montesinos,10 Mariana Bastos-Oreiro,11 María-Paz Queipo de Llano,12 Pilar Bravo,13 Anna Torrent,1 Pilar Herrera,14 Antoni Garcia-Guiñon,15 Ferran Vall-llovera, Josefina Serrano, María-José Terol, Juan-Miguel Bergua, Ana García16

Noblejas,20 Cristina Barrenetxea,21 Laura Llorente,22 Daniel García-Belmonte,23 Eva Gimeno,24 Antonia Cladera,25 Santiago Mercadal26 and Juan-Manuel Sancho1 on behalf of the PETHEMA and GELTAMO groups

Correspondence: J-M. Ribera jribera@iconcologia.net Received: Accepted: Early view:

April 19, 2023. July 31, 2023. August 10, 2023.

Department of Hematology, Institut Català d’Oncologia-Hospital Germans Trias i Pujol, Josep

Carreras Research Institute-Badalona, Universitat Autònoma de Barcelona, Barcelona; Department of Hematology, Hospital Universitario de Donostia, Donostia; 3Department of

Hematology, Hospital Universitario 12 de Octubre, Madrid; 4Department of Hematology, Institut Català d’Oncologia-Hospital Joan XXIII, Tarragona; 5Department of Hematology, Hospital Universitario de Gran Canaria Dr. Negrín, Las Palmas de Gran Canaria; 6Department of Hematology, IBSAL, IBMCC, Centro de Investigación del Cáncer, CIBERONC, Universidad de Salamanca-CSIC, Hospital Universitario de Salamanca, Salamanca; 7Department of Hematology, Institut Català d’Oncologia-Hospital Josep Trueta, Girona; 8Department of Hematology, Hospital de la Santa Creu i Sant Pau, Barcelona; 9Department of Hematology, Hospital Universitari Vall d’Hebron, Barcelona; 10Department of Hematology, Hospital Universitari i Politècnic La Fe, Valencia; 11Department of Hematology, Hospital General Universitario Gregorio Marañón, IiSGM, Madrid; 12Department of Hematology, Hospital Universitario Virgen de la Victoria, Málaga; 13Department of Hematology, Hospital Universitario de Fuenlabrada, Madrid; 14Department of Hematology, Hospital Universitario Ramón y Cajal, Madrid; 15Department of Hematology, Hospital Universitari Arnau de Vilanova de Lleida, Lleida; Department of Hematology, Hospital Universitari Mútua de Terrassa, Terrassa; 17Department

of Hematology, Hospital Universitario Reina Sofía, IMIBIC, Córdoba; 18Department of Hematology, Hospital Clínico Universitario de Valencia, Valencia; 19Department of Hematology, Hospital San Pedro de Alcántara, Cáceres; 20Department of Hematology, Hospital Universitario de La Princesa, Madrid; 21Department of Hematology, Hospital Universitario Basurto, Bilbao; Department of Hematology, Hospital Universitario HM Sanchinarro, Madrid; 23Department of

Hematology, Hospital Universitario La Zarzuela, Madrid; 24Department of Hematology, Hospital del Mar, Barcelona; 25Department of Hematology, Hospital Universitari Son Llàtzer, Palma de Mallorca and 26Department of Hematology, Institut Català d’Oncologia-Hospital Duran i Reynals, L’Hospitalet de Llobregat, Spain

Abstract High dose-intensive or infusional intermediate-dose immunochemotherapy is highly effective treatment for Burkitt lymphoma irrespective of human immunodeficiency virus (HIV) infection. However, toxicities of these regimens are relevant, especially in older adults and elderly patients. The prospective multicenter BURKIMAB14 trial included four to six blocks of immunochemotherapy according to stage (localized: 1 and 2 non-bulky; advanced: 2 bulky, 3, 4) and age, with dose reduction in patients >55 years old. Dose-intensity of chemotherapy was reduced in patients ≤55 years old after achieving complete metabolic response (CMR). Their outcomes were compared with those of similar patients included in the former BURKIMAB08 trial, in which there was no dose reduction. CMR was attained in 86 of 107 (80%) patients (17/19 in localized stages and 69/88 in advanced stages). Patients from the BURKIMAB14 trial ≤55 years old showed similar overall survival (OS), fewer Haematologica | 109 February 2024

543

J.M. Ribera et al.

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

infections and cytopenias than patients from the BURKIMAB08 trial. Patients >55 years old had a significantly higher treatment-related mortality despite dose reduction of chemotherapy. With a median follow-up of 3.61 years the 4-year OS probability was 73% (range, 63-81%). Age (≤55 vs. >55 years) and stage (localized vs. advanced) had prognostic significance. No significant differences in OS were observed in HIV-positive versus HIV-negative patients. The results of BURKIMAB14 are similar to those of other dose-intensive immunochemotherapy trials. Age >55 years and advanced stage, but not HIV infection, were associated with poor survival. Dose reduction of chemotherapy in young adults in CMR is safe and does not impact outcomes (clinicaltrials gov. Identifier: NCT05049473).

Introduction Burkitt’s lymphoma or leukemia (BL) is a highly aggressive mature B-cell non-Hodgkin lymphoma (NHL), that represents 1-2% of NHL in adults from western countries.1 BL accounts for nearly 40% of lymphomas that arise in human immunodeficiency virus (HIV)-infected patients and occurs in those with relatively normal CD4 lymphocyte counts. Whereas more than 90% of children and adolescents are cured with high-dose intensive chemotherapy, the cure rate in adults ranges from 75% to 85% in prospective trials using high-dose2-13 or infusional intermediate-dose immunochemotherapy.14,15 In HIV-associated BL the use of combination antiretroviral therapy (cART) has led to improved tolerance to full-dose chemotherapy, which has translated into outcomes similar to those of non-immunosuppressed patients. Despite highly effective front-line therapy, the incidence of treatment failure among adults may be as high as 20-30%, especially in older adults and elderly people.16,17 To date, there are no effective rescue therapies for patients with refractory disease or relapse.18,19 The BURKIMAB08 trial of the Spanish PETHEMA (Programa Español de Tratamientos en Hematología) and GELTAMO (Grupo Español de Linfomas y Trasplantes de Medula Ósea) cooperative groups included 118 evaluable adult patients with BL irrespective of their HIV infection status.8 The intensity of the therapy was adapted to age, with a cutoff point of 55 years (yrs). The response rate was 85% and the 4-year overall survival (OS) was 73%. Toxicity was relevant especially in older as well as in HIV-infected patients. In the BURKIMAB14 trial, the dose of cyclophosphamide, methotrexate and cytarabine was reduced in young adults aged up to 55 yrs after achieving response, whereas no modifications were performed for patients older than 55 yrs. Herein we report the efficacy and toxicity of this trial and compare these with the results observed in the BURKIMAB08 trial in young adults.

Methods Eligibility criteria included age ≥18 yrs, confirmed histologic and immune cytologic or histochemical diagnosis, including cytogenetics and/or MYC gene rearrangement. HIV-infected patients had to be on treatment with or had

to begin combined cART at BL diagnosis. Exclusion criteria included patients with t(8;14), t(2;8) or t(8;22) with additional cytogenetic abnormalities, patients with BCL-2 and/or BCL-6 gene rearrangements, organ failure not due to BL, previous organ transplant and previous radiotherapy or chemotherapy. The study was approved by the Spanish Health Ministry on May 23, 2014 (reference 14155/RG44063). Accrual began in January 2014, and the study was closed for follow-up in June 2022. The study was registered as clinicaltrials gov. Identifier: NCT05049473. The Ann Arbor staging system was used to define stage. The International Prognostic Index (IPI) and the Burkitt lymphoma International Prognostic Index (BL-IPI)20 were used to stratify risk. Table 1 shows the therapy of the BURKIMAB14 trial. Patients in advanced stage (2 with bulky disease or 3-4) ≤55 yrs received two courses of alternating cycles A, B and C, whereas patients >55 yrs received three courses of alternating cycles A and B, for a total of six cycles in both groups. Patients with localized stage (non-bulky stages 1-2) received four cycles of treatment (A, B, C and A for younger adults and A, B, A and B for older patients). A single dose of rituximab was administered before each cycle. Compared with the BURKIMAB08 trial, after achievement of complete response (CR) for patients ≤55 yrs intravenous methotrexate for cycles A, B and C was reduced 33%, iphosphamide 28% in cycle A and cytarabine 25% in cycle C. Central nervous system (CNS) prophylaxis consisted of triple intrathecal therapy (TIT) administered in the pre-phase and twice in each cycle A and B, for a total of nine doses in all patients. Positron-emission tomography-computed tomography (PETCT) scans were performed after cycle 2 in all patients and after cycle 4 for patients in localized stages and after cycle 6 for patients in advanced stages. Early complete metabolic response (CMR) was defined as the disappearance of marrow and extramedullary disease after two cycles. Patients in partial metabolic remission (PMR, defined as >50% reduction of all measurable lesions) after two cycles continued therapy and were re-evaluated at the end of the fourth or the sixth cycle, in patients in early or advanced stages, respectively. The overall CMR rate was calculated as the sum of early CMR and CMR attained at the end of the fourth or sixth cycle, respectively. Treatment failure was defined when a patient did not achieve at least PMR after the first two cycles or CMR after six, respectively. Toxicity was evaluated according to the Common Terminology Cri-

Haematologica | 109 February 2024

544

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

J.M. Ribera et al.

teria for Adverse Events (CTCAE v 6.0) criteria. The primary objective of the study was to assess OS and to compare the OS and toxicity with those observed in patients aged 18 to 55 yrs from the BURKIMAB08 trial. Survival curves were plotted according to the Kaplan-Meier method and were compared by the log-rank test. Median follow-up was calculated for alive patients.

higher frequency of BL, high serum lactate dehydrogenase (LDH) level and higher frequency of poor IPI and BL-IPI risk groups, whereas the cohorts of HIV-positive and HIV-negative patients were comparable for the main clinical and biological parameters at diagnosis. In HIV-positive patients, the CD4 lymphocyte count was < 200/µL in nine of 20 cases (45%) (median 204; range, 0-773/µL), the HIV viral load was detectable in 12 of 22 patients (55%) and BL led to the diagnosis of HIV infection in 11 patients (46%).

Results Patients One hundred eleven patients were registered at 26 Spanish centers, of whom 107 were valid for this study. Reasons for exclusion of the remaining four patients were previous therapy (n=2) and treatment not given according to the age range (n=2). Sixty-eight patients were aged ≤55 yrs and 24 were HIV-positive. Table 2 shows the characteristics of the whole series and compares them according to age and HIV infection. Patients >55 yrs showed a significantly

Response to treatment, survival and toxicity Figure 1 shows the study flow chart. Following two cycles of chemotherapy early CMR and PMR were attained in 15 and four patients with localized stages and in 54 and 23 patients with advanced stages, respectively. At the end of therapy, the overall CMR rates were 17 of 19 (89%) and 69 of 88 (78%), respectively. For the whole series, the overall CMR rate was 80%. No relapses or deaths by toxicity were observed in patients with localized stages, whereas the main events registered in patients in advanced stages

Table 1. BURKIMAB08 and BURKIMAB14 protocol treatmenta. Dose (mg/m2) BURKIMAB14 18-55 yrs

Dose (mg/m2) BURKIMAB08 & BURKIMAB14 >55 yrs

Route (time in hrs)

Cycle Day

Drug

Dose (mg/m2) BURKIMAB08 15-55 yrs

Pre-phase 1-5 1-5

Cyclophosphamide Prednisone

200 60

IV (1) IV bolus

Cycle A 7 8 8 8-12 8-12 11-12 11-12

Rituximab Vincristine Methotrexate Iphosphamide Dexamethasone Etoposide Cytarabine

375 2 (absolute) 1,500b 800 10 100 150

375 2 (absolute) 1,000b 500 10 100 150

375 500b 400 10 60 60

IV (4) IV bolus IV (24)b IV (1) IV bolus IV (1) IV (1 every 12)

Rituximab Vincristine Methotrexate Cyclophosphamide Dexamethasone Doxorubicin

375 2 (absolute) 1,500b 200 10 25

375 2 (absolute) 1,000b 200 10 25

375 1 (absolute) 500b 200 10 25

IV (4) IV bolus IV (24)b IV (1) IV bolus IV (15 min)

Rituximab Vindesine Methotrexate Dexamethasone Etoposide Cytarabine

375 3 (max. 5) 1,500b 10 250 2,000

375 3 (max. 5) 1,000b 10 250 1,500

IV (4) IV bolus IV (24)b IV bolus IV (1) IV (3 every 12)

Methotrexate Cytarabine Dexamethasone

15 40 20

15 -

IT IT IT

Cycle B 28 29 29 29-33 29-33 32-33

Cycle Cc 49 50 50 50-54 53-54 54

CNS prophylaxis 1, 8, 12, 29, 33

Patients with localized stage (non-bulky stages 1-2I) received 4 cycles of treatment (A, B, C and A for younger adults and A, B, A and B for older patients). bFolinic acid rescue from 12 hours after the end of infusion. cOmitted for patients >55 years (yrs) old. IV: intravenous; IT: intrathecal; hrs: hours; min: minutes; CNS: central nervous system.

Haematologica | 109 February 2024

545

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

J.M. Ribera et al.

were resistance (n=2), death by toxicity (n=13), withdrawal for toxicity (n=5) and relapse or progression during therapy (n=8, 4 patients each). Sixty-three of 69 patients who completed the therapy remain alive in CMR. The main events in patients off therapy were relapse (n=4) and non-relapse mortality (n=2, neuroblastoma and infection, 1 patient each). With a median follow-up of 3.61 (range, 0.1-7.42) yrs the 4-year disease-free survival was 86% (95% confidence interval [CI]: 76-92) and the 4-year OS was 73% (95% CI: 63-81) (Figure 2A, B). No differences in OS were observed in patients who attained CMR (n=69) versus those in PMR (n=17) after two cycles of therapy (4-year OS: 87% [95% CI: 76-93] vs. 88% [95% CI: 59-97]). The death rate by toxicity was significantly higher in patients with advanced age (3/68 patients ≤55 yrs vs. 11/39 patients >55 yrs; P<0.001). In patients >55 yrs, death by toxicity was most frequent in the early phase of the therapy (pre-phase and cycle A, 8/11 deaths). The overall causes of death (n=28) were relapse/resistance (n=7 and n=5, respectively), infection (n=13), tumor lysis syndrome (n=1), hemorrhage (n=1) and second cancer (n=1). Both the toxic deaths in general and

the deaths by infection were significantly higher in older versus younger patients (4/68 vs. 11/38; P=0.001 for toxic deaths and 3/68 vs. 10/38; P=0.002 for infections). On the contrary, the deaths by disease progression were not significantly different according to age (7/68 vs. 5/38; P=0.752). The deaths by infections were due to bacteria (E. coli n=2, Enterobacter cloacae n=3, Klebsiella pneumoniae n=1, Pseudomonas aeruginosa n=1, Serratia marcesens n=1) followed by fungal infections (Candida spp n=2, Aspergillus spp n=1). There was one death due to syncytial respiratory virus and another by COVID-19 infection. Relapses (n=8) were localized in the CNS (n=4) and in bone marrow combined with lymph nodes (n=3) or CNS (n=1). The median time to CNS relapses was 4 (range, 2-9) months. Age (≤55 yrs vs. >55 yrs), stage (localized vs. advanced), IPI (low-low intermediate vs. intermediate-high) and BL-IPI (low vs. intermediate vs. high) had prognostic significance in this series (Figure 3A-D). No significant differences in OS were observed in HIV-positive versus HIV-negative patients (62% [95% CI: 38-79] vs. 76% [95% CI: 65-84], respectively) or in patients with Burkitt leukemia versus those with BL in stages 3 and 4 (62% [95%

Table 2. Patient characteristics in the series. Patients Whole series ≤55 yrs old N=107 N=68 Sex: male, N (%)

Patients >55 yrs old N=39

HIV-infected N=24

Non-HIVinfected N=83

82 (77)

55 (81)

27 (69)

0.170

20 (83)

62 (75)

0.379

51 (18, 80)

46 (30, 74)

52 (18, 80)

0.258

Age in yrs, N (%) ≤55 >55

68 (64) 39 (36)

17 (71) 7 (29)

61 (61) 32 (39)

0.400

Diagnosis, N (%) Burkitt’s leukemia Burkitt Lymphoma

38 (35) 69 (65)

19 (28) 49 (72)

19 (49) 20 (51)

0.031

5 (21) 19 (79)

33 (40) 50 (60)

0.088

Ann Arbor stage, N (%) 1-2 3-4

29 (27) 78 (73)

21 (31) 47 (69)

8 (20) 31 (80)

0.245

5 (21) 19 (79)

24 (29) 59 (71)

0.433

HIV-infected, N (%)

24 (22)

17 (25)

7 (18)

0.400

ECOG <2, N (%)

71/105 (68)

48/66 (73)

23 (59)

0.146

15 (65)

56/82 (68)

0.781

≥2 extranodal involvements, N (%)

54 (51)

34 (50)

20 (51)

0.898

16 (67)

38 (46)

0.072

CNS involvement, N (%)

19 (18)

10 (15)

9 (23)

0.275

5 (21)

14 (17)

0.762

Bulky mass (>10 cm), N (%)

31/102 (30)

24/65 (37)

7/37 (19)

0.057

5 (21)

26/78 (33)

0.244

High LDH, N (%)

80/105 (76)

46/66 (70)

34 (87)

0.042

19/23 (83)

61/82 (74)

0.413

IPI, N (%) Low / Low-intermediate Intermediate / Intermediate-high

44/106 (41) 62/106 (59)

35/67 (52) 32/67 (48)

9 (23) 30 (77)

0.003

7 (29) 17 (71)

37/82 (45) 45/82 (55)

0.163

BL-IPI, N (%) Low Intermediate High

16/105 (15) 31/105 (30) 58/105 (55)

16/66 (24) 21/66 (32) 29/66 (44)

0 10 (26) 29 (74)

0.001

3/23 (13) 8/23 (35) 12/23 (52)

13/82 (16) 23/82 (28) 46/82 (56)

0.811

Age in yrs, median (min, max)

HIV: human immunodeficiency virus; ECOG: Eastern Cooperative Oncology Group; CNS: central nervous system; LDH: lactate dehydrogenase; IPI: International Prognostic Index; min: minimum; max: maximum; BL-IPI: Burkitt Lymphoma International Prognostic Index; yrs: years.

Haematologica | 109 February 2024

546

J.M. Ribera et al.

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

CI: 44-75] vs. 72% [95% CI: 57-83], respectively). Table 3 shows the main grade ≥3 toxic events according to age and stage of BL. Neutropenia and thrombocytopenia were the most frequent events, with a significantly longer duration in patients >55 yrs in advanced stages. Hepatic and renal toxicity were also higher in the latter group of patients. Outcomes and toxicity of the BURKIMAB14 versus BURKIMAB08 trials in patients aged 18-55 years Online Supplementary Table S1 shows the comparison of the main clinical and biologic characteristics of patients included in both protocols. Compared with the BURKIMAB08 trial patients from the BURKIMAB14 trial showed a lower frequency of HIV infection and a higher frequency of patients belonging to the low-risk category. Table 4 compares the main outcomes of these patients. No significant differences were found for any outcome measure. However, there were fewer deaths in CMR in the BURKIMAB14 trial (3/60 [5%] vs. 8/74 [11%]). The 4-year OS probabilities in the BURKIMAB 08 and BURKIMAB14 trails were 72% (95% CI: 61-80) and 82% (95% CI: 70-90), respectively (Figure 4). Online Supplementary Table S2 compares the main toxic events in both trials. The duration of thrombocytopenia as well as the number of grade>2 infections and the deaths by infection were significantly lower in the BURKIMAB14 trial.

Discussion This study shows that the results of BURKIMAB14 are similar to those of other dose-intensive immunochemotherapy trials. Age >55 years and advanced stage were associated with poor survival. In contrast, HIV infection did not show different outcome. It is of note that the reduction of chemotherapy in young adults in CMR did not negatively impact outcome and was associated with a reduced frequency and mortality by infections. Prospective studies on the treatment of BL used at least six different regimens (LMB, GMALL, CALGB, R-CODOX-M/RIVAC, DA-EPOCH-R, HD-MTX-CHP [BASIC]).2-15 Treatment schedules using high-dose intensive immunochemotherapy (R-CODOX-M/R-IVAC being the most frequently used) show considerable differences in the drugs and dosages used, age of patients included, HIV infection status, risk definition, days of hospitalization and follow-up. The response rate and the OS probability in the different trials range from 80-90% and 70-95%, respectively. Only one study randomly compared R-CODOX-M/R-IVAC versus DA-EPOCH-R in newly diagnosed patients with high-risk BL, and reported a similar CMR (79% vs. 73%) and 2-year OS probability (75% vs. 76%) but a lower frequency of infections/ febrile neutropenia, platelet transfusions, and fewer days of hospitalization were observed in patients treated with

Figure 1. Flowchart of the study. This figure summarizes the distribution of patients according to their stage and the metabolic response. Most patients were diagnosed in advanced-stage disease and showed early complete metabolic response (CMR). PMR: partial metabolic remission; NMR: no metabolic response. Haematologica | 109 February 2024

547

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

J.M. Ribera et al.

DA-EPOCH-R.21 However, this trial was prematurely closed due to low accrual rate when only 89 out of planned 260 patients were included. On the other hand, a retrospective study of 641 adult patients treated in 30 centers from US showed a non-significant advantage of the CODOX-M/RIVAC regimen over DA-EPOCH-R or hyperCVAD/MA regimens in terms of progression-free survival and OS.16 The BURKIMAB14 trial showed CMR rates of 80% and a 4-year OS probability of 73%. Early versus late attainment of CMR did not impact the OS. The OS probability was influenced by age and stage but not by HIV infection status. Despite a significant dose reduction for patients >55 yrs old, a high mortality rate was observed, being concentrated in the pre-phase and the first cycle of chemotherapy. This could be explained by the poorer BL characteristics in patients >55 yrs old and by the lack of delay between the pre-phase and the first block of chemotherapy, a feature inducing an accumulation of toxicity. Dose reduction of drugs included in the first block A for older patients (e.g., eliminating iphosphamide [as cyclophosphamide has been given in the prephase] and also eliminating the doses of cytarabine and etoposide) could reduce the death rate, mainly due to infections in the neutropenic period, and simplify the implementation of the trial. The BURKIMAB14 trial aimed to evaluate the efficacy and tolerability of dose reduction of chemotherapy in young patients in CMR, in order to reduce toxicity and mortality compared with those of the BURKIMAB08 trial without hampering the efficacy. In the BURKIMAB14 trial, the mortality of CMR patients (5% vs. 11%), the frequency of grade >2 infections and the duration of thrombocytopenia

were reduced, and the OS showed a non-significant increase (82% vs. 72%). This difference could be explained by the reduction of mortality as well as by better disease characteristics of young patients in the BURKIMAB14 trial. These differences in patients’ inclusion in both trials could be due in part by chance and in part by unknown biases from physicians after having acquired experience with the BURKIMAB08 trial. Overall, these data suggest that a dose reduction is feasible and safe for young BL patients after CMR achievement. CNS relapse is a matter of concern in this and other BL trials, although was not apparently related with the reduction of chemotherapy in older patients from our trial. The possible inclusion of cranial irradiation could increase the toxicity and could influence the timely delivery of the chemotherapy. Better efforts are needed to provide adequate protection of this sanctuary, as alternative approached such as DA-EPOCH-R do not protect adequately from CNS relapse. This study has some limitations; first, the selection of the chemotherapy according only to stage and age, without considering other risk factors such as lactate dehydrogenase levels, CNS involvement or performance status, which are variables included in BL-IPI; second, the lack of centralized review of the PET-CT scans. The value of complete metabolic response by PET-TC imaging in Burkitt lymphoma not involving the bone marrow is uncertain. However, this is currently the only available tool for response evaluation. Data on the value of circulating tumor DNA could represent a promising tool for residual disease assessment but solid data are not available for BL; third, the lack of review of pathologic samples, although stringent criteria for BL

Figure 2. Disease-free survival and overall survival in the whole series. Panel (A) shows the disease-free survival of 86 patients in complete response (CR), the probability of disease-free survival at 4 years was 86% (95% confidence interval [CI]: 76-92). Panel (B) shows the overall survival of 107 patients, the probability of overall survival at 4 years was 73% (95% CI: 63-81). Haematologica | 109 February 2024

548

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

J.M. Ribera et al.

diagnosis were used in this trial. Additional limitations include the lack of sample size calculation (the comparison was established when similar number of patients were recruited in BURKIMAB 08 and BURKIMAB14 trials) and the

lack of multivariable analysis of prognostic factors, as we validated the well-established BL-IPI as a prognostic index for BL in our series. The immunochemotherapy results of BL should be improved

Figure 3. Overall survival according to age, stage, International Prognostic Index and the Burkitt lymphoma International Prognostic Index. Panel (A) shows a significantly poorer survival of patients aged >55 years. Panel (B) shows the prognostic influence of stage at diagnosis, with poorer survival of patients in advanced stages. Panel (C) shows the prognostic significance of the International Prognostic Index (IPI) with poorer outcome of patients with intermediate-high and high IPI score. Panel (D) shows the reproducibility of the three groups of the Burkitt lymphoma IPI in this series of patients. interm: intermediate. Haematologica | 109 February 2024

549

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

J.M. Ribera et al.

in older patients and in those who relapse. In our series, patients >55 years old showed a significantly higher frequency of Burkitt leukemia and a higher frequency of poor IPI and BL-IPI risk groups. These features, together with the high mortality rate, explain their poor prognosis. DA-EPOCH-R is best tolerated in patients with advanced age and should probably be the treatment of choice.1 The frequent CNS relapse also points out the need for improvement of CNS prophylaxis.22 To date, no rescue chemotherapies have demonstrated efficacy, and allogeneic or autologous stem cell transplantation do not seem to be effective, with a survival rate less than 20%. Patients with relapsed or refractory disease should be included in clinical trials.

Therapies based on the improved knowledge of the genetic background of BL23 including those targeting the PI3Kδ, AKT, mTOR complex 1 (mTORC1) or CDK6 inhibitors, among others, are currently being evaluated. Immunotherapeutic approaches such as bispecific anti CD20-CD3 monoclonal antibodies or chimeric antigen receptor T-cell therapy24,25 are also under investigation. Disclosures JMR receives consultancy fees, research funding and speaker’s bureau fees from Pfizer and Amgen; receives consultancy and speakers bureau fees from Ariad and Novartis. PA receives honoraria from Janssen, Celgene, AbbVie, AstraZeneca, Gilead

Figure 4. Overall survival of patients aged 18-55 years in the BURKIMAB14 and BURKIMAB08 trials. In the BURKIMAB14 trial the 4-year overall survival probability was 82% (95% confidence interval [CI]: 70-90) versus 72% (95% CI: 61-80) for patients included in the BURKIMAB08 trial.

Table 3. Grade >2 toxicity according to age and stage in patients from the BURKIMAB14 trial. Patients ≤55 yrs old Localized stage N=12

Advanced stage N=56

Localized stage N=7

Advanced stage N=32

Neutropenia, N/N (%) Median days (min, max)

11/12 (92) 4 (2, 7)

53/54 (98) 7 (1, 21)1

6/7 (86) 7 (3, 13)

30/31 (97) 10.5 (2, 24)1

Thrombocytopenia, N/N (%) Median days (min, max)

7/12 (58) 2 (1, 5)

41/54 (76) 3 (1, 31)2

4/7 (57) 4 (2, 16)

25/31 (81) 11 (1, 28)2

Hepatic toxicity, N (%)

1 (8)

4 (7)3

6 (19)3

Renal toxicity, N (%)

3 (5)4

9 (28)4

4 (33)

25 (45)

4 (57)

19 (59)

2 (4)

1 (14)

2 (6)

3 (25)

42 (75)

3 (43)

28 (88)

2 (4)5

10 (31)5

Mucositis, N (%) Neurologic toxicity, N (%) Infections, N (%) Death by infection, N (%) 1

Patients >55 yrs old

P=0.227; 2P=0.004; 3P=0.160; 4P=0.007; 5P=0.001. yrs: years; min: minimum; max: maximum.

Haematologica | 109 February 2024

550

J.M. Ribera et al.

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

Table 4. Main outcomes of patients from the BURKIMAB08 and BURKIMAB14 trials in patients aged 18-55 years. BURKIMAB08 N=87

BURKIMAB14 N=68

Early death, N (%)

6 (7)

1 (2)

0.136

Failure, N (%)

6 (7)

4 (6)

1.000

CMR, N (%)

74 (85)

60 (88)

0.566

Relapse, N/N (%)

5/74 (7)

5/60 (8)

0.752

Death in CMR, N/N (%)

8/74 (11)

3/60 (5)

0.344

4-year OS, % (95% CI)

72 (61-80)

82 (70-90)

0.160

4-year DFS, % (95% CI)

83 (73-90)

87 (74-93)

0.495

The final response was not evaluated (due to withdrawal without response assessment or partial response on treatment or death in partial response) in 1 and 3 patients in the BURKIMAB08 and BURKIMAB14 trials, respectively. CMR: complete metabolic response; OS: overall survival; DFS: disease-free survival; CI: confidence interval.

and Incyte; has a consulting/advisory role at Janssen, Celgene, AbbVie, AstraZeneca; is part of the speakers’ bureau of Janssen, Celgene, AbbVie, AstraZeneca and Gilead. MB receives research funding from Roche and Kite/Gilead; receives honoraria from Roche, Kite/Gilead, Novartis, Janssen, Incyte and BMS/Celgene; is on the advisory board of Roche, Kite/Gilead, Novartis, Janssen, Incyte and BMS/Celgene. MJT has as consultancy (advisory) at Roche, Gilead, Janssen, Takeda, Abbvie and Lilly; discloses educational activities for Roche, Gilead, Janssen, Takeda, Abbvie and Lilly. EG serves as consultant and/or on the speaker’s bureau for/of AbbVie, Janssen-Cilag and AstraZeneca; has received travel grants from AbbVie, Janssen-Cilag and AstraZeneca. SM receives research funding from Roche and honoraria from Roche, Gilead, Janssen and Servier. JMS receives honoraria for a consultancy or advisory role from Roche, Janssen, Gilead-Kite, Novartis, Celgene-BMS, Incyte, Beigene, Miltenyi Biomedicine, Celltrion and Kern-Pharma; receives honoraria as a speaker in educational events from Roche, Janssen, Gilead-Kite, Novartis, Celgene-BMS, Incyte, Takeda and Kern-Pharma. The remaining authors have no conflicts of interest to disclose.

Contributions JMR and JMS designed the trial and contributed to the analysis. JMR wrote the paper. MM and OGC collected the data, created the database and performed the statistical analysis. MS, BB, MC, HL, JMHR, MS, IGC, PA, PM, MB, MPQL, PB, AT, PH, AGG, FVLJS, MJT, JMB, AGN, CB, LL, DGB, EG, AC and SM qualify for authorship according to the WAME criteria, and were involved in patient inclusion, data collection and data acquisition. All authors approved the final version of the manuscript. Funding The research was supported in part by the grant 2021 SGR 00771, AGAUR, Generalitat de Catalunya, Spain. Data-sharing statement For original data, please contact the corresponding author or mmorgades@iconcologia.net. De-identified individual participant data are available indefinitely at www.pethema.org. The study protocol is also available at the same website.

References 1. Roschewski M, Staudt LM, Wilson WH. Burkitt’s lymphoma. N Engl J Med. 2022;387(12):1111-1122. 2. Ribrag V, Koscielny S, Bosq J, et al. Rituximab and dose-dense chemotherapy for adults with Burkitt’s lymphoma: a randomised, controlled, open-label, phase 3 trial. Lancet. 2016;387(10036):2402-2411. 3. Mead GM, Barrans SL, Qian W, et al. A prospective clinicopathologic study of dose-modified CODOX-M/IVAC in patients with sporadic Burkitt lymphoma defined using cytogenetic and immunophenotypic criteria (MRC/NCRI LY10 trial). Blood. 2008;112(6):2248-2260. 4. Corazzelli G, Frigeri F, Russo F, et al. RD-CODOX-M/IVAC with rituximab and intrathecal liposomal cytarabine in adult Burkitt

lymphoma and ‘unclassifiable’ highly aggressive B-cell lymphoma. Br J Haematol. 2012;156(2):234-244. 5. Evens AM, Carson KR, Kolesar J, et al. A multicenter phase II study incorporating high-dose rituximab and liposomal doxorubicin into the CODOX-M/IVAC regimen for untreated Burkitt’s lymphoma. Ann Oncol. 2013; 24(12):3076-3081. 6. Noy A, Lee JY, Cesarman E, et al. AMC 048: modified CODOX-M/ IVAC-rituximab is safe and effective for HIV-associated Burkitt lymphoma. Blood. 2015;126(2):160-166. 7. Intermesoli T, Rambaldi A, Rossi G, et al. High cure rates in Burkitt lymphoma and leukemia: a Northern Italy Leukemia Group study of the German short intensive rituximabchemotherapy program. Haematologica. 2013;98(11):1718-1725.

Haematologica | 109 February 2024

551

ARTICLE - Dose reduction of therapy in Burkitt lymphoma

J.M. Ribera et al.

8. Ribera JM, García O, Grande C, et al. Dose-intensive chemotherapy including rituximab in Burkitt’s leukemia or lymphoma regardless of human immunodeficiency virus infection status: final results of a phase 2 study (Burkimab). Cancer. 2013;119(9):1660-1668. 9. Hoelzer D, Walewski J, Döhner H, et al. Improved outcome of adult Burkitt lymphoma/leukemia with rituximab and chemotherapy: report of a large prospective multicenter trial. Blood. 2014;124(26):3870-3879. 10. Rizzieri DA, Johnson JL, Byrd JC, et al. Improved efficacy using rituximab and brief duration, high intensity chemotherapy with filgrastim support for Burkitt or aggressive lymphomas: cancer and Leukemia Group B study 10002. Br J Haematol. 2014;165(1):102-111. 11. Kasamon YL, Brodsky RA, Borowitz MJ, et al. Brief intensive therapy for older adults with newly diagnosed Burkitt or atypical Burkitt lymphoma/leukemia. Leuk Lymphoma. 2013;54(3):483-490. 12. Kujawski LA, Longo WL, Williams EC, et al. A 5-drug regimen maximizing the dose of cyclophosphamide is effective therapy for adult Burkitt or Burkitt-like lymphomas. Cancer Invest. 2007;25(2):87-93. 13. Ferreri AJM, Angelillo P, Erbella F, et al. Safety and efficacy of a dose-dense short-term therapy in patients with MYC-translocated aggressive lymphoma. Blood Adv. 2022;6(22):5811-5820. 14. Dunleavy K, Pittaluga S, Shovlin M, et al. Low-intensity therapy in adults with Burkitt’s lymphoma. N Engl J Med. 2013;369(20):1915-1925. 15. Roschewski M, Dunleavy K, Abramson JS, et al. Multicenter study of risk-adapted therapy with dose-adjusted EPOCH-R in adults with untreated Burkitt lymphoma. J Clin Oncol. 2020;38(22):2519-2529. 16. Evens AM, Danilov A, Jagadeesh D, et al. Burkitt lymphoma in the modern era: real-world outcomes and prognostication

across 30 US cancer centers. Blood. 2021;137(3):374-386. 17. Crombie J, LaCasce A. The treatment of Burkitt lymphoma in adults. Blood. 2021;137(6):743-750. 18. Woessmann W, Zimmermann M, Meinhardt A, et al. Progressive or relapsed Burkitt lymphoma or leukemia in children and adolescents after BFM-type first-line therapy. Blood. 2020;135(14):1124-1132. 19. Cremer M, Schwarzbich MA, Schöning T, Lisenko K, Ho AD, Witzens-Harig M. In Burkitt lymphoma patients who relapse after induction with a short-intensive chemoimmunotherapy protocol, aggressive salvage chemotherapy therapy is ineffective: a single-center retrospective study. Ann Hematol. 2017;96(9):1573-1575. 20. Olszewski AJ, Jakobsen LH, Collins GP, et al. Burkitt Lymphoma International Prognostic Index. J Clin Oncol. 2021;39(10):11291138. 21. Chamuleau M, Stenner F, Chityu D, et al. R-CODOX-M/R-IVAC versus dose-adjusted (DA)-EPOCH-R in patients with newly diagnosed high-risk Burkitt lymphoma: first results of a multi-center randomized HOVON/SAKK trial. Hemasphere. 2022;6:S3:LB2370. 22. Zayac AS, Evens AM, Danilov A, et al. Outcomes of Burkitt lymphoma with central nervous system involvement: evidence from a large multicenter cohort study. Haematologica. 2021;106(7):1932-1942. 23. Thomas N, Dreval K, Gerhard DS, et al. Genetic subgroups inform on pathobiology in adult and pediatric Burkitt lymphoma. Blood. 2023;141(8):904-916. 24. Liu Y, Deng B, Hu B, et al. Sequential different B-cell antigentargeted CAR T-cell therapy for pediatric refractory/relapsed Burkitt lymphoma. Blood Adv. 2022;6(3):717-730. 25. Zhang W, Yang J, Zhou C, et al. Early response observed in pediatric patients with relapsed/refractory Burkitt lymphoma treated with chimeric antigen receptor T cells. Blood. 2020;135(26):2425-2427.

Haematologica | 109 February 2024

552

ARTICLE - Non-Hodgkin Lymphoma

Tafasitamab for patients with relapsed or refractory diffuse large B-cell lymphoma: final 5-year efficacy and safety findings in the phase II L-MIND study Johannes Duell,1 Pau Abrisqueta,2 Marc Andre,3 Gianluca Gaidano,4 Eva Gonzales-Barca,5 Wojciech Jurczak,6 Nagesh Kalakonda,7 Anna Marina Liberati,8 Kami J. Maddocks,9 Tobias Menne,10 Zsolt Nagy,11 Olivier Tournilhac,12 Christian Kuffer,13 Abhishek Bakuli,13 Aasim Amin,13 Konstantin Gurbanov13 and Gilles Salles14 Medizinische Klinik und Poliklinik II, Universitätsklinik Würzburg, Würzburg, Germany;

Correspondence: J. Duell Duell_J@ukw.de Received: Accepted: Early view:

May 12, 2023. August 23, 2023. August 31, 2023.

Department of Hematology, Vall d’Hebron Institute of Oncology (VHIO), Vall d’Hebron

https://doi.org/10.3324/haematol.2023.283480

University Hospital, Barcelona, Spain; 3Centre Hospitalier Universitaire CHU UCL Namur,

Published under a CC BY license

Namur, Belgium; 4Division of Hematology, Department of Translational Medicine University of Eastern Piedmont and Ospedale Maggiore della Carità, Novara, Italy; 5Department of Hematology, Institut Català d’Oncologia, Hospitalet de Llobregat, IDIBELL, Universitat de Barcelona, Barcelona, Spain; 6Department of Clinical Oncology, Maria Sklodowska-Curie National Research Institute of Oncology, Kraków, Poland; 7Department of Molecular and Clinical Cancer, University of Liverpool, Liverpool, UK; 8Università degli Studi di Perugia, Azienda Ospedaliera Santa Maria di Terni, Terni, Italy; 9Department of Internal Medicine, Arthur G James Comprehensive Cancer Center, Ohio State University Wexner Medical Center, Columbus, OH, USA; 10Freeman Hospital, The Newcastle upon Tyne Hospitals, Newcastle upon Tyne, UK; 11Semmelweis University, Budapest, Hungary; 12Centre Hospitalier Universitaire de Clermont-Ferrand, Lyon, France; 13MorphoSys AG, Planegg, Germany and 14Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA

Abstract Tafasitamab, an anti-CD19 immunotherapy, is used with lenalidomide for patients with autologous stem cell transplant-ineligible relapsed/refractory diffuse large B-cell lymphoma based on the results of the phase II L-MIND study (NCT02399085). We report the final 5-year analysis of this study. Eighty patients ≥18 years who had received one to three prior systemic therapies, and had Eastern Cooperative Oncology Group performance status 0-2 received up to 12 cycles of co-administered tafasitamab and lenalidomide, followed by tafasitamab monotherapy until disease progression or unacceptable toxicity. The primary endpoint was the best objective response rate. Secondary endpoints included duration of response, progression-free survival, overall survival, and safety. Exploratory analyses evaluated efficacy endpoints by prior lines of therapy. At data cutoff on November 14, 2022, the objective response rate was 57.5%, with a complete response rate of 41.3% (n=33), which was consistent with prior analyses. With a median follow-up of 44.0 months, the median duration of response was not reached. The median progression-free survival was 11.6 months (95% confidence interval [95% CI]: 5.7-45.7) with a median follow-up of 45.6 months. The median overall survival was 33.5 months (95% CI: 18.3-not reached) with a median follow-up of 65.6 months. Patients who had received one prior line of therapy (n=40) showed a higher objective response rate (67.5%; 52.5% complete responses) compared to patients who had received two or more prior lines of therapy (n=40; 47.5%; 30% complete responses), but the median duration of response was not reached in either subgroup. Other exploratory analyses revealed consistent long-term efficacy results across subgroups. Adverse events were consistent with those described in previous reports, were manageable, and their frequency decreased during tafasitamab monotherapy, with no new safety concerns. This final 5-year analysis of L-MIND demonstrates that the immunotherapy combination of tafasitamab and lenalidomide is well tolerated and has long-term clinical benefit with durable responses.

Haematologica | 109 February 2024

553

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

Introduction Diffuse large B-cell lymphoma (DLBCL) is the most common subtype of non-Hodgkin lymphoma.1 First-line standard-of-care immunotherapy for newly diagnosed DLBCL consists primarily of rituximab, an anti-CD20 antibody, with cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) or variations thereof, which may be curative in up to 60-70% of patients.2,3 However, 30-40% of patients experience relapsed or refractory (R/R) disease after firstline R-CHOP.4,5 Second-line options depend on patients’ responses to first-line therapy; those with refractory disease or who experience early relapse (≤12 months) can receive chimeric antigen receptor T-cell (CAR-T) therapy, with event-free survival in 40-62% of patients.4,5 Patients relapsing after more than 1 year are potential candidates for high-dose chemotherapy and autologous stem cell transplant (ASCT).6 Many patients, however, are ineligible for intensive treatment due to advanced age and/or comorbidities,3,7 and about 50% do not proceed to ASCT because of failure of salvage therapy. A further 40-65% relapse following ASCT.6,8,9 Tafasitamab is a CD19-targeted immunotherapy that elicits direct cytotoxicity, antibody-dependent cellular cytotoxicity, and antibody-dependent cellular phagocytosis via Fc domain interactions.10 The primary analysis of the phase II L-MIND study (NCT02399085) showed that the combination of tafasitamab and lenalidomide resulted in an objective response rate (ORR) of 60%, a complete response (CR) rate of 43%, and a median duration of response of 21.7 months in patients with R/R DLBCL ineligible for ASCT.11 Based on these data, tafasitamab was approved in combination with lenalidomide followed by tafasitamab monotherapy under accelerated approval in the USA (July 2020) and received conditional marketing authorization in Europe (August 2021) for the treatment of adult patients with R/R DLBCL (detailed in the USA label as not otherwise specified, including DLBCL arising from low-grade lymphoma) who are ineligible for ASCT, and is now a standard second-line therapy in this setting.12,13 Durable responses were seen after approximately 3 years of follow-up of the L-MIND study, with an ORR of 57.5%, a CR rate of 40%, and median duration of response of 43.9 months.14 We now present the final 5-year efficacy and safety outcomes from the L-MIND study.

Methods L-MIND was an open-label, single-arm, global, multicenter, phase II study (NCT02399085).11 The study was approved by institutional review boards at each study site and conducted in accordance with the International Council for Harmonization Good Clinical Practice guidelines and the Declaration of Helsinki. All patients provided written informed consent to participation in the study.

J. Duell et al.

Study design and patients Full details of the L-MIND study methods and patients’ eligibility criteria have been described previously.11,14 Briefly, patients were aged ≥18 years with ASCT-ineligible R/R DLBCL, had received one to three prior systemic therapies (including ≥1 targeting CD20) and had an Eastern Cooperative Oncology Group performance status of 0-2. Patients with primary refractory disease were excluded, but because the definition changed while the study was active, some were eligible and included. At study set-up, primary refractory disease was defined as no response to, or progression/ relapse within 3 months of a previous anti-CD20-containing regimen. In a protocol amendment in June 2016, this definition was updated to within 6 months of a previous anti-CD20-containing regimen. Therefore, patients who relapsed within the first 3-6 months after completing prior therapy (and had primary refractory disease according to the updated definition) were initially eligible to enroll in the study. Patients received tafasitamab and lenalidomide for up to 12 cycles (28 days each), followed by tafasitamab monotherapy (once every 2 weeks) in patients with stable disease or better, until progressive disease. Tafasitamab (12 mg/kg intravenously) was administered according to the label.12,13 Lenalidomide (25 mg orally) was self-administered on days 1-21 of each 28-day cycle. We now present data following 5 years of follow-up from enrollment of the last patient. Study outcomes The primary endpoint was the ORR (CR plus partial response [PR]), assessed by an independent review committee, according to the 2007 International Working Group response criteria for malignant lymphoma.15 Secondary endpoints included duration of response (time from initial CR or PR to first observation of progressive disease), progression-free survival (PFS; time from first dose to progressive disease or death), overall survival (OS; time from first dose to date of death), time to progression, time to next treatment, and incidence and severity of adverse events (AE). Statistical analyses The primary analysis occurred when all patients had completed a minimum of 12 months follow-up (data cutoff: November 30, 2018);11 the 3-year follow-up had a data cutoff date of October 30, 2020.14 The data cutoff for the present, pre-specified 5-year analysis was November 14, 2022. Efficacy outcomes were analyzed in the full analysis set (patients who received ≥1 dose of both tafasitamab and lenalidomide), and safety was assessed in those who received any study medication. The frequency of treatment-emergent adverse events (TEAE) per unit of treatment exposure time was analyzed across three periods; details are provided in the Online Supplementary Methods.

Haematologica | 109 February 2024

554

J. Duell et al.

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

Exploratory subgroup analyses Efficacy outcomes (ORR, PFS, OS, duration of response, and/or duration of CR) were also evaluated in exploratory analyses in subgroups of clinical interest, defined by the number of prior lines of therapy (1 vs. ≥2); time to progression after first-line therapy (<1 vs. ≥1 year, in patients who had received only 1 prior line of therapy); patients’ age (≤70 vs. >70 years); International Prognostic Index (IPI) score at baseline (0-2 vs. 3-5); presence of bulky disease (longest lesion diameter ≥7.5 cm, by central radiological assessment) at screening; cell of origin (germinal center B cell vs. non-germinal center B cell); and natural killer (NK) cell count (<100 cells/µL vs. ≥100 cells/µL peripheral blood). NK cell counts were analyzed at baseline by flow

cytometry; details are provided in the Online Supplementary Methods. We also examined outcomes in patients who ended treatment while they had a CR or PR, in patients who received tafasitamab ≥2 years, in patients with OS >5 years, and according to the best response experienced during the study. Regression analyses were used to explore associations with the likelihood of ORR (CR or PR vs. no response) and duration of OS or PFS after adjusting for important covariates of interest; details are provided in the Online Supplementary Methods. Efficacy outcomes were also analyzed in a subset of patients with centrally confirmed diagnoses of DLBCL, which aligns with the population according to the US prescribing information (USPI population).

Table 1. Baseline characteristics of the patients in the full analysis set and by prior lines of therapy. All patients: full analysis set

1 prior line of therapy

≥2 prior lines of therapy

72.0 (41.0-86.0)

72.0 (53.0-86.0)

70.5 (41.0-82.0)

Age >70 years, N (%)

45 (56.3)

25 (62.5)

20 (50.0)

Sex, N (%) Female Male

37 (46.3) 43 (53.8)

19 (47.5) 21 (52.5)

18 (45.0) 22 (55.0)

Ann Arbor stage, N (%) I-II III-IV

20 (25.0) 60 (75.0)

11 (27.5) 29 (72.5)

9 (22.5) 31 (77.5)

IPI score, N (%) 0-2 3-5

40 (50.0) 40 (50.0)

25 (62.5) 15 (37.5)

15 (37.5) 25 (62.5)

Elevated LDH, N (%) Yes No

44 (55.0) 36 (45.0)

18 (45.0) 22 (55.0)

26 (65.0) 14 (35.0)

Prior lines, N (%) 1 2 3 4

40 (50.0) 34 (42.5) 5 (6.3) 1 (1.3)

Primary refractory*, N (%) Yes No

15 (18.8) 65 (81.3)

6 (15.0) 34 (85.0)

9 (22.5) 31 (77.5)

Refractory to previous line of therapy, N (%) Yes No

35 (43.8) 45 (56.3)

6 (15.0) 34 (85.0)

29 (72.5) 11 (27.5)

Prior ASCT, N (%) Yes No

9 (11.3) 71 (88.8)

2 (5.0) 38 (95.0)

7 (17.5) 33 (82.5)

Cell of origin (by IHC), N (%) GCB Non-GCB Unknown/NE

38 (47.5) 22 (27.5) 20 (25.0)

16 (40.0) 14 (35.0) 10 (25.0)

22 (55.0) 8 (20.0) 10 (25.0)

Characteristics N Age in years, median (range)

*Patients with primary refractory disease had a response lasting 3-6 months after first-line therapy. IPI: International Prognostic Index; LDH: lactate dehydrogenase; ASCT: autologous stem cell transplant; IHC: immunohistochemistry; GCB: germinal center B cell; NE: not evaluated.

Haematologica | 109 February 2024

555

J. Duell et al.

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

Results Patients and treatment The full analysis set comprised 80 patients while 81 were included in the safety analysis set. The patients’ disposition for treatment is shown in Online Supplementary Figure S1. The median age of enrolled patients was 72 years (range, 41-86). Fifty percent of patients in the full analysis set had received one prior line of therapy at study entry; half had been given two or more prior lines of therapy. Fifty percent of patients had an IPI score of 3-5. Nine patients had undergone prior stem-cell transplantation. The baseline characteristics of patients in the full analysis set and according to the number of prior lines of therapy are shown in Table 1 (equivalent data in the USPI population are shown in Online Supplementary Table S1). Fifteen (18.8%) patients had primary refractory disease, with a duration of response to first-line therapy of 3-6 months. The 5-year follow-up period commenced from when the last enrolled patient began screening (maximum screening period 28 days). Some patients had slightly less than 60 months’ follow-up available before data cutoff. Twenty-six patients had more than 59 months of follow-up for survival (23 reached the end of the study, 3 withdrew because of AE) and 21 had more than 60 months (18 reached the end of the study, 3 withdrew because of AE). Among those with long-term follow-up, eight received tafasitamab until the end of the study as per protocol. In total, 27 patients received tafasitamab therapy for 2 or more years (see Safety outcomes for treatment duration).

Efficacy outcomes At this 5-year analysis, the best ORR, assessed by an independent review committee, was 57.5% (46/80; 95% confidence interval [95% CI]: 45.9-68.5), with a CR rate of 41.3% (n=33) and a PR rate of 16.3% (n=13), which was consistent with prior analyses (Table 2). Stable disease as the best response was observed in 16.3% of patients (n=13). Overall, five best responses altered in the 5-year versus 3-year analyses (t2 CR to PR, 3 PR to CR). One response deepened from PR to CR, whereas the other changes resulted from re-adjudications following inter-reader variance and/or change of personnel. After a median follow-up of 44.0 months (95% CI: 29.957.0), the median duration of response was not reached (NR) (Figure 1A); the curve suggests a plateau after approximately 12 months, although the number of patients at risk is limited. The median PFS was 11.6 months (95% CI: 5.7-45.7) with a median follow-up of 45.6 months (95% CI: 22.9-57.6) (Figure 1B), and the median OS was 33.5 months (95% CI:18.3-NR) following a median follow-up of 65.6 months (95% CI: 59.9-70.3) (Figure 1C). After a median follow-up of 32.7 months (95% CI: 24.4-53.6), the median duration of CR was not reached (Figure 1D); the 5-year duration of CR was estimated to be 80.7% (95% CI: 59.1-91.6). Figure 1E shows the impact of response quality on OS; whereas the median OS was not reached in patients with CR or PR (95% CI: 45.5 months-NR) and in patients with a best response of CR (NR-NR), it was 18.6 months (95% CI: 8.6-45.5) in patients with PR as their best response. The median time to response was 2.0 months (range, 1.7-16.8), which coincided with the first evaluation as per

Table 2. Efficacy outcomes in the primary, 3-year and 5-year follow-up analyses in the full analysis set (N=80)

Characteristics

Primary analysis

3-year follow-up

Final 5-year data

5-year data for patients with 1 prior line of therapy, N=40

Data cut-off date

Nov 30, 2018

Oct 30, 2020

Nov 14, 2022

Best ORR, N (%) [95% CI]

48 (60.0) [48.4-70.9]

46 (57.5) [45.9-68.5]

27 (67.5) [50.9-81.4]

19 (47.5) [31.5-63.9]

CR rate, N (%) [95% CI]

34 (42.5) [32.0-54.0]

32 (40.0) [29.2-51.6]

33 (41.3) [30.4-52.8]

21 (52.5) [36.1-68.5]

12 (30.0) [16.6-46.5]

PR rate, N (%) [95% CI]

14 (17.5) [10.0-28.0]

14 (17.5) [9.9-27.6]

13 (16.3) [8.9-26.2]

6 (15.0) [5.7-29.8]

7 (17.5) [7.3-32.8]

Median DoR in months [95% CI]

21.7 [21.7-NR]

43.9 [26.1-NR]

NR [33.8-NR]

NR [9.1-NR]

NR [26.1-NR]

Median PFS in months [95% CI]

12.1 [5.7-NR]

11.6 [6.3-45.7]

11.6 [5.7-45.7]

23.5 [7.4-NR]

7.6 [2.7-45.5]

Median OS in months [95% CI]

NR [18.3-NR]

33.5 [18.3-NR]

NR [24.6-NR]

15.5 [8.6-45.5]

5-year data for patients with ≥2 prior lines of therapy, N=40

ORR: objective response rate; 95% CI: 95% confidence interval; CR: complete response; PR: partial response; DoR: duration of response; NR: not reached; PFS: progression-free survival; OS: overall survival.

Haematologica | 109 February 2024

556

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

J. Duell et al.

Continued on following page.

Haematologica | 109 February 2024

557

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

J. Duell et al.

Continued on following page.

Haematologica | 109 February 2024

558

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

J. Duell et al.

Figure 1. Kaplan-Meier curves of time-to-event endpoints. (A) Duration of response in the overall group and in subgroups divided according to the number of prior lines of therapy. (B) Progression-free survival in the overall group and in subgroups divided according to the number of prior lines of therapy. (C) Overall survival in the overall group and in subgroups divided according to the number of prior lines of therapy. (D) Duration of complete response in the overall group and in subgroups divided according to the number of prior lines of therapy. (E) Overall survival according to best response in patients with a best response of complete response or partial response in the overall group. (F) Overall survival according to natural killer cell count < or ≥100 cells/μL of peripheral blood. DoR: duration of response; mFU: median follow-up; 95% CI: 95% confidence interval; pLoT: prior line(s) of therapy; mDoR: median DoR; NR: not reached; PFS: progression-free survival; mPFS: median PFS; OS: overall survival; mOS: median OS; DoCR: duration of complete response; mDoCR: median DoCR; CR: complete response; PR: partial response; NK: natural killer cell count.

protocol. The median time to CR was 8.1 months (range, 1.7-64.9). Twenty-six patients stopped treatment while their disease was in response, 23 with CR and three with PR (Figure 2). Among them, 19 patients were alive with their disease in response at the end of the study (8 were on treatment until the end of the study, while 11 had previously discontinued tafasitamab), two patients later died from progressive disease (both had previous CR but <12 months of treatment, both remained off therapy for over 1 year after study treatment and thereafter went on to receive another anti-lymphoma therapy before they died), and three died from other causes (2 with PR and 1 with CR as best response). Two patients experienced disease relapse after discontinuation of treatment and were alive at the end of participation in the study. Among the nine patients who had previously undergone stem-cell transplantation, four had a best response of CR. Three were alive at the ~60-month follow-up (all had previously discontinued tafasitamab), the fourth withdrew after less than 2 months because of an AE and died after 31 months. Three of the nine patients had a best response of PR; two of them later had progressive disease and died, the third was alive and on treatment at the end of the study. Two of the nine had a best response of progressive disease; one died and one was lost to follow-up. Outcomes and response assessments for the 26 patients with a follow-up for survival of more than 59 months are shown in Online Supplementary Figure S2A. Twenty-two had CR as their best response, two PR, and one each had stable disease and progressive disease. Efficacy outcomes in the USPI population are shown in Online Supplementary Table S2 and Online Supplementary Figure S3. Exploratory subgroup analyses The median time on study treatment was 11.4 months (range, 0.7-78.0) for patients who had received one prior line of therapy and 6.1 months (range, 0.1-65.9) for those who had received two or more prior lines of therapy in the full analysis set. Patients who had received only one prior line of therapy had a higher ORR (67.5%, with 52.5% CR and 15.0% PR) compared to patients who had received two or more prior lines of therapy (47.5%, with 30.0% CR and 17.5% PR). Similarly, and as expected, the median PFS

and median OS were longer in patients who had received one prior line of therapy compared with those who had received two or more prior lines of therapy (Table 2, Figure 1B, C). However, the median duration of response and duration of CR were not reached in either subgroup, indicating comparable long-term efficacy in patients who received the combination of tafasitamab + lenalidomide as second or later lines of therapy (Table 2, Figure 1A, D). Time-to-response curves for the 27 patients who received tafasitamab treatment for 2 or more years are shown in Online Supplementary Figure S2B, with outcomes and response assessments for this subgroup in Online Supplementary Figure S2C. The best response was a CR in 24 and PR in three patients. The ORR was generally comparable between subgroups of clinical interest (Figure 3), although numerically favorable in all patients with positive prognostic factors, such as lack of bulky disease, lower IPI score, only one prior line of therapy, and late relapse (defined as time to relapse/ progressive disease ≥12 months after first-line therapy, and its influence was investigated only in the subgroup of patients given 1 prior line of therapy). The ORR in the USPI population between subgroups of clinical interest was also favorable in all patients with positive prognostic factors (Online Supplementary Figure S4). Similarly, 5-year rate estimates for duration of response, PFS and OS suggest long-term clinical activity in all subgroups of patients (Online Supplementary Table S3). NK cell count at baseline was significantly related to survival; the median OS was not reached (95% CI: 19.3-NR) in patients with a NK cell count of ≥100 cells/µL, compared with 18.3 months (95% CI: 8.6-45.5) in patients with a NK cell count of <100 cells/µL (Figure 1F). In regression analysis, NK cell count was not significantly associated with the odds of ORR in univariate or multivariate models, but NK cell count ≥100 cells/µL at baseline was significantly associated with both longer PFS and longer OS (Online Supplementary Table S4). Reflecting the small sample size, none of the factors in the regression analysis was significantly associated with ORR. IPI score was significantly associated with PFS and OS in univariate analysis but was excluded from the multivariate model as it is derived from other included factors. For PFS and OS, a few known prognostic factors besides NK cells remained significant in the multivariate models: low lactate dehydrogenase levels were associated

Haematologica | 109 February 2024

559

J. Duell et al.

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

Figure 2. Time under treatment and outcomes in patients who ended treatment with a response (N=26). Per protocol, the first computed tomography or magnetic resonance imaging scan for tumor measurement and disease assessment (local) was on day 1 of cycle 3 (~2 months), and the first disease and disease response assessment (computed tomograpy/positron emission tomography) was on day 28 of cycle 12. COO: cell of origin; Refract.: refractory disease; GCB: germinal center B cell; LastL: disease refractory to last line of therapy (but not primary refractory); miss.: missing; Prim.: primary refractory disease; PET: positron emission tomography.

with longer PFS and younger age was associated with longer OS. Although limited in power, this analysis supports further investigation of potential patient profiles combining specific characteristics in larger studies to determine differential outcomes following tafasitamab therapy. Safety outcomes The median duration of exposure to study treatment (either lenalidomide or tafasitamab) in the safety analysis set was 9.2 months (range, 0.23-78.46). The median duration of exposure to tafasitamab monotherapy (following discontinuation of lenalidomide at any time [n=52]) was 13.9 months [range, 0.23-67.2]) versus median tafasitamab exposures of 4.1 months (range, 0.1-20.8) in the primary analysis11 and 9.2 months (range, 0.2-54.7) at the 3-year analysis.14

An overview of the exposure-adjusted frequency of all TEAE, hematologic and non-hematologic TEAE, and important TEAE of interest for three periods of the study (combination therapy, tafasitamab monotherapy up to 2 years, and tafasitamab monotherapy >2 years) is presented in Figure 4, showing that for all these categories of TEAE, frequencies were less common with tafasitamab monotherapy than with tafasitamab + lenalidomide combination therapy. Most TEAE of special interest during the tafasitamab + lenalidomide combination period were hematologic events; the incidences of infusion-related reactions and grade ≥3 infections and infestations were low. Nine patients experienced at least one secondary primary malignancy: one with grade 1 and two with grade 2 basal cell carcinoma; one with grade 2 Bowen’s disease; one with grade 2 breast

Haematologica | 109 February 2024

560

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

J. Duell et al.

Figure 3. Five-year objective response rate in subgroups of clinical interest. FAS: full analysis set; GCB: germinal center B cell; IPI: International Prognostic Index; LDH: lactate dehydrogenase; NK: natural killer; Prim. Refr.: primary refractory; mths: months; ORR: objective response rate.

cancer; one with grade 3 lung adenocarcinoma; one with grade 3 recurrent marginal zone lymphoma; one with grade 3 prostate cancer and grade 2 squamous cell carcinoma; and one other with grade 2 squamous cell carcinoma. The most common non-hematologic TEAE were diarrhea and peripheral edema during the combination therapy phase, and most TEAE were grade 1/2. TEAE are summarized by incidence and severity in Table 3. Treatment-emergent serious AE were reported in 47 (58.0%) patients, with the most frequent being pneumonia (7 patients, 8.6%), febrile neutropenia (5 patients, 6.2%), neoplasms (4 patients, 4.9%), pulmonary embolism and COVID-19 infections (3 patients, 3.7%, each), bronchitis, lower respiratory tract infection, dyspnea, atrial fibrillation and congestive cardiac failure (2 patients, 2.5%, each). Among these events, COVID-19 infections, dyspnea, and benign neoplasms were newly observed compared with those recorded at the 3-year analysis. The median duration of neutropenia and thrombocytopenia (all grades) was 9 and 13 days, respectively. Infections and infestations had a median duration of 12 days (18 days for infectious pneumonia, 10.5 days for urinary tract infection, 9 days for sepsis, and 20 days for opportunistic infections). A total of 45 patients had died at the time of this analysis,

among whom 32 (39.5%) died due to progressive disease, 12 (14.8%) for reasons unrelated to progression, and one (1.2%) in whom the role of progressive disease was not established. Eight of the 45 patients (9.9%) who died did so while on treatment, among whom five (6.2%) died due to progressive disease, and three (3.7%) for reasons unrelated to progression. Among the 37 patients (45.7%) who died after treatment completion, 27 (33.3%) died due to progressive disease, nine (11.1%) died due to reasons unrelated to progression, and one (1.2%) died with the role of progressive disease not established. Six (7.4%) patients died due to AE, none of which were considered related to the study drugs. Temporary tafasitamab interruptions occurred in 65 patients, of whom 50 patients (76.9%) had 171 interruptions (52.5%) due to TEAE. Twenty-eight patients required dose interruption of lenalidomide from the starting dose of 25 mg during combination therapy, with these interruptions being due to AE in 25 (89.3%) patients. Interruptions due to unacceptable toxicity occurred in one patient for each of the study drugs. A total of 16 (19.8%) and 18 (22.2%) patients discontinued tafasitamab and lenalidomide, respectively, due to AE. AE leading to tafasitamab discontinuations were COVID-19,

Haematologica | 109 February 2024

561

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

J. Duell et al.

Continued on following page.

Haematologica | 109 February 2024

562

J. Duell et al.

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

Figure 4. Exposure-adjusted comparison of the frequencies of treatment-emergent adverse events during the combined tafasitamab + lenolamide treatment period, during tafasitamab monotherapy up to 2 years, and during tafasitamab monotherapy beyond 2 years. (A) All treatment-emergent adverse events (TEAE). (B) Hematologic TEAA. (C) Non-hematologic TEAE (cutoff: ≥10 events in any treatment period. (D) Important TEAE of interest. LEN: lenalidomide.

bronchitis, pneumonia, progressive multifocal leukoencephalopathy, varicella zoster virus infection, pyrexia, sudden death, recurrent marginal zone lymphoma, prostate cancer, cerebrovascular accident, cognitive disorder, pulmonary embolism, respiratory failure, thrombocytopenia, congestive cardiac failure, allergic dermatitis, and deep vein thrombosis.

Table 3. Five-year safety summary: treatment-emergent adverse events occurring in ≥10% of patients or grade 3-5 treatment-emergent adverse events in >1 patient (safety analysis set)

Discussion The final 5-year analysis of the phase II L-MIND study continues to demonstrate clinical benefit from the tafasitamab + lenalidomide combination therapy followed by long-term tafasitamab monotherapy, in patients with R/R DLBCL ineligible for ASCT, in all subgroups of clinical interest. The ORR of 57.5% and other efficacy outcomes are consistent with previous results.14 The median duration of response was not reached after a median follow-up of 44.0 months. Long-term tafasitamab therapy was well tolerated, consistent with the drug’s established safety profile, and no new safety concerns were observed at the 5-year analysis. The incidence of all-grade and grade ≥3 AE decreased as patients transitioned from combination therapy to tafasitamab monotherapy and decreased further in the tafasitamab monotherapy phase from 2 years onwards. The importance of depth of response was apparent in the 5-year probability of OS of 72.7% in patients whose best response was CR, compared with 18.3% in those whose best response was PR. As would be expected, analyses by prior lines of therapy showed better outcomes in patients receiving tafasitamab + lenalidomide as second-line therapy rather than third or later lines of therapy. Nevertheless, the median duration of response was not reached in either subgroup; that is, durable responses were also seen with use in second and later lines of therapy. Other subgroup

All grades, N (%)

Grade ≥3, N (%)

Any TEAE

74 (91.4)

52 (64.2)

Hematologic Neutropenia Anemia Thrombocytopenia Febrile neutropenia Leukopenia

40 (49.4) 30 (37.0) 23 (28.4) 10 (12.3) 10 (12.3)

39 (48.1) 6 (7.4) 13 (16.0) 10 (12.3) 8 (9.9)

Non-hematologic Asthenia Peripheral edema Pyrexia Fatigue Diarrhea Constipation Nausea Vomiting Bronchitis Urinary tract infection Pneumonia Respiratory tract infection Decreased appetite Hypokalemia Cough Dyspnea Back pain Muscle spasms C-reactive protein increased

21 (25.9) 20 (24.7) 19 (23.5) 14 (17.3) 30 (37.0) 15 (18.5) 12 (14.8) 12 (14.8) 13 (16.0) 11 (13.6) 10 (12.3) 9 (11.1) 18 (22.2) 15 (18.5) 24 (29.6) 11 (13.6) 16 (19.8) 12 (14.8) 9 (11.1)

2 (2.5) 0 1 (1.2) 2 (2.5) 1 (1.2) 0 0 0 1 (1.2) 2 (2.4) 8 (9.9) 0 0 5 (6.2) 1 (1.2) 2 (2.5) 3 (3.7) 0 0

TEAE: treatment-emergent adverse event.

analyses indicated that durable remissions can be achieved in patients with a range of poor prognostic factors, albeit at lower rates than in those with favorable prognostic factors

Haematologica | 109 February 2024

563

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

including lower IPI score. Exploratory analysis by NK cell count at baseline showed a survival benefit and better odds of PFS and OS for patients with ≥100 cells/µL peripheral blood compared with <100 cells/µL. Together with the suggestion of a plateau in the Kaplan-Meier curves for duration of response, PFS and OS after approximately 12-18 months, these results are consistent with an immunotherapeutic mode of action for tafasitamab + lenalidomide. Outcomes in the subgroup of patients who ended treatment while in response build further on these findings, and suggest that this immunotherapy may have curative potential for patients with R/R DLBCL ineligible for ASCT. Treatment in this setting has not historically been with curative intent, and these 5-year data in a limited number of patients do not definitively confirm that tafasitamab + lenalidomide may be curative. Furthermore, no comparison with age- and sex-matched general population data has been performed. Nevertheless, durable responses were maintained in many patients after discontinuing treatment (including at least 8 who discontinued tafasitamab more than 6 months before the end of the study). Positron emission tomography and computed tomography scans are not sufficiently sensitive to fully ascertain disease eradication; novel assessment methods are needed to better understand whether patients whose disease remains in long-term remission without treatment are cured. Similarly, patients with long-term CR while continuing treatment are of unknown status with regard to being cured. Long-term follow-up data are also emerging from studies of CAR-T therapy. The phase II JULIET study of tisagenlecleucel was conducted in 115 patients with R/R DLBCL ineligible for or progressing after ASCT;16 14 of 24 (58%) patients maintained a response at the 5-year analysis, including 46% with a CR, and the median duration of response was 61.4 months (95% CI: 3.2-not estimable).17 The 5-year follow-up of ZUMA-1 (a phase I/II study of axicabtagene ciloleucel in 101 patients with refractory large B-cell lymphoma18), was recently published under the title of ‘curative potential’.19 Data supporting this claim included the 30% of patients with ongoing CR at data cutoff, after a median follow-up of 63.1 months, with a median duration of CR of 62.2 months (95% CI: 12.9-not estimable), and an estimated 51.0% rate of disease-specific survival, which excluded deaths unrelated to disease progression.19 Data on next anti-lymphoma therapy were not systematically collected as part of the L-MIND 5-year survival follow-up, but two patients were documented to have later received CAR-T therapy. One with 44.7 months on treatment, a best response of CR and subsequent investigator-assessed progressive disease received CAR-T therapy approximately 10 months after the end of treatment and was alive at the OS follow-up at 59.9 months. Another patient, who had a best response of PR and subsequent centrally-confirmed progressive disease with 7 months on treatment, received CAR-T therapy approximately 12 months after the end of treatment and died 4 months after CAR-T therapy. Pre-

J. Duell et al.

viously, a case report was published of a patient with a best response of stable disease in L-MIND who went on to experience CR with axicabtagene ciloleucel.20 In a real-world study of 82 patients from nine academic institutions, 91% of patients would not have been eligible for L-MIND, including 23% who had received prior anti-CD19 therapy (21% with prior CAR-T therapy).21 The population had challenging disease characteristics, substantial comorbidity, and were heavily pretreated with best possible care including experimental treatments (28% with ≥3 prior lines of therapy, 51% with IPI scores of 3-5, 46% with primary refractory disease and 33% with renal dysfunction); accordingly, clinical outcomes with tafasitamab + lenalidomide were lower than in L-MIND. Nevertheless, one of six patients with refractory disease to CAR-T therapy had CR with tafasitamab + lenalidomide, and four of 11 with relapsed disease after CAR-T therapy achieved a CR, suggesting that the combination immunotherapy can provide a feasible approach in a post-CAR-T therapy setting.21 Safety and tolerability are important factors when considering the eligibility for and selection of second-line and later therapies, with some patients being ineligible for CAR-T therapies because of the associated toxicity, including adverse hematologic events. The 5-year safety data from L-MIND are reassuring and prolonged responses observed in the study were not offset by any detrimental long-term treatment-related AE. The reduced exposure-adjusted incidences of hematologic and non-hematologic TEAE that were previously reported with the transition from tafasitamab + lenalidomide combination therapy to tafasitamab monotherapy up to 2 years were maintained or further reduced with tafasitamab monotherapy beyond 2 years. These outcomes of second-line treatment and beyond indicate a potential first-line applicability, which is supported by results from the phase Ib firstMIND study of tafasitamab + lenalidomide combined with standard R-CHOP therapy (NCT04134936). An objective response at the end of treatment was documented in 25/33 patients, with an ORR of 75.8% (95% CI: 57.7-88.9).22 Accrual into the phase III frontMIND study of tafasitamab + lenalidomide + R-CHOP versus R-CHOP in newly diagnosed patients with high-intermediate and high-risk DLBCL (NCT04824092) is now finished, with primary completion of the study due in 2025.23 Testing for minimal residual disease based on circulating tumor DNA is among promising new avenues for optimizing future outcomes in DLBCL, especially in the light of new advanced methods.24 Preliminary data suggest that negativity for minimal residual disease may be a surrogate biomarker for clinical benefit of tafasitamab + lenalidomide + R-CHOP,25 but it remains to be seen whether it will have potential applications beyond the first-line setting, including whether it could support a decision to stop therapy in patients with durable CR in the R/R DLBCL setting.24 In conclusion, the final 5-year analysis of L-MIND showed clinically significant and enduring responses to tafasitamab

Haematologica | 109 February 2024

564

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

+ lenalidomide combination therapy, followed by long-term tafasitamab monotherapy, in patients with R/R DLBCL ineligible for ASCT. The median duration of response had not been reached after a median of 44.0 months follow-up, with the appearance of a plateau in survival curves after 12-18 months, although the numbers of patients at later timepoints were limited. Long-term clinical benefit was observed across subgroups of clinical interest, including patients with poor prognosis risk factors. No new safety signals were identified, confirming the tolerability profile observed with earlier data cutpoints, and the incidence of AE declined during prolonged tafasitamab monotherapy. Together, these long-term results suggest that this immunotherapy combination may have curative potential, which is being explored in further studies. Disclosures JD has received research funding from MorphoSys AG and Regeneron. PA has received honoraria from Janssen, Celgene, AbbVie, AstraZeneca, Gilead, and Incyte; has played a consulting or advisory role for Janssen, Celgene, AbbVie, and AstraZeneca; and participated in speakers’ bureau for Janssen, Celgene, AbbVie, AstraZeneca, and Gilead. MA has sat on advisory boards for Takeda, Bristol Myers Squibb, Karyopharm, Gilead, and Incyte; has received research grants from Roche, Johnson & Johnson, and Takeda; and has received travel grants from Roche, Bristol Myers Squib, Celgene, Gilead, AbbVie, and AstraZeneca. GG has sat on advisory boards for AbbVie, AstraZeneca, BeiGene, Incyte, Janssen, and Roche and participated in speakers’ bureau for AbbVie and Janssen. EG-B has provided consultancy for Janssen, AbbVie, Gilead, Kiowa, EUSAPharma, Incyte, Lilly, and BeiGene; participated in speakers’ bureau for Janssen, AbbVie, Takeda, Kiowa, Roche, EUSAPharma, Incyte, and BeiGene; and received travel costs from Janssen, AbbVie, and EUSAPharma. WJ has provided consulting/advisory services for Mei Pharma, Debiopharm, Loxo, Takeda, AstraZeneca, BeiGene; and has received research funding from GSK, Acerta, BeiGene, Nordic Nanovector, Incyte, Debiopharm, Incyte, Genentech, Janssen, Loxo, Mei Pharma, MorphoSys AG, Takeda, and TG Therapeutics. AML has received honoraria from Bristol Myers Squibb, Servier, Celgene, AbbVie, and Amgen; provided consulting or advisory services for Incyte; and received research funding from Novartis, Janssen, AbbVie, Roche, Amgen, Sanofi Genzyme, Celgene, Bristol Myers Squibb, Servier, Incyte, Pfizer, IQVIA, Doxopharma, Verastem, BeiGene, Oncopeptides, Karyopharm, Archigen, CTI BioPharma, Debiopharm, MorphoSys AG, FibroGen, Mei Pharma, Regeneron, and Dr Reddy’s Laboratories Spa. KJM has received honoraria from Pharmacyclics, MorphoSys AG, Bristol Myers Squibb, Karyopharm Therapeutics, Kite Pharma/ Gilead Company, ADC Therapeutics, AbbVie, AstraZeneca, BeiGene, Genmab, Genentech, Janssen, Lilly, Incyte; and research funding from Pharmacyclics, Merck, and Bristol

J. Duell et al.

Myers Squibb. TM has received travel grants from Amgen, Jazz, Pfizer, Bayer, Kyowa Kirin, Celgene/BMS, Kite/Gilead, Janssen, and Takeda; honoraria for advisory board meetings from Kite/Gilead, Amgen, Novartis, Pfizer, Celgene/BMS, Daiichi Sankyo, Atara, Roche, and Janssen; honoraria for lectures from Kite/Gilead, Takeda, Janssen, Roche, Servier, Novartis, Celgene/BMS, and Pfizer; and research funding from Janssen, AstraZeneca, and Novartis. ZN has provided consulting/advisory services for Takeda, Janssen, AbbVie, Roche, Amgen, Servier, and Astellas. OT has provided consulting/advisory services for Takeda, AstraZeneca, BeiGene, Incyte, Janssen, Gilead, AbbVie, Roche, Sandoz, and Blueprint. CK and KG are employees of MorphoSys AG. AB is an employee of MorphoSys AG and a statistical consultant for Ludwig-Maximilians-University Hospital, Munich, Germany. AA is an employee of MorphoSys AG and holds stock in Paion AG. GS has provided consulting services for Roche/Genentech, Gilead Sciences, Janssen, Celgene, Novartis, MorphoSys AG, Epizyme, Alimera Sciences, Genmab, Debiopharm Group, VelosBio, Bristol Myers Squibb, BeiGene, Miltenyi Biotec, and Ipsen; and has received honoraria from Roche/Genentech, Janssen, Celgene, Gilead Sciences, Novartis, AbbVie, and MorphoSys AG. Contributions JD, PA, MA, GG, EG-B, WJ, NK, AML, KJM, TM, ZN, OT, CK, AB, AA, KG, and GS conceived the study. JD, PA, MA, GG, EG-B, WJ, NK, AML, KJM, TM, ZN, OT, KG and GS were responsible for the investigation. AB, CK, AA, and KG were responsible for methodology and resources. AA was the project administrator. AA, AB, JD, CK, and KG supervised the study, and curated and analyzed the data. AB was responsible for validation. JD, AB, CK, AA, KG, and NK wrote the original draft of the article. JD, PA, MA, GG, EG-B, WJ, NK, AML, KJM, TM, ZN, OT, CK, AB, AA, KG, and GS reviewed and edited the original draft. Acknowledgments The authors thank the patients, caregivers, and study investigators. Medical writing assistance was provided by Pavitra Joshi, MS, and Emma Leah, PhD, of Syneos Health. Funding This study was sponsored by MorphoSys AG. Medical writing assistance was funded by MorphoSys AG, in accordance with Good Publication Practice. Data-sharing statement Data-sharing requests by qualified researchers pertaining to the L-MIND study will be considered only for non-commercial use on a case-by-case basis (to be approved by MorphoSys; contact Daniel.Moik@morphosys.com), starting 12 months after acceptance of the manuscript and until 36 months thereafter. Approval may be subject to a data access agreement.

Haematologica | 109 February 2024

565

ARTICLE - Tafasitamab + lenalidomide in R/R DLBCL: 5-year data

J. Duell et al.

References 1. World Health Organization. World Cancer Report: Cancer Research for Cancer Prevention. IARC Press; 2020. 2. Sarkozy C, Sehn LH. New drugs for the management of relapsed or refractory diffuse large B-cell lymphoma. Ann Lymphoma. 2019;3:310. 3. Crump M, Neelapu SS, Farooq U, et al. Outcomes in refractory diffuse large B-cell lymphoma: results from the international SCHOLAR-1 study. Blood. 2017;130(16):1800-1808. 4. Locke F, Miklos DB, Jacobson C, et al. Primary analysis of ZUMA 7: a phase 3 randomized trial of axicabtagene ciloleucel (axicel) versus standard of care therapy in patients with relapsed/ refractory large B-cell lymphoma. Blood. 2021;138(Suppl 1):2. 5. Kamdar M, Solomon SR, Arnason J, et al. Lisocabtagene maraleucel versus standard of care with salvage chemotherapy followed by autologous stem cell transplantation as secondline treatment in patients with relapsed or refractory large B-cell lymphoma (TRANSFORM): results from an interim analysis. Lancet. 2022;399(10343):2294-2308. 6. González-Barca E, Boumendil A, Blaise D, et al. Outcome in patients with diffuse large B-cell lymphoma who relapse after autologous stem cell transplantation and receive active therapy. A retrospective analysis of the Lymphoma Working Party of the European Society for Blood and Marrow Transplantation. Bone Marrow Transplant. 2020;55(2):393-399. 7. Sarkozy C, Coiffier B. Diffuse large B-cell lymphoma in the elderly: a review of potential difficulties. Clinical Cancer Res. 2013;19(7):1660-1669. 8. Gisselbrecht C, Glass B, Mounier N, et al. Salvage regimens with autologous transplantation for relapsed large B-cell lymphoma in the rituximab era. J Clin Oncol. 2010;28(27):4184-4190. 9. Chihara D, Izutsu K, Kondo E, et al. High-dose chemotherapy with autologous stem cell transplantation for elderly patients with relapsed/refractory diffuse large B cell lymphoma: a nationwide retrospective study. Biol Blood Marrow Transplant. 2014;20(5):684-689. 10. Horton HM, Bernett MJ, Pong E, et al. Potent in vitro and in vivo activity of an Fc-engineered anti-CD19 monoclonal antibody against lymphoma and leukemia. Cancer Res. 2008;68(19):8049-8057. 11. Salles G, Duell J, González Barca E, et al. Tafasitamab plus lenalidomide in relapsed or refractory diffuse large B-cell lymphoma (L-MIND): a multicentre, prospective, single-arm, phase 2 study. Lancet Oncol. 2020;21(7):978-988. 12. MONJUVI. Prescribing information. Boston, MA: MorphoSys. 2020. https://www.monjuvi.com/pi/monjuvi-pi.pdf. Accessed January 19, 2023. 13. European Medicines Agency. Minjuvi, SmPC 2022. https://www. ema.europa.eu/en/medicines/human/EPAR/minjuvi. Accessed January 19, 2023.

14. Duell J, Maddocks KJ, González-Barca E, et al. Long-term outcomes from the phase II L-MIND study of tafasitamab (MOR208) plus lenalidomide in patients with relapsed or refractory diffuse large B-cell lymphoma. Haematologica. 2021;106(9):2417-2426. 15. Cheson BD, Pfistner B, Juweid ME, et al. Revised response criteria for malignant lymphoma. J Clin Oncol. 2007;25(5):579-586. 16. Schuster SJ, Tam CS, Borchmann P, et al. Long-term clinical outcomes of tisagenlecleucel in patients with relapsed or refractory aggressive B-cell lymphomas (JULIET): a multicentre, open-label, single-arm, phase 2 study. Lancet Oncol. 2021;22(10):1403-1415. 17. Chong EA, Ruella M, Schuster SJ. Five-year outcomes for refractory B-cell lymphomas with CAR T-cell therapy. N Engl J Med. 2021;384(7):673-674. 18. Locke FL, Ghobadi A, Jacobson CA, et al. Long-term safety and activity of axicabtagene ciloleucel in refractory large B-cell lymphoma (ZUMA-1): a single-arm, multicentre, phase 1–2 trial. Lancet Oncol. 2019;20(1):31-42. 19. Neelapu SS, Jacobson CA, Ghobadi A, et al. Five-year follow-up supports curative potential of axicabtagene ciloleucel in refractory large B-cell lymphoma (ZUMA-1). Blood. 2023;141(19):2307-2315. 20. Oliai C, de Vos S. Case report: Sustained remission achieved from anti-CD19 CAR T cell therapy despite prior treatment with anti-CD19 antibody tafasitamab (MOR208) in a patient with relapsed and refractory diffuse large B-cell lymphoma. Blood. 2019;134(Suppl 1):5360. 21. ClinicalTrials.gov. Qualls D, Buege MJ, Dao P, et al. Tafasitamab and lenalidomide in relapsed/refractory large B cell lymphoma (R/R LBCL): real world outcomes in a multicenter retrospective study. Blood. 2022;140(Suppl 1):787-789. 22. Belada D, Kopeckova K, Bergua Burgues JM, et al. First-MIND: primary analysis from a phase Ib, open-label, randomized study to assess safety of tafasitamab or tafasitamab + lenalidomide in addition to R-CHOP in patients with newly diagnosed diffuse large B-cell lymphoma. Blood. 2021;138(Suppl 1):3556. 23. Tafasitamab + lenalidomide + R-CHOP versus R-CHOP in newly diagnosed high-intermediate and high risk DLBCL patients (frontMIND). https://clinicaltrials.gov/ct2/show/NCT04824092. Accessed January 19, 2023. 24. Galimberti S, Genuardi E, Mazziotta F, et al. The minimal residual disease in non-Hodgkin’s lymphomas: from the laboratory to the clinical practice. Front Oncol. 2019;9:528. 25. Kurtz DM, Hogan GJ, Schultz A, et al. Ultrasensitive MRD profiling predicts outcomes in DLBCL after frontline therapy with tafasitamab in combination with lenalidomide and R-CHOP. Blood. 2022;140(Suppl 1):3498-3499.

Haematologica | 109 February 2024

566

ARTICLE - Plasma Cell DIsorders

PVT1 interacts with polycomb repressive complex 2 to suppress genomic regions with pro-apoptotic and tumour suppressor functions in multiple myeloma Patrick Nylund,1 Berta Garrido-Zabala,1 Alba Atienza Párraga,1 Louella Vasquez,2 Paul Theodor Pyl,2,3 George Mickhael Harinck,1 Anqi Ma,4 Jian Jin,4 Fredrik Öberg,1 Antonia Kalushkova1 and Helena Jernberg Wiklund1 Science for Life Laboratory, Department of Immunology, Genetic and Pathology, Rudbeck

Laboratory, Uppsala University, Uppsala, Sweden; 2Department of Laboratory Medicine, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Lund University, Lund, Sweden; 3Department of Clinical Sciences, National Bioinformatics Infrastructure Sweden, Science for Life Laboratory, Lund University, Lund, Sweden and 4Mount Sinai Center for Therapeutics Discovery, Departments of Pharmacological Sciences and Oncological Sciences, Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai, New

Correspondence: H. Jernberg Wiklund helena.jernberg_wiklund@igp.uu.se P. Nylund patrick.nylund@igp.uu.se Received: Accepted: Early view:

March 1, 2023. July 20, 2023. July 27, 2023.

York, NY, USA

Abstract Multiple myeloma is a heterogeneous hematological disease that originates from the bone marrow and is characterized by the monoclonal expansion of malignant plasma cells. Despite novel therapies, multiple myeloma remains clinically challenging. A common feature among patients with poor prognosis is the increased activity of the epigenetic silencer EZH2, which is the catalytic subunit of the PRC2. Interestingly, the recruitment of PRC2 lacks sequence specificity and, to date, the molecular mechanisms that define which genomic locations are destined for PRC2-mediated silencing remain unknown. The presence of a long non-coding RNA (lncRNA)-binding pocket on EZH2 suggests that lncRNA could potentially mediate PRC2 recruitment to specific genomic regions. Here, we coupled RNA immunoprecipitation sequencing, RNA-sequencing and chromatin immunoprecipitation-sequencing analysis of human multiple myeloma primary cells and cell lines to identify potential lncRNA partners to EZH2. We found that the lncRNA plasmacytoma variant translocation 1 (PVT1) directly interacts with EZH2 and is overexpressed in patients with a poor prognosis. Moreover, genes predicted to be targets of PVT1 exhibited H3K27me3 enrichment and were associated with pro-apoptotic and tumor suppressor functions. In fact, PVT1 inhibition independently promotes the expression of the PRC2 target genes ZBTB7C, RNF144A and CCDC136. Altogether, our work suggests that PVT1 is an interacting partner in PRC2-mediated silencing of tumor suppressor and pro-apoptotic genes in multiple myeloma, making it a highly interesting potential therapeutic target.

Introduction Multiple myeloma (MM) is a hematological malignancy characterized by aberrant monoclonal expansion of malignant plasma cells (PC) within the bone marrow.1 Despite advances in treatment, disease management and therapeutic interventions remain challenging and non-curative. Global multi-omics analyses have revealed that MM cells exhibit complex intra-tumoral heterogeneity, have a diverse mutational landscape and undergo large scale epigenetic and metabolic reconfiguration during disease progression.2,3 As a result, patients undergoing conventional treatment eventually develop drug resistance and relapse in disease.4 In

recent years, we and others have presented data suggesting that epigenetic regulatory mechanisms are key features in MM pathogenesis, and numerous drugs targeting epigenetic regulators have been developed and tested clinically or preclinically.5 These drugs target epigenetic modifiers such as histone deacetylases (HDAC),6 DNA methyltransferases (DNMT)7 and histone metyltransferases (HMT).8 For instance, transcriptional repression through epigenetic redistribution of histone H3 lysine 27 tri-methylation (H3K27me3), catalyzed by the polycomb repressive complex 2 (PRC2), has been reported to be a common feature of MM by us and others.9-11 Accordingly, the catalytic component of PRC2, enhancer of zeste homolog 2 (EZH2), is overexpressed in

Haematologica | 109 February 2024

567

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

MM,3 and increased methylation of H3K27 correlates with tumor progression according to the MM International Staging System (ISS).1 We have previously demonstrated that EZH2 inhibition (EZH2i) reduces the viability of MM cells through the re-activation of microRNA (miRNA) which silence methionine cycling-associated genes and oncogenes involved in MM proliferation.8 While the mechanisms behind enzyme-mediated epigenetic regulation of MM are in the process of being unravelled, potential regulatory mechanisms of a class of non-enzymatic epigenetic regulators, such as long non-coding RNA (lncRNA), remain largely unexplored. Large-scale transcriptome efforts have identified an extensive number of lncRNA, that are involved in vital cellular processes for disease development such as malignant transformation, early tumor onset, chromatin reorganization, cell differentiation and gene expression modulation.12 Recently, the first Cancer LncRNA Census was generated to effectively identify lncRNA with a putative causal role in cancers of various origin.13 The lack of sequence specificity and the presence of a lncRNA binding pocket14 on the enzymatic subunit EZH2 of the PRC2 complex suggests that lncRNA should be a highly interesting partner for PRC2 recruitment to specific genomic regions. One of the most interesting candidates among the list of potentially relevant lncRNA for MM is PVT1.13 In lung cancer and other diseases PVT1 has been attributed the ability to promote transcriptional repression in a context-dependent manner by facilitating the deposition of H3K27me3 on various promoter regions through the recruitment of EZH2.15 Moreover, PVT1 has also been suggested to modulate gene expression patterns by stabilizing PRC2 in various cancers,16-18 and its inhibition resulted in decreased EZH2 expression, promoted apoptosis and reduced tumor cell proliferation in a number of cancer types, including other hematological malignancies.16,19 PVT1 has previously been associated with relapse and drug resistance in MM20 and its overexpression is connected to increased genomic stability in MM cells, providing enhanced protection against DNA damage.21 In addition, PVT1 expression can be transcriptionally activated by c-Myc binding to the PVT1 promoter.22 Interestingly, a recent single-cell RNA sequencing (scRNA-seq)-based gene fusion analysis of immunoglobulin in MM reported immunoglobulin (Ig) loci fusion with either MYC, a known oncogene, or its downstream neighbor PVT1, resulting in gain of MYC expression.23 Moreover, patients harboring PVT1-IGL translocation had worse prognosis than patients with MYC-IGL translocation.23 However, a functional role of PVT1 as a PRC2 collaborator has not yet been demonstrated in MM. To date, the PVT1-EZH2 interaction has not been described in MM, and a comprehensive genome-wide understanding of the relationship between PVT1 and PRC2-mediated silencing is lacking. In this study we determined that PVT1 is overexpressed in MM patients and its expression is as-

sociated with a poor prognosis. Moreover, we determined that a physical interaction between PVT1-EZH2 exists in MM cells. This interaction occurs at specific gene locations and regulates genes associated with apoptosis as well as tumor suppressor genes (TSG) such as CXCL14, RNF144A and ZBTB7C, which are linked to oncogenic function and immune system evasion in MM. Taken together, our study identifies PVT1-mediated PRC2 targeting as a regulator of gene silencing in MM, thus highlighting PVT1 as a potentially interesting therapeutic target for patients affected by this malignancy.

Methods Cell culture Human MM authenticated cell lines were cultured and supplemented as previously described.8 Potential mycoplasma infections were investigated before the start of each experimental procedure utilizing MycoAlertTM Mycoplasma Detection Kit (Lonza; Basel, Switzerland; cat. no. LT07118). RNA immunoprecipitation sequencing RNA immunoprecipitation was conducted utilizing Magna Nuclear RIP kit (Millipore; Billerica, MA, USA; cat. no. 1710520 and 17-700) as described by the manufacturer. In brief, 1.0x107 INA-6 cells were collected and cross-linked with formaldehyde with a final concentration of 0.3% for 10 minutes at room temperature. Excess formaldehyde was quenched with glycine. Post outer membrane and nuclear membrane lysis, the cells were sonicated for ten cycles on Pico Bioruptor™ (Diagenode) (30 seconds on/30 seconds off). Immunoprecipitation of EZH2 targets was conducted by incubating the samples with 5 μg of anti-EZH2 (cat. no. 17-662, Millipore) and anti-IgG Mouse (cat. no. CS200621, Millipore) antibodies over night at 4°C. RNA was purified and cleaned by using RNeasy Micro Kit (Netherlands, Qiagen; cat. no. 74004). Complementary DNA conversion and quantitative polymerase chain reaction analysis were performed as previously described8 with primers found in the Online Supplementary Table S6. RNA immunoprecipitation sequencing library preparation and analysis RNA concentration was measured using QubitTM (Thermo Scientific). One hundred ng of RNA was used for sample library preparation using TruSeq Stranded Total RNA Gold (Illumina) with non-poly-A selection. Samples were then sequenced 50 cycles pair-end on one lane of a SP flow cell on NovaSeq 6000 system and v1 sequencing chemistry (Illumina). The fastq files from three biological replicates of RNA immunoprecipitation sequencing (RIP-seq) were concatenated per read pair to generate one pooled fastq file. The read mapping was then carried out using the nf-core24 RNA sequencing (RNA-seq) pipeline (https://doi.

Haematologica | 109 February 2024

568

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

org/10.5281/zenodo.3503887) in version 1.4.2 using default parameters for paired-end sequencing, but with additional flags -reverseStranded - removeRiboRNA.25 The BAMS were used by RIPSeeker26 to statistically infer RIP regions for each strand given the background of input RNA, with parameter setting minBinSize=200 and maxBinSize=10,000. The RIP regions were selected at an estimated false discovery rate (eFDR) of 5%. Transfection One hundred and sixty thousand cells/mL MM.1S cells were seeded in Opti-MEMTM Reduced Serum Media (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA; cat. no. 31985070) and were allowed to attach over 24 hours (h) before transfection. HiPerFect transfection reagent (Netherlands, Qiagen; cat. no. 301704) and PVT1 GapmeR (200 nM) (Qiagen, Netherlands, cat. no. 339517) were added (1:3,000) and the cells were incubated for 72 h at 37°C in a humidified 5% CO2 in-air atmosphere. Transfection efficiency was evaluated 72 h post transfection using 5-FAM-labeled positive/negative control GapmeR (Qiagen, Netherlands, cat. no. 339515) on Cytoflex LX (Beckman Coulter, Brea, CA, USA). Data was analyzed utilizing CytExpert v.2.4.0.28 (Beckman Coulter). Validation of PVT1 inhibition was evaluated by real-time quantitative polymerase chain reaction. Other methods Additional methods are described in the Online Supplemental Appendix.

Results Long non-coding RNA PVT1 directly binds EZH2 in multiple myeloma cells The expression of the catalytic subunit of PRC2 (EZH2) has been associated with disease progression from premalignant to malignant MM,27 and patients exhibiting high expression of EZH2 have a significantly poorer prognosis (Online Supplementary Figure S1A). In addition, high expression of EZH2 correlates with poor survival in patients treated with bortezomib, dexamethasone, lenalidomide as monotherapies, or in combination therapies (Online Supplementary Figure S1B-F). To date, the mechanisms which mediate the recruitment of PRC2 to specific genomic regions have not been elucidated, however, prior studies have suggested that lncRNA may be putative partners contributing to PRC2 genomic binding and consequently to silencing of genes in selected cancer types.14,28,29 In order to evaluate whether dysregulation of lncRNA is a contributing mechanism to PRC2 targeting in MM, we first analyzed lncRNA expression in MM patients using transcriptomic data from the Blueprint Consortium Cohort (BCC) and identified 67 dysregulated lncRNA (5% false discovery rate [FDR]) with a putative role in the tumor (Figure 1A; Online

Supplementary Table S1). In order to determine whether a functional and direct interaction may exist between PRC2 and the identified lncRNA in MM, we performed RNA immunoprecipitation coupled with sequencing (RIP-seq) against EZH2 in the INA-6 MM cell line and found 101 lncRNA (5% FDR) that physically interacted with EZH2 (Figure 1B; Online Supplementary Table S2). By overlapping the list of EZH2-bound lncRNA with the list of lncRNA overexpressed in MM patients, we identified the lncRNA PVT1, PCAT1 and SAMD12-AS1 as potential mediators of PRC2 targeting to chromatin in MM (Figure 1C). One interesting candidate among the list of potentially relevant lncRNA for MM is PVT1, which has been previously reported to regulate the expression of EZH2 and has been associated with relapse and drug resistance in various cancers.29,30 However, its interaction to EZH2 has not yet been resolved in MM cells. In order to elucidate whether EZH2 and PVT1 interact in MM, we first validated PVT1 overexpression in MM cell lines compared to peripheral blood mononuclear cells (PBMC) (Online Supplementary Figure S1G). Interestingly, EZH2i resulted in decreased PVT1 expression in the EZH2i-sensitive MM cell line INA-6, but not in the EZH2i-resistant U1996 MM cells (Online Supplementary Figure S1H). Accordingly, pull-down of EZH2 by RIP-quantitative polymerase chain reaction (RIP-qPCR) confirmed a direct PVT1-EZH2 interaction in the MM cell lines INA-6, KMS-28PE, MM.1S and U1996 (Online Supplementary Figure S1I). PVT1 expression is associated with disease progression and poor prognosis in multiple myeloma patients In order to assess the clinical relevance of PVT1 in MM, we analyzed expression data collected from three independent data sets of MM patients. Using the BCC, we found that MM cells expressed higher levels of PVT1 than normal plasma cells (Figure 1D). Interestingly, PVT1 expression levels were heterogeneous across MM patients (Online Supplementary Figure S1J) and positively correlated with increased ISS stage of MM (CoMMpass cohort) (Figure 1E). In accordance with previously published data,20 we found that premalignant stages of MM such as monoclonal gammopathy of undetermined significance (MGUS) and smouldering MM (sMM) harbor increased levels of PVT1 as compared to normal bone marrow plasma cells (BMPC) (Online Supplementary Figure S1K), suggesting a potential role of PVT1 already in the early stages of tumor development. Finally, high expression of PVT1 was associated with poor overall survival in newly diagnosed MM patients (CoMMpass cohort; Figure 1F), and patients that were resistant to conventional bortezomib treatment (GSE97582 cohort; Online Supplementary Figure S1L). Stratification of patients based on molecular classification revealed an increased expression of PVT1 in patients with a hyperdiploid karyotype (GSE4581 cohort) (Online Supplementary Figure S1M), while patient grouping based on cytogenetics revealed that patients with

Haematologica | 109 February 2024

569

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

Figure 1. RNA immunoprecipitation sequencing analysis determined a physical interaction between EZH2 and the long non-coding RNA PVT1. (A) Expression profile of upregulated long non-coding RNA (lncRNA) in multiple myeloma (MM) patients from RNA-sequencing (RNA-seq) data with P<0.05 and 5% false discovery rate (FDR) cutoff, collected from the Blueprint Consortium (Nmm=3, NtPC=3). (B) EZH2-RNA interactome collected from RNA immunoprecipitation sequencing (RIP-seq) in the MM cell line INA-6 with 5% estimated FDR (eFDR) cutoff. Samples collected from 3 biological replicates. Red circles indicate EZH2-PVT1 interaction. (C) Overlap between lncRNA overexpressed in MM patients and lncRNA interacting with EZH2. (D) Log2 normalized expression data of PVT1 in MM patients compared to tonsil plasma cells (tPC) from the Blueprint Consortium dataset. Statistical analysis was performed with student t test. (E) PVT1 expression categorized by Internation Staging System (ISS) stage from the MMRF-CoMMpass dataset. Statistical analysis was performed with one-way ANOVA. Values are presented with standard error of the mean. (F) Overall survival data associated with PVT1 expression (MMRF-CoMMpass, N=667). Statistical test was performed with log rank (Mantel-Cox test). *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001. Haematologica | 109 February 2024

570

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

17p deletion exhibit the lowest PVT1 expression (Online Supplementary Figure S1N). In summary, PVT1 expression increases gradually across premalignant and progressive stages of MM, suggesting a potential role of PVT1 in MM disease progression. PVT1-PRC2-mediated silencing of genes involved in tumor suppression and apoptosis signaling can be reversed in multiple myeloma We then sought to study the relationship between PVT1 and PRC2. First, we investigated potential genomic binding sites for PVT1 using LongTarget,31 a lncRNA-genomic DNA interaction tool which predicts potential triplex-forming oligos and triplex target sites. We identified 8,976 potential PVT1-binding sites and their closest corresponding genes (Online Supplementary Figure S2; Online Supplementary Table S3).32 Next, we overlapped the obtained list of predicted

PVT1 gene targets with genes that fulfilled the requirements of being downregulated in MM patients (Blueprint cohort, 5% FDR) and genes that are enriched for H3K27me3 in MM patients (Blueprint cohort, 5% FDR). The analysis resulted in a list of 141 genes which are predicted to be both PRC2 targets (H3K27me3-enriched) and PVT1 targets, suggesting that this subset of genes may be subjected to PVT1-PRC2 regulation in MM (Figure 2A, B; Online Supplementary Table S4). In order to validate which of these genes could be regulated by the PVT1-PRC2 complex, we treated INA-6 MM cells with the EZH2 inhibitor UNC1999 and performed RNA-seq. We identified 713 genes that gained expression post-EZH2 inhibition (Figure 3A), 270 of which were also predicted PVT1 genomic binding sites (Online Supplementary Figure S3A; Online Supplementary Table S5). Interestingly, 21 of the identified genes had a known TSG function based on

Figure 2. EZH2-PVT1 axis potentially regulates 141 genes in multiple myeloma patients. (A) Overlap between PVT1 genomic targets (LongTarget), as well as downregulated genes and H3K27me3-enriched genes identified in multiple myeloma (MM) patients (Blueprint Consortium). (B) Differential expression of selected genes identified as overlapping in (A). CPM: counts per million.

Haematologica | 109 February 2024

571

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

TSGene 2.0 (https://bioinfo.uth.edu/TSGene/) (Online Supplementary Figure S3B). Among these, CXCL14 and ZBTB7C showed de novo activation and RNF144A together with CCDC136 demonstrated an increase in expression post EZH2 inhibition (Figure 3B). Low expression profiles were observed for CXCL14, ZBTB7C, RNF144A and CCDC136 in primary MM samples (Online Supplementary Figure S3C-F) and were associated with poor prognosis in MM patients (Figure 3C-F). Similarly, decreased expression of these genes was identified in MGUS and sMM, the asymptomatic prestages

of MM (Figure 4A-D). In order to evaluate the functional relationship between EZH2 and PVT1, we knocked down PVT1 expression by transfecting MM cells with GapmeR directly targeting the PVT1 transcripts (Online Supplementary Table S7). Inhibition of PVT1 expression was successful after 72 h of transfection (Figure 4E). Interestingly, PVT1 inhibition (PVT1i) promoted a gain of expression of ZBTB7C, RNF144A and CCDC136, consistent with EZH2i in MM cells (Figure 4F). In addition, gene set enrichment analysis (GSEA) of the overall list of PRC2 target genes showed significant enrichment of

Figure 3. Inhibition of EZH2 results in de novo activation of the tumor suppressor genes CXCL14 and ZBTB7C. (A) Differential expressed genes in UNC1999-treated MM cells (log2 fold change [FC] >1.5 and adjusted (adj.) P value <0.05). (B) Log2 TPM difference of 21 tumor suppressor genes upregulated in multiple myeloma (MM) cells post-UNC1999 treatment as compared to dimethyl sulfoxide control (DMSO). Samples collected from 3 biological replicates per condition. (C-F) Survival analysis of MM patients (GSE9782, N=264) after normalization (MAS5) and expression-based stratification of the following genes: CXCL14, ZBTB7C, RNF144A and CCDC136. Samples for expression analysis consisted of bone marrow CD138+ cells. RNA-seq: RNA-sequencing.

Haematologica | 109 February 2024

572

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

EZH2 is the catalytic subunit of PRC2 and is responsible for the deposition of methyl groups to histone H3 lysine 27. We and others have previously demonstrated the clinical

relevance of targeting PRC2 in MM and that PRC2-mediated gene silencing is a key feature of MM pathogenesis.2,3 One challenging aspect of defining target genes of PRC2 is that this complex lacks sequence specificity; thus, the molecular mechanisms of its genomic localization are largely unknown. Prior studies have suggested that EZH2-lncRNA interactions could promote PRC2’s functional capacity to bind chromatin. Indeed, while EZH2 does not contain a conventional RNA-binding motif, it includes a RNA-binding domain in residues 342–368 of the protein,28 as well as a major RNA-binding site within its N-terminal helix.14 Increasing evidence highlights the physiological and pathological impact that lncRNA have on cancer cell proliferation, metastasis, invasion, relapse, resistance, and genomic stability.21 Dysregulation of lncRNA has been observed in various cancers, including MM, and numerous studies have provided insight into the diversity of the biological functions that lncRNA can have an impact on during cancer patho-

genes regulating apoptosis in INA-6 cells that underwent EZH2 inhibition (Figure 5A). Among these, we identified a subset of PRC2-PVT1 targets, such as TNF, CCNA1, IGFBP6, SATB1 and PLCB2 (Figure 5B). Importantly, we found that decreased expression of these genes was associated with a poor prognosis (Figure 5C-F) and advanced stages of the disease (Online Supplementary Figure S3G-K), while SATB1 showed no correlation to poor prognosis in MM patients. Taken together, our data suggests that PVT1-mediated PRC2 targeting regulates apoptosis and mediates the silencing of a selected number of TSG in MM.

Discussion

Figure 4. Loss of the EZH2-PVT1 regulatory axis promotes expression of the tumor suppressor genes ZBTB7C, RNF144A and CCDC136. (A-D) Normalized (MAS5) CXCL14, ZBTB7C, RNF144A and CCDC136 expression data from bone marrow-collected CD138+ cells from monoclonal gammopathy of undetermined significance (MGUS) and smoldering multiple myeloma (sMM) (GSE5900) compared to bone marrow plasma cells (BMPC). (E) Normalized relative expression of PVT1 post 72 hours transfection of GapmeR targeting PVT1 in MM.1S MM cells. (F) Normalized relative expression of ZBTB7C, RNF144A and CCDC136 post 72 hours PVT1i in MM.1S MM cells. Statistical analysis was performed with one-way ANOVA with Tukey test for multiple comparisons or multiple t test. Values are presented with standard error of the mean.*P<0.05; **P<0.01; ***P<0.001; ****P<0.0001. GSEA: gene set enrichment analysis. Haematologica | 109 February 2024

573

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

Figure 5. PRC2-PVT1 axis regulates genes associated with apoptosis signaling. (A) Gene set enrichment analysis (GSEA) of PRC2 target genes in INA-6 multiple myeloma (MM) cells. (B) GSEA of PRC2-PVT1 target genes in INA-6 MM cells. (C-F) Survival analysis of MM patients (GSE9782, N=264; GSE9782, N=262) after normalization (MAS5) and expression-based stratification of the following genes: TNF, CCNA1, IGFBP6 and PLCB2. Samples for expression analysis consisted of bone marrow CD138+ cells.

genesis.20,33-35 Therefore, we sought to evaluate a potential lncRNA-mediated targeting mechanism of PRC2 in MM. Herein, three of the 67 lncRNA upregulated in MM primary

cells PVT1, PCAT1 and SAMD12-AS1 were found to interact with the EZH2 protein. PVT1 was overexpressed in MGUS, sMM and MM compared to normal PC, and its expression

Haematologica | 109 February 2024

574

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

Figure 6. Suggested mechanism of action of EZH2-PVT1-mediated silencing. Adapted from “Epigenetic Deregulation in Cancer” by BioRender.com (2023). Retrieved from https://app.biorender.com/biorender-templates. IncRNA: long non-coding RNA.

gradually increased with ISS staging. In this paper, we show in two independent large MM cohorts that PVT1 expression is also associated with poor prognosis. In our previous work we found no correlation between Polycomb-mediated expression signatures and specific genetic alterations,3,11 which suggests that the epigenetically regulated signature mediated by PVT1 is likely independent from these genetic alterations. The functional implication of EZH2-PVT1 interaction has not been fully investigated in the context of hematological malignancies, including MM. Thus, we sought to unravel the relationship between PRC2 and PVT1 target genes in this malignancy. EZH2 inhibition in MM cells resulted in downregulation of PVT1 expression, further solidifying the functional relationship between EZH2 and PVT1. Moreover, we found that 270 PRC2 target regions overlapped with genomic targets for PVT1. This suggests that PVT1 plays a pivotal role in mediating EZH2 targeting in MM, which is in line with what has been reported for non-small cell lung cancer.15 Previous studies reported that treatment of MM cells with a PVT1 inhibitor resulted in decreased cell proliferation and induction of apoptosis.36 In line with this finding, we now suggest an important role for PVT1 as an interacting partner to PRC2 by showing that EZH2 inhibition resulted in increased expression of PRC2-PVT1 target genes associated with apoptosis regulation, such as TNF, IGFBP6, CCNA1, PLCB2 and SATB1. Induced TNF expression has been reported to promote cell death in MM cell lines through the NFκB pathway,37 and PLCB2 expression has been associated with a favorable prognosis in other hematological malignancies such as AML.38 CCNA1 has previously been identified as a PRC2 target in AML, and decreased expression of SATB1 resulted in increased cell proliferation in AML.39,40 Importantly, we also show that these genes are downregulated in more advanced stages of MM, highlighting the potential relevance of their silencing as the disease progresses. Loss of the EZH2-PVT1 axis was also associated with the upregulation of 21 tumor suppressor genes. Interestingly,

CXCL14 and ZBTB7C showed de novo activation. Downregulation of CXCL14 is an important step in malignancy transformation within the bone marrow.41,42 Indeed, previous studies have suggested that CXCL14 is needed for trafficking natural killer cells to sites of inflammation or oncogenesis as well as for the inhibition of the CXCL12-CXCR4 axis, which is critical for the migration of malignant cells.41,42 Strikingly, similar to the effects observed with EZH2 inhibition, PVT1 inhibition in MM cells resulted in the gain of expression of ZBTB7C, RNF144A and CCDC136, suggesting a co-regulatory relationship of these genes by the proposed PRC2-PVT1 functional axis. ZBTB7C binds to p53 in solid tumors to prevent p53-mediated activation of CDKN1A, a known oncogene in both Burkitt lymphoma and MM, suggesting that ZBTB7C repression is of importance for MM oncogenesis.43,45 Interestingly, suppression of the E3 ligase RNF144A has previously been described to increase survival of glioblastoma cells in stressful microenvironments and disruption of EZH2-mediated silencing in these cells assisted in overcoming drug resistance.45 CCDC136 has been identified as a putative TSG and is frequently deleted in various malignancies such as gastric cancer,46 however, its exact function in the cancer setting remains unclear. Studies in zebrafish have shown that CCDC136 promotes enhanced Wnt/β-catenin activity during zebrafish development.47 In summary, our study demonstrates that the PVT1-mediated EZH2 recruitment to genomic loci is responsible for the targeted silencing of genes associated with apoptosis (Figure 6) and regulates the expression of important oncogenes in MM. This makes PVT1 an attractive candidate for targeted therapy in MM. Disclosures No conflicts of interest to disclose. Contributions PN, AAP, AK and HJW conceptualized the project. PN, BGZ, GMH and AAP acquired data. PN, PTP and LV performed formal analysis of the data. JJ and AM provided reagents.

Haematologica | 109 February 2024

575

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

PN, BGZ, AAP, AK and HJW assisted in project investigation. HJW provided acquisition of funding. FÖ, AK and HJW supervised the project. PN and HJW administrated the project. PN visualized all the data. PN organized and integrated the data. PN wrote the original manuscript draft. All authors read and approved the final manuscript. Acknowledgments We are grateful to Charlotta Sandberg Blixt for the excellent technical assistance with the cell laboratory work. Flow cytometry analysis was performed at BioVis - Biological Visualization, Science for Life Laboratory, Department of Immunology, Genetics and Pathology, Rudbeck Laboratory, Uppsala University, Sweden. The data handling was enabled by resources provided by the Swedish National Infrastructure for Computing (SNIC) at UPPMAX. We are grateful for the assistance provided by The SciLifeLab Bioinformatics platform NBIS (National Bioinformatics Infrastructure Sweden) for their bioinformatics support related to the RIP-sequencing, ChIP-sequencing and RNA-sequencing data. We would also like to thank the participants in CoMMpass study and the MMRF for sharing sequencing and clinical data through the MMRF genomics portal. These data

were generated as part of the Multiple Myeloma Research Foundation Personalized Medicine Initiatives (https://research. themmrf.org and www.themmrf.org). This study makes use of data generated by the Blueprint Consortium. A full list of investigators who contributed to the generation of the data is available from www.blueprint-epigenome.eu. Funding for the project was provided by the European Union’s Seventh Framework Programme (FP7/2007-2013) under grant agreement no. 282510 - BLUEPRINT. Funding The project was supported by grants from the Swedish Cancer Society (CAN 2016/458, 200727 PjVSF) and the Swedish Research Council (K2019-64X-20102-13-3/KDB 1335/17). Additional support was provided by the Knut and Alice Wallenberg Foundation (KAW 2017.0003) as part of the National Bioinformatics Infrastructure Sweden at SciLifeLab. Data-sharing statement RIP-sequencing and RNA-sequencing data have been deposited at the ArrayExpress platform with the accession numbers E-MTAB-13135 and E-MTAB-13136, respectively.

References 1. Walker BA, Wardell CP, Melchor L, et al. Intraclonal heterogeneity is a critical early event in the development of myeloma and precedes the development of clinical symptoms. Leukemia. 2014;28(2):384-390. 2. Kalushkova A, Nylund P, Parraga AA, Lennartsson A, JernbergWiklund H. One omics approach does not rule them all: the metabolome and the epigenome join forces in haematological malignancies. Epigenomes. 2021;5(4):22. 3. Agarwal P, Alzrigat M, Parraga AA, et al. Genome-wide profiling of histone H3 lysine 27 and lysine 4 trimethylation in multiple myeloma reveals the importance of Polycomb gene targeting and highlights EZH2 as a potential therapeutic target. Oncotarget. 2016;7(6):6809-6823. 4. Walker BA, Boyle EM, Wardell CP, et al. Mutational spectrum, copy number changes, and outcome: results of a sequencing study of patients with newly diagnosed myeloma. J Clin Oncol. 2015;33(33):3911-3920. 5. De Smedt E, Lui H, Maes K, et al. The epigenome in multiple myeloma: impact on tumor cell plasticity and drug response. Front Oncol. 2018;8:566. 6. Tandon N, Ramakrishnan V, Kumar SK. Clinical use and applications of histone deacetylase inhibitors in multiple myeloma. Clin Pharmacol. 2016;8:35-44. 7. Kiziltepe T, Hideshima T, Catley L, et al. 5-Azacytidine, a DNA methyltransferase inhibitor, induces ATR-mediated DNA double-strand break responses, apoptosis, and synergistic cytotoxicity with doxorubicin and bortezomib against multiple myeloma cells. Mol Cancer Ther. 2007;6(6):1718-1727. 8. Nylund P, Atienza Parraga A, Haglof J, et al. A distinct metabolic response characterizes sensitivity to EZH2 inhibition in multiple myeloma. Cell Death Dis. 2021;12(2):167. 9. Croonquist PA, Van Ness B. The polycomb group protein enhancer of zeste homolog 2 (EZH 2) is an oncogene that

influences myeloma cell growth and the mutant ras phenotype. Oncogene. 2005;24(41):6269-6280. 10. Zhan F, Hardin J, Kordsmeier B, et al. Global gene expression profiling of multiple myeloma, monoclonal gammopathy of undetermined significance, and normal bone marrow plasma cells. Blood. 2002;99(5):1745-1757. 11. Kalushkova A, Fryknas M, Lemaire M, et al. Polycomb target genes are silenced in multiple myeloma. PLoS One. 2010;5(7):e11483. 12. Iyer MK, Niknafs YS, Malik R, et al. The landscape of long noncoding RNAs in the human transcriptome. Nat Genet. 2015;47(3):199-208. 13. Carlevaro-Fita J, Lanzos A, Feuerbach L, et al. Cancer lncRNA census reveals evidence for deep functional conservation of long noncoding RNAs in tumorigenesis. Commun Biol. 2020;3(1):56. 14. Long Y, Bolanos B, Gong L, et al. Conserved RNA-binding specificity of polycomb repressive complex 2 is achieved by dispersed amino acid patches in EZH2. Elife. 2017;6:e31558. 15. Wan L, Sun M, Liu GJ, et al. Long Noncoding RNA PVT1 promotes non-small cell lung cancer cell proliferation through epigenetically regulating LATS2 expression. Mol Cancer Ther. 2016;15(5):1082-1094. 16. Huang XM, Shi SS, Jian TM, Tang DR, Wu T, Sun FY. LncRNA PVT1 knockdown affects proliferation and apoptosis of uveal melanoma cells by inhibiting EZH2. Eur Rev Med Pharmacol Sci. 2019;23(7):2880-2887. 17. Guo J, Hao C, Wang C, Li L. Long noncoding RNA PVT1 modulates hepatocellular carcinoma cell proliferation and apoptosis by recruiting EZH2. Cancer Cell Int. 2018;18:98. 18. Sun Y, Ren D, Zhou Y, Shen J, Wu H, Jin X. Histone acetyltransferase 1 promotes gemcitabine resistance by regulating the PVT1/EZH2 complex in pancreatic cancer. Cell

Haematologica | 109 February 2024

576

ARTICLE - PVT1 is an interaction partner of PRC2 in MM

P. Nylund et al.

Death Dis. 2021;12(10):878. 19. Houshmand M, Yazdi N, Kazemi A, et al. Long non-coding RNA PVT1 as a novel candidate for targeted therapy in hematologic malignancies. Int J Biochem Cell Biol. 2018;98:54-64. 20. Handa H, Honma K, Oda T, et al. Long noncoding RNA PVT1 is regulated by bromodomain protein BRD4 in multiple myeloma and is associated with disease progression. Int J Mol Sci. 2020;21:19. 21. Saltarella I, Apollonio B, Lamanuzzi A, et al. The landscape of lncRNAs in multiple myeloma: implications in the “hallmarks of cancer”, clinical perspectives and therapeutic opportunities. Cancers (Basel). 2022;14(8):1963. 22. Carramusa L, Contino F, Ferro A, et al. The PVT-1 oncogene is a Myc protein target that is overexpressed in transformed cells. J Cell Physiol. 2007;213(2):511-518. 23. Foltz SM, Gao Q, Yoon CJ, et al. Evolution and structure of clinically relevant gene fusions in multiple myeloma. Nat Commun. 2020;11(1):2666. 24. Ewels PA, Peltzer A, Fillinger S, et al. The nf-core framework for community-curated bioinformatics pipelines. Nat Biotechnol. 2020;38(3):276-278. 25. Kopylova E, Noe L, Touzet H. SortMeRNA: fast and accurate filtering of ribosomal RNAs in metatranscriptomic data. Bioinformatics. 2012;28(24):3211-3217. 26. Li Y, Zhao DY, Greenblatt JF, Zhang Z. RIPSeeker: a statistical package for identifying protein-associated transcripts from RIPseq experiments. Nucleic Acids Res. 2013;41(8):e94. 27. Ishiguro K, Kitajima H, Niinuma T, et al. Dual EZH2 and G9a inhibition suppresses multiple myeloma cell proliferation by regulating the interferon signal and IRF4-MYC axis. Cell Death Discov. 2021;7(1):7. 28. Kaneko S, Li G, Son J, et al. Phosphorylation of the PRC2 component Ezh2 is cell cycle-regulated and up-regulates its binding to ncRNA. Genes Dev. 2010;24(23):2615-2620. 29. Zheng S, Cherniack AD, Dewal N, et al. Comprehensive pangenomic characterization of adrenocortical carcinoma. Cancer Cell. 2016;29(5):723-736. 30. Jiang B, Yang B, Wang Q, Zheng X, Guo Y, Lu W. lncRNA PVT1 promotes hepatitis B virus-positive liver cancer progression by disturbing histone methylation on the c-Myc promoter. Oncol Rep. 2020;43(2):718-726. 31. He S, Zhang H, Liu H, Zhu H. LongTarget: a tool to predict lncRNA DNA-binding motifs and binding sites via Hoogsteen base-pairing analysis. Bioinformatics. 2015;31(2):178-186. 32. Yu Y, Ouyang Y, Yao W. shinyCircos: an R/Shiny application for interactive creation of Circos plot. Bioinformatics. 2018;34(7):1229-1231. 33. Liu J, Yang L, Liu X, et al. lncRNA HOTTIP recruits EZH2 to inhibit PTEN expression and participates in IM resistance in chronic myeloid leukemia. Stem Cells Int. 2022;2022:9993393.

34. Zang X, Wang J, Xia Y, et al. LncRNA MEG3 promotes the sensitivity of bortezomib by inhibiting autophagy in multiple myeloma. Leuk Res. 2022;123:106967. 35. Ma T, Chen Y, Yi ZG, et al. NORAD promotes multiple myeloma cell progression via BMP6/P-ERK1/2 axis. Cell Signal. 2022;100:110474. 36. Zhang M, Zhao X, Cai X, Wang P, Yu M, Wei Z. Knockdown of long non-coding RNA plasmacytoma variant translocation 1 inhibits cell proliferation while promotes cell apoptosis via regulating miR-486-mediated CDK4 and BCAS2 in multiple myeloma. Ir J Med Sci. 2020;189(3):825-834. 37. El-Mesery M, Rosenthal T, Rauert-Wunderlich H, et al. The NEDD8-activating enzyme inhibitor MLN4924 sensitizes a TNFR1(+) subgroup of multiple myeloma cells for TNF-induced cell death. Cell Death Dis. 2019;10(8):611. 38. Park MS, Lee YE, Kim HR, et al. Phospholipase C beta 2 protein overexpression is a favorable prognostic indicator in newly diagnosed normal karyotype acute myeloid leukemia. Ann Lab Med. 2021;41(4):409-413. 39. Yang X, Wan M, Yu F, Wu X. Histone methyltransferase EZH2 epigenetically affects CCNA1 expression in acute myeloid leukemia. Cell Signal. 2021;87:110144. 40. Luo X, Xu L, Wu X, Tan H, Liu L. Decreased SATB1 expression promotes AML cell proliferation through NF-kappaB activation. Cancer Cell Int. 2019;19:134. 41. Starnes T, Rasila KK, Robertson MJ, et al. The chemokine CXCL14 (BRAK) stimulates activated NK cell migration: implications for the downregulation of CXCL14 in malignancy. Exp Hematol. 2006;34(8):1101-1105. 42. Tanegashima K, Suzuki K, Nakayama Y, et al. CXCL14 is a natural inhibitor of the CXCL12-CXCR4 signaling axis. FEBS Lett. 2013;587(12):1731-1735. 43. Jeon BN, Kim MK, Choi WI, et al. KR-POK interacts with p53 and represses its ability to activate transcription of p21WAF1/ CDKN1A. Cancer Res. 2012;72(5):1137-1148. 44. Han SS, Tompkins VS, Son DJ, et al. CDKN1A and FANCD2 are potential oncotargets in Burkitt lymphoma and multiple myeloma. Exp Hematol Oncol. 2015;4:9. 45. Jin X, Kim LJY, Wu Q, et al. Targeting glioma stem cells through combined BMI1 and EZH2 inhibition. Nat Med. 2017;23(11):1352-1361. 46. Zhang XM, Sheng SR, Wang XY, Bin LH, Wang JR, Li GY. Expression of tumor related gene NAG6 in gastric cancer and restriction fragment length polymorphism analysis. World J Gastroenterol. 2004;10(9):1361-1364. 47. Wei S, Shang H, Cao Y, Wang Q. The coiled-coil domain containing protein Ccdc136b antagonizes maternal Wnt/betacatenin activity during zebrafish dorsoventral axial patterning. J Genet Genomics. 2016;43(7):431-438.

Haematologica | 109 February 2024

577

ARTICLE - Plasma Cell DIsorders

TTK/MPS1 inhibitor OSU-13 targets the mitotic checkpoint and is a potential therapeutic strategy for myeloma Larissa Valle Guilhen Longo,1,2 Tiffany Hughes,1,2 Betina McNeil-Laidley,1,2 Francesca Cottini,1,2 Gerard Hilinski,3 Elizabeth Merritt2 and Don M. Benson Jr.1,2

Correspondence: D.M. Benson Don.Benson@osumc.edu

Division of Hematology, Department of Internal Medicine, Ohio State University;

Received: Accepted: Early view:

Comprehensive Cancer Center and James Cancer Hospital and Solove Research Institute

and 3Drug Development Institute, Comprehensive Cancer Center, James Cancer Hospital and Solove Research Institute, Columbus, OH, USA

January 27, 2023. July 20, 2023. July 27, 2023.

Abstract Despite substantial recent advances in treatment, multiple myeloma (MM) remains an incurable disease, with a shortage of treatment options for patients with high-risk disease, warranting the need for novel therapeutic targets and treatment approaches. Threonine and tyrosine kinase (TTK), also known as monopolar spindle 1 (MPS1), is a kinase essential for the mitotic spindle checkpoint whose expression correlates to unfavorable prognosis in several cancers. Here, we report the importance of TTK in MM, and the effects of the TTK inhibitor OSU-13. Elevated TTK expression correlated with amplification/gain of 1q21 and decreased overall and event-free survival in MM. Treatment with OSU-13 inhibited TTK activity efficiently and selectively at a similar concentration range to other TTK inhibitor clinical candidates. OSU-13 reduced proliferation and viability of primary human MM cells and cell lines, especially those with high 1q21 copy numbers, and triggered apoptosis through caspase 3 and 7 activation. In addition, OSU-13 induced DNA damage and severe defects in chromosome alignment and segregation, generating aneuploidy. In vivo, OSU-13 decreased tumor growth in mice with NCI-H929 xenografts. Collectively, our findings reveal that inhibiting TTK with OSU-13 is a potential therapeutic strategy for MM, particularly for a subset of high-risk patients with poor outcome.

Introduction Multiple myeloma (MM) is a plasma cell disorder that accounts for more than 10% of hematologic cancers.1 Despite the recent improvement in overall survival of patients with MM due to novel treatment options,2,3 MM is still a mostly incurable cancer,4 warranting the need for novel therapeutic targets, especially for patients with high-risk disease. The presence of genetic alterations is an important hallmark of cancer.5,6 In MM, structural and numerical genetic abnormalities are usually associated with disease development and progression.7 Aneuploidy is present in about 70% of MM cases.8 In addition, deletions, duplications, or translocations are very common events in MM,9,10 some of which - t(4;14)(p16;q32), t(14;16)(q32;q23), and del(17p) - are considered poor prognostic indicators.11 Possible mechanisms leading to genomic instability in MM include elevated homologous recombination activity, which causes an increased mutation rate and accumulation of genetic variation,12 and centrosome amplification, which is frequent

in MM and has been associated with high-risk disease and poor prognosis.13 Adequate chromosome segregation is essential for genomic stability and relies on a group of proteins known as the spindle assembly checkpoint (SAC).14 The SAC blocks cell cycle progression until the chromosomes are correctly attached to the spindle microtubules.15 Failure of this process can cause uneven chromosomal segregation and aneuploidy, resulting in chromosomal instability.16 Manipulation of SAC proteins leads to tumor formation in animal models.17,18 Accordingly, drugs that interfere with SAC, such as inhibitors of Aurora kinases, Polo-like kinases, CENP-E, and threonine and tyrosine kinase (TTK) have been investigated or are being tested in clinical trials for several cancers including MM.19-21 TTK, also known as monopolar spindle 1 (MPS1), participates in the regulation of the DNA damage checkpoint response,22 centrosome duplication,23 and mitosis arrest until proper chromosome alignment, playing an essential role in SAC.24 High levels of TTK expression correlate to unfavorable prognosis in several cancers,25-27 and TTK inhibition shows

Haematologica | 109 February 2024

578

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

potential utility for the treatment of glioma,28 melanoma,29 and colon,30 breast,31,32 lung,32 ovarian,29,32 and cervical33 cancers. Consequently, many clinical trials have tested TTK inhibitors - either as monotherapy or combinations - for the treatment of solid tumors, especially breast cancer. However, to date, there have been no clinical trials with TTK inhibitors involving patients with any kind of hematologic malignancies. In 2017, the TTK/MPS1 inhibitor OSU-13 was identified as a potential therapy for breast cancer.31 Here, we provide new data establishing the relevance of TTK expression in MM prognosis and perform the first comprehensive study using a TTK/MPS1 inhibitor as a therapeutic strategy for a hematologic malignancy.

Methods OSU-13 drug OSU-13 was provided by the Drug Development Institute of the Ohio State University and stored at -20°C at 10 mM in dimethyl sulfoxide (DMSO). Prognostic survival analysis Overall survival and event-free survival outcome studies were performed in 769 patients using the GSE2658 dataset. The median value of TTK levels was used as a cut-off to define patients with low (n=386) or high (n=383) TTK expression. Prognostic value of TTK was evaluated by the Kaplan-Meier curve, obtained from www.canevolve.org. TTK expression analysis in high-risk MM with specific genetic alterations (17p del, t(4;14), t(11;14), t(8;14), and 1q21 gain) was performed in the MMRF CoMMpass database. Viability, apoptosis and necroptosis assays For viability measurement, cells were plated at 100,000 cells/mL in medium containing DMSO or 0.5 μM OSU-13 and incubated for 24, 48, and 72 hours (h) at 37°C and 5% CO2. Afterwards, cells were stained with Zombie-aqua (Life Technologies; Carlsbad, CA, USA) according to manufacturer’s instructions. Cells were then analyzed using an Attune Nxt cytometer (Invitrogen; Waltham, MA, USA), and FlowJo software (FlowJo LLC; Ashland, OR, USA). For apoptosis and necroptosis inhibition assays, OPM-2 cells (250,000 cells/mL) were pre-incubated for 1 h at 37oC in medium containing 100 μM of the general caspase inhibitor Z-VAD-FMK (Sigma-Aldrich; St Louis, MO, USA) or the necroptosis inhibitor necrostatin-1s (Cell Signaling Inc.; Danvers, MA, USA). Then, cells were washed and incubated for 72 h in medium containing DMSO or 1 μM OSU-13, stained with Zombie-aqua (Life Technologies), and cell viability was assessed as previously described. Fluorescence microscopy of chromosome segregation OPM-2 cells (1x106) were synchronized at G1/S by double

L. Valle Guilhen Longo et al.

thymidine block. Cells were incubated for 16 h in medium containing 2 mM thymidine (Sigma-Aldrich), released for 8 h in fresh medium, and incubated in 2 mM thymidine for additional 16 h. Next, cells were washed, and cell cycle release was induced with 24 μM 2’-deoxycytidine (Sigma-Aldrich) plus DMSO or 1 μM OSU-13 for 9 h. Cells were resuspended in 100 μL phosphate-buffered saline (PBS), immobilized onto a microscope slide using a Cytospin (Thermo Fisher Scientific; Waltham, MA, USA), mounted with Prolong Glass Antifade Mountant with NucBlue (Thermo Fisher Scientific), and imaged using a Nikon DM5000 B microscope (Tokyo, Japan) equipped with fluorescence optics with a Leica X63 oil immersion objective (Wetzlar, Germany). Images were analyzed using ImageJ (NIH; Bethesda, MD, USA), and the percentage of anaphase and telophase cells with and without lagging chromosomes was manually calculated. Statistical considerations Unless specified otherwise, all data are presented as mean values ± standard deviation (SD) from independent experiments. Student’s t test was used to evaluate differences between conditions with P value <0.05 considered to be statistically significant. *P<0.05, **P<0.01, ***P<0.001 and ****P<0.0001. The correlation between sensitivity to OSU-13 and 1q21 copy number in the cell lines was analyzed using the non-parametric Spearman’s rank correlation coefficient. Additional methods Cell lines, primary cell experiments, small interfering RNA (siRNA) knockdown, real-time polymerase chain reaction (PCR), NanoBRET™, kinase profiling, co-crystallization, cell viability and proliferation assays, caspases 3 and 7 activity assay, western blotting, TUNEL, cell cycle analysis, metaphase chromosome spread, and studies in murine model are described in the Online Supplementary Appendix. Experiments involving human subject samples were conducted with the approval of The Ohio State University Institutional Review Board (IRB 2023C0065).

Results TTK expression correlates with multiple myeloma prognosis and plays a role in the survival of multiple myeloma cell lines High TTK expression correlates to unfavorable prognosis in several cancers.25-28,34 In order to determine the relevance of TTK expression in MM prognosis, we examined the GSE2658 dataset and the CoMMpass database, analyzing the association of TTK expression with clinical outcome and chromosomal alterations in MM patients (n=769). We found that elevated TTK expression correlated with decreased overall (P=0.0001) and event-free (P<0.0001) survival (Figure 1A, B). Furthermore, TTK expression was higher in patients

Haematologica | 109 February 2024

579

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

L. Valle Guilhen Longo et al.

with amplification or gain of 1q21, a genetic alteration found in high-risk MM (P<0.0001) (Figure 1C). No correlation was seen between TTK expression and other high-risk genetic alterations, such as deletion of 17p or t(4;14), t(8;14), or t(11;14) translocations (Online Supplementary Figure S1). We also compared TTK expression levels in eight human MM cell lines and primary CD138+ plasma cells isolated from bone marrow (BM) of four recently diagnosed, untreated MM patients (Figure 1D). MM cell lines exhibited higher TTK

expression than primary CD138+ cells. MM.1S showed the highest expression (16.2-fold higher than CD138+ cells), whereas L363 had the lowest expression (3.2-fold increase). In order to investigate the role of TTK in MM cell lines, we specifically knocked down TTK expression in OPM-2 and NCI-H929 using Alexa Fluor 647-conjugated non-targeting scrambled or TTK-specific siRNA (Online Supplementary Figure S2). Streptolysin was used to facilitate siRNA uptake, and cell viability was evaluated using Zombie Aqua

Figure 1 TTK expression correlates with multiple myeloma prognosis and varies in human multiple myeloma cell lines. (A) Overall and (B) event-free survival outcome studies from GSE2658 dataset. Median value of TTK levels were used as a cut-off to define “low TTK” and “high TTK” patients. Kaplan Meier curves were obtained from www.canevolve.org; ***P<0.001; ****P<0.0001. (C) Analysis from MMRF CoMMpass database of the TTK expression in multiple myeloma (MM) patients with (N=99) or without +1q21 (N=171); ****P<0.0001. (D) TTK expression in human MM cell lines. Values were normalized to the average of TTK expression in CD138+ cells from four patients with MM. FPKM: fragments per kilobase per million. Haematologica | 109 February 2024

580

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

staining. Knock-down of TTK was confirmed by quantitative PCR (data not shown). Our results exhibited a significant decrease in viability in TTK-siRNA+ cells compared to cells transfected with non-targeting scrambled siRNA in both cell lines, with a more pronounced effect in NCI-H929. Following treatment with 20 units (U) of streptolysin, 51±5.8% of TTK-siRNA+ NCI-H929 cells were dead (Zombie+), in contrast to only 16±6.2% of scrambled siRNA+ cells (P<0.001). These findings highlight the critical role of TTK in MM cell survival and suggest that targeting TTK could serve as a potential therapeutic strategy for MM. OSU-13 efficiently and selectively inhibits TTK In order to analyze the interaction between TTK and its novel small molecule inhibitor OSU-13, their co-crystal

L. Valle Guilhen Longo et al.

structure was obtained at a 2.37 Å resolution (Figure 2A). We determined that OSU-13 binds to the ATP-binding pocket, forming multiple interactions with the protein kinase hinge region, P-loop, and A-loop. OSU-13 inserts between Val539, Ala551 and Ile531 on one side and Leu654, Ile663 and Ile586 on the other, as well as the gatekeeper residue Met602 thiol deep in the pocket. The pyrrolopyrimidine bicyclic scaffold donates a H-bond to the Glu603 main chain carbonyl oxygen and accepts a H-bond from the Gly605 amide nitrogen. Additionally, OSU-13 forms a H-bond with the Gly605 main chain carbonyl oxygen and establishes van der Waals contacts with the ribose-binding pocket and the solvent-exposed channel. OSU-13’s activity was initially measured by a TTK target engagement assay (NanoBRET™). We compared its relative

Figure 2 OSU-13 interacts with TTK via hydrogen bonds and inhibits it in low doses and different ATP concentrations. (A) Co-crystal structure of the small molecule inhibitor OSU-13 interacting with TTK. The hydrogen bonds that anchor the adenine binding pocket of OSU-13 to the TTK ATP-binding region are represented by yellow dotted lines. Resolution =2.37 Å, R cryst =20.2, R free =23.3. (B) TTK target engagement assay (NanoBRET™) in HEK293 cells. Relative levels of OSU-13-mediated inhibition of TTK-NanoLuc binding to a fluorescent tracer was measured in comparison to other TTK inhibitors and the half maximal effective concentration (EC50) (nM) was calculated. Graph represents the mean of 2 independent experiments ± standard deviation. (C, D) ADP-Glo kinase assay comparing OSU-13 and BOS-172722 activities at physiologic (1 mM) and low (10 µM) ATP concentrations against the human and mouse TTK enzymes. Inhibition reaction was performed at room temperature for 30 minutes. The graphs show the mean of 2 independent experiments ± standard deviation. Haematologica | 109 February 2024

581

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

inhibition of TTK-NanoLuc binding to a fluorescent tracer with other reported TTK inhibitors. OSU-13 showed inhibition within the same concentration range as other compounds, with a half maximal effective concentration (EC50) =10 nM, nearly 10-fold lower than the clinical candidate CFI-402257 (EC50=94.3 nM) (Figure 2B). We also verified that OSU-13 inhibits both mouse and human TTK at physiologic (1 mM) and low (10 µM) ATP concentrations, demonstrating activity comparable to the clinical candidate BOS-172722 (Figure 2C, D). Additionally, OSU-13 showed high stability, with a half-life exceeding 90 minutes in human hepatocyte stability assays. In order to evaluate its selectivity, OSU-13 was profiled against a panel of human kinases in a cell-free kinase activity inhibition assay (Online Supplementary Table S1). Among the 399 kinases tested, only seven showed >80% inhibition when incubated with 1 μM OSU-13. Of these, only LRRK2 exhibited a half maximal inhibitory concentration (IC50) in a similar concentration range to TTK/MPS1 (7.5 nM vs. 4.3 nM), whereas the other kinases (ALK, IR, LTK, INSRR, and FAK) had IC50 values at least 36-fold higher than TTK/MPS1 (Table 1). Importantly, OSU-13 did not significantly inhibit other mitosis-related kinases, such as Aurora kinase, PLK, or Cyclin-dependent kinase family members. Overall, these results indicate that OSU-13 is a biologically stable molecule that selectively and efficiently inhibits TTK kinase activity, comparable to other TTK/MPS1 inhibitor clinical candidates. OSU-13 selectively decreases proliferation and viability of multiple myeloma cells We investigated the effects of OSU-13 on primary cells derived from the bone marrow of a MM patient and the peripheral blood mononuclear cells (PBMC) of a plasma cell leukemia (PCL) patient. Compared to DMSO, OSU-13 treatment for 72 h reduced viable CD138+ plasma cells from MM and PCL patients by 44% and 18%, respectively (Figure 3A). Only a small effect was observed in viable B cells (reduced by 37% in MM and 6% in PCL) and natural killer (NK) cells (reduced by 11% in MM and 1% in PCL). No significant effect was observed in other lymphocyte populations (Online Supplementary Figure S3A). Similarly, PBMC from three healthy donors showed only a 22% decrease in viable B cells in one of the three donors after OSU-13 treatment (Online Supplementary Figure S3B). These findings suggest that OSU-13 displays relative selectivity for plasma cells, with mild effects on B-cell lineage lymphocytes and no substantial adverse effects on other PBMC populations. In order to study OSU-13’s effects on human MM cell lines, we evaluated proliferation and viability. Cells were treated with various concentrations of OSU-13 or DMSO, and cell division was monitored using a dye dilution experiment (Figure 3B). OSU-13 decreased proliferation of OPM-2 and NCI-H929 cell lines in a dose-dependent manner. At 5 µM OSU-13, proliferation decreased to 54±10% and 49±4%

L. Valle Guilhen Longo et al.

Table 1. OSU-13 inhibitor IC50 against select kinases.* Kinase

OSU-13 IC50 (nM)

TTK/MPS1

4.3

LRRK2

7.5

ALK

156.0

262.0

TYK1/LTK

313.0

INSRR

424.0

FAK/PTK2

1009.0

OSU-13 half maximal inhibitory concentration (IC50) was calculated against the kinases previously inhibited ≥80% at 1 µM OSU-13 during the profiling test (Online Supplementary Table S1). Values represent the average of 2 independent experiments. *

Table 2. Cytotoxic effect of OSU-13 against human multiple myeloma cell lines with different 1q21 copy numbers.* Cell line

OSU-13 IC50 (nM)

1q21 copy number

EJM

5,870±1,125

U266

7,636±791

MM.1S

1,106±188

RPMI-8226

10,448±1184

NCI-H929

867±52

JJN3

1,051±127

L363

132±1

OPM-2

643±109

KMS11

309±18

OSU-13 half maximal inhibitory concentration (IC50) values for each cell line represent the mean ± standard deviation from 3 independent MTS assay experiments. Each experiment was performed in triplicate or quadruplicate, and the mean of the replicates was used to generate the curve and calculate the IC50. The 1q21 copy numbers were previously determined by fluorescense in situ hybridization.51 *

compared to DMSO control in OPM-2 and NCI-H929, respectively (P<0.01). RPMI-8226 and U266 showed a less pronounced effect on proliferation (data not shown). The cytotoxic potential of OSU-13 was evaluated via MTS assay in nine human MM cell lines with different 1q21 copy numbers (Table 2). L363 was the most sensitive (IC50=132±1 nM) cell line, whereas RPMI-8226 was the least sensitive (IC50=10,448±1,184 nM) to OSU-13. Further analysis revealed a significant inverse correlation (rs=-0.76; P=0.017) between the number of 1q21 copies and the in vitro IC50 of OSU-13. MM cell lines with a higher number of 1q21 copies exhibited increased sensitivity to OSU-13. This finding suggests that OSU-13, through TTK inhibition, is particularly effective in a subgroup of high-risk MM patients with a poor prognosis. Next, we assessed viability of NCI-H929 and OPM-2 cells using Zombie staining (Figure 3C). DMSO-treated cells

Haematologica | 109 February 2024

582

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

showed an increase in viable cell numbers over time, while OSU-13-treated cells remained constant. After 72 h, the quantity of viable DMSO-treated cells was 5-fold and 3.6-fold higher (P<0.05) than OSU-13-treated cells in OPM-2 and NCI-H929, respectively (Figure 3C, left). The percentage of dead cells (Zombie+) peaked after 72 h, reaching 35±4.5% of OPM-2 (P<0.01) and 21±6.3% of NCI-H929 cells (P<0.05). Meanwhile, DMSO-treated cells did not show increase in Zombie+ cells over time (Figure 3C, right). Moreover, some

L. Valle Guilhen Longo et al.

dead cells turned into debris, resulting in a smaller Zombie+ population than expected. OSU-13 induces apoptosis in human multiple myeloma cell lines through caspase activation In order to investigate whether OSU-13-induced cell death was apoptotic, caspase 3 and 7 activity was assessed in NCI-H929 cells at different time points after treatment with varying concentrations of OSU-13. Caspase 3 and 7

Continued on following page.

Haematologica | 109 February 2024

583

L. Valle Guilhen Longo et al.

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

Figure 3. OSU-13 decreases viability, induces apoptosis, and causes DNA damage in human multiple myeloma cells. (A) Primary cells derived from the bone marrow of a multiple myeloma (MM) patient and the peripheral blood mononuclear cells of a plasma cell leukemia (PCL) patient were cultured in vitro at 37oC in dimethyl sulfoxide (DMSO) ±0.5 μM of OSU-13 for 24, 48, and 72 hours (h). The graph shows the comparative viability of CD138+ plasma cells (Zombie-) treated with OSU-13 relative to DMSO. (B) Relative proliferation of OPM-2 (left) and NCI-H929 (right) MM cell lines after incubation with 0, 0.1, 0.65, 1.25, 2.5, and 5 µM of OSU-13 for 72 h. The average of DMSO control data (concentration 0) was set to 100%. Proliferation was calculated using Cell Trace Far Red. Data shown are the mean of 3 independent experiments ± standard deviation (SD); *P<0.05; **P<0.01 in relation to DMSO. (C) OPM2 (left) and NCI-H929 (right) cells were incubated in DMSO or 0.5 µM OSU-13 for 24, 48, and 72 h, and assessed for viability through Zombie staining. Viable cell counts (Zombie-; left panel) and percentage of dead cells (Zombie+; right panel) were analyzed by flow cytometry. Data represent the mean of 3 independent experiments ± SD; *P<0.05; **P<0.01. (D) Activity of caspases 3 and 7 in NCI-H929 after 24, 48, and 72 h treatment with 0, 0.1, and 0.5 µM OSU-13 assessed by ApoTox-Glo™Triplex assay. Data represent the mean ± SD (N=3); *P<0.05; **P<0.01; ****P<0.0001. (E) Western blot analysis of caspase 3, cleaved caspase 3, caspase 7, and cleaved caspase 7 in lysates from OPM-2 (left) and NCI-H929 (right) cells incubated in DMSO or 0.5 µM OSU-13 for 72 h. α-tubulin was used as loading control. Images depict a representative experiment from 3 independent biological replicates. (F) Effect of Z-VAD-FMK in OSU-13-induced apoptosis. OPM-2 cells were pre-incubated for 1 h with 100 μM Z-VAD-FMK or DMSO, washed with phosphate-buffered saline, and incubated for 72 h in 1 μM OSU-13. Cell viability was assessed with Zombie-aqua dye staining by flow cytometry. Data represent the mean of 3 independent experiments ± SD; *P<0.05. (G) TUNEL staining analysis of OPM2 and NCI-H929 cells treated with DMSO or OSU-13 (2.5, 5, and 10 μM) for 48 h. Representative fluorescence microscopy images from a spinning-disk confocal system (UltraVIEW) on a Nikon Ti-E microscope show TUNEL+ staining indicating DNA damage in OPM-2 cells. Scale bars, 20 μm. (H) TUNEL images were manually quantified in ImageJ and TUNEL positivity rate (%) was calculated. Results are representative of 2 independent experiments for each cell line; *P<0.05. (I) Western blot analysis of cleaved PARP1 and p-H2AX in lysates from OPM-2 and NCI-H929 cells incubated in DMSO or 0.5 μM OSU-13 for 72 h. GAPDH was used as loading control. Images depict a representative experiment from 4 independent experiments. RLU: relative luminescence unit.

activation started at 24 h with 0.1 µM (P<0.05) and peaked after 72 h with 0.5 µM OSU-13 (P<0.0001) (Figure 3D). Accordingly, western blot analysis of cell lysates showed increased levels of cleaved caspases 3 and 7 in OPM-2 and NCI-H929 cells after 72 h of OSU-13 treatment (Figure 3E), and pretreatment with the pan-caspase inhibitor Z-VADFMK decreased OSU-13-mediated cell death by 20±6.7% in OPM-2 cells (Figure 3F). In contrast, pretreatment with the necroptosis inhibitor Necrostatin-1s had no measurable effect on cell death, and there was no increase in the autophagy marker LC3B following OSU-13 treatment (Online

Supplementary Figure S4A, B, respectively). Collectively, these results indicate that OSU-13 triggers cell death in MM cell lines through caspase activation, and that apoptosis - but not necroptosis or autophagy - is partially responsible for this OSU-13-mediated cell death. OSU-13 induces DNA damage in multiple myeloma cell lines Apoptosis has been reported to induce DNA fragmentation. In order to measure internucleosomal DNA degradation, we treated OPM-2 and NCI-H929 cells with various concen-

Haematologica | 109 February 2024

584

L. Valle Guilhen Longo et al.

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

Figure 4. OSU-13 causes cell cycle and chromosome segregation abnormalities in OPM-2 cells. (A) DNA content analysis of OPM2 cells treated with dimethyl sulfoxide (DMSO) (red) or 1 μM OSU-13 (blue) for 72 hours (h). After treatment, DNA was stained with the intercalating agent propidium iodide (PI) and analyzed by flow cytometry. 2N and 4N populations are indicated. Data are representative of 4 independent experiments. (B) Graphical representation of the cell cycle analysis from data depicted in panel (A). Analysis was performed by the Cell Cycle tool in FlowJo_V10 software using the Watson model. Results are mean ± standard deviation of 4 independent experimental replicates; *P<0.05. (C, D) Chromosome spread analysis of OPM-2 cells treated with Continued on following page.

Haematologica | 109 February 2024

585

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

L. Valle Guilhen Longo et al.

DMSO or 1 μM OSU-13 for 24 h. Slides were mounted in Prolong Glass Antifade Mountant with NucBlue to stain the chromosomes and imaged in a Nikon DM5000 B microscope. (C) representative fluorescence microscopy images depict metaphase spreads. Scale bars, 10 μm. (D) graph shows distribution of the cell population according to the number of chromosomes, manually quantified in ImageJ. Results are representative of 3 independent experiments; *P<0.05; **P<0.01. (E, F) Analysis of chromosome segregation during metaphase/anaphase. OPM-2 cells were synchronized by double thymidine block, released, and treated (1 μM OSU-13 or DMSO) for 9 h. Samples were mounted with NucBlue to stain the chromosomes. (E) representative fluorescence images from cells during chromosome alignment and segregation. Scale bars, 10 μm. (F) graph shows the mean percentage of cells with lagging chromosomes ± standard deviation from 2 independent experiments; **P<0.01.

trations of OSU-13 for 48 h and performed TUNEL analysis. DNA damage increased in a dose-dependent manner in both cell lines, as indicated by the correlation between OSU-13 concentration and percentage of TUNEL+ cells (Figure 3G, H). In OPM-2 cells, TUNEL positivity rates were 22±5.4%, 25±3.6% (P<0.05), and 43±3.1% (P<0.05) with 2.5 µM, 5 µM, and 10 µM OSU-13 treatment, respectively, compared to DMSO control (TUNEL+ =7±0.4%). In NCI-H929, TUNEL positivity increased with 2.5 µM OSU-13 treatment (P<0.05) (Figure 3H). Western blot analysis confirmed the presence of DNA damage, showing increased levels of p-H2AX and cleaved PARP1 after 72 hours of OSU-13 treatment (Figure 3I). OSU-13 interferes in the cell cycle and ploidy of multiple myeloma cell lines TTK/MPS1 inhibitors affect the cell cycle in a variety of cancer cells.30,33,35 In our study, treatment with OSU-13 induced significant alterations in DNA content in OPM-2 cells, with an increase in tetraploid (4N) and hyperploid (>4N) cells (Figure 4A). Accordingly, flow cytometric analysis revealed a substantial increase in the proportion of OPM-2 cells in the G2 phase (4N) after OSU-13 treatment, with a 1.9-fold increase compared to DMSO-treated cells (P<0.05). Additionally, there was a higher proportion of cells in the >G2 phase (>4N), although this difference was not statistically significant (Figure 4B). This increase in the hyperploid population was confirmed by fluorescence microscopy of OPM-2 chromosome spreads (Figure 4C, D). There was a higher quantity of chromosomes (Figure 4C), with a significant increase in cells with more than 91 chromosomes (13.2±2.9% in OSU-13 vs. 0.9±0.5% in DMSO; P<0.01). In addition, cells with 81-90 chromosomes also increased (9.7±0.8% in OSU-13 vs. 4.4±2.7% in DMSO; P<0.05) (Figure 4D). Similar effects were observed in NCI-H929 cells (Online Supplementary Figure S5A-C). Chromosome segregation was analyzed by microscopy to understand the mechanism behind this increase in the hyperploid population. OPM-2 cells treated with OSU-13 exhibited severe defects in chromosome alignment and segregation (Figure 4E), with 86.6±7.7% of the cells (P<0.01) presenting lagging chromosomes during anaphase (Figure 4F). OSU-13 has in vivo effect against multiple myeloma in mice The therapeutic potential of OSU-13 was evaluated in an NCI-H929 subcutaneous MM xenograft model in immuno-

deficient CB.17 SCID mice (Figure 5). Mice were treated daily per oral gavage starting 14 days after tumor implantation (day -14). Vehicle control or 10 mg/kg dose of OSU-13 were administered for 21 days, with a dosing holiday on days 14-17 due to unexpected weight loss in the OSU-13-treated group (Figure 5A). Treatment with OSU-13 produced a significant delay in tumor growth (Figure 5B). On average, vehicle-treated mice reached the endpoint tumor volume (≥2,000 mm3) by day 18, whereas OSU-13-treated mice only achieved this size on day 25. At day 18, 50% of vehicle-treated mice reached the threshold, compared to only 10% of OSU-13-treated mice (P<0.05). In addition, the average tumor volume in the OSU-13-treated group was half of the vehicle-treated group (1,084 mm3 vs. 1,980 mm3). Treatment ended on day 21, justifying the increased tumor growth after this date. Altogether, treatment with OSU-13 resulted in a 22% tumor growth delay (P<0.05) and showed superior efficacy compared to the clinical candidate CFI402257 in a parallel experiment (data not shown). In another MM xenograft model using NOD SCID mice (Figure 5C), OSU-13 treatment also led to a significant delay in tumor growth. By the end of the treatment period (day 14), the average tumor volume in the OSU-13-treated group was 453 mm3, significantly lower than the control group (681 mm3; P<0.05) (Figure 5D). In addition, western blot analysis of tumor lysates showed increased expression of the apoptosis markers cleaved-caspase 3 and p-H2AX in the OSU-13-treated group (Figure 5E, F). These findings demonstrate that OSU-13 effectively reduces MM tumor burden and induces apoptosis in MM cells in vivo.

Discussion TTK/Mps1 plays an essential role in SAC,24 and high levels of TTK expression correlate to unfavorable prognosis in several cancers.25-28,34 Accordingly, TTK inhibition has been explored in solid tumors as a therapeutic strategy to halt cell cycle arrest and induce genomic instability.28,30-33 However, the impact of TTK expression and TTK/MPS1 inhibition in hematologic cancers remains poorly understood.36,37 Here, we show that elevated TTK expression correlates with reduced overall and event-free survival along with amplification or gain of 1q21, found in high-risk MM. Indeed, TTK expression is significantly increased in high-risk compared to low-risk MM,36 and TTK is included in the MM Kinome

Haematologica | 109 February 2024

586

L. Valle Guilhen Longo et al.

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

Figure 5 OSU-13 shows therapeutic potential in mouse model. (A, B) Human multiple myeloma (MM) NCI-H929 cells were subcutaneously inoculated in immunodeficient CB.17 SCID mice. Fourteen days post-implantation, mice were treated orally daily with the vehicle control or 10 mg/kg of OSU-13 for 21 days, with a dosing holiday from days 14-17 due to weight loss. (B) Graphical representation of the mean ± standard error the mean tumor volume of the vehicle- (blue) and OSU-13-treated (red) groups over time (N=10 mice per group). The black dashed line indicates the 2,000 mm3 tumor volume threshold, the blue dashed line highlights the 18th day of experiment, and the red dashed line stresses the day when treatment was stopped; *P<0.05. The number of animals remaining in the study (tumor volume <2,000 mm3) are indicated for the vehicle- (blue) and OSU-13-treated (red) groups. (C) NCI-H929 cells were subcutaneously injected into immunodeficient NOD SCID mice. Thirteen days after implantation, the mice were subjected to daily oral treatment with either the vehicle control or OSU-13 (10 mg/kg on days 1-7 and 5 mg/kg on days 8-14). After 24 hours of treatment, the mice were euthanized, and the tumors were collected for further analysis. (D) Graphical representation of the mean ± standard error the mean tumor volume of the vehicle- (blue) and OSU-13-treated (red) groups over time (N=5 mice per group); *P<0.05. (E) Western blot analysis of tumor lysates from the experiment described in (C). Expression of cleaved caspase 3, cleaved PARP1 and p-H2AX are shown with GAPDH as a loading control. Individual results from each mouse are shown in the images. (F) Quantification of the western blot analysis depicted in panel (E). Results are mean ± standard deviation; *P<0.05; **P<0.01. Haematologica | 109 February 2024

587

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

Index, a gene expression profile risk score that predicts poor prognosis in MM.36 We report the first comprehensive analysis of a TTK/MPS1 inhibitor, OSU-13, in MM. Co-crystal structures reveal its binding to the ATP-binding pocket of TTK similarly to other TTK/MPS1 inhibitors,33,38 suggesting a conserved mechanism of inhibition. OSU-13 is stable and selectively inhibits TTK, showing comparable activity to the TTK/MPS1 inhibitor clinical candidates CFI-402257 and BOS-172722, which reinforces its clinical potential. OSU-13 exhibited similar effects in MM as previously observed in breast cancer31 or with other TTK inhibitors in several cancers.29,30,32,33 First, OSU-13 showed cytotoxicity and anti-proliferative activity across MM cell lines. Sensitivity to OSU-13 varied among cell lines and correlated with 1q21 copy number. Nonetheless, sensitivity to TTK/MPS1 inhibitors seems to be multifactorial as observed in other cancer types,29,35 and identifying additional predictive biomarkers would further define patients who might benefit from TTK/MPS1 inhibitor-based therapies, expanding the potential applications of these treatments. Second, OSU-13 induced apoptosis in MM cells, evidenced by caspase 3 and 7 activation and DNA damage, indicated by PARP1 cleavage, phosphorylation of H2AX, and TUNEL assay. However, partial rescue by the caspase inhibitor Z-VAD-FMK suggested involvement of caspase-independent mechanisms in OSU-13-induced cell death. Unlike the TTK/MPS1 inhibitor reversine that triggers autophagy by increasing LC3-B and Beclin 1 via the AKT pathway in cholangiocarcinoma cells,39 it seems that OSU-13 does not induce autophagy or necroptosis in MM cell lines. Third, OSU-13 triggered abnormal mitosis, leading to chromosome missegregation, aneuploidy, and elevated chromosome numbers. Comparable effects have been reported in hematologic cancers with the TTK/MPS1 inhibitor AZ3146, inducing chromosome instability and DNA damage in MM36 and acute myeloid leukemia.37 Lastly, OSU-13 achieved significant tumor growth delay in NCI-H929 xenografts in two immunodeficient mouse models, despite a 3-day treatment interruption and a dose reduction due to weight loss. In a breast cancer mouse model, OSU-13 previously reduced tumor growth without affecting body weight.31 However, the aforementioned study used athymic nude mice, potentially less sensitive to OSU13 than the CB.17 SCID and NOD SCID mice we used. OSU-13 showed modest efficacy in our experiment as a single agent. Combination strategies for TTK/MPS1 inhibitors have been explored in various cancer models. Combination with the anti-programmed cell death 1 (PD-1) antibody was more effective than monotherapy in a colon cancer model.35 TTK inhibition also sensitized cells to paclitaxel treatment in colon,33 breast, and lung cancers.32 Accordingly, clinical trials with CFI-402257 are currently investigating combination therapies with paclitaxel and fulvestrant for breast cancer treatment.

L. Valle Guilhen Longo et al.

Standard-of-care therapies for MM involve combinations of different agents to enhance response and prevent resistance. However, the combination strategies used for TTK inhibitors in solid tumors are unlikely to succeed in MM. Clinical trials with PD-1/PD-L1 inhibitors have shown limited results in MM - probably due to the immunosuppressive environment in MM - and have been halted due to severe adverse effects.40 Similarly, studies with the albumin-bound formulation Nab-paclitaxel41 did not achieve sustained responses in MM.42 In acute myeloid leukemia, the TTK inhibitor AZ3146 induced robust upregulation of the interferon gene,37 which has both direct anti-cancer effects43,44 and immune-activation potential.45-47 It would be interesting to investigate whether TTK inhibition triggers similar immunological effects in MM, potentially enhancing the effect of immunomodulatory drugs. In addition, MM cells may be sensitized to TTK inhibition by DNA-damaging agents such as proteasome inhibitors, alkylating agents, or experimental drugs that induce DNA double-strand breaks.48-50 Indeed, our initial findings suggest synergistic effects of OSU-13 with melphalan, a DNA-damaging agent. Identifying optimal combination strategies for OSU-13 will provide maximum efficacy and tolerability in MM treatment, providing new treatment options for relapsed/refractory MM patients. In summary, our findings emphasize the importance of TTK in MM prognosis and show that OSU-13-mediated TTK inhibition induces cell death in MM cell lines and reduces tumor growth in vivo. We propose that inhibition of TTK using OSU-13 is an effective approach for treating multiple myeloma, particularly in a significant subgroup of highrisk patients with poor prognosis. Importantly, this is the first time that TTK inhibition has been comprehensively explored as a potential therapeutic strategy for a hematologic malignancy. Disclosures No conflicts of interest to disclose. Contributions LVGL, BM-L, FC, TH and GH conceptualized and designed experiments, validated and analyzed data. LVGL, BM-L, TH and FC performed experiments. LVGL, TH and FC designed and created figures for the manuscript. LVGL wrote the manuscript, and FC, TH, GH and EM reviewed and edited the manuscript. DB advised on experimental design and data analysis, supervised, and provided resources for the study. Acknowledgments The authors would like to thank the OSU Comprehensive Cancer Center Leukemia Tissue Bank, the OSUCCC support (core) grant (CCSG) 2P30CA016058-45, Dr. Jian-Qiu Wu and the OSU Molecular Genetics Department for the use of the fluorescence and confocal microscopes, and the OSU

Haematologica | 109 February 2024

588

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma

Comprehensive Cancer Center Drug Development Institute for providing the OSU-13 drug and collaborating with experiments and discussions. Funding This work was supported by OSU divisional funds (to DB and FC), the Pelotonia Foundation (to FC and BM), the Paula and Rodger Riney Foundation (to DB and FC), the Elsa U. Pardee

L. Valle Guilhen Longo et al.

Foundation (to FC), the International Myeloma Society (to FC), the National Cancer Institute (1K08CA263476-01A1 to FC) and the OSU Comprehensive Cancer Center Drug Development Institute. Data-sharing statement The data are available from the corresponding author upon request.

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin. 2019;69(1):7-34. 2. Attal M, Lauwers-Cances V, Hulin C, et al. Lenalidomide, bortezomib, and dexamethasone with transplantation for myeloma. N Engl J Med. 2017;376(14):1311-1320. 3. Mikkilineni L, Kochenderfer JN. Chimeric antigen receptor T-cell therapies for multiple myeloma. Blood. 2017;130(24):2594-2602. 4. Robak P, Drozdz I, Szemraj J, Robak T. Drug resistance in multiple myeloma. Cancer Treat Rev. 2018;70:199-208. 5. Bharadwaj R, Yu H. The spindle checkpoint, aneuploidy, and cancer. Oncogene. 2004;23(11):2016-2027. 6. Kops GJ, Weaver BA, Cleveland DW. On the road to cancer: aneuploidy and the mitotic checkpoint. Nat Rev Cancer. 2005;5(10):773-785. 7. Neuse CJ, Lomas OC, Schliemann C, et al. Genome instability in multiple myeloma. Leukemia. 2020;34(11):2887-2897. 8. Drach J, Schuster J, Nowotny H, et al. Multiple-myeloma - highincidence of chromosomal aneuploidy as detected by interphase fluorescence in-situ hybridization. Cancer Res. 1995;55(17):3854-3859. 9. Bergsagel PL, Kuehl WM. Chromosome translocations in multiple myeloma. Oncogene. 2001;20(40):5611-5622. 10. Bergsagel PL, Kuehl WM. Molecular pathogenesis and a consequent classification of multiple myeloma. J Clin Oncol. 2005;23(26):6333-6338. 11. Sawyer JR. The prognostic significance of cytogenetics and molecular profiling in multiple myeloma. Cancer Genet. 2011;204(1):3-12. 12. Shammas MA, Shmookler Reis RJ, Koley H, Batchu RB, Li C, Munshi NC. Dysfunctional homologous recombination mediates genomic instability and progression in myeloma. Blood. 2009;113(10):2290-2297. 13. Chng WJ, Fonseca R. Centrosomes and myeloma; aneuploidy and proliferation. Environ Mol Mutagen. 2009;50(8):697-707. 14. Lara-Gonzalez P, Westhorpe FG, Taylor SS. The spindle assembly checkpoint. Curr Biol. 2012;22(22):R966-980. 15. Musacchio A, Salmon ED. The spindle-assembly checkpoint in space and time. Nat Rev Mol Cell Biol. 2007;8(5):379-393. 16. Thompson SL, Bakhoum SF, Compton DA. Mechanisms of chromosomal instability. Curr Biol. 2010;20(6):R285-295. 17. Diaz-Rodriguez E, Sotillo R, Schvartzman JM, Benezra R. Hec1 overexpression hyperactivates the mitotic checkpoint and induces tumor formation in vivo. Proc Natl Acad Sci U S A. 2008;105(43):16719-16724. 18. Sotillo R, Hernando E, Diaz-Rodriguez E, et al. Mad2 overexpression promotes aneuploidy and tumorigenesis in mice. Cancer Cell. 2007;11(1):9-23. 19. Borisa AC, Bhatt HG. A comprehensive review on aurora kinase:

small molecule inhibitors and clinical trial studies. Eur J Med Chem. 2017;140:1-19. 20. Talati C, Griffiths EA, Wetzler M, Wang ES. Polo-like kinase inhibitors in hematologic malignancies. Crit Rev Oncol Hematol. 2016;98:200-210. 21. Chung V, Heath EI, Schelman WR, et al. First-time-in-human study of GSK923295, a novel antimitotic inhibitor of centromere-associated protein E (CENP-E), in patients with refractory cancer. Cancer Chemother Pharmacol. 2012;69(3):733-741. 22. Wei JH, Chou YF, Ou YH, et al. TTK/hMps1 participates in the regulation of DNA damage checkpoint response by phosphorylating CHK2 on threonine 68. J Biol Chem. 2005;280(9):7748-7757. 23. Fisk HA, Mattison CP, Winey M. Human Mps1 protein kinase is required for centrosome duplication and normal mitotic progression. Proc Natl Acad Sci U S A. 2003;100(25):14875-14880. 24. Liu X, Winey M. The MPS1 family of protein kinases. Annu Rev Biochem. 2012;81:561-585. 25. Slee RB, Grimes BR, Bansal R, et al. Selective inhibition of pancreatic ductal adenocarcinoma cell growth by the mitotic MPS1 kinase inhibitor NMS-P715. Mol Cancer Ther. 2014;13(2):307-315. 26. Zhang L, Jiang B, Zhu N, et al. Mitotic checkpoint kinase Mps1/ TTK predicts prognosis of colon cancer patients and regulates tumor proliferation and differentiation via PKCalpha/ERK1/2 and PI3K/Akt pathway. Med Oncol. 2019;37(1):5. 27. Al-Ejeh F, Simpson PT, Saunus JM, et al. Meta-analysis of the global gene expression profile of triple-negative breast cancer identifies genes for the prognostication and treatment of aggressive breast cancer. Oncogenesis. 2014;3:e124. 28. Tannous BA, Kerami M, Van der Stoop PM, et al. Effects of the selective MPS1 inhibitor MPS1-IN-3 on glioblastoma sensitivity to antimitotic drugs. J Natl Cancer Inst. 2013;105(17):1322-1331. 29. Colombo R, Caldarelli M, Mennecozzi M, et al. Targeting the mitotic checkpoint for cancer therapy with NMS-P715, an inhibitor of MPS1 kinase. Cancer Res. 2010;70(24):10255-10264. 30. Tardif KD, Rogers A, Cassiano J, et al. Characterization of the cellular and antitumor effects of MPI-0479605, a smallmolecule inhibitor of the mitotic kinase Mps1. Mol Cancer Ther. 2011;10(12):2267-2275. 31. Sugimoto Y, Sawant DB, Fisk HA, et al. Novel pyrrolopyrimidines as Mps1/TTK kinase inhibitors for breast cancer. Bioorg Med Chem. 2017;25(7):2156-2166. 32. Wengner AM, Siemeister G, Koppitz M, et al. Novel Mps1 kinase inhibitors with potent antitumor activity. Mol Cancer Ther. 2016;15(4):583-592. 33. Jemaa M, Galluzzi L, Kepp O, et al. Characterization of novel

Haematologica | 109 February 2024

589

ARTICLE - OSU-13 targets the mitotic checkpoint in myeloma MPS1 inhibitors with preclinical anticancer activity. Cell Death Differ. 2013;20(11):1532-1545. 34. Choi M, Min YH, Pyo J, Lee CW, Jang CY, Kim JE. TC Mps1 12, a novel Mps1 inhibitor, suppresses the growth of hepatocellular carcinoma cells via the accumulation of chromosomal instability. Br J Pharmacol. 2017;174(12):1810-1825. 35. Mason JM, Wei X, Fletcher GC, et al. Functional characterization of CFI-402257, a potent and selective Mps1/TTK kinase inhibitor, for the treatment of cancer. Proc Natl Acad Sci U S A. 2017;114(12):3127-3132. 36. de Boussac H, Bruyer A, Jourdan M, et al. Kinome expression profiling to target new therapeutic avenues in multiple myeloma. Haematologica. 2020;105(3):784-795. 37. Jin N, Lera RF, Yan RE, et al. Chromosomal instability upregulates interferon in acute myeloid leukemia. Genes Chromosomes Cancer. 2020;59(11):627-638. 38. Kwiatkowski N, Jelluma N, Filippakopoulos P, et al. Smallmolecule kinase inhibitors provide insight into Mps1 cell cycle function. Nat Chem Biol. 2010;6(5):359-368. 39. Prajumwongs P, Waenphimai O, Vaeteewoottacharn K, Wongkham S, Sawanyawisuth K. Reversine, a selective MPS1 inhibitor, induced autophagic cell death via diminished glucose uptake and ATP production in cholangiocarcinoma cells. PeerJ. 2021;9:e10637. 40. Oliva S, Troia R, D’Agostino M, Boccadoro M, Gay F. Promises and pitfalls in the use of PD-1/PD-L1 inhibitors in multiple myeloma. Front Immunol. 2018;9:2749. 41. Desai N, Trieu V, Yao Z, et al. Increased antitumor activity, intratumor paclitaxel concentrations, and endothelial cell transport of cremophor-free, albumin-bound paclitaxel, ABI007, compared with cremophor-based paclitaxel. Clin Cancer Res. 2006;12(4):1317-1324. 42. Jain T, Dueck AC, Kosiorek HE, et al. Phase II trial of nab-

paclitaxel in patients with relapsed or refractory multiple myeloma. Am J Hematol. 2016;91(12):E504-E505. 43. Takaoka A, Hayakawa S, Yanai H, et al. Integration of interferonalpha/beta signalling to p53 responses in tumour suppression and antiviral defence. Nature. 2003;424(6948):516-523. 44. Thyrell L, Erickson S, Zhivotovsky B, et al. Mechanisms of interferon-alpha induced apoptosis in malignant cells. Oncogene. 2002;21(8):1251-1262. 45. Prchal M, Pilz A, Simma O, et al. Type I interferons as mediators of immune adjuvants for T- and B cell-dependent acquired immunity. Vaccine. 2009;27 Suppl 6:G17-20. 46. Tough DF, Kamath AT. Interferon with dendritic cells? Nat Immunol. 2001;2(12):1098-1100. 47. Tudor D, Riffault S, Carrat C, Lefevre F, Bernoin M, Charley B. Type I IFN modulates the immune response induced by DNA vaccination to pseudorabies virus glycoprotein C. Virology. 2001;286(1):197-205. 48. Bergsagel DE, Sprague CC, Austin C, Griffith KM. Evaluation of new chemotherapeutic agents in the treatment of multiple myeloma. IV. L-Phenylalanine mustard (NSC-8806). Cancer Chemother Rep. 1962;21:87-99. 49. Cottini F, Hideshima T, Suzuki R, et al. Synthetic lethal approaches exploiting DNA damage in aggressive myeloma. Cancer Discov. 2015;5(9):972-987. 50. Neri P, Ren L, Gratton K, et al. Bortezomib-induced “BRCAness” sensitizes multiple myeloma cells to PARP inhibitors. Blood. 2011;118(24):6368-6379. 51. Hanamura I, Stewart JP, Huang Y, et al. Frequent gain of chromosome band 1q21 in plasma-cell dyscrasias detected by fluorescence in situ hybridization: incidence increases from MGUS to relapsed myeloma and is related to prognosis and disease progression following tandem stem-cell transplantation. Blood. 2006;108(5):1724-1732.

Haematologica | 109 February 2024

590

L. Valle Guilhen Longo et al.

ARTICLE - Plasma Cell DIsorders

Minor clone of del(17p) provides a reservoir for relapse in multiple myeloma Jian Cui,1,2 Rui Lv,1,2 Tengteng Yu,1,2,3 Wenqiang Yan,1,2 Jingyu Xu,1,2 Huishou Fan,1,2 Lingna Li,1,2 Yuntong Liu,1,2 Chenxing Du,1,2 Shuhui Deng,1,2 Weiwei Sui,1,2 Yan Xu,1,2 Shuhua Yi,1,2 Dehui Zou,1,2 Lugui Qiu1,2# and Gang An1,2# State Key Laboratory of Experimental Hematology, National Clinical Research Center for

Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Science & Peking Union Medical College, Tianjin, China; 2Tianjin Institutes of Health Science, Tianjin, China and 3LeBow Institute for Myeloma Therapeutics and Jerome Lipper Center for Multiple Myeloma Center, Dana-Farber

Correspondence: G. An angang@ihcams.ac.cn L. Qiu qiulg@ihcams.ac.cn Received: Accepted: Early view:

May 14, 2023. July 27, 2023. August 3, 2023.

Cancer Institute, Harvard Medical School, Boston, MA, USA

Abstract The deletion of chromosome 17p (del(17p)) is considered a crucial prognostic factor at the time of diagnosis in patients with multiple myeloma (MM). However, the impact of del(17p) on survival at different clonal sizes at relapse, as well as the patterns of clonal evolution between diagnosis and relapse and their prognostic value, has not been well described. To address these issues, we analyzed the interphase fluorescence in situ hybridization (iFISH) results of 995 newly diagnosed MM (NDMM) patients and 293 patients with MM at their first relapse. Among these patients, 197 had paired iFISH data at diagnosis and first relapse. Our analysis of paired iFISH revealed that a minor clone of del(17p) at relapse but not at diagnosis was associated with poor prognosis in MM (hazard ratio for median overall survival 1.64 vs. 1.44). Fifty-six and 12 patients developed one or more new cytogenetic abnormalities at relapse, mainly del(17p) and gain/amp(1q), respectively. We classified the patients into six groups based on the change patterns in the clonal size of del(17p) between the two time points. Patients who did not have del(17p) during follow-up showed the best outcomes, whereas those who acquired del(17p) during their disease course, experienced compromised survival (median overall survival: 61.3 vs. 49.4 months; hazard ratio =1.64; 95% confidence interval: 1.06-2.56; P<0.05). In conclusion, our data confirmed the adverse impact of a minor clone of del(17p) at relapse and highlighted the importance of designing optimal therapeutic strategies to eliminate high-risk cytogenetic abnormalities (clinicaltrials gov. identifier: NCT04645199).

Introduction Although there have been significant improvements in the survival of patients with multiple myeloma (MM) over the past decade, patient outcomes still vary, and high-risk patients do not fully benefit from novel drugs.1-3 This can be attributed, in part, to intra-tumor heterogeneity within MM, where treatment only targets sensitive clones, and chemo-resistant clones cannot be eliminated.4,5 In addition, clonal evolution induced by therapy or disease progression is a crucial determinant of patient outcomes in MM.6,7 Recent single-cell studies have further revealed that subclonal secondary genetic events, which are previously undetectable at baseline, may become detectable during follow-up.8,9 Cytogenetic abnormalities (CA), particularly those detected by interphase fluorescence in situ hybridization (iFISH) at diagnosis, have become a crucial aspect

of risk stratification in MM.10,11 However, iFISH-based risk stratification is often used as a static prognostic indicator, and little attention has been paid to examining dynamic changes in the genetic status, such as the number of CA and risk status, from diagnosis to relapse in MM. As a secondary high-risk CA, deletion of chromosome 17p (del(17p)), especially in the high subclonal fraction, is associated with poor prognosis in MM.12,13 Although del(17p) is detected in approximately 5-10% of newly diagnosed MM (NDMM) patients,8 its prevalence increases to more than 10% in patients at relapse,13 mainly due to the emergence of new clones with acquired del(17p) during follow-up.14,15 However, the impact of del(17p) at relapse on survival at different clonal sizes remains unclear, despite the fact that patients who acquire del(17p) during follow-up have significantly shorter overall survival (OS) than controls.8,14,16 Furthermore, the patterns of clonal evolution of del(17p)

Haematologica | 109 February 2024

591

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

between diagnosis and relapse, and its prognostic value are not fully understood. In order to address these questions, we analyzed the paired genetic profiles of 197 patients with MM at diagnosis and first relapse, characterizing the impact of risk status, number of CA, and clonal evolution on their prognostic significance between the two time points. We also assessed the prognostic value of del(17p) at different clonal sizes, both at baseline and relapse. In addition, we identified different patterns of clonal evolution of del(17p) and evaluated their influence on patient outcomes.

Methods Patient database and study population The patients included in this study were sourced from the MM database of the National Longitudinal Cohort of Hematological Diseases (NICHE, clinicaltrials gov. Identifier: NCT04645199). The inclusion criteria required patients to have MM, as defined by the International Myeloma Working Group consensus17 and to have the necessary iFISH data, including testing for gain/amp(1q), del(17p), del(13q), and immunoglobulin heavy chain (IgH) rearrangement. MM patients diagnosed between January 2014 and June 2021 were included in this study. A total of 995 patients with NDMM and 293 patients with their first relapse were identified, with median follow-up times of 76 and 85 months from diagnosis, respectively. For acquired CA, we identified 197 patients with paired iFISH results at diagnosis and first relapse. Patients who did not experience relapse by the end of the follow-up period were excluded from the paired dataset. Patients were allocated to either immunomodulating drug-based or proteasome inhibitor-based induction, as previously described.18 After at least four cycles of induction with a minimum partial response, patients underwent either first-line autologous stem cell transplant (ASCT) or two additional cycles of consolidation treatment. Response assessments were performed according to International Myeloma Working Group consensus criteria.19 Post-induction minimal residual disease (MRD) was assessed by multiparameter flow cytometry as previously reported.7,18 MRD sensitivity threshold was between 10-4 to 10-5. All the patients provided informed consent in compliance with the Declaration of Helsinki. This study was approved by the local Institutional Ethics Committees of the Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical Science & Peking Union Medical College (certificate: IIT2020023-EC-1). Interphase fluorescence in situ hybridization testing at diagnosis and relapse The iFISH technique used in this study has been previously described. Bone marrow (BM) aspirate samples anticoagulated with EDTA were collected, and CD138+ plasma cells

(PC) were isolated using CD138+ magnetic beads (Miltenyi Biotec, Paris, France). iFISH analysis for CA included del(13q), del(17p), gain/amp(1q), t(11;14)(q13;q32), t(4;14)(p16.3;q32), and t(14;16)(q32;q23) in 200 interphase nuclei. The cut-off values for del(17p), gain/amp(1q), del(13q), and translocations were previously reported to be 50%, 20%, 10%, and 10%, respectively.13 Patients with del(17p), t(4;14), or t(14;16) were categorized as having high-risk CA,10 whereas those without these CA were considered standard-risk. Statistical analysis This study aimed to investigate the association between CA and survival outcomes in patients with MM. We defined progression-free survival (PFS) as the duration from diagnosis to the date of death, first progression, or last follow-up. OS was calculated from diagnosis to the date of death or last follow-up. In order to account for time bias, we conducted post-relapse landmark PFS and OS analysis. PFS2, the time from diagnosis to progression of second-line treatment, was defined based on previous studies.20 We used the Kaplan-Meier method to analyze survival data, and differences in survival were evaluated using the log-rank test. Hazard ratios (HR) and 95% confidence intervals (CI) were calculated using the Cox regression model. Multivariable Cox stepwise proportional models were developed to assess the variables with significant impact on survival in the univariable analyses, including age, post-induction response, International Staging System (ISS) stage, post-induction MRD status, transplantation, and del(17p) at relapse. Continuous variables were compared using either Student’s t test or Mann-Whitney U test based on the variables’ distributional statistics. The χ2 test or Fisher’s exact test was used to assess the statistical significance of categorical variables between the different groups. A two-sided P value of less than 0.05 was considered statistically significant. All statistical analyses were performed using SPSS (version 26.0; IBM, Chicago, IL, USA) and R (version 4.2.0; R Foundation, Vienna, Austria).

Results The presence of high-risk cytogenetic abnormalities at relapse exerts a greater adverse impact on multiple myeloma compared with their presence at diagnosis This study enrolled 995 NDMM patients and 293 MM patients experienced their first relapse. Patients in these two cohorts are shown in the Online Supplementary Table S1. All patients underwent iFISH testing for gain/amp(1q), del(17p), del(13q), and IgH rearrangements. Gain/amp(1q) was observed in ≥20% of malignant PC in nearly half of the patients at diagnosis (457/995, 46%), making it the second most frequent cytogenetic event (Figure 1A). However, at relapse, gain/amp(1q) was observed in 63% of the patients and was the most frequent event (Figure 1B). The distri-

Haematologica | 109 February 2024

592

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

bution of each cytogenetic event in the two datasets is summarized in Table 1. We also observed del(17p) (present in at least 50% of malignant PC) in 6% (63/995) of patients at the time of diagnosis (Figure 1A), but this percentage increased to 17% at relapse (Figure 1B). However, the frequency of del(13q) and IgH translocations was comparable between the two groups (Table 2). In order to further investigate this, we compared the number of CA detected by iFISH in each patient. The results demonstrated that patients at relapse carried more CA than those at diagnosis, especially for two or more CA detected using iFISH (69% vs. 54%; P<0.001; Figure 1C). Additionally, when comparing patients with fewer than two CA to those carrying more than two CA, it was observed that patients with fewer than two CA exhibited a longer OS from the time of diagnosis (at diagnosis: 68.5 vs. 41.0 months, HR =1.83; 95% CI: 1.44-2.32; P<0.001; at relapse: 62.3 vs. 38.7 months, HR=1.60; 95% CI: 1.15-2.23; P=0.005) (Figure 1D). In order to gain a better understanding of the prognostic relevance of high-risk CA at relapse, we examined 197 patients

with paired iFISH results at both diagnosis and first relapse (Figure 2A). The baseline characteristics of the patients are presented in the Online Supplementary Table S2. In summary, 45% of the patients had ISS stage 3, 88% had at least one CA detected by iFISH, and 33% of patients in this cohort exhibited high-risk aberrations at baseline. Additionally, del(17p) was observed in 7% (14/197) of patients at diagnosis and in 18% (36/197) of the patients at first relapse, using a cutoff value of 50% (Online Supplementary Table S2). Consistent with our previous findings,6 patients with high-risk CA, whether detected early at diagnosis or later at relapse, experienced inferior outcomes (Figure 2B). Moreover, the presence of highrisk aberrations at relapse had a greater adverse effect on MM than those present at diagnosis (first OS: HR=1.79; 95% CI: 1.25-2.57 vs. HR=1.56; 95% CI: 1.08-2.25).

Longitudinal interphase fluorescence in situ hybridization reveals a clonal selection of secondary cytogenetic abnormalities The paired iFISH results of each patient at diagnosis and

Figure 1. The prognostic significance of high-risk cytogenetic abnormalities that are present at diagnosis or at relapse. (A) Upset plots of cytogenetic abnormalites (CA) detected by interphase fluorescence in situ hybridization (iFISH) at diagnosis. (B) Upset plots of CA detected by iFISH at relapse. (C) Rates of the number of CA in multiple myeloma (MM) patients detected at diagnosis and relapse. ***P<0.001, by two-sided χ² test. (D) Kaplan-Meier analysis of overall survival (OS) for patients with ≤2 CA or >2 CA at diagnosis or at relapse. NS: not significant; ***P<0.001, by two-sided log-rank test. Haematologica | 109 February 2024

593

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

relapse were evaluated, and our findings demonstrated that the newly acquired CA at relapse were primarily secondary cytogenetic events, including del(17p) and gain/amp(1q), whereas no obvious changes in the clonal architecture of del(13q) were observed (Figure 3A). Compared with the

number of CA detected by iFISH at diagnosis, MM showed an increased number of CA at relapse (Figure 3B). While most patients had the same number of CA between the two time points, few patients (56, 11, and 1) developed one, two, and three new CA, respectively, at relapse (Figure 3C).

Table 1. The distributions of cytogenetic abnormalities by probes of 995 patients in the diagnosis dataset and 293 patients in the relapse dataset. At diagnosis N=995

CA, N/N (%)

At relapse N=293

0-10%

10.5-20%

20.5-50%

>50.5%

0-10%

10.5-20%

20.5-50%

>50.5%

Del(13q)

517/995 (52.0)

39/995 (3.9)

81/995 (8.1)

358/995 (36.0)

136/293 (46.4)

12/293 (4.1)

30/293 (10.2)

115/293 (39.2)

Gain/amp(1q)

382/995 (48.8)

52/995 (5.2)

75/995 (7.5)

382/995 (38.4)

94/293 (32.1)

15/293 (5.1)

25/293 (8.5)

159/293 (54.3)

Del(17p)

884/995 (88.8)

17/995 (1.7)

31/995 (3.1)

63/995 (6.3)

215/293 (73.4)

12/293 (4.1)

15/293 (5.1)

51/293 (17.4)

IgH rearrangement

416/995 (41.8)

21/995 (2.1)

53/995 (5.3)

505/995 (50.8)

116/293 (39.6)

4/293 (1.4)

15/293 (5.1)

158/293 (53.9)

t(4;14)

687/835 (82.2)

4/835 (0.4)

13/835 (1.6)

131/835 (15.7)

212/259 (81.9)

0/259 (0)

4/259 (1.5)

43/259 (16.6)

t(11;14)

694/827 (83.9)

8/827 (1.0)

10/827 (1.2)

115/827 (13.9)

211/257 (82.1)

2/257 (0.8)

4/257 (1.6)

40/257 (15.6)

t(14;16)

806/832 (96.8)

0/832 (0)

3/832 (0.4)

23/832 (2.8)

245/259 (94.6)

0/259 (0)

1/259 (0.4)

13/259 (5.0)

t(14; undefined)a

726/824 (88.1)

4/824 0.5)

7/824 (0.8)

87/824 (10.6)

225/257 (87.6)

0/257 (0)

3/257 (1.2)

29/257 (11.3)

t(14; undefined): patients with an undefined abnormality of the 14q32 loci not corresponding to one of the above 3 described common translocations. CA: cytogenetic abnormality; Del: deletion, amp: amplification; IgH: immunoglogulin heavy chain. a

Table 2. Patient characteristics of 995 patients in the diagnosis dataset and 293 patients in the relapse dataset. Characteristics

NDMM N=995

RRMM N=293

Male, N (%)

587 (59)

167 (57)

0.587

Age in years, median (range)

60 (29-83)

57 (34-77)

0.687

ISS stage III, N (%)

390 (39)

132 (45)

0.084

478/995 (48) 457/995 (46) 63/995 (6) 579/995 (58) 148/835 (18) 133/827 (16) 26/832 (3) 98/824 (12) 807/995 (81) 231/836 (28)

157/293 (54) 184/293 (63) 51/293 (17) 177/293 (60) 47/259 (18) 46/257 (18) 14/259 (5) 32/257 (12) 263/293 (90) 105/266 (39)

0.109 < 0.001 < 0.001 0.542 0.950 0.556 0.130 0.896 0.004 <0.001

Cytogenetics, N/N (%) Del(13q)a Gain/amp(1q)b Del(17p)c IgH rearrangementd t(4;14) t(11;14) t(14;16) t(14;undefined)e At least 1 CA by iFISH High-risk CA

The cutoff value for del(17p), gain/amp(1q), del(13q), and IgH translocations were set at 50%, 20%, 10% and 10%, respectively. et(14; undefined): patients with an undefined abnormality of the 14q32 locus that did not correspond to 1 of the above 3 described common translocations. fHigh-risk CA: presence of t (4;14), t(14;16), and/or del(17p). CA: cytogenetic abnormality; ISS: International Staging System; Del: deletion; amp: amplification; IgH: immunoglogulin heavy chain; iFISH: interphase fluorescence in situ hybridization; NDMM: newly diagnosed multiple myeloma; RRMM: relapsed/refractory multiple myeloma.

a,b,c,d

Haematologica | 109 February 2024

594

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

Figure 2. Survival outcomes in patients with different risk statuses identified by paired interphase fluorescence in situ hybridization examinations at diagnosis and relapse. (A) Diagram of different clinical endpoints used in the study. (B) Forest plots of hazard ratio (HR) for median survival in patients with standard-risk versus high-risk cytogenetic abnormalites (CA) at diagnosis (upper) and relapse (lower). PFS: progression-free survival; OS: overall survival; CI: confidence interval.

Furthermore, IgH rearrangement and chromosomal translocations into Ig loci are the founder cytogenetic events in MM. And our results did not indicate any significant changes in IgH-related CA at relapse. We identified a total of 62 patients who maintained at standard-risk, 20 patients who evolved to high-risk, and 112 patients who maintained at high-risk during follow-up, respectively (Figure 3D). When comparing the survival outcomes, the median OS for the three patient groups was as follows: 64.2 months for those who maintained at stand-

ard-risk, 49.4 months for those who evolved to high-risk, and 34.1 months for those who maintained at high-risk (Figure 3E). Additionally, the median second OS for the three groups was 27.2 months, 23.5 months, and 17.4 months, respectively (Figure 3F). Although the log-rank test did not show statistical differences in survival between patients who maintained at high-risk and those who evolved to high-risk, our result showed that patients who evolved to high-risk experienced a relatively longer survival (1st OS: HR=0.91, 95% CI: 0.51-1.63, P=0.751; 2nd OS: HR=0.85, 95%

Haematologica | 109 February 2024

595

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

Continued on following page.

Haematologica | 109 February 2024

596

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

Figure 3. The cytogenetic abnormality profiles between two time points and their prognostic relevance. (A) Heatmap of cell fraction of del(17p), gain/amp(1q) and del(13q) detected by interphase fluorescence in situ hybridization (iFISH) at diagnosis and relapse. Each row represents a specific cytogenetic abnormalites (CA), and each column represents a patient, color coded according to the fraction of plasma cells (PC) detected with a specific CA. (B) Rates of the number of CA in multiple myeloma (MM) patients detected at diagnosis and relapse. *P<0.05, by two-sided χ² test. (C) Sankey diagram showing the distribution and migration of patients’ number of CA between 2 time points. (D) Sankey diagram showing the distribution and migration of patients’ risk status between 2 time points. (E, F) Kaplan-Meier curves in patients with different risk statuses and evolutionary patterns identified by iFISH. Different landmarks are used: overall survival (OS) from diagnosis (E) and OS from relapse (F). NS: not significant; *P<0.05, **P< 0.01, by two-sided log-rank test. Del: deletion; amp: amplification.

CI: 0.47-1.56, P=0.61). Further analysis revealed that all patients who evolved to high-risk at relapse exhibited the acquisition of del(17p) (Online Supplementary Table S3). Minor clone of del(17p) at relapse is associated with poor prognosis in multiple myeloma In order to evaluate the prognostic impact of del(17p) at different clonal sizes, we divided the patients with this CA into three clusters based on the percentage of PC involved: 0-10%, 10-20%, 20-50%, and ≥50%. Using cutoffs ranging from 10% to 50%, the median OS at diagnosis ranged from 34.1 to 29.1 months, and at first relapse, it ranged from 38.7 to 35.5 months (Figure 4A). This highlights the additional prognostic significance of the clonal size of del(17p). Furthermore, there was no significant difference in the first PFS between patients with and without del(17p) at relapse at different clonal sizes (Online Supplementary Figure S1A), suggesting that poor outcomes associated with del(17p) at relapse were mainly due to reduced survival after the first relapse. We then classified the clonal size of del(17p) into three groups: ≤10% (no del(17p)), 10-50% (minor clone of del(17p)), and >50% (major clone of del(17p)). Survival analyses revealed that patients with a minor clone of del(17p) (10-50%) at relapse experienced significantly shorter survival compared to those without del(17p) (≤10%) (1st OS: 43.9 months vs. 63.5 months, HR=1.64, 95% CI: 1.03-2.81, P=0.044; 2nd OS: 28.1 months vs. 17.1 months, HR=1.98, 95% CI: 1.15-3.41, P=0.008). Moreover, our findings indicated no significant difference in survival between patients carrying a major or a minor clone of del(17p) at relapse (Figure 4D, E; Online Supplementary Figure S1D, E). In order to investigate whether a minor clone of del(17p) at relapse remained an independent predictor of outcome when taking account of other prognostic markers including age ≥65 years, post-induction response, ISS stage, post-induction MRD status, transplantation and del(17p) at relapse, we included del(17p) in a multivariable analysis. After univariable analysis, age ≥65 versus <65 years and del(17p) were included in multivariable analysis. Using multivariable Cox stepwise proportional model, the presence of a minor clone of del(17p) at relapse predicted shorter second OS with a hazard ratio of 1.90 (95% CI: 1.10-3.29; P=0.021) (Online Supplementary Tables S4 and S5). Therefore, 10% may be the proper cutoff value for del(17p) at relapse.

Clonal evolution of del(17p) Subsequently, we classified the patients into six groups according to the change patterns in the clonal size of del(17p) between the two time points (Figure 5A). Patients in group A, who experienced the loss of del(17p) at relapse, those in group B, who had a decreasing clonal size from the major to the minor clone at relapse, and those in group C, who did not have del(17p) at both time points, had similar superior outcomes (with a second OS of 50.3 months, 16.6 months, and 26.9 months, respectively). In contrast, patients in group D, who had newly acquired del(17p) at relapse, had a relatively worse survival (with a second OS of 20.2 months). Of the remaining 16 patients, those with a stable clone of del(17p) between the two time points (group E) and those with an obvious increase in clonal size of del(17p) (group F) had the poorest outcomes (with a second OS of 12.5 months and 12.8 months, respectively; Figure 5B). These six del(17p) evolutionary groups were subsequently combined into three patterns, based on the survival curve. Although there was no significant difference in the sampling time between the two time points (Figure 5C), survival analysis revealed that the different evolutionary patterns of del(17p) were able to distinguish the survival curves of OS from diagnosis and post-relapse survival (Figure 5D-F; Online Supplementary Figure S2A). Longitudinal analyses were conducted to investigate the minor clone of del(17p) at diagnosis, with a focus on patients in groups D and F. For patients in group D, nine and 27 patients without del(17p) at diagnosis evolved into a minor clone or major clone of del(17p) at relapse, respectively (Figure 6A; Online Supplementary Table S6). Within our cohort, we also observed two patients who had del(17p) present in less than 10% of PC at baseline, but who subsequently acquired a major clone of del(17p) during follow-up (Figure 6B). Despite the relatively low incidence of del(17p) at diagnosis (14% at the 10% cutoff value), we observed that 18% (36/197) of cases acquired del(17p) during follow-up and 3% (5/197) had a significant increase in clonal size at relapse (Online Supplementary Table S6). Our findings suggest that clonal selection might occur on minor clones of del(17p), which is indicative of poor prognosis in MM.

Discussion This retrospective analysis involved the examination of 995

Haematologica | 109 February 2024

597

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

Figure 4. The prognostic significance of del(17p) that are present at diagnosis or at relapse. (A) Forest plots of hazard ratio (HR) for median survival in patients with different cell fractions of del(17p) at diagnosis (upper) or at relapse (lower). (B, C) Kaplan-Meier curves in patients at diagnosis with no del(17p), a minor clone of del(17p) or a major clone of del(17p). Different landmarks are used: overall survival (OS) from diagnosis (B) and OS from relapse (C). NS: not significant; *P<0.05, by two-sided log-rank test. (D, E) Kaplan-Meier curves in patients at relapse with no del(17p), a minor clone of del(17p) or a major clone of del(17p). Different landmarks are used: OS from diagnosis (D) and OS from relapse (E). NS: not significant; *P<0.05, **P<0.01, ***P<0.001, by two-sided log-rank test. CF: cell fraction. CI: confidence interval.

patients with NDMM and 293 patients with first-relapse MM, all of whom had cytogenetic data available. Among them, 197 patients had paired iFISH results at both diagnosis and the first relapse. Our study led to five main conclusions. First, risk status was dynamic, and routine iFISH should be performed at the first relapse to re-evaluate patients’ risk statuses. Second, clonal evolution caused by disease progression resulted in a higher incidence of secondary CA, specifically del(17p) and gain/amp(1q), at relapse than at

diagnosis. Third, our findings demonstrated that patients who experienced changes in risk status or acquired new CA during follow-up had poorer survival rates, both from diagnosis and post-relapse, compared to patients who maintained standard risk status or the same number of CA between the two time points. Fourth, a minor clone of del(17p) at relapse, but not at diagnosis, was associated with poor prognosis in MM. Finally, patients who never had del(17p) during follow-up had the best outcomes, whereas

Haematologica | 109 February 2024

598

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

Figure 5. Clonal evolution of del(17p) in 197 patients with paired interphase fluorescence in situ hybridization results. (A) The change in cell fraction (CF) of del(17p) between 2 time points. Different colors demonstrate 6 different evolutionary patterns. (B) Overall survival (OS) from second sampling among patients with different del(17p) evolutionary patterns. (C-F) Six del(17p) evolutionary patterns are merged into 3 groups according to the survival curves in (B). Kaplan-Meier curves for the first progression-free survival (PFS) (C), second PFS (D), first OS (E), and second OS (F) are presented. NS: not significant; *P<0.05, **P<0.01, ***P<0.001, by two-sided log-rank test. (G) Diagram of 6 different evolutionary patterns of del(17p) between 2 time points. mOS: median OS. Haematologica | 109 February 2024

599

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

Figure 6. Representative patterns of clonal evolution of del(17p) in relapsed patients. Fish plots visualizing 5 representative patterns of clonal evolution of del(17p) in relapse patients according to the cell fraction of cytogenetic abnormalities (CA) detected using interphase fluorescence in situ hybridization (iFISH) at diagnosis. The vertical line highlights sampling points at diagnosis, post-induction, and relapse. (A) Without del(17p) at diagnosis and with major clones at the first progression (73%) and second progression (100%). (B) Without del(17p) at diagnosis (8%) and with a major clone at first progression (74%). (C) Without del(17p) at diagnosis, at first progression (3%) and post-first progression (7%), evolved into a major clone at the second progression (58%). (D) A minor clone of del(17p) at diagnosis (10.5) evolved into a major clone at the first progression (98%) and second progression (100%). (E) Minor clone of del(17p) at diagnosis (12.5%), eliminated at remission, regrew to major clone at first progression (54%). MM: multiple myeloma; MRD: minimal residual disease. Haematologica | 109 February 2024

600

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

those who had newly acquired del(17p) had compromised survival. The survival analyses in our study revealed that high-risk CA were associated with reduced survival compared to standard-risk CA at diagnosis for all endpoints examined. These findings remained consistent for patients with highrisk CA at first relapse, which was in line with the results of previous studies.21,22 Additionally, patient outcomes were more significantly affected by the presence of high-risk CA at the time of relapse than at diagnosis. Clonal evolution has been widely recognized as inherent mechanism driving the progression of MM,23,24 and extensive research has investigated various patterns of clonal evolution from diagnosis to the first relapse.6,16 In our study, patients who evolved to high-risk experienced relatively longer survival compared to those maintained at high-risk during follow-up, while patients who maintained a standard-risk status demonstrated the best survival outcome. However, the elimination of high-risk CA during follow-up was infrequent, as demonstrated by only three (2%) patients in our cohort. These findings further supported the cumulative nature of CA, especially high-risk CA, which had a negative impact on the prognosis of MM. While there is no universally agreed upon definition for high-risk myeloma,25,26 previous studies have consistently demonstrated that del(17p) is a strong predictor of poor prognosis in patients.27-29 Depending on the specific cutoff value employed, iFISH-based detection of del(17p) has been reported in 5-20% of NDMM patients,27-33 with those with aggressive forms of the disease, such as PC leukemia, having significantly higher incidence rates of del(17p).34 In previous reports on NDMM patients treated with bortezomib and dexamethasone, the 4-year OS rates were 50% and 79% for patients with and without del(17p), respectively.33 A phase III trial of ixazomib or placebo, in combination with lenalidomide and dexamethasone, for relapsed/refractory multiple myeloma (RRMM), has shown that patients without and with del(17p) (observed in ≥5% of malignant PC) have a median PFS of 21.4 and 9.7 months, respectively.35 These results highlight the prognostic value of del(17p) in both NDMM and RRMM patients. In our study, a cutoff of 50% was established for del(17p) at diagnosis, based on the findings from our previous study.13 Del(17p) was detected in 7% (14/197) of the patients at diagnosis and 18% (36/197) of the patients at first relapse, using the cutoff value of 50%, respectively. Additionally, our further analysis revealed that a minor clone of del(17p) at relapse, but not at diagnosis, was associated with a poor prognosis in MM. Therefore, a cutoff value of 10% may be appropriate for del(17p) at relapse. However, laboratories often prefer to use the mean + standard deviation from normal BM controls as the cutoff value. And the choice of the cutoff value for del(17p) at diagnosis remains a topic of debate, with ongoing discussions on how conservative it should be. Hence, further validation of our results is necessary to determine whether

a lower cutoff value should be at relapse. Several factors may contribute to the poor prognosis of patients with a minor clone of del(17p) at relapse. Firstly, studies have shown that therapy-induced clonal evolution can occur as early as the post-induction stage.7,36,37 The residual PC not only undergo clonal evolution at the cytogenetic level, but also adapt to treatment at the transcriptional level. The upregulation of antioxidative genes,36 and protein-folding response genes37 has been observed in residual PC. Consequently, despite the small number of remaining tumor cells after treatment, their adaptation to therapy makes it difficult to eliminate these cells. From this perspective, minor clones of del(17p) can be considered “smart” tumor cells that possess an adaptive response to treatment. Secondly, a previous study reported that inflammation in the BM of MM patients persists after anti-tumor therapy.38 And the abundances of tumor-associated macrophages, natural killer cells, and inflammatory classical dendritic cells has been linked to subclonal (10-80%) or dominant (>80%) gain/amp(1q).9 The interactions between tumor cells and the MM tumor microenvironment contribute to the immune escape of tumor cells. It can be hypothesized that both tumor-intrinsic factors and external microenvironmental factors simultaneously contribute to the drug resistance observed in the minor clone of del(17p), which ultimately resulting in a poor prognosis for these patients. The acquisition of del(17p) during follow-up is considered a rare event in MM, as recently reported in a study of 52 patients with MM who underwent paired targeted sequencing at diagnosis and first relapse. In this study, only 3.8% (2/52) of patients acquired del(17p).16 In a more recent study of 76 patients who acquired del(17p) later during the disease course, the median PFS was 30.1 and 23.0 months (P=0.032), and the median OS was 106.1 and 68.2 months (P<0.001) for controls and patients with acquired del(17p), respectively.14 In another study of 956 patients who were tested for CA by iFISH at diagnosis and first relapse, acquired del(17p) was observed in 38 patients.8 In our cohort, 36 patients had newly acquired del(17p) at the time of relapse. Among these patients, nine and 27 patients without del(17p) at diagnosis developed minor or major clones of del(17p) at relapse, respectively. Consistent with previous studies,8,14 patients with acquired del(17p) had significantly shorter OS than those without del(17p) at both time points (49.4 vs. 63.5 months; P<0.05). Our study had some limitations owing to its retrospective nature. Data on post-relapse survival were not available for all the patients. As patients were not enrolled in a prospectively designed trial, iFISH was not performed at regular time intervals or at every relapse. Additionally, although patients received a relatively homogeneous induction treatment, there was considerable heterogeneity in their post-relapse treatment. Additionally, as the incidence of del(17p) at relapse was low (<10%), and the number of patients with del(17p) in each group was limited, this result

Haematologica | 109 February 2024

601

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM

should be validated in future studies in larger cohorts of patients. Furthermore, recent data have shown that some patients carry micro-subclones of secondary CA that may be missed by bulk analyses such as iFISH.8,14 Therefore, advanced techniques such as next-generation sequencing and single-cell RNA sequencing should be used to monitor the clonal evolution of MM with higher resolution. Besides, the lack of next-generation sequencing data also results in our inability to assess the TP53 allelic state. Finally, survival analysis, other than PFS1, depends on possibly high-risk-enriched patients since they are all selected as relapse patients, thus the interpretation of the results of our study needs to take into account that the populations of interest in our study are for relapsed patients. In conclusion, our data confirmed the poor prognosis of MM associated with high-risk CA. Our findings suggest that even a small subclone of del(17p) at diagnosis should be treated as a high-risk MM. Acquisition of del(17p) or a significant increase in the clonal size of del(17p) during disease progression, though rare events in MM, were associated with a marked reduction in patients’ survival outcomes. We recommend that prospectively designed clinical studies be conducted to regularly monitor the clonal evolution of MM and develop optimal therapeutic strategies to eliminate high-risk CA. Disclosures No conflicts of interest to disclose.

Contributions JC, LQ, and GA developed the concept of the project. GA developed the methodology. RL, TY, WY, JX, HF, LL, YL, SD, CD, WS and YX acquired data, recruited and managed patients, and provided facilities. JC and GA analyzed and interpreted data (i.e., statistical and computational analyses). JC, LQ, and GA reviewed the data, wrote and revised the manuscript. JC, WY, JX, HF, and GA provided administrative, technical, or material support (i.e., reporting or organizing data and constructing databases). SY, DZ, LQ, and GA supervised the study. Acknowledgments We thank all MM patients who participated in this study. Funding This research was supported by the CAMS Innovation Fund for Medical Sciences (CIFMS) (grants 2022-I2M-1-022, to LQ; grants 2021-I2M-1-041), the National Natural Science Foundation (grants 82270218, 81670202, to GA; grants 81900214, to SD; grants 81630007, to LQ), the International Cooperation Projects of National Natural Science Foundation (grant 81920108006, to LQ). Data-sharing statement: The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request.

References 1. Moreau P, Attal M, Hulin C, et al. Bortezomib, thalidomide, and dexamethasone with or without daratumumab before and after autologous stem-cell transplantation for newly diagnosed multiple myeloma (CASSIOPEIA): a randomised, open-label, phase 3 study. Lancet. 2019;394(10192):29-38. 2. Voorhees PM, Kaufman JL, Laubach J, et al. Daratumumab, lenalidomide, bortezomib, and dexamethasone for transplanteligible newly diagnosed multiple myeloma: the GRIFFIN trial. Blood. 2020;136(8):936-945. 3. Attal M, Lauwers-Cances V, Hulin C, et al. Lenalidomide, bortezomib, and dexamethasone with transplantation for myeloma. N Engl J Med. 2017;376(14):1311-1320. 4. Miething CC. Clonal evolution in myeloma: a narrow road to remission. Haematologica. 2019;104(7):1292-1293. 5. Adashek JJ, Subbiah V, Westphalen CB, Naing A, Kato S, Kurzrock R. Cancer: slaying the 9-headed hydra. Ann Oncol. 2022;34(1):61-69. 6. Yan Y, Qin X, Liu J, et al. Clonal phylogeny and evolution of critical cytogenetic aberrations in multiple myeloma at singlecell level by QM-FISH. Blood Adv. 2022;6(2):441-451. 7. An G, Yan Y, Xu Y, et al. Monitoring the cytogenetic architecture of minimal residual plasma cells indicates therapy-induced clonal selection in multiple myeloma. Leukemia. 2020;34(2):578-588. 8. Lannes R, Samur M, Perrot A, et al. In multiple myeloma, highrisk secondary genetic events observed at relapse are present

from diagnosis in tiny, undetectable subclonal populations. J Clin Oncol. 2022;41(9):1695-1702. 9. Tirier SM, Mallm JP, Steiger S, et al. Subclone-specific microenvironmental impact and drug response in refractory multiple myeloma revealed by single-cell transcriptomics. Nat Commun. 2021;12(1):6960. 10. Palumbo A, Avet-Loiseau H, Oliva S, et al. Revised International Staging System for multiple myeloma: a report from International Myeloma Working Group. J Clin Oncol. 2015;33(26):2863-2869. 11. Costa LJ, Usmani SZ. Defining and managing high-risk multiple myeloma: current concepts. J Natl Compr Canc Netw. 2020;18(12):1730-1737. 12. Thakurta A, Ortiz M, Blecua P, et al. High subclonal fraction of 17p deletion is associated with poor prognosis in multiple myeloma. Blood. 2019;133(11):1217-1221. 13. An G, Li Z, Tai YT, et al. The impact of clone size on the prognostic value of chromosome aberrations by fluorescence in situ hybridization in multiple myeloma. Clin Cancer Res. 2015;21(9):2148-2156. 14. Lakshman A, Painuly U, Rajkumar SV, et al. Impact of acquired del(17p) in multiple myeloma. Blood Adv. 2019;3(13):1930-1938. 15. Walker BA, Mavrommatis K, Wardell CP, et al. A high-risk, double-hit, group of newly diagnosed myeloma identified by genomic analysis. Leukemia. 2019;33(1):159-170. 16. Corre J, Cleynen A, Robiou du Pont S, et al. Multiple myeloma

Haematologica | 109 February 2024

602

J. Cui et al.

ARTICLE - Minor clone of del(17p) in MM clonal evolution in homogeneously treated patients. Leukemia. 2018;32(12):2636-2647. 17. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International Myeloma Working Group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 2014;15(12):e538-e548. 18. Yan Y, Mao X, Liu J, et al. The impact of response kinetics for multiple myeloma in the era of novel agents. Blood Adv. 2019;3(19):2895-2904. 19. Kumar S, Paiva B, Anderson KC, et al. International Myeloma Working Group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 2016;17(8):e328-e346. 20. Mbanya Z, Chadda S. Time to second objective disease progression (PFS2): an emerging clinical trial endpoint with regulatory and reimbursement implications. Blood. 2014;124(21):6005-6005. 21. Lonial S, Dimopoulos M, Palumbo A, et al. Elotuzumab therapy for relapsed or refractory multiple myeloma. N Engl J Med. 2015;373(7):621-631. 22. Dimopoulos MA, Moreau P, Palumbo A, et al. Carfilzomib and dexamethasone versus bortezomib and dexamethasone for patients with relapsed or refractory multiple myeloma (ENDEAVOR): a randomised, phase 3, open-label, multicentre study. Lancet Oncol. 2016;17(1):27-38. 23. Dutta AK, Alberge J, Sklavenitis-Pistofidis R, Lightbody ED, Getz G, Ghobrial IM. Single-cell profiling of tumour evolution in multiple myeloma - opportunities for precision medicine. Nat Rev Clin Oncol. 2022;19(4):223-236. 24. van Nieuwenhuijzen N, Spaan I, Raymakers R, Peperzak V. From MGUS to multiple myeloma, a paradigm for clonal evolution of premalignant cells. Cancer Res. 2018;78(10):2449-2456. 25. Zamagni E, Barbato S, Cavo M. How I treat high-risk multiple myeloma. Blood. 2022;139(19):2889-2903. 26. Corre J, Munshi NC, Avet-Loiseau H. Risk factors in multiple myeloma: is it time for a revision. Blood. 2021;137(1):16-19. 27. Avet-Loiseau H, Attal M, Moreau P, et al. Genetic abnormalities and survival in multiple myeloma: the experience of the Intergroupe Francophone du Myélome. Blood. 2007;109(8):3489-3495. 28. Avet-Loiseau H, Attal M, Campion L, et al. Long-term analysis of

the IFM 99 trials for myeloma: cytogenetic abnormalities [t(4;14), del(17p), 1q gains] play a major role in defining longterm survival. J Clin Oncol. 2012;30(16):1949-1952. 29. Lakshman A, Painuly U, Rajkumar SV, et al. Natural history of multiple myeloma with de novo del(17p). Blood Cancer J. 2019;9(3):32. 30. Avet-Loiseau H, Li JY, Godon C, et al. P53 deletion is not a frequent event in multiple myeloma. Br J Haematol. 1999;106(3):717-719. 31. Thanendrarajan S, Tian E, Qu P, et al. The level of deletion 17p and bi-allelic inactivation of TP53 has a significant impact on clinical outcome in multiple myeloma. Haematologica. 2017;102(9):e364-e367. 32. Cohen YC, Saranga A, Gatt ME, et al. Treatment patterns and clinical outcomes in high-risk newly diagnosed multiple myeloma patients carrying the 17p deletion: An observational multi-center retrospective study. Am J Hematol. 2018;93(6):810-815. 33. Avet-Loiseau H, Leleu X, Roussel M, et al. Bortezomib plus dexamethasone induction improves outcome of patients with t(4;14) myeloma but not outcome of patients with del(17p). J Clin Oncol. 2010;28(30):4630-4634. 34. Schinke C, Boyle EM, Ashby C, et al. Genomic analysis of primary plasma cell leukemia reveals complex structural alterations and high-risk mutational patterns. Blood Cancer J. 2020;10(6):70. 35. Avet-Loiseau H, Bahlis NJ, Chng WJ, et al. Ixazomib significantly prolongs progression-free survival in high-risk relapsed/ refractory myeloma patients. Blood. 2017;130(24):2610-2618. 36. Goicoechea I, Puig N, Cedena M, et al. Deep MRD profiling defines outcome and unveils different modes of treatment resistance in standard- and high-risk myeloma. Blood. 2021;137(1):49-60. 37. Cohen YC, Zada M, Wang SY, et al. Identification of resistance pathways and therapeutic targets in relapsed multiple myeloma patients through single-cell sequencing. Nat Med. 2021;27(3):491-503. 38. de Jong M, Kellermayer Z, Papazian N, et al. The multiple myeloma microenvironment is defined by an inflammatory stromal cell landscape. Nat Immunol. 2021;22(6):769-780.

Haematologica | 109 February 2024

603

ARTICLE - Plasma Cell DIsorders

Isatuximab plus carfilzomib and dexamethasone in patients with early versus late relapsed multiple myeloma: IKEMA subgroup analysis Thierry Facon,1 Philippe Moreau,2 Ross Baker,3 Chang-Ki Min,4 Xavier Leleu,5 Mohamad Mohty,6 Lionel Karlin,7 Nicole M. Armstrong,8 Christina Tekle,8 Sandrine Schwab,9 Marie-Laure Risse10 and Thomas Martin

Department of Hematology, Lille University Hospital, Lille, France; 2Department of

Hematology, University Hospital Hôtel-Dieu, Nantes, France; 3Perth Blood Institute, Murdoch

Correspondence: T. Facon Thierry.FACON@chu-lille.fr Received: Accepted: Early view:

March 22, 2023. August 8, 2023. August 17, 2023.

University, Perth, Western Australia, Australia; 4Department of Hematology, Seoul St. Mary’s

https://doi.org/10.3324/haematol.2023.283073

Hospital, College of Medicine, the Catholic University of Korea, Seoul, Republic of Korea;

Published under a CC BY license

Service d’Hématologie et Thérapie Cellulaire, CHU and CIC INSERM 1402, Poitiers Cedex,

France; 6Department of Hematology, Hôpital Saint-Antoine, Sorbonne University, INSERM UMRS 938, Paris, France; 7Department of Hematology, Hôpital Lyon Sud, Hospices Civils de Lyon, Pierre-Bénite, France; 8Sanofi, Cambridge, MA, USA; 9Sanofi R&D, Chilly-Mazarin, France; 10Sanofi R&D, Vitry-sur-Seine, France and 11Department of Hematology, University of California at San Francisco, San Francisco, CA, USA

Abstract Patients with multiple myeloma (MM) who experience early relapse within 12 months of therapy initiation are considered functional high-risk and represent an unmet need, needing better therapies to improve outcomes. The final IKEMA (clinicaltrials gov. identifier: NCT03275285) progression-free survival (PFS) analysis confirmed the significant PFS improvement reported at interim analysis with isatuximab (Isa) plus carfilzomib and dexamethasone (Kd; Isa-Kd) versus Kd in patients with relapsed MM (updated median PFS: 35.7 vs. 19.2 months; hazard ratio [HR] =0.58, 95% confidence interval [CI]: 0.420.79). This IKEMA subgroup analysis examined efficacy and safety of Isa-Kd versus Kd in patients who experienced early (n=61 [Isa-Kd], n=46 [Kd]) vs. late relapse (n=104 [Isa-Kd], n=72 [Kd]). As expected, more aggressive features in baseline characteristics were observed in early relapse patients. Consistent with IKEMA overall population results, median PFS (early relapse: 24.7 vs. 17.2 months, HR=0.662, 95% CI: 0.407-1.077; late relapse: 42.7 vs. 21.9 months, HR=0.542, 95% CI: 0.3550.826), minimal residual disease negativity (MRD−) (early relapse: 24.6% vs. 15.2%; late relapse: 37.5% vs. 16.7%), and MRD− complete response (≥CR) rates (early relapse: 18.0% vs. 10.9%; late relapse: 30.8% vs. 13.9%) were higher with Isa-Kd versus Kd, respectively, in both early and late relapse patients. Grade ≥3, serious treatment-emergent adverse events, and death rates were higher in the late relapse Isa-Kd arm. However, the numbers of deaths were low and treatment exposure was significantly longer in Isa-Kd versus Kd late relapse patients. These results support the addition of Isa to Kd as standardof-care therapy for relapsed and/or refractory MM regardless of relapse timing.

Introduction The availability of novel treatment options, such as immunomodulatory drugs (e.g., lenalidomide, pomalidomide, thalidomide), proteasome inhibitors (e.g., bortezomib, carfilzomib, ixazomib), targeted monoclonal antibodies (e.g., isatuximab, daratumumab, elotuzumab), and combination regimens, has improved treatment outcomes for patients with multiple myeloma (MM); however, MM is still associated with a significant patient burden.1-3 Patients with MM frequently relapse and experience shorter duration of response

with each successive regimen.4 Those who experience early relapse within 1 year of initiating therapy with novel agents have worse prognosis, with significantly reduced median overall survival (21.0 months [early relapse] vs. not reached [late relapse]), and are classified as functional high-risk patients.5,6 The survival disadvantage in early relapse patients is observed regardless of depth of response to initial therapy.5 Furthermore, poor survival outcomes have been reported among patients who experienced early relapse within 12 months of autologous stem cell transplantation (ASCT), even in the era of novel agents.5,7,8 Thus, more ef-

Haematologica | 109 February 2024

604

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

fective agents and combinations are needed for the early relapse subgroup of patients who are at higher risk of more aggressive disease. Anti-CD38 monoclonal antibodies have demonstrated synergistic antitumor effects in combination with backbone therapies that include immunomodulatory drugs and proteasome inhibitors and are being increasingly used in the MM treatment continuum to improve patient outcomes.9,10 Isatuximab (Isa), an anti-CD38 monoclonal antibody that targets a specific CD38 epitope, induces myeloma cell death via multiple mechanisms of action including antibody-dependent cellular-mediated cytotoxicity, complement-dependent cytotoxicity, antibody-dependent cellular phagocytosis, direct apoptosis without crosslinking, and direct inhibition of CD38 ectoenzyme activity.11-13 Based on the results of the phase III ICARIA-MM study (clinicaltrials gov. Identifier: NCT02990338), Isa is approved in a number of countries in combination with pomalidomide and dexamethasone for the treatment of adult patients with relapsed and refractory MM who have received ≥2 prior therapies, including lenalidomide and a proteasome inhibitor.11,14,15 The phase III IKEMA study (clinicaltrials gov. Identifier: NCT03275285) compared Isa in combination with carfilzomib (K) and dexamethasone (d) (Isa-Kd) versus Kd in patients with relapsed MM.16,17 Based on the primary interim analysis results of IKEMA, Isa-Kd is approved in the United States for the treatment of adult patients with relapsed or refractory MM who have received 1-3 prior lines of therapy, in the European Union and other countries for the treatment of adult patients with relapsed MM who have received ≥1 prior therapy, and in Japan for the treatment of adult patients with relapsed or refractory MM who have received one prior treatment.11,14,17,18 The final progression-free survival (PFS) analysis of the IKEMA study, performed 2 years after the prespecified interim analysis, at a median follow-up of 44 months, was recently published.19 The results of this analysis confirmed the significant improvement in PFS reported at the time of the interim analysis with Isa-Kd versus Kd in patients with relapsed MM (updated median PFS 35.7 [Isa-Kd] vs. 19.2 months [Kd]; hazard ratio [HR] =0.58, 95% confidence interval [CI]: 0.42-0.79), with a clinically meaningful increase in minimal residual disease negativity (MRD−) (33.5% vs. 15.4%) and complete response (CR) (44.1% vs. 28.5%) rates in the intent-to-treat population, and a manageable safety profile.19 This subgroup analysis of IKEMA examined updated efficacy and safety of Isa-Kd versus Kd in patients with functional high-risk MM as defined by early relapse from the most recent prior line of therapy versus those who experienced late relapse.

Methods Study design and participants IKEMA was a prospective, multinational, randomized, open-la-

bel, phase III study.16,17 The study was conducted according to the Declaration of Helsinki and the International Conference on Harmonization Guidelines for Good Clinical Practice. The study protocol was approved by an institutional ethics committee or independent review board at all participating centers. All patients provided written informed consent. The IKEMA study design and dosing schedule of the study drugs were previously described in detail16,17 and are summarized in the Online Supplementary Appendix. Briefly, 302 eligible patients with relapsed and/or refractory MM who had received one to three prior lines of therapy, were randomized 3:2 to receive Isa-Kd or Kd. Patients were classified into early or late relapse subgroups based on previously established definitions.20,21 Early relapse was defined as relapse that occurred <12 months from initiation of the most recent line of therapy for patients with ≥2 prior lines of therapy, <18 months for patients with one prior line of therapy, or <12 months following frontline ASCT. Late relapse subgroup included patients who relapsed ≥12 months from initiation of the most recent line of therapy for those with ≥2 prior lines of therapy and ≥18 months for patients with one prior line of therapy. A few patients (n=14 [Isa-Kd], n=5 [Kd]) from the IKEMA overall population were not categorized into either early or late relapse because dates of relapse/progression and latest prior line or ASCT initiation were either not available or incomplete (only year reported) for these patients, and these patients have, therefore, been omitted from the current analysis. MRD was assessed by next-generation sequencing at a central laboratory using the ClonoSEQ Assay (Adaptive Biotechnologies, Seattle, WA, USA) with a sensitivity of 10-5. Adverse events (AE) and laboratory abnormalities were graded according to the National Cancer Information Center Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 4.03. Outcomes The primary endpoint was PFS, defined using the International Myeloma Working Group criteria for progression and disease response evaluation.22 Key secondary endpoints included overall response rate (ORR), rates of very good partial response or better (≥VGPR), MRD−, and CR, and safety. PFS was assessed by a blinded independent review committee based on central laboratory M-protein quantification, local bone marrow aspiration when needed, and central radiologic review. In order to determine the CR rate, the Hydrashift 2/4 Isa immunofixation assay (Sebia, Lisses Evry Cedex, France)23 was used to correct for M-protein interference, when needed. Statistical analysis A prespecified final PFS analysis was conducted on the IKEMA intent-to-treat population (n=179 [Isa-Kd], n=123 [Kd]) and utilized for this post hoc subgroup analysis. The median PFS and CI were calculated by the Kaplan-Meier method.

Haematologica | 109 February 2024

605

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

Non-stratified Cox proportional hazards models including treatment as a covariate were used to estimate HR.24,25 The safety population included all treated patients (n=177 [Isa-Kd], n=122 [Kd]) and analyses of the safety variables were descriptive.

Results Patient characteristics Baseline characteristics for early (n=107) and late relapse (n=176) patients are shown in Table 1. Some imbalances in these characteristics were observed between treatment arms as well as between early and late relapse patients. Imbalances between early and late relapse patients were noted in International Staging System (ISS) stage at study entry and high-risk cytogenetics, with more aggressive features observed in early relapse patients. Additional imbalances were noted in prior lines of treatment and refractoriness. Patients with early relapse had more prior lines, fewer prior ASCT, and were more frequently refractory than those classified as late relapse. Early relapse In total, 61 of 179 (34.1%) patients in the Isa-Kd arm and 46 of 123 (37.4%) patients in the Kd arm were classified as early relapse. The Isa-Kd arm had a higher proportion of patients who had renal impairment (31.0% vs. 15.4%), prior ASCT (49.2% vs. 30.4%), or prior proteasome inhibitors (93.4% vs. 82.6%), and a lower proportion of patients aged ≥75 years old (11.5% vs. 17.4%), patients with ISS stage I at study entry (31.1% vs. 54.3%), or with chromosomal abnormality 1q21+ (41.0% vs. 56.5%) versus Kd arm, respectively. The median number of prior lines of therapy was two for both treatment arms; 32.8% of patients had one prior line of therapy with Isa-Kd versus 41.3% with Kd. Late relapse A total of 104 of 179 (58.1%) patients in the Isa-Kd arm and 72 of 123 (58.5%) patients in the Kd arm were classified as late relapse. The median number of prior lines of therapy was one in the Isa-Kd arm and two in the Kd arm; 55.8% of patients had one prior line of therapy with Isa-Kd versus 48.6% of patients with Kd. More patients were aged ≥75 years (8.7% vs. 2.8%), had renal impairment (21.7% vs. 16.7%), 1q21+ (44.2% vs. 33.3%), or two cytogenetic abnormalities (11.5% vs. 6.9%), and fewer patients were relapsed and refractory (52.9% vs. 68.1%) with Isa-Kd versus Kd, respectively. Treatment exposure At data cutoff (January 14, 2022), the median follow-up was 44 months. The duration of study treatment was longer in patients who received Isa-Kd versus Kd, regardless of early (median [min–max]: 79.0 [2–209] weeks [Isa-Kd]; 52.6

[4–208] weeks [Kd]) or late relapse (median [min–max]: 102.6 [6–206] weeks [Isa-Kd]; 64.9 [2–194] weeks [Kd]) (Online Supplementary Table S1). In addition, treatment duration was longer in late relapse patients compared with that in early relapse patients across both treatment arms. Notably, exposure to Isa-Kd was significantly longer in late relapse patients than that observed in early relapse patients. Among early relapse patients, 16.4% of patients in the IsaKd arm and 6.5% of patients in the Kd arm were still on treatment at data cutoff. The median (min–max) number of cycles was 19.0 (1-49) cycles with Isa-Kd versus 13.5 (1-42) cycles with Kd. The median relative dose intensity (RDI) for Isa was 94.1%. The median RDI for carflizomib was similar in both arms (93.1%, Isa-Kd vs. 91.3%, Kd). The median RDI for dexamethasone was 83.1% in the Isa-Kd arm versus 87.2% in the Kd arm. In late relapse patients, 32.7% of patients in the Isa-Kd arm and 11.1% of patients in the Kd arm were still on treatment at data cutoff. The median (min–max) number of cycles was 24.0 (2-50) cycles with Isa-Kd versus 16.0 (1-47) cycles with Kd. The median RDI for Isa was 91.9%. The median RDI for carfilzomib was 86.5% with Isa-Kd versus 90.5% with Kd. The median RDI for dexamethasone was 77.4% in the Isa-Kd arm versus 88.0% in the Kd arm. Efficacy Progression-free survival At data cutoff, the PFS was longer for patients treated with Isa-Kd versus Kd, respectively, in both early relapse (median 24.7 vs. 17.2 months; HR=0.662, 95% CI: 0.407-1.077) and late relapse patients (median 42.7 vs. 21.9 months; HR=0.542, 95% CI: 0.355-0.826) (Figure 1). Among patients refractory to the last regimen, there was a similar treatment effect favoring Isa-Kd over Kd in early (HR=0.544, 95% CI: 0.313-0.944) and late relapse (HR=0.552, 95% CI: 0.2791.093) patients (Figure 2). Similar treatment effect was also observed in early (median 21.5 vs. 14.8 months; HR=0.620, 95% CI: 0.359-1.069) and in late relapse (median PFS 42.7 vs. 16.2 months; HR=0.634, 95% CI: 0.334-1.202) patients who were refractory to an immunomodulatory agent or a proteasome inhibitor. Depth of response The ORR were 82.0% versus 82.6% in early relapse patients, and 90.4% versus 86.1% in late relapse patients with Isa-Kd versus Kd, respectively (Figure 3). We assessed depth of response by rates of ≥VGPR, ≥CR, MRD−, MRD− ≥VGPR, and MRD− ≥CR. More patients achieved ≥VGPR (early relapse: 67.2% vs. 52.2%; late relapse: 76.0% vs. 58.3%), ≥CR (early relapse: 31.1% vs. 23.9%; late relapse: 52.9% vs. 30.6%), MRD− (early relapse: 24.6% vs. 15.2%; late relapse: 37.5% vs. 16.7%), MRD− ≥VGPR (early relapse: 24.6% vs. 13.0%; late relapse: 37.5% vs. 15.3%), MRD− ≥CR rates (early relapse: 18.0% vs. 10.9%; late relapse: 30.8% vs. 13.9%) with Isa-Kd versus Kd, respectively, regardless of early or late relapse.

Haematologica | 109 February 2024

606

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

Table 1. Key patient demographics and baseline characteristics in IKEMA early relapse and late relapse patients (intent-to-treat population). Early relapse

Age in years, median (range)

Age in years, by category, N (%) <65 65-74 ≥75

Sex, female, N (%)

β2 microglobulin in mg/L, by category, N (%) <3.5 ≥3.5 to <5.5 ≥5.5

Albumin g/L, by category, N (%) <35 ≥35 ISS stage at study entry, N (%) Stage I Stage II Stage III Unknown

Cytogenetics at study entryb,c, N (%) High-risk Standard-risk Missing del(17p) t(4;14) t(14;16) 1q21+ Gain 1q21 1 cytogenetic abnormality 2 cytogenetic abnormalities Bone marrow plasma cells % at baseline, by category, N (%) 0 >0 to <5 ≥5 to <20 ≥20 to <50 ≥50 Missing Prior lines of therapy, median (min-max) 1 2 3 >3 Prior ASCT, N (%)

Prior proteasome inhibitors, N (%) Refractory status, N (%) Relapsed and refractory Relapsed Refractory to IMiD agent Refractory to PI Refractory to IMiD agent and PI Refractory to last regimen

Isa-Kd N=61

Kd N=46

Isa-Kd N=104

Kd N=72

65.0 (39-83)

66.0 (33-90)

64.5 (37-86)

63.0 (40-78)

30 (49.2) 24 (39.3) 7 (11.5)

21 (45.7) 17 (37.0) 8 (17.4)

52 (50.0) 43 (41.3) 9 (8.7)

41 (56.9) 29 (40.3) 2 (2.8)

30 (49.2)

CrCl <60 mL/min/1.73 m , MDRD , N (%) 2

Late relapse

21 (45.7)

46 (44.2)

33 (45.8)

18/58 (31.0)

6/39 (15.4)

20/92 (21.7)

11/66 (16.7)

21 (34.4) 26 (42.6) 14 (23.0)

29 (63.0) 8 (17.4) 9 (19.6)

74 (71.2) 21 (20.2) 9 (8.7)

48 (66.7) 15 (20.8) 9 (12.5)

17 (27.9) 43 (70.5)

9 (19.6) 37 (80.4)

18 (17.3) 84 (80.8)

10 (13.9) 60 (83.3)

19 (31.1) 28 (45.9) 14 (23.0) 0

25 (54.3) 12 (26.1) 9 (19.6) 0

63 (60.6) 31 (29.8) 9 (8.7) 1 (1.0)

44 (61.1) 18 (25.0) 9 (12.5) 1 (1.4)

21 (34.4) 33 (54.1) 7 (11.5) 10 (16.4) 7 (11.5) 5 (8.2) 25 (41.0) 15 (27.8) 26 (42.6) 9 (14.8)

16 (34.8) 28 (60.9) 2 (4.3) 8 (17.4) 11 (23.9) 0 26 (56.5) 18 (41.9) 19 (41.3) 8 (17.4)

19 (18.3) 71 (68.3) 14 (13.5) 6 (5.8) 14 (13.5) 1 (1.0) 46 (44.2) 26 (28.0) 34 (32.7) 12 (11.5)

13 (18.1) 48 (66.7) 11 (15.3) 8 (11.1) 7 (9.7) 0 24 (33.3) 18 (30.0) 20 (27.8) 5 (6.9)

0 13 (21.3) 14 (23.0) 20 (32.8) 13 (21.3) 1 (1.6) 2.0 (1-4) 20 (32.8) 24 (39.3) 16 (26.2) 1 (1.6)

2 (4.3) 7 (15.2) 15 (32.6) 12 (26.1) 9 (19.6) 1 (2.2) 2.0 (1-4) 19 (41.3) 12 (26.1) 14 (30.4) 1 (2.2)

1 (1.0) 19 (18.3) 35 (33.7) 29 (27.9) 18 (17.3) 2 (1.9) 1.0 (1-4) 58 (55.8) 34 (32.7) 11 (10.6) 1 (1.0)

0 11 (15.3) 22 (30.6) 28 (38.9) 9 (12.5) 2 (2.8) 2.0 (1-4) 35 (48.6) 22 (30.6) 14 (19.4) 1 (1.4)

30 (49.2)

14 (30.4)

81 (77.9)

53 (73.6)

57 (93.4)

38 (82.6)

96 (92.3)

63 (87.5)

54 (88.5) 7 (11.5) 33 (54.1) 34 (55.7) 21 (34.4) 49 (80.3)

41 (89.1) 5 (10.9) 27 (58.7) 24 (52.2) 14 (30.4) 39 (84.8)

55 (52.9) 49 (47.1) 34 (32.7) 15 (14.4) 8 (7.7) 32 (30.8)

49 (68.1) 23 (31.9) 27 (37.5) 17 (23.6) 11 (15.3) 29 (40.3)

Incidence calculated in patients with race reported in case report form: 165 patients in Isa-Kd arm, 111 patients in Kd arm in the overall IKEMA ITT population. bHigh risk was defined as the presence of del(17p), or t(4;14), or translocation t(14;16) by fluorescence in situ hybridization. c Cytogenetics was performed by a central laboratory with cutoffs of 50% for del(17p), 30% for t(4;14), t(14;16), and 1q21+. ASCT: autologous stem cell transplantation; CrCl: creatinine clearance; d: dexamethasone; IMiD: immunomodulatory drug; Isa: isatuximab; ISS: International Staging System; ITT: intent-to-treat; Kd: carfilzomib and dexamethasone; MDRD: modification of diet in renal disease; PI: proteasome inhibitor. a

Haematologica | 109 February 2024

607

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

Depth of response in patients who were refractory to the last regimen was also in favor of Isa-Kd in both early and late relapse patients (Figure 4). Consistent with these results, depth of response was improved with Isa-Kd versus Kd after one or ≥2 prior lines of therapy, or after prior ASCT in both early and late relapse patients (Figure 5). Notably, for

patients with one prior line of therapy, MRD− ≥CR rate with Isa-Kd was similar between early (30.0%) and late relapse (34.5%). Depth of response benefit with Isa-Kd versus Kd, regardless of relapse timing, was also consistent in patients who were refractory to an immunomodulatory agent or a proteasome inhibitor (Online Supplementary Figure S1).

Figure 1. Median progression-free survival of early and late relapse patients in the IKEMA intent-to-treat population. (A) Median progression-free survival (mPFS) of early relapse patients. (B) mPFS of late relapse patients. Cut-off date: January 14, 2022. Median follow-up time: 44 months. *As per Independent Review Committee. †mPFS and 95% confidence interval (CI) were calculated by the Kaplan-Meier method. †† Non-stratified Cox proportional hazards models using treatment as a covariate were used to estimate hazard ratios (HR). For adjusted HR estimates, the confounding factors - age, renal impairment, International Staging System (ISS) stage at study entry, 1q21+, and number of prior lines - were used as adjustment covariates. When adjusted for confounding factors, the PFS HR was similar between early (0.577) and late relapse (0.527) patients and in favor of the isatuximab plus carfilzomib and dexamethasone (Isa-Kd) arm. Haematologica | 109 February 2024

608

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

Figure 2. Median progression-free survival of early and late relapse patients refractory to the last regimen. (A) Median progression-free survival (mPFS) of early relapse patients refractory to the last regimen. (B) mPFS of late relapse patients refractory to the last regimen. Cut-off date: January 14, 2022. Median follow-up time: 44 months. *As per Independent Review Committee. † mPFS and 95% confidence interval (CI) were calculated by the Kaplan-Meier method. ††Non-stratified Cox proportional hazards models using treatment as a covariate were used to estimate hazard ratios (HR). Isa: isatuximab; Kd: carfilzomib and dexamethasone.

Safety Among early relapse patients, rates of all-grade (98.4% [IsaKd], 97.8% [Kd]), grade ≥3 (83.6% [Isa-Kd], 80.4% [Kd]), and serious (68.9% [Isa-Kd], 65.2% [Kd]) treatment-emergent adverse events (TEAE) were similar between treatment arms (Table 2). In late relapse patients, rates of all-grade TEAE (99.0% [Isa-Kd], 97.2% [Kd]) were similar between treatment

arms, but rates of grade ≥3 (82.4% [Isa-Kd], 70.4% [Kd]) and serious TEAE (66.7% [Isa-Kd], 54.9% [Kd]) were higher in the Isa-Kd arm. Rates of TEAE leading to definitive treatment discontinuation (early relapse: 11.5% [Isa-Kd] vs. 13.0% [Kd]; late relapse: 13.7% [Isa-Kd] vs. 19.7% [Kd]) were similar in both treatment arms across both early and late relapse patients. The rates of death were 4.9% [Isa-Kd] versus 6.5%

Haematologica | 109 February 2024

609

ARTICLE - IKEMA subgroup analysis by early or late relapse

T. Facon et al.

Figure 3. Depth of response of early and late relapse patients in the IKEMA intent-to-treat population. Cut-off date: January 14, 2022. Median follow-up time: 44 months. Minimal residual disease negativity (MRD−) was assessed by next-generation sequencing Adaptive ClonoSEQ Assay (Adaptive Biotechnologies) at 10-5 sensitivity. For analysis purpose, subjects in the intent-to-treat population but without MRD assessment were considered as having positive MRD. ≥Complete response (≥CR) rate is the proportion of patients who achieved stringent complete response (sCR) or CR as the best overall response according to the International Myeloma Working Group response criteria. Isa: isatuximab; Kd: carfilzomib and dexamethasone; ORR: overall response rate; VGPR: very good partial response.

Figure 4. Depth of response of early and late relapse patients refractory to the last regimen. Cut-off date: January 14, 2022. Median follow-up time: 44 months. Minimal residual disease negativity (MRD−) was assessed by next-generation sequencing Adaptive ClonoSEQ Assay (Adaptive Biotechnologies) at 10-5 sensitivity. For analysis purpose, subjects in the intent-to-treat population but without MRD assessment were considered as having positive MRD. ≥Complete response (≥CR) is the proportion of patients who achieved stringent complete response (sCR) or CR as the best overall response according to the International Myeloma Working Group response criteria. Isa: isatuximab; Kd: carfilzomib and dexamethasone; ORR: overall response rate; VGPR: very good partial response. Haematologica | 109 February 2024

610

ARTICLE - IKEMA subgroup analysis by early or late relapse

T. Facon et al.

Figure 5. Depth of response of early and late relapse patients according to the number of prior lines of treatment or prior transplant. Depth of response after (A) 1 prior line of treament (LOT), or (B) ≥2 prior LOT, or (C) prior autologous stem cell transplant. Cut-off date: January 14, 2022. Median follow-up time: 44 months. Minimal residual disease negativity (MRD−) was assessed by next-generation sequencing Adaptive ClonoSEQ Assay (Adaptive Biotechnologies) at 10-5 sensitivity. For analysis purpose, subjects in the intent-to-treat population but without MRD assessment were considered as having positive MRD. ≥Complete response (≥CR) is the proportion of patients who achieved stringent complete response (sCR) or CR as the best overall response according to the International Myeloma Working Group response criteria. Isa: isatuximab; Kd: carfilzomib and dexamethasone; ORR: overall response rate; VGPR: very good partial response. Haematologica | 109 February 2024

611

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

[Kd] in early relapse patients and 5.9% [Isa-Kd] versus 2.8% [Kd] in late relapse patients. Significantly longer treatment duration in the Isa-Kd arm than in the Kd arm in late relapse patients may have contributed to the increased frequency of grade ≥3, serious TEAE, and deaths in this subgroup. The most common all-grade TEAE were infusion reactions, and less than 10% of patients had all-grade cardiac failure across early (3.3% [Isa-Kd]; 8.7% [Kd]) and late relapse (4.9% [Isa-Kd]; 4.2% [Kd]) patients (Table 3). All-grade TEAE reported more frequently with Isa-Kd (≥10% difference vs. Kd) included infusion reactions in early relapse (41.0% vs. 6.5%) and late relapse patients (50.0% vs. 1.4%), and upper respiratory tract infection (38.2% vs. 26.8%), fatigue (32.4% vs. 19.7%), dyspnea (36.3% vs. 22.5%), bronchitis (30.4% vs. 12.7%), cough (23.5% vs. 11.3%), and gastroenteritis (14.7% vs. 4.2%) in late relapse patients. Grade ≥3 TEAE with different incidences between treatment arms included hypertension in early relapse (19.7% [Isa-Kd] vs. 28.3% [Kd]) and pneumonia in late relapse (18.6% [Isa-Kd] vs. 9.9% [Kd]) (Table 3). Fatal (grade 5) TEAE during study treatment period in early relapse patients included cardiac failure in one (1.6%) patient, disease progression in one (1.6%) patient, and pneumonia and multiple non-site-specific injuries in one (1.6%) patient in the Isa-Kd arm; and acute myocardial infarction in one (2.2%) patient, disease progression in one (2.2%) patient, and COVID-19 in one (2.2%) patient in the Kd arm. Fatal TEAE during study treatment period in late relapse patients included pneumonia in one (1.0%) patient, atypical pneumonia in one (1.0%) patient, asthma in one (1.0%) patient, cardiac failure and acute kidney injury in one (1.0%) patient, and COVID-19 infections in two (2.0%) patients in the Isa-Kd arm; and cardiac failure and acute kidney injury in one (1.4%) patient, and sudden death in one (1.4%) patient in the Kd arm. Hematologic laboratory abnormalities reported more fre-

quently in the Isa-Kd arm included grade 3 anemia (42.6% vs. 30.4%) in early relapse patients and grade 3 neutropenia in early (18.0% vs. 4.3%) and late relapse (13.7% vs. 8.5%) patients, and thrombocytopenia in early relapse patients (21.3% vs. 15.2%) with Isa-Kd versus Kd, respectively (Table 3).

Discussion Patients with MM frequently relapse, requiring successive lines of therapy; those who experience early relapse within 12 months of therapy initiation have worse outcomes and are considered functional high-risk patients.5-8 In this post hoc subgroup analysis of IKEMA, the addition of Isa to Kd resulted in clinically meaningful improvement in PFS (early relapse: HR=0.662, 95% CI: 0.407-1.077; late relapse: HR=0.542, 95% CI: 0.355-0.826) and depth of response, with a manageable safety profile in both early and late relapse patients, consistent with the benefit observed in the overall IKEMA study population.17,19 The benefit with Isa-Kd versus Kd was also observed in early and late relapse patients who were refractory to the last regimen. Similar to the observations in the IKEMA intent-to-treat population, the ORR in the current analysis were comparable between treatment arms, but deeper responses were seen with Isa-Kd versus Kd regardless of the timing of relapse, favoring Isa-Kd over Kd.17,19 Notably, the depth of response (≥VGPR, ≥CR, MRD−, MRD− ≥VGPR, and MRD− ≥CR rates) benefit with Isa-Kd versus Kd in early and late relapse patients was consistent across different subpopulations regardless of prior lines of therapy or prior transplant. Among patients who had received only one prior line of therapy, MRD− ≥CR rates with Isa-Kd were similar regardless of early (30.0%) or late (34.5%) relapse, suggesting Isa-Kd as an effective treatment regimen for salvage in these patients and may lead to deep responses

Table 2. Safety overview with Isa-Kd versus Kd in IKEMA early and late relapse patients (safety population). Early relapse

Late relapse

TEAEa, N (%)

Isa-Kd N=61

Kd N=46

Isa-Kd N=102

Kd N=71

Any TEAE Grade ≥3 TEAE Serious TEAE

60 (98.4) 51 (83.6) 42 (68.9)

45 (97.8) 37 (80.4) 30 (65.2)

101 (99.0) 84 (82.4) 68 (66.7)

69 (97.2) 50 (70.4) 39 (54.9)

7 (11.5)

6 (13.0)

14 (13.7)

14 (19.7)

1 (1.6) 10 (16.4) 9 (14.8)

0 0 5 (10.9)

0 19 (18.6) 13 (12.7)

0 1 (1.4) 2 (2.8)

Any TEAE leading to definitive treatment discontinuation Any TEAE leading to premature discontinuation Isatuximab Carfilzomib Dexamethasone

Fatal TEAE during study treatment

3 (4.9)

3 (6.5)

6 (5.9)

2 (2.8)

Treatment-emergent adverse events (TEAE) were assessed according to National Cancer Institute Common Toxicity Criteria for Adverse Events version 4.03. Isa: isatuximab; Kd: carfilzomib and dexamethasone. a

Haematologica | 109 February 2024

612

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

despite patients’ functional high-risk status. The depth of response results in the current study are consistent with our observations reported at IKEMA prespecified interim subgroup analysis, where deeper responses were observed with Isa-Kd versus Kd, regardless of number of prior lines of therapy or refractory status.26 Clinically meaningful higher ≥VGPR (1 prior line: 75.0% vs. 61.8%; >1 prior line: 70.7% vs. 51.5%) and MRD− (1 prior line: 33.8% vs. 18.2%; >1 prior line: 26.3% vs. 8.8%) rates were observed with Isa-Kd versus Kd, regardless of number of prior lines of therapy. Deeper

responses with Isa-Kd versus Kd were also reported for patients who were refractory to lenalidomide (≥VGPR: 66.7% vs. 35.7%; MRD−: 24.6% vs. 9.5%), refractory to lenalidomide at last regimen (≥VGPR: 72.2% vs. 38.7%; MRD−: 27.8% vs. 9.7%), refractory to bortezomib (≥VGPR: 55.8% vs. 51.3%; MRD−: 17.3% vs. 10.3%), or refractory to bortezomib at last regimen (≥VGPR: 62.5% vs. 47.8%; MRD−: 25.0% vs. 8.7%). Grade ≥3, serious TEAE, and deaths were higher in the IsaKd arm in late relapse patients. TEAE leading to definitive treatment discontinuation were similar between treatment

Table 3. Most common treatment-emergent adverse events, selected treatment-emergent adverse events, and hematologic laboratory abnormalities with Isa-Kd versus Kd in IKEMA early and late relapse patients (safety population). Early relapse Isa-Kd N=61

Selected TEAE Preferred term, N (%)

Late relapse Kd N=46

Isa-Kd N=102

Kd N=71

All grades

Grade ≥3

All grades

Grade ≥3

All grades

Grade ≥3

All grades

Grade ≥3

Infusion reaction

25 (41.0)

3 (6.5)

51 (50.0)

1 (1.0)

1 (1.4)

Hypertension

23 (37.7)

12 (19.7)

17 (37.0)

13 (28.3)

37 (36.3)

22 (21.6)

25 (35.2)

15 (21.1)

Diarrhea

21 (34.4)

2 (3.3)

14 (30.4)

1 (2.2)

44 (43.1)

3 (2.9)

24 (33.8)

2 (2.8)

URTI

20 (32.8)

2 (3.3)

12 (26.1)

1 (2.2)

39 (38.2)

3 (2.9)

19 (26.8)

1 (1.4)

Fatigue

20 (32.8)

3 (4.9)

11 (23.9)

1 (2.2)

33 (32.4)

7 (6.9)

14 (19.7)

Dyspnea

14 (23.0)

2 (3.3)

9 (19.6)

37 (36.3)

8 (7.8)

16 (22.5)

1 (1.4)

Pneumonia

14 (23.0)

11 (18.0)

9 (19.6)

7 (15.2)

29 (28.4)

19 (18.6)

15 (21.1)

7 (9.9)

Cough

11 (18.0)

9 (19.6)

24 (23.5)

8 (11.3)

Bronchitis

10 (16.4)

5 (10.9)

31 (30.4)

3 (2.9)

9 (12.7)

1 (1.4)

Gastroenteritis

3 (4.9)

2 (3.3)

6 (13.0)

2 (4.3)

15 (14.7)

3 (4.2)

2 (3.3)

4 (8.7)

3 (6.5)

5 (4.9)

2 (2.0)

3 (4.2)

1 (1.4)

Cardiac failure events, N (%) Cardiac failure, any class

2 (3.3)

Hematologic laboratory abnormalities, N (%)

All grades

Grade 3

Grade 4

All grades

Grade 3

Grade 4

All grades

Grade 3

Grade 4

All grades

Grade 3

Grade 4

Anemia

61 (100)

26 (42.6)

45 (97.8)

14 (30.4)

102 (100)

14 (13.7)

71 (100)

10 (14.1)

Lymphopenia

18 (29.5)

2 (3.3)

16 (34.8)

3 (6.5)

1 (2.2)

25 (24.5)

2 (2.0)

1 (1.0)

15 (21.1)

1 (1.4)

Neutropenia

35 (57.4)

11 (18.0)

2 (3.3)

18 (39.1)

2 (4.3)

1 (2.2)

53 (52.0)

14 (13.7)

2 (2.0)

33 (46.5)

6 (8.5)

Thrombocytopenia

57 (93.4)

13 (21.3)

11 (18.0)

39 (84.8)

7 (15.2)

6 (13.0)

99 (97.1)

17 (16.7)

8 (7.8)

66 (93.0)

11 (15.5)

3 (4.2)

Isa: isatuximab; Kd: carfilzomib and dexamethasone; TEAE: treatment-emergent adverse event; URTI: upper respiratory tract infection.

Haematologica | 109 February 2024

613

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

arms across early and late relapse patients. Significantly longer exposure to Isa-Kd in late relapse patients may also explain the increased frequency of grade ≥3, serious TEAE, and deaths in this subgroup. The numbers of deaths were low, so differences may be due to chance. Other subgroup analyses with triplet regimens containing an anti-CD38 antibody in a similar patient subpopulation also showed efficacy benefit versus doublet control regimen. The definitions used to classify early and late relapse patients in the current study were the same as those used in the phase III CANDOR, CASTOR, and POLLUX subgroup analyses that evaluated triplet regimens based on another anti-CD38 monoclonal antibody, daratumumab.20,27 Post hoc subgroup analyses of the CANDOR (daratumumab plus Kd vs. Kd), CASTOR (daratumumab plus bortezomib and dexamethasone [D-Vd] vs. Vd), and POLLUX (daratumumab plus lenalidomide and dexamethasone [D-Rd] vs. Rd) studies reported PFS HR=0.4-0.7, and ≥CR rates of 16.0-53.0% versus 0-17.0% in early relapse patients who had received daratumumab-based regimens versus control regimens, respectively.20,27 MRD− (10-5) rates for early relapse patients were 13-30% with D-Vd /D-Rd versus 0-4% with Vd/Rd in CASTOR/POLLUX and were not reported for CANDOR. Consistent with observations in the IKEMA overall population, the efficacy outcomes of patients in the Kd arm in the current subgroup analysis were favorable, indicating that the benefit observed with the addition of Isa is not due to suboptimal outcomes in the control group.17,19 Our results align with a previous post hoc subgroup analysis of the phase III ASPIRE (carfilzomib, lenalidomide, and dexamethasone [KRd] vs. Rd) and ENDEAVOR (Kd vs. Vd) studies, which demonstrated improved PFS and ORR in patients receiving carfilzomib-based treatment compared with control arm, regardless of early (relapse ≤1 year after initiating most recent prior line of therapy) or late relapse (relapse after >1 year following initiation of most recent prior line of therapy).28 A prospective observational study across Europe and Israel reported similar ORR benefit with KRd regardless of early relapse (83.3%; included patients who relapsed ≤12 months [≤18 months with 1 prior line of therapy] from start of most recent prior line of therapy) or late relapse (77.1%; included patients who relapsed >12 months [>18 months with 1 prior line of therapy] from start of most recent prior line of therapy), but ≥CR rates were lower in early relapse (16.7%) than in late relapse patients (22.9%).21 A common theme that is evident across all studies summarized above is that the outcomes in early relapse patients are generally worse than in late relapse patients. Consistent with these reports, the PFS and depth of response in the current study were lower in early relapse patients than in late relapse patients across both treatment arms, in the intent-to-treat population as well as in patients refractory to the last regimen, confirming the unmet need in the early relapse subgroup of patients. The only exception was the depth of response with one prior line of therapy, which was

similar between early and late relapse patients in the IsaKd arm. Nevertheless, the PFS and the depth of response were in favor of Isa-Kd over Kd in both early and late relapse patients. Importantly, the median PFS of 24.7 months in the Isa-Kd arm of early relapse patients compares favorably to data recently reported for early relapse patients who had progression <18 months after frontline ASCT in the KarMMa-2 phase II trial of the BCMA-directed chimeric antigen receptor T-cell therapy, idecabtagene vicleucel (median PFS 11.4 months at a median follow-up of 21.5 months).29 However, cross-trial comparisons should be interpreted with caution given the inherent differences between the study populations, and differences in follow-up duration, outcomes within the control arms, definitions used to classify early and late relapse patients, as well as limitations within each study. A limitation of the current study includes small numbers of patients in the subgroup analyses. However, the PFS, depth of response, and safety profile of Isa-Kd versus Kd observed in the overall IKEMA population were consistent across early and late relapse patients, and in subgroups that were refractory to last regimen as well as those who received one or ≥2 prior lines of therapy or prior ASCT, favoring IsaKd over Kd. These results showed improved median PFS and depth of response with Isa in combination with Kd and support the use of Isa-Kd as a standard of care in patients with relapsed MM regardless of early or late relapse. Disclosures PM has received honoraria from and is part of the advisory board of AbbVie, Amgen, Celgene, GlaxoSmithKline, Janssen and Sanofi. RB has received grants from AbbVie, Acerta Pharma, Alexion, Amgen, Bayer, Biegene, BMS, Boehringer Ingelheim, Celgene, CSL Behring, Daiichi Sankyo, Jansen-Cilag, MorphoSys, Pfizer, Pharmaxis, Portola, Rigel Pharmaceuticals, Roche, Sanofi, Takeda and Technoclone; has received honoraria from Bayer, Cardinal Health, BMS, Jansen-Cilag and Roche; is part of the advisory board of Pharmaxis, Jansen-Cilag and Roche. MM has received grants from Janssen and Sanofi; consults for Adaptive Biotechnologies, Janssen, Oncopeptides and Sanofi; has received honoraria from Amgen, Astellas, BMS, Gilead, GlaxoSmithKline, Janssen, Novartis, Pfizer, Sanofi and Takeda; discloses other board/society/ committee leadership at EBMT, IACH and IFM. LK has received honoraria from Amgen, Celgene, Sanofi, AbbVie, Takeda and Janssen; discloses other board/society/committee leadership at Amgen, Celgene, GlaxoSmithKline, Janssen and Takeda. TM has received research funding from Sanofi. NMA, CT, SS, and M-LR are employed by Sanofi; may hold stock and/or stock options in the company. All other authors have no conflicts of interest to disclose. Contributions TF, PM, RB, LP, C-KM, XL, MM, LK and TM were investigators in the study and contributed to data acquisition and analysis. CT, M-LR, NMA and SS contributed to study design, data

Haematologica | 109 February 2024

614

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse

analysis, and interpretation. All authors revised the work for important intellectual content and assume responsibility for data integrity and the decision to submit this manuscript for publication, had full access to the study data, edited and reviewed manuscript drafts, and approved the final version for submission. Acknowledgments We thank the participating patients and their families, and the study centers and investigators, for their contributions to the study. We are grateful to Dr. Ludek Pour and Dr. Andreea M. Rawlings for their support and contributions. Coordination of the development of this manuscript, facilitation of author discussion, and critical review was provided by Aidee Ayala Camargo, PhD, Sci Comms Director at Sanofi. Medical writing support was provided by Smitha Reddy, PhD, of Envision

Pharma Group, funded by Sanofi. Funding The IKEMA study was funded by Sanofi. Data-sharing statement Qualified researchers can request access to patient-level data and related study documents including the clinical study report, study protocol with any amendments, blank case report forms, statistical analysis plan, and dataset specifications. Patient-level data will be anonymized, and study documents will be redacted to protect the privacy of trial participants. Further details on Sanofi’s data sharing criteria, eligible studies, and process for requesting access are at: https://www.vivli.org.

References 1. Jagannath S, Abonour R, Durie BGM, et al. Heterogeneity of second-line treatment for patients with multiple myeloma in the Connect MM Registry (2010-2016). Clin Lymphoma Myeloma Leuk. 2018;18(7):480-485. 2. Raab MS, Fink L, Schoen P, et al. Evolution of multiple myeloma treatment practices in Europe from 2014 to 2016. Br J Haematol. 2019;185(5):981-984. 3. Song X, Cong Z, Wilson K. Real-world treatment patterns, comorbidities, and disease-related complications in patients with multiple myeloma in the United States. Curr Med Res Opin. 2016;32(1):95-103. 4. Kumar SK, Therneau TM, Gertz MA, et al. Clinical course of patients with relapsed multiple myeloma. Mayo Clin Proc. 2004;79(7):867-874. 5. Majithia N, Rajkumar SV, Lacy MQ, et al. Early relapse following initial therapy for multiple myeloma predicts poor outcomes in the era of novel agents. Leukemia. 2016;30(11):2208-2213. 6. Soekojo CY, Chung TH, Furqan MS, Chng WJ. Genomic characterization of functional high-risk multiple myeloma patients. Blood Cancer J. 2022;12(1):24. 7. Kumar S, Mahmood ST, Lacy MQ, et al. Impact of early relapse after auto-SCT for multiple myeloma. Bone Marrow Transplant. 2008;42(6):413-420. 8. Jimenez-Zepeda VH, Reece DE, Trudel S, Chen C, Tiedemann R, Kukreti V. Early relapse after single auto-SCT for multiple myeloma is a major predictor of survival in the era of novel agents. Bone Marrow Transplant. 2015;50(2):204-208. 9. Martin TG, Corzo K, Chiron M, et al. Therapeutic opportunities with pharmacological inhibition of CD38 with isatuximab. Cells. 2019;8(12):1522. 10. Leleu X, Martin T, Weisel K, et al. Anti-CD38 antibody therapy for patients with relapsed/refractory multiple myeloma: differential mechanisms of action and recent clinical trial outcomes. Ann Hematol. 2022;101(10):2123-2137. 11. Sanofi. Sarclisa (isatuximab-irfc) [package insert]. Bridgewater, NJ: Sanofi-Aventis U.S. LLC; 2021. http://products.sanofi.us/ Sarclisa/sarclisa.pdf. Accessed August 3, 2023. 12. Jiang H, Acharya C, An G, et al. SAR650984 directly induces multiple myeloma cell death via lysosomal-associated and

apoptotic pathways, which is further enhanced by pomalidomide. Leukemia. 2016;30(2):399-408. 13. Tai YT, Anderson KC. Targeting CD38 alleviates tumor-induced immunosuppression. Oncotarget. 2017;8(68):112166-112167. 14. Sanofi. Sarclisa (isatuximab) [summary of product characteristics]. Paris, France: Sanofi-Aventis; 2021. https:// www.ema.europa.eu/en/documents/product-information/ sarclisa-epar-product-information_en.pdf. Accessed August 3, 2023. 15. Attal M, Richardson PG, Rajkumar SV, et al. Isatuximab plus pomalidomide and low-dose dexamethasone versus pomalidomide and low-dose dexamethasone in patients with relapsed and refractory multiple myeloma (ICARIA-MM): a randomised, multicentre, open-label, phase 3 study. Lancet. 2019;394(10214):2096-2107. 16. Moreau P, Dimopoulos MA, Yong K, et al. Isatuximab plus carfilzomib/dexamethasone versus carfilzomib/dexamethasone in patients with relapsed/refractory multiple myeloma: IKEMA Phase III study design. Future Oncol. 2020;16(2):4347-4358. 17. Moreau P, Dimopoulos MA, Mikhael J, et al. Isatuximab, carfilzomib, and dexamethasone in relapsed multiple myeloma (IKEMA): a multicentre, open-label, randomised phase 3 trial. Lancet. 2021;397(10292):2361-2371. 18. Sanofi. Sarclisa (isatuximab) [Prescribing Information]. Nishi Shinjuku, Tokyo: Sanofi-Aventis; 2021. https://www.pmda.go.jp/ PmdaSearch/iyakuDetail/ ResultDataSetPDF/780069_4291454A1021_1_02. Accessed August 3, 2023. 19. Moreau P, Dimopoulos MA, Mikhael J, et al. Updated progression-free survival (PFS) and depth of response in IKEMA, a randomized phase III trial of isatuximab, carfilzomib and dexamethasone (Isa-Kd) vs Kd in relapsed multiple myeloma (MM). Ann Oncol. 2022;33(6):P664-665. 20. Weisel K, Geils G, Karlin L, et al. Carfilzomib, dexamethasone, and daratumumab versus carfilzomib and dexamethasone in relapsed or refractory multiple myeloma: subgroup analysis of the phase 3 CANDOR study in patients with early or late relapse. Blood. 2020;136;(Suppl 1):S37-38. 21. Terpos E, Caers J, Gamberi B, et al. Response to carfilzomib

Haematologica | 109 February 2024

615

T. Facon et al.

ARTICLE - IKEMA subgroup analysis by early or late relapse regimens among patients with early or late relapse following prior multiple myeloma therapy: a subgroup analysis from a prospective observational study across Europe and Israel. In: European Hematology Association; 2020 Virtual. p. Abstract EP1010. 22. Kumar S, Paiva B, Anderson KC, et al. International Myeloma Working Group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 2016;17(8):e328-e346. 23. Finn F, Macé, S, Chu, R, et al. Development of a Hydrashift 2/4 isatuximab assay to mitigate interference with monoclonal protein detection on immunofixation electrophoresis in vitro diagnostic tests in multiple myeloma. Blood. 2020;136;(Suppl 1):S15. 24. Kaplan E, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc. 1958;282(53):457-481. 25. Cox D. Regression models and life-tables. J R Stat Soc Series B Stat Methodol. 1972;34(2):187-202. 26. Dimopoulos MA, Moreau P, Augustson B, et al. Isatuximab plus carfilzomib and dexamethasone in patients with relapsed

multiple myeloma based on prior lines of treatment and refractory status: IKEMA subgroup analysis. Am J Hematol. 2023;98(1):E15-E19. 27. Spencer A, Moreau P, Mateos MV, et al. Daratumumab (DARA) in combination with bortezomib plus dexamethasone (D-Vd) or lenalidomide plus dexamethasone (D-Rd) in relapsed or refractory multiple myeloma (RRMM): subgroup analysis of the phase 3 CASTOR and POLLUX studies in patients (pts) with early or late relapse after initial therapy. J Clin Oncol. 2022;40;(Suppl 16):S8052. 28. Mateos MV, Goldschmidt H, San-Miguel J, et al. Carfilzomib in relapsed or refractory multiple myeloma patients with early or late relapse following prior therapy: A subgroup analysis of the randomized phase 3 ASPIRE and ENDEAVOR trials. Hematol Oncol. 2018;36(2):463-470. 29. Usmani S, Patel K, Kari P, et al. KarMMa-2 Cohort 2a: Efficacy and safety of idecabtagene vicleucel in clinical high-risk multiple myeloma patients with early relapse after frontline autologous stem cell transplantation. Blood. 2022;140;(Suppl 1):S875-877.

Haematologica | 109 February 2024

616

LETTER TO THE EDITOR

Efficacy and feasibility of pharmacoscopy-guided treatment for acute myeloid leukemia patients who have exhausted all registered therapeutic options Elderly or unfit patients with relapsed acute myeloid leukemia (AML) are frequently unable to undergo intensive chemotherapy and allogeneic stem cell transplantation, the only curative treatment approach for this entity. They often rapidly exhaust the few approved and reimbursable (registered) treatment options for AML at relapse, thus facing a poor prognosis.1,2 The situation is aggravated by the rapidly progressive nature of the disease, severely restricting the time frame for therapy selection.3 Recent studies have demonstrated the effectiveness of pharmacoscopy - an image-based ex vivo functional drug testing platform - to provide guidance when selecting therapies for hematological malignancies lacking standard treatment protocols.4–7 However, no previous study has focused exclusively on AML patients that have exhausted all registered therapeutic options. Furthermore, the issue of obtaining financial coverage for such therapy plans remains unaddressed. We aimed to establish whether pharmacoscopy can be employed for therapy selection in AML patients that have exhausted the registered treatment options. We evaluated whether such therapies can be made available within a suitable time frame and adequate financial coverage and whether they provide a clinical benefit in heavily pretreated and frail patients. In our prospective, non-randomized, single-center observational study (DARTT-1; clinicaltrials gov. Identifier: NCT05732688; BASEC-ID: 2021-01294, Department of Medical Oncology, University Hospital Bern, Switzerland), we enrolled and screened 24 adult AML patients at relapse who had exhausted all registered treatment options. Of those, five patients were screened at least a second time after relapsing. In total, 30 screening events with subsequent treatment and follow-up from 24 patients were included in our intention-to-treat population. We successfully collected sufficient samples for subsequent drug screening from all intended patients, of which 14 were from bone marrow origin, 14 from peripheral blood, and two from skin tissue. Excluded were patients who had not undergone previous treatment or still had registered therapy options available. All participants signed informed consent, and the study was executed in accordance with good clinical practice and approved by the relevant review boards and regulatory agencies. The 24 patients were pretreated with a median of two previous treatment lines (interquartile range (IQR), 1-3). The median patient age was 68.5 years (IQR, 66-73). The sex distribution was 62.5% male versus 37.5% female. According to the European LeukemiaNet (ELN) risk category,1 12.5% of the patients were classified as favorable, 37.5% as

intermediate, and 50.0% as adverse. Further demographics are listed in Table 1. For statistical analyses, each screening event was treated as a separate data point. The complete drug screening results used in this study, including more detailed patient information, prior lines of treatments, and other relevant clinical parameters, as well as more detailed descriptions of the methodologies and statistical approaches employed in this study, are available at https:// www.snijderlab.org/trials/DARTT-1/. For each patient and screening event, we performed pharmacoscopy using real-time patient-derived leukemic samples to generate treatment recommendations (Figure 1), as previously described.4–6 The results of the drug screen were communicated in the form of a short list of drugs recommended for the treatment of the respective patient (pharmacoscopy report). If the compounds in these reports could not be made available within a reasonable time frame and adequate financial coverage, patients were provided with a therapy based on in-house guidelines. Mandatory health insurance in Switzerland reimburses pharmaceuticals if they are on the national list of reimbursable specialties. Other drugs may be reimbursed upon reasoned request if they are expected to have a substantial effect against a serious illness for which no other approved treatment is available.8 Patients were provided with best supportive care if they could not undergo further treatment attempts. Online Supplementary Table S1 lists a summary of ex vivo drug screening results, waiting times, and characteristics of treatment choice. The median waiting time for screening results was 5.0 days (IQR, 4.0-6.0). The pharmacoscopy reports listed a median of 5.5 drugs per patient (IQR, 4.3-8.8), ranked by their predicted efficacy. In 17 (56.7%) instances, the patient received one of the recommended therapy options, while no recommended drug was administered in 13 (43.3%) instances. Nine screening instances (30.0%) resulted in patients receiving an AML-specific therapy not listed in the pharmacoscopy reports. Four patients were switched to best supportive care due to their deteriorating condition or refusal of further therapy. The duration from receiving the drug recommendations to starting a new therapy was a median of 11 days (IQR, 6-24). The top six most frequently recommended drugs were navitoclax, venetoclax, omacetaxine, cladribine, carfilzomib, and panobinostat. Whereas omacetaxine and panobinostat were never administered to patients in the study due to the difficulty of obtaining financial coverage, venetoclax was the most frequently administered drug.

Haematologica | 109 February 2024

617

LETTER TO THE EDITOR

In order to monitor response and outcome, we used clinical assessments, as well as blood and bone marrow testing during routine follow-ups. Our primary endpoints were how frequently pharmacoscopy derived treatments could be started, the frequency of patients realizing a complete remission (CR) in the bone marrow, overall survival (OS), and event-free survival (EFS). We assessed these endpoints by comparing patients who received a therapy recommended by pharmacoscopy versus patients receiving a therapy based on in-house guidelines. In order to compare different treatment regimens (pharmacoscopy based vs. in-house guidelines) in terms of their performance during the drug screen, we calculated the integrated peritoneal cancer index (i-PCY) score of each regimen. The i-PCY score is a previously established score calculated during pharmacoscopy and indicates a treatment regimen’s capacity to selectively eliminate leukemic blasts from a patient-derived sample.4 We divided patients into two groups based on their therapy regimen’s i-PCY score (above or below the median of the study population) and separately assessed the frequency of CR, OS, and EFS. Neither of our stratification strategies differed significantly in patient characteristics or tissue origin of the sample (Online Supplementary Table S2). The response to the chosen therapy regimen in the bone marrow could be assessed for 25 screening instances (Online Supplementary Table S3). We found that a significantly

higher percentage (45.5% vs. 21.4%) of patients receiving a drug combination with an i-PCY score above the median of the study population (i.e., a regimen that performed exceptionally well during the respective drug screen) achieved a CR than patients receiving a regimen with a lower i-PCY Table 1. Summary of study group. Characteristics Age in years at start of the study Median (IQR)

68.5 (66-73)

Sex, N (%) Female Male

9.0 (37.5) 15.0 (62.5)

ELN risk category, N (%) Favorable Intermediate Adverse

3.0 (12.5) 9.0 (37.5) 12.0 (50.0)

Time in months from diagnosis to study Median (IQR)

12.5 (4.5-24.8)

Number of previous therapy lines Median (IQR)

2.0 (1.0-3.0)

Previous high dose chemotherapy Patients, N (%)

5.0 (20.8)

Most common previous therapy protocol

azacitidine + venetoclax

IQR: interquartile range; N: number; ELN: European LeukemiaNet.

Figure 1. Pharmacoscopy workflow for acute myeloid leukemia at relapse. Patient samples (bone marrow draws, peripheral blood, or subcutaneous/skin samples) were shipped by courier to the pharmacoscopy laboratory. There, the cells from the samples were processed by either density centrifugation (blood and bone marrow) or tissue dissociation (skin) and seeded into 384-well imaging plates, with each well containing a chemo- or immunotherapeutic compound from our test library. The plates were then incubated overnight. Immunofluorescence stainings against specific surface antigen characteristics of the patient’s leukemic cells were used to distinguish between healthy cells and malignant blasts. The cells were then subjected to automated confocal microscopy (Opera Phenix, Perkin Elmer) and image analysis using nuclear morphology to quantify the viability of malignant and healthy cells, respectively. Based on this readout, the ex vivo blast reduction capacity (peritoneal cancer index [PCY] score) was calculated for each compound. Pharmacoscopy reports were provided to the treating oncologists in the form of a short list of top-scoring drugs ranked by their predicted efficacy, as well as complete drug response profiles. The selection of treatment regimens was subsequently based on the pharmacoscopy report and the availability of compounds listed therein. If none of the listed compounds could be made available within a reasonable time frame, therapy regimens were chosen based on previously established in-house guidelines at our department. Haematologica | 109 February 2024

618

LETTER TO THE EDITOR

Figure 2. Patients outcome. (A) Kaplan-Meier estimates of overall survival (OS) in the intention-to-treat population stratified by whether patients received a pharmacoscopy-derived treatment regimen (17 screening instances) or a regimen based on in-house guidelines (13 screening instances). (B) Kaplan-Meier estimates of event-free survival (EFS) in the intention-to-treat population stratified by whether patients received a pharmacoscopy-derived treatment regimen or not (17 vs. 13 screening instances). (C) Kaplan-Meier estimates of OS in the intention-to-treat population stratified by the ex vivo blast reduction capacity of their respective treatment regimen (integrated peritoneal cancer index [i-PCY] score) (14 screening instances with above median i-PCY score and 16 below). (D) Kaplan-Meier estimates of EFS in the intention-to-treat population stratified by the i-PCY score of their respective treatment regimen (14 vs. 16 screening instances). (E) Kaplan-Meier estimates of OS in patients receiving an acute myeloid leukemia (AML)-specific therapy (excluding patients receiving best supportive care) stratified by the i-PCY score of their respective treatment regimen (12 screening instances with above median i-PCY score and 14 below). (F) Kaplan-Meier estimates of EFS in patients receiving an AML-specific therapy stratified by the i-PCY score of their respective treatment regimen (12 vs. 14 screening instances). All P values were calculated by Gehan-Breslow-Wilcoxon test. CI: confidence interval; NS: not significant. Haematologica | 109 February 2024

619

LETTER TO THE EDITOR

score (odds ratio [OR] =3.078; P=0.0005). Conversely, patients that achieved a complete remission tended to receive treatment regimens with a higher i-PCY score (mean 0.278) than patients that did not achieve a CR (mean 0.155). When comparing patients receiving a pharmacoscopy-recommended therapy with patients receiving a regimen-based on in-house guidelines, the difference in the relative number of complete responders was 35.3% versus 25.0% (OR=1.615; P=0.1646). We next evaluated EFS and OS using standard outcome definitions in AML.1 Patients receiving a pharmacoscopy recommended therapy had a median OS of 18.0 weeks versus 8.0 weeks in patients receiving a regimen based on in-house guidelines (OS ratio =2.250; 95% confidence interval [CI]: 1.021-4.958) (Figure 2A). Median EFS was 11.1 weeks in patients receiving a recommended therapy compared to 6.3 weeks in the rest of the cohort (EFS ratio =1.773; 95% CI: 0.805- 3.906) (Figure 2B). When stratifying the intention-to-treat population by i-PCY score of their therapy regimen (i.e., its performance in the drug screen), we found that patients receiving a regimen with an i-PCY score above the median of the study population showed a median OS of 28.6 weeks as opposed to 8.4 weeks for the other half of the cohort (ratio 3.390; 95% CI: 1.5067.632; P=0.006) (Figure 2C). Median EFS for patients receiving a regimen with an i-PCY score above the median was 12.4 in comparison to 6.4 weeks for the other half of the cohort (EFS ratio =1.933; 95% CI: 0.859-4.353; P= 0.0446) (Figure 2D). The significant effect of a treatment regimen’s i-PCY score on OS and a strong effect on EFS was also present when focusing only on patients receiving an AML-specific therapy (Figure 2E, F). The inherently poor prognosis of AML patients at relapse is aggravated by the limited availability of registered therapy options at this stage and the rapid progression of the disease. Thus, improving the prognosis of AML patients that have exhausted all registered therapy options remains an unsolved issue.3 We evaluated a novel approach, basing treatment decisions on the recommendations of pharmacoscopy, an automated imaging-based ex vivo drug screening platform. We could demonstrate that integrating pharmacoscopy into the clinical decision-making AML treatment at relapse is technically feasible and appears beneficial to the patients. The screening procedure is fast, taking a median of only 16 days from sample taking to the start of a new protocol. Thus, clinical decision-making is not significantly delayed, a key concern in AML at relapse.1,3 Financial coverage for pharmacoscopy-derived treatment plans could be obtained for most patients in our cohort (56.7%), and such therapy regimens led to promising trends in response and survival rates. Additionally, we found that the i-PCY score of a treatment regimen (i.e., its performance during the drug screen) is an excellent predictor of response and survival. Patients receiving a therapy regimen with an above-average i-PCY score showed significantly higher rates of CR and significantly longer OS than the rest of the cohort. Thus

the i-PCY score can be a useful cue to prioritize between readily available standard therapy options even when the highest-scoring compounds in the screen prove unavailable. Previous studies have evaluated integrating ex vivo drug screening into therapy selection for AML patients.4,5,7 However, direct comparisons with these studies are inherently complex, given that previous research either adopted an observational approach,9 employed different inclusion criteria (e.g., including patients for whom standard treatment options were still viable4,5,7), or lacked control groups.7 To the best of our understanding, this is the first research effort concentrating solely on patients who have exhausted all registered treatment methodologies while including a control group consisting of patients for whom pharmacoscopy-guided treatment was not feasible. Additionally, our investigation is the inaugural one to address the crucial clinical issue of securing cost coverage for AML treatment protocols chosen via drug screening. The major limitation of our study resides in the absence of patient randomization into distinct treatment groups. Nevertheless, significant stratification of the intention-to-treat population, based on the ex vivo blast reduction capacity, signals potential for future randomized trials. In such studies, patients could be divided into an intervention group, which would receive a treatment regimen optimized by a summed RBF, and a control group, which would adhere to either physician-chosen treatments or treatments grounded in established guidelines. We conclude that pharmacoscopy can rapidly provide valuable decision-making cues for therapy selection in late-stage AML patients, helping to choose between established therapies and to design novel treatment plans. We, therefore, suggest that it may standardly be employed to provide patients with optimized treatment plans.

Authors Jonas Andreas Schmid,1* Yasmin Festl,2* Yannik Severin,2* Ulrike Bacher,3 Marie-Noëlle Kronig,4 Berend Snijder2# and Thomas Pabst4# Faculty of Medicine University of Bern, Bern; 2Department of Biology,

Institute of Molecular Systems Biology, ETH Zürich, Zürich; Department of Hematology, Inselspital, University Hospital Bern,

University of Bern, Bern and 4Department of Medical Oncology, Inselspital, University Hospital Bern, University of Bern, Bern, Switzerland JAS, YF and YS contributed equally as first authors.

BS and TB contributed equally as senior authors.

Correspondence: B. SNIJDER - snijder@imsb.biol.ethz.ch https://doi.org/10.3324/haematol.2023.283224

Haematologica | 109 February 2024

620

LETTER TO THE EDITOR

Received: March 27, 2023.

assisted in the enrollment of the patients and wrote the initial draft

Accepted: July 6, 2023.

of the manuscript. YF and YS performed all pharmacoscopy

Early view: July 13, 2023.

experiments and the associated computational analysis, helped to write the final manuscript and, together with BS, provided

treatment recommendations. MNK recruited and enrolled the

Published under a CC BY-NC license

patients. UB assisted in conceptualizing the study and writing of the manuscript. BS and TB conceived, supervised, and funded the study,

Disclosures

advised on experimental design, data interpretation and helped

BS was a scientific co-founder of Allcyte GmbH, which has been

writing the manuscript. All authors read and approved the final

acquired by Exscientia; is a shareholder of Exscientia and a co-

version of the manuscript.

inventor on U.S. patent application 15/514,045 relevant to the study; declares research funding from Roche and speaker fees from

Data-sharing statement

Novartis, GSK and AbbVie. All other authors have no potential

The data required to reproduce our results are included in the

conflicts of interest to disclose.

tables and the Online Supplementary Appendix. Complete drug screening results generated in this study and more detailed patient

Contributions

information are furthermore available at https://www.snijderlab.org/

JAS recorded clinical response, carried out statistical analysis,

trials/DARTT-1/.

References 1. Döhner H, Estey EH, Amadori S, et al. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European LeukemiaNet. Blood. 2010;115(3):453-474. 2. Laribi K, Sobh M, Ghez D, Baugier de Materre A. Impact of age, functional status, and comorbidities on quality of life and outcomes in elderly patients with AML: review. Ann Hematol. 2021;100(6):1359-1376. 3. Webster JA, Luznik L, Gojo I. Treatment of AML relapse after allo-HCT. Front Oncol. 2021;11:812207. 4. Snijder B, Vladimer GI, Krall N, et al. Image-based ex-vivo drug screening for patients with aggressive haematological malignancies: interim results from a single-arm, open-label, pilot study. Lancet Haematol. 2017;4(12):e595-e606. 5. Kornauth C, Pemovska T, Vladimer GI, et al. Functional precision

medicine provides clinical benefit in advanced aggressive hematologic cancers and identifies exceptional responders. Cancer Discov. 2022;12(2):372-387. 6. Heinemann T, Kornauth C, Severin Y, et al. Deep morphology learning enhances ex vivo drug profiling-based precision medicine. Blood Cancer Discov. 2022;3(6):502-515. 7. Malani D, Kumar A, Brück O, et al. Implementing a functional precision medicine tumor board for acute myeloid leukemia. Cancer Discov. 2022;12(2):388-401. 8. De Pietro C, Camenzind P, Sturny I, et al. Switzerland: health system review. Health Syst Transit. 2015;17(4):1-288, xix. 9. Kuusanmäki H, Kytölä S, Vänttinen I, et al. Ex vivo venetoclax sensitivity testing predicts treatment response in acute myeloid leukemia. Haematologica. 2023;108(7):1768-1781.

Haematologica | 109 February 2024

621

LETTER TO THE EDITOR

Vaccine utilization and overwhelming post-splenectomy infection risk factors in two asplenia cohorts meningitidis, and H. influenzae OPSI, the overall time spent unvaccinated, and risk factors for individuals with OPSI in two asplenia registries. A retrospective analysis of two independent cohorts of Tricare beneficiaries with asplenia was performed. The first cohort was derived from the Department of Defense Joint Trauma Registry (DoDTR) of the Joint Trauma System (JTS).13 The JTS was developed to provide evidence-based improvement of combat casualty care, and the DoDTR facilitates analysis of combat casualties. The registry incorporates data from the point of injury forward in time or until the patient no longer has Tricare insurance. The DoDTR was queried on February 27, 2020 for service members who underwent splenectomy following combat trauma from January 1, 2000 to January 1, 2020 using ICD codes (see Online Supplementary Table S1). Most individuals identified from the DoDTR were not active Tricare beneficiaries at the time of the query as only 44 of 256 subjects from the DoDTR had a most recent encounter within 6 months of the query.

The spleen is a secondary lymphoid organ involved with immune surveillance of blood. Asplenia following splenectomy and functional asplenia due to sickle cell disease (HbSS disease) are associated with invasive infections known as overwhelming post-splenectomy infections (OPSI).1,2 OPSI can be caused by several bacterial species including gram-positive Streptococcus pneumoniae as well as the gram-negative organisms Neisseria meningitidis and Haemophilus influenzae type B2–4 for which the American College of Immunizations Practices (ACIP) recommends vaccination.5–7 The rate of OPSI in patients with asplenia in an Australian clinical registry was lower (1 OPSI every 2,778 patient years) than the rate of OPSI prior to joining the registry (1 OPSI every 667 patient years) suggesting a benefit to immunization.8,9 However, there are few other studies assessing which patients are most likely to develop OPSI though risk factors may include failure to vaccinate, infection with S. pneumoniae non-vaccine serotypes, inadequate response to immunizations, and immunodeficiency.8,10–12 We aimed to assess the rates of S. pneumoniae, N.

Table 1. Demographics of National Capital Region Registry (NCRR) and Department of Defense Joint Trauma Registry (DoDTR) Cohorts. Characteristic

NCRR

DoDTR

Subjects, N

171

256

Average age in years

49.65

25.17

Male, N (%)

60 (35)

251 (98.0)

HJ bodies, N (%)

45 (26.3)

92 (35.9)

Etiology of asplenia, N (%) HbSS disease Splenectomy, N (%) Other neoplasms Pancreatic neoplasms Trauma Immune thrombocytopenia Other

40 (23.4) 131 (76.6) 29 (17) 26 (15.2) 18 (10.5) 18 (10.5) 40 (23.3)

0 256 (100) 0 0 256 (100) 0 0

Total patient years with asplenia

3,344.66

1,374.56

19.45 (0.58-64.42)

5.37 (0.5-16)

145/171 (84) 121/171 (71) 116/171 (68) 142/171 (83) 65/171 (38)

17/256 (15) 117/256 (69) 168/256 (65) 234/256 (91) 4/256 (5)

Average patient years with asplenia (range) OPSI events, N Immunizations, N/N (%) PCV13 immunized PPSV23 immunized HiB immunized MCV4/MPS4 immunized MenB immunized

HJ bodies: Howell–Jolly bodies; HbSS: sickle cell disease; OPSI: overwhelming post-splenectomy infections; PCV13: pneumococcal conjugate vaccine 13; PPSV23: pneumococcal polysaccharide vaccine 23; HiB: Haemophilus influenzae type B vaccine; MCV4/MPS4: meningococcal vaccines; MenB: meningococcal B vaccine.

Haematologica | 109 February 2024

622

LETTER TO THE EDITOR

The second cohort was derived from a National Capital Region Registry (NCRR) linked to the Military Health System (MHS) electronic medical record (EMR) and claims data accessed from the MHS population health portal. The NCRR includes military personnel, dependents, foreign nationals, and retirees who receive care in the national capital region. The NCRR was queried on April 30, 2021 for individuals with any form of asplenia including those with HbSS disease using ICD and CPT codes (see Online Supplementary Table S1). All NCRR individuals at the time of the query were active Tricare beneficiaries. Individuals in the DoDTR and NCRR were included if there was documentation of splenectomy or asplenia in a clinical encounter and the duration of follow up was greater than 30 days. Coded data were collected by review of subject charts. Data collected included demographics, microbiology reports, peripheral blood smear interpretations, immunology laboratory assessments, and vaccinations. OPSI were defined by a S. pneumoniae, N. meningitidis, or H. influenzae positive cultures from immunologically protected sites such as blood and cerebral spinal fluid. Immunizations collected included pneumococcal polysaccharide (PPSV23), pneumococcal conjugate (PCV13), meningococcal conjugate (MCV4), meningococcal polysaccharide (MPSV4), meningococcal B (MenB), and H. influenzae type b (Hib). In the DoDTR, 552 individuals were identified with splenic injury. Of the 552, 239 were excluded since no documentation of splenectomy was identified in their EMR, nine of 552 were excluded due to partial splenectomy, 14 of 552 were excluded due to splenorrhaphy, and 34 of 552 were excluded due to lack of follow-up beyond 30 days after

splenectomy. Of the 552, 256 were confirmed to have asplenia with at least 30 days of follow-up and were included in the analysis. In the NCRR, 338 individuals were identified with an asplenia ICD or CPT code. Of the 338, 167 were excluded since no documentation of asplenia was identified in the EMR. The remaining 171 of 338 were confirmed to have asplenia. All subjects had at least 30 days of follow-up after splenectomy or 1 year of age if they had HbSS disease without splenectomy since nearly all patients with HbSS disease have splenic dysfunction by age 1.1 Demographic data for both cohorts are shown in Table 1. Ninety-eight percent of subjects were men and 65% of subjects were women in the DoDTR and NCRR, respectively. Subjects in the DoDTR had a lower average age than subjects in the NCRR (25.17 vs. 49.65 years old respectively). One hundred percent of subjects in the DoDTR had asplenia from splenectomy following trauma compared to only 10.5% in the NCRR; 23.4% of subjects in the NCRR had asplenia related to HbSS disease. The average duration of clinical follow-up after development of asplenia was nearly three times lower in the DoDTR compared to the NCRR cohort (5.37 vs. 19.45 years, respectively). There were five OPSI events in the NCRR cohort corresponding to a prevalence of one OPSI per 669 patient years. There were no OPSI events in the DoDTR cohort in 1,375 patient years. The rate of immunization was higher in the NCRR compared to DoDTR for PCV13 (84% vs. 15%), PPSV23 (71% vs. 69%), HiB (68% vs. 65%), and MenB (38% vs. 5%). More subjects in the DoDTR compared to NCRR were immunized for MCV4/MPS4 (91% vs. 83%). The total time spent unvaccinated was higher in the NCRR

Table 2. Description of subjects with overwhelming post-splenectomy infections. Age in years^/Sex Asplenia etiology

Years to OPSI

Infection

Immunizations prior to OPSI* PCV13

PPSV23

HiB

Immunodeficiency diagnoses

58/M

Splenectomy Pancreatic Neoplasm

4.6

H. influenzae bacteremia

Yes

53/F

Splenectomy Pancreatic Neoplasm

S. pneumoniae meningitis

Yes

Selective IgM deficiency

13/F

Splenectomy Hematologic Malignancy

S. pneumoniae bacteremia

Yes**

Selective IgM deficiency

1/M

HbSS

28.4

S. pneumoniae bacteremia and olecranon bursitis

Yes**

1/F

HbSS

4.9

S. pneumoniae bacteremia

Yes

Yes**

Age corresponds to age at splenectomy or age 1 year old where asplenia is present in sickle cell disease (HbSS). *Prior immunizations are listed for the organism underlying the overwhelming post-splenectomy infections (OPSI). **PPSV23 was administered within 5 years of the OPSI. PCV13: pneumococcal conjugate vaccine 13; PPSV23: pneumococcal polysaccharide vaccine 23; HiB: Haemophilus influenzae type B vaccine; F: female; M: male; IgM: immunoglobulin M; NA: not applicable. ^

Haematologica | 109 February 2024

623

LETTER TO THE EDITOR

cohort as compared to the DoDTR for all vaccines assessed (Figure 1). In the NCRR cohort, four subjects had an invasive pneumococcal and one subject had invasive H. influenzae infection (Table 2). For the pneumococcal OPSI, there was a prevalence of one OPSI per 695 PCV13-unvaccinated patient years and per 526 PPSV23-unvaccinated patient years. One of the four subjects with an invasive pneumococcal infection was immunized with PCV13 and four of four were immunized with PPSV23 prior to developing the OPSI. Three of the four vaccinated with PPSV23 were immunized within 5 years of developing their OPSI. For H. influenzae OPSI, there was a prevalence of one OPSI per 434 Hib-unvaccinated patient years. The subject with H. influenzae OPSI was immunized with HiB prior to the OPSI and demonstrated H. influenzae B antibody IgG titer of 5.8 mcg/ mL several years later (protective titer >0.15 mcg/mL). Two

of the four subjects with an invasive pneumococcal OPSI had evidence of selective IgM deficiency (SIGMD). SIGMD was associated with reduced responses to polysaccharide PPSV23 antigens in both subjects and low isohemagglutinin titers in one subject (see Online Supplementary Table S2). There was no documentation in the EMR that the S. pneumoniae isolates had been serotyped. Limitations of this study included the retrospective design and lack of pneumococcal serotyping data. In addition, the small number of OPSI events make it difficult to draw firm conclusions on OPSI risk factors in patients undergoing splenectomy or HbSS disease. Nonetheless, there are several interesting findings. First, OPSI in patients with asplenia are rare, consistent with prior reports,8,9 with one OPSI per 669 patient years in the NCRR cohort and zero OPSI per 1,375 patient years in the DoDTR cohort. The lower rate

Figure 1. Cumulative days spent unvaccinated for each vaccine in each cohort. The sum of the days spent unvaccinated following development of asplenia for all subjects in each cohort for each vaccine was determined. The Y-axis corresponds to the total days spent unvaccinated and the X-axis corresponds to each vaccine. The bars corresponding to the Department of Defense Joint Trauma Registry (DoDTR) cohort are in dark gray and the bars corresponding to the National Capital Region Registry (NCRR) cohort are in light gray. Conjugate and polysaccharide meningococcal vaccines rates were combined since the American College of Immunizations Practices has not recommended one over another. Time spent unvaccinated for each vaccine or vaccine series was defined by the difference in days between splenectomy and either the date of first immunization (PCV13, PPSV23, Hib), last immunization in series (MCV4/MPSV4, MenB), or last encounter date in the electronic medical record if no vaccine was administered. In patients with sickle cell (HbSS) disease, the date of the subject’s first birthday was used in place of the date of splenectomy. PCV13: pneumococcal conjugate vaccine 13; PPSV23: pneumococcal polysaccharide vaccine 23; HiB: Haemophilus influenzae type B vaccine; MCV4/MPS4: meningococcal vaccines; MenB: meningococcal B vaccine. Haematologica | 109 February 2024

624

LETTER TO THE EDITOR

of OPSI in the DoDTR compared to the NCRR may reflect the relative health of the active-duty military population as well as the fewer total days spent unimmunized. The low OPSI rate in both cohorts is notable especially when considering the total number of days spent unvaccinated for each vaccine. Second, the data we present suggests a possible benefit to PCV13 immunization. For instance, the cumulative time spent unvaccinated was 2.5 times longer for PCV13 in the NCRR cohort (1,190,862 days) compared to the DoDTR cohort (482,511days), and three of the four subjects with invasive pneumococcal infection had not been immunized with PCV13 prior to the OPSI. Third, most patients with an S. pneumoniae OPSI are not eligible under CDC and state rules for pneumococcal serotyping. Therefore, there is a need to develop a national resource for pneumococcal serotyping for patients with OPSI to differentiate between vaccine non-response and emergence of pneumococcal non-vaccine serotype infection. Finally, two of the five patients with OPSI in the NCRR cohort demonstrated SIGMD suggesting a possible link between SIGMD and OPSI. Prior reports have shown a high rate of poor response to pneumococcal polysaccharide antigens in patients with SIGMD compared to the rates of poor response in the general population.14,15 Also, a link between OPSI and problems in humoral immunity has been suggested for autoimmune lymphoproliferative syndrome10 as well as in two patients with lack of response to pneumococcal polysaccharide antigens.11 In summary, OPSI are rare in patients with asplenia, even in those who are unvaccinated suggesting the number needed to immunize is high to prevent one OPSI infection, pneumococcal isolate serotyping would benefit patients with OPSI, and the rare patients with OPSI should be evaluated for occult humoral immunodeficiency.

Surgery, Walter Reed National Military Medical Center, Bethesda, MD; 4

Department of Research Programs, Walter Reed National Military

Medical Center, Bethesda, MD; 5Enterprise Intelligence and Data Solutions Program Office, Program Executive Office, Defense Healthcare Management Systems, San Antonio, TX; 6Laboratory of Clinical Immunology and Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD and 7Department of Medicine, Uniformed Services University, Bethesda, MD, USA Correspondence: N. A. BOGGS - nathan.boggs@usuhs.edu https://doi.org/10.3324/haematol.2023.283419 Received: May 3, 2023. Accepted: July 12, 2023. Early view: July 20, 2023. ©2024 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose. Contributions NB, MS, VKR, MB and RC designed the study. MS, MM, QW, WH, and NB collected the data. MS, MM, QW, NW and NB analyzed the data. MS and NB wrote the manuscript with input from all co-authors. The view(s) expressed herein are those of the author(s) and do not reflect the official policy or position of Uniformed Services University, National Institutes of Health or the Department of Health and Human Services, Walter Reed National Military Medical Center, Brooke Army Medical Center, the US Army Medical Department, the US Army Office of the Surgeon General, the Defense Health Agency, the Departments of the Air Force, Navy, Army, or the Department of Defense, or the US Government.

Authors

Funding Matthew A. Soderstrom, Mechelle A. Miller M.D., Qing Wang,

This research was funded in whole or in part by the Division of Intramural

William P. Hennrikus, Nora L. Watson, Ryan C. Costantino, Matthew

Research of the National Institute of Allergy and Infectious Diseases,

J. Bradley, V. Koneti Rao and Nathan A. Boggs

National Institutes of Health.

Department of Internal Medicine, Brooke Army Medical Center, San

Data-sharing statement

Antonio, TX; Allergy & Immunology Service, Walter Reed National

The data that support the findings of this study are available upon

Military Medical Center, Bethesda, MD; Department of General

reasonable request to the corresponding author.

2,7

References 1. Rogers ZR, Wang WC, Luo Z, et al. Biomarkers of splenic function in infants with sickle cell anemia: baseline data from the BABY HUG Trial. Blood. 2011;117(9):2614-2617. 2. King H, Shumacker HB. Splenic studies. I. Susceptibility to infection after splenectomy performed in infancy. Ann Surg. 1952;136(2):239-242. 3. Smith CH, Erlandson M, Schulman I, Stern G. Hazard of severe

infections in splenectomized infants and children. Am J Med. 1957;22(3):390-404. 4. Smith CH, Schulman I, Ando RE, Stern G. Studies in Mediterranean (Cooley’s) anemia. I. Clinical and hematologic aspects of splenectomy, with special reference to fetal hemoglobin synthesis. Blood. 1955;10(6):582-599. 5. Kobayashi M, Farrar JL, Gierke R, et al. Use of 15-valent

Haematologica | 109 February 2024

625

LETTER TO THE EDITOR pneumococcal conjugate vaccine and 20-valent pneumococcal conjugate vaccine among U.S. adults: updated recommendations of the Advisory Committee on Immunization Practices, United States, 2022. MMWR Morb Mortal Wkly Rep. 2022;71(4):109-117. 6. Mbaeyi SA, Bozio CH, Duffy J, et al. Meningococcal vaccination: recommendations of the Advisory Committee on Immunization Practices, United states, 2020. MMWR Recomm Rep. 2020;69(9):1-41. 7. Briere EC, Rubin L, Moro PL, et al. Prevention and control of haemophilus influenzae type b disease: recommendations of the advisory committee on immunization practices (ACIP). MMWR Recomm Rep. 2014;63(RR-01):1-14. 8. Arnott A, Jones P, Franklin LJ, et al. A registry for patients with asplenia/hyposplenism reduces the risk of infections with encapsulated organisms. Clin Infect Dis. 2018;67(4):557-561. 9. Wang J, Jones P, Cheng AC, Leder K. Adherence to infection prevention measures in a statewide spleen registry. Med J Aust. 2014;200(9):538-540. 10. Neven B, Bruneau J, Stolzenberg M-C, et al. Defective anti-

polysaccharide response and splenic marginal zone disorganization in ALPS patients. Blood. 2014;124(10):1597-1609. 11. Musher DM, Ceasar H, Kojic EM, et al. Administration of proteinconjugate pneumococcal vaccine to patients who have invasive disease after splenectomy despite their having received 23-valent pneumococcal polysaccharide vaccine. J Infect Dis. 2005;191(7):1063-1067. 12. Kristinsson SY, Gridley G, Hoover RN, Check D, Landgren O. Long-term risks after splenectomy among 8,149 cancer-free American veterans: a cohort study with up to 27 years followup. Haematologica. 2014;99(2):392-398. 13. Eastridge BJ, Jenkins D, Flaherty S, Schiller H, Holcomb JB. Trauma system development in a theater of war: experiences from Operation Iraqi Freedom and Operation Enduring Freedom. J Trauma. 2006;61(6):1366-72; discussion 1372-1373. 14. Musher DM, Groover JE, Watson DA, et al. Genetic regulation of the capacity to make immunoglobulin G to pneumococcal capsular polysaccharides. J Investig Med. 1997;45(2):57-68. 15. Gupta S, Gupta A. Selective IgM deficiency - an underestimated primary immunodeficiency. Front Immunol. 2017;8:1056.

Haematologica | 109 February 2024

626

LETTER TO THE EDITOR

Identification of PSMB4 and PSMD4 as novel target genes correlated with 1q21 amplification in patients with smoldering myeloma and multiple myeloma Multiple myeloma (MM) is a malignant plasma cell (PC) dyscrasia characterized by heterogeneous biological features and genetic alterations, resulting in a wide range of disease courses.1,2 Despite all the therapeutic strategies developed in the last three decades, MM is still incurable, and almost all patients will inevitably experience disease progression and eventually relapse.3 Among all the genetic abnormalities, the amplification of the 1q21 region is one of the most frequent cytogenetic abnormalities occurring in malignant PC and it has become a new prognostic factor in MM patients.4,5 The incidence of gain and/or amplification of the 1q21 locus (1q21+) increases with disease progression. It can be detected in around 30-45% of patients with smoldering MM (SMM) and newly diagnosed MM (NDMM), and in around 70% of relapsed/refractory MM patients (RRMM).6 The impact of 1q21 on disease progression at an early stage has not been widely investigated. A few studies have suggested that the acquisition of extra 1q21 copies may play a role in disease progression.7,8 In fact, SMM patients with 1q21+ may be more likely to progress to MM than patients without 1q21+.8 Recent studies have demonstrated that the 1q21 copy number has a different impact on the responsiveness to MM treatments, especially proteasome inhibition (PI).9 PI is a well-established anti-cancer treatment approach used in MM. Throughout the years, the implementation of PI drugs as part of standard MM therapy has continued to improve the quality of life and clinical outcomes of MM patients. Furthermore, additional copies of 1q21 have been associated with PI resistance and recurrence of the disease in patients with 1q21+, limiting the long-term medical utility of PI.9,10 Recent studies have demonstrated that patients with 1q21+ treated with combination treatment with bortezomib (Bor) have inferior progression-free survival and overall survival compared to patients who do not present 1q21+.11 Similar results were observed when patients harboring 1q21 amplification were treated with second-generation PI; however, this study showed that patients with 1q21 gain can greatly benefit from second-generation PI treatment upfront.10 Several genes are known to be deregulated upon the amplification of the 1q21 locus;9 nonetheless, the pathogenic mechanism of how these genes drive disease progression and contribute to the poor outcome in patients with 1q21+ and their possible role as druggable targets is not fully understood. In our study, we analyzed primary MM bone marrow (BM) PC from both SMM and NDMM patients to identify genes whose expressions are deregulated in patients with 1q21+

in correlation with the number of copies and their putative role in drug response. This study was conducted in line with the Declaration of Helsinki. Written consent was obtained from all patients for sample collection and clinical analysis. The Institutional Ethics Committee of Parma Hospital (Parma, Italy) reviewed and approved the study. We evaluated purified CD138+ BM PC from 11 SMM and 18 NDMM patients. The cytogenetic features of all the patients are summarized in Table 1. All the patients underwent fluorescence in situ hybridization (FISH) analysis to detect 1q21 copy number alteration (CNA); 48% of patients presented 1q21+ at the 1q21 locus. Based on the hybridization pattern of each patient, we generated a score representing the 1q21 copy number in each PC sample. The transcriptional profiles of the 29 BM samples were obtained using Gene CHIP ClairomD Arrays (Affymetrix Inc., Santa Clara, CA, USA) as previously described.12 The analyses were performed using R (v4.0.2 in Rstudio v1.3.959). The global expression profiles of 19,012 protein-coding were obtained, analyzed using RMA normalization techniques, and annotated based on the Gencode project (v26) as previously described.13 Annotation data were extracted from Ensembl v102 using the biomaRt package. Unprocessed sequence data from this study have been submitted to the Gene Expression Omnibus (GEO) of the National Center for Biotechnology Information under the accession number GSE227907. We performed a regional amplification analysis using LOESS smoothing of the W statistic and the P value from the comparison between 1q21+ and control samples. Our analysis revealed that the Table 1. Cytogenetic features of patients.

Female, N (%) Male, N (%)

Median age in years (range) Del13q, N (%)

Hyperdiploid, N (%)

SMM N=11

NDMM N=18

3 (27)

8 (44)

8 (73)

67 (38-86)

72 (53-86)

5 (45)

8 (44.4)

6 (55)

del17p, N (%)

4 (36.3)

t(4:14), N (%)

3 (27.2)

Chr14 translocation, N (%) 1q21+, N (%)

del1p32, N (%)

10 (56)

6 (45.4)

8 (72.7) 0 (0)

11 (61.1) 4 (22.2)

7 (38.8)

3 (16.6)

6 (33.3) 3(16.6)

N: number; SMM: smoldering multiple myeloma; NDMM: newly diagnosed multiple myeloma; Del: deletion; Chr: chromosome; t: translocation.

Haematologica | 109 February 2024

627

LETTER TO THE EDITOR

most up-regulated genes were located in the 1q21 region, after regrouping all the genes according to their corresponding positions on chromosome 1 (Figure 1A). The samr package was used in R for differential analyses to identify genes that were differentially expressed between 1q-amplified and control samples. The correlation between the expected numbers of 1q21 copy number was performed using globaltest function in the same R package. Combining these two analyses (most differentially expressed gene between 1q21+ vs. control samples and most correlated gene

expression with 1q21 copy number) we decided to focus on the PSMB4 and PSMD4 genes. The expression analysis from our bioinformatics studies revealed a significant increase in RNA expression of the proteasome subunits PSMB4 and PSMD4 in MM patients with 1q21+ (Figure 1B, D). Consistently, the expression of both subunits was positively correlated with the 1q21 copy number determined by FISH (Figure 1C, E). Next, we characterized the functional roles of PSMB4 and PSMD4 in MM in vitro. We evaluated the RNA expression

Figure 1. Regional amplification analysis across chromosome 1. Regional amplification was evaluated across chromosome 1 (Chr1). The genes with the most significant changes in expression were located in the 1q21 locus (red circle), followed by adjacent regions in the q arm of Chr1. A smoothed color density representation of the scatterplot obtained through a 2D-kernel density estimate (dark blue denotes most significant genes in terms of P value) (A). Expression levels of PSMB4 and PSMD4 in primary multiple myeloma (MM) bone marrow (BM) plasma cells (PC). PSMB4 and PSMD4 expression analysis reveals that both subunits are significantly up-regulated in purified BM PC from patients with 1q21+ when compared with controls (B and D). The gene expression profile was correlated with the 1q21 copy number determined by FISH analysis (C and E). Data were analyzed by Mann-Whitney test. Haematologica | 109 February 2024

628

LETTER TO THE EDITOR

levels of the proteasome subunits in a panel of human myeloma cell lines (HMCL) previously characterized by FISH (data not shown). Notably, our FISH analysis revealed that the HMCL (H929, RPMI, U266, OPM2, and JJN3) carried different degrees of CNA. Our results showed that the RNA levels of PSMB4 and PSMD4 were higher in cell lines with 1q21+ when compared with control cell line OCI (Figure 2A, B). Remarkably, cell lines with a higher number of amplified 1q clones have greater RNA expression. These findings suggest that proteasome subunits PSMB4 and PSMD4 are both up-regulated in relation to the HMCL copy number. To investigate the effects of PI Bor in HMCL, we treated cell lines with 1q21+ (JJN3, U266) with 2nM and 5nM of Bor for 48 hours. Western blot analysis showed that the protein expression of PSMB4 was unaffected by Bor treatment (Figure 2C), while the expression of PSMD4 was down-regulated after treatment (Figure 2C). Furthermore, a MTT cell viability assay performed on the same cell lines after Bor treatment showed that cells treated at 5nM have a remarkable cell mortality when compared with cells treated with 2nM of Bor (Figure 2D).

To further investigate the role of both proteasome subunits in the pathogenesis of MM, we knocked down the expression of PSMB4 and PSMD4 in JJN3, a cell line carrying 1q21+, using short hairpin RNA (shRNA) lentivectors targeting both subunits. PSMB4 knockdown led to a reduction in RNA transcript levels and a drastically increased cell death, presumably associated with the high toxicity accumulated in the cells due to the lack of PSMB4 (data not shown). Moreover, we found that, like PSMB4, the RNA transcripts of PSMD4 were also significantly down-regulated upon the knockdown of PSMD4 (Online Supplementary Figure S1A). Notably, when cells with shPSMD4 were treated with PI Bor, we observed an increase in cell death when compared with the scramble cell line, though our statistical analysis determined that the increase in apoptosis was not significant (Online Supplementary Figure S1C). Similar results were obtained when PSMD4 knockdown cells were treated with carfilzomib (Online Supplementary Figure S1D). These results could be explained by the possible synergistic or additive effect of a not fully functional proteasome due to a partial inhibition of the PSMD4 (Online Supplementary

Figure 2. RNA expression levels of PSMB4 and PSMD4 in human myeloma cell lines. RNA expression of PSMB4 and PSMD4 in myeloma cell lines. OCI was used as a control. Cell lines with a greater number of 1q21 clones have higher RNA expression of PSMB4 and PSMD4 (A and B). Western blot analysis of 1q21+ U266 and JJN3 cell lines treated with bortezomib (Bor) at 2nM and 5nM for 48 hours (C). MTT viability assay of 1q21+ cells treated for 48 hours with 2nM and 5nM of Bor (D). Haematologica | 109 February 2024

629

LETTER TO THE EDITOR

Figure S1B) and the inhibitory effect of the PI in the MM cell that leads to increased cell death when compared with the scramble cell line. In conclusion, our results showed that proteasome subunits PSMB4 and PSMD4 are up-regulated in 1q21+ patients, and that this upregulation is positively correlated to the 1q21 copy number. Interestingly, this correlation was independent of the disease stage (SMM vs. NDMM). Our functional analysis showed that inhibition of PSMD4 in cells with 1q21+ results in increased cell death after treatment with PI when compared with scramble control. The increased drug resistance to all available MM therapies is a significant barrier to long-term patient survival in MM, particularly those with 1q21+. Our findings suggest that PSMD4 can be used as a potential target for the treatment of 1q21+ patients. Combination therapies with next-generation agents such as cereblon E3 ligase modulators (CELMoD), which demonstrated remarkable in vitro potency and enhanced efficacy in RRMM patients,14,15 new-generation PI, and other immunomodulatory drugs could represent an ideal partner for combination therapy. However, further studies are needed to decipher the molecular mechanisms by which MM patients with 1q21+ fail to respond to PI drugs.

Correspondence: N. GIULIANI - nicola.giuliani@unipr.it https://doi.org/10.3324/haematol.2023.283200 Received: March 24, 2023. Accepted: August 14, 2023. Early view: August 24, 2023. ©2024 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures NG received research funding and honoraria from Amgen, BristolMyers Squibb, Takeda, Celgene, Millennium Pharmaceuticals, and Janssen Pharmaceuticals. The other authors have no conflicts of interest to disclose. Contributions JBG wrote the manuscript. LN, MS and BDP provided clinical data and enrolled patients. GS and GT performed the cytogenetic analysis. LA and PS performed the statistical analysis. VF and GD generated the viral vector. JBG, PS, VM and NTI performed the experiments. JBG, DT, VM, PS, VR, OL and NTI collected and processed the samples. NG, GP, PS and GD reviewed the manuscript. NG approved the final version of

Authors

the manuscript. All authors contributed to the article and approved the final version for publication.

Jessica Burroughs Garcia, Paola Storti, Nicolas Thomas Iannozzi, 1

Valentina Marchica,2 Luca Agnelli,3 Denise Toscani,2 Valentina

Acknowledgments

Franceschi, Giannalisa Todaro, Gabriella Sammarelli, Laura

We would like to thank the Associazione Italiana Contro Leucemie,

Linfomi e Mielomi ONLUS, and ParmAIL for their support.

Notarfranchi, Matteo Scita, Benedetta Dalla Palma, Vincenzo 2

Raimondi, Oxana Lungu, Giancarlo Pruneri, Gaetano Donofrio, and 2

Nicola Giuliani1,2

Funding This work was supported by a grant from Ricerca Finalizzata del

Ministero della Salute Italiana (PE-2016-02361261).

Hematology and BMT Unit, “Azienda Ospedaliero-Universitaria di

Parma”, Parma; Department of Medicine and Surgery, University of 2

Parma, Parma; 3Istituto Nazionale dei Tumori Foundation, Milan and

Data-sharing statement

Department of Medical-Veterinary Science, University of Parma,

The data supporting the findings of this study are available from the

Parma, Italy

corresponding author upon reasonable request.

References 1. Pawlyn C, Morgan GJ. Evolutionary biology of high-risk multiple myeloma. Nat Rev Cancer. 2017;17(9):543-556. 2. Zhan F, Huang Y, Colla S, et al. The molecular classification of multiple myeloma. Blood. 2006;108(6):2020-2028. 3. Perrot A, Lauwers-Cances V, Tournay E, et al. Development and validation of a cytogenetic prognostic index predicting survival in multiple myeloma. J Clin Oncol. 2019;37(19):1657-1665. 4. Avet-Loiseau H, Attal M, Campion L, et al. Long-term analysis of the IFM 99 trials for myeloma: cytogenetic abnormalities [t(4;14), del(17p), 1q gains] play a major role in defining longterm survival. J Clin Oncol. 2012;30(16):1949-1952. 5. Hanamura I, Stewart JP, Huang Y, et al. Frequent gain of chromosome band 1q21 in plasma-cell dyscrasias detected by fluorescence in situ hybridization: incidence increases from

MGUS to relapsed myeloma and is related to prognosis and disease progression following tandem stem-cell transplantation. Blood. 2006;108(5):1724-1732. 6. An G, Li Z, Tai YT, et al. The impact of clone size on the prognostic value of chromosome aberrations by fluorescence in situ hybridization in multiple myeloma. Clin Cancer Res. 2015;21(9):2148-2156. 7. Sawyer JR, Tian E, Heuck CJ, et al. Evidence of an epigenetic origin for high-risk 1q21 copy number aberrations in multiple myeloma. Blood. 2015;125(24):3756-3759. 8. Neben K, Jauch A, Hielscher T, et al. Progression in smoldering myeloma is independently determined by the chromosomal abnormalities del(17p), t(4;14), gain 1q, hyperdiploidy, and tumor load. J Clin Oncol. 2013;31(34):4325-4332.

Haematologica | 109 February 2024

630

LETTER TO THE EDITOR 9. Shaughnessy JD Jr, Qu P, Usmani S, et al. Pharmacogenomics of bortezomib test-dosing identifies hyperexpression of proteasome genes, especially PSMD4, as novel high-risk feature in myeloma treated with Total Therapy 3. Blood. 2011;118(13):3512-3524. 10. D’Agostino M, Ruggeri M, Aquino S, et al. Impact of gain and amplification of 1q in newly diagnosed multiple myeloma patients receiving carfilzomib based treatment in the Forte trial. Blood. 2020;136(Suppl 1):38-40. 11. Schmidt TM, Barwick BG, Joseph N, et al. Gain of chromosome 1q is associated with early progression in multiple myeloma patients treated with lenalidomide, bortezomib, and dexamethasone. Blood Cancer J. 2019;9(12):94. 12. Chiu M, Toscani D, Marchica V, et al. Myeloma cells deplete bone marrow glutamine and inhibit osteoblast differentiation

limiting asparagine availability. Cancers (Basel). 2020;12(11):3267. 13. Storti P, Agnelli L, Palma BD, et al. The transcriptomic profile of CD138(+) cells from patients with early progression from smoldering to active multiple myeloma remains substantially unchanged. Haematologica. 2019;104(10):e465-e469. 14. Hansen JD, Correa M, Nagy MA, et al. Discovery of CRBN E3 ligase modulator CC-92480 for the treatment of relapsed and refractory multiple myeloma. J Med Chem. 2020;63(13):6648-6676. 15. Richardson PG, Trudel S, Quach H, et al. Mezigdomide (CC92480), a potent, novel Cereblon E3 ligase modulator (CELMoD), combined with dexamethasone (DEX) in patients (pts) with relapsed/refractory multiple myeloma (RRMM): preliminary results from the dose-expansion phase of the CC-92480MM-001 Trial. Blood. 2022;140:(Suppl 1):1366-1368.

Haematologica | 109 February 2024

631

LETTER TO THE EDITOR

Quantitative evaluation of the clinical severity of hemoglobin H disease in a cohort of 591 patients using a scoring system based on regression analysis α-thalassemia, a common genetic disorder characterized by decreased or absent synthesis of α-globin chains, frequently occurs in Southeast Asia (SEA) with an estimated carrier rate of 17.3-51.5%, meaning that 74-663 of every 10,000 newborns in the region may be affected by α-thalassemia major or intermedia and will have birth defects without intervention.1-4 α-thalassemia intermedia, also termed hemoglobin H (Hb H) disease, is typically caused by genetic defects in three of the four α-globin genes, which can be divided into deletional and non-deletional forms.1,5 The most common deletional Hb H disease genotypes in SEA and southern China are - -SEA/-α3.7 and - -SEA/-α4.2, while the main non-deletional genotype is - -SEA/αCSα (HCS).3-7 In this study, we first aimed to evaluate clinical symptoms by collecting multiple phenotypic indicators from a large cohort of Hb H patients. By referring to a validated scoring system for β-thalassemia intermedia and adjusting some parameters for the phenotypic analysis of Hb H patients,8 we were able to quantitatively determine the clinical severity of Hb H patients’ conditions, based on the phenotypic differences among patients with varied α-thalassemic genotypes. The proposed scoring system may support the evaluation of Hb H disease progression and treatment decisions in clinical practice. A total of 591 patients were recruited via standard sampling based on their basic information, medical history (transfusion dependence, chelation history, etc.), physical examination data, laboratory examination results, and abdominal ultrasound results. We validated the thalassemia (HBA and HBB) genotypes using conventional molecular diagnostic approaches, including gap-Polymerase Chain Reaction (gap-PCR), Sanger sequencing, and multiplex ligation-dependent probe amplification (MLPA).7 All patients provided their informed consent to participate in the original study in accordance with the Declaration of Helsinki. The final cohort comprised seven ethnic groups, of which the southern Chinese Han population was the largest. Of the 591 patients, 224 (38%) had deletional Hb H disease, mainly with - -SEA/-α3.7 (73.21%) and - -SEA/-α4.2 (26.34%) genotypes. The remaining 367 patients (62%) had non-deletional Hb H disease, primarily with HCS (90.74%) and Hb H Quong Sze (HQS) (5.18%) genotypes. The proportion of young patients (aged <18 years) with the non-deletional HCS genotype was significantly higher than that of young patients with the deletional Hb H genotype, indicating that HCS has an earlier age of onset and more severe symptoms. Our analysis of the core indicators of thalassemia

symptoms revealed that HCS patients had an earlier age at first blood transfusion, higher serum ferritin levels, etc. (Table 1). Patients with Hb H Westmead (HWS; a non-deletional type of α-thalassemia7) exhibited relatively mild clinical symptoms as well as higher Hb levels compared to individuals with the - -SEA/-α3.7 and - -SEA/-α4.2 genotypes. These results largely support previous findings regarding the heterogeneity of Hb H patients with non-deletional (- -SEA/αCSα) or deletional (- -SEA/-α3.7 or - -SEA/-α4.2) α-thalassemia genotypes.9,10 After observing the heterogeneity of Hb H patients in this cohort, we further classified the patients into transfusion-dependent thalassemia (TDT) and non-transfusion-dependent thalassemia (NTDT) groups according to their age at first transfusion and annual transfusion frequency, according to the guidelines of the Thalassemia International Federation.11,12 Of the 544 patients, 70 were classified as TDT (including 68 HCS patients and 2 deletional Hb H patients) and 474 were classified as NTDT (including 216 deletional Hb H patients and 258 non-deletional Hb H patients). Notably, of the 47 patients who underwent splenectomy, 39 had a history of blood transfusions before surgery. Of those 39 patients, 13 depended on regular blood transfusions, with severe cases requiring transfusions as frequently as every 20 days. Following surgery, only 6 patients received blood transfusions, 4 of whom received transfusions due to unexpected fevers, and 2 of whom required occasional transfusions during menstruation. These results suggest that splenectomy generally reduces the transfusion dependence of Hb H patients. These results demonstrate a high degree of heterogeneity among Hb H diseases and suggest that their clinical severity cannot be solely predicted by genotype or transfusion dependence. We next attempted to quantitatively evaluate the disease severity of Hb H patients by considering the phenotypic complexity of the disease. To do this, three hematologists first classified the patients’ symptoms as mild, moderate, or severe according to the following six key parameters: 1) Hb at a steady state; 2) age at first blood transfusion; 3) requirement for blood transfusion; 4) spleen size; 5) age at thalassemia presentation; and 6) growth and development.8 To establish and evaluate the scoring system, we randomly divided the 591 patients into a training cohort with 298 cases and a validation cohort with 293 cases (Online Supplementary Figure S1A, B). We used a univariate logistic regression model to assess the associations between the severity classifications and 18 candidate parameters,

Haematologica | 109 February 2024

632

633

Haematologica | 109 February 2024

Laboratory values, mean±SD Hemoglobin, g/L Serum ferritin, ng/mL < 800, N (%) 800-2,500, N (%) > 2,500, N (%) Total bilirubin, μmol/L RBC, ×106/mm3 MCV,b μm3 MCH,c pg MCHC,d g/dL

Growth retardation, N (%)

Age at first blood transfusion in months, mean±SD

2 0

46 (77.97) 11 (18.64) 2 (3.39) 0 0 0 0

28 / 31

32 (54.24) 23 (38.98) 4 (6.78)

59 (26.34)

- -SEA/-α4.2

12 (20.34)

86.06±110.63

99.23±10.95 98.64±9.58 266.93±369.25 266.40±306.25 149 (90.85) 55 (93.22) 15 (9.15) 4 (6.78) 0 0 21.32±12.67 24.89±12.94 5.69±0.71 5.63±0.66 57.34±5.36 57.96±6.30 17.55±1.51 17.63±1.81 306.34±9.18 304.71±10.58

34 (20.73)

101.82±128.45

128.15±142.54 117.55±132.80

2 2

HBB genotype, N β0/β β+/β

Age at thalassemia presentation in months, mean±SD

134 (81.71) 26 (15.85) 3 (1.83) 1 (0.61) 0 0 0

Ethnic group, N (%) Han Zhuang Yao Buyi She Li Miao

79 / 85

94 (57.32) 56 (34.15) 14 (8.54)

Age, N (%) < 18 years 18-35 years > 35 years

Gender, female / male, N

164 (73.21)

- -SEA/-α3.7

N (%)

Parameter

95 42.5 1 (100) 0 0 16 6.07 50.5 15.6 310

0 0

0 1 (100) 0 0 0 0 0

0/1

1 (100) 0 0

1 (0.45)

- -THAI/-α3.7

Deletional Hb H disease, N=224

4 (21.05)

76.94±116.76

55.84±82.12

0 0

15 (78.95) 4 (21.05) 0 0 0 0 0

12 / 7

15 (78.95) 3 (15.79) 1 (5.26)

19 (5.18)

- -SEA/αQSα

1 (10)

159.6±158.9

1 0

10 (100) 0 0 0 0 0 0

5/5

4 (40) 4 (40) 2 (20)

10 (2.73)

- -SEA/αWSα

14.00±14.14

0 0

1 (50) 1 (50) 0 0 0 0 0

1/1

1 (50) 1 (50) 0

2 (0.54)

- -THAI/αCSα

1 (50)

270.00±296.98

0 0

2 (100) 0 0 0 0 0 0

0/2

0 1 (50) 1 (50)

2 (0.54)

- -SEA/αCD30α

0 0

1 (100) 0 0 0 0 0 0

0/1

1 (100) 0 0

1 (0.27)

- -SEA/ αintA-Gα

<0.001 <0.001 <0.001 <0.001 <0.001

<0.001 <0.001 <0.001

0.052

0.38

<0.001

0.031

<0.001

0.598

<0.001

Continued on following page.

93.65±12.37 90.68±8.54 120.90±13.03 83.50±2.12 89.50±13.44 101 1,002.69±989.76 557.28±522.43 133.02±171.53 652.50±762.97 1,689.89±291.96 1,104.5 185 (55.56) 15 (78.95) 10 (100) 1 (50) 0 0 126 (37.84) 4 (21.05) 0 1 (50) 2 (100) 1 (100) 22 (6.60) 0 0 0 0 0 42.85±21.47 41.88±19.81 12.80±5.67 38.70±17.11 56.45±2.47 20.6 4.44±0.55 4.86±0.55 5.88±0.53 4.01±0.18 4.35±0.97 4.55 75.01±5.91 65.74±5.56 65.90±4.24 78.90±2.40 75.15±3.61 75.4 21.34±3.60 18.77±1.63 20.57±1.11 20.80±0.42 20.75±1.49 22.3 282.35±16.07 285.42±10.23 312.40±4.40 263.50±2.12 276.50±6.36 295

111 (33.33)

44.36±54.70

36.72±55.45

2 4

193 (57.96) 133 (39.94) 4 (1.20) 0 1 (0.30) 1 (0.30) 1 (0.30)

175 / 158

280 (84.10) 45 (13.50) 8 (2.40)

333 (90.74)

- -SEA/αCSα

Non-deletional Hb H disease, N=367

Table 1. Phenotypic characterization of 591 hemoglobin H patients based on two major genotypic categories.

LETTER TO THE EDITOR

- -SEA/-α4.2

Haematologica | 109 February 2024

634

141 (89.24) 7 (4.43) 10 (6.33) 5 (3.05) 40 (24.4) 4 (2.44)

Treatment,f N (%) None/Rare transfusion Occasional transfusion Regular transfusion Splenectomy Iron chelation Other drugs 51 (87.93) 4 (6.90) 3 (5.17) 1 (1.69) 17 (28.81) 0

10.47±3.74 1.18±1.91 6.92±2.06 1.58±1.91 0 1 (100) 0 0 0 1 0

7.40 0 6.1 1.1 0

363 0.2 0.89 2.35 0 0 0.73 2.66 4.35

- -THAI/-α3.7

- -SEA/αQSα

- -SEA/αWSα

- -THAI/αCSα

70 (23.89) 45 (15.36) 178 (60.75) 39 (11.71) 99 (29.73) 7 (2.10)

11.29±6.77 3.00±2.19 7.50±1.45 3.28±1.63 30 (9.01) 13 (68.42) 2 (10.53) 4 (21.05) 2 (15.53) 10 (52.63) 0

10.30±5.05 2.86±3.20 7.07±1.30 2.96±1.78 1 (5.26)

10 (100) 0 0 0 0 0

8.80±1.32 0 7.42±1.69 0.68±1.45 0

1 (50) 1 (50) 0 0 1 (50) 0

5.34±7.55 1.75±2.47 8.01±0.70 5.03±0.24 0

322.90±145.22 349.68±129.11 301.00±64.09 474.00±268.70 0.41±0.17 0.41±0.11 0.10±0.03 0.68±0.20 0.82±0.31 0.69±0.18 0.82±0.48 0.72±0.04 1.26±0.74 1.36±0.34 3.05±1.83 0.67±0.14 2.05±1.53 0 0 2.63±1.39 7.17±4.92 20.78±5.36 0 11.61±7.88 5.86±3.78 3.04±4.36 0 4.51±6.39 1.55±0.35 1.83±0.34 2.29±0.30 1.48±0.23 9.47±3.13 6.36±1.83 1.66±0.59 7.10±3.08

- -SEA/αCSα

1 (50) 0 1 (50) 0 1 (50) 0

17.7±4.95 5.73±2.02 8.75±0.73 3.37±3.80 1(50)

152.50±34.65 0.36±0.14 0.75±0.19 1.09±0.16 0 18.65±5.18 3.25±4.60 1.59±0.23 5.78±2.07

- -SEA/αCD30α

Non-deletional Hb H disease (N=367)

0.081 <0.001 0.0046 <0.001 0.011

0.060 <0.001 0.039 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001

0 <0.001 0 1 (100) 0 0.035 0 0.059 0 0.988

14.4 3.6 9.20 3.5 0

390 0.28 0.88 2.2 0 8.79 11.47 2.06 3.93

- -SEA/ αintA-Gα

N: number; Hb: hemoglobin; RBC: red blood cells; SD: standard deviation; SEA: Southeast Asia; THAI: Thailand; CS: Constant Spring; QS: Quong Sze; WS: Westmead; −α3.7: Single-gene deletion; −α4.2: Single-gene deletion; αCD30(-GAG): HBA2:c.91_93del; αintA-G: initiation codon(ATG>GTG): HBA2:c.1A>G. aP: χ2 test for categorical variables and one-way ANOVA for continuous variables or discrete variables. bMCV: mean corpuscular volume. cMCH: mean corpuscular Hb. dMCHC: mean corpuscular Hb concentration. esTfR: serum transferrin receptor. f Regular transfusion: >4 times/year; occasional transfusion: 1-3 times/year; None/Rare transfusion: blood transfusion therapy was never given, or transfusions were given every few years. Patients with splenectomy were excluded from blood transfusion events. Since this was a cross-sectional observation study, the transfusion treatment events of patients with splenectomy before surgery could not be fully counted.

9.86±3.47 0.84±1.58 6.83±2.13 1.73±1.75 11 (6.71)

329.93±113.08 319.39±103.43 0.24±0.08 0.25±0.09 0.88±0.41 0.86±0.22 2.09±1.18 2.02±0.96 0 0 3.00±3.13 4.39±3.98 1.36±2.23 2.33±3.27 2.01±0.31 1.94±0.38 3.84±0.99 4.10±1.05

- -SEA/-α3.7

Abdominal ultrasonography, mean±SD Spleen size, cm Subcostal spleen size, cm Liver size, cm Subcostal liver size, cm Cholelithiasis, N (/%)

Platelets, x109/L Reticulocytes, x1012/L Fetal Hb, % Hemoglobin A2, % Hemoglobin CS, % Hemoglobin H, % Hemoglobin Bart’s, % Transferrin, g/L STfR,e mg/L

Parameter

Deletional Hb H disease (N=224)

LETTER TO THE EDITOR

estimating Odds Ratios (OR) with 95% Confidence Intervals (CI) for the training cohort (Online Supplementary Table S1). We selected variables with P values <0.1 in the univariate regression analysis and used an ordinal logistic regression model to determine the extent to which independent variables affected the dependent variable. P<0.05 was considered statistically significant. The results of the multivariate logistic regression analysis showed that the best model combined seven parameters: Hb level, age at thalassemia presentation, age at first blood transfusion, transfusion frequency, subcostal spleen size, growth and development, and soluble transferrin receptor (sTfR) level (Online Supplementary Table S2). There was no evidence of multicollinearity among these parameters, suggesting that each criterion was independently associated with disease severity. To simplify the scoring process and avoid using complex mathematical formulae, we classified each of the seven significant criteria into three levels based on the severity of the phenotype. We then assigned a score of 0, 1, or 2 points to each criterion to reflect increasing levels of severity (Table 2). This allowed us to establish a final clinical scoring system for assessing disease severity in patients with Hb H disease, with total scores ranging from 0-14 (Table 2). To determine the relevant total score thresholds, we tested various combinations of cut-off values, including 4, 5, 6, 8, 9, and 10. We found that the patients’ disease severity was best distinguished by a severity cut-off score of 8. Based on these results, we assigned the following Hb H disease severity categories: mild (0-5), moderate (6-8), and severe (9-14) (Table 2). We then used the validation cohort to test the performance of the scoring system. The scoring system correctly identified 114 mild cases, 101 moderate cases, and 78 severe cases, and the

consistency rates according to the 6 objective parameters were 89.47% (mild), 85.15% (moderate), and 87.18% (severe), respectively (Figure 1A-C), suggesting the practicability of this scoring system for the clinical assessment of Hb H diseases. In addition, 130 out of the 591 patients had blood transfusions more than 8 times a year, of which 95 (73%) were classified in the severe group, indicating that our proposed model could sensitively distinguish those patients who needed regular transfusions as having “severe” conditions. Our results support the use of this newly established scoring system for the clinical assessment and management of Hb H disease. Based on our scoring system, we categorized 219 of 591 (37%) Hb H patients as mildly affected, 210 (36%) as moderately affected, and 162 (27%) as severely affected. Our results indicate that most of the deletional Hb H disease patients had a relatively mild phenotype, with scores ranging from 0-9 and a mean value of 3.5. In comparison, the mean value for non-deletional Hb H patients was 8.1, significantly higher than for the former group (P<0.0001) (Figure 1D). Notably, 13 of the 591 Hb H patients were also β-thalassemia carriers (Table 1). Coinheritance of β-thalassemia mutations seemingly reduced the severity scores, especially among the HCS patients with critical P values (P=0.054) (Online Supplementary Figure S1C). Of the 224 deletional Hb H patients, 185 cases (83%) were classified as mild, whereas only 3 cases (1%) were classified as severe (Online Supplementary Figure S1D). We found no difference in disease severity between - -SEA/-α3.7 patients and - -SEA/-α4.2 patients in the deletional Hb H disease group (P=0.42), with mean scores of 3.4 and 3.7, respectively (Figure 1E). However, all 3 cases of deletional Hb H patients classified as severe were - -SEA/-α3.7 patients, with

Table 2. Scoring criteria and weighted effects of these on the severity outcome according to the regression model. Clinical criteria

Total severity score

Points scoreda 0

Age at thalassemia presentation in months

>120

60-120

<60

Age at receiving first blood transfusion in months

>120

36-120

<36

Transfusion frequency, times/year

0-3

4-8

>25th

3rd–25th

<3rd

Subcostal spleen size, cm

03/05/23

>5 or splenectomized

Hemoglobin, g/L

>90

75-90

<75

sTfRc, mg/L

0-4.32

4.32-4.92

>4.92

Growth and developmentb

Severity category Mild Moderate Severe

0-5 6-8 9-14

The weighted-score was obtained by dividing the Odds Ratio of the criteria by the smallest significant Odds Ratio obtained from the multivariable logistic regression model and rounding the resulting number to the 1 or 2. bPercentile of growth development was assessed based on weight and height measurements plotted on a China standard growth chart. CsTfR: soluble transferrin receptor.

Haematologica | 109 February 2024

635

LETTER TO THE EDITOR

Figure 1. Distribution of disease severity in 591 hemoglobin H patients with different HBA genotypes. (A-C) Pie chart showing the distribution of clinical disease severity for patient classification by hematologists into mild, moderate, and severe groups using a severity scoring system, respectively. (D) Bar-scatter plot showing the severity score comparison between the deletional hemoglobin (Hb) H disease group and non-deletional Hb H disease group. (E) Bar-scatter plot showing the distribution of severity score for each genotype in the deletional Hb H disease cohort. (F) Bar-scatter plot showing the distribution of severity score for each genotype in the non-deletional Hb H disease cohort. (Mann-Whitney U test: P<0.05).

a score of 9. These patients, aged between 4 and 5 years old, exhibited severe delays in growth and development. In contrast, non-deletional Hb H disease patients generally had a more severe phenotype, with 159 out of 367 cases (43%) classified as severe (Online Supplementary Figure S1D). In the non-deletional Hb H disease group, HCS patients displayed the most severe symptoms and HWS patients the mildest (P<0.0001) (Figure 1F). Interestingly, we identified a subset of HCS patients (N=19, 6%) with mild symptoms and a score range of 3-5. None of these patients had severe growth retardation, and their Hb levels ranged from 71-122 g/L (93g/L on average). Notably, none of these patients had received regular blood transfusions in the past year, and 6 had never received blood transfusions. These findings suggest that other potential genetic factors may contribute to these unusually mild symptoms. The heterogeneity and phenotype predictions of Hb H patients have not been comprehensively documented. To address this gap, we first established a clinical scoring system for Hb H disease severity that considered various

aspects of the disease.8 Through these efforts, we confirmed less severe clinical symptoms in the deletional than in the non-deletional Hb H patients. Meanwhile, the clinical severity of these patients’ conditions was highly heterogeneous, including some cases that defied the conventional understanding of the disease (Online Supplementary Figure S1E, F). Our study highlights the need to consider the complexity of Hb H disease when investigating its genetic modifiers. Despite these findings, our study has some limitations, such as the lack of a long-term patient follow-up, and the limited genotypes of Hb H patients due to the sampling including only Chinese ethnic groups. Nevertheless, we were able to leverage the available clinical records to extract valuable information on patients’ growth, development, and treatment history, and to reflect on the understanding of disease progression and clinical outcomes. For example, we identified a group of HCS patients (N=57) with marginal values between the “severe” and “moderate” groups. Their scores were among the highest in the moderate group, but

Haematologica | 109 February 2024

636

LETTER TO THE EDITOR

few of them received regular transfusions (score = 8). They were assigned “warning” scores due to their low pre-transfusion Hb levels, high levels of serum ferritin representing iron overload, bilirubin (an indicator of hemolysis), and hepatosplenomegaly. The scoring system for Hb H disease proposed in this paper can serve as an effective tool for accurate classification and phenotype prediction of Hb H diseases, which may facilitate better treatment decisions and prognostic predictions. In addition, the scoring system can be used as a tool for prenatal diagnosis and genetic counseling, enabling families with a history of Hb H disease to make more informed decisions. In summary, our study contributes to the understanding of Hb H disease phenotypes and provides a framework for further genetic and clinical investigations. By addressing the limitations of previous research, we hope to promote more comprehensive and personalized approaches to Hb H disease diagnosis, treatment, and prevention.

Hospital, Affiliated to Fujian Medical University, Longyan, Fujian, China; 12Department of Medical Dispute, Maternal and Child Health Hospital, Heyuan China Heyuan, Guangdong, China; 13Maternal and Child Health Hospital of Yongzhou City, Yongzhou, Hunan, China and Biologics, Gene and Cell Therapy, Frontage Laboratories, Exton, PA,

USA Correspondence: X. XU - xixm@smu.edu.cn; gzxuxm@pub.guangzhou.gd.cn Y. YE - yeyuhua_genetics@163.com X. ZHANG - zxh303xy@163.com https://doi.org/10.3324/haematol.2023.283211 Received: March 24, 2023. Accepted: August 21, 2023. Early view: August 31, 2023. ©2024 Ferrata Storti Foundation Published under a CC BY-NC license

Authors

Disclosures Yumeng Liu, Yuan Zhuang, Jianhong Chen, Zeyan Zhong, Jianpei 1,2

1,2

Fang, Xinyu Li, Bin Xiao, Pingping Li, Bin Lin, 4

No conflicts of interest to disclose.

6,7

Zhenzhong Tao,

6,7

Yidan Liang, Peng Lin, Xingmin Wang, Mengyang Song,1,2 Hualei 1,2

1,2

Contributions

1,2

Luo, Lang Qin, Li Huang, Jufang Tan, Hailiang Li, Tianyu Zhong,

YL, YY, YZ and XX designed the study. YL and YZ organized the

Lian Yu, Zhixiang Liu, Deguo Tang, Yongzhong Zhao, Xinhua

clinical data and conducted the analyses. XZ, JC and JF assessed

Zhang, Yuhua Ye

the patients’ clinical symptoms. BL and ZT conducted genetic

1,2

and Xiangmin Xu

1,2

testing. YL, PiL, XW, MS, HuL, LQ and LH collected the clinical and Innovation Center for Diagnostics and Treatment of Thalassemia,

genetic data, and prepared the blood DNA samples. ZZ, XL, BX, PeL,

Nanfang Hospital, Southern Medical University, Guangzhou,

JT, HaL, TZ, LY, ZL and LH performed clinical hematologic and

Guangdong, China; Department of Medical Genetics, School of

ultrasound examinations. YL, YY and XX drafted the paper. XX

Basic Medical Sciences, Southern Medical University, Guangzhou,

supervised the study.

Guangdong, China; Department of Medical Genetics and Prenatal 3

Diagnosis, Huizhou First Maternal and Child Health Care Hospital,

Acknowledgments

Huizhou, Guangdong, China; Department of Pediatric Hematology/

The authors would like to thank Dr. Zhiyu Peng, Dr. Jiguang Peng

Oncology, Children’s Medical Center, Sun Yat-sen Memorial Hospital,

and Dr. Lipei Liu for helpful discussion of the statistical method.

Sun Yat-sen University, Guangzhou, Guangdong, China; Department 5

of Hematology, 923rd Hospital of the People’s Liberation Army,

Funding

Nanning, Guangxi, China; Guangzhou Huayin Healthcare Group Co.

This study was supported by National Natural Science Foundation of

Ltd., Guangzhou, Guangdong, China; Guangzhou Jiexu Gene

China (Grant N. 31871265), the National Key Research and

Technology Co. Ltd., Guangzhou, Guangdong, China; Prenatal

Development Program of China (grant N. 2018YFA0507800 and

Diagnosis Center, Chenzhou First People’s Hospital, Chenzhou,

2018YFA0507803) and Shanghai Municipal Science and Technology

Hunan, China; Department of Laboratory Hematology, The First

Major Project (Grant N. 2017SHZDZX01).

Affiliated Hospital of Gannan Medical University, Ganzhou, Jiangxi, China; 10Department of Laboratory Medicine, The First Affiliated

Data-sharing statement

Hospital of Gannan Medical University, Ganzhou, Jiangxi, China;

The data that support the findings of this study are available on

Department of Hematology and Rheumatology, Longyan First

request from the corresponding author.

References 1. Taher AT, Weatherall DJ, Cappellini MD. Thalassaemia. Lancet. 2018;391(10116):155-167. 2. Williams TN, Weatherall DJ. World distribution, population

genetics, and health burden of the hemoglobinopathies. Cold Spring Harb Perspect Med. 2012;2(9):a011692. 3. Weatherall DJ. The evolving spectrum of the epidemiology of

Haematologica | 109 February 2024

637

LETTER TO THE EDITOR thalassemia. Hematol Oncol Clin North Am. 2018;32(2):165-175. 4. Goh LPW, Chong ETJ, Lee PC. Prevalence of alpha(α)thalassemia in Southeast Asia (2010-2020): a meta-analysis involving 83,674 subjects. Int J Environ Res Public Health. 2020;17(20):7354. 5. Piel FB, Weatherall DJ. The α-thalassemias. N Engl J Med. 2014;371(20):1908-1916. 6. Shang X, Peng Z, Ye Y, et al. Rapid targeted next-generation sequencing platform for molecular screening and clinical genotyping in subjects with hemoglobinopathies. EBioMedicine. 2017;23:150-159. 7. Xiong F, Sun M, Zhang X, et al. Molecular epidemiological survey of haemoglobinopathies in the Guangxi Zhuang Autonomous Region of southern China. Clin Genet. 2010;78(2):139-148. 8. Sripichai O, Makarasara W, Munkongdee T, et al. A scoring system for the classification of beta-thalassemia/Hb E disease

severity. Am J Hematol. 2008;83(6):482-484. 9. Lal A, Goldrich ML, Haines DA, Azimi M, Singer ST, Vichinsky EP. Heterogeneity of hemoglobin H disease in childhood. N Engl J Med. 2011;364(8):710-718. 10. Chen FE, Ooi C, Ha SY, et al. Genetic and clinical features of hemoglobin H disease in Chinese patients. N Engl J Med. 2000;343(8):544-550. 11. Cappellini MD, Cohen A, Porter J, Taher A, Viprakasit V, eds. Guidelines for the management of transfusion dependent thalassaemia (TDT) [Internet]. 3rd ed. Nicosia (CY): Thalassaemia International Federation; 2014. 12. Taher A, Vichinsky E, Musallam K, Cappellini MD, Viprakasit V, Weatherall D, eds. Guidelines for the management of non transfusion dependent thalassaemia (NTDT) [Internet]. Nicosia (Cyprus): Thalassaemia International Federation; 2013.

Haematologica | 109 February 2024

638

LETTER TO THE EDITOR

Thiostrepton induces cell death of acute myeloid leukemia blasts and the associated macrophage population It is increasingly becoming clear that the tumor immune microenvironment (TME) plays a critical role in tumor progression and drug resistance. For instance, we and others have identified that the presence of tumor supportive M2-polarized macrophages associates with the poorest prognosis in patients with acute myeloid leukemia (AML).1,2 Functionally, we could show for the first time that M2-polarized macrophages were able to increase homing, self-renewal potential, alter the metabolome, and increase transformation capacity of leukemic blasts in vivo.1 Thus, it would be of clinical interest to target this tumor-supportive cell population in order to improve outcome in AML patients. Given the inherent plasticity of macrophages, several studies have aimed to identify factors that would repolarize tumor-supportive M2 macrophages into tumor-suppressive M1 macrophages. Blockage of TIGIT, as well as knockout of GFI1 or upregulation of IRF7 were shown to promote an M2 to M1 repolarization.3-5 In the context of AML, however, it is possible that macrophages also carry leukemic mutations,1,6 thereby potentially impacting on macrophage function even in case M1 repolarization would be successful. Thus, an alternative to macrophage repolarization would be to target this subpopulation via drug-induced apoptosis. Previously, we identified the NAMPT inhibitor KPT-9274 as well as the antibiotic thiostrepton as potential candidates. Thiostrepton targets the oncogene FOXM1,7 but was recently also suggested to have M2 to M1 macrophage-repolarizing activity.8 Furthermore, thiostrepton was reported to have anti-cancer effects in solid tumors9,10 and lymphoid malignancies, such as multiple myeloma.11 Therefore, we investigated the efficacy of thiostrepton in AML and focused on its mode of action. First, we evaluated whether thiostrepton would drive M2 to M1 macrophage repolarization. Human peripheral blood mononuclear cell (PBMC)-derived monocytes were isolated from healthy PBMC by adherence and cultured for 6 days in M-CSF (50 ng/mL). Thereafter, macrophages were either kept in M-CSF as non-polarized M0 macrophages, or they were differentiated towards M1 (20 ng/mL interferon γ [IFNγ] and 100 ng/mL lipoplysaccharide [LPS]) or M2 macrophages (20 ng/mL interleukin [IL] 4 for M2a and 20 ng/mL IL6 for M2d) for 24 hours (hrs). Macrophage phenotypes were characterized morphologically (Figure 1A), by flow cytometry (Figure 1B, C; Online Supplementary Figure 1A; whereby M1-polarized macrophages expressed CD80 and CD86, and M2-polarized macrophages expressed CD163 and CD206) and by gene expression profiles (Figure 1E). Both CD80, CD86 as well as CD163, CD206 are well established markers for M1 and M2 macrophages, respectively, whose expression is

induced upon cytokine polarization.12 Upregulation of CD80 and CD86 is associated with bactericidal activities, while increased expression of CD163 and CD206 is linked to immunomodulation and tissue repair.12,13 In order to study the effects of thiostrepton, non-polarized M0 macrophages, or M2a- or M2d-polarized macrophages were treated for 24 hrs and results were compared to vehicle-treated (0.01% dimethyl sulfoxide [DMSO]) macrophages. When non-polarized M0 macrophages were treated with thiostrepton, we observed a slight decrease in CD206 expression in a dose-dependent manner (Online Supplementary Figure S1B). At a dose of 10 µM, thiostrepton did not upregulate any of the M1 markers in M2-polarized macrophages. The only significant differences that were observed were slight reductions in CD86 mean fluorescence intensity (MFI) and CD206 MFI in M2a-polarized macrophages as well as CD163 MFI in M2d-polarized macrophages (Figure 1D; Online Supplementary Figure S1C). Gene expression analysis of a wide range of M1 and M2 markers (16 in total) in macrophages treated with thiostrepton provided no clear indication of an M2 to M1 repolarization. Instead several M1 markers (e.g., CD86, IL1B) as well as M2 markers (e.g., CD206, CD163, MMP9) were downregulated upon thiostrepton treatment (Figure 1E). While we are aware that these repolarization studies were performed on healthy PBMC-derived macrophages and not on AML associated macrophages (AAM) macrophages, similar to previous studies,8 these data do not support the notion that thiostrepton acts as an M1-repolarizing agent. Next, we wished to study how thiostrepton would impact on M2 macrophage function. AML and lymphoma cell lines were cultured on murine and human M2- polarized macrophages that were pretreated with thiostrepton for 12 hrs or 9 days, to study direct and long-term effects. As controls, leukemic cells were cultured on M1-polarized macrophages, which impaired cell proliferation, in particular of AML cell lines (Online Supplementary Figure S1D). However, thiostrepton pretreatment of M2-polarized macrophages did not convert these into tumor-suppressive macrophages (Online Supplementary Figure S1D). No apoptosis was induced when AML cells were grown on thiostrepton-pretreated human M2d macrophages (Online Supplementary Figure S1E), and proliferation was also not impaired when murine thiostrepton-pretreated macrophages were used (Online Supplementary Figure S1F). Previous studies showed that the secretion of lactate can repolarize M1 to M2 macrophages.14 Extracellular flux measurements confirmed that lymphoma cell lines were low in oxygen consumption rates (OCR) and tended to be more glycolytic, possibly explain-

Haematologica | 109 February 2024

639

LETTER TO THE EDITOR

ing their increased resistance to M1 macrophages (Online Supplementary Figure S1G). While we could not find compelling evidence for M2 to M1 macrophage reprogramming by thiostrepton, we did note that the viability of M2 macrophages was affected at the highest dose (10 µM) of thiostrepton (Figure 1F, G). Loss in

viability was also accompanied by morphological changes, whereby thiostrepton-treated macrophages displayed a round shaped appearance compared to the long stretchedout macrophages treated with vehicle (Figure 1F). Next, we wondered whether thiostrepton would also exhibit antileukemic potential towards leukemic cells directly.

A B

E D

Continued on following page. Haematologica | 109 February 2024

640

LETTER TO THE EDITOR

Figure 1. Thiostrepton treatment induces macrophage cytotoxicity but not M2 to M1 repolarization. (A) Human macrophages were cultured for 24 hours (hrs) in the presence of lipopolysaccharide (LPS) (100 ng/mL) and interferon γ (IFNγ) (20 ng/mL) for M1 polarization, interleukin (IL)4 (20 ng/mL) for M2a polarization, and IL6 (20 ng/mL) for M2d polarization. Representative pictures of M1, M2a and M2d macrophage morphology. Scale bar: 200 µm. (B) Bar plot comparing the levels (mean fluorescence intensity [MFI]) of M2 (CD163 and CD206) and M1 (CD80 and CD86) markers measured by flow cytometery measurements (fluorescence-activated cell sorting [FACS] in healthy macrophages activated with respective cytokines for 24 hrs. M0 macrophages were generated with 50 ng/ mL of MCSF (upper panels). (C) Representative histograms of macrophage marker expression in different macrophage subtypes. (D) Bar plot comparing the levels (MFI) of M2 (CD163 and CD206) and M1 (CD80 and CD86) markers measured by FACS in healthy activated M2a and M2d macrophages treated with dimethyl sulfoxide (DMSO) or thiostrepton (Thio) (10 µM) for 24 hrs. (E) Heatmap displaying the transcript levels of the indicated genes in M1, M0, M2a and M2d macrophages treated with vehicle (DMSO) or thiostrepton (10 µM) for 24 hrs. **P<0.01, ***P<0.001 indicate significance compared to all other groups. Genes in blue indicate M2-associated genes and genes in red indicate M1-associated genes. (F) Cell morphologies of M2a and M2d activated macrophages treated with vehicle or thiostrepton (10 µM) for 24 hrs. Scale bar: 200 µm. (G) Bar plot comparing the viability (DAPI- cells) of M0, M1, M2a, and M2d macrophages treated with thiostrepton and vehicle for 24 hrs. One-way (F, left panel) or two-way (B-D) analysis of variance (ANOVA); *P<0.05, **P<0.01, ***P<0.001. (F, right panel) Wilcoxon signed rank test (two-sided), *P<0.05.

Indeed, thiostrepton also displayed cytotoxic effects in leukemic cell lines with median effective dose (ED50) values ranging from 1.91 µM to 38.67 µM (n=16 AML cell lines; n=2 B-cell acute lymphoblastic leukemia [B-ALL]/lymphoma cell lines; Figure 2A). Similar observations were made by a recent publication showing thiostrepton-mediated apoptosis in a panel of B-pre-ALL cell lines.15 In order to unravel the molecular mechanisms underlying thiostrepton sensitivity in AML, we ranked AML cells treated with thiostrepton from most sensitive to most resistant based on the continuous ED50 values obtained, which we then correlated with the gene expression data retrieved from the CCLE RNA sequencing of the treated AML cell lines (Figure 2B). Gene ontology (GO) and gene set enrichment analysis (GSEA) indicated that cells resistant to thiostrepton displayed increased expression of genes regulated by the SREBP family of transcription factor, whereas sensitivity was linked to increased mitochondrial metabolism, fatty acid metabolism and oxidative phosphorylation (OXPHOS) terms (Online Supplementary Figure S1H; Figure 2C). In order to investigate whether thiostrepton treatment would target both the tumor cells as well as their supportive microenvironment, we used our previously published workflow to investigate cytotoxicity in ex vivo treated primary AML samples1 with which efficacy of drugs can be evaluated on all compartments (Online Supplementary

Figure S2A). Thiostrepton induced a significant increase in cell death in a dose-dependent manner in ex vivo treated primary AML blast cells as well as in the myeloid/monocytic tumor supportive subpopulation (n=22 patients) (Figure 2D), while cytarabine (AraC) and venetoclax (VEN) mainly targeted the leukemic blast population. Our results indicate that thiostrepton targets both the M2 (CD163+ and CD206+) as well as the M1 (CD80+) macrophage population in AML patients. Ideally, M1 macrophages would be spared to promote tumor suppression. However, patients with a large population of M2 macrophages typically have low levels of M1 macrophages and it is this patient group that displayed the poorest prognosis,1 and we, therefore, propose that in this patient group thiostrepton treatment could provide therapeutic benefits. Evaluation of healthy CD34+ showed no signs of impaired viability upon thiostrepton treatment and progenitor frequencies were not affected, suggesting that thiostrepton treatment might provide an attractive therapeutic approach (Online Supplementary Figure S2B). We wished to identify AML patients that would benefit most from thiostrepton treatment. Therefore, we correlated various AML features such as the presence, size and polarization state of macrophages, as well as metabolic features such as basal oxygen consumption rate (OCR) and the amount of mitochondria, with sensitivity to thiostrepton. The macrophage/monocytic tumor supportive

Haematologica | 109 February 2024

641

LETTER TO THE EDITOR

Figure 2. Thiostrepton can target both the leukemic cells and the tumor microenvironment in acute myeloid leukemia. (A) Median effective dose (ED50) of thiostrepton in a panel of acute myeloid leukemia (AML) and B-cell acute lymphoblastic leukemia (B-ALL)/lymphoma cell lines. (B) Hockey stick plot displaying the correlation between RNA-sequencing data from AML cell lines (CCLE dataset) and thiostrepton sensitivity. Expression of SREBP target genes was correlated with thiostrepton insensitivity. (C) Gene set enrichment analysis (GSEA) using the Pearson correlations depicted in (B) normalized enrichment score (NES) and false discover rate (FDR-q) are indicated. (D) Ex vivo drug-induced apoptosis of thiostrepton in a set of 22 primary AML samples (left panel). Cells were treated for 72 hours, and apoptosis was evaluated by Annexin-V/ DAPI staining by flow cytometry. Leukemic blast cells were analyzed in the CD45dimCD34+CD117+ population, while macrophages were analyzed in the CD45highCD14+CD163+ population. (E, F) Volcano plot demonstrating the Spearman correlation between the ex vivo thiostrepton-induced apoptosis (reported as reverse area under the curve [rAUC]) on M2 macrophage (E) and AML blasts (F) treated ex vivo in our set of primary AML samples (N=22 samples). Blue dots indicate significant negative correlation and red dot indicate a positive correlation. Bar graphs represent the mean ± standard error of the mean of at least 3 independent experiments. *P<0.05, **P<0.01, ***P<0.001. UMCG: University Medical Center Groningen.

Haematologica | 109 February 2024

642

LETTER TO THE EDITOR

compartment (Figure 2E) as well as the leukemic blasts themselves (Figure 2F) were most sensitive to thiostrepton-induced apoptosis in AML samples with high M2 macrophage content, increased mitochondrial mass and higher OCR, potentially highlighting an increased sen-

sitivity in AML patients that are more OXPHOS-driven.16 These data are in line with our cell line data in which we observed that the most sensitive cell lines like MV411, MOLM13 and THP1 are rather high in OCR, while the least sensitive lines like HL60 are glycolytic.17 We then

Figure 3. Decreased mitochondrial membrane potential and increased reactive oxygen species upon thiostrepton treatment. (A) Mitochondrial membrane potential was detected by flow cytometry in acute myeloid leukemia (AML) blasts, gated on CD45+CD34+ (or CD117+ for NPM1 mutant AML)/Annexin V- and the myeloid-associated monocytic population (gated on SSChighCD45highCD163+ cells) using the TMRE staining method. Cells were treated with vehicle dimethyl sulfoxide [DMSO], cytarabine (AraC, 250 nM), venetoclax (VEN, 500 nM), or thiostrepton (1-5-10 μM) for 72 hours (hrs). Bar graphs represent the mean ± standard error of the mean, each point represents a patient sample. (B) Fold change in lipid reactive oxygen species (ROS) (C11-BODIPY), cytoplasmic ROS (DCFDA), and mitochondrial superoxide levels (MitoSOXTM) in MV4-11, MOLM13 and HL60 cells treated with thiostrepton (5-10 μM), cytarabine (AraC, 750 nM) or venetoclax (VEN, 200 nM) for 72 hrs; N=4 independent experiments. (C) Apoptosis levels (left panel) and viable cell counts (relative to vehicle control, right panel) measured by flow cytometry in HL60 cells treated with vehicle or with increasing concentrations of thiostrepton (0.5-1 μM) and/or dipyridamole (DP, 5 μM) or KPT-9274 (750 nM) for 72 hrs using an APC-Annexin V/DAPI staining method. Bar graphs represent the mean ± standard error of the mean of at least 3 independent experiments. *P<0.05, ** P<0.01, ***P<0.001. UMCG: University Medical Center Groningen.

Haematologica | 109 February 2024

643

LETTER TO THE EDITOR

correlated thiostrepton-sensitivity to RNA-sequencing data we generated for the ex vivo treated primary AML at diagnosis, which again revealed that OXPHOS-driven cells were more sensitive to thiostrepton, while terms such ferroptosis control and SREBP-controlled genes were associated with thiostrepton resistance (Online Supplementary Figure S2C). Subsequent analysis of the mitochondrial membrane potential (measured by TMRE staining) indicated that both thiostrepton (10 µM) and AraC decreased the mitochondrial membrane potential in leukemic blasts, but only thiostrepton was also able to reduce the mitochondrial membrane potential of the macrophage/monocytic subpopulation (Figure 3A). Given the link between thiostrepton insensitivity and increased ferroptosis and gene expression control of processes associated with lipid peroxidation, we questioned whether thiostrepton would induce cell death via the induction of reactive oxygen species (ROS), ultimately resulting in lipid ROS formation and induction of ferroptosis.18 In order to study the role of oxidative stress, we measured the level of mitochondrial superoxide, cytoplasmatic ROS, and lipid ROS using compartment-specific and redox-sensitive fluorescent dyes. Short-term exposure to thiostrepton (24 hrs) increased ROS levels in all three compartments in MV4-11 cells, while in MOLM13 cells we observed a particularly strong upregulation of cytoplasmic ROS (Figure 3B). In stark contrast, thiostrepton was not able to induce ROS in thiostrepton-resistant HL60 cells, while treatment with AraC did induce cytoplasmic ROS (Figure 3B). Finally, considering that resistance to thiostrepton was associated with the activation of SREBF/ SREBP pathway, we decided to test if the Food and Drug Administration-approved phosphodiesterase inhibitor dipyridamole (DP) and/or the NAMPT inhibitor (KPT-9274) could promote cytotoxicity in leukemic cells resistant to thiostrepton. Indeed treatment of HL60 cells with thiostrepton in combination with DP and KPT-9274 induced a significant increase in cell death compared to thiostrepton alone (Figure 3C). SREBP act as key transcriptional regulators of genes involved in lipid homeostasis, associated with lipid detoxification by refueling the monosaturated fatty acid pool. Inhibition of such pathways sensitizes cells to thiostrepton-induced cell death. Therapeutic approaches aimed at eradicating the tumor-supportive microenvironment are gaining interest. Thiostrepton treatment provides such a therapeutic opportunity, by inducing cytotoxicity in the leukemic blasts but also in the supportive macrophage/monocytic compartment of AML patients. While OXPHOS-driven AML subtypes harboring relatively large M2-macrophage compartments are most sensitive, the thiostrepton-resistant AML subtypes can be sensitized by co-treatment with SREBF/SREBP pathway inhibition.

Authors Isabel Weinhäuser,1,2,3* Diego A. Pereira-Martins,1,2,3* Jacobien R. Hilberink,1* Annet Brouwers-Vos,1 Eduardo M. Rego,3 Gerwin Huls1 and Jan Jacob Schuringa1 1

Department of Experimental Hematology, University Medical Center

Groningen, University of Groningen, Groningen, the Netherlands; 2

Department of Internal Medicine, Medical School of Ribeirao Preto,

University of São Paulo, Ribeirao Preto, Brazil and 3Center for Cell Based Therapy, University of São Paulo, Ribeirao Preto, Brazil *

IW, DAP-M and JRH contributed equally as first authors.

Correspondence: J.J. SCHURINGA - j.j.schuringa@umcg.nl https://doi.org/10.3324/haematol.2023.283621 Received: May 26, 2023. Accepted: August 21, 2023. Early view: August 31, 2023. ©2024 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose. Contributions IW, DAP-M, JRH, ABV and JJS conceived and designed the study, performed experiments, analyzed, and interpreted data, performed the statistical analyses, and drafted the article. EMR and GH provided patient samples and clinical data and reviewed the paper. All authors gave final approval of the submitted manuscript. Acknowledgments The authors would like to thank Dr Emmanuel F. Griessinger for kindly providing the MS5, RS4;11, MOLM14 and RAJI cells used in the coculture assays, as well as some of the leukemic lines used in the study. Funding This investigation was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP, grant no. 2013/08135-2). DAP-M received a fellowship from FAPESP (grant no. 2017/23117-1). IW received a fellowship from FAPESP (grant no. 2015/09228-0). IW and DAP-M were sponsored by the Abel Tasman Talent Program (ATTP) of the Graduate School of Medical Sciences of the University of Groningen/ University Medical Center Groningen (UG/UMCG), the Netherlands. Data-sharing statement All data is presented in the manuscript and is also available upon request to the corresponding author.

Haematologica | 109 February 2024

644

LETTER TO THE EDITOR

References 1. Weinhäuser I, Pereira-Martins DA, Almeida LY, et al. M2 macrophages drive leukemic transformation by imposing resistance to phagocytosis and improving mitochondrial metabolism. Sci Adv. 2023;9(15):eadf8522. 2. Xu ZJ, Gu Y, Wang CZ, et al. The M2 macrophage marker CD206: a novel prognostic indicator for acute myeloid leukemia. Oncoimmunology. 2020;9(1):1683347. 3. Yang X, Feng W, Wang R, et al. Repolarizing heterogeneous leukemia-associated macrophages with more M1 characteristics eliminates their pro-leukemic effects. Oncoimmunology. 2018;7(4):e1412910. 4. Brauneck F, Fischer B, Witt M, et al. TIGIT blockade repolarizes AML-associated TIGIT(+) M2 macrophages to an M1 phenotype and increases CD47-mediated phagocytosis. J Immunother Cancer. 2022;10(12):e004794. 5. Al-Matary YS, Botezatu L, Opalka B, et al. Acute myeloid leukemia cells polarize macrophages towards a leukemia supporting state in a growth factor independence 1 dependent manner. Haematologica. 2016;101(10):1216-1227. 6. Klco JM, Spencer DH, Miller CA, et al. Functional heterogeneity of genetically defined subclones in acute myeloid leukemia. Cancer Cell. 2014;25(3):379-392. 7. Kim TH, Hanh BTB, Kim G, et al. Thiostrepton: a novel therapeutic drug candidate for mycobacterium abscessus infection. Molecules. 2019;24(24):4511. 8. Hu G, Su Y, Kang BH, et al. High-throughput phenotypic screen and transcriptional analysis identify new compounds and targets for macrophage reprogramming. Nat Commun. 2021;12(1):773. 9. Liu SX, Zhou Y, Zhao L, et al. Thiostrepton confers protection against reactive oxygen species-related apoptosis by restraining FOXM1-triggerred development of gastric cancer. Free Radic Biol Med. 2022;193(Pt 1):385-404.

10. Takeshita H, Yoshida R, Inoue J, et al. FOXM1-mediated regulation of reactive oxygen species and radioresistance in oral squamous cell carcinoma cells. Lab Invest. 2023;103(5):100060. 11. Trasanidis N, Katsarou A, Ponnusamy K, et al. Systems medicine dissection of chr1q-amp reveals a novel PBX1-FOXM1 axis for targeted therapy in multiple myeloma. Blood. 2022;139(13):1939-1953. 12. Takiguchi H, Yang CX, Yang CWT, et al. Macrophages with reduced expressions of classical M1 and M2 surface markers in human bronchoalveolar lavage fluid exhibit pro-inflammatory gene signatures. Sci Rep. 2021;11(1):8282. 13. Sica A, Mantovani A. Macrophage plasticity and polarization: in vivo veritas. J Clin Invest. 2012;122(3):787-795. 14. Noe JT, Rendon BE, Geller AE, et al. Lactate supports a metabolic-epigenetic link in macrophage polarization. Sci Adv. 2021;7(46):eabi8602. 15. Kuttikrishnan S, Prabhu KS, Khan AQ, Alali FQ, Ahmad A, Uddin S. Thiostrepton inhibits growth and induces apoptosis by targeting FoxM1/SKP2/MTH1 axis in B-precursor acute lymphoblastic leukemia cells. Leuk Lymphoma. 2021;62(13):3170-3180. 16. Erdem A, Marin S, Pereira-Martins DA, et al. Inhibition of the succinyl dehydrogenase complex in acute myeloid leukemia leads to a lactate-fuelled respiratory metabolic vulnerability. Nat Commun. 2022;13(1):2013. 17. Erdem A, Marin S, Pereira-Martins DA, et al. The glycolytic gatekeeper PDK1 defines different metabolic states between genetically distinct subtypes of human acute myeloid leukemia. Nat Commun. 2022;13(1):1105. 18. Stockwell BR, Friedmann Angeli JP, Bayir H, et al. Ferroptosis: a regulated cell death nexus linking metabolism, redox biology, and disease. Cell. 2017;171(2):273-285.

Haematologica | 109 February 2024

645

LETTER TO THE EDITOR

Systemic and mucosal adaptive immunity to SARS-CoV-2 during the Omicron wave in patients with chronic lymphocytic leukemia The COVID-19 pandemic has significantly impacted patients with chronic lymphocytic leukemia (CLL),1 with many failing to seroconvert2 or mediate variable T-cell immunity3 after mRNA vaccination. The emergence of the B1.1.529 (Omicron) variant of SARS-CoV-2 has altered the development of the COVID-19 pandemic due to its less severe clinical course and associated reduced risk of hospitalization.4 However, the impact of Omicron on immunosuppressed subgroups, such as patients who have received CD20 monoclonal antibodies (mAb)5 remains uncertain. Moreover, the observed decrease in severe disease cases within the general population may be influenced by the high number of infected individuals.6 In addition to the systemic immunoglobulin (Ig) G response, SARS-CoV-2 infection induces production of specific secretory IgA in mucosal secretions from local plasma cells, and serum IgA from plasma cells homing to the bone marrow.7 Whether this occurs after Omicron infection in patients with hematological or solid cancer remains elusive. We report here on serological, cellular, and mucosal immune response in a cohort of patients with CLL diagnosed with symptomatic SARS-CoV-2 infection during the Omicron BA.1 and BA.2 wave. Twenty-six patients with CLL who had symptoms of COVID-19 and tested positive for SARS-CoV-2 between January 9, 2022 and April 29, 2022, were included. Ninety-nine percent of all sequenced SARS-CoV-2 samples in Sweden taken on January 17, 2022 or later were Omicron variants.8 Patients diagnosed earlier than January 17, 2023 were only included if viral sequencing confirmed Omicron. The national ethics authority approved the study. Written informed consent was obtained from each patient before samples were obtained. The clinical characteristics are summarized in Table 1. Three patients had had a previous polymerase chain reaction (PCR)-verified SARS-CoV-2 infection at a median time of 20 months earlier (range, 13-21). Their immunological outcomes were similar to those who had Omicron as their first-time infection (data not shown). Patients had either early-stage, untreated CLL (n=11) or had ongoing CLL treatment (n=12), either Burton tyrosine kinase inhibitor (BTKi) therapy (n=11) or venetoclax + CD20 mAb (n=1). Four patients paused their BTKi treatment for a few days during the infection, and their immunological outcomes were similar to those who continued (data not shown). Five additional patients had completed various prior CLL therapies (including CD20 mAb) at a median time of 26 months before infection

(range, 8-74). Five patients had ongoing immunoglobulin supplemental treatment (IVIG). Total Ab levels against SARS-CoV-2 Spike receptor-binding domain (RBD) protein were analyzed in 14 of 26 patients at the time point when they had just been diagnosed with active, symptomatic COVID-19 infection, using Elecsys® anti-SARS-CoV-2 S immunoassay (Roche Diagnostics) (positive test was defined as >0.8 U/mL, patients with IVIG treatment were not included). Fifty percent (7/14) were seronegative, and of these, four had received a third vaccine dose 2-4 months before the infection and three had received 1-2 doses 9-11 months before. Two to three weeks after clinical recovery, a positive Elecsys® total anti-RBD test was noted in 81% of analyzed patients (13/16, 1 missing sample, 9 samples were excluded from analysis due to treatment with IVIG or the anti-SARS-CoV-2 mAb sotrovimab). We next used the V-PLEX Panel 25 assay (Meso Scale Discovery9) to differentiate IgG and IgA reactivities against ten different SARS-CoV-2 Spike variants in serum (n=24, 2 missing samples) respectively in saliva (n=25, 1 missing sample) from the convalescence follow-up. The serum was analyzed according to the manufacturer’s instructions, and saliva collection has been described elsewhere.10 Cutoff levels for positive saliva reactivity was defined for each antigen using pre-pandemic samples from healthy donors. Serum IgG was not analyzed in samples from patients who had received IVIG or sotrovimab treatment (n=9). Results against all SARS-CoV-2 variants are shown in the Online Supplementary Figure S1. Positive IgG levels against the Wuhan-Hu-1 (wild-type) SARS-CoV-2 variant (defined by the manufacturer as >1,960 AU/mL) were noted in all but one convalescent serum sample (Figure 1A). Generally, IgG reactivity against the three main variants (wild-type, Omicron BA.1 and Omicron BA.2 variants) varied substantially between individuals, and no significant differences were noted between the CLL treatment subgroups. Congruent with the serum findings, IgG reactivity against any SARS-CoV-2 Spike variant was observed in 88% of convalescent saliva samples (22/25; Online Supplementary Figure S1C), without difference in frequency or magnitude between the CLL treatment subgroups when comparing reactivity to the three main variants (Figure 1B). In contrast to the IgG reactivity, the serum IgA (i.e., mucosa-derived) responses to BA.2 Spike were significantly lower in BTKi/BCL-2i treated patients than in early-stage untreated patients (P=0.012) with a similar trend for responses against the wild-type variant (P=0.051) (Figure 1C).

Haematologica | 109 February 2024

646

LETTER TO THE EDITOR Table 1. Clinical characteristics of patients with chronic lymphocytic leukemia (N=26) at the time of SARS-CoV-2 Omicron infection. Early-stage untreated N=11

Previously treateda N=5

Ongoing BTKi/BCL2ib N=12

Entire cohort N=26

71 (53-82)

75 (63-87)

70.5 (42-82)

70.5 (42-87)

5/6

3/2

10/2

17/9

Time in years since CLL diagnosis, median (range)

3 (1-16)

8 (4-24)

9.5 (0-19)

7.5 (0-24)

CLL stage (Rai), N (%) 0 I-II III-IV

8 (73) 2 (18) 1 (9)

3 (60) 0 2 (40)

11 (92) 1 (8) 0

20 (77) 3 (12) 3 (12)

CLL remission status (iwCLL), 24 evaluated, N (%) PR/CR SD PD

7/9 (78) 2/9 (22)

3 (60) 2 (40)

12 (100) 0 0

12/24 (50) 9/24 (38) 3/24 (13)

1 (20)

4 (33))

5 (19)

7 (64) 2 (18)) 3 (27)

4 (80) 0) 1 (20)

5 (42) 2 (17) 2 (17)

14 (54) 4 (15) 5 (19)

2 (18) 8 (73) 1 (9) 0 0

1 (20) 2 (40) 1 (20) 0 1 (20)

1 (8) 7 (58) 3 (25) 1 (8) 0

3 (12) 16 (62) 5 (19) 1 (4) 1 (4)

2.1 (1-6.6)

2.6 (0.5-9)

3.4 (1-11)

2.7 (0.5-11)

Unknown

4/0

3/5

7/5

Median age in years (range) Male/Female, N

Ongoing Ig supplement, N (%) Comorbidities, N (%) Hypertension COPD/asthma Diabetes

Vaccination statusc, N (%) 4 doses 3 doses 2 doses 1 dose Unvaccinated Time in months since last vaccine dose, median (range)

Omicron variant, N=12 of sequenced samples, N BA.1/BA.2 Admitted to hospital, N (%)

Length in days of hospital stay, median (range) Omicron treatment, N (%) Supplementary oxygen Corticosteroids Sotrovimab Remdesivir

Secondary bacterial infection, N (%)

2 (18)

4 (80)

5 (42)

9 (35)

12 (8-16)

7.5 (1-18)

5 (3-9)

7 (1-18)

1 (9) 2 (18) 1 (9) 1 (9)

3 (60) 3 (60) 3 (60)) 2 (40)

2 (17) 1 (8) 5 (42) 4 (33)

5 (19) 4 (15) 8 (31) 6 (23)

2 (18)

4 (80)

1 (8)

7 (27)

All >6 months ago. Chemoimmunotherapy (N=3), ibrutinib (N=1), rituximab (N=2). bBruton´s tyrosine kinase inhibitor (BTKi) (N=14) or venetoclax (BCL2i) (N=2). cSix patients received Vaxzevria (AstraZeneca) at dose 1 and 2, all other were mRNA vaccinations (Comirnaty, Pfizer BioNTech or Spikevax, Moderna). CLL: chronic lymphocytic leukemia; iwCLL: International Workshop on Chronic Lymphocytic leukemia; PR: partial remission; CR: complete remission; SD: stable disease; PD: progressive disease; COPD: chronic obstructive pulmonary disease. Ig: immunoglobulin. a

Furthermore, salivary Spike-specific IgA against any variant was detected only in 40% (10/25; Online Supplementary Figure S1D) of patients. In line with the serum findings, IgA response was more rarely detected in saliva in patients with ongoing BTKi/BCL-2i therapy compared to early-stage untreated patients (2/12 vs. 6/9; P=0.032). The magnitude of the IgA salivary response was also significantly lower in BTKi/BCL-2i treated patients than the early-stage untreated patients when comparing the three main variants separately (Figure 1D; wild-type P=0.010; BA.1 P=0.016; BA.2 P=0.038). The ability of the convalescence sera to block Spike-protein binding to ACE2, a measure of viral neutralization capacity,11

was measured in 15 samples (9 samples were excluded due to sotrovimab or IVIG treatment) using the V-PLEX SARS-CoV-2 Panel 25. Fifty-three percent (8/15) were able to neutralize at least one Spike variant to 50% inhibition or higher (Online Supplementary Figure S2). Conversely, only 16% of saliva samples (4/25; 2 early-stage untreated, 1 previously treated, and 1 with ongoing BTKi/BCL-2i therapy) were able to neutralize at least one Spike variant (Online Supplementary Figure S2). The neutralization magnitude did not differ significantly between the patient subgroups (data not shown). The correlation between IgG and IgA levels and the cor-

Haematologica | 109 February 2024

647

LETTER TO THE EDITOR

Figure 1. Spike-specific antibodies in serum and saliva after clinical recovery from Omicron infection. Anti-Spike immunoglobuln (Ig) G in convalescent sera (A) and saliva (B) specific for SARS-CoV-2 wild-type, Omicron BA.1, and Omicron BA.2. Patients who had received IVIG or sotrovimab are excluded from the serum analyses and highlighted (red) in the saliva panel (B). The corresponding anti-Spike IgA levels are shown in (C) (serum) and (D) (saliva). Cutoff levels (dotted lines) for positive responses against wild-type in serum were determined by the manufacturer (1,960 AU/mL) and against all antigens in saliva using prepandemic saliva samples (defined as the mean plus 6x standard deviation of the intensity signals of 27 negative prepandemic saliva samples) and were as follows: anti-wild-type IgG: 4.01 AU/mL; anti-BA.1 IgG: 4.98 AU/mL; anti-BA.2 IgG: 7.33 AU/mL; anti-wild-type IgA: 226.72 AU/mL; anti-BA.1 IgA: 81.77 AU/mL; anti-BA.2 IgA: 203.18 AU/mL. Median and interquartile range are indicated in the panels. Statistics was assessed with non-parametric Kruskal-Wallis’ test with Dunn’s multiple comparison correction. *P<0.05, **P<0.01 and NS P>0.05: not statistically significant. Haematologica | 109 February 2024

648

LETTER TO THE EDITOR

responding neutralization capacity was stronger in serum than in saliva, and more pronounced for the wild-type variant compared to BA.1 (Online Supplementary Figure S3). The serum and salivary neutralization capacity of Omicron BA.2 was generally low, and correlation with corresponding Ab levels was hence not done. Next, we measured SARS-CoV-2-specific T-cell responses to wild-type and Omicron Spike-specific peptides using an AIM assay (Figure 2A), as previously described.12 PBMC were collected after clinical recovery from 22 patients (8 with untreated early-stage CLL, 4 previously treated, 9 with ongoing BTKi, and 1 with venetoclax + CD20 mAb treatment).

Eight otherwise healthy and previously vaccinated individuals who had recovered from Omicron infection served as controls. Specific CD4+ T cells against wild-type Spike were detected in 95% (21/22) of patients and against Omicron BA.1 Spike in 91% (20/22) (Figure 2B). Marginally lower frequencies of Spike-specific CD8+ T cells (Figure 2C) were observed, as 77% (17/22) and 73% (16/22) of patients had a response against wild-type and Omicron BA.1, respectively. No significant correlation was found between T-cell responses and serum or saliva reactivities, and also seronegative convalescents had measurable T-cell responses (data not shown). The magnitudes of the T-cell responses

Figure 2. SARS-CoV-2 reactive T cells in chronic lymphocytic leukemia patients and healthy controls after clinical recovery from Omicron infection. (A) Representative flow cytometry plot of antigen-specific CD4+ (CD69+CD154+) and CD8+ (CD69+CD137+) T cells after peptide stimulation. Frequencies of Spike-specific CD4+ (B) and CD8+ (C) T cells against SARS-CoV-2 wild-type and Omicron BA.1 peptides. A positive response was defined with a cutoff level of 0.05%. Median and interquartile range are indicated in the panels. Statistics was assessed with non-parametric Kruskal-Wallis’ test with Dunn’s multiple comparison correction. NS P>0.05: not statistically significant. DMSO: dimethyl sulfoxide. Haematologica | 109 February 2024

649

LETTER TO THE EDITOR

were similar in all CLL treatment subgroups and comparable to those of the healthy individuals (Figures 2B, C). Taken together, many patients mounted high post-infection IgG levels and T-cell responses. Notably, the T-cell responses were similar to those of healthy donors, also in patients with B-cell inhibiting therapy or low or absent convalescent Ab levels, which is most likely of clinical importance.12 However, we found an impaired IgA reactivity against all three virus variants in patients with ongoing BTKi/BCL-2i therapy in saliva, with a similar trend in serum, suggesting a previously not yet described negative effect of precision B-cell inhibiting treatment on mucosal immunity. Whether this is related to impaired mucosal memory B cells13 remains to be shown. Healthy individuals have significantly better protection against SARS-CoV-2 infection with higher mucosal IgA levels,14 and further studies are required on how the decreased IgA levels and generally low neutralization capacity of saliva Ab affect the risk of re-infection, particularly in BTKi-treated individuals. Notably, a significant reduction in the risk of grade 3-4 bacterial infections, mainly pneumonia, has been reported when the administration of BTKi is temporarily ceased in patients with CLL.15 This observation suggests a more widespread impairment of mucosal immunity post-BTKi, which also extends to other pathogens. The major limitations of our study are the small number of included patients and the heterogeneity of both previous CLL treatment, number of vaccine doses and antiviral treatment, including short-term use of corticosteroids, which might have impacted the immunological response. Also, the use of immunoglobulin treatment limited the number of IgG analyses. We provide a comprehensive analysis of both systemic and mucosal immunity to ten SARS-CoV-2 variants after Omicron infection in patients with CLL. Our data indicate that patients on BTKi/BCL-2i therapy exhibit compromised mucosal immunity, potentially increasing the susceptibility of this already vulnerable population to recurrent episodes of SARS-CoV-2 infection.

Huddinge, Karolinska Institutet, Stockholm, Sweden; 5Department of Dental Medicine, Karolinska Institutet, Huddinge, Sweden; 6

Department of Clinical Microbiology, Karolinska University

Hospital, Stockholm, Sweden; 7Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden; 8

Department of Medical Sciences, Clinical Chemistry and Science

for Life Laboratory, Uppsala University, Uppsala, Sweden; 9

Department of Internal Medicine, Capio St Göran Hospital,

Stockholm, Sweden; 10Department of Laboratory Medicine, Karolinska Institutet, Stockholm, Sweden; 11Center for Infectious Disease and Vaccine Research, La Jolla Institute for Immunology, La Jolla, CA, USA; 12Department of Medicine, Division of Infectious Diseases and Global Public Health, University of California, San Diego (UCSD), La Jolla, CA, USA; 13Department of Medicine Huddinge, Infectious Diseases, Karolinska Institutet, Stockholm, Sweden; 14Department of Infectious Diseases, Karolinska University Hospital, Stockholm, Sweden; 15Department of Infectious Diseases, Department of Transplantation, Karolinska University Hospital, Stockholm, Sweden and 16Department of Clinical Science, Intervention and Technology, Karolinska Institutet, Stockholm, Sweden +

LB, DW, JW and YD contributed equally.

Correspondence: H.M. INGELMAN-SUNDBERG - hanna.muren-ingelman-sundberg@ regionstockholm.se https://doi.org/10.3324/haematol.2023.282894 Received: February 13, 2023. Accepted: August 24, 2023. Early view: August 31, 2023. ©2024 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures MB is a consultant and has received honoraria from MSD, Pfizer, Mabtech, BMS, and Oxford Immunotec. AS is a consultant for Gritstone Bio, Flow Pharma, Moderna, AstraZeneca, Qiagen,

Authors

Fortress, Gilead, Sanofi, Merck, RiverVest, MedaCorp, Turnstone, NA Vaccine Institute, Emervax, Gerson Lehrman Group and

Hanna M. Ingelman-Sundberg,

1,2

Lisa Blixt,

1,3+

David Wullimann,

Jinghua Wu,

Yu Gao,

6,7

Bogdanovic,

Mikael Åberg, Christian Kjellander, Alba Grifoni,

Katie Healy, Sandra Muschiol, 5

6,7

Alessandro Sette,

11,12

Lotta Hansson,

1,3

Gordana

Soo Aleman,

13,14

protection for various aspects of T-cell epitope and vaccine design work. All other authors have no conflicts of interest to disclose.

Puran Chen, Ola Blennow, 4

15,16

Hans-Gustaf Ljunggren,4 Margaret Sällberg Chen,5

Marcus Buggert and Anders Österborg 4

Guggenheim. La Jolla Institute for Immunology has filed for patent

Contributions HMIS, LB, AÖ, GB, SA, MSC, HGL, and MB contributed to the

1,3

conceptualization, funding acquisition, and discussion of data. DW, Department of Oncology-Pathology, Karolinska Institutet,

JW, YG, KH, PC, MÅ, and SM performed experiments and analyzed

Stockholm, Sweden; Department of Oncology, Karolinska

data. LB, HMIS, CK, LH, and AÖ recruited study participants,

University Hospital Solna, Stockholm, Sweden; Department of

conducted the management of participants during the study, and

Haematology, Karolinska University Hospital Solna, Stockholm,

analyzed data. AG and AS provided peptide pools to measure the

Sweden; Center for Infectious Medicine, Department of Medicine

Spike-specific T-cell responses. OB provided information on

Haematologica | 109 February 2024

650

LETTER TO THE EDITOR

sequencing. HMIS, LB, DW, KH, MSC, AÖ, HGL, and MB wrote the

Stockholm, the Swedish Cancer Society, the Cancer Society in

original draft of the manuscript. All authors reviewed and edited

Stockholm, the Cancer and Allergy Foundation, the Swedish Blood

revisions of the manuscript and had final responsibility for the

Cancer Foundation, Center for Innovative Medicine (CIMED) and

decision to submit for publication.

Karolinska Institutet. This project has been funded in whole or in part with Federal funds from the National Institute of Allergy and

Acknowledgments

Infectious Diseases, National Institutes of Health, Department of

We thank all patients who donated blood for this study and A.

Health and Human Services, under contract no. 75N93021C00016 (to

Fahlén, K. Heimersson, and L. Relander for technical assistance.

AS).

Funding

Data-sharing statement

This study was supported by grants from the SciLifeLab National

The datasets analyzed during the current study are available from

COVID-19 Research Program, the Swedish Research Council, Region

the corresponding author on reasonable request.

References 1. Langerbeins P, Eichhorst B. Immune dysfunction in patients with chronic lymphocytic leukemia and challenges during COVID-19 pandemic. Acta Haematol. 2021;144(5):508-518. 2. Herishanu Y, Avivi I, Aharon A, Shefer G, Levi S, Bronstein Y, et al. Efficacy of the BNT162b2 mRNA COVID-19 vaccine in patients with chronic lymphocytic leukemia. Blood. 2021;137(23):3165-3173. 3. Blixt L, Wullimann D, Aleman S, et al. T-cell immune responses following vaccination with mRNA BNT162b2 against SARS-CoV-2 in patients with chronic lymphocytic leukemia: results from a prospective open-label clinical trial. Haematologica. 2022;107(4):1000-1003. 4. Davies MA, Kassanjee R, Rousseau P, et al. Outcomes of laboratory-confirmed SARS-CoV-2 infection in the Omicrondriven fourth wave compared with previous waves in the Western Cape Province, South Africa. Trop Med Int Health. 2022;27(6):564-573. 5. Cattaneo C, Masina L, Pagani C, et al. High mortality in fully vaccinated hematologic patients treated with anti-CD20 antibodies during the “Omicron wave” of COVID-19 pandemic. Hematol Oncol. 2023;41(1):205-207. 6. Bhattacharyya RP, Hanage WP. Challenges in inferring intrinsic severity of the SARS-CoV-2 omicron variant. N Engl J Med. 2022;386(7):e14. 7. Russell MW, Mestecky J. Mucosal immunity: the missing link in comprehending SARS-CoV-2 infection and transmission. Front Immunol. 2022;13:957107. 8. The Public Health Agency of Sweden. Covid-19 Veckorapporter [Internet]. Sweden; 2022. Updated: 2022.04.20; accessed:

2022.05.01. https://www.folkhalsomyndigheten.se/ folkhalsorapportering-statistik/statistik-a-o/sjukdomsstatistik/ covid-19-veckorapporter/. 9. Blom K, Marking U, Havervall S, et al. Immune responses after omicron infection in triple-vaccinated health-care workers with and without previous SARS-CoV-2 infection. Lancet Infect Dis. 2022;22(7):943-945. 10. Healy K, Pin E, Chen P, et al. Salivary IgG to SARS-CoV-2 indicates seroconversion and correlates to serum neutralization in mRNA-vaccinated immunocompromised individuals. Med. 2022;3(2):137-153.e3. 11. Narowski TM, Raphel K, Adams LE, et al. SARS-CoV-2 mRNA vaccine induces robust specific and cross-reactive IgG and unequal neutralizing antibodies in naive and previously infected people. Cell Rep. 2022;38(5):110336. 12. Gao Y, Cai C, Grifoni A, et al. Ancestral SARS-CoV-2-specific T cells cross-recognize the Omicron variant. Nat Med. 2022;28(3):472-476. 13. Mahanonda R, Champaiboon C, Subbalekha K, et al. Human memory B cells in healthy gingiva, gingivitis, and geriodontitis. J Immunol. 2016;197(3):715-725. 14. Havervall S, Marking U, Svensson J, et al. Anti-spike mucosal IgA protection against SARS-CoV-2 Omicron infection. N Engl J Med. 2022;387(14):1333-1336. 15. Lundin J, Mulder TA, Kattstrom M, et al. Temporary cessation of ibrutinib results in reduced grade 3-4 infections and durable remissions-Interim analysis of an on-off-repeat phase 1b/2 study in patients with chronic lymphocytic leukemia. EJHaem. 2021;2(3):525-529.

Haematologica | 109 February 2024

651

LETTER TO THE EDITOR

Prognostic value of minimal disseminated disease assessed using digital polymerase chain reaction for 3' ALK assays in pediatric anaplastic lymphoma kinasepositive anaplastic large cell lymphoma Approximately 90% of pediatric and 50% of adult anaplastic large cell lymphoma (ALCL) cases are anaplastic lymphoma kinase (ALK)-positive,1,2 with >80% of these cases carrying the NPM::ALK variant.3 In ALK-positive ALCL, minimal disseminated disease (MDD) has mainly been quantified using quantitative real-time polymerase chain reaction (qPCR) for NPM::ALK transcripts in the bone marrow or peripheral blood at diagnosis and is associated with a poor prognosis.4-6 MDD assessment is highly promising as a prognostic tool for ALK-positive ALCL; however, comparing the results of MDD assessment using qPCR between laboratories is difficult because qPCR requires standard curve calibration. Although MDD assessment using qualitative reverse-transcriptase PCR (RT-PCR) for NPM::ALK has been established and is used internationally for patient stratification, it is challenging to interpret the results of borderline cases.4,7,8 Recently, the efficacy of MDD assessment using digital PCR (dPCR) for NPM::ALK has been reported.7,9 dPCR does not require standard curve calibration and its reproducibility may be better than that of qPCR.7 dPCR can be used for MDD assessment; however, it requires the reference ABL gene assay to verify the quality of a given cDNA, as well as appropriate positive and negative controls to guarantee its performance for measurements of clinical samples. Although the efficacy of MDD has been established for the NPM:ALK variant, this is not currently the case for other ALK variants. A previous report10 revealed that a 3′ALK assay using a universal ALK probe works the same as an NPM::ALK assay; however, the correlation between MDD assessment using the 3′ALK assay and prognosis has not yet been evaluated. We, therefore, examined the prognostic value of MDD assessed using dPCR to establish a 3′ALK MDD assay as a prognostic tool for future clinical trials. Patients aged <20 years were enrolled in the ALCL99 trial in Japan between January 2000 and April 2012. Bone marrow and/or peripheral blood samples collected from these patients at diagnosis were used. ALK-positive ALCL was confirmed using the pathological review system of the Japanese Pediatric Leukemia/Lymphoma Study Group (JPLSG) for all patients. This study was approved by the institutional ethics committee of Yamaguchi University Hospital and the National Hospital Organization Nagoya Medical Center. RNA was isolated from mononuclear cells using Trizol Reagent (Sigma, St. Louis, MO, USA). The protocols applied for cDNA synthesis and quantification of NPM::ALK were

as previously reported.4,6 cDNA synthesis was performed using SuperScript II (Invitrogen Corporation, Carlsbad, CA, USA) with 1 μg of RNA per 20 μL of reaction volume with random primers (Invitrogen). dPCR assays for 3′ALK were performed according to a previous report.10 The primers and probe were designed to amplify the 3′ part of ALK and included the exon 20/21 boundary to avoid genomic amplification. Furthermore, we performed dPCR assays for NPM::ALK according to the previous report for patients who were confirmed as being positive for NPM::ALK using genetic or immunohistological findings.9 The copy numbers of 3′ALK and NPM::ALK were calculated per 10,000 copies of ABL and are reported as normalized copy number (NCN). The dPCR details for 3′ALK and NPM::ALK, including primer and probe sequences, are shown in Online Supplementary Table S1. Differences in the distribution of variables were assessed using the χ2 test. Survival rates were estimated with the Kaplan-Meier method, and differences were compared using the log-rank test. Progression-free survival was estimated from the date of diagnosis until the last follow-up visit or the date of documented failure (progression, relapse, or death). The quantification of 3′ALK dPCR and NPM::ALK dPCR was compared using Spearman correlation. The prognostic effects of variables were compared by Cox regression analysis. All analyses were performed using SPSS Statistics 27 software (IBM, NY, USA). We collected 50 bone marrow and 52 peripheral blood samples from 61 patients. We have previously reported the NPM::ALK MDD qPCR assay results of 34 patients.6 The ALK-staining patterns were “cytoplasmic and nuclear”, and confirmed as NPM::ALK variants, in 50 patients, while other patterns were observed in four patients. As with the other ALK variants, the MYH9::ALK fusion variant was confirmed using karyotyping in one patient. For six of the 61 patients, ALK-staining patterns or karyotyping data were not available. When both bone marrow and peripheral blood samples were available for a patient, we used the MDD results for the sample with a higher NCN in the analysis of prognosis. The 3-year progression-free and overall survival rates of the 61 patients analyzed for the 3′ALK MDD assay were 83.5% (95% confidence interval [95% CI]: 75.528-91.821) and 96.6% (95% CI: 94.368-99.149), respectively. The clinical and biological characteristics of the patients who underwent the 3′ALK MDD assay are shown in Table 1.

Haematologica | 109 February 2024

652

LETTER TO THE EDITOR

For 3′ALK analysis, we initially determined whether the 3′ALK universal probe and primers could work using NPM::ALK samples and MYH9::ALK variants. For NPM::ALK variants, copy number amplifications were observed in both NPM::ALK and 3′ALK assays. However, only the 3′ALK assay could detect copy number amplification for MYH9::ALK variants (Figure 1A, B). Low-amplitude droplets, differentiated from high-amplitude positive droplets, were observed around an amplitude of 2,000 in the 3′ALK assays. These low-amplitude droplets were also observed in negative control samples from HL-60 cells and in peripheral blood from healthy controls (6 samples), and did not exceed an amplitude of 2,500. We considered these low-amplitude droplets to be non-specific amplifications caused by the 3′ALK probe and set the threshold at an amplitude of 2,500. Subsequently, we assessed the correlation between the NCN of the NPM::ALK dPCR and 3′ALK dPCR assays for 87 samples from 50 patients with an NPM::ALK fusion variant confirmed by their ALK-staining pattern. Forty-four samples were negative for NPM::ALK. Of these 44 samples, 15 were negative for NPM::ALK and 3′ALK assays whereas 29 were negative for the NPM::ALK assay and showed <10 NCN for 3′ALK. Of all 87 samples, only one showed >30 and ≤30 NCN for the NPM::ALK and 3′ALK assays, respectively. The NCN of the NPM::ALK and 3′ALK assays were highly concordant (r=0.855) (Figure 1C). Next, we analyzed the correlation between MDD and patients’ survival. As for the NPM::ALK MDD dPCR assay, we defined a 30 NCN cutoff for the 3′ALK MDD dPCR assay.9 The 3-year progression-free survival rates of the 21 patients with a NCN >30 and 40 patients with a NCN ≤30 were 56.7% (95% CI: 38.812-72.117) and 97.5% (95% CI: 91.322-100.128), respectively (P<0.001) (Figure 1D). The 3-year overall survival rate of patients with a NCN >30 was not significantly different from that of patients with a NCN ≤30 (100% vs. 100%, P=0.168) (Figure 1E). MDD with NCN >30 alone exhibited a significant prognostic value in both univariate (P=0.004) and multivariate analyses (P=0.024) (Table 2). The stage at diagnosis and histological subtype of an “uncommon” component tended to have prognostic value in the univariate analysis; however, these two features showed no significant influence in the multivariate analysis. In our study, we confirmed that the 3′ALK MDD assay using dPCR can be a prognostic tool for newly diagnosed ALK-positive ALCL. We determined a NCN cutoff of 30 based on previous reports on the NPM::ALK dPCR assay.7,9 Non-specific amplification was a concern in the 3′ALK MDD assay as the probe is not designed for the fusion gene part; however, we confirmed that the NCN cutoff of 30 is appropriate for this assay. On the 3′ALK MDD assay using dPCR, we found that 3-year progression-free survival rates were significantly lower in patients with MDD showing a NCN >30 than in those with a NCN ≤30, consistent with the previous report on the NPM::ALK MDD assay using qPCR.4-6 The 3-year progression-free survival rates for pa-

tients with NCN >30 and ≤30 using the 3′ALK dPCR assay in our study were higher than those reported in a previous study on MDD using the NPM::ALK dPCR assay (NCN >30 and ≤30: 57% and 98% vs. 33% and 79%, respectively).9 This difference could be attributed to the differences in 3-year progression-free survival rates for the entire cohort between our study (84%) and the previous NPM::ALK dPCR report (74%).9 It has been reported that there is complete concordance between the ALK staining pattern and ALK fusion partner variant.11,12 ALK is expressed in the nucleus in NPM::ALK-positive ALCL, and is absent in ALCL with other ALK fusion variants. The presence of an NPM::ALK fusion transcript Table 1. Patients’ characteristics according to minimal disseminated disease in the 3′ALK droplet polymerase chain reaction assay. Minimal disseminated disease All patients NCN ≤30 NCN >30 N of patients

Gender, N (%) Male Female

42 19

27 (68) 13 (32)

15 (71) 6 (29)

1.000

Age, N (%) <10 years ≥10 years

28 33

17 (43) 23 (57)

11 (52) 10 (48)

0.590

7 18 31 5

7 (18) 16 (40) 15 (37) 2 (5)

0 (0) 2 (10) 16 (76) 3 (14)

0.004

33 28

27 (68) 13 (32)

6 (29) 15 (71)

0.006

8 (20) 32 (80)

6 (29) 15 (71)

0.527

Mediastinum, N (%) No Yes

14 47 41 20

32 (80) 8 (20)

9 (43) 12 (57)

0.005

Skin, N (%) No Yes

52 9

33 (83) 7 (17)

19 (90) 2 (10)

0.479

Stage, N (%) I II III IV B symptom, N (%) No Yes

Peripheral LN, N (%) No Yes

Bone, N (%) No Yes Bone marrow, N (%) No Yes

45 16

31 (78) 9 (22)

14 (67) 7 (33)

0.376

56 5

38 (95) 2 (5)

18 (86) 3 (14)

0.329

Visceral organs, N (%) No Yes

53 8

36 (90) 4 (10)

17 (81) 4 (19)

0.429

Histology, N (%) Common Non-common Not available

40 15 6

29 (73) 7 (17) 4 (10)

11 (52) 8 (38) 2 (10)

0.199

NCN: normalized copy number, LN: lymph nodes.

Haematologica | 109 February 2024

653

LETTER TO THE EDITOR

can also be confirmed using two-color fluorescence in situ hybridization (FISH) or RT-PCR; however, other ALK fusion variants cannot be evaluated for MDD using the NPM::ALK method. The 3′ALK MDD method can be used for all cases

of ALK-positive ALCL after confirming ALK positivity by immunostaining; thus, this method is promising and suitable for MDD assessment at diagnosis. In the current study, only one of the fusion partners in ALCL with cytoplasmic ALK

Figure 1. 3′ALK digital polymerase chain reaction assay for ALK-positive anaplastic large cell lymphoma. (A) In NPM::ALK-positive cases of anaplastic large cell lymphoma (ALCL), copy number amplifications were observed in both the NPM::ALK and 3′ALK assays. (B) In MYH9::ALK-positive ALCL cases, only the 3′ALK assay could detect copy number amplification. (C) Correlation between the normalized copy number (NCN) analyzed using the NPM::ALK digital polymerase chain reaction (dPCR) and 3′ALK dPCR assays. NCN detected using NPM::ALK dPCR and 3′ALK dPCR were highly concordant in 87 samples. (D) Progression-free survival according to minimal disseminated disease (MDD) analyzed by dPCR in the 3′ALK assay with a cutoff NCN of 30. (E) Overall survival according to MDD analyzed by dPCR in the 3′ALK assay with a cutoff NCN of 30. PFS: progression-free survival; OS: overall survival.

Haematologica | 109 February 2024

654

LETTER TO THE EDITOR

Table 2. Cox proportional hazard model of progression-free survival rates.

Oncology and Laboratory Medicine, Yamaguchi University Graduate School of Medicine, Yamaguchi and 8Department of Pediatrics, St.

Hazard ratio

95% CI

Stage III/IV

6.883

0.872-54.346

0.067

B symptom

2.963

0.766-12.461

0.116

Mediastinum

2.139

0.619-7.393

0.229

Skin

1.484

0.315-6.995

0.618

Visceral organs

0.707

0.090-5.578

0.742

Histology: non-common type

3.612

0.969-13.464

0.056

3′ALK MDD NCN >30

21.160

2.675-167.411 0.004

Univariate analysis

of Pediatrics, NHO Nagoya Medical Center, Aichi; 7Department of

Marianna University School of Medicine, Kanagawa, Japan Correspondence: R. FUKANO - fukano.r@yamaguchi-u.ac.jp https://doi.org/10.3324/haematol.2023.282812 Received: February 1, 2023. Accepted: August 28, 2023. Early view: September 7, 2023.

Multivariate analysis Stage III/IV

1.596

0.158-16.110

0.692

Histology: non-common type

2.159

0.571-8.168

0.257

3′ALK MDD NCN >30

14.823

1.436-153.044 0.024

95% CI: 95% confidence interval; ALK: anaplastic lymphoma kinase; MDD: minimal disseminated disease; NCN: normalized copy number.

Disclosures RF has received research funding from Pfizer and received speakers’ honoraria from Chugai and Takeda. TM has received research funding from Pfizer and received speakers’ honoraria from Chugai

expression could be confirmed. To establish the efficacy of the 3′ALK universal probe, further evaluation of the 3′ALK assay for other ALK fusion variants is required. Our findings suggest that 3′ALK MDD assessments are useful for patients with ALK-positive ALCL as 3′ALK MDD can be assessed without the need to confirm the ALK fusion partners by karyotyping, FISH, RT-PCR, or sequencing. In this era of targeted therapy, stratification of patients with ALK-positive ALCL based on MDD assessment using dPCR can help establish new standard treatments.

and Takeda. KH is a consultant for Pfizer and Kyowa Kirin and has received honoraria for speakers’ bureaus from Amgen, Chugai, and Novartis. All other authors declare that they have no competing interests to disclose. Contributions RF and YIY designed the study. RF, YIY and YS performed the experiments. RF, YIY, YS, TY, SH, and TM analyzed and interpreted the data. RF and YIY wrote the manuscript. RF, YIY, AMS, TT, MS, TM, and KH collected the clinical data and samples. HI and AN performed the pathological review. All authors read and approved the final manuscript.

Authors

Acknowledgments We thank the patients, their families, and clinicians for their

Reiji Fukano, Yuka Iijima-Yamashita, Hideto Iwafuchi, Atsuko 1

participation in the ALCL99 study in Japan. This study was partly

Nakazawa, Akiko M. Saito, Tetsuya Takimoto, Masahiro Sekimizu, 4

Yutaka Suehiro, Takahiro Yamasaki, Shunji Hasegawa, Tetsuya Mori

presented at the 64th American Society of Hematology Annual

and Keizo Horibe2,6

Meeting, New Orleans, LA, USA on December 10, 2022.

Department of Pediatrics, Yamaguchi University Graduate School of

Funding

Medicine, Yamaguchi; Clinical Research Center, NHO Nagoya Medical

This work was supported by Grants-in-Aid for Translational Research

Center, Aichi; Department of Pathology, Shizuoka Children’s Hospital,

of Yamaguchi University Hospital 2021.

Shizuoka; Department of Clinical Research, Saitama Children’s Medical 4

Center, Saitama; 5Department of Childhood Cancer Data Management,

Data-sharing statement

National Center for Child Health and Development, Tokyo; Department

For original data, please contact fukano.r@yamaguchi-u.ac.jp.

References 1. Turner SD, Lamant L, Kenner L, Brugieres L. Anaplastic large cell lymphoma in paediatric and young adult patients. Br J Haematol. 2016;173(4):560-572. 2. Stein H, Foss H-D, Dürkop H, et al. CD30(+) anaplastic large cell lymphoma: a review of its histopathologic, genetic, and clinical

features. Blood. 2000;96(12):3681-3695. 3. Morris SW, Xue L, Ma Z, Kinney MC. ALK+ CD30+ lymphomas: a distinct molecular genetic subtype of non-Hodgkin’s lymphoma. Br J Haematol. 2001;113(2):275-295. 4. Damm-Welk C, Busch K, Burkhardt B, et al. Prognostic

Haematologica | 109 February 2024

655

LETTER TO THE EDITOR significance of circulating tumor cells in bone marrow or peripheral blood as detected by qualitative and quantitative PCR in pediatric NPM-ALK-positive anaplastic large-cell lymphoma. Blood. 2007;110(2):670-677. 5. Mussolin L, Damm-Welk C, Pillon M, et al. Use of minimal disseminated disease and immunity to NPM-ALK antigen to stratify ALK-positive ALCL patients with different prognosis. Leukemia. 2013;27(2):416-422. 6. Iijima-Yamashita Y, Mori T, Nakazawa A, et al. Prognostic impact of minimal disseminated disease and immune response to NPM-ALK in Japanese children with ALK-positive anaplastic large cell lymphoma. Int J Hematol. 2018;107(2):244-250. 7. Damm-Welk C, Lovisa F, Contarini G, et al. Quantification of minimal disease by digital PCR in ALK-positive anaplastic large cell lymphoma: a step towards risk stratification in international trials? Cancers (Basel). 2022;14(7):1703. 8. Mussolin L, Pillon M, d’Amore ES, et al. Prevalence and clinical implications of bone marrow involvement in pediatric anaplastic large cell lymphoma. Leukemia. 2005;19(9):1643-1647. 9. Damm-Welk C, Kutscher N, Zimmermann M, et al.

Quantification of minimal disseminated disease by quantitative polymerase chain reaction and digital polymerase chain reaction for NPM-ALK as a prognostic factor in children with anaplastic large cell lymphoma. Haematologica. 2020;105(8):2141-2149. 10. Quelen C, Grand D, Sarot E, et al. Minimal residual disease monitoring using a 3′ALK universal probe assay in ALK-positive anaplastic large-cell lymphoma: ddPCR, an attractive alternative method to real-time quantitative PCR. J Mol Diagn. 2021;23(2):131-139. 11. Perkins SL, Pickering D, Lowe EJ, et al. Childhood anaplastic large cell lymphoma has a high incidence of ALK gene rearrangement as determined by immunohistochemical staining and fluorescent in situ hybridisation: a genetic and pathological correlation. Br J Haematol. 2005;131(5):624-627. 12. Damm-Welk C, Klapper W, Oschlies I, et al. Distribution of NPM1-ALK and X-ALK fusion transcripts in paediatric anaplastic large cell lymphoma: a molecular-histological correlation. Br J Haematol. 2009;146(3):306-309.

Haematologica | 109 February 2024

656

LETTER TO THE EDITOR

Uniform conditioning regardless of donor in bone marrow transplantation for severe aplastic anemia For nearly 40 years, the treatment of choice for young patients with severe aplastic anemia (SAA) has been matched sibling donor (MSD) bone marrow transplantation (BMT). This medical preference was related to rapid recovery of hematopoiesis, minimal complications, mitigation of clonal evolution rates, and impressive rates of overall survival.1 Cyclophosphamide (50 mg/kg/day × 4 days) with or without ATG, has traditionally been used as conditioning before MSD BMT. Although this regimen is non-myeloablative, the immunosuppression is sufficient to allow engraftment in most cases. Avoidance of total body irradiation and busulfan has been continued for MSD BMT to reduce transplantrelated complications such as mucositis, graft-versus-host disease (GVHD), second malignancies, and infertility. However, several conditioning regimens as well as varied GVHD approaches can be used for BMT in AA. The increasingly chosen donor is often a haploidentical, given they are more common, but a common question often arises as to how to condition and prevent GVHD in the patient with a fully matched sibling or fully matched unrelated donor. Additionally, mixed chimerism or late secondary graft failure remain obstacles to longer-term successful outcomes in BMT for SAA.3 Survival rates following MSD allogeneic BMT have steadily improved since the 1970s largely because of improved supportive care, refined HLA-typing, and better GVHD prophylaxis.2 However, late BMT-related complications such as chronic GVHD occur in up to one-third of patients, with many of these patients requiring long-term therapy for their GVHD. In patients under 30 years of age, the event-free survival after HLA-matched sibling BMT ranges from 70% to 90%. These reduced survival rates are predominantly due to GVHD (which steadily increases with age) and late graft failures.3 These outcomes remain ripe for improvement. More recent research efforts in SAA BMT have focused on transplant feasibility using mismatched donors to expand the donor pool. The most promising approach to facilitate engraftment and mitigate the risk of GVHD is the use of post-transplant cyclophosphamide (PTCy) for GVHD prophylaxis. These regimens have also augmented the total body irradiation dosing from 200 cGy to 400 cGy in 2019 to enhance engraftment as learned from sickle disease and myelodysplastic syndrome regimens. These studies demonstrated success with standardized conditioning and intensive PTCy-based GVHD prophylaxis for haploidentical donors with >90% full donor chimerism and <10% GVHD.4,5 Given the disease rarity, there have been no comparative studies of conditioning or GVHD prophylaxis, but

an important issue remains in 2023 with matched donors, especially in adult patients, where we still need to minimize any complications with BMT, namely cGVHD, mixed chimerism, relapse or clonal evolution, and other transplant-related complications. Here, we report the outcomes using the established haploidentical approach to BMT in SAA,4,6 for patients with related or unrelated fully matched donors. The numbers are few, given haploidentical donors are more often selected, but we show both feasibility and efficacy. We treated 11 adult patients with treatment-naïve or refractory SAA patients using matched donor marrow grafts (related and unrelated) with the Baltimore regimen as previously published7 (Table 1). Rabbit anti-thymocyte globulin (rATG, Thymoglobulin©) dosed at 0.5 mg/kg on day -9 and 2 mg/ kg on days -8 and -7 intravenously (IV). Fludarabine was administered at 30 mg/m2 IV daily for 5 days, from day -6 to day -2 (total dose received 150 mg/m2). Cyclophosphamide was given at 14.5 mg/kg IV daily for 2 days from day -6 to day -5 and administered as a 1-2-hour infusion (total dose received 29 mg/kg) and total body irradiation was delivered in a single fraction of 200 cGy on day -1 until augmented this to a single fraction of 400 cGy on day -1 institutionally after first eight patients. This was an institutional change based on noted graft failures in an upfront study of haploidentical donors using the same conditioning regimen.5 The marrow graft was infused on day 0. Granulocyte colony stimulating factor was given on day +5 at 5 mg/kg/day and continued until absolute neutrophil count was greater than 1.5x109/L for 3 days. Posttransplant GVHD prophylaxis included PTCy administered at 50 mg/kg/day IV on days +3 and +4, mycophenolate mofetil orally given at a dose of 15 mg/kg three times a day up to 1 gm three times a day starting on day -3 (max dose 3,000 mg/day) from day 5 through 35 and tacrolimus orally or IV was given starting day 5 to maintain a level of 10-15 ng/mL. Tacrolimus was discontinued in patients without GVHD initially on day 365 in the first ten patients and then stopped day 180 in the last patient as is now done for all haploidentical patients. There was not standardization of Cytomegalovirus prophylaxis, rather initiation of therapy when viremia present. The median follow-up of this cohort is 53.8 (range, 6-99) months. The overall survival for 11 patients is 100% at 1, and 2 years, and 90% (95% confidence interval [CI]: 73100) at 3 years. There was no primary graft failure. One patient died from complications of gastric adenocarcinoma diagnosed 2 years post BMT. At the time of death,

Haematologica | 109 February 2024

657

LETTER TO THE EDITOR Table 1. Demographics of 11 patients and donors. Patient characteristics Sex, N (%) Female Male Self-identified as minority, N (%) Age in years Mean (SD) Median (range) Very severe aplastic anemia diagnosis (absolute neutrophil <0.2x109/L), N (%) Severe aplastic anemia diagnosis (absolute neutrophil <0.5x109/L), N (%) Treatment-naïve, N (%) Refractory, N (%) Clonality at baseline (PNH clone or molecular data including karyotype), N (%) Total body irradiation dose, N (%) 200 cGY 400 cGy Donor characteristics Age in years Mean (SD) Median (range) Relationship, N (%) Sibling Unrelated 10/10 Unrelated 9/10 Sex, N (%) Female Male Reason for use of non-haploidentical related donor, N (%) Donor specific antibodies to related donors No available related donors (i.e., adopted, parent with illness, childlessness) Patient above age 25 years with available matched sibling

Total N=11 5 (45) 6 (55) 3 (27) 32 (12) 25 (24-38) 6 (55) 5 (45) 5 (45) 6 (55) 11 (100) 4 (36) 7 (64) Total N=11 29 (6) 29 (28-32) 4 (36) 4 (36) 3 (27) 2 (18) 9 (82) 0 7 (64) 4 (36)

SD: standard deviation; PNH: paroxysmal nocturnal hemoglobinuria.

he was transfusion independent without GVHD and had >80% donor chimerism in the whole blood. This mixed chimerism was considered secondary graft loss after adenovirus infection but did not result in relapse of his SAA. The median CD34+ cell count of all grafts was 3.86x106/kg recipient ideal body weight (range, 1.908.30x106). The median time to neutrophil recovery was 17 (range, 14-41) days. The day 28 cumulative incidence of neutrophil recovery was 91% (95% CI: 69-100). The median time to platelet recovery was 26 days with 91% transfusion independence by day 100. The median time to red cell recovery was 31 days with 91% transfusion independence by day 100. This is quite consistent with published timelines from other donor sources.7 Ten of the 11 patients had sustained >95% donor chimerism in both whole blood and CD3 compartments through 1 year. Nine (81%) patients experienced infections post-transplant. Of a total of 14 infection events, 11 were grade 2 and three were grade 3. Two of the documented grade 3 infections occurred in the patient with secondary graft failure from adenovirus. Two patients experienced Cytomegalovirus reactivation evidenced by viremia treated to undetectable by day 100, and no patients had Epstein-Barr virus. The

cumulative incidence of grade 2-4 acute GVHD at day 100 was 9% (95% CI: not applicable [NA]-27) while the cumulative incidence of chronic GVHD at 2 years is 19% (95% CI: NA-45) and none was beyond score 2 skin and mouth and lacked any other organ scoring. The two patients with chronic GVHD were off all treatment by months 13 and 17 post BMT, respectively. As all other patients are beyond 6 months follow-up, they are all off immunosuppression. We followed patients clinically for early events so lack longer-term data on fertility, other secondary malignancies, and extended chimerism data. The role of the BMT for the patient that died of a solid malignancy was less clear but could have contributed. This patient had an unrevealing work-up for inherited causes of disease as there was the presence of a paroxysmal nocturnal hemoglobinuria clone, his telomere length in the lymphocytes was at the tenth percentile when white blood cell count was very low, and his next-generation sequencing did not identify targeted mutations. He had a history of significant gastroesophageal reflux since infancy and esophageal evaluations akin to Barrett’s disease so it is possible this contributed to his malignancy risk. The choice to use a matched donor in each of these pa-

Haematologica | 109 February 2024

658

LETTER TO THE EDITOR tients was related to a lack of suitable haploidentical donor (Table 1). All patients were believed to have acquired disease. Anecdotally, another refractory acquired SAA patient was treated at our institution with this identical platform using an 8/10 unrelated donor peripheral blood graft, given very high levels of donor-specific antibodies to all available related donors, and this 55 years SAA patient also had full engraftment and no GVHD and is doing well at 3 years post transplant. Thus, this platform may even be further extended to mismatched unrelated donors and a peripheral blood stem cell source. However, augmentation to 400 cGY was first used in SAA in the treatment-naïve haploidentical setting5 to ensure engraftment and 200 cGy in the matched donor setting may be sufficient to allow engraftment and preserve fertility, as has been seen in the refractory setting.4 Given that more than 40% of patients initially treated with immunosuppression will ultimately need a transplant8 due to relapse or clonal evolution, the use of BMT as the therapeutic approach should be consistent. There is merit in a uniform condition approach in all patients, regardless of the marrow donor’s relationship or matched status, to maintain clinical expertise and consistent outcomes. Experience9 and consistency of a regimen as described here will allow uniform management and follow-up for relevant clinical outcomes. Upfront use of this regimen was suggested by the BMT CTN State of the Science Symposium 2021 as the next needed innovation in the non-malignant field,10 and a trial using this exact approach for both haploidentical and unrelated donors is anticipated later in 2024. In summary, the Hopkins group uses ATG, fludarabine, cyclophosphamide, and PTCy always with 400 cGy in the upfront and refractory settings as well as with mismatched and matched donors.7 Rarely do we use immunosuppressive therapy for SAA patients anymore, given this platform has durability, no early transplant-related mortality, faster hematopoietic recovery, low rates of acute and chronic GvHD, and higher treatment-free remissions. Additional investigations will provide valuable data to determine the effects on fertility and secondary malignancies as well as relevance of noted

differences in financial cost in comparison to IST.11 The forthcoming multicenter BMT CTN 2207 study may establish this intensive GVHD prophylaxis and augmented conditioning as the standard in all SAA bone marrow transplantation, regardless of donor.

Authors Amy E. DeZern,1,2 Marianna Zahurak,1,3 Richard J. Jones1,2 and Robert A. Brodsky1,2 Department of Oncology, Sidney Kimmel Cancer Center;

Department of Medicine, Johns Hopkins University, Division of

Hematology and 3Department of Oncology Biostatistics, Sidney Kimmel Cancer Center, Baltimore, MD, USA Correspondence: A.E. DEZERN - adezern1@jhmi.edu https://doi.org/10.3324/haematol.2023.284022 Received: August 3, 2023. Accepted: August 31, 2023. Early view: September 7, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose. Contributions AED, RJJ and RAB performed research. MZ performed data analysis. AED wrote the manuscript and supervised the study. All authors reviewed, edited, and approved the manuscript. Data-sharing statement The data that support the findings of this study are available from the corresponding author upon reasonable request.

References 1. Bacigalupo A. How I treat acquired aplastic anemia. Blood. 2017;129(11):1428-1436. 2. Bacigalupo A, Socie G, Schrezenmeier H, et al. Bone marrow versus peripheral blood as the stem cell source for sibling transplants in acquired aplastic anemia: survival advantage for bone marrow in all age groups. Haematologica. 2012;97(8):1142-1148. 3. Xu ZL, Xu LP, Zhang YY, et al. Mixed chimaerism is associated with poorer long-term failure-free survival among aplastic anaemia patients receiving HLA-matched donor transplantation. Bone Marrow Transplant. 2023;58(7):832-834. 4. DeZern AE, Eapen M, Wu J, et al. Haploidentical bone marrow

transplantation in patients with relapsed or refractory severe aplastic anaemia in the USA (BMT CTN 1502): a multicentre, single-arm, phase 2 trial. Lancet Haematol. 2022;9(9):e660-e669. 5. DeZern AE, Zahurak ML, Symons HJ, et al. Alternative donor BMT with post-transplant cyclophosphamide as initial therapy for acquired severe aplastic anemia. Blood. 2023;141(25):3031-3038. 6. DeZern AE, Brodsky RA. Combining PTCy and ATG for GvHD prophylaxis in non-malignant diseases. Blood Rev. 2022;101016.

Haematologica | 109 February 2024

659

LETTER TO THE EDITOR 7. DeZern AE, Zahurak ML, Symons HJ, et al. Haploidentical BMT for severe aplastic anemia with intensive GVHD prophylaxis including posttransplant cyclophosphamide. Blood Adv. 2020;4(8):1770-1779. 8. Tichelli A, Schrezenmeier H, Socie G, et al. A randomized controlled study in patients with newly diagnosed severe aplastic anemia receiving antithymocyte globulin (ATG), cyclosporine, with or without G-CSF: a study of the SAA Working Party of the European Group for Blood and Marrow Transplantation. Blood. 2011;117(17):4434-4441. 9. Arcuri LJ, Nabhan SK, Loth G, et al. A case series of post-

yransplantation cyclophosphamide in unrelated donor hematopoietic cell transplantation for aplastic anemia. Biol Blood Marrow Transplant. 2020;26(9):e222-e226. 10. Heslop HE, Stadtmauer EA, Levine JE, et al. Blood and Marrow Transplant Clinical Trials Network State of the Science Symposium 2021: looking forward as the network celebrates its 20th year. Transplant Cell Ther. 2021;27(11):885-907. 11. Zhang MX, Wang Q, Wang XQ. Hematopoietic stem-cell transplantation versus immunosuppressive therapy in patients with adult acquired severe aplastic anemia: a costeffectiveness analysis. Int J Gen Med. 2021;14:3529-3537.

Haematologica | 109 February 2024

660

LETTER TO THE EDITOR

Frequent, high density expression of surface CD38 as a potential therapeutic target in adult T-lineage acute lymphoblastic leukemia Polychemotherapy is the fundamental backbone in the treatment of acute lymphoblastic leukemia (ALL). However, the spectrum of treatment modalities has been expanded in B-cell lineage (i.e., monoclonal antibodies, chimeric antibody receptor T cells), but not in T-cell lineage ALL (T-ALL). This discrepancy is likely due to a limited number of identified potential target antigens in T-ALL. In addition, T-ALL is a less frequent subtype, which further limits the evaluation of novel treatment strategies.1 Patients with an immature T-ALL subtype exhibit a particular poor prognosis across various studies and age groups.2-4 Thus, the integration of innovative immunotherapeutic treatment modalities into the treatment strategy of poor-risk T-ALL appears sensible. The CD38 molecule is predominantly expressed on T cells and is associated with activation.5 Expression of CD38 is not T-cell lineage-specific, but has also been found on a variety of other hematopoietic and some non-hematopoietic cells. Therapeutic antibodies, which target CD38, have successfully entered clinical routine in the treatment of multiple myeloma. Although CD38 expression is not restricted to multiple myeloma cells, the use of these antibodies is not limited by off-target effects. It is unknown whether this is in part due to differences in cell surface expression on target and off-target cells. However, published data indicate towards improved responses to daratumumab in patients with multiple myeloma and higher expression levels of CD38 on their myeloma cells.6 Expression of CD38 has been demonstrated in patients with T-ALL,7 and clinical efficacy of daratumumab has been reported in a limited number of ALL patients,8 suggesting that CD38 could serve as a useful therapeutic target. In order to broaden the available knowledge about CD38 in ALL, we evaluated the frequency of its expression across various T-ALL subtypes and in a large number of patients. In addition, we compared the densities of CD38 molecules on the surface of T-ALL cells with those on residual, normal T cells. We evaluated diagnostic specimens of adult patients with ALL at the central reference laboratory of the German Multicenter Study Group on Adult Acute Lymphoblastic Leukemia (GMALL). The GMALL studies (clinicaltrials gov. identifier: NCT02881086, NCT02872987) were approved by the ethical committees of the participating centers and patients gave written informed consent for biological research, including research on their archived samples. Study procedures were in accordance with the Declaration of Helsinki of 1975, as revised in 2008. Cell samples from 401 consecutive patients were investigated

(Table 1). Of these, 121 samples were derived from patients with T-ALL, including 33 patients with an ETP immunophenotype. In addition, 280 samples from patients with various B-cell lineage ALL subtypes were assessed (Table 1). Expression of CD38 was determined by using the antibody clone LS198.4.3 (Beckmann Coulter, Krefeld, Germany) during routine flow cytometry.9 Quantification of CD38 cell surface molecules was done by bead-calibrated flow cytometry using a commercially available kit (QuantiBRITETM Beads, BD Bioscience, San Jose, California, USA), the antibody clone HB7 (BD Bioscience), and a different flow cytometer (FACSCanto II and FlowJo version 10.1, BD Bioscience), following the manufacturer’s instructions (Online Supplementary Figure S2). Expression of CD38 was almost uniformly present on T-ALL leukemic cells with only two (1.7%) negative cases and at least 80% of cells positive for CD38 in 108 (89%) of all tested 121 T-ALL samples (Figure 1A). In B-cell lineage ALL, 266 (95%) of 280 samples were positive for CD38, but the median percentage of CD38-positive leukemic cells was significantly lower compared to T-ALL (84% vs. 98%, respectively; P<0.0001), and CD38 was present on at least 80% of cells in only 161 (58%) of 280 B-cell lineage ALL samples (Figure 1A). In T-ALL, the median percentage of CD38-positive cells was similar in pre- and thymic T-ALL, but lower in mature T-ALL (98%, 98% and 92%, respectively; P≤0.02; Figure 1B). Eight samples containing ≥10% of residual, normal T cells Table 1. Patient and sample characteristics. Characteristics ALL subtype

Sex M/F Age in years range (median) Cell sample type Bone marrow Peripheral blood Pleural effusion First diagnosis/relapse

N=121

N=280

T-cell lineage 53 pre T (29 ETP-ALL) 47 thymic 21 mature (4 ETP-ALL)

B-cell lineage 18 pro B 204 common 47 pre B 11 mature

90/31

147/133

18-85 (33)

18-86 (55)

92* 26 3 118/3

202** 75 280/-

*1 and **3 undeclared specimens, most likely bone marrow aspirates. ALL: acute lymphoblastic leukemia; ETP: early T-cell precursor; M: male, F: female.

Haematologica | 109 February 2024

661

LETTER TO THE EDITOR

were analyzed in more detail. The median percentage of cells with CD38 expression was higher among T-ALL cells compared to normal residual T cells (Figure 1C). In one pre T-ALL sample, the percentage of CD38 expression was slightly lower on T-ALL cells versus residual, normal T cells (81% vs. 93%; Figure 1C). Likewise, the median fluorescence intensity (MFI) of T-ALL cells was significantly higher compared to residual, normal T lymphocytes in all 121 analyzed samples (Online Supplementary Figure S1). The numbers of CD38 molecules per cell were quantified in 21 T-ALL samples (pre T-ALL n=8, thymic T-ALL n=8, mature n=5) using bead-calibrated flow cytometry (11 fresh samples, 10 samples thawed from liquid nitrogen). The number of CD38 molecules on the surface of T-ALL cells widely ranged from 6,406 to 80,122 (median 19,520) with significant lower median numbers in mature T-ALL compared to pre- and thymic T-ALL (8,839 vs. 25,563 and 25,723; P=0.0194; Figure 2A). The numbers of CD38 molecules per normal, residual T cell were considerably lower compared to T-ALL cells

across all T-ALL subgroups and ranged from 374 to 5,980 only (median 1,321 vs. 19,520; P<0.0001; Figure 2B). The ratios of CD38 molecules on the surface of T-ALL cells versus normal, residual T cells in respective samples ranged from 2.4 to 49.9 (median 16.7). Treatment optimization in T-ALL is urgently needed in patients with high-risk T-ALL or adult patients with advanced age and deemed less suitable candidates for intensification of conventional therapies. The CD38 molecule represents a potential target for integration of respective monoclonal antibodies into the therapy algorithms of patients with ALL. We comprehensively evaluated a large number of T-ALL samples within the framework of the GMALL study group and were able to demonstrate that CD38 is almost uniformly expressed across all T-ALL subtypes. Noteworthy, expression of CD38 was more robust in T-ALL compared to B-cell lineage ALL with higher percentages of positive leukemic cells in T-ALL. Previously published studies investigated lower sample numbers from adult patients and did not

Figure 1. Expression of CD38 in acute lymphoblastic leukemia and residual T cells. A) T- versus B-cell lineage acute lymphoblastic leukemia (ALL). The median percentage of CD38-positive cells was lower in B-cell lineage ALL compared to T-ALL (84% vs. 98%; P<0.0001). Each circle represents the percentage of CD38-positive cells assessed by routine flow cytometry in a given patient sample (- median). (B) CD38 expression in T-ALL subtypes. The median percentage of CD38-positive cells was higher in pre and thymic T-ALL compared to mature T-ALL (98%, 98% and 92%, respectively; P≤0.02). Each circle represents the percentage of CD38-positive cells assessed by routine flow cytometry in a given patient sample (- median). (C) CD38 expression in T-ALL versus residual T cells. The median percentage of cells with CD38 expression was 97% among T-ALL cells compared to 45% among normal, residual T cells (P=0.0156). Each circle (T-ALL) and triangle (T cells) represents the percentage of differential CD38 expression within 1 patient sample. Haematologica | 109 February 2024

662

LETTER TO THE EDITOR

differentiate the T-ALL subtype beyond ETP and non-ETP ALL. One of these studies evaluated the expression of CD38 in 83 pediatric and 105 adult patients with newly diagnosed T-ALL. Using a similar 20% cutoff for positivity, 184 (98%) of 188 diagnostic samples from these patients were categorized positive for CD38 expression. However, the median percentage of CD38-positive cells in this previously published study was less than 90% with higher numbers of samples in the low range compared to 98% in the present study.10 The confounding factors for these observed differences are likely diverse (e.g., use of different antibody clones and flow cytometers with variable setups and gating strategies), but these diverging results underscore that use of a more refined T-ALL classification and application of techniques to assess antigen densities on the target cells should be employed before considering CD38 targeted immunotherapies in T-ALL. Published data comparing the densities of CD38 on malignant and non-malignant hematopoietic cells are scarce and evaluated only a limited number of patient samples.11 We observed a remarkable wide range of CD38 expression across T-ALL cells and normal T cells. It has been shown that CD38 expression is associated with T-cell activation and plays a key role in the interaction of T cells with antigen-presenting cells, which might explain at least in part the variable expression in normal T cells.12 The numbers of cells positive for CD38 in the present study were mostly higher among leukemic cells compared to residual T cells, but there was an overlap in a subset of samples with respect to percentages of positive cells and MFI values. However, the antigen densities measured by bead-calibrated flow cytometry were consistently higher on T-ALL cells. Furthermore, the ratios

of CD38 densities on T-ALL cells versus normal T cells were always >2 and ranged up to 49.9. This indicates that targeting of CD38 on T-ALL cells by therapeutic antibodies is likely more efficient compared to normal T cells. First promising results from a phase II trial evaluating daratumumab together with low-intensity chemotherapy in pediatric patients and young adults with relapsed or refractory ALL (clinicaltrials gov. Identifier: NCT03384654) were released recently. In this study, the rate of complete responses ranged from 60% in young adult patients to 83% in pediatric patients with T-ALL.13 Various immune escape mechanisms could limit the efficacy of CD38 directed immunotherapy. Among these, myeloid checkpoint blockade through inhibition of the interaction between CD47 and its ligands has gained interest. It has recently been shown that a combined targeting of CD38 and CD47 could enhance antibody-dependent phagocytosis of T-ALL cell lines and patient derived T-ALL cells in vitro and prolong survival in a patient-derived T-ALL xenograft model.14 In another study, an IgA2-type variant of daratumumab exhibited more effective cell killing by neutrophils, which was further enhanced by blockade of CD47.15 Interestingly, CD38 expression was shown to be upregulated in the T-ALL cell line HSB-2 after exposure to all-trans retinoic acid, which was associated with enhanced cell killing, especially when combined with CD47 blockade. This suggests that the density of CD38 on target cells and the myeloid checkpoint blockade mediated through CD47 are likely of relevance. Our data demonstrate that CD38 is abundantly expressed in adult T-ALL, with very few exceptions, making this molecule an attractive target in adjunct immunotherapeutic approaches in future clinical trials.

Figure 2. Densities of CD38 molecules in T-cell lineage acute lymphoblastic leukemia and normal T cells. (A) Molecule densities in T-cell lineage acute lymphoblastic leukemia (T-ALL) subtypes. The median densities of CD38 molecules on the surface of pre T, thymic and mature T-ALL cells were 25,563, 25,723 and 8,839, respectively (P=0.0194). (B) Molecule densities on T-ALL versus normal, residual T cells. The median density of CD38 molecules on the surface of T-ALL cells was 19,518 compared to 1,321 for normal, residual T cells (P<0.0001). Haematologica | 109 February 2024

663

LETTER TO THE EDITOR

Authors

Disclosures SH has received travel grants from Sanofi; and honoraria from

Sebastian Koslowski, Rainer Glauben, Stefan Habringer, Thomas

Pentixapharm unrelated to this study. TB has received speakers’

Burmeister, Ulrich Keller,

honoraria from Novartis and Pfizer unrelated to this study. MB has

Gökbuget,

3,8

1,3,5

1,3

Monika Brüggemann,

6,7

Nicola

and Stefan Schwartz

received research grants from Affimed, Amgen and Regeneron;

1,3

consulting fees from Amgen; honoraria from Amgen and Janssen; Department of Hematology, Oncology and Cancer Immunology

and travel grants from Janssen unrelated to this study. SS has

(Campus Benjamin Franklin), Charité - Universitätsmedizin Berlin,

received a research grant from MorphoSys AG related to this study.

corporate member of Freie Universität and Humboldt-Universität zu

SS has received consulting fees from AMGEN, Gilead, Pfizer and

Berlin, Berlin; Department of Gastroenterology, Infectious Diseases

SERB SAS; honoraria from the Akademie für Infektionsmedizin e.V.,

and Rheumatology (Campus Benjamin Franklin), Charité -

AMGEN, AVIR Pharma, CSi Hamburg GmbH, Gilead, Labor28,

Universitätsmedizin Berlin, corporate member of Freie Universität

Novartis, Persberg Group GmbH/DGIM e.V., Pfizer and Vivantes

and Humboldt-Universität zu Berlin, Berlin; German Cancer

GmbH; financial support for research projects from Protherics

Consortium (DKTK) and German Cancer Research Center (DKFZ),

Medicines Development Ltd; and travel grants from Gilead, and

Heidelberg; Department of Hematology, Oncology and Cancer

Novartis, all unrelated to this study. All other authors have no

Immunology (Campus Virchow-Klinikum), Charité -

conflicts of interest to disclose.

Universitätsmedizin Berlin, corporate member of Freie Universität Berlin and Humboldt-Universität zu Berlin, Berlin; 5Max-Delbrück-

Contributions

Center, Berlin; University Cancer Center Schleswig-Holstein

SK, RG and SS performed research. SK, RG, TB, SH, MB and SS

(UCCSH), University Hospital Schleswig-Holstein, Kiel; Department

performed data analysis. SK, RG, UK and SS wrote the manuscript.

of Medicine II, Hematology and Oncology, University Hospital

MB and NG provided essential data and have made substantial

Schleswig-Holstein, Kiel and

contributions to the conception and design. NG, UK and SS

Medizinische Klinik II,

Universitätsklinikum der Johann-Wolfgang-Goethe Universität,

supervised the study. All authors have drafted the work and

Frankfurt/Main, Germany

approved the final manuscript version.

Correspondence:

Acknowledgments

S. SCHWARTZ - stefan.schwartz@charite.de

SK is a MD candidate at the Charité. This work is submitted in partial fulfillment of the requirement for the MD.

https://doi.org/10.3324/haematol.2023.283814 Funding Received: June 22, 2023.

This work was supported in part by a grant to SS from MorphoSys

Accepted: September 1, 2023.

AG, Planegg, Germany.

The laboratory protocols and original data might be obtained upon

Published under a CC BY-NC license

request from the corresponding author.

References 1. Marks DI, Rowntree C. Management of adults with T-cell lymphoblastic leukemia. Blood. 2017;129(9):1134-1142. 2. Hoelzer D, Thiel E, Arnold R, et al. Successful subtype oriented treatment strategies in adult T-All; results of 744 patients treated in three consecutive GMALL studies. Blood. 2009;114(22):324. 3. Coustan-Smith E, Mullighan CG, Onciu M, et al. Early T-cell precursor leukaemia: a subtype of very high-risk acute lymphoblastic leukaemia. Lancet Oncol. 2009;10(2):147-156. 4. Jain N, Lamb AV, O’Brien S, et al. Early T-cell precursor acute lymphoblastic leukemia/lymphoma (ETP-ALL/LBL) in adolescents and adults: a high-risk subtype. Blood. 2016;127(15):1863-1869. 5. Li Y, Yang R, Chen L, Wu S. CD38 as an immunomodulator in cancer. Future Oncol. 2020;16(34):2853-2861. 6. Nijhof IS, Casneuf T, van Velzen J, et al. CD38 expression and complement inhibitors affect response and resistance to

daratumumab therapy in myeloma. Blood. 2016;128(7):959-970. 7. Bride KL, Vincent TL, Im SY, et al. Preclinical efficacy of daratumumab in T-cell acute lymphoblastic leukemia. Blood. 2018;131(9):995-999. 8. Cerrano M, Bonifacio M, Olivi M, et al. Daratumumab with or without chemotherapy in relapsed and refractory acute lymphoblastic leukemia. a retrospective observational Campus ALL study. Haematologica. 2022;107(4):996-999. 9. Baldus CD, Martus P, Burmeister T, et al. Low ERG and BAALC expression identifies a new subgroup of adult acute T-lymphoblastic leukemia with a highly favorable outcome. J Clin Oncol. 2007;25(24):3739-3745. 10. Tembhare PR, Sriram H, Khanka T, et al. Flow cytometric evaluation of CD38 expression levels in the newly diagnosed T-cell acute lymphoblastic leukemia and the effect of chemotherapy on its expression in measurable residual disease, refractory disease and relapsed disease: an implication for

Haematologica | 109 February 2024

664

LETTER TO THE EDITOR anti-CD38 immunotherapy. J Immunother Cancer. 2020;8(1):e000630. 11. Naik J, Themeli M, de Jong-Korlaar R, et al. CD38 as a therapeutic target for adult acute myeloid leukemia and T-cell acute lymphoblastic leukemia. Haematologica. 2019;104(3):e100-e103. 12. Munoz P, Mittelbrunn M, de la Fuente H, et al. Antigen-induced clustering of surface CD38 and recruitment of intracellular CD38 to the immunologic synapse. Blood. 2008;111(7):3653-3664. 13. Hogan LE, Bhatla T, Teachey DT, et al. Efficacy and safety of daratumumab (DARA) in pediatric and young adult patients with

relapsed/refractory T-cell acute lymphoblastic leukemia (ALL) or lymphoblastic lymphoma (LL): initial results from the phase 2 DELPHINUS study. J Clin Oncol. 2022;40(s16):abstract 10001. 14. Muller K, Vogiatzi F, Winterberg D, et al. Combining daratumumab with CD47 blockade prolongs survival in preclinical models of pediatric T-ALL. Blood. 2022;140(1):45-57. 15. Baumann N, Arndt C, Petersen J, et al. Myeloid checkpoint blockade improves killing of T-acute lymphoblastic leukemia cells by an IgA2 variant of daratumumab. Front Immunol. 2022;13:949140.

Haematologica | 109 February 2024

665

LETTER TO THE EDITOR

Oligosecretory Waldenstrom macroglobulinemia exhibits excellent treatment response and outcomes Lymphoplasmacytic lymphoma (LPL) is characterized by a monoclonal expansion of predominantly small B lymphocytes with variable differentiation from plasmacytoid lymphocytes to plasma cells.1 Waldenstrom macroglobulinemia (WM) is a rare B-cell malignancy characterized as an immunoglobulin M-secreting LPL, while it does not require the quantification of immunoglobulin M (IgM).2,3 Most WM patients present with elevated IgM and 15% of them may present with hyperviscosity at diagnosis.4 Moreover, in many clinical trials and studies, some patients have very low IgM levels, or even levels in the normal IgM range.5-7 According to the latest response criteria consensus from the 11th International Workshop on Waldenstrom’s macroglobulinemia and National Comprehensive Cancer Network (NCCN) guidelines (Version 2.2022),8,9 a reduction in serum IgM ≥90% and ≥50% is defined as very good partial response (VGPR) and partial response (PR), respectively.10 When the IgM level is within twice the upper limit of normal (ULN), it is impossible to make an accurate efficacy evaluation. Therefore, in our study, patients with an initial IgM quantification lower than or equal to two times the ULN were defined as “oligosecretory WM”. Those with twice higher than the ULN value were defined as “measurable WM”. The characteristics and prognosis of these patients have not been studied. Here, we aimed to: 1) present the clinical features of oligosecretory WM; 2) evaluate the tumor burden of oligosecretory WM; and 3) explore the treatment response and outcomes of oligosecretory WM. We, therefore, performed a retrospective study based on the database of the Chinese Registration Network for Waldenstrom Macroglobulinemia (CRNWM) which included 1,420 LPL/WM patients diagnosed between January 2003 and September 2020 in 35 hematologic centers in China.11 A total of 1,274 patients with newly diagnosed WM were included in the analysis. Patients receiving two or more courses of treatment were defined as systemic treatment. The treatment response in measurable WM patients was assessed according to the NCCN guidelines (Version 2.2022) of Waldenstrom Macroglobulinemia/ Lymphoplasmacytic Lymphoma and the latest consensus from the 11th International Workshop on Waldenstrom’s macroglobulinemia.8,9 A complete response (CR) requires the absence of serum monoclonal IgM protein by immunofixation, complete resolution of extramedullary disease, and morphologically normal bone marrow (BM) aspirate. This CR criterion was applied to all the WM patients. For oligosecretory WM patients, bone marrow biopsy (BMB), flow cytometry (FCM), extramedullary disease, and clinical manifestations were combined to make a comprehensive judgment of PR. When

the tumor cells of BM were reduced by ≥50% (by BMB and/ or FCM), accompanied by a reduction in spleen volume and lymph node size, and improvement in clinical symptoms, it was defined as ≥PR. When BM tumor cells were increased by more than 50% (by BMB or FCM) from nadir (requires confirmation) and/or progression in clinical features attributed to the disease, it was defined as progressive disease (PD). Informed consent was obtained from each patient, and the study was approved by the institutional ethics committee of each center (IIT2021030-EC-1). Patients’ characteristics were summed using median and interquartile range (IQR) for continuous variables, and absolute and relative frequencies for categorical variables. The association between two categorical variables was analyzed using χ2 or Fisher’s exact test for the qualitative variables and independent sample t test for quantitative variables. Among the 1,274 enrolled patients, 80 (6.3%) were classified as oligosecretory WM based on our definition; median serum IgM level was 3.52g/L (range: 0.15-5.92 g/L). The clinical characteristics of oligosecretory WM and measurable WM are described in Table 1. Median age of oligosecretory WM patients was 65 years (range: 31-88), with a male-to-female ratio of 2.5:1, and median hemoglobin (Hb) of 8.4 g/dL (interquartile range [IQR]: 6.4-10.5). Age and sex distribution, and Hb levels were similar between the two groups. Using allele-specific polymerase chain reaction, 47 patients in the oligosecretory WM group and 607 patients in the measurable WM group had MYD88L265P mutation, with positivity rates of 83.0% and 70.3% (P=0.065) respectively, although differences between the two groups were not significant. Importantly, compared with the measurable WM, oligosecretory WM had a higher proportion of thrombocytopenia (41.2% vs. 27.4%, P=0.008) and a lower proportion of hypoalbuminemia (32.9% vs. 64.6%, P<0.001) and elevated serum β2-microglobulin (57.1% vs. 73.8%, P=0.002). Information on the International Prognostic Scoring System for Waldenstrom macroglobulinemia (IPSSWM) was available for 75 patients in the oligosecretory WM group and 1,042 patients in the measurable WM group. In the oligosecretory WM group, the proportion of high-risk patients was lower than that of the measurable WM group (40.0% vs. 51.5%, P=0.054) (Table 1). Besides IgM level, we also evaluated patient tumor burden according to: malignant cells by FCM in BM, malignant cells by BMB, splenomegaly, and lymphadenopathy. No difference was observed in median malignant BM cells by FCM (9.9% vs. 8.9%, P=0.114), splenomegaly (38.4% vs. 35.2%, P=0.583), or lymphadenopathy (44.8% vs. 39.1%,

Haematologica | 109 February 2024

666

LETTER TO THE EDITOR Table 1. Baseline characteristics of newly diagnosed Waldenstrom macroglobulinemia with oligosecretory Waldenstrom macroglobulinemia and measurable disease. Oligosecretory WM N=80

Measurable WM N=1,194

Age in years Median (range) ≥65, N (%)

65 (31-88) 37 (46.3)

64 (27-90) 538 (45.1)

0.831 0.836

B symptoms, N (%)

17 (23.6)

252 (22.9)

0.891

Characteristic

Gender, male, N (%) Lymphadenopathy ≥1.5 cm, N (%) Splenomegaly ≥15 cm, N (%) MYD88

L265P

mutation, N (%)

Hemoglobin, g/L Median (IQR) ≤115, N (%) Platelets, x109/L Median (IQR) ≤100, N (%)

ALC, x109/L, median (IQR)

Serum β2-microglobulin, mg/L Median (IQR) >3 mg/L, N (%)

LDH, U/L Median (IQR) ≥250, N (%)

Serum albumin, g/L Median (IQR) <35 g/L, N (%)

Malignant cells by FCM of BM, %, median (IQR) Malignant cells by BMB >50%, N (%) IPSSWM Low-risk, N (%) Intermediate-risk, N (%) High-risk, N (%)

57 (71.3)

26 (44.8)

28 (38.4)

874 (73.2)

296 (39.1)

321 (35.2)

0.685

0.390

0.583

39/47 (83.0)

427/607 (70.3)

0.065

84 (67-105) 69 (86.2)

84 (69-103) 1,010 (85.0)

0.998 0.764

137 (61-237) 33 (41.2)

164 (95-250) 316 (27.4)

0.045 0.008

1.69 (0.94-2.93)

1.63 (1.13-2.54)

0.766

3.25 (2.43-3.54) 40 (57.1)

4.1 (2.91-5.80) 736 (73.8)

0.874 0.002

175 (142.3-236.5) 13 (17.1)

146 (112.4-197.5) 138 (12.7)

0.040 0.275

37.7 (32.1-42.0) 26 (32.9)

32.0 (27.5-36.7) 749 (64.6)

0.000 0.000

14/26 (53.8)

496/912 (54.4)

0.957

14 (18.7) 31 (41.3) 30 (40.0)

139 (13.2) 366 (35.1) 537 (51.5)

0.195 0.278 0.054

9.9 (3.6-44.5)

8.9 (2.5-26.0)

0.114

WM: Waldenstrom macroglobulinemia; N: number; IQR: interquartile range; ALC: absolute lymphocyte count; LDH: lactic dehydrogenase; FCM: flow cytometry; BM: bone marrow; BMB: bone marrow biopsy; IPSSWM: International Prognostic Scoring System for Waldenstrom macroglobulinemia.

P=0.390) between the two groups. Malignant cells by BMB were given in four groups: percentage of malignant cells ≤5%, 5-20%, 20-50%, and ≥50%. The percentage of malignant cells by BMB of different ranges were also similar between the two groups (malignant cells ≤5%, 11.5% vs. 6.9%, P=0.645; 5-20%, 19.2% vs. 15.2%, P=0.305; 20-50%, 15.4% vs. 23.5%, P=0.336; ≥50%, 53.8% vs. 54.4%, P=0.957). Furthermore, there was a significantly higher proportion of patients with >50% abnormal cells in oligosecretory WM patients compared with the measurable WM group (22.7% vs. 12.4%, P=0.048) by FCM, suggesting that some patients with low IgM levels still had high tumor infiltration of BM (Figure 1). Overall, the BM tumor burden for oligosecretory WM was comparable with the measurable WM. Treatment information was available for 45 (56.3%) and 682 (57.1%) patients in the oligosecretory WM group and measurable WM group, respectively. We summarized the treatment options for all patients into four regimens: rit-

Figure 1. The percentage of malignant cells in bone marrow by flow cytometry. WM: Waldenstrom macroglobulinemia; FCM: flow cytometry; ns: not significant.

Haematologica | 109 February 2024

667

LETTER TO THE EDITOR

Figure 2. Kaplan-Meier curves for survival of oligosecretory Waldenstrom macroglobulinemia patients and measurable Waldenstrom macroglobulinemia patients. (A) Overall and (B) progression-free survival of the two groups.

uximab-based regimens (R-based), bortezomib-based regimens (B-based), BTK inhibitors, and traditional cytotoxic drug regimens. The treatment regimens were comparable between the two groups (Online Supplementary Table S1). Thirty-four (42.5%) oligosecretory WM patients had a post-treatment BM biopsy, which could evaluate the treatment response, while treatment response was evaluated in 559 patients (46.8%) in the measurable group. At a median follow-up of 21.1 months, 8 patients had died in the oligosecretory WM group and 173 in the measurable WM group. Overall, the 3-year overall survival (OS) rates in the oligosecretory WM group and measurable WM group were 83.4% and 87.3%, respectively (P=0.890) (Figure 2A). Interesting, the CR rate and 3-year PFS rate were both higher in the oligosecretory group, but with highly heterogeneous treatments used in both groups. Among the 34 evaluable oligosecretory patients, 21 (61.8%) achieved ≥PR, and 8 (23.5%) achieved <PR. Overall response rates were 85.3% and 76.0% in the oligosecretory and measurable group (CR 5.4%, ≥PR 57.1%, minor response 13.6%), respectively. Importantly, 5 oligosecretory WM patients achieved CR, which was significantly higher than the measurable WM group (14.7% vs. 5.4%, P=0.043). During the follow-up period, 8 patients in the oligosecretory WM group and 286 patients in the measurable WM group experienced disease progression. The 3-year PFS rate was 59.6% in the measurable WM group and 78.8% in the oligosecretory WM group (P=0.001) (Figure 2B). Subsequently, we validated the predictive performance of the IPSSWM for survival in oligosecretory WM patients. The IPSSWM had a prognostic role for PFS (P=0.070) but not for OS (P=0.280) in oligosecretory WM (Online Supplementary Figure S1). This study is the first to study the characteristics and survival of a large cohort of patients with very low IgM levels. Treon et al.12 have defined 10 g/L as the cut-off value for low IgM WM. We found that WM patients with IgM levels lower than 10 g/L also showed better PFS with no signifi-

cant differences found (Online Supplementary Figure S2). We believe that initial IgM status lower than two times the ULN was a reasonable and clinically useful threshold. Interestingly, we found that patients with oligosecretory WM had different clinical characteristics. Although oligosecretory WM had low IgM, the tumor load was actually not low. Importantly, this group of patients had a good treatment response and PFS. However, the study has some limitations. Firstly, the retrospective design may lead to potential bias. Secondly, the number of patients available for treatment regimens and efficacy was limited, so we could not perform a subgroup analysis of different treatment responses and meta-regression analyses of the PFS of oligosecretory WM patients. Lastly, the follow-up period was not long enough to evaluate long-term survival outcomes, which may be one of the causes for the negligible differences in OS between two groups. In the future, we will expand the cohort and extend the follow-up time for more detailed subgroup analysis. Previous studies have shown that Chinese WM patients have unique genetic characteristics.13,14 We also expect that genetic information such as gene mutations, chromosome karyotypes, and IGHV gene repertoire will further deepen our knowledge of this rare entity.

Authors Wenjie Xiong,1,2* Ying Yu,1,2* Chunyan Sun,3* Juan Du,4* Zhen Cai,5* Zanzan Wang,6 Xinxin Cao,7 Yuting Yan,1,2 Jiawen Chen,1,2 Yanshan Huang,1,2 Zhongxing Jiang,8 Huihan Wang,9 Ting Niu,10 Guangzhong Yang,11 Hua Xue,12 Bingzong Li,13 Honghui Huang,14 Zhenling Li,15 Qinhua Liu,16 Fei Li,17 Ou Bai,18 Min Mao,19 Rong Fu,20 Ling Wang,21 Chunrui Li,3 Xiaoxia Chu,22 Lihong Liu,23 Yujun Dong,24 Luqun Wang,25 Jun Luo,26 Yongqiang Wei,27 Rui Cui,28 Lugui Qiu,1,2 Jian Li7 and Shuhua Yi;1,2 on behalf of the Chinese Workshop on Waldenström Macroglobulinemia (CWWM)

Haematologica | 109 February 2024

668

LETTER TO THE EDITOR

State Key Laboratory of Experimental Hematology, National Clinical

WX, YYu, CS, JD and Zc contributed equally as first authors.

Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology Blood Diseases Hospital,

Correspondence:

Chinese Academy of Medical Sciences Peking Union Medical College,

S. YI - yishuhua@ihcams.ac.cn

Tianjin; Tianjin Institutes of Health Science, Tianjin; Institute of

J. LI - lijian@pumch.cn

Hematology, Union Hospital, Tongji Medical College, Huazhong https://doi.org/10.3324/haematol.2023.283402

University of Science and Technology, Hubei; 4Department of Hematology, The Myeloma and Lymphoma Center, Shanghai Changzheng Hospital, Naval Medical University, Shanghai; 5Bone

Received: April 27, 2023.

Marrow Transplantation Center, The First Affiliated Hospital, School

Accepted: September 1, 2023.

of Medicine, Zhejiang University, Hangzhou; Department of

Early view: September 14, 2023.

Hematology, Ningbo First Hospital, Zhejiang; Department of 7

Hematology, Peking Union Medical College Hospital, Chinese

Academy of Medical Sciences Peking Union Medical College, Beijing;

Published under a CC BY-NC license

The First Affiliated Hospital of Zhengzhou University, Henan;

Shengjing Hospital of China Medical University, Liaoning; 10West

Disclosures

China Hospital Sichuan University, Chengdu, Sichuan; Department

No conflicts of interest to disclose.

of Hematology, Beijing Chao-Yang Hospital, Capital Medical University, Beijing; 12Affiliated Hospital of Hebei University, Hebei;

Contributions

Department of Hematology, The Second Affiliated Hospital of

SY conceptualized the study. WX, YYu, CS, JD and ZC analyzed data,

Soochow University, Suzhou, Jiangsu; Department of Hematology,

performed statistical analyses, and wrote the manuscript. ZW, XCa,

Renji Hospital, Shanghai Jiaotong University School of Medicine,

YYa, JC, YH, ZJ, HW, TN, GY, HX, BL, HH, ZL, QL, FL, OB, MM, RF, LiW,

Shanghai; Department of Hematology, China-Japan Friendship

CL, XCh, LL, YD, LuW, JL, YW and RC acquired data and managed

Hospital, Beijing; Department of Hematology, The First Affiliated

patients. LQ, JL and SY revised the manuscript critically and

Hospital of Anhui Medical University, Anhui; The First Affiliated

approved the final version.

Hospital of Nanchang University, Jiangxi; Department of 18

Hematology, The First Hospital of Jilin University, Jilin; 19Department

Funding

of Hematology, Xinjiang Uiger Municipal People’s Hospital, Xinjiang;

This work was supported by grants from the National Nature

Department of Hematology, Tianjin Medical University General

Science Foundation of China (81970187, 82170193, 81920108006,

Hospital, Tianjin; Qingdao Central Hospital, Shandong;

The

81900203, 82200215), the Chinese Academy of Medical Sciences

Affiliated Yantai Yuhuangding Hospital of Qingdao University,

Innovation Fund for Medical Sciences (2022-I2M-1-022), Tianjin

Shandong;

Health Science and Technology Project (TJWJ2022XK021), and

Department of Hematology, The Fourth Hospital of

Hebei Medical University, Hebei;

University First Hospital, Beijing;

Department of Hematology, Peking

Tianjin Health Research Project (TJSQNYXXR-D2-152).

Department of Hematology, Qilu

Hospital of Shandong University, Shandong; 26Department of

Data-sharing statement

Hematology, The First Affiliated Hospital of Guangxi Medical

Data are available from the corresponding author on reasonable

University, Guangxi; Department of Hematology, Nanfang Hospital,

request.

Southern Medical University, Guangdong and

Department of

Hematology, Tianjin First Center Hospital, Tianjin, China

References 1. Cazzola M. Introduction to a review series: the 2016 revision of the WHO classification of tumors of hematopoietic and lymphoid tissues. Blood. 2016;127(20):2361-2364. 2. Sabattini E, Bacci F, Sagramoso C, Pileri SA. WHO classification of tumours of haematopoietic and lymphoid tissues in 2008: an overview. Pathologica. 2010;102(3):83-87. 3. Owen RG, Treon SP, Al-Katib A, et al. Clinicopathological definition of Waldenstrom’s macroglobulinemia: consensus panel recommendations from the Second International Workshop on Waldenstrom’s Macroglobulinemia. Semin Oncol. 2003;30(2):110-115. 4. Dimopoulos MA, Kyle RA, Anagnostopoulos A, Treon SP. Diagnosis and management of Waldenstrom’s

macroglobulinemia. J Clin Oncol. 2005;23(7):1564-1577. 5. Buske C, Tedeschi A, Trotman J, et al. Ibrutinib plus rituximab versus placebo plus rituximab for Waldenstrom’s macroglobulinemia: final analysis from the randomized phase III iNNOVATE study. J Clin Oncol. 2022;40(1):52-62. 6. Kastritis E, Morel P, Duhamel A, et al. A revised international prognostic score system for Waldenstrom’s macroglobulinemia. Leukemia. 2019;33(11):2654-2661. 7. Tam CS, Opat S, D’Sa S, et al. A randomized phase 3 trial of zanubrutinib vs ibrutinib in symptomatic Waldenstrom macroglobulinemia: the ASPEN study. Blood. 2020;136(18):2038-2050. 8. Treon SP, Tedeschi A, San-Miguel J, et al. Report of consensus

Haematologica | 109 February 2024

669

LETTER TO THE EDITOR Panel 4 from the 11th International Workshop on Waldenstrom’s macroglobulinemia on diagnostic and response criteria. Semin Hematol. 2023;60(2):97-106. 9. NCCN Clinical Practice Guidelines in Oncology: Waldenström macroglobulinemia/lymphoplasmacytic lymphoma Version 2.2022. http://www.nccn.org/patients. Accessed December 2022. 10. Owen RG, Kyle RA, Stone MJ, et al. Response assessment in Waldenström macroglobulinaemia: update from the VIth International Workshop. Br J Haematol. 2013;160(2):171-176. 11. Cao XX, Yi SH, Jiang ZX, et al. Treatment and outcome patterns of patients with Waldenstrom’s macroglobulinemia: a large, multicenter retrospective review in China. Leuk Lymphoma. 2021;62(11):2657-2664.

12. Tripsas CK, Patterson CJ, Uljon SN, Lindeman NI, Turnbull B, Treon SP. Comparative response assessment by serum immunoglobulin M M-protein and total serum immunoglobulin M after treatment of patients with Waldenstrom macroglobulinemia. Clin Lymphoma Myeloma Leuk. 2013;13(2):250-252. 13. Xiong W, Wang T, Yu Y, et al. Cytogenetic aberrations of lymphoplasmacytic lymphoma/Waldenstrom’s macroglobulinemia in Chinese patients. Chin Med J (Engl). 2023;136(10):1240-1242. 14. Wang J, Yan Y, Xiong W, et al. Landscape of immunoglobulin heavy chain gene repertoire and its clinical relevance to LPL/ WM. Blood Adv. 2022;6(13):4049-4059.

Haematologica | 109 February 2024

670

LETTER TO THE EDITOR

Exclusion of persistent mutations in splicing factor genes and isocitrate dehydrogenase 2 improves the prognostic power of molecular measurable residual disease assessment in acute myeloid leukemia Accurate risk assessment is crucial for the management of patients with acute myeloid leukemia (AML).1 The detection of measurable residual disease (MRD) after remission induction therapies has been shown to be an independent risk factor for relapse and death.2 The use of next-generation sequencing (NGS)-based techniques to detect mutations found in leukemic cells has emerged as a promising approach for MRD assessment.3,4 One of the main challenges of this approach is differentiating between mutations that are found only in the leukemic cell population (henceforth termed “AML-related”) and those associated with clonal hematopoiesis (CH). The persistence of CH during remission has not been associated with inferior clinical outcomes.5 Approaches involving genotyping of sorted populations or single cells are required to identify the cellular origins of the mutations, but they are not yet practical for routine clinical use. To overcome this challenge, a common practice is to exclude mutations in three genes, namely DNMT3A, TET2, and ASXL1 (collectively known as DTA), from molecular MRD assessment,1,3,4 because they are among the most frequently mutated genes in people with clonal hematopoiesis of indeterminate potential (CHIP).6,7 However, mutations in other genes are also found in CHIP carriers.6-8 Moreover, the relative frequencies of CH-related mutations in AML patients differ from those of CHIP carriers who, by definition, do not have any other hematologic abnormalities.5 This discordance is likely because the risk of AML development varies between different CH-related mutations.8 Thus, it is unclear whether DTA mutations are the optimal ones for exclusion in molecular MRD analysis in AML. To address the above uncertainty, we systematically analyzed the impact of exclusion of mutations in 22 myeloid malignancy-associated genes on the difference in clinical outcomes between patients stratified as MRD-positive (MRDPOS) and MRD-negative (MRDNEG). To perform this analysis, we studied 114 newly diagnosed AML patients who received high-intensity induction chemotherapy and achieved a complete remission. The clinical characteristics of the patients are listed in Online Supplementary Table S1. We performed targeted conventional NGS analysis on DNA extracted from their diagnostic peripheral blood or bone marrow samples. Variants classified as “be-

nign” or “likely benign” based on American College of Medical Genetics criteria were excluded from further analysis.9 During the remission phase, we collected a total of 223 peripheral blood samples upon count recovery at a median of 36 days after induction (n=93) or after one (n=93), two (n=35), or three (n=2) cycles of consolidation chemotherapy. Remission samples were collected at two different timepoints (designated as T1 and T2) for 95.6% (n=109) of the patients. The remaining patients (n=5) had one remission sample collected at T1. To detect mutations in the remission samples, we used a custom 37-gene hybrid capture panel and error-corrected NGS based on the duplex sequencing approach.10,11 A total of 336 mutations in 35 genes were identified and passed the American College of Medical Genetics filtering step in the diagnostic samples. Of those, we excluded 26 mutations in genes that were not covered by the custom panel for MRD detection from further analysis. A further 13 mutations in five genes (PHF6, KDM6A, JAK2, KIT, CBL) were excluded because of a low number of mutational events per gene (≤4 events). The remaining 297 mutations were distributed across 22 genes in 101 patients. Mutations in genes that share a common pathogenic mechanism were analyzed as a group. The groups were splicing factor mutations (SRSF2, SF3B1, and U2AF1), RAS mutations (KRAS and NRAS), and cohesin complex mutations (RAD21 and STAG2). Mutations in IDH1 and IDH2 were analyzed separately because of recent evidence demonstrating distinct clinical and co-mutational patterns between the two types of mutations in patients with myeloid malignancies.12 The use of duplex sequencing enabled sensitive and accurate measurement of the allele frequency of each mutation in the remission samples. The distribution of mutant allele frequencies at T1 (Online Supplementary Figure S1A) and stability of the mutations between T1 and T2 were highly variable across the genes (Online Supplementary Figure S1B). At one end of the spectrum were mutations that demonstrated high levels of persistence and stability during remission, such as DNMT3A mutations. At the other end were mutations characterized by lower levels of persistence and higher probability of clearance with chemotherapy, such as NPM1 mutations. The characteristics of most mutations fell somewhere between

Haematologica | 109 February 2024

671

LETTER TO THE EDITOR the two extremes. It is noteworthy that some mutations, including mutations in the splicing factor genes and IDH2, demonstrated a comparable level of persistence and stability as DNMT3A mutations. To evaluate the impact of mutations in each gene on MRD analysis in an unbiased manner, we generated 2,500 unique permutations in which each of the 15 genes or three gene groups was randomly assigned to be included or excluded from MRD assessment. For each permutation, we calculated the hazard ratio for overall survival between MRDPOS and MRDNEG patients in our study cohort. Patients with mutations in any of the included genes above a mutant allele frequency cutoff of 0.01 (1%) at T1 or T2 were considered MRDPOS. Permutations that excluded DNMT3A and TET2 mutations were associated

with higher hazard ratios, whereas permutations that excluded well-characterized AML-related mutations, such as NPM1 and RAS mutations, correlated with lower hazard ratios (Figure 1A). To determine the significance and magnitude of these associations, we used the KolmogorovSmirnov test to compare the distribution of hazard ratios among the subset of permutations in which a specific gene or gene group was excluded with the reference distribution of all 2,500 permutations. This analysis showed that exclusion of DNMT3A or TET2 mutations significantly shifted the hazard ratio distribution higher relative to the reference distribution (Figure 1B). Intriguingly, exclusion of mutations in the splicing factor genes or IDH2 also significantly shifted the hazard ratio distribution higher (Figure 1B), and their inclusion eliminated the highest hazard

Haematologica | 109 February 2024

672

Figure 1. Evaluation of the impact of exclusion or inclusion of myeloid malignancy-associated mutations on the prognostic power of molecular measurable residual disease assessment. (A) Heatmap showing each of the 2,500 unique permutations ordered according to their associated hazard ratios for overall survival. A yellow cell indicates exclusion of the indicated gene mutation for measurable residual disease assessment, whereas a blue cell indicates inclusion. (B) Volcano plot showing statistical significance plotted against the D statistic from the Kolmogorov-Smirnov test comparing the distribution of hazard ratios of permutations in which the indicated gene mutation is excluded versus the reference distribution. See text for details. (C) Violin plots showing the distribution of hazard ratios of permutations in which the indicated gene mutation is included (+) for measurable residual disease assessment. HR: hazard ratio.

LETTER TO THE EDITOR

Figure 2. Exclusion of mutations in splicing factor genes and IDH2 in addition to DNMT3A, TET2, and ASXL1 mutations improves the prognostic power of molecular measurable residual disease assessment. Kaplan-Meier plots for (A) overall survival (OS), (B) relapse-free survival (RFS), and (C) cumulative incidence of relapse (CIR) of patients classified as measurable residual disease (MRD)-positive (MRD+) or MRD-negative (MRD-) and based on whether DNMT3A, TET2, and ASXL1 (DTA) mutations or mutations in DTA, splicing factor genes and IDH2 (DTAS12) were excluded from MRD determination. A mutant allele frequency of >0.01 (1%) at timepoints 1 and 2 was used to define MRD positivity. Comparison of the hazard ratios for OS, RFS, and CIR in the Morita et al. (D) or Ahn et al. (E) validation cohort between exclusion of mutations in DTA alone versus DTAS12 for determination of MRD status. A mutant allele frequency of >0.01 (1%) was used to define MRD positivity. The hazard ratios and P values shown in all panels were calculated using the Cox proportional-hazards model.

Haematologica | 109 February 2024

673

LETTER TO THE EDITOR ratio values found in the reference distribution (Figure 1C). These findings, along with their high level of persistence and stability during remission, suggest that mutations in the splicing factor genes and IDH2 should be considered CH-related for the purpose of molecular MRD assessment. Indeed, exclusion of these mutations in addition to DTA (henceforth referred to as “DTASI2”) led to a greater difference in overall survival, relapse-free survival, and cumulative incidence of relapse between MRDPOS and MRDNEG patients using a mutant allele frequency cutoff at 0.01 (1%) or 0.005 (0.5%) when compared with exclusion of DTA mutations alone or no mutations (Figure 2A-C, Online Supplementary Figure S2A-C). To ensure that these observations were not restricted to our cohort of patients and the analysis of peripheral blood samples, we analyzed two independent datasets that used panel-based targeted NGS for MRD monitoring of bone marrow samples in AML patients in remission.4,13 The total numbers of patients in the cohorts studied by Morita et al.4 and Ahn et al.13 were 131 and 124, respectively. In both validation cohorts, exclusion of DTASI2 mutations increased the hazard ratio for overall survival, relapse-free survival, and cumulative incidence of relapse in MRDPOS patients compared with exclusion of DTA mutations alone (Figure 2D, E). Our findings demonstrate that the exclusion of mutations in splicing factor genes and IDH2 improves the prognostic power of NGS-based MRD assessment. This effect is likely attributable to their involvement in CH. Our results are consistent with those of a recent study showing that the persistence of IDH1/2 and SRSF2 mutations had no impact on survival in NPM1-mutated AML patients.14 Due to the size of our study cohort, the potential impact of mutations in other genes with low representation could have been missed. Notably, our analysis did not identify ASXL1 mutations as CH-related. However, the number of patients with ASXL1 mutations in our cohort was small (n=9). Further studies are required to clarify the significance of persistence of ASXL1 mutations in MRD analysis. In addition, it is important to emphasize that patients in our study cohort were treated with intensive chemotherapy and thus our findings may not be applicable to patients treated with other therapies (e.g., venetoclax-based regimens). Our work highlights the importance of optimizing the definition of CH-related gene mutations for molecular MRD assessment.

Zhao,1 Yangqiao Zheng,1 Vikas Gupta,1 Dawn Maze,1 Caroline D. McNamara,2 Mark D. Minden,1 Aaron D. Schimmer,1 Hassan Sibai,1 Karen W. Y. Yee,1 Jose-Mario Capo-Chichi,1,3 Tracy L. Stockley,1,3 Andre C. Schuh,1 Scott V. Bratman1# and Steven M. Chan1# Princess Margaret Cancer Center, University Health Network,

Toronto, Ontario, Canada; 2Royal Brisbane and Women’s Hospital, Herston, Queensland, Australia and 3Laboratory Medicine Program, University Health Network, Toronto, Ontario, Canada *

TM and JZ contributed equally as first authors.

SVB and SMC contributed equally as senior authors.

Correspondence: STEVEN M. CHAN - steven.chan@uhnresearch.ca https://doi.org/10.3324/haematol.2023.283510 Received: May 10, 2023. Accepted: June 14, 2023. Early view: June 22, 2023. ©2023 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures No conflicts of interest to disclose. Contributions TM, JZ, SVB, and SMC designed and performed the research, analyzed the data, and wrote the paper. VG, DM, CJM, MDM, ADS, HS, KWLY, and ACS managed and treated the patients. AA, TTW, ZZ, YZ, J-MC-C, and TLS performed the research. Acknowledgments The authors thank the leukemia tissue bank and the Liquid Biopsy Evaluation and Repository Development (LIBERATE) study team at Princess Margaret Cancer Centre for the collection and storage of samples. The authors also thank Dr. Dennis Kim at Princess Margaret Cancer Centre and Dr. Koichi Takashashi at MD Anderson Cancer Center for providing clinical data of the independent cohorts. Funding This work was supported by a Collaborative Personalized Cancer Medicine Team Grant from the Princess Margaret Cancer Foundation and an Acute Leukemia Translational Research Initiative Grant from the Ontario Institute for Cancer Research.

Authors

Data-sharing statement Sequencing and mutation data are available upon request to the cor-

Tracy Murphy, Jinfeng Zou, Andrea Arruda, Ting Ting Wang, Zhen

responding author.

Haematologica | 109 February 2024

674

LETTER TO THE EDITOR

References 1. Dohner H, Wei AH, Appelbaum FR, et al. Diagnosis and management of AML in adults: 2022 ELN recommendations from an international expert panel. Blood. 2022;129(4):424-447. 2. Walter RB, Ofran Y, Wierzbowska A, et al. Measurable residual disease as a biomarker in acute myeloid leukemia: theoretical and practical considerations. Leukemia. 2021;35(6):1529-1538. 3. Jongen-Lavrencic M, Grob T, Hanekamp D, et al. Molecular minimal residual disease in acute myeloid leukemia. N Engl J Med. 2018;378(13):1189-1199. 4. Morita K, Kantarjian HM, Wang F, et al. Clearance of somatic mutations at remission and the risk of relapse in acute myeloid leukemia. J Clin Oncol. 2018;36(18):1788-1797. 5. Tanaka T, Morita K, Loghavi S, et al. Clonal dynamics and clinical implications of postremission clonal hematopoiesis in acute myeloid leukemia. Blood. 2021;138(18):1733-1739. 6. Jaiswal S, Fontanillas P, Flannick J, et al. Age-related clonal hematopoiesis associated with adverse outcomes. N Engl J Med. 2014;371(26):2488-2498. 7. Genovese G, Kahler AK, Handsaker RE, et al. Clonal hematopoiesis and blood-cancer risk inferred from blood DNA sequence. N Engl J Med. 2014;371(26):2477-2487. 8. Fabre MA, de Almeida JG, Fiorillo E, et al. The longitudinal

dynamics and natural history of clonal haematopoiesis. Nature. 2022;606(7913):335-342. 9. Kopanos C, Tsiolkas V, Kouris A, et al. VarSome: the human genomic variant search engine. Bioinformatics. 2019;35(11):1978-1980. 10. Kennedy SR, Schmitt MW, Fox EJ, et al. Detecting ultralowfrequency mutations by duplex sequencing. Nat Protoc. 2014;9(11):2586-2606. 11. Murphy T, Zou J, Daher-Reyes GS, et al. Impact of preleukemic mutations and their persistence on hematologic recovery after induction chemotherapy for AML. Blood Adv. 2019;3(15):2307-2311. 12. Middeke JM, Metzeler KH, Rollig C, et al. Differential impact of IDH1/2 mutational subclasses on outcome in adult AML: results from a large multicenter study. Blood Adv. 2022;6(5):1394-1405. 13. Ahn JS, Kim T, Jung SH, et al. Allogeneic transplant can abrogate the risk of relapse in the patients of first remission acute myeloid leukemia with detectable measurable residual disease by next-generation sequencing. Bone Marrow Transplant. 2021;56(5):1159-1170. 14. Cappelli LV, Meggendorfer M, Baer C, et al. Indeterminate and oncogenic potential: CHIP vs CHOP mutations in AML with NPM1 alteration. Leukemia. 2022;36(2):394-402.

Haematologica | 109 February 2024

675

LETTER TO THE EDITOR

Clinical outcomes of patients with myelofibrosis after immediate transition to momelotinib from ruxolitinib Janus kinase inhibitors (JAKi) such as ruxolitinib, approved for the treatment of myelofibrosis (MF), confer symptom and spleen improvements but can induce or worsen anemia and thrombocytopenia.1-4 Although there is no consensus on the definition of JAKi treatment failure in MF, anemia and thrombocytopenia may require a reduction in JAKi dosing or discontinuation, which are associated with poor overall survival.5-7 In addition, discontinuation from ruxolitinib is complicated by the potential for discontinuation syndrome characterized by acute relapse of symptoms, splenomegaly, anemia, thrombocytopenia, and risk of hemodynamic decompensation,5,8 with approximately 40% of the cases being moderate or severe according to real-world evidence.9 Given that discontinuation rates with ruxolitinib are high (up to 89% at 3 years) and dose modifications of ruxolitinib are associated with lower survival,10,11 we sought to examine how transitioning directly from ruxolitinib to another therapy may be beneficial to patients with MF. Here, we present data from a retrospective analysis of a phase III clinical study (ClinicalTrials.gov identifier: NCT01969838) demonstrating that patients may be better served by a timely transition from ruxolitinib to momelotinib that can help improve anemia while maintaining or improving splenic and symptom responses. Momelotinib is a potent and selective small-molecule inhibitor of JAK1, JAK2, and activin A receptor type 1 (ACVR1); the inhibition of JAK1 and JAK2 drives symptomatic and splenic benefits while the inhibition of ACVR1 promotes restoration of iron homeostasis and erythropoiesis, resulting in anemia benefits including increased hemoglobin (Hb) levels and reduced need for transfusions.12-16 Notably, transfusion-independence response with momelotinib has been associated with improved overall survival.6 Three phase III clinical studies of momelotinib in MF have provided extensive experience with momelotinib administered in more than 500 patients previously treated with ruxolitinib.12-14 In the SIMPLIFY-1 study, patients in the ruxolitinib-randomized group who crossed over to receive momelotinib at week 24 were immediately administered momelotinib without ruxolitinib tapering or washout.12 Here, we conducted a retrospective analysis to evaluate the clinical outcomes (i.e., dosing, spleen volume, frequency of transfusions, Hb levels, and occurrence of adverse events) of patients with MF who immediately transitioned from ruxolitinib to momelotinib in SIMPLIFY-1. In SIMPLIFY-1, JAKi-naïve intermediate- and high-risk pa-

tients with primary MF, post-essential thrombocythemia MF, or post-polycythemia vera MF (N=432) were randomized 1:1 to receive momelotinib at 200 mg once daily or ruxolitinib twice daily across four starting doses (5, 10, 15, and 20 mg twice daily) based on baseline platelet counts and other laboratory values. After the 24-week (6-month) randomized treatment period, patients in the momelotinib-randomized group could continue momelotinib (momelotinib→momelotinib), and patients in the ruxolitinib-randomized group could crossover to open-label momelotinib (ruxolitinib→momelotinib) immediately without tapering or washout.12 After the week 24 crossover into open-label treatment, clinical data including dosing, spleen volume, transfusions, and Hb levels, collected at weeks 4 and 8 after crossover and every 12 weeks thereafter, were analyzed to characterize the transition from ruxolitinib→momelotinib. Transfusion independence was defined as the absence of red blood cell (RBC) transfusion and no Hb level below 8 g/dL in the prior 12 weeks; transfusion dependence was defined as at least four units of RBC transfusions, or a Hb level below 8 g/dL in the previous eight weeks. In addition, safety assessments including recording of adverse events continued throughout open-label treatment. During randomized treatment, there was no significant difference in mean spleen volume reduction between the momelotinib and ruxolitinib arms (P=0.9853 at week 24), whereas mean Hb level increased with momelotinib and decreased with ruxolitinib (Figure 1A). After 24 weeks of randomized treatment, 197 patients transitioned from ruxolitinib→momelotinib and 171 continued momelotinib→momelotinib. At the first assessment four weeks after crossover from ruxolitinib→momelotinib, mean Hb levels improved rapidly (approx. 1 g/dL), and mean spleen volume was maintained (approx. 1700 cm3), similar to the mean spleen volume for momelotinib→momelotinib patients (Figure 1A). Patients continuing momelotinib treatment in the open-label phase maintained Hb levels that increased after two weeks of momelotinib treatment in the randomized phase. Mean platelet counts were generally maintained in patients randomized to momelotinib during both randomized and open-label treatment. For patients randomized to ruxolitinib, the mean platelet counts decreased by approximately 100x109/L during the first four weeks of treatment from a mean baseline platelet count of 301x109/L and remained at lower levels throughout the randomized phase; after crossover from ruxolitinib→momelotinib, mean platelet counts improved throughout open-label

Haematologica | 109 February 2024

676

LETTER TO THE EDITOR momelotinib treatment and converged with momelotinib→momelotinib by week 48 (Online Supplementary Figure S1). Of the patients in the ruxolitinib-randomized group, 70% were transfusion independent at baseline, which dropped to 49% at week 24.12 Of the 92 ruxolitinib-randomized patients who were not transfusion independent at week 24 who crossed over to receive momelotinib, 42 (46%) became transfusion independent by week 12 after crossover (Figure 1B). Among the 197 patients who completed 24 weeks of ruxolitinib treatment, 112 (57%) required a ruxolitinib dose modification (Figure 2A). Among patients who crossed over to receive open-label momelotinib from ruxolitinib

after randomized treatment, 90% (177/197) initiated momelotinib at the 200 mg daily dose (Figure 2B), with the majority of patients maintaining full-dose treatment at 200 mg momelotinib after 12 weeks (Figure 2C). Notably, of the 71 patients who received a mean of ≤10 mg twice daily ruxolitinib over the four weeks before crossover, only 10% achieved a spleen response (≥35% volume reduction from baseline) at week 24 (before crossover); following crossover, 23% achieved or maintained spleen response at week 48. Safety observations during the immediate 2-week period after ruxolitinib→momelotinib crossover revealed that the transition was well tolerated (Table 1); new onset grade 3/4 anemia and thrombocytopenia were experienced by

Figure 1. Clinical efficacy of momelotinib after immediate crossover from ruxolitinib in the SIMPLIFY-1 study. (A) Hemoglobin (Hb) and spleen volume dynamics in patients randomized to momelotinib→momelotinib or ruxolitinib→momelotinib. (B) Transfusion-independence rate after transition to open-label momelotinib at week 24 in non–transfusion-independent ruxolitinibrandomized patients (N=92). MMB: momelotinib; RUX: ruxolitinib; XO: crossover. Haematologica | 109 February 2024

677

LETTER TO THE EDITOR only 3% and 2% of patients, respectively, with no cases of ruxolitinib discontinuation syndrome, namely, no acute relapse of symptoms or splenomegaly, worsening of cytopenias, or hemodynamic decompensation, including acute respiratory distress syndrome and shock.8 More broadly, the new onset adverse events (by preferred term) of any

grade experienced within two weeks of ruxolitinib→momelotinib transition occurred at a rate of ≤7% each. Weight gain was higher with ruxolitinib than momelotinib during the randomized treatment period (weight change 0.9 ± 3.28 kg for momelotinib group vs. 3.3 ± 3.82 kg for ruxolitinib group [mean ± standard deviation]) but body weight remained

Figure 2. Dosing in ruxolitinib-randomized patients in the SIMPLIFY-1 study. (A) Dosing from baseline to week 24 of ruxolitinib treatment. (B) Dosing at crossover from ruxolitinib→momelotinib. (C) Dosing from baseline momelotinib at crossover to week 12 of open-label momelotinib treatment. MMB: momelotinib; OL: open label; RUX: ruxolitinib; XO: crossover.

Haematologica | 109 February 2024

678

LETTER TO THE EDITOR Table 1. Adverse events in the two weeks after crossover at week 24 in the SIMPLIFY-1 study.

Adverse events, N (%)

Ruxolitinib→Momelotinib N=197 Maximum grade Maximum grade Overall 1/2 3/4

Momelotinib→Momelotinib N=171 Maximum grade Maximum grade Overall 1/2 3/4

Overall

88 (44.7)

69 (35.0)

19 (9.6)

49 (28.7)

43 (25.1)

6 (3.5)

Nausea

14 (7.1)

13 (6.6)

1 (0.5)

3 (1.8)

Diarrhea

12 (6.1)

11 (5.6)

1 (0.5)

3 (1.8)

Fatigue

12 (6.1)

10 (5.1)

2 (1.0)

1 (0.6)

Dizziness

9 (4.6)

1 (0.6)

Headache

9 (4.6)

8 (4.1)

1 (0.5)

Pruritus

9 (4.6)

2 (1.2)

Anemia

8 (4.1)

2 (1.0)

6 (3.0)

4 (2.3)

1 (0.6)

3 (1.8)

Cough

8 (4.1)

Rash

6 (3.0)

1 (0.6)

Vitamin B1 deficiency

5 (2.5)

Back pain

4 (2.0)

Night sweats

4 (2.0)

2 (1.2)

Thrombocytopenia

4 (2.0)

4 (2.3)

stable and did not increase further after ruxolitinib→momelotinib crossover (Online Supplementary Figure S2). Momelotinib is a promising new therapy for MF. Data from the completed, randomized, phase III SIMPLIFY-1 study of momelotinib versus ruxolitinib provide a unique opportunity to evaluate transition to open-label momelotinib therapy in the extended treatment phase without tapering or washout of prior randomized treatment with ruxolitinib. Transition to momelotinib from ruxolitinib did not result in symptoms associated with ruxolitinib withdrawal, and control of spleen volume was maintained. Most patients tolerated full-dose momelotinib including those previously on low-dose ruxolitinib. In addition, transition to momelotinib was associated with rapid improvement in anemia and a shift toward transfusion independence. These data are consistent with those of SIMPLIFY-2, an international, randomized, openlabel, phase III study conducted to evaluate the efficacy and safety of momelotinib versus best available therapy (ruxolitinib accounting for 88.5% of best available therapy) in patients with intermediate- or high-risk primary MF, post-essential thrombocythemia MF, or post-polycythemia vera MF whose prior treatment with ruxolitinib was associated with anemia or thrombocytopenia.13 Washout was prohibited for patients receiving active MF therapy at screening; 72% of those randomized to momelotinib (75 of 104) continued ruxolitinib until the day of randomization. Similar to SIMPLIFY-1, spleen volume control was maintained with transition to momelotinib treatment (Online Supplementary Figure S3); transition

to momelotinib also provided symptom and anemia improvements in conjunction with an acceptable safety profile.13 These analyses provide confidence in an immediate transition to momelotinib from ruxolitinib without washout or tapering, which is likely to rapidly improve anemia without compromising safety or control of symptoms and spleen. The recently published Response to Ruxolitinib After 6 Months criteria modeled predictors of survival in patients with MF after six months of ruxolitinib.11 This multivariate model included negative risk factors of spleen length, ruxolitinib dose reduction, and RBC transfusion requirement; in this analysis, 45% were considered at intermediate risk and 36% at high risk of poor survival after six months of ruxolitinib therapy. These findings suggest that most patients with anemia on ruxolitinib therapy or those receiving low-dose ruxolitinib therapy should transition to a different therapy that can improve anemia and maintain recommended dose levels while also maintaining or improving on splenic and symptom responses.

Authors Ruben Mesa,1 Srdan Verstovsek,2 Uwe Platzbecker,3 Vikas Gupta,4 David Lavie,5 Pilar Giraldo,6 Christian Recher,7 Jean-Jacques Kiladjian,8 Stephen T. Oh,9 Aaron T. Gerds,10 Timothy Devos,11 Francesco Passamonti,12 Alessandro M. Vannucchi,13 Miklos Egyed,14

Haematologica | 109 February 2024

679

LETTER TO THE EDITOR Ewa Lech-Maranda,15 Andrzej Pluta,16 Lars Nilsson,17 Kazuya

PharmaEssentia, and Promedior. UP reports consulting fees from

Shimoda,18 Donal McLornan,19 Jun Kawashima,20 Barbara Klencke,20

AbbVie, Bristol Myers Squibb/Celgene, Janssen, and Novartis;

Mei Huang,

honoraria from Amgen, Jazz Pharmaceuticals, and Takeda; and

Bryan Strouse

and Claire Harrison

participation on data safety monitoring board or advisory board Atrium Health Wake Forest Baptist Comprehensive Cancer Center,

for AbbVie and Novartis. VG reports consulting fees from AbbVie,

Wake Forest University School of Medicine, Winston Salem, NC,

Bristol Myers Squibb/Celgene, Constellation Biopharma, Novartis,

USA; The University of Texas MD Anderson Cancer Center, Houston,

Pfizer, and Sierra Oncology; honoraria from Bristol Myers

TX, USA; Leipzig University Hospital, Leipzig, Germany; Princess

Squibb/Celgene, Constellation Biopharma, and Novartis; and

Margaret Cancer Centre, Toronto, Ontario, Canada; 5Hadassah-

participation on data safety monitoring board or advisory board

Hebrew University Medical Center, Jerusalem, Israel; 6Miguel Servet

for AbbVie, Bristol Myers Squibb/Celgene, Pfizer, and Roche. CR

University Hospital and Centro de Investigación Biomédica en Red

reports grants or contracts from AbbVie, Astellas, Bristol Myers

de Enfermedades Raras (CIBERER), Zaragoza, Spain; Institut

Squibb, Jazz Pharmaceuticals, and IQVIA; honoraria and travel

Universitaire du Cancer de Toulouse, Université de Toulouse III,

support from AbbVie, Astellas, Bristol Myers Squibb, Jazz

Toulouse, France; Universite Paris Cité, AP-HP, Hopital Saint-Louis,

Pharmaceuticals, Novartis, and Servier; and participation on a

Centre d’Investigations Cliniques, INSERM, CIC1427, Paris, France;

data safety monitoring board or advisory board for AbbVie,

Washington University School of Medicine, St. Louis, MO, USA;

Astellas, Bristol Myers Squibb, Jazz Pharmaceuticals, Novartis,

Cleveland Clinic Taussig Cancer Institute, Cleveland, OH, USA;

Servier, and Takeda. J-JK reports honoraria from Novartis, and

Department of Hematology, University Hospitals Leuven and

participation on a data safety monitoring board or advisory board

Department of Microbiology and Immunology, Laboratory of

for AbbVie, AOP Orphan, Bristol Myers Squibb, Incyte, and

Molecular Immunology (Rega Institute), KU Leuven, Leuven,

Novartis. STO reports consulting fees from AbbVie, Blueprint

Belgium; University of Insubria, Varese, Italy; University of

Medicines, Bristol Myers Squibb/Celgene, Constellation

Florence and AOU Careggi, Florence, Italy; Somogy County Mór

Pharmaceuticals, CTI BioPharma, Disc Medicine, Incyte, Kartos

Kaposi General Hospital, Kaposvár, Hungary; Institute of

Therapeutics, PharmaEssentia, and Sierra Oncology. ATG reports

Hematology and Transfusion Medicine, Warsaw, Poland;

consulting fees from AbbVie, Bristol Myers Squibb,

Department of Hematological Oncology, Oncology Specialist

Constellation/MorphoSys, CTI Biopharma, Novartis,

Hospital, Brzozow, Poland; 17Department of Hematology, Oncology

PharmaEssentia, and Sierra Oncology. TD reports consulting fees

and Radiation Physics, Skåne University Hospital, Lund, Sweden;

from AOP Health, Bristol Myers Squibb/Celgene, Incyte, and

University of Miyazaki, Miyazaki, Japan; Guy's and St Thomas'

MorphoSys, and honoraria from Novartis and Sobi. FP reports

NHS Foundation Trust, London, UK and

Sierra Oncology Inc., San

grants or contracts from Bristol Myers Squibb; consulting fees

Mateo, CA, USA

from AbbVie, AOP, Bristol Myers Squibb/Celgene, Janssen, Karyopharm Therapeutics, Kyowa Kirin, MEI Pharma, Novartis,

Correspondence:

Roche, and Sierra Oncology; and honoraria from AbbVie, Bristol

R. MESA - rmesa@wakehealth.edu

Myers Squibb/Celgene, Janssen, Novartis, and Sierra Oncology. AMV reports honoraria from AbbVie, Blueprint Medicines, Bristol

https://doi.org/10.3324/haematol.2023.283106

Myers Squibb, GSK, Incyte, and Novartis, and participation on a data safety monitoring board or advisory board for AbbVie,

Received: March 9, 2023.

Blueprint Medicines, Bristol Myers Squibb, GSK, Incyte,

Accepted: May 24, 2023.

MorphoSys, Novartis, and Roche. AP reports honoraria from

Early view: June 1, 2023.

Kedrion Biopharma. KS reports honoraria from Novartis and Takeda. DM reports grants or contracts from CPI, and honoraria

from AbbVie, Bristol Myers Squibb/Celgene, Jazz Pharmaceuticals,

Published under a CC BY-NC license

and Novartis. JK reports employment at Sierra Oncology, and stock or stock options at Gilead Sciences and Sierra Oncology. BK

Disclosures

and MH report employment and stock options at Sierra Oncology.

RM reports grants or contracts from AbbVie, Celgene, CTI

BS reports employment at Sierra Oncology. CH reports grants or

Biopharma, Constellation Biopharma, Genotech, Incyte,

contracts from Bristol Myers Squibb/Celgene, Constellation

Promedior, Samus Therapeutics, and the Mays Cancer Center P30

Pharmaceuticals, and Novartis; consulting fees from AOP, Galecto,

Cancer Center Support Grant from the National Cancer Institute

Keros, and Roche; honoraria from AbbVie, Celgene, Constellation

(CA054174), and consulting fees from Constellation Biopharma, La

Pharmaceuticals, CTI BioPharma, Janssen, and Novartis;

Jolla, Novartis, and Sierra Oncology. SV reports consulting fees

participation on data safety monitoring board or advisory board

from Bristol Myers Squibb/Celgene, Incyte, Novartis, and Sierra

for AbbVie, AOP, CTI BioPharma, Geron, Promedior, Roche, and

Oncology, and research funding from AstraZeneca, Blueprint

Sierra Oncology; and leadership or fiduciary role in the European

Medicines, Bristol Myers Squibb/Celgene, CTI BioPharma,

Hematology Association and MPN Voice. DL, PG, ME, EL-M and LN

Genentech, Gilead, Incyte, Italfarmaco, Novartis, NS Pharma,

have no conflicts of interest to disclose.

Haematologica | 109 February 2024

680

LETTER TO THE EDITOR Contributions

law or regulation. Qualified scientific and medical researchers can

JK, BK, MH and BS contributed to the study design. RM, SV, UP, VG,

request patient-level data for studies of Sierra pharmaceutical

DL, PG, CR, J-JK, STO, ATG, TD, FP, AMV, ME, EL-M, AP, LN, DM, MH

substances listed on ClinicalTrials.gov and approved by health

and CH contributed to data acquisition. JK, BK, MH and BS

authorities in the USA and the EU. Patient-level data for studies of

conducted the data analysis. MH performed the statistical analysis.

newly approved pharmaceutical substances or indications can be

All authors contributed to data interpretation, reviewed and

requested 9 months after US Food and Drug Administration and

provided important intellectual contributions to the manuscript,

European Medicines Agency approvals. Such requests are assessed

and approved the final version for publication.

at Sierra’s discretion, and the decisions depend on the scientific merit of the proposed request, data availability, and the purpose of

Acknowledgments

the proposal. If Sierra agrees to share clinical data for research

We thank the patients and families who participated in the trial and

purposes, the applicant is required to sign an agreement for data

all study investigators. Medical writing and editorial support were

sharing before data release, to ensure that the patient data are

provided, based on the authors’ input and in accordance with

deidentified. In case of any risk of reidentification on anonymized

ICMJE and GPP3 guidelines, by Yaeko Hiyama, PhD, of Second City

data despite measures to protect patient confidentiality, the data

Science, who was supported by Sierra Oncology, a GSK company.

will not be shared. The patients’ informed consent will always be respected. If the anonymization process will provides futile data,

Funding

Sierra will have the right to refuse the request. Sierra will provide

This study was sponsored by Sierra Oncology, a GSK company.

access to patient-level clinical trial analysis datasets in a secured environment upon execution of the data-sharing agreement. Sierra

Data-sharing statement

will also provide the protocol, statistical analysis plan, and the

Sierra Oncology commits to sharing clinical study data with qualified

clinical study report synopsis if needed. For additional information or

researchers to enable enhancement of public health. As such, Sierra

requests for access to Sierra clinical trial data for research

will share anonymized patient-level data on request or if required by

purposes, please contact us at GSKClinicalSupportHD@gsk.com.

References 1. Verstovsek S, Mesa RA, Gotlib J, et al. A double-blind, placebocontrolled trial of ruxolitinib for myelofibrosis. N Engl J Med. 2012;366(9):799-807. 2. Pardanani A, Harrison C, Cortes JE, et al. Safety and efficacy of fedratinib in patients with primary or secondary myelofibrosis: a randomized clinical trial. JAMA Oncol. 2015;1(5):643-651. 3. Mascarenhas J, Hoffman R, Talpaz M, et al. Pacritinib vs best available therapy, including ruxolitinib, in patients with myelofibrosis: a randomized clinical trial. JAMA Oncol. 2018;4(5):652-659. 4. Gupta V, Harrison C, Hexner EO, et al. The impact of anemia on overall survival in patients with myelofibrosis treated with ruxolitinib in the COMFORT studies. Haematologica. 2016;101(12):e482-e484. 5. Harrison CN, Schaap N, Mesa RA. Management of myelofibrosis after ruxolitinib failure. Ann Hematol. 2020;99(6):1177-1191. 6. Mesa R, Harrison C, Oh ST, et al. Overall survival in the SIMPLIFY-1 and SIMPLIFY-2 phase 3 trials of momelotinib in patients with myelofibrosis. Leukemia. 2022;36(9):2261-2268. 7. Palandri F, Breccia M, Bonifacio M, et al. Life after ruxolitinib: reasons for discontinuation, impact of disease phase, and outcomes in 218 patients with myelofibrosis. Cancer. 2020;126(6):1243-1252. 8. Tefferi A, Pardanani A. Serious adverse events during ruxolitinib treatment discontinuation in patients with myelofibrosis. Mayo Clin Proc. 2011;86(12):1188-1191. 9. Palandri F, Palumbo GA, Elli EM, et al. Ruxolitinib discontinuation syndrome: incidence, risk factors, and management in 251

patients with myelofibrosis. Blood Cancer J. 2021;11(1):4. 10. Passamonti F, Heidel FH, Parikh RC, et al. Real-world clinical outcomes of patients with myelofibrosis treated with ruxolitinib: a medical record review. Future Oncol. 2022;18(18):2217-2231. 11. Maffioli M, Mora B, Ball S, et al. A prognostic model to predict survival after 6 months of ruxolitinib in patients with myelofibrosis. Blood Adv. 2022;6(6):1855-1864. 12. Mesa RA, Kiladjian JJ, Catalano JV, et al. SIMPLIFY-1: a phase III randomized trial of momelotinib versus ruxolitinib in Janus kinase inhibitor-naive patients with myelofibrosis. J Clin Oncol. 2017;35(34):3844-3850. 13. Harrison CN, Vannucchi AM, Platzbecker U, et al. Momelotinib versus best available therapy in patients with myelofibrosis previously treated with ruxolitinib (SIMPLIFY 2): a randomised, open-label, phase 3 trial. Lancet Haematol. 2018;5(2):e73-e81. 14. Verstovsek S, Gerds AT, Vannucchi AM, et al. Momelotinib versus danazol in symptomatic patients with anaemia and myelofibrosis (MOMENTUM): results from an international, double-blind, randomised, controlled, phase 3 study. Lancet. 2023;401(10373):269-280. 15. Mesa R, Oh ST, Gerds AT, et al. Momelotinib reduces transfusion requirements in patients with myelofibrosis. Leuk Lymphoma. 2022;63(7):1718-1722. 16. Oh ST, Talpaz M, Gerds AT, et al. ACVR1/JAK1/JAK2 inhibitor momelotinib reverses transfusion dependency and suppresses hepcidin in myelofibrosis phase 2 trial. Blood Adv. 2020;4(18):4282-4291.

Haematologica | 109 February 2024

681

CASE REPORT

Immune-mediated facial nerve paralysis in a myeloma patient post B-cell maturation antigen-targeted chimeric antigen receptor T cells Chimeric antigen receptor (CAR) T-cell therapies targeting B-cell maturation antigen (BCMA) have changed the standard of care for multiple myeloma (MM).1 Recently, the Food and Drug Administration approved idecabtagene vicleucel (ide-cel) and ciltacabtagene autoleucel (cilta-cel) for MM,2,3 however, further research is needed to fully understand the long-term safety and efficacy of these treatments. Neurotoxicity including its most common form - immune effector cell-associated neurotoxicity syndrome (ICANS) - is a potential side effect of CAR T-cell therapies. ICANS can range from confusion or headaches to seizures, coma, or death.4 The exact cause of CAR T-mediated neurotoxicity is not fully understood, but it is thought to be related to cytokine release by CAR T in and outside of the central nervous system (CNS). In addition to ICANS, other CAR T-related neurotoxicities, such as movement disorders, cognitive impairment, and personality changes, have been described.4 However, the pathophysiology of these is even less well understood. Here, we describe an MM patient, without prior history of CNS involvement by her myeloma, who developed bilateral facial nerve palsy (facial diplegia) following cilta-cel. The focal neurologic deficiency correlated with a marked expansion of BCMA-targeted CAR T in the peripheral blood (PB) and recruitment of central memory-type CAR T cells into the CNS. Our study suggests mechanisms potentially resulting in CAR T CNS infiltration with neurotoxicity and ways to prevent/treat these off-tumor effects. Our patient was diagnosed with immunoglobulin (Ig)G κ MM approximately 4.5 years prior to receiving CAR T. She received multiple prior treatments (induction with bortezomib/lenalidomide/dexamethasone followed by carfilzomib/lenalidomide/ dexamethasone, high-dose melphalan/autologous stem cell transplant, a clinical study with an MM-dendritic cell (DC) fusion vaccine + lenalidomide as maintenance, carfilzomib/ daratumumab/dexamethasone). After her most recent line of treatment, she was found to have another relapse of her myeloma with 15% plasma cells in the BM with 1q amplification and TP53 deletion. We performed leukapheresis and decided to use daratumumab/pomalidomide/dexamethasone for post-apheresis bridging in an effort to control the myeloma during manufacturing, avoid significant myelosuppression and to allow the patient to minimize the frequency of office visits prior to admission for CAR T-cell therapy. Leukapheresis and lymphodepleting chemotherapy with cyclophosphamide/ fludarabine were performed and cilta-cel CAR T were given in November of 2022 (Figure 1A-C).

Following CAR T, she developed grade 1 cytokine release syndrome (CRS) and Enterobacter cloacae urinary tract infection. As part of our broad CAR T-related research efforts, serum concentrations of different cytokines/chemokines were measured and we found the development of CRS to coincide with peak levels of interferon (IFN)γ, interleukin (IL)10, and monocyte chemoattractant protein-1 (MCP-1) followed by increases in C-reactive protein and ferritin (Figure 1A, E). The CRS was treated by giving one dose of tocilizumab on day +6 and two doses on day +7. The patient also received dexamethasone 10 mg twice daily on days +7 to +8 for ongoing fevers and antibiotics for her infection. Subsequently, the patient showed recovery of her absolute lymphocyte counts (Figure 1B) paralleled by a marked expansion of BCMA-targeted CD4+ and CD8+ CAR T cells in her PB (Figure 1C, D). Shortly before her CAR T cells reached peak levels, the patient showed a substantial increase in granzyme B, IL13, and MIP-1α serum concentrations (Figure 1C, E). Only 2 days later (day +17), she started to complain of sudden difficulties speaking, chewing and puckering her lips. We consulted with our neurologists and on exam, she was noted to have bilateral cranial nerve VII palsy being unable to smile, puff her cheeks, frown, and form words due to facial weakness. The remainder of her neurological exam was normal; computed tomography and magnetic resonance imaging of the brain showed no pathology. Cerebrospinal fluid (CSF) collected by our neurologists on day +18 was unremarkable for infection or acute inflammation but evidenced lymphocytes consisting of large amounts of BCMA-targeted CAR T infiltrating the neuroaxis (Figure 2A). On the same day, PB levels of BCMA CAR T peaked (Figure 1C, D). Both PB (Figure 1F) and CNS (Figure 2A) CAR T consisted primarily of central memory-type T cells. PB CAR T were composed of equal proportions of CD4+ and CD8+ T cells (Figure 1C) while CNS-infiltrating CD4+ T cells by far outnumbered CD8+ T cells (Figure 2A), indicating a specific recruitment of T-helper cells. On the same day, the patient showed very high concentrations of IP-10 in the CSF (Figure 2C) which markedly exceeded IP-10 concentrations in the patient’s PB (Figure 1E). Chemokine receptor CXCR3 is a ligand for IP-10 and the patient’s PB CAR T showed surface expression of CXCR3 similar to their own non-CAR T (Figure 2D). However, expression levels of CCR6, which is involved in recruiting activated T cells to the brain,5 were higher on the patient’s PB CAR T cells compared to their non-CAR T (Figure 2D). Within the same

Haematologica | 109 February 2024

682

CASE REPORT

CAR T cells, expression of effector molecule granzyme B was more pronounced than within non-CAR T (Figure 2D). Finally, the patient’s CAR T were uniformly positive

for α4β1 integrin (Figure 2B), a receptor supporting T-cell migration across the blood-brain barrier (BBB).5,6 As per recommendation by Neurology, the patient was

Continued on following page.

Haematologica | 109 February 2024

683

CASE REPORT

Figure 1. CAR T-cell expansion and persistence in a myeloma patient with immune-mediated facial nerve paralysis. Time course of (A) serum ferritin and C-reactive protein (CRP) levels, (B) reconstitution of white blood cell, neutrophil, and lymphocyte counts, and (C) chimeric antigen receptor (CAR) T-cell numbers and proportions of CD4+ versus CD8+ CAR T cells after lymphodepleting chemotherapy and CAR T-cell infusion. Time points with peak serum cytokine/chemokine levels are indicated by green or yellow dots, respectively. Occurrence of facial nerve paralysis is indicated by a dotted line. (D) Dot plots showing peripheral blood CAR T cells at different time points post CAR T infusion. CAR T cells were identified by staining of the expression of the CAR on the cell surface and co-staining with anti-CD3 and other T-cell markers. (E) Serum concentrations of 22 different T-cell-related cytokines/chemokines were determined in our patient at different time points post CAR T treatment using CodePlex Secretome technology. Cytokine concentrations were quantified using the CodePlex Secretome Human Adaptive Immune Panel kit (IsoPlexis; #CODEPLEX-2L01). Analysis was performed using the Isolight instrument (Isoplexis, Branfold, CT). Automated analysis of raw data was performed using IsoSpeak software (Isoplexis, Branfold, CT). Results are shown as fold change from baseline at pre-lymphodepleting chemotherapy. (F) CAR T-cell memory subtypes were determined on day +18 post CAR T treatment by co-staining for CD45RO and CD62L. Central memory CAR T cells are shown in the right upper quadrant. Samples were collected under Institutional Review Board-approved protocol 2043GCCC (IRB HP-00091736). Plasma was generated by centrifugation at 400 G and frozen immediately at -80°C. Peripheral blood mononuclear cells were isolated using density gradient centrifugation and frozen in liquid nitrogen. Staining for flow cytometry was performed using monoclonal antibodies following manufacturer’s instructions. Samples were acquired using a Miltenyi MACSQuant Analyzer 10 Flow Cytometer. Analysis of flow cytometry data was performed using FlowJo software (BD Biosciences, San Jose, CA). D: day; PB: peripheral blood.

initially started on dexamethasone 10 mg every 12 hours but was transitioned to solumedrol 1 g daily for 3 days due to worsening speech. As a result, her facial movements and speech returned to baseline. Dexamethasone was tapered slowly over the next week. By discharge, her exam had improved as compared to admission, but her symptoms had not resolved completely. Subsequently, she had a temporary re-emergence of her symptoms, requiring restarting steroids. There were no obvious correlations with serum markers for CNS damage over time, however, there seemed to be a transient post-CAR T increase in NSE and GFAP. At 1 month post CAR T, the patient’s neurologic symptoms had significantly improved. Importantly, levels of PB CAR

T cells had persisted over time (Figure 1C, D) and her BM evidenced substantial infiltration by CD4+ and CD8+ BCMA-targeted central memory CAR T (Figure 3C). Apart from some expression of TIM3, the BM-residing CAR T did not express any co-inhibitory molecules (Figure 3D), however, they were CD27-positive7 and CD127-positive8 (Figure 3E), indicating full functionality. Importantly, at that point in time all myeloma cells had been eradicated from the patient’s BM (Figure 3A, B) and serum free light chains had normalized. At her most recent visit, approximately 6 months after CAR T, her neurologic symptoms had completely resolved without any sequalae. Here, we describe the case of a MM patient who developed bilateral facial diplegia following cilta-cel infusion. In the

Haematologica | 109 February 2024

684

CASE REPORT

CARTITUDE-1 clinical trial using cilta-cel, neurotoxicity occurred in 21% of patients and one patient had facial nerve paralysis.2 In the phase II KarMMa trial using idecabtagene vicleucel, neurotoxicity was reported in 18%.3 Recently, a case of progressive movement disorder with features of parkinsonism was described after cilta-cel, associated with CAR T-cell persistence in the blood and CSF. BCMA

was found to be present on neurons and astrocytes in the patient’s basal ganglia, suggesting an on-target effect.9 Our patient responded to treatment with high-dose steroids suppressing CAR T activity, however, whether the CNS toxicity was based on immediate CAR T on-target/off-tumor cytotoxicity remains to be evaluated. CAR T, especially CD4+ CAR T, have previously been shown

Figure 2. Central nervous system infiltration by BCMA-targeted CAR T cells in a myeloma patient with post-treatment facial nerve paralysis. (A) Proportions of CD4+ and CD8+ chimeric antigen receptor (CAR) T cells were determined in the cerebrospinal fluid (CSF) of our patient at onset of potentially immune-mediated central nervous system (CNS) toxicity using flow cytometry. Dot plots show CNS-infiltrating CAR T cells at day +18. CAR T cells were identified by staining of the expression of the CAR on the cell surface and co-staining with anti-CD3 and other T-cell markers. CAR T cell memory subtypes were determined on day +18 post CAR-T treatment by co-staining for CD45RO and CD62L. Central memory (CM) CAR T-cell are shown in the right upper quadrant. (B) Proportions of peripheral blood CAR T cells expressing α4β1 integrin required for the entry of T cells into the CNS. (C) CSF concentrations of 22 different T-cell-related cytokines/chemokines were determined in our patient onset of potentially immune-mediated CNS toxicity using CodePlex Secretome technology. Results are shown as absolute concentrations in pg/mL. (D) Surface expression of receptors involved in CNS-directed homing of T cells on peripheral blood (PB) CAR T cells (red histograms) and non-CAR T (gray histograms) from our patient. In addition, cytoplasmic granzyme B was stained on day +14 post CAR T-cell treatment in both PB CAR T cells (red histogram) and non-CAR T (gray histogram) from the same patient/time point. Haematologica | 109 February 2024

685

CASE REPORT

capable of infiltrating the CNS, e.g., in the case of CD19-targeted CAR T in patients with CNS lymphoma.10 In this context, we consider it possible that increased numbers of CD4+ CAR T could be due to a survival advantage of these cells over CD8+ CAR T and that these cells could represented regulatory-type T cells.11 We show here that BCMA-targeted CAR T are able to cross the BBB even without CNS malignancies. Our data indicate that IP-10 produced by CNS-residing cells such as astrocytes12 could play a role in the CNS recruitment of activated CAR T, either through CXCR3 or an alternative ligand. CXCR3 has extensively been studied with regard to T-cell

recruitment during neuroinflammation. It is abundantly expressed on CNS-infiltrating T lymphocytes in multiple sclerosis patients13 and co-ordinates migration in response to its three ligands, CXCL9/CXCL10/CXCL11.14 Our data suggest that inhibition of IP-10 (CXCL10) could potentially represent a way to prevent CAR T neurotoxicity. Our patient’s CAR T showed high levels of CCR6 and the vasculature of the choroid plexus expresses adhesion molecules and chemokines including CCL20, the only known ligand for CCR6. CCL20/CCR6 interactions influence immune cell adhesion, rolling, and extravasation across the endothelium and pia mater. As a result, CCR6+ leukocytes

C Figure 3. Infiltration by BCMA CAR T cells and eradication of myeloma cells and in the bone marrow of a patient with immune-mediated facial nerve palsy. (A) Sections from pre-chimeric antigen receptor (CAR) T-cell therapy showed a hypocellular bone marrow (BM) with trilineage hematopoiesis (hematoxilin & eosin [H&E], 400x magnification) and small clusters of plasma cells (CD138, 400x magnification) accounting for 15% of BM cellularity overall. (B) Post CAR T-cell therapy, the BM demonstrated a mild decrease in cellularity with maturing trilineage hematopoiesis (H&E, 400x magnification) and a near total absence of plasma cells (CD138, 400x magnification). (C) At the same time point, the BM showed infiltration by central memory-type CAR T cells. (D) BM-infiltrating CAR T cells did not show any significant surface expression of exhaustion markers (red histogram) by flow cytometry when compared to unstained controls (gray histogram). (E) In contrast, BM-infiltrating CAR T cells showed high levels of CD27 and CD127. Histologic sections from formalin-fixed, paraffin-embedded tissue samples underwent immunohistochemical and in situ hybridization staining using standard techniques.

Haematologica | 109 February 2024

686

CASE REPORT

enter CSF-containing ventricles and circulate through the CNS, surveying for antigen and signs of inflammation.15 Accordingly, neutralizing CCL20 or CCR6 in mice with neuroinflammation decreased disease severity, highlighting the role of the CCL20/CCR6 axis in CNS-damaging autoimmune processes.16 Interactions between α4β1 integrin on effector T cells and its ligand VCAM-1 on BBB endothelial cells is a requirement for the entry of T cells into the CNS and neutralization of α4 integrin inhibits neuroinflammation and prevents T-cell recruitment into the CNS parenchyma.17 Our findings suggest that inhibiting the function of α4β1, IP-10 and/or CCL20/CCR6 could potentially help to avoid CNS toxicity by CAR T. Future studies will further delineate the most relevant pathophysiologic mechanisms behind CAR T-related neurotoxicity and develop targeted methods to prevent and/or treat these immune-mediated side effects.

Transplant Program, Stanford University, Stanford, CA, USA Correspondence: D. ATANACKOVIC - datanackovic@som.umaryland.edu https://doi.org/10.3324/haematol.2023.283296 Received: April 6, 2023. Accepted: August 30, 2023. Early view: September 7, 2023. ©2024 Ferrata Storti Foundation Published under a CC BY-NC license Disclosures SD serves on advisory boards for Bristol-Myers Squibb, Incyte, and Atara Biotherapeutics. NMH serves on advisory boards for InCyte and Kite-Gilead; and is a member of the DSMB for American Gene Technologies. The remaining authors have no conflicts of interest to disclose.

Authors

Contributions DA designed the study, performed experiments, analyzed the data,

Yamini K. Kathari, Haroon Ahmad, Michael E. Kallen, Rima Koka,

made figures, and wrote the manuscript. PL and EP collected and

Destiny Omili, Thierry Iraguha, Jean Clement, Lily Pham, Mazhar

processed patient samples. TI, DO, EG, and ND processed patient

Khalid, Xiaoxuan Fan, Etse Gebru, Patricia Lesho, Esther Park,

samples and performed experiments. YKH, HA, RK, JC, LP, JMB, KAD,

Nishanthini Dishanthan, Jillian M. Baker, Kenneth A. Dietze, Kim G.

KGH, AB, JAY, SD, NMH, HK, and APR analyzed data and wrote the

Hankey, Ashraf Badros,

manuscript. MEK, MK, and XF performed experiments, analyzed the

1,2

1,3

1,5

1,3

1,2

1,6

Nancy M. Hardy, Rapoport

1,2,5

1,3

1,5

1,2

Jean A. Yared,

1,2,5

Hakan Kocoglu,

1,2,5

Saurabh Dahiya,

1,2,5,7

Tim Luetkens,

1,5,6

data, and wrote the manuscript. TL analyzed the data, prepared

Aaron P.

figures, and wrote the manuscript.

and Djordje Atanackovic

1,2,5,6

University of Maryland Greenebaum Comprehensive Cancer Center,

Funding

Baltimore, MD; Department of Medicine, University of Maryland

This study was funded by two grants from the Kahlert Foundation

School of Medicine, Baltimore, MD; Department of Neurology,

(to DA) and by the Maryland Department of Health’s Cigarette

University of Maryland School of Medicine, Baltimore, MD;

Restitution Fund Program (to DA and XF) and by the National Cancer

Department of Pathology, University of Maryland School of

Institute - Cancer Center support grant P30CA134274.

Medicine, Baltimore, MD; Transplant and Cellular Therapy Program, 5

University of Maryland Greenebaum Comprehensive Cancer Center,

Data-sharing statement

Baltimore, MD; Department of Microbiology and Immunology,

Original data and protocols will be made available by the authors to

University of Maryland, Baltimore, MD and Blood and Marrow

other investigators upon request.

References 1. Carpenter RO, Evbuomwan MO, Pittaluga S, et al. B-cell maturation antigen is a promising target for adoptive T-cell therapy of multiple myeloma. Clin Cancer Res. 2013;19(8):2048-2060. 2. Berdeja JG, Madduri D, Usmani SZ, et al. Ciltacabtagene autoleucel, a B-cell maturation antigen-directed chimeric antigen receptor T-cell therapy in patients with relapsed or refractory multiple myeloma (CARTITUDE-1): a phase 1b/2 open-label study. Lancet. 2021;398(10297):314-324. 3. Munshi NC, Anderson LD, Jr., Shah N, et al. Idecabtagene vicleucel in relapsed and refractory multiple myeloma. N Engl J Med. 2021;384(8):705-716. 4. Cohen AD, Parekh S, Santomasso BD, et al. Incidence and management of CAR-T neurotoxicity in patients with multiple myeloma treated with ciltacabtagene autoleucel in CARTITUDE

studies. Blood Cancer J. 2022;12(2):32. 5. Heng AHS, Han CW, Abbott C, McColl SR, Comerford I. Chemokine-driven migration of pro-inflammatory CD4(+) T cells in CNS autoimmune disease. Front Immunol. 2022;13:817473. 6. Glatigny S, Duhen R, Oukka M, Bettelli E. Cutting edge: loss of alpha4 integrin expression differentially affects the homing of Th1 and Th17 cells. J Immunol. 2011;187(12):6176-6179. 7. Fraietta JA, Lacey SF, Orlando EJ, et al. Determinants of response and resistance to CD19 chimeric antigen receptor (CAR) T cell therapy of chronic lymphocytic leukemia. Nat Med. 2018;24(5):563-571. 8. Liu W, Putnam AL, Xu-Yu Z, et al. CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ T reg cells. J Exp Med. 2006;203(7):1701-1711.

Haematologica | 109 February 2024

687

CASE REPORT 9. Van Oekelen O, Aleman A, Upadhyaya B, et al. Neurocognitive and hypokinetic movement disorder with features of parkinsonism after BCMA-targeting CAR-T cell therapy. Nat Med. 2021;27(12):2099-2103. 10. Li T, Zhao L, Zhang Y, et al. CAR T-cell therapy is effective but not long-lasting in B-cell lymphoma of the brain. Front Oncol. 2020;10:1306. 11. Upreti D, Bakhshinyan D, Bloemberg D, et al. Strategies to enhance the efficacy of T-cell therapy for central nervous system tumors. Front Immunol. 2020;11:599253. 12. Wang K, Wang H, Lou W, et al. IP-10 promotes blood-brain barrier damage by inducing tumor necrosis factor alpha production in Japanese encephalitis. Front Immunol. 2018;9:1148. 13. Mahad DJ, Howell SJ, Woodroofe MN. Expression of chemokines in the CSF and correlation with clinical disease activity in

patients with multiple sclerosis. J Neurol Neurosurg Psychiatry. 2002;72(4):498-502. 14. Loetscher M, Gerber B, Loetscher P, et al. Chemokine receptor specific for IP10 and mig: structure, function, and expression in activated T-lymphocytes. J Exp Med. 1996;184(3):963-969. 15. Williams JL, Holman DW, Klein RS. Chemokines in the balance: maintenance of homeostasis and protection at CNS barriers. Front Cell Neurosci. 2014;8:154. 16. Liston A, Kohler RE, Townley S, et al. Inhibition of CCR6 function reduces the severity of experimental autoimmune encephalomyelitis via effects on the priming phase of the immune response. J Immunol. 2009;182(5):3121-3130. 17. Yednock TA, Cannon C, Fritz LC, et al. Prevention of experimental autoimmune encephalomyelitis by antibodies against alpha 4 beta 1 integrin. Nature. 1992;356(6364):63-66.

Haematologica | 109 February 2024

688

CASE REPORT

Daratumumab and brentuximab vedotin combination therapy in T-cell acute lymphoblastic leukemia refractory to conventional chemotherapy and allogeneic stem cell transplant T-cell acute lymphoblastic leukemia/lymphoma (T-ALL) is an aggressive bone marrow neoplasm that accounts for 20% of acute lymphoblastic leukemias cases in adults.1 Despite high rates of complete remission with first-line combination chemotherapy, with approximately 50% survival at 5 years, the outcomes for patients with relapsed disease are dismal.2 There are limited salvage treatment options for patients with relapsed/refractory (R/R) disease.3,4 This highlights the need for effective therapeutic options with novel mechanisms of action. Daratumumab is a CD38-targeted IgG1κ human monoclonal antibody that is used in the treatment of newly diagnosed and R/R multiple myeloma.5 CD38 has been shown to be uniformly expressed on T-ALL blasts with persistent expression after treatment with chemotherapy.6 Preclinical studies have demonstrated efficacy of daratumumab using patient-derived xenograft models of T-ALL.6,7 There have been a few reports of daratumumab use in patients with R/R T-ALL and in patients who achieved a complete remission but were measurable residual disease (MRD)-positive. Brentuximab vedotin is an antibody drug conjugate composed of a CD30-targeting chimeric IgG antibody and the microtubule inhibitor monomethyl auristatin E. It is currently approved for use in the first line setting in combination with chemotherapy in classical Hodgkin lymphoma8 and CD30-positive peripheral T-cell lymphoma,9 and as monotherapy in the R/R setting.10 One study demonstrated CD30 expression in 13 of 34 (38%) of cases of T-ALL by multicolor flow cytometry using a 20% cutoff for positivity; and upregulation of CD30 expression during the course of high-dose chemotherapy in some cases, suggesting that CD30 may be a potential therapeutic target for T-ALL.11 However, there have not been prior clinical reports of brentuximab use in ALL. Here we report the use of daratumumab and brentuximab vedotin combination in a patient with heavily pretreated R/R T-ALL. A 45-year-old female was diagnosed with T-cell acute lymphoblastic leukemia (T-ALL) after presenting with an enlarging anterior chest wall mass (Figure 1A) and leukocytosis (white blood cell count: 110.4x109/L, 84% blasts) with bicytopenia. Bone marrow immunophenotyping showed blasts expressing CD3, CD45(dim), CD7, CD38, nTDT, cCD3, TCR-γ/δ(dim), CD5(dim), and CD58. Fluorescence in situ hybridization testing on the bone marrow showed a TRB rearrangement at 7q34 that was not fused with the common translocation partners (MYB, TLX1, LMO1 or LMO2) and

trisomy of chromosome 9. Next-generation sequencing (FoundationOne heme) showed JAK3 M511I, MYCN R357C, NOTCH1 L1600P, and PHF6 Y301fs*1 mutations. She was started on treatment with hyperfractionated cyclophosphamide, vincristine, doxorubicin, and dexamethasone (Hyper-CVAD) regimen and achieved a complete response after 1 cycle with MRD estimated at 0.7% by flow cytometry at a sensitivity of 0.1%. Positron emission tomography/ computed tomography (PET/CT) showed resolution of the hypermetabolic soft tissue mass and improvement in splenomegaly with normal 18F-fluorodeoxglucose (FDG) activity (Figure 1B). After completing five cycles, she remained MRD-positive at 0.05%. She subsequently received conditioning chemotherapy with cyclophosphamide and total body irradiation 1,200 cGy followed by allogeneic bone marrow transplantation with 11/12 HLA-matched (DP-permissive) unrelated donor. She experienced disease relapse, 3 months following transplant, and was enrolled on the clinical trial EA9152 of liposomal vincristine 2.25 mg/m2 (days 1, 8, 15, and 22) and venetoclax (600 mg days 1-28). She experienced severe autonomic and sensory neuropathy after the first dose of liposomal vincristine leading to its discontinuation. She had persistent disease after completing one cycle of venetoclax, proceeded with salvage treatment with venetoclax combined with fludarabine, cytarabine, granulocyte colony-stimulating factor, and idarubicin (FLAG-IDA). Bone marrow biopsy at day 21 showed persistent T-ALL with >95% bone marrow blasts. A repeat FoundationOne heme revealed similar mutational profile with subclonal JAK1 R724C and STAT5B N642H mutations. She was subsequently started on nelarabine (1,000 mg/m2) (days 1, 3 and 5, reduced dose due to neuropathy) and achieved a MRD-negative state after two cycles. This was followed by consolidation with a donor lymphocyte infusion (DLI) at a dose of 1x107 CD3 cells/kg from the original donor. However, approximately 1 month following DLI, she experienced another relapse with 72% lymphoblasts and immunophenotype showed blasts expressing CD10, CD45(dim), CD5, CD7, CD38, nTdT, cCD3 but did not express CD3, TCR-γ/δ. Furthermore, immunohistochemistry also revealed blasts expressing diffuse CD38 (>80% of blasts) and strong CD30 staining (Figure 3A-D). There were new FDG-avid retroperitoneal and axillary lymph nodes and enlargement of the spleen with increased uptake on PET/ CT (Figure 2A). Based on the CD38 and CD30 expression,

Haematologica | 109 February 2024

689

CASE REPORT

she was started on treatment with weekly daratumumab 16 mg/kg (multiple myeloma schedule) and brentuximab vedotin 1.8 mg/kg every 21 days (Hodgkin lymphoma schedule). After one cycle, the bone marrow was markedly hypocellular with no morphologic features of involvement by T-ALL. MRD by flow cytometry was negative (a representative marrow after cycle 3 is shown in Figure 3E-H). PET/CT showed interval resolution of lymphadenopathy and decrease in the size of the spleen with normal uptake (Figure 2B). Due to persistent pancytopenia, cycle 2 was delayed by 1 month. The dose of brentuximab vedotin was reduced to 1.4 mg/kg starting with the third cycle due to grade 2 sensory peripheral neuropathy. The fourth cycle was complicated by grade 4 neutropenia with bacteremia, grade 3 diarrhea, and grade 3 sensory peripheral neuropathy. Treatment-emergent toxicities were likely a result of prior exposure to multiple chemotherapies, and presence of pre-existing peripheral neuropathy. Therefore, brentuximab vedotin was discontinued and single agent daratumumab was maintained. To date, she has completed eight weekly doses and six biweekly doses of daratumumab and remains in MRD-negative complete remission for a duration of 8 months. Here we demonstrate the efficacy of daratumumab in combination with brentuximab vedotin in a patient with R/R T-ALL, who had received four prior lines of conventional chemotherapy, including allogeneic stem cell transplant with DLI, with high disease burden who achieved deep and durable remission. At this time, patients with R/R T-ALL have limited therapeutic options and poor survival outcomes, especially when relapse occurs early after first-line induction chemotherapy.12 Despite evidence for uniform CD38 expression in T-ALL blasts, clinical experience with daratumumab in T-ALL has been limited to case reports and small case series. Ofran et al. reported clinical efficacy of daratumumab monotherapy in three patients with heavily pretreated T-ALL who achieved complete remission but were MRD-positive after intensive chemotherapy prior to allogeneic stem cell transplantation. Daratumumab use was associated with eradication of residual disease after three to four doses, with disease-free survival of about 10 months.13 In a more recent case series of daratumumab in 20 patients with ALL including 13 with T-ALL with R/R or MRD-positive disease after a median of three lines of therapy, overall response rate was 20% including MRD-negative complete remission in two patients. One patient with high disease burden achieved a response when daratumumab was used in combination with chemotherapy. The median time to response in the overall cohort was 4 weeks and median survival was 4 weeks. Treatment was safe with no unexpected toxicities.14 Furthermore, the efficacy of daratumumab in CD38-positive B-ALL has also been demonstrated in a multiple relapsed pediatric patient with Philadelphia chromosome-positive B-ALL, prior treatments included DLI, two allogeneic stem cell transplants and anti-CD19

Figure 1. Metabolic response by positron emission tomography scan showing changes over the course of therapy. (A) Patient at diagnosis of T-cell acute lymphoblastic leukemia showing hypermetabolic anterior right chest wall mass and splenomegaly; and (B) after the first cycle of chemotherapy showing resolution of chest wall mass and improvement in splenomegaly.

Figure 2. Metabolic response by positron emission tomography scan showing changes over the course of therapy after relapse. (A) Prior to starting treatment with daratumumab and brentuximab vedotin showing hypermetabolic retroperitoneal and axillary lymph nodes and splenomegaly with increased 18F-fluorodeoxglucose avidity; and (B) after 1 cycle of daratumumab and brentuximab vedotin showing interval resolution of lymphadenopathy and decrease in the size of the spleen with normal uptake.

Haematologica | 109 February 2024

690

CASE REPORT

Figure 3. Bone marrow aspirate and core biopsy findings showing changes pre- and post-treatment with daratumumab and brentuximab vedotin. Sheets of blasts prior to daratumumab and brentuximab vedotin on (A) diagnostic bone marrow aspirate (Wright Giemsa, original magnification x400); (B) core biopsy (hematoxylin and eosin, original magnification x400); (C) strong CD30 positivity (immunohistochemistry, antibodies to CD30, original magnification x400) and (D) diffuse CD38 positivity on the blasts (immunohistochemistry, antibodies to CD38, original magnification x400); (E) post treatment bone marrow aspirate (Wright Giemsa, original magnification x600); (F) core biopsy in remission (hematoxylin and eosin, original magnification x100); (G) negative CD30 stain (immunohistochemistry, antibodies to CD30, original magnification x400) and (H) negative CD38 stain (immunohistochemistry, antibodies to CD38, original magnification x400).

chimeric antigen receptor T cells, who rapidly achieved morpholigical remission following intitation of daratumumab, together with weekly vincristine and ponatinib.15 Whether the deep and durable remission achieved in our patient was due to daratumumab, brentuixmab or combination of the two agents is unclear, although it is to be noted that our patient had previously received liposomal vincristine, which has a similar mechanism to brentuximab. To our knowledge, there are no reports for the use of brentuximab vedotin in T-ALL, but the expression of CD30 in a subset of patients with T-ALL provides a rationale for its evaluation in this setting. About one third of patients with T-ALL may express CD30 with increased expression level following chemotherapy.11 Brentuximab may lead to worsening of pre-existing peripheral neuropathy and prolonged cytopenias in heavily pretreated patients which may limit the duration of treatment or necessitate dose reductions as in our patient. Daratumumab is well tolerated and can be continued on a maintenance schedule (every 28 days). Large studies are needed to confirm the efficacy of this combination in the R/R setting and define the optimal doses, schedule, and treatment duration. Daratumumab in combination with chemotherapy is under investigation in relapsed/ refractory pediatric and young adults with ALL (clinicaltrials gov. Identifier: NCT03384654). Preliminary findings from 24 patients enrolled on the phase II DELPHINUS study with daratumumab and vincristine, prednisone, PEG-asparaginase, and doxorubicin (VPLD) reinduction backbone, showed complete remission in ten patients (41.7%) at the end of cycle 1.16 We also await with interest the results of EA9213, a phase II study of daratumumab-hyaluronidase for chemotherapy-relapsed/refractory minimal residual disease

in T-cell ALL (clinicaltrials gov. Identifier: NCT05289687). To our knowledge this is the first report to demonstrate daratumumab with brentuximab vedotin combination as an effective and safe therapy in relapsed/refractory T-ALL. Additional studies are required to elucidate the possible synergistic mechanism of daratumumab and brentuximab.

Authors Kebede H. Begna,1 Nadine H. Abdallah,1 Michelle Janania-Martinez,2 Abhishek A. Mangaonkar,1 Aruna Rangan,3 Jennifer L. Herrick,3 and Naseema Gangat1 1

Department of Internal Medicine, Division of Hematology, Mayo

Clinic, Rochester, MN; 2Division of Hematology/Oncology, Sanford Cancer Center, Sioux Falls, SD and 3Department of Laboratory Medicine and Pathology, Division of Hematopathology, Mayo Clinic, Rochester, MN USA Correspondence: K.H. BEGNA - begna.kebede@mayo.edu https://doi.org/10.3324/haematol.2023.283740 Received: June 12, 2023. Accepted: September 7, 2023. Early view: September 14, 2023. ©2024 Ferrata Storti Foundation Published under a CC BY-NC license

Haematologica | 109 February 2024

691

CASE REPORT

All authors reviewed and approved the final draft of the paper.

Disclosures No conflicts of interest to disclose.

Data-sharing statement Contributions

Data will be shared upon reasonable request addressed to the

KB and NA wrote the paper. KB, NA, MJ, AM, and NG participated in

corresponding author.

patient care. AR and JH performed review of bone marrow biopsies.

References 1. Marks DI, Paietta EM, Moorman AV, et al. T-cell acute lymphoblastic leukemia in adults: clinical features, immunophenotype, cytogenetics, and outcome from the large randomized prospective trial (UKALL XII/ECOG 2993). Blood. 2009;114(25):5136. 2. Fielding AK, Richards SM, Chopra R, et al. Outcome of 609 adults after relapse of acute lymphoblastic leukemia (ALL); an MRC UKALL12/ECOG 2993 study. Blood. 2007;109(3):944-950. 3. McMahon CM, Luger SM. Relapsed T cell ALL: current approaches and new directions. Curr Hematol Malig Rep. 2019;14(2):83-93. 4. Samra B, Alotaibi AS, Short NJ, et al. Outcome of adults with relapsed/refractory T-cell acute lymphoblastic leukemia or lymphoblastic lymphoma. Am J Hematol. 2020;95(9):E245-E247. 5. Moreau P, Kumar SK, San Miguel J, et al. Treatment of relapsed and refractory multiple myeloma: recommendations from the International Myeloma Working Group. Lancet Oncol. 2021;22(3):e105-e118. 6. Bride KL, Vincent TL, Im SY, et al. Preclinical efficacy of daratumumab in T-cell acute lymphoblastic leukemia. Blood. 2018;131(9):995-999. 7. Naik J, Themeli M, de Jong-Korlaar R, et al. CD38 as a therapeutic target for adult acute myeloid leukemia and T-cell acute lymphoblastic leukemia. Haematologica. 2019;104(3):e100-e103. 8. Ansell SM, Radford J, Connors JM, et al. Overall survival with brentuximab vedotin in stage III or IV Hodgkin’s lymphoma. N Engl J Med. 2022;387(4):310-320. 9. Horwitz S, O’Connor OA, Pro B, et al. Brentuximab vedotin with chemotherapy for CD30-positive peripheral T-cell lymphoma

(ECHELON-2): a global, double-blind, randomised, phase 3 trial. Lancet. 2019;393(10168):229-240. 10. Chen R, Gopal AK, Smith SE, et al. Five-year survival and durability results of brentuximab vedotin in patients with relapsed or refractory Hodgkin lymphoma. Blood. 2016;128(12):1562-1566. 11. Zheng W, Medeiros LJ, Young KH, et al. CD30 expression in acute lymphoblastic leukemia as assessed by flow cytometry analysis. Leuk Lymphoma. 2014;55(3):624-627. 12. Ribera JM, Morgades M, Genesca E, et al. Outcomes and prognostic factors of adults with refractory or relapsed T-cell acute lymphoblastic leukemia included in measurable residual disease-oriented trials. Hematol Oncol. 2021;39(4):529-538. 13. Ofran Y, Ringelstein-Harlev S, Slouzkey I, et al. Daratumumab for eradication of minimal residual disease in high-risk advanced relapse of T-cell/CD19/CD22-negative acute lymphoblastic leukemia. Leukemia. 2020;34(1):293-295. 14. Cerrano M, Bonifacio M, Olivi M, et al. Daratumumab with or without chemotherapy in relapsed and refractory acute lymphoblastic leukemia. A retrospective observational Campus ALL study. Haematologica. 2022;107(4):996-999. 15. Ganzel C, Kharit M, Duksin C, et al. Daratumumab for relapsed/ refractory Philadelphia-positive acute lymphoblastic leukemia. Haematologica. 2018;103(10):e489-e490. 16. Hogan LE, Bhatla T, Teachey DT, et al. Efficacy and safety of daratumumab (DARA) in pediatric and young adult patients (pts) with relapsed/refractory T-cell acute lymphoblastic leukemia (ALL) or lymphoblastic lymphoma (LL): results from the phase 2 DELPHINUS study. J Clin Oncol. 2022;40(Suppl 16):S1001.

Haematologica | 109 February 2024

692

ERRATUM

Erratum to: Introduction to the peripheral T-cell lymphoma review series: advances in molecular characterization, classification refinement and treatment optimization Kerry J. Savage1 and Laurence de Leval2

Correspondence: K.J. Savage

Center for Lymphoid Cancer, Division of Medical Oncology, BC Cancer and the University of British Columbia, Vancouver, British Columbia, Canada and 2Institute of Pathology, Department of Laboratory Medicine and Pathology, Lausanne University Hospital and Lausanne University, Lausanne, Switzerland 1

ksavage@bccancer.bc.ca Received: Accepted:

January 8, 2024. January 8, 2024.

After the publication of the article "Introduction to the peripheral T-cell lymphoma review series: advances in molecular characterization, classification refinement and treatment optimization" by Kerry J. Savage and Laurence de

Leval in the December 2023 issue,1 we realized that an error had been made in Table 1. The correct table is shown below. The Editorial Office apologize for the inconvenience.

Table 1. Mature T- and NK-cell neoplasms in the International Consensus Classification (ICC) and World Health Organization (WHO)-HAEM5 classification (2022) in reference to the WHO-HAEM4R classification (2017) (adapted from Alaggio et al.1 and Campo et al.2). WHO-HAEM4R 2017

ICC 2022

WHO-HAEM5 2022

T-cell prolymphocytic leukemia

T-prolymphocytic leukemia

T-cell large granular lymphocytic leukemia

T-large granular lymphocytic leukemia

Chronic lymphoproliferative disorder of NK cells

NK-large granular lymphocytic leukemia

Adult T-cell leukemia/lymphoma

EBV-positive T-cell/NK-cell lymphoproliferative disorders of childhood

EBV-positive T-cell and NK-cell lymphoid proliferations and lymphomas of childhood

Hydroa vacciniforme-like lymphoproliferative disorder

Hydroa vacciniforme lymphoproliferative disorder, classic type and systemic type

Hydroa vacciniforme lymphoproliferative disorder

Severe mosquito bite allergy

Chronic active EBV infection of T- and NKcell type, systemic form

Chronic active EBV disease, systemic (Tcell and NK-cell phenotype)

Systemic chronic active EBV disease

Systemic EBV-positive T-cell lymphoma of childhood

Extranodal NK/T-cell lymphoma, nasal type

Extranodal NK/T-cell lymphoma

Aggressive NK-cell leukemia

Not listed as an entity, subtype of peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS)

Primary nodal EBV-positive T-cell/NK-cell lymphoma

EBV-positive nodal T- and NK-cell lymphoma

Continued on following page. Haematologica | 109 February 2024

693

ERRATUM

K.J. Savage and L. de Leval WHO-HAEM4R 2017

ICC 2022

WHO-HAEM5 2022

Enteropathy-associated T-cell lymphoma

Not listed as an entity

Type II refractory celiac disease

Not listed as an entity

Monomorphic epitheliotropic intestinal T-cell lymphoma

Intestinal T-cell lymphoma, NOS

Indolent T-cell lymphoproliferative disorder of Indolent clonal T-cell lymphoproliferative the gastrointestinal tract disorder of the gastrointestinal tract

Indolent T-cell lymphoma of the gastrointestinal tract

Not listed

Indolent NK-cell lymphoproliferative disorder of the gastrointestinal tract

Hepatosplenic T-cell lymphoma

Mycosis fungoides

Sézary syndrome

Primary cutaneous CD30+ T-cell lymphoproliferative disorders Lymphomatoid papulosis

Primary cutaneous CD30+ T-cell lymphoproliferative disorders

Primary cutaneous CD30+ T-cell lymphoproliferative disorder: Lymphomatoid papulosis

Primary cutaneous anaplastic large cell lymphoma

Primary cutaneous CD30+ T-cell lymphoproliferative disorder Primary cutaneous anaplastic large cell lymphoma

Primary cutaneous CD4+ small/medium T-cell lymphoproliferative disorder

Primary cutaneous small/medium CD4+ T-cell lymphoproliferative disorder

Subcutaneous panniculitis-like T-cell lymphoma Primary cutaneous gamma-delta T-cell lymphoma

Subcutaneous panniculitis-like T-cell lymphoma

Primary cutaneous gamma-delta T-cell lymphoma

Primary cutaneous acral CD8+ T-cell lymphoma

Primary cutaneous acral CD8+ T-cell lymphoproliferative disorder

Primary cutaneous CD8+ aggressive epidermotropic cytotoxic T-cell lymphoma

Not listed

Primary cutaneous peripheral T-cell lymphoma, NOS

Peripheral T-cell lymphoma, NOS

Nodal lymphomas of T follicular helper origin

Follicular helper T-cell lymphoma

Nodal T-follicular helper (TFH) cell lymphoma

Lymphomatoid papulosis

Angioimmunoblastic T-cell lymphoma

Follicular helper T-cell lymphoma, angioimmunoblastic type (angioimmunoblastic Tcell lymphoma)

Nodal TFH cell lymphoma, angioimmunoblastic-type

Follicular T-cell lymphoma

Follicular helper T-cell lymphoma, follicular type

Nodal TFH cell lymphoma, follicular-type

Nodal peripheral T-cell lymphoma with T follicular helper phenotype

Follicular helper T-cell lymphoma, NOS

Nodal TFH cell lymphoma, NOS

Anaplastic large cell lymphoma, ALK-positive Anaplastic large cell lymphoma, ALK-positive ALK-positive anaplastic large cell lymphoma Anaplastic large cell lymphoma, ALK-negative Anaplastic large cell lymphoma, ALK-negative ALK-negative anaplastic large cell lymphoma Breast implant-associated anaplastic large cell lymphoma

Breast implant-associated anaplastic large cell lymphoma

The entities are listed according to the order in which they appear in the ICC 2022. Shading/no shading denotes groups of entities. Provisional entities in WHO-HAEM4R and the ICC are shown in italics. EBV: Epstein-Barr virus; NOS: not otherwise specified; ALCL: anaplastic large cell lymphoma.

References 1. Savage KJ and de Leval L. Introduction to the peripheral T-cell lymphoma review series: advances in molecular

characterization, classification refinement and treatment optimization. Haematologica 2023;108(12)3204-3210.

Haematologica | 109 February 2024

694

Journal of the Ferrata Storti Foundation

haematologica.org

ISSN 0390 - 6078