Issuu

Peer-Reviewed Articles Theme 3 - Topographic Tumor Heterogeneity Salvador J. Diaz-Cano, LMS, MD, PhD, FRCPath Research Project • Clonal Evolution and Topographic Tumor Heterogeneity •

Salvador J. Diaz-Cano • Peer-Reviewed Articles • Topographic Tumor Heterogeneity

Contents Topographic Tumor Heterogeneity

Molecular and kinetic features of transitional cell carcinomas of the bladder: biological and clinical implications. Virchows Arch 2001;438:289-297! 4 Suhail I. Baithun · Mahmoud Naase · Alfredo Blanes · Salvador J. Diaz-Cano 4

Molecular Evolution and Intratumor Heterogeneity by Topographic Compartments in Muscle-Invasive Transitional Cell Carcinoma of the Urinary Bladder. Lab Invest 2000;80(3):279-289! 4 Salvador J. Diaz-Cano, Alfredo Blanes, Javier Rubio, Alfredo Matilla, and Hubert J. Wolfe 4

Kinetic profiles by topographic compartments in muscle-invasive transitional cell carcinomas of the bladder. Role of TP53 and NF1 genes. Am J Clin Pathol 2002;118:93-100! 4 Alfredo Blanes, Javier Rubio, Armando Martinez, Hubert J. Wolfe, and Salvador J. Diaz-Cano 4

Topographic Molecular Profile of Pheochromocytomas: Role of Somatic DownRegulation of Mismatch Repair. J Clin Endocrinol Metab 2006;91:1150-1158! 4 Alfredo Blanes, Juan J. Sanchez-Carrillo, and Salvador J. Diaz-Cano

Complementary analysis of microsatellite tumor profile and mismatch repair defects in colorectal carcinomas. World J Gastroenterol 2006; 12(37): 59325940.! 4 Alfredo Blanes, and Salvador J. Diaz-Cano Salvador J. Diaz-Cano • Peer-Reviewed Articles • Topographic Tumor Heterogeneity

4 2

Coexistent intraurothelial carcinoma and muscle-invasive urothelial carcinoma of the bladder: clonality and somatic down-regulation of DNA mismatch repair. HUM PATHOL 2009;40:988-997! 4 Alfredo Blanes, Javier Rubio, Juan J. Sanchez-Carrillo, Salvador J. Diaz-Cano 5

Salvador J. Diaz-Cano â€˘ Peer-Reviewed Articles â€˘ Topographic Tumor Heterogeneity

MOLECULAR AND KINETIC FEATURES OF TRANSITIONAL CELL CARCINOMAS OF THE BLADDER: BIOLOGICAL AND CLINICAL IMPLICATIONS. Virchows Arch 2001;438:289-297 Suhail I. Baithun · Mahmoud Naase · Alfredo Blanes · Salvador J. Diaz-Cano

MOLECULAR EVOLUTION AND INTRATUMOR HETEROGENEITY BY TOPOGRAPHIC COMPARTMENTS IN MUSCLE-INVASIVE TRANSITIONAL CELL CARCINOMA OF THE URINARY BLADDER. Lab Invest 2000;80(3):279-289 Salvador J. Diaz-Cano, Alfredo Blanes, Javier Rubio, Alfredo Matilla, and Hubert J. Wolfe

KINETIC PROFILES BY TOPOGRAPHIC COMPARTMENTS IN MUSCLE-INVASIVE TRANSITIONAL CELL CARCINOMAS OF THE BLADDER. ROLE OF TP53 AND NF1 GENES. Am J Clin Pathol 2002;118:93-100

Alfredo Blanes, Javier Rubio, Armando Martinez, Hubert J. Wolfe, and Salvador J. Diaz-Cano

TOPOGRAPHIC MOLECULAR PROFILE OF PHEOCHROMOCYTOMAS: ROLE OF SOMATIC DOWN-REGULATION OF MISMATCH REPAIR. J Clin Endocrinol Metab 2006;91:1150-1158 Alfredo Blanes, Juan J. Sanchez-Carrillo, and Salvador J. Diaz-Cano

COMPLEMENTARY ANALYSIS OF MICROSATELLITE TUMOR PROFILE AND MISMATCH REPAIR DEFECTS IN COLORECTAL CARCINOMAS. WORLD J GASTROENTEROL 2006; 12(37): 5932-5940. Alfredo Blanes, and Salvador J. Diaz-Cano

COEXISTENT INTRAUROTHELIAL CARCINOMA AND MUSCLE-INVASIVE UROTHELIAL CARCINOMA OF THE BLADDER: CLONALITY AND SOMATIC DOWNREGULATION OF DNA MISMATCH REPAIR. HUM PATHOL 2009;40:988-997 Salvador J. Diaz-Cano • Peer-Reviewed Articles • Topographic Tumor Heterogeneity

Alfredo Blanes, Javier Rubio, Juan J. Sanchez-Carrillo, Salvador J. DiazCano

Salvador J. Diaz-Cano • Peer-Reviewed Articles • Topographic Tumor Heterogeneity

Virchows Arch (2001) 438:289–297 DOI 10.1007/s004280000289

O R I G I N A L A RT I C L E

Suhail I. Baithun · Mahmoud Naase · Alfredo Blanes Salvador J. Diaz-Cano

Molecular and kinetic features of transitional cell carcinomas of the bladder: biological and clinical implications Received: 20 March 2000 / Accepted: 21 June 2000 / Published online: 1 November 2000 © Springer-Verlag 2000

Abstract Molecular and kinetic analyses have contributed to our understanding of the biology of transitional cell carcinomas (TCC) of the bladder. The concordant pattern of X-chromosome inactivation of multiple TCCs appearing at different times and at different sites and concordant genetic abnormalities in a subset of muscleinvasive TCC strongly support a monoclonal origin and a homogeneous tumor cell selection throughout the neoplasm. However, topographic intratumor heterogeneity results from the accumulation of genetic lesions in tumor suppressor genes, predominantly neurofibromatosis (NF)1-defective in the superficial compartment and tumor protein p53 (TP53)-defective in the deep one, with lower proliferation and down-regulation of apoptosis in the latter. TCCs follow the general concept of multistep carcinogenesis and proceed through two distinct genetic pathways responsible for generating different TCC morphologies. These are the inactivation of cyclin-dependent kinase inhibitors (p15, p16, and p21WAF/CIP1) in lowgrade TCC and early TP53-mediated abnormalities in high-grade TCC. TCC progression correlates with genetic instability and accumulation of collaborative genetic lesions mainly involving TP53, retinoblastoma (RB)-1, and growth factors. Distinctive genetic (low incidence of RB-1 and NF-1 abnormalities) and kinetic (slower cell turnover) profiles also correlate with a “single-file” infiltration pattern and poor survival in muscle-invasive TCCs. The underlying molecular changes of carcinoma in situ involve multiple and more extensive deletions S.I. Baithun · M. Naase · S.J. Diaz-Cano (✉) Department of Pathology, St Bartholomew’s and the Royal London School of Medicine and Dentistry, Queen Mary, University of London, London, UK e-mail: s.j.diaz-cano@mds.qmw.ac.uk Tel.: +44-171-3777348, Fax: +44-171-3777030 S.J. Diaz-Cano Department of Histopathology and Morbid Anatomy, The Royal London Hospital, Whitechapel, London, E1 1BB, UK A. Blanes Departamento de Anatomia Patologica, Facultad de Medicina, Universidad de Malaga, Malaga, Spain

(normally TP53-defective) than coexistent invasive TCC, suggesting an independent genetic evolution, while lowgrade dysplasia is mainly polyclonal and shows a low rate of gene deletions. Keywords Bladder · Transitional cell carcinoma · Oncogene · Tumor Suppressor Gene · Cell kinetics · Tumor progression · Tumor heterogeneity

Introduction The heterogeneity and complexity of transitional cell carcinomas (TCCs) of the urinary bladder are reflected in the number of publications. More than 700 references in the last 5 years can be retrieved from bibliography databases regarding the molecular and kinetic features of these tumors. Two opposing theories explain multiple TCCs as multicentric (field carcinogenesis) or multifocal (spread from a single tumor). Clonality analysis has helped to resolve this disparity, but the results of any analysis must be interpreted in the appropriate setting of neoplastic transformation and cellular kinetics (tumor cell selection) [15, 17, 18]. Additionally, the association between clonality and cell kinetics closely correlates with intratumor heterogeneity, depth of invasion, and the molecular pathways [51, 52, 61]. Bladder TCCs have been classified into low-grade papillary TCC (usually superficial), high-grade TCC (normally muscle-invasive), and flat intraurothelial lesions [dysplasia, carcinoma in situ (CIS)] [19] However, this grading–staging system does not correlate with the molecular patterns and leaves undefined the relationship between low-grade dysplasia, primary CIS, and secondary CIS. We review the role and relationship of clonality and cell kinetics in tumor progression of bladder TCC with special emphasis on intratumor heterogeneity. The molecular pathways of bladder TCCs are then considered for both invasive (low- and high-grade TCC) and intraurothelial neoplasms, with attention focusing on the ge-

290

netic control of proliferation and apoptosis, the two key elements in TCC development.

Tumor cell selection and kinetics in invasive TCC Neoplasms are defined as abnormal, self-maintained proliferations, which share a common kinetic advantage and result from the accumulation of genetic lesions (multistep tumorigenesis) [7, 15, 17, 44, 53, 59, 66]. In that setting, the results of any molecular test depend on the timing between the biological feature evaluated and the specific marker and are positive only if the target participates in the molecular pathway (Fig. 1) [15, 17, 18, 39]. In addition, one must consider cell kinetics. Some mutations are, however, more important than others because they facilitate additional phenotypic or genotypic changes, thus explaining certain trends in the sequence of mutations [15, 17]. Cooperative genetic alterations in kinetically active tumor cells lead to progressive clone selection and enhance genetic diversity within a neoplasm, resulting in tumor outgrowth only if they bypass the cell repairing systems, do not activate the apoptotic pathway, and maintain proliferation [15, 18]. The lack of a distinct sequence and the unpredictability of genetic changes in malignancies preclude the extensive clinical use of those genetic markers for diagnostic and prognostic purposes.

Low-grade superficial TCC often present as multiple tumors, appearing at different times and at different sites in the bladder, indicating that a “field defect” has occurred. The concordant pattern of X chromosome inactivation and microsatellite markers in multiple TCC from a single patient suggests that different TCC in a single patient derive from the same progenitor cell [59, 63]. Allele typing in each tumor from a given patient has shown loss of the same allele on chromosome 9q, suggesting that this loss preceded the spread of neoplastic cells in the bladder. Losses of 17p and 18q alleles, however, are not common to different tumors, suggesting late genetic events [59]. Similarly, the superficial and deep compartments of muscle-invasive TCCs have been demonstrated to show a concordant microsatellite pattern of tumor suppressor genes (TSG) and the same X-chromosome inactivated [4]. Coexistent genetic abnormalities involving two or more TSG loci in several samples strongly support a monoclonal origin [15, 17]. The background level of loss of heterozygosity (LOH) reported in tumors and normal tissues ranges between 4% and 20% [14, 69]. Considering all genetic lesions to be equally important and frequent, the probability of finding coexistent genetic alterations randomly would be 0.22 for two genetic loci, 0.23 for three genetic loci, and so on. Applying this principle to two separately microdissected samples (superficial and deep) from a given tumor, the probability of getting the same locus involved in both samples would be (0.22)2 for two genetic loci or (0.23)2 for three genetic loci. We found the same inactivated X-chromosome and concordant TSG microsatellite patterns in both compartments in 68.2% of muscle-invasive TCC, strongly supporting a monoclonal origin and a homogeneous tumor cell selection throughout the neoplasm. This finding suggests that this subset of muscle-invasive TCC arose from the uncontrolled spread of a single transformed cell and then grew through a multistep tumorigenesis that involves “common” TSG, such as tumor protein p53 (TP53), retinoblastoma (RB-1), and neurofibromatosis (NF)-1 [16].

Intratumor heterogeneity: correlation between molecular and kinetic profiles in TCC Histological and genetic heterogeneity is well documented in TCCs [2, 5, 28, 56, 63, 67]. After malignant transformation, tumor cells can grow independently with variable subsequent genetic alterations in each tumor comFig. 2 Molecular pathways in bladder carcinogenesis. Both pathways can only be distinguished by the relative timing of genetic abnormalities. Genetic instability (shaded area) is mainly related to TP53 and/or RB-1 dysfunction that leads to tumor progression; it has been described in late stages for low-grade superficial transitional cell carcinomas (TCC) and in early phases for high-grade TCC. The biologic process of progression is determined by blockage of cellular senescence, resulting in topographic cell selection by compartments in muscle-invasive TCC [16]

▲

Fig. 1 Clonality analyses and tumor progression. Tumors evolve in multiple steps of initiation and promotion resulting in clone selection and eventually progression. Two main stages can be identified (preneoplastic and neoplastic), and these are separated by an irreversible genetic change. After the initial mutation, the kinetic advantage determines a clonal proliferation detectable with tests, such as X-chromosome inactivation. However, specific chromosome markers only provide positive information if mutated; they are negative before that event. This reason justifies the dual approach for clonality assays in neoplastic conditions of the bladder

291

292

partment, explaining the heterogeneity [16, 63]. In that sense, the presence of additional clonal abnormalities in different cultures illustrates intratumoral heterogeneity [43], but tumor cell topography in the bladder wall has rarely been considered [16]. Multiple genetic lesions can be expected in muscleinvasive TCC if we consider the accumulation of genetic alterations to be an expression of molecular progression [15]. It would help to explain the relatively high incidence of genetic abnormalities reported in RB-1 and Wilms Tumor (WT)-1 loci [16]. Inactivated retinoblastoma protein (pRB-1) has been found in one-third of randomly selected bladder cancers [29], and RB-1 abnormalities in 51.5% of muscle-invasive TCCs [16]. WT-1 exon mutations were not found in any of 11 randomly selected bladder cancers using polymerase chain reaction (PCR)-single strand conformation polymorphism analysis and restriction fragment length polymorphism analysis [30]. However, more sensitive microsatellite analysis using PCR/denaturing gradient gel electrophoresis (DGGE) demonstrated WT-1 abnormalities in 53.8% of muscle-invasive TCCs [16]. Topographic genetic heterogeneity has been reported in 31.8% of muscle-invasive TCC [16], but it did not occur at the phenotypic level: the pattern of protein expression showed no relationship with LOH or single nucleotide polymorphism [4, 16]. The comparison of topographic compartments of muscle-invasive TCCs has suggested that two processes of tumor cell selection should be responsible for the topographic heterogeneity. NF-1 was more frequently altered in tumors in the superficial compartment (75% of cases), while the deep compartment always revealed TP53 abnormalities (Fig. 2) [16]. The NF1 gene has rarely been implicated in bladder carcinogenesis, and no mutations were initially reported [65]. However, the presence of NF-1 gene abnormalities in 63.3% of muscle-invasive TCCs [16], and the decreased NF-1 mRNA and protein levels in high-grade TCCs [1] support an important role for NF-1 in bladder carcinogenesis, especially superficial TCC compartments [16]. The NF-1 gene product has an effect on ras inhibition, and ras protein is expressed with the highest levels in immature and proliferating cells. The absence of the NF-1 inhibitory effect will favor increased proliferation. TP53 abnormalities have been reported in 67.4% of the muscle-invasive TCCs [16] and 61.1% of the invasive TCCs [58], frequently showing coexistent allelic deletions and mutations [58]. TP53 abnormalities are more frequently found in the deep tumor compartment, a finding that may represent the consequence of tumor cell selection [16]. Similarly, the loss of chromosome 17p has been reported as a late event in tumor progression in superficial TCC [59]. Figure 2 summarizes this position [16]. The differential genetic profile by tumor compartments is also reflected on the tumor proliferation and differentiation grade. Kinetic markers (including proliferation and apoptosis) have been used as diagnostic and prognostic tools in TCC [31, 34, 37] but have revealed variability as a result of intratumor heterogeneity. Using slide cytome-

try, the topographic evaluation of kinetic features in 72 muscle-invasive TCCs showed significantly higher mitotic figure counts, Ki-67 index, and proliferation index in the superficial compartment than in the deep compartment. Apoptosis based on the in situ end labeling (ISEL) of fragmented DNA was revealed to be less than 1% in 63% of the superficial compartments and 86% of the deep compartments (P=0.05). These results suggest that lower proliferation activity and downregulation of apoptosis mainly define the kinetic profile in the deep compartment of muscle-invasive TCC of the bladder and correlate with the genetic profile described above [16, 51].

Molecular and kinetic patterns in urothelial dysplasiaâ€“CIS Urothelial dysplasia is frequently identified as being associated with TCC of the urinary tract and is assumed to be the redundant precursor of invasive TCC. The Consensus Classification of Urothelial Neoplasms distinguishes low-grade urothelial dysplasia (LGUD) and high-grade urothelial dysplasia (HGUD)â€“CIS only [19]. Urothelial dysplasia confers a significant risk for the development of CIS and invasive TCC; cytological progression is documented but with different topography reported for LGUD and CIS in a given patient [9]. This finding supports the multifocal distribution of TCC, although does not prove the clonal identity of LGUD and CIS to sustain the sequence LGUDâ†’CIS. Fluorescence in situ hybridization using gene locusspecific probes for chromosomes 9q22 (FACC), 9p21 (p16/CDKI2), and 17p13 (TP53) has shown the same chromosome nine deletions in urothelial hyperplasias (10 of 14), coexistent low-grade papillary superficial TCCs (8 of 9), and the surrounding histologically normal urothelium (6 of 12) [27]. In contrast, 17p13 hemizygous deletion was found in the urothelial hyperplasia and papillary TCC from 1 of 12 patients, whereas the normal urothelium was always normal [27]. This genetic profile suggests a neoplastic potential for flat urothelial lesions, regardless of whether or not cytologic atypia is present. It also suggests that the earliest molecular alterations of low-grade TCC involve chromosome 9 (p16/CDKI2) but not chromosome 17 (TP53) [27]. However, this study does not analyze urothelial dysplasia and high-grade muscle-invasive TCC, which have been reported to evolve through a different molecular pathway [52, 61]. The biologic and kinetic patterns of LGUD and CIS have been revealed to be different. The analysis of dysplastic lesions associated with muscle-invasive TCC revealed monoclonal patterns in CIS (6, 100%), invasive TCC (13, 100%), and LGUD (2, 20%), whereas polyclonal patterns were observed in LGUD only (8, 80%). CIS showed aneuploid DNA content and more microsatellite loci altered (PCR/DGGE analysis) than the corresponding invasive compartments, always involving TP53 loci and expressing abnormal p53 (12 cases) [52, 61]. In contrast, LGUD (18 cases) revealed diploid DNA content

293

and microsatellite abnormalities in only two cases, one monoclonal (RB-1) and one polyclonal (WT-1 and NF-1). Opposite kinetic patterns were observed for CIS (higher Ki-67 index, lower ISEL index) and LGUD (lower Ki-67 index, higher ISEL index). These genetic findings suggest that CIS evolution is independent of muscle-invasive TCC, whereas LGUD should not be closely connected with this molecular progression. These kinetic patterns would contribute to the accumulation of genetic abnormalities in CIS but not in LGUD. LOH involving several loci have been shown in CIS, especially 9p (77%), 14q (70%), 8p (65%), 17p (60%), 13q (56%), 11p (54%), and 4q (52%), and slightly less frequently 11q (36%), 4p (32%), 3p (31%), 18q (29%), and 5q (20%) [6, 8, 23, 24, 25, 26, 57, 70]. The key molecular changes are largely unknown, but both p53 over-expression and LOH of chromosome 9 have been implicated in the progression into muscle-invasive TCC [52]. Therefore, only CIS displays the genetic alterations reported in invasive TCC, potentially accounting for the aggressive nature of these lesions [52]. However, both the differential LOH pattern found in the superficial TCC compartment (NF-1 defective) and the accumulation of genetic damages in CIS does not support the sequence CIS→superficial TCC→deep TCC [16]. This incidence of genetic abnormalities has been revealed to be higher in CIS than in invasive TCCs, suggesting an advanced molecular stage for CIS. As presented above, several and coexistent deletions, generally involving TP53 gene, characterize secondary CIS and suggest that CIS follows an independent progression from the invasive TCC rather than its precursor lesion.

Molecular findings of use in the classification and pathogenesis of TCC The molecular analysis of tumors on the basis of gene expression can identify previously undetected and clinically significant subtypes of cancer. DNA microarrays are valuable in tumor classification and prognosis [21], as demonstrated in gastrointestinal tumors. This strategy will be useful in the future for discovering and predicting cancer classes, but it will require certain identification of cell populations, because it is based on gene expression. The heterogeneous differentiation normally observed in epithelial neoplasms would also limit its general application in carcinomas, although the technique can reveal useful genes for diagnosis and prognosis. In contrast, in situ hybridization allows efficient recognition of the chromosomes involved in the process of imbalance and is a valid method to assess structural chromosome aberrations in interphase nuclei of TCC [46, 50, 68].

Molecular pathways in TCC TCC is believed to arise through a series of genetic changes affecting proliferation and apoptosis. TCC occur

via multistep carcinogenesis and proceed through two distinct pathways of genetic alterations. One is associated with papillary low-grade TCC, where the initial key factor is inactivation of CDK inhibitors (p15, p16, and p21WAF/CIP1). The second is associated with high-grade lesions, TP53 abnormalities being the key element. TCC progression correlates with genetic instability and accumulation of collaborative genetic lesions, mainly involving TP53, RB-1, and growth factors (Fig. 2) [32, 61]. Papillary pTa-pT1 TCCs are often multifocal and only occasionally progress, whereas CIS frequently progress to invasive disease [19]. This distinctive morphology reflects the differential molecular background: LOH of chromosome 9 has been observed mainly in papillary low-grade TCC, while TP53 alterations characterize high-grade TCCs. Both TP53 and RB-1 alterations are also known to occur in early stage bladder CIS where they are thought to indicate a poor prognosis through their propensity to progress, since these mutations are known to destabilize the genome. The presence of TP53 mutations has been similarly reported in CIS and muscle-invasive TCCs [16, 51, 61]. Analysis of several tumor pairs involving a CIS and an invasive cancer provided evidence that chromosome 9 alteration may in some cases be involved in the progression of CIS to invasive TCC, in addition to its role in the initiation of pTa TCCs [63]. Alterations in TP53 and RB-1 occur in approximately 50% and 33% of bladder cancers, respectively, and are associated with high-stage, high-grade TCC [12, 13]. Amplification and/or over-expression of growth factor receptors, such as c-erbB-2 and loss of 17p and 18q sequences are associated with advanced TCC [59]. Finally, recent findings generated using in vitro transformation systems with human urothelial cells provide strong evidence that the loss of genes on 3p, reported in approximately 20% of TCC, and/or the gain of genes on 20q play an important role in blocking cellular senescence. This feature should represent a critical step in oncogenesis, because cells that do not senesce can survive to accumulate multiple genetic alterations [15], such as those associated with invasive TCC (Fig. 2) [51]. Additionally, the genetic profile has also been correlated with the infiltration pattern of muscleinvasive TCC [16]. TSG microsatellite analysis using PCR/DGGE in muscle-invasive TCCs has revealed a distinctive genetic profile in “single-file” TCCs, characterized by a low incidence of genetic abnormalities not involving the RB-1 locus (P=0.0003) and very occasionally involving the NF-1 locus (two cases, 13%; P=0.0023). “Single-file” TCCs showed lower cell turnover (Ki-67 index 14.94±4.28, ISEL 14.1±10.0), lower incidence of aneuploid DNA content, and shorter mean survival (20 months) than solid TCCs (Ki-67 index 20.65±4.94, ISEL 20.2±22.7, 37-month survival, respectively; unpublished observations). The relatively low cell turnover would be consistent with the low incidence of genetic alterations, especially RB-1 and NF-1, suggesting that “single-file” muscle-invasive TCCs are

294

related by having an alternative molecular pathway. It is not known yet whether specific alterations targeting cell adhesion molecules, thus disturbing the growth pattern, are involved.

Deregulation of the genetic control of cell kinetics in TCC Those pathways mainly involve three basic types of genes: oncogenes, TSG, and genes involved in DNA repair. Most studies in bladder TCC have centered the attention on cell kinetic regulators of proliferation-apoptosis [TP53, MDM-2, RB-1, E2F-1 transcription factor, and cyclin-dependent kinase (CDK) inhibitors] and telomerase (Fig. 3). TP53, probably the most extensively studied gene in TCC [11, 20, 51, 54, 55, 60, 61, 64], shows diverse biological functions and a heterogeneous molecular mechanism of inactivation [10]. Alterations in TP53 have been reported to be associated with bad prognostic factors such, as high grade or stage [20, 60], or higher proliferation rates [40, 41, 48]. However, TP53 is only one of the cell cycle regulators at the restriction point (Fig. 3) and must be considered together with other regulators, especially MDM-2. Its interaction with RB-1 and its associated transcription factor (E2F-1) is also important. The MDM-2 gene is located on the long arm of chromosome 12 (12q13–14) and encodes for a 90 kDa nuclear protein (Mdm2) [10]. MDM-2 links the two main regulators of the restriction point, TP53 and RB-1. Mdm2 binds pRB-1 and is shown to be essential for RB-1 to overcome both the anti-apoptotic function of Mdm2 and the MDM-2-dependent degradation of p53. The

Fig. 3 Main cell cycle regulators of the G1-S transition (‘restriction point’). The final outcome (cell cycle progression vs G1 arrest – apoptosis) is determined by the balance of several gene products

RB-MDM-2 interaction does not prevent MDM-2 from inhibiting p53-dependent transcription, but the RBMDM-2 complex still binds to p53. Since RB-1 specifically rescues the apoptotic function [22], but not the transcriptional activity of p53 from negative regulation by MDM-2, transactivation by wild-type p53 is not required for the apoptotic function of p53. However, a pRB-Mdm2-p53 trimeric complex is active in p53mediated transrepression. These data link directly the function of p53 and pRB-1 and demonstrate a novel role of RB-1 in regulating the apoptotic function of p53. TCC have demonstrated a strong statistical association between Mdm2 and p53 overexpression, with Mdm2 overexpressed in low-stage, low-grade TCC. Those results suggest that aberrant Mdm2 and p53 phenotypes are frequent events in bladder cancer and may be involved in tumorigenesis or tumor progression [35]. RB-1 plays a role similar to TP53 in TCC: losses of functional protein correlate with aggressive behavior, high grade, and stage [38]. Both TP53 and RB-1 have frequently been reported to be abnormal in TCC and have prognostic significance, suggesting a cooperative mechanism in tumor progression [12, 13]. Altered p53 and undetectable pRB-1 are commonly found together and correlate with a marked increase in progression and decreased overall survival, after stratifying cases for tumor stage, tumor grade, and suspicion of vascular invasion [12]. Likewise, TCC with altered p53 and pRB-1 have shown significant increased rates of recurrence and survival, relative to patients with no alterations in either p53 or pRB-1; patients with alterations in only one of these proteins had intermediate rates of recurrence and survival [13]. These data suggest that alterations of p53 and pRB-1 have a cooperative or synergistic negative effect to promote tumor progression and in decreasing survival in primary TCC. It may be postulated that aberrant p53 and pRB-1 expression deregulates cell cycle control at the G1 checkpoint and engenders tumor cells with reduced response to programmed cell death (Fig. 3), resulting in the aggressive clinical course of TCC harboring both p53 and pRB-1 alterations [12]. In addition, cases with undetectable and high pRB-1 reactivity had identical rates of recurrence, indicating that high levels of pRB-1 expression may reflect a dysfunctional (altered) RB-1 pathway and do not reflect the tumor suppressor effects of the protein [13]. In the complex regulation of the restriction point, cyclin D1 and the transcription factor E2F-1 (Fig. 3) play a role and have been implicated in bladder carcinogenesis. Cyclin D1 can also cooperate to dysregulate the G1 checkpoint, which becomes completely abolished only if RB-1 is lost, removing any selective advantage for cells that alter additional cell cycle proteins [42]. E2F-1 is a transcription factor that binds to pRB-1 [10]. The results reported on TCC have demonstrated occasional nucleotide polymorphism and no bandshifts in the nuclearlocalization or DNA-binding domains.[49]. E2F-1 alterations occur at the phenotypic level, rather than at the genotypic level, in TCC: the pattern of E2F-1 protein ex-

295

pression has shown no relationship with nucleotide polymorphism but correlated inversely with the percentage of cells showing pRB-1 reactivity. Rabbani et al. also reported that patients with lower E2F-1 reactivity had statistically significant increased risks of progression to metastases and death, suggesting a possible tumor suppressor role for E2F-1 in TCC [49]. CDK inhibitors implicated in bladder carcinogenesis include p15, p16, and p21WAF/CIP1. p16 (also known as CDKN2, INK4a, or MTS1) and p15 (also described as INK4b or MTS2) are found in tandem on chromosome 9p21. They encode proteins that function as negative cell cycle regulators encoding inactivating polypeptides that inactivate specific cyclin-protein kinase complexes required for progression through the cell cycle (Fig. 3) [10]. The overall frequency of alteration reported for these genes in TCC is approximately 18% for each one, being significantly associated with low stage and grade TCC for p16 gene and low stage TCC for p15 gene alterations [45]. Overall, 70% of the TCCs showed abnormalities in one or more of the intrinsic proteins of the G1 checkpoint [42], supporting the cooperative role of p16 and p15 gene alterations (especially coincident homozygous deletion) as a common event in bladder carcinogenesis [42, 45]. These CDK inhibitors cross-talk several molecular pathways, as demonstrated for the INK4a gene [47]. The INK4a gene encodes two distinct growth inhibitors – the CDK inhibitor p16Ink4a, which is a component of the RB-1 pathway and the tumor suppressor p19Arf, which has been functionally linked to p53. p19Arf potently suppresses the oncogenic transformation in primary cells, and this function is abrogated when TP53 is neutralized by viral oncoproteins and dominantnegative mutants but not by the TP53 antagonist MDM-2. Coupled with the findings that p19Arf and Mdm2 physically interact and that p19Rrf blocks MDM-2-induced p53 degradation and transactivational silencing, the results reported suggest that p19Arf functions mechanistically to prevent neutralization of p53 by MDM-2. All of these findings together ascribe the potent tumor suppressor activity of INK4a to the cooperative actions of its two protein products and their relation to the two central growth control pathways, RB-1 and TP53 [47]. The action of TP53 on cell cycle regulation is mediated, in part, through the expression of p21WAF/CIP1, as suggested by the significant association between p21WAF/CIP1 expression and p53 status [62]. p21WAF/CIP1 expression has been demonstrated to be an independent predictor of TCC recurrence and of survival when assessed with tumor grade, tumor stage, lymph node status, and p53 status: patients with p53-altered/p21-negative TCC demonstrated a higher rate of recurrence and worse survival when compared with those with p53-altered/p21-positive tumors [62]. Those results also suggest that maintenance of p21WAF/CIP1 expression appears to abrogate the deleterious effects of TP53 alterations on TCC progression. Telomerase activation may be a critical step in TCC pathogenesis.[33, 36, 71]. Unexpectedly, no significant correlation was observed between levels of telomerase

expression and the clinicopathologic features of the tumors, including clinical stage, pathologic grade, tumor multiplicity, and status of recurrence [33]. However, proliferating human urothelial cells in tissue culture reveal telomerase activity, which is readily detected usually at lower levels than in TCC, and telomeres did not shorten [3]. Notably, telomerase activity was relatively low or undetectable in non-proliferating cultures. These data support a model in which the detection of telomerase in TCC biopsies reflects differences in proliferation between tumor and normal cells in vivo [3]. In conclusion, the molecular analysis of bladder TCC has improved the classification and our knowledge on the pathogenesis of these neoplasms. Two main molecular pathways have been clearly demonstrated, which closely correlate with the morphological appearance. These are the inactivation of cyclin-dependent kinase inhibitors in low-grade TCC and early TP53-mediated abnormalities in high-grade TCC. However, the heterogeneity associated with advanced neoplasms needs a careful correlation of genetic and kinetic features to understand cellular progression and tumor cell selection in TCC. This combined analysis becomes more powerful when coupled with reliable topographic features, because it will allow the identification of aggressive tumor cell clones susceptible of genetic therapy.

References 1. Aaltonen V, Bostrom PJ, Soderstrom KO, Hirvonen O, Tuukkanen J, Nurmi M, Laato M, Peltonen J (1999) Urinary bladder transitional cell carcinogenesis is associated with down-regulation of NF1 tumor suppressor gene in vivo and in vitro. Am J Pathol 154:755–765 2. Baithun S, Daruwala P, Oliver RTD (1998) Squamous change in bladder cancer and its relevance to understanding clonal evolution in development of bladder cancer. Cancer Surveys 31:17–27 3. Belair CD, Yeager TR, Lopez PM, Reznikoff CA (1997) Telomerase activity: a biomarker of cell proliferation, not malignant transformation. Proc Natl Acad Sci USA 94:13677– 13682 4. Blanes A, Rubio J, Garcia I, Sanchez-Carrillo JJ, Matilla A, Diaz-Cano SJ (1999) Genetic changes of cell cycle regulators define topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder (abstract). J Pathol 187:12 5. Brown JL, Russell PJ, Philips J, Wotherspoon J, Raghavan D (1990) Clonal analysis of a bladder cancer cell line: an experimental model of tumour heterogeneity. Br J Cancer 61:369–376 6. Cairns P, Proctor AJ, Knowles MA (1991) Loss of heterozygosity at the RB locus is frequent and correlates with muscle invasion in bladder carcinoma. Oncogene 6:2305–2309 7. Califano J, van der Riet P, Westra W, Nawroz H, Clayman G, Piantadosi S, Corio R, Lee D, Greenberg B, Koch W, Sidransky D (1996) Genetic progression model for head and neck cancer: implications for field cancerization. Cancer Res 56: 2488–2492 8. Cappellen D, Gil Diez de Medina S, Chopin D, Thiery JP, Radvanyi F (1997) Frequent loss of heterozygosity on chromosome 10q in muscle-invasive transitional cell carcinomas of the bladder. Oncogene 14:3059–3066 9. Cheng L, Cheville JC, Neumann RM, Bostwick DG (1999) Natural history of urothelial dysplasia of the bladder. Am J Surg Pathol 23:443–447

296 10. Cordon-Cardo C (1995) Mutations of cell cycle regulators. Biological and clinical implications for human neoplasia. Am J Pathol 147:545–560 11. Cordon-Cardo C, Reuter VE (1997) Alterations of tumor suppressor genes in bladder cancer. Semin Diagn Pathol 14: 123–132 12. Cordon-Cardo C, Zhang ZF, Dalbagni G, Drobnjak M, Charytonowicz E, Hu SX, Xu HJ, Reuter VE, Benedict WF (1997) Cooperative effects of p53 and pRB alterations in primary superficial bladder tumors. Cancer Res 57:1217–1221 13. Cote RJ, Dunn MD, Chatterjee SJ, Stein JP, Shi SR, Tran QC, Hu SX, Xu HJ, Groshen S, Taylor CR, Skinner DG, Benedict WF (1998) Elevated and absent pRb expression is associated with bladder cancer progression and has cooperative effects with p53. Cancer Res 58:1090–1094 14. Deng G, Lu Y, Zlotnikov G, Thor AD, Smith HS (1996) Loss of heterozygosity in normal tissue adjacent to breast carcinomas. Science 274:2057–2059 15. Diaz-Cano SJ (1998) Clonality studies in the analysis of adrenal medullary proliferations: application principles and limitations. Endocr Pathol 9:301–316 16. Diaz-Cano SJ, Blanes A, Rubio J, Matilla A, Wolfe HJ (2000) Molecular evolution and intratumor heterogeneity by topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder. Lab Invest 80:279–289 17. Diaz-Cano SJ, Blanes A, Wolfe HJ (2000) PCR-based techniques for clonality analysis of neoplastic progression. Bases for its appropriate application in paraffin-embedded tissues. Diagn Mol Pathol (in press) 18. Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ (2000) Clonality as expression of distinctive cell kinetic patterns in nodular hyperplasias and adenomas of the adrenal cortex. Am J Pathol 156:311–319 19. Epstein JI, Amin MB, Reuter VR, Mostofi FK (1998) The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference Committee. Am J Surg Pathol 22:1435–1448 20. Esrig D, Spruck CHd, Nichols PW, Chaiwun B, Steven K, Groshen S, Chen SC, Skinner DG, Jones PA, Cote RJ (1993) p53 nuclear protein accumulation correlates with mutations in the p53 gene, tumor grade, and stage in bladder cancer. Am J Pathol 143:1389–1397 21. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, Bloomfield CD, Lander ES (1999) Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286:531–537 22. Haas-Kogan DA, Kogan SC, Levi D, Dazin P, T’Ang A, Fung YK, Israel MA (1995) Inhibition of apoptosis by the retinoblastoma gene product. Embo J 14:461–472 23. Habuchi T, Ogawa O, Kakehi Y, Ogura K, Koshiba M, Sugiyama T, Yoshida O (1992) Allelic loss of chromosome 17p in urothelial cancer: strong association with invasive phenotype. J Urol 148:1595–1599 24. Habuchi T, Ogawa O, Kakehi Y, Ogura K, Koshiba M, Hamazaki S, Takahashi R, Sugiyama T, Yoshida O (1993) Accumulated allelic losses in the development of invasive urothelial cancer. Int J Cancer 53:579–584 25. Habuchi T, Devlin J, Elder PA, Knowles MA (1995) Detailed deletion mapping of chromosome 9q in bladder cancer: evidence for two tumour suppressor loci. Oncogene 11:1671– 1674 26. Habuchi T, Yoshida O, Knowles MA (1997) A novel candidate tumour suppressor locus at 9q32–33 in bladder cancer: localization of the candidate region within a single 840 kb YAC. Hum Mol Genet 6:913–919 27. Hartmann A, Moser K, Kriegmair M, Hofstetter A, Hofstaedter F, Knuechel R (1999) Frequent genetic alterations in simple urothelial hyperplasias of the bladder in patients with papillary urothelial carcinoma. Am J Pathol 154:721–727 28. Hierro Martin I, Alvarez Perez M, Blanes Berenguel A, Quinonero Diaz A, Diaz Cano S, Garcia Gonzalez I, Matilla

29.

30. 31.

32. 33. 34. 35.

36.

37. 38.

39.

40.

41.

42.

43. 44. 45.

46.

47.

Vicente A (1999) Contribucion a la categorizacion de los CCT de vejiga, sistematizacion de la gradacion y utilidad de la ploidia de ADN. Actas Urol Esp 23:489–496 Horowitz JM, Park SH, Bogenmann E, Cheng JC, Yandell DW, Kaye FJ, Minna JD, Dryja TP, Weinberg RA (1990) Frequent inactivation of the retinoblastoma anti-oncogene is restricted to a subset of human tumor cells. Proc Natl Acad Sci USA 87:2775–2779 Kageyama Y, Yamamura Y, Oshima H, Ikawa Y (1995) Infrequent mutations of the WT1 gene in primary cancers of the adult urinary tract. Jpn J Clin Oncol 25:173–178 King ED, Matteson J, Jacobs SC, Kyprianou N (1996) Incidence of apoptosis, cell proliferation and bcl-2 expression in transitional cell carcinoma of the bladder: association with tumor progression. J Urol 155:316–320 Kroft SH, Oyasu R (1994) Urinary bladder cancer: mechanisms of development and progression. Lab Invest 71:158– 174 Kyo S, Kunimi K, Uchibayashi T, Namiki M, Inoue M (1997) Telomerase activity in human urothelial tumors. Am J Clin Pathol 107:555–560 Lara PC, Perez S, Rey A, Santana C (1999) Apoptosis in carcinoma of the bladder: relation with radiation treatment results. Int J Radiat Oncol Biol Phys 43:1015–1019 Lianes P, Orlow I, Zhang ZF, Oliva MR, Sarkis AS, Reuter VE, Cordon-Cardo C (1994) Altered patterns of MDM2 and TP53 expression in human bladder cancer. J Natl Cancer Inst 86:1325–1330 Linn JF, Lango M, Halachmi S, Schoenberg MP, Sidransky D (1997) Microsatellite analysis and telomerase activity in archived tissue and urine samples of bladder cancer patients. Int J Cancer 74:625–629 Lipponen PK, Aaltomaa S (1994) Apoptosis in bladder cancer as related to standard prognostic factors and prognosis. J Pathol 173:333–339 Lipponen PK, Liukkonen TJ (1995) Reduced expression of retinoblastoma (Rb) gene protein is related to cell proliferation and prognosis in transitional-cell bladder cancer. J Cancer Res Clin Oncol 121:44–50 Mao L, Lee DJ, Tockman MS, Erozan YS, Askin F, Sidransky D (1994) Microsatellite alterations as clonal markers for the detection of human cancer. Proc Natl Acad Sci USA 91: 9871–9875 Moch H, Sauter G, Mihatsch MJ, Gudat F, Epper R, Waldman FM (1994) p53 but not erbB-2 expression is associated with rapid tumor proliferation in urinary bladder cancer. Hum Pathol 25:1346–1351 Nakopoulou L, Constantinides C, Papandropoulos J, Theodoropoulos G, Tzonou A, Giannopoulos A, Zervas A, Dimopoulos C (1995) Evaluation of overexpression of p53 tumor suppressor protein in superficial and invasive transitional cell bladder cancer: comparison with DNA ploidy. Urology 46:334–340 Niehans GA, Kratzke RA, Frober g MK, Aeppli DM, Nguyen PL, Geradts J (1999) G1 checkpoint protein and p53 abnormalities occur in most invasive transitional cell carcinomas of the urinary bladder. Br J Cancer 80:1175–1184 Nordenson I, Ljungberg B, Roos G (1988) Chromosomes in renal carcinoma with reference to intratumor heterogeneity. Cancer Genet Cytogenet 32:35–41 Nowell PC (1976) The clonal evolution of tumor cell populations. Science 194:23–28 Orlow I, Lacombe L, Hannon GJ, Serrano M, Pellicer I, Dalbagni G, Reuter VE, Zhang ZF, Beach D, Cordon-Cardo C (1995) Deletion of the p16 and p15 genes in human bladder tumors. J Natl Cancer Inst 87:1524–1529 Poddighe PJ, Ramaekers FC, Smeets AW, Vooijs GP, Hopman AH (1992) Structural chromosome 1 aberrations in transitional cell carcinoma of the bladder: interphase cytogenetics combining a centromeric, telomeric, and library DNA probe. Cancer Res 52:4929–4934 Pomerantz J, Schreiber-Agus N, Liegeois NJ, Silverman A, Alland L, Chin L, Potes J, Chen K, Orlow I, Lee HW, CordonCardo C, DePinho RA (1998) The Ink4a tumor suppressor

297

48.

49.

50.

51.

52. 53. 54.

55.

56.

57. 58.

59.

gene product, p19Arf, interacts with MDM2 and neutralizes MDM2’s inhibition of p53. Cell 92:713–723 Popov Z, Hoznek A, Colombel M, Bastuji-Garin S, LefrereBelda MA, Bellot J, Abboh CC, Mazerolles C, Chopin DK (1997) The prognostic value of p53 nuclear overexpression and MIB-1 as a proliferative marker in transitional cell carcinoma of the bladder. Cancer 80:1472–1481 Rabbani F, Richon VM, Orlow I, Lu ML, Drobnjak M, Dudas M, Charytonowicz E, Dalbagni G, Cordon-Cardo C (1999) Prognostic significance of transcription factor E2F-1 in bladder cancer: genotypic and phenotypic characterization. J Natl Cancer Inst 91:874–881 Reeder JE, O’Connell MJ, Yang Z, Morreale JF, Collins L, Frank IN, Messing EM, Cockett AT, Cox C, Robinson RD, Wheeless LL (1998) DNA cytometry and chromosome 9 aberrations by fluorescence in situ hybridization of irrigation specimens from bladder cancer patients. Urology 51:58–61 Reznikoff CA, Belair CD, Yeager TR, Savelieva E, Blelloch RH, Puthenveettil JA, Cuthill S (1996) A molecular genetic model of human bladder cancer pathogenesis. Semin Oncol 23:571–584 Rosin MP, Cairns P, Epstein JI, Schoenberg MP, Sidransky D (1995) Partial allelotype of carcinoma in situ of the human bladder. Cancer Res 55:5213–5216 Ruppert JM, Tokino K, Sidransky D (1993) Evidence for two bladder cancer suppressor loci on human chromosome 9. Cancer Res 53:5093–5095 Sarkis AS, Dalbagni G, Cordon-Cardo C, Zhang ZF, Sheinfeld J, Fair WR, Herr HW, Reuter VE (1993) Nuclear overexpression of p53 protein in transitional cell bladder carcinoma: a marker for disease progression. J Natl Cancer Inst 85:53–59 Sarkis AS, Dalbagni G, Cordon-Cardo C, Melamed J, Zhang ZF, Sheinfeld J, Fair WR, Herr HW, Reuter VE (1994) Association of P53 nuclear overexpression and tumor progression in carcinoma in situ of the bladder. J Urol 152:388–392 Schapers R, Smeets W, Hopman A, Pauwels R, Geraedts J, Ramaekers F (1993) Heterogeneity in bladder cancer as detected by conventional chromosome analysis and interphase cytogenetics. Cancer Genet Cytogenet 70:56–61 Shipman R, Schraml P, Colombi M, Raefle G, Ludwig CU (1993) Loss of heterozygosity on chromosome 11p13 in primary bladder carcinoma. Hum Genet 91:455–458 Sidransky D, Von Eschenbach A, Tsai YC, Jones P, Summerhayes I, Marshall F, Paul M, Green P, Hamilton SR, Frost P, Vogelstein B (1991) Identification of p53 gene mutations in bladder cancers and urine samples. Science 252:706–709 Sidransky D, Frost P, Von Eschenbach A, Oyasu R, Preisinger AC, Vogelstein B (1992) Clonal origin bladder cancer. N Engl J Med 326:737–740

60. Soini Y, Turpeenniemi-Hujanen T, Kamel D, Autio-Harmainen H, Risteli J, Risteli L, Nuorva K, Paakko P, Vahakangas K (1993) p53 immunohistochemistry in transitional cell carcinoma and dysplasia of the urinary bladder correlates with disease progression. Br J Cancer 68:1029–1035 61. Spruck CH 3rd, Ohneseit PF, Gonzalez-Zulueta M, Esrig D, Miyao N, Tsai YC, Lerner SP, Schmutte C, Yang AS, Cote R, et al. (1994) Two molecular pathways to transitional cell carcinoma of the bladder. Cancer Res 54:784–788 62. Stein JP, Ginsberg DA, Grossfeld GD, Chatterjee SJ, Esrig D, Dickinson MG, Groshen S, Taylor CR, Jones PA, Skinner DG, Cote RJ (1998) Effect of p21WAF1/CIP1 expression on tumor progression in bladder cancer. J Natl Cancer Inst 90:1072– 1079 63. Takahashi T, Habuchi T, Kakehi Y, Mitsumori K, Akao T, Terachi T, Yoshida O (1998) Clonal and chronological genetic analysis of multifocal cancers of the bladder and upper urinary tract. Cancer Res 58:5835–5841 64. Tetu B, Fradet Y, Allard P, Veilleux C, Roberge N, Bernard P (1996) Prevalence and clinical significance of HER/2neu, p53 and Rb expression in primary superficial bladder cancer. J Urol 155:1784–1788 65. Uchida T, Wada C, Ishida H, Egawa S, Ao T, Yokoyama E, Koshiba K (1995) Infrequent involvement of mutations on neurofibromatosis type 1, H-ras, K-ras and N-ras in urothelial tumors. Urol Int 55:63–67 66. Vogelstein B, Fearon ER, Kern SE, Hamilton SR, Preisinger AC, Nakamura Y, White R (1989) Allelotype of colorectal carcinomas. Science 244:207–211 67. Voorter C, Joos S, Bringuier PP, Vallinga M, Poddighe P, Schalken J, du Manoir S, Ramaekers F, Lichter P, Hopman A (1995) Detection of chromosomal imbalances in transitional cell carcinoma of the bladder by comparative genomic hybridization. Am J Pathol 146:1341–1354 68. Watters AD, Ballantyne SA, Going JJ, Grigor KM, Bartlett JM (2000) Aneusomy of chromosomes 7 and 17 predicts the recurrence of transitional cell carcinoma of the urinary bladder. BJU Int 85:42–47 69. Wolman SR, Heppner GH (1992) Genetic heterogeneity in breast cancer. J Natl Cancer Inst 84:469–470 70. Xu HJ, Cairns P, Hu SX, Knowles MA, Benedict WF (1993) Loss of RB protein expression in primary bladder cancer correlates with loss of heterozygosity at the RB locus and tumor progression. Int J Cancer 53:781–784 71. Yoshida K, Sugino T, Tahara H, Woodman A, Bolodeoku J, Nargund V, Fellows G, Goodison S, Tahara E, Tarin D (1997) Telomerase activity in bladder carcinoma and its implication for noninvasive diagnosis by detection of exfoliated cancer cells in urine. Cancer 79:362–369

Vol. 80, No. 3, p. 279, 2000 Printed in U.S.A.

ARTICLES Molecular Evolution and Intratumor Heterogeneity by Topographic Compartments in Muscle-Invasive Transitional Cell Carcinoma of the Urinary Bladder Salvador J. Diaz-Cano, Alfredo Blanes, Javier Rubio, Alfredo Matilla, and Hubert J. Wolfe Department of Pathology (SJD-C), St Bartholomew’s and the Royal London School of Medicine and Dentistry, London, United Kingdom; Tufts University-New England Medical Center (SJD-C, HJW), Boston, Massachusetts; and the University Hospital of Ma´laga (AB, JR, AM), Ma´laga, Spain SUMMARY: Superficial transitional cell carcinomas (TCC) of the urinary bladder have been shown to be monoclonal. However, no combined study of clonality and tumor suppressor genes (TSG) is available to date for muscle-invasive TCC. Forty-four muscle-invasive TCC of the urinary bladder selected from women were included in this study. Tumor cells located above and below the muscularis mucosa zone were systematically microdissected and used for DNA extraction. Hha-I digested and undigested samples were used to study the methylation pattern of androgen receptor alleles and undigested samples were used for microsatellite analysis of TSG (TP53, RB1, WT1, and NF1). Both loss of heterozygosity (LOH) and single nucleotide polymorphism (SNP) analyses were performed using optimized denaturing gradient gel electrophoresis. The expression of p53, pRB, and p21WAF1 was assessed by immunohistochemistry. Appropriate controls were run in every case. All except two TCC showed a monoclonal pattern with the same allele inactivated in both compartments. Microsatellite analysis of TSG revealed the same LOH/SNP pattern in both tumor compartments in 30 cases (involving more than 1 TSG locus in 8) and genetic heterogeneity in 14 cases. From the latter group, 9 cases expressed more genetic changes in the deep compartment (involving TP53 gene in all cases, WT1 gene in 2, and NF1 in 1), whereas in 4 cases the superficial compartment showed more genetic changes (three involving NF1 and one involving both RB and TP53). No statistical difference in the immunoexpression was detected, although it tended to be higher in the superficial compartment than in the deep compartment. These concordant data in polymorphic DNA regions indicate that bladder-muscle-invasive TCC are monoclonal proliferations with homogeneous tumor cell selection. Heterogeneous tumor cell selection by topography defined two different genetic compartments: superficial, NF1-defective, and deep, TP53-defective. No differences in the immunohistochemical expression were observed, precluding a more extensive clinical application. (Lab Invest 2000, 80:279–289).

uperficial cancer of the urinary bladder often presents as multiple tumors, appearing at different times and at different sites in the bladder. This observation has been attributed to a “field defect” in the bladder that allows the independent transformation of epithelial cells at a number of sites. Previous studies have shown the same X chromosome inactivated in all tumors from a single patient, whereas normal bladder mucosa cells had random patterns of inactivation. Moreover, each tumor that could be evaluated from a given patient had lost the same allele on

Received July 20, 1999. Presented in part as abstract at the USCAP Meetings in Boston, Massachusetts, 1998 and San Francisco, California, 1999. Address reprint requests to: Dr. S. J. Diaz-Cano, Department of Histopathology and Morbid Anatomy, The Royal London Hospital, Whitechapel, London E1 1BB, United Kingdom. Fax: 44 171 377 7030; E-mail: s.j.diaz-cano@mds.qmw.ac.uk

chromosome 9q, but with heterogeneous losses of chromosomes 17p and 18q alleles (Sidransky et al, 1992). Point mutations, or single nucleotide polymorphism if located in introns, within all cells imply a common progenitor contributing that mutation (Knudson, 1995; Nowell, 1976), and have been found associated with loss of heterozygosity (LOH) of certain loci (Sternlicht et al, 1994). Its demonstration applies only to those cases carrying that marker and fails to identify clonal proliferations occurring before the creation of a specific genetic lesion (Diaz-Cano et al, in press; Sternlicht et al, 1994). Nevertheless, certain genetic markers can be used to test clonal expansions within a tumor cell sample (Diaz-Cano, 1998; Diaz-Cano et al, in press). The LOH of a given genetic marker should be linked to loss of tumor suppressor genes (TSG) by DNA deletions, one of the key components of Knudson’s hypothesis (Knudson, 1995). The TP53/RB mutation status of recurrent bladder cancers has comLaboratory Investigation • March 2000 • Volume 80 • Number 3

279

Diaz-Cano et al

pletely matched their corresponding primary bladder cancer, suggesting a monoclonal origin of recurrent superficial bladder cancer (Chern et al, 1996). Histologic heterogeneity is well documented in transitional cell carcinoma (TCC), even for xenograft tumors that have been reported to express transitional, squamous, and glandular elements in different clones. The DNA content and lectin binding profiles of the clones also reflects the heterogeneity of the line (Brown et al, 1990). Cytogenetic heterogeneity has also been demonstrated by comparative genomic hybridization (Voorter et al, 1995) and, in radiationinduced tumors, the presence of cytogenetically abnormal and unrelated clones has been assumed the result of heterogeneity of the tumor cell population (Fadl-Elmula et al, 1998). The analysis of marker chromosomes in tumors with two or more cultures has shown, besides a primary cytogenetic change, additional clonal abnormalities illustrating intratumoral heterogeneity (Nordenson et al, 1988). All these studies on tumor heterogeneity do not take into consideration the tumor cell topography in the urinary bladder wall. The staging system in urinary bladder tumors has been improved after recognizing the relevance of the muscularis mucosa (Ro et al, 1987). The data show that the extent of lamina propria invasion is a clinically relevant prognostic factor for progression of pT1 TCC of the bladder (Smits et al, 1998; Younes et al, 1990). Tumors extending beyond the muscularis mucosa behave in a way similar to muscle-invasive TCC, especially if they are high-grade, reveal associated carcinoma in situ, or express nuclear TP53 (Hermann et al, 1998; Smits et al, 1998; Younes et al, 1990). However, no combined study on the methylation pattern of androgen receptor alleles and microsatellite (MS) pattern of tumor suppressor genes (TSG) is available to date for muscle-invasive TCC. Likewise, there is no reference about the differences in the molecular profile of tumor cells located above and below the muscularis mucosa. The main aim of this study is to examine the molecular evolution and tumor heterogeneity by topographic compartments in a series of 44 muscle-invasive TCC, considering that tumor cell depth in the bladder wall would express the potential of cellular progression in TCC.

Results Five cases were considered non-informative and were excluded from the clonality analysis because of the unbalanced methylation pattern of the undigested control samples (four cases) or MS instability involving the androgen receptor locus (one case). The remaining 39 cases revealed 37 TCC with a monoclonal pattern and the same X chromosome inactivated in samples from a single patient and 2 TCC with a polyclonal pattern (Fig. 1). The MS analysis of TSG revealed TP53 alterations in 29 of 43 informative cases (67.4%, screening two introns), RB abnormalities in 17 of 33 informative cases (51.5%), WT1 genes lesions in 21 of 38 informative cases (55.3%), and NF1 alterations in 19 of 30

280

Laboratory Investigation â€˘ March 2000 â€˘ Volume 80 â€˘ Number 3

informative cases (63.3%). Fourteen TCC (31.8%) revealed no genetic abnormalities in the TSG introns evaluated and included 3 non-informative cases, 1 case with MS instability, and 2 polyclonal cases in the test of X chromosome inactivation. Eight additional TCC (18.2%) revealed MS alterations involving the same locus in both tumor compartments: TP53 locus in 4 cases (9.1%), WT1 locus in 3 cases (6.8%), and RB1 locus in 1 case (2.3%). Case TCC 7 showed TP53 LOH in the superficial compartment and WT1 LOH in the deep compartment; 3 TCC revealed only one TSG locus in one tumor compartment, either superficial (TP53, 1 case) or deep (TP53, 2 cases). The remaining TCC had two TSG loci altered in 13 cases (29.5%, 7 with superficial-deep concordance), three TSG loci in 4 cases (9.1%, 1 with superficial-deep concordance), and four TSG loci in 1 case (2.3%). Concordant MS patterns of TSG in both tumor compartments were observed in 30 muscle-invasive TCC (68.2%) and topographically related genetic heterogeneity (LOH and/or SNP) in 14 cases (31.8%) (Table 1). In the latter group, 9 TCC (20.5%) expressed more genetic changes in the deep compartment (TP53 gene in all cases, WT1 gene in 2, and NF1 in 1), and 4 cases (9.1%) at the superficial compartment (3 involving NF1 and 1 involving both RB and TP53 genes) (Fig. 2, Table 1). One case (TCC 7) showed a discordant MS pattern of TSG, involving TP53 locus in the superficial compartment and WT1 locus in the deep compartment. The group of muscle-invasive TCC with concordant MS patterns of TSG showed a subset of tumors with demonstrable genetic alterations in at least one TSG locus (16 cases, 36.4%) and another subset with no demonstrable alteration in TSG loci (14 cases, 31.8%). A heterogeneous immunohistochemical expression of cell cycle regulators was observed with high variability of expression from field to field (high standard deviation). The antibodies used in this study detect the normal pRB1 and p21WAF1 proteins, and both normal and abnormal p53 protein (clone DO1), but they were not able to differentiate groups of muscle-invasive TCC or topographic compartments. In any case, the protein expression tended to be higher in the superficial compartment than in the deep compartment (Figs. 3 and 4, Table 2). No statistically significant differences in the immunohistochemical expression of markers were detected by topography or tumor group.

Discussion The concordant MS pattern of TSG and the inactivation of the same X chromosome in both superficial and deep compartments of muscle-invasive TCC support a monoclonal origin and a homogeneous selection of tumor cell throughout the neoplasm. In approximately 70% of muscle-invasive TCC in this series, each patient had inactivation of the same X chromosome and concordant MS pattern of TSG in both tumor compartments; in all these cases, normal bladder mucosa cells had random X-inactivation patterns and retained the constitutional heterozygosity. On the

Intratumor Heterogeneity in Bladder TCC

Figure 1. Example of tumor cell and control sampling in muscle-invasive transitional cell carcinoma (TCC) of the urinary bladder (top panel) and clonality results (bottom panel). Note the unbalanced methylation pattern of androgen receptor alleles in the superficial tumor compartment (Sup TCC) and carcinoma in situ (CIS). The balanced pattern in the deep compartment (Deep TCC) was related to inflammatory cell contamination. The intraurothelial lesions are not part of this article. Uroth, histologically normal urothelium; LGD, low-grade dysplasia; SqM, squamous metaplasia; SmM, smooth muscle.

other hand, tumor cell heterogeneity by topographic compartments was demonstrated in 14 of 44 muscleinvasive TCC (31.8%). Three main groups of molecular evolution could be drawn in muscle-invasive TCC: two in the first genetically homogeneous group and one for the heterogeneous group.

We found coexistent genetic abnormalities involving two or more TSG loci in the first group of muscleinvasive TCC (16 of 44 cases, 36.4%). The background level of LOH in normal tissues has been reported between 4% and 20%, regardless of the detection system used (Chen et al, 1992; Deng et al, Laboratory Investigation • March 2000 • Volume 80 • Number 3

281

Diaz-Cano et al

Table 1. Muscle-Invasive Transitional Cell Carcinoma of the Urinary Bladder with Discordant Genetic Abnormalities in the Superficial and Deep Compartments Case/Sample TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC TCC

1-Sup. 1-Deep 7-Sup. 7-Deep 10-Sup. 10-Deep 13-Sup. 13-Deep 16-Sup. 16-Deep 21-Sup. 21-Deep 23-Sup. 23-Deep 26-Sup. 26-Deep 30-Sup. 30-Deep 31-Sup. 31-Deep 35-Sup. 35-Deep 37-Sup. 37-Deep 43-Sup. 43-Deep 44-Sup. 44-Deep

Methylation of AR Alleles*

TP53–1‡

TP53–2‡

RB1‡

WT1‡

NF1‡

Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced Unbalanced MSI MSI Unbalanced Unbalanced NI NI Balanced Unbalanced Unbalanced Unbalanced Unbalanced Balanced Unbalanced Unbalanced Unbalanced Unbalanced

LOH–S SNP–SL LOH–L ROH SNP–S SNP–S LOH–L LOH–L ROH ROH ROH LOH–L ROH ROH ROH LOH–S LOH–L ROH ROH/SNP–SL LOH/SNP–S ROH LOH–S ROH LOH–S ROH ROH ROH ROH

NI NI ROH ROH ROH ROH ROH ROH LOH–L LOH–L NI NI ROH LOH–L ROH LOH–S NI NI ROH ROH ROH ROH ROH ROH ROH LOH–L ROH LOH–S

ROH ROH ROH ROH LOH–S ROH LOH–S LOH–S ROH ROH ROH ROH NI NI — — ROH ROH NI NI NI NI LOH–S LOH–S ROH ROH ROH ROH

LOH–S LOH–S ROH LOH–L ROH ROH LOH–L SNP–SL ROH SNP SNP–SL SNP–SL ROH LOH–L ROH ROH ROH ROH ROH LOH–L ROH ROH ROH ROH ROH ROH ROH ROH

LOH–L ROH ROH ROH SNP–S ROH LOH–L ROH ROH ROH NI NI ROH LOH–S ROH ROH — — NI NI ROH ROH ROH ROH ROH ROH ROH ROH

* AR, androgen receptor; NI, non informative; MSI, microsatellite instability. ‡ Tumor suppressor genes were evaluated for the presence of loss of heterozygosity (LOH), retention of heterozygosity (ROH), and single nucleotide polymorphism (SNP) in the small (S) and large (L) alleles. NI, noninformative.

1996; Sager, 1989; Wolman and Heppner, 1992). Similar LOH frequency must be assumed as background in the evaluation of tumor tissues (Sager, 1989). Considering the worst-case scenario of all genetic lesions being equally important and frequent (Diaz-Cano and Wolfe, 1997), the probability of randomly finding coexisting genetic alterations in normal tissues would be 0.22 ⫽ 4.0 ⫻ 10⫺2 for two genetic loci, 0.23 ⫽ 8.0 ⫻ 10⫺3 for three genetic loci, and so on. Applying this principle to two separately microdissected samples (superficial and deep) from any given single tumor, the probability of having the same locus involved in both samples would be (0.21)2 ⫽ 4.0 ⫻ 10⫺2 for one genetic locus, (0.22)2 ⫽ 1.6 ⫻ 10⫺3 for two genetic loci, or (0.23)2 ⫽ 6.4 ⫻ 10⫺5 for three genetic loci. Under these circumstances, the probability of randomly finding concordant genetic abnormalities in both superficial and deep compartments in 16 patients (8 involving one TSG locus, 7 involving two TSG loci, and 1 involving three TSG loci) would be [(0.21)2]8 [(0.22)2]7 [(0.23)2]1 ⫽ 1.13 ⫻ 10⫺35. That result strongly supports the monoclonal origin and a homogeneous selection of tumor cells throughout the neoplasm. On the other hand, it also suggests that this

282

Laboratory Investigation • March 2000 • Volume 80 • Number 3

subset of muscle-invasive TCC can arise from the uncontrolled spread of a single transformed cell and evolve through a multistep tumorigenesis involving those “common” TSG. In general, the accumulation of genetic lesions in TSG supports a monoclonal origin of tumors (DiazCano, 1998). Similar findings have been reported in coexistent superficial bladder TCC, but comparing different tumors instead of tumor compartments (Sidransky et al, 1992). Muscle-invasive TCC represent advanced neoplasms, and an accumulation of genetic abnormalities can be expected when compared with reference TCC, if we consider that accumulation of genetic alterations is an expression of molecular progression. It would help to explain the relatively high incidence of genetic abnormalities found in RB1 and WT1 loci. Horowitz et al (1990) found inactivated pRB in one-third of randomly selected bladder cancers, compared with our 51.5% incidence of RB1 abnormalities. Kageyama et al (1995) did not find WT1 mutations in any of 11 randomly selected bladder cancers using polymerase chain reactionsingle strand conformation polymorphism analysis and restriction fragment length polymorphism analysis

Intratumor Heterogeneity in Bladder TCC

Figure 2. Comparison of allele patterns of tumor suppressor genes in the superficial (Sup. TCC) and deep (Deep TCC) compartments from two examples of muscle-invasive transitional cell carcinoma (TCC) of the urinary bladder. Deep compartments show more genetic abnormalities in tumor suppressor gene introns than superficial compartments involving especially TP53 (arrows in lane 1 of panel a and lane 2 of panel b point to allelic imbalance as demonstrated in the densitometric analysis). Panel a also shows single nucleotide polymorphism of NF1 in the superficial compartment, and panel b shows allelic imbalance of WT1 in the deep compartment. Lanes 1, TP53(1); 2, TP53(2); 3, RB1; 4, WT1; 5, NF1.

of the WT1 locus. In addition, they only performed exon analysis for point mutation and no LOH analysis or intron analysis was carried out. The second group of muscle-invasive TCC also showed homogeneous cell selection by tumor compartment but no demonstrable genetic alteration in the TSG introns evaluated. Fourteen muscle-invasive TCC (31.8%) did not reveal intron deletions or single nucleotide polymorphisms, suggesting a reduced incidence of genetic damage in those TSG loci. Some alternative explanations could be offered for those cases. (A) Genetic alterations below the detection threshold could explain that result. We previously optimized the denaturing gradient gel electrophoresis protocol with appropriate controls, including positive, negative, and sensitivity (not shown); the progressive dilution of a known positive case in a background of germline DNA gave us a sensitivity threshold of 1% for positive detection. We systematically microdissected at least 100 cells from each tumor compartment. With a sensitivity threshold of 1%, the technique would miss the positive results from the DNA equivalent of less than one cell in the sampling, which is probably clinically irrelevant. Potential reasons could be normal tissue contamination, which could be excluded by repeated microdissection with the same results. (B) Genetic alterations in those TSG could be present outside of the screened introns. Even in that situation, a statistical approach would prove that it is unlikely to find that association randomly, confirming

that this group is a true subset of muscle-invasive TCC. If we consider 0.8 to be the probability of finding no-LOH for a given marker [p(no-LOH) ⫽ 1 ⫺ p(LOH) ⫽ 1 ⫺ 0.2 ⫽ 0.8], then p(no-LOH) for five markers from a single sample would be 0.85 ⫽ 0.33. We microdissected from the superficial and deep areas of TCC; the p(no-LOH) for two samples would be (0.85)2 ⫽ 0.11. The probability that such consistent results would be randomly found in 14 patients would be [(0.85)2]14 ⫽ 2.71 ⫻ 10⫺14. Supporting that hypothesis, preliminary results of the histopathologic evaluation of those tumors revealed a particular growth pattern in the deep compartment, predominantly single-file infiltration (not shown). We can speculate that the malignant transformation of those tumors should target genes involved in cellular adhesion, thus disturbing the growth pattern. This particular aspect would need additional investigation. The last subset of muscle-invasive TCC showed intratumor heterogeneity in the selection of tumor cells by topographic compartments (14 cases, 31.8%). After malignant transformation, tumor cells can grow independently with variable subsequent genetic alterations in each tumor compartment, explaining the heterogeneity. This group of tumors would also result from the accumulation of genetic damage like the first subgroup. The number of cases is not large, but two processes of tumor cell selection seem to involve different TSG and be responsible for that topographic Laboratory Investigation • March 2000 • Volume 80 • Number 3

283

Diaz-Cano et al

Figure 3. The immunohistochemical expression of p53, p21WAF1, and pRb was more extensive in the superficial compartment of muscle-invasive transitional cell carcinoma (arrows) than in the deep compartment. Panel a shows the same field stained with hematoxylin-eosin.

284

Laboratory Investigation • March 2000 • Volume 80 • Number 3

Intratumor Heterogeneity in Bladder TCC

Table 2. Immunohistochemical Expression of p53, p21WAF1, and pRB1 by Topographic Compartments in Muscle-Invasive Transitional Cell Carcinoma of the Urinary Bladder Superficial compartment Deep compartment (Av. ⫾ SD)* (Av. ⫾ SD)* 60.36 ⫾ 27.32 42.13 ⫾ 21.38 52.25 ⫾ 19.95

p53 (%) p21WAF1 (%) pRB (%) * Av, average;

Figure 4. Immunohistochemical expression of p53, p21WAF1, and pRb in the deep compartment of muscle-invasive transitional cell carcinoma (arrows) from the case shown in Figure 3.

SD,

52.49 ⫾ 26.94 35.95 ⫾ 15.56 47.59 ⫾ 14.76

standard deviation.

heterogeneity. The differential TSG most frequently altered was NF1 in the superficial compartment (75% of cases), and TP53 in the deep compartment (all cases). The NF1 gene has been rarely implicated in bladder carcinogenesis, and no mutations have been observed in a series of 31 bladder cancers studied by Uchida et al (1995). However, the presence of NF1 gene abnormalities in 63.3% of muscle-invasive TCC in our series (19 of 30 informative cases) supports an important role for NF1 in this malignant pathway, especially in the superficial tumor compartment. NF1 gene product has an effect on ras inhibition, protein expressed with the highest levels in immature and proliferating cells (Furth et al, 1987). The absence of the NF1 inhibitory effect will favor increased cell proliferation, as found in the superficial tumor compartments of muscle-invasive TCC (Blanes et al, 1999). Likewise, Aaltonen et al (1999) have reported a decreased NF1 mRNA and protein levels in high-grade TCC, suggesting that alterations in NF1 gene expression may be involved in bladder carcinogenesis. On the other hand, we found TP53 abnormalities in 29 of 43 informative cases (67.4%), after screening two introns. This proportion is similar to that originally reported by Sidransky et al (1991) who found TP53 alterations in 11 of 18 invasive TCC, associated with 17p allelic deletions in all but 1 case, and leaving cells with only mutant forms of the p53 gene product. TP53 abnormalities tended to concentrate in the deep tumor compartment, suggesting it can represent the consequence of tumor cell selection. Similarly, the loss of chromosome 17p has been reported as a late event in tumor progression in superficial TCC (Sidransky et al, 1992). In conclusion, three main groups of molecular evolution could be drawn in muscle-invasive TCC of the urinary bladder. The first two groups result in homogeneous cell selection in the tumor progression, expressed by concordant patterns of TSG microsatellite and inactivation of the same X chromosome, but with different molecular pathways. In one group, the progression takes place through the accumulation of genetic lesions in TSG, and in the other group, probably through a different genetic target. The third group is characterized by topographic tumor heterogeneity and by the accumulation of genetic lesions in TSG. Two genetically different topographic compartments were then apparent: superficial, NF1-defective, and deep, TP53-defective. Laboratory Investigation • March 2000 • Volume 80 • Number 3

285

Diaz-Cano et al

Materials and Methods Case Selection and Sampling We reviewed all muscle-invasive TCC of the urinary bladder (pT2 and pT3) diagnosed in women during a time period of 5 years from three reference hospitals. Forty-four cases had properly preserved archival material available for further analyses, including both tumor and control tissues (histologically normal urothelium, lamina propria, and smooth muscle from the same patient). All surgical specimens were completely embedded for histopathologic diagnosis. Two topographic compartments were analyzed in each TCC regarding their relationship with the muscularis mucosa (MM), revealed by its ectatic vascular plexus (Fig. 1) (Ro et al, 1987). Tumor cells above MM were labeled superficial, whereas those neoplastic cells located below the MM were considered deep. The same areas in consecutive sections were used in each study and their cellular composition was confirmed in adjacent hematoxylineosin stained sections.

Clonality Analysis DNA was extracted from the selected areas using two 20-␮m unstained paraffin sections and a modified phenol-chloroform protocol (Diaz-Cano and Brady, 1997). Appropriate controls were included for each test (histologically normal urothelium, stroma from the lamina propria, and smooth muscle). All samples were divided for restriction endonuclease digestion with Hha-I (New England Biolabs, Beverly, Massachusetts). Half of each sample underwent enzymatic digestion (0.8 unit/␮l); the remaining half was kept as undigested control. Both samples were equally processed, but

excluding Hha-I in the undigested one (Allen et al, 1992; Diaz-Cano et al, in press; Mutter and Boynton, 1995a; Mutter et al, 1995). A mimicker (0.3 ␮g of double stranded and XhoI-linearized ␾X174-RII phage; Gibco-BRL, Gaithersburg, Maryland) was included in each reaction for digestion testing. Complete digestion was checked by gel electrophoresis; incompletely digested samples were repurified and redigested with higher Hha-I concentration. Hha-I was then inactivated by phenol-chloroform extraction (Diaz-Cano and Brady, 1997). DNA was precipitated with ice-cold absolute ethanol in the presence of 0.3 M sodium acetate, pH 5.2, and resuspended in 10 ␮l of PCR buffer (10 mM Tris-HCl pH 8.4, 50 mM KCl, 1.5 mM MgCl2, and 100 ␮g/ml BSA). The hypervariable CAG repeat in the first exon of the human androgen receptor gene (HUMARA) was then amplified using both digested and undigested DNA templates (Mutter and Boynton, 1995a; Mutter et al, 1995). The tests were run in a Perkin-Elmer thermal cycler model 480 (Perkin-Elmer, Norwalk, Connecticut), according to the conditions shown in Table 3. The whole PCR volume (10 ␮l) was subjected to electrophoresis in non-denaturing polyacrylamide gels (8%, 0.75 mm). The gels were run at 5 volt/cm until the xylene cyanol band was within the bottom gel inch. The gels were then fixed with 7% acetic acid (5 minutes), dried under vacuum (40 minutes, 80° C), and put inside a developing cassette containing one intensifying screen and preflashed films (Kodak XAR; Kodak, Rochester, New York) facing the intensifying screen (16 to 48 hours, ⫺70° C). The autoradiograms were developed using an automated processor Kodak-Omat 100 (Kodak). Interpretation and inclusion criteria in each sample were as reported (Diaz-Cano et al, in press; Mutter and

Table 3. Primer Sequences and PCR Cycling Conditions for the Amplification of Polymorphic DNA Regions Primers

Primer sequences

AR-a* AR-b* TP53(1)-a‡ TP53(1)-b‡ TP53(2)-a‡ TP53(2)-b‡ RB1-a‡ RB1-b‡ WT1-a‡ WT1-b‡ NF1-a‡ NF1-b‡

5⬘-CCG AGG AGC TTT CCA GAA TC-3⬘ 5⬘-TAC GAT GGG CTT GGG GAG AA-3⬘ 5⬘-AGG GAT ACT ATT CAG CCC-3⬘ 5⬘-ACT GCC ACT CCT TGC CCC ATT C-3⬘ 5⬘-GAA TCC GGG AGG AGG TTG-3⬘ 5⬘-AAC AGC TCC TTT AAT GGC AG-3⬘ 5⬘-CTC CTC CCC TAC TTA CTT GT-3⬘ 5⬘-AAT TAA CAA GGT GTG GTG GTA CAC G-3⬘ 5⬘-AAT GAG ACT TAC TGG GTG AGG-3⬘ 5⬘-TTA CAC AGT AAT TTC AAG CAA CGG-3⬘ 5⬘-CAG AGC AAG ACC CTG TCT-3⬘ 5⬘-CTC CTA ACA TTT ATT AAC CTT A-3⬘

Tandem repeat/PCR product CAG repeat/215–300 bp CA repeat/103–135 bp AAAAT repeat/140 –175 bp CTTT(T) repeat/266 –306 bp CA repeat/⬃144 bp CA repeat/171–187 bp

All reactions were run in duplicate using 1.5 mM of MgCl2 and 1 ␮l of template. A long denaturation (4 minutes) was used in the first 3 cycles for each set of primers. * The HUMARA tests were run using 0.3 ␮M of each primer and 200 ␮M of each dNTP (including 7-deaza-dGTP instead of dGTP) (Boehringer-Mannheim, Indianapolis, Indiana). The amplicon was internally labeled with 0.3 ␮Ci ␣[32P]-dTTP (800 Ci/mmol, 10 mCi/ml) (New England Nucleotide, Boston, Massachusetts). A “hot start” protocol was also used, completing 28 cycles with an annealing temperature of 55° C. ‡ The polymorphic regions of TSG were amplified using 0.25 ␮M of each primer, 50 ␮M of each dNTP (Boehringer-Mannheim), and internally labeled with 0.3 ␮Ci ␣[32P]-dCTP (3000 Ci/mmol, 10 mCi/ml) (New England Nucleotide). The annealing temperature was 55° C for all primer sets (except for NF1, it was 52° C), and the number of cycles was experimentally optimized to 26.

286

Laboratory Investigation • March 2000 • Volume 80 • Number 3

Intratumor Heterogeneity in Bladder TCC

Boynton, 1995b). Only informative cases (two different alleles in undigested and digested control samples) were included in the final analysis (Diaz-Cano et al, in press; Mutter and Boynton, 1995b; Mutter et al, 1995) and lanes were normalized in relation to the corresponding undigested sample and controls. Allelic imbalance was densitometrically evaluated (EC model 910 optical densitometer; EC Apparatus, St Petersburg, Florida), considering evidence of monoclonality allele ratios ⱖ 4:1 in the normalized digested lanes. The presence of additional allele bands in the tumor samples was considered positive evidence of microsatellite instability if not present in the corresponding control.

LOH/SNP analyses of Tumor Suppressor Genes DNA was extracted from two 20-␮m unstained paraffin sections per tumor compartment and control samples, including at least 100 cells (approximately 0.4 mm2) per sample. Appropriate controls were included for each test (histologically normal urothelium, stroma from the lamina propria, and smooth muscle). DNA was extracted using a modified phenolchloroform protocol (Diaz-Cano and Brady, 1997). DNA was precipitated with ice-cold absolute ethanol in the presence of 0.3 M sodium acetate pH 5.2 and resuspended in 10 ␮l of PCR buffer (10 mM Tris-HCl pH 8.4, 50 mM KCl, 1.5 mM MgCl2, and 100 ␮g/ml BSA). DNA was then used for PCR amplification of polymorphic DNA regions of TSG (TP53, RB, WT1, and NF1), using the primers and conditions shown in Table 3 (Cawkwell et al, 1993, 1994). The tests were run in duplicate in a Perkin-Elmer thermal cycler model 480 (Perkin-Elmer). The whole PCR volume (10 ␮l) was subjected to electrophoresis in 8% denaturing gradient polyacrylamide gels (0.75 mm, 20% to 80% denaturing conditions from top to bottom). The gels were run at 5 volt/cm until the xylene cyanol band was within the bottom gel inch. The gels were then fixed with 7% acetic acid (5 minutes), dried under vacuum (40 minutes, 80° C), and put inside a developing cassette containing one intensifying screen and preflashed films (Kodak XAR) facing the intensifying screen (16 to 48 hours, ⫺70° C). The autoradiograms were developed using an automated processor Kodak-Omat 100. Interpretation and inclusion criteria in each sample were according to Diaz-Cano et al, (in press) and Mutter and Boynton (1995b). Only informative cases (two different alleles in control samples) were included in the final analysis (Diaz-Cano et al, in press; Mutter and Boynton, 1995b; Mutter et al, 1995). Allelic imbalance was densitometrically evaluated (EC model 910 optical densitometer; EC Apparatus). Only allele ratios ⱖ 4:1 in any TSG were considered evidence of loss of heterozygosity (LOH); otherwise retention of heterozygosity (ROH) was assigned. Additional allele bands in the tumor samples were considered positive evidence of single nucleotide polymorphism (SNP) in

denaturing gradient gels if they were not present in the corresponding control.

Immunohistochemical Expression of p53, pRB, and p21WAF1 The sections were mounted on positively charged microscope slides (Superfrost Plus; Fisher Scientific, Fair Lawn, New Jersey) and baked at 60° C for 2 hours. The slides were routinely dewaxed and rehydrated. The endogenous peroxidase activity was then quenched with 0.5% H2O2 in methanol, 10 minutes). A microwave antigen retrieval method (20 minutes in 10 mM citrate buffer, pH 6.0, at 600 watts) was used, followed by incubation with polyclonal horse serum (20 minutes, 1:100 dilution; Dako, Glostrup, Denmark) and with monoclonal primary antibodies (overnight, 4° C), at 2 ␮g/ml for p53 and p21WAF1 and 5 ␮g/ml for pRB1 (Calbiochem, Cambridge, Massachusetts). Then sections were serially incubated with biotinylated antimouse antibody (30 minutes, 1:200 dilution; Dako), and peroxidase-labeled avidin-biotin complex (60 minutes, 1:100 dilution; Dako). All incubations were performed in moist chamber at room temperature unless otherwise specified. The reaction was developed under microscopic control, using 3,3⬘diaminobenzidine tetrahydrochloride with 0.3% H2O2 as chromogen (Sigma Chemical, St. Louis, Missouri), and the sections counterstained with hematoxylin. Both positive (reactive lymph node) and negative (omitting the primary antibody) controls were simultaneously run.

Quantification of Positive Nuclei The threshold of positivity was experimentally established at the positive control in each staining batch. Only those nuclei with staining features similar to those of their corresponding positive control were considered positive for a given marker. Reactivity for each marker was assessed and scored by three independent observers (SDC, AB, and JR). At least 50 high-power fields (HPF), or the complete lesion if smaller (50 HPF ⫽ 7.6 mm2), were screened in each compartment; the screening began in the most cellular area. Both the number of positive nuclei per HPF and the number of neoplastic cells intercepted by the microscope field diameter were registered. The last score was used to estimate the number of neoplastic cells per HPF using the formula N ⫽ (n␲/4)2, where N is the number of estimated cells per HPF and n the number of cells intercepted by the microscope field diameter (Diaz-Cano et al, 1996; Simpson et al, 1992). The number of positive nuclei was always expressed per HPF and per 1,000 proliferating cells. Both the average and the standard deviation (SD) values were calculated as representative scores per compartment and patient.

Statistical Analysis Analysis of variance (ANOVA) and Student t tests were applied to assess the differences, by tumor compartLaboratory Investigation • March 2000 • Volume 80 • Number 3

287

Diaz-Cano et al

ment, of average and SD values of every quantitative variable. Differences were considered statistically significant if p ⬍ 0.05.

References Aaltonen V, Bostrom PJ, Soderstrom KO, Hirvonen O, Tuukkanen J, Nurmi M, Laato M, and Peltonen J (1999). Urinary bladder transitional cell carcinogenesis is associated with down-regulation of NF1 tumor suppressor gene in vivo and in vitro. Am J Pathol 154:755–765. Allen RC, Zoghbi HY, Moseley AB, Rosenblatt HM, and Belmont JW (1992). Methylation of HpaII and HhaI sites near the polymorphic CAG repeat in the human androgenreceptor gene correlates with X chromosome inactivation. Am J Hum Genet 51:1229 –1239. Blanes A, Rubio J, Martinez A, Matilla A, Diaz-Cano SJ, and Wolfe HJ (1999). Kinetic profile by topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder. Lab Invest 79:90A. Brown JL, Russell PJ, Philips J, Wotherspoon J, and Raghavan D (1990). Clonal analysis of a bladder cancer cell line: an experimental model of tumour heterogeneity. Br J Cancer 61:369 –376. Cawkwell L, Bell SM, Lewis FA, Dixon MF, Taylor GR, and Quirke P (1993). Rapid detection of allele loss in colorectal tumours using microsatellites and fluorescent DNA technology. Br J Cancer 67:1262–1267. Cawkwell L, Lewis FA, and Quirke P (1994). Frequency of allele loss of DCC, p53, RBI, WT1, NF1, NM23 and APC/MCC in colorectal cancer assayed by fluorescent multiplex polymerase chain reaction. Br J Cancer 70:813– 818. Chen LC, Kurisu W, Ljung BM, Goldman ES, Moore DI, and Smith HS (1992). Heterogeneity for allelic loss in human breast cancer. J Natl Cancer Inst 84:506 –510. Chern HD, Becich MJ, Persad RA, Romkes M, Smith P, Collins C, Li YH, and Branch RA (1996). Clonal analysis of human recurrent superficial bladder cancer by immunohistochemistry of P53 and retinoblastoma proteins. J Urol 156: 1846 –1849. Deng G, Lu Y, Zlotnikov G, Thor AD, and Smith HS (1996). Loss of heterozygosity in normal tissue adjacent to breast carcinomas. Science 274:2057–2059. Diaz-Cano SJ (1998). Clonality studies in the analysis of adrenal medullary proliferations: Application principles and limitations. Endocr Pathol 9:301–316. Diaz-Cano SJ, Blanes A, and Wolfe HJ (In press, 2000). PCR-based techniques for clonality analysis of neoplastic progression. Bases for its appropriate application in paraffinembedded tissues. Diagn Mol Pathol. Diaz-Cano SJ and Brady SP (1997). DNA extraction from formalin-fixed, paraffin-embedded tissues: protein digestion as a limiting step for retrieval of high-quality DNA. Diagn Mol Pathol 6:342–346. Diaz-Cano SJ, Leon MM, de Miguel M, Galera Davidson H, and Wolfe HJ (1996). Mitotic index quantification: Different approaches and their value in adrenocortical proliferative lesions (Abstract). Lab Invest 74:170A.

288

Laboratory Investigation • March 2000 • Volume 80 • Number 3

Diaz-Cano SJ and Wolfe HJ (1997). Clonality in Kaposi’s sarcoma [letter; comment]. N Engl J Med 337:571–572. Fadl-Elmula I, Bonaldi L, Gorunova L, Mandahl N, Elfving P, and Heim S (1998). Cytogenetic heterogeneity in a second primary radiation-induced bladder carcinoma: ten karyotypically unrelated clones. Cancer Genet Cytogenet 105:134 – 137. Furth ME, Aldrich TH, and Cordon-Cardo C (1987). Expression of ras proto-oncogene proteins in normal human tissues. Oncogene 1:47–58. Hermann GG, Horn T, and Steven K (1998). The influence of the level of lamina propria invasion and the prevalence of p53 nuclear accumulation on survival in stage T1 transitional cell bladder cancer. J Urol 159:91–94. Horowitz JM, Park SH, Bogenmann E, Cheng JC, Yandell DW, Kaye FJ, Minna JD, Dryja TP, and Weinberg RA (1990). Frequent inactivation of the retinoblastoma anti-oncogene is restricted to a subset of human tumor cells. Proc Natl Acad Sci USA 87:2775–2779. Kageyama Y, Yamamura Y, Oshima H, and Ikawa Y (1995). Infrequent mutations of the WT1 gene in primary cancers of the adult urinary tract. Jpn J Clin Oncol 25:173–178. Knudson AG (1995). Mutation and cancer: A personal odyssey. Adv Cancer Res 67:1–23. Mutter GL and Boynton KA (1995a). PCR bias in amplification of androgen receptor alleles, a trinucleotide repeat marker used in clonality studies. Nucleic Acids Res 23:1411–1418. Mutter GL and Boynton KA (1995b). X chromosome inactivation in the normal female genital tract: implications for identification of neoplasia. Cancer Res 55:5080 –5084. Mutter GL, Chaponot ML, and Fletcher JA (1995). A polymerase chain reaction assay for non-random X chromosome inactivation identifies monoclonal endometrial cancers and precancers. Am J Pathol 146:501–508. Nordenson I, Ljungberg B, and Roos G (1988). Chromosomes in renal carcinoma with reference to intratumor heterogeneity. Cancer Genet Cytogenet 32:35– 41. Nowell PC (1976). The clonal evolution of tumor cell populations. Science 194:23–28. Ro JY, Ayala AG, and el-Naggar A (1987). Muscularis mucosa of urinary bladder. Importance for staging and treatment. Am J Surg Pathol 11:668 – 673. Sager R (1989). Tumor suppressor genes: The puzzle and the promise. Science 246:1406 –1412. Sidransky D, Frost P, Von Eschenbach A, Oyasu R, Preisinger AC, and Vogelstein B (1992). Clonal origin bladder cancer. N Engl J Med 326:737–740. Sidransky D, Von Eschenbach A, Tsai YC, Jones P, Summerhayes I, Marshall F, Paul M, Green P, Hamilton SR, Frost P, and Vogelstein B (1991). Identification of p53 gene mutations in bladder cancers and urine samples. Science 252:706 –709. Simpson JF, Dutt PL, and Page DL (1992). Expression of mitoses per thousand cells and cell density in breast carcinomas: a proposal. Hum Pathol 23:608 – 611. Smits G, Schaafsma E, Kiemeney L, Caris C, Debruyne F, and Witjes JA (1998). Microstaging of pT1 transitional cell carcinoma of the bladder: identification of subgroups with distinct risks of progression. Urology 52:1009 –1013.

Intratumor Heterogeneity in Bladder TCC

Sternlicht M, Mirell C, Safarians S, and Barsky SH (1994). A novel strategy for the investigation of clonality in precancerous disease states and early stages of tumor progression. Biochem Biophys Res Commun 199:511–518. Uchida T, Wada C, Ishida H, Egawa S, Ao T, Yokoyama E, and Koshiba K (1995). Infrequent involvement of mutations on neurofibromatosis type 1, H-ras, K-ras and N-ras in urothelial tumors. Urol Int 55:63– 67.

Wolman SR and Heppner GH (1992). Genetic heterogeneity in breast cancer. J Natl Cancer Inst 84:469 – 470. Younes M, Sussman J, and True LD (1990). The usefulness of the level of the muscularis mucosae in the staging of invasive transitional cell carcinoma of the urinary bladder. Cancer 66:543–548.

Voorter C, Joos S, Bringuier PP, Vallinga M, Poddighe P, Schalken J, du Manoir S, Ramaekers F, Lichter P, and Hopman A (1995). Detection of chromosomal imbalances in transitional cell carcinoma of the bladder by comparative genomic hybridization. Am J Pathol 146:1341–1354.

Laboratory Investigation • March 2000 • Volume 80 • Number 3

289

Anatomic Pathology / TOPOGRAPHIC KINETICS IN TRANSITIONAL CELL CARCINOMAS

Kinetic Profiles by Topographic Compartments in MuscleInvasive Transitional Cell Carcinomas of the Bladder Role of TP53 and NF1 Genes Alfredo Blanes, MD, PhD,1 Javier Rubio, MD, PhD,1 Armando Martinez, MD, PhD,2 Hubert J. Wolfe, MD,3† and Salvador J. Diaz-Cano, MD, PhD, MRCPath3,4 Key Words: Bladder; Transitional cell carcinoma; Proliferation; Apoptosis; Tumor heterogeneity

Abstract We evaluated 71 muscle-invasive transitional cell carcinomas (TCCs) of the bladder by tumor compartments. Kinetic parameters included mitotic figure counting, Ki-67 index, proliferation rate (DNA slide cytometry), and apoptotic index (in situ end labeling [ISEL] of fragmented DNA using digoxigeninlabeled deoxyuridine triphosphate and Escherichia coli DNA polymerase [Klenow fragment]). At least 50 highpower fields per compartment were screened from the same tumor areas; results are expressed as percentage of positive neoplastic cells. Mean and SD were compared by tumor compartment. DNA was extracted from microdissected samples (superficial and deep) and used for microsatellite analysis of TP53 and NF1 by polymerase chain reaction–denaturing gradient gel electrophoresis. Significantly higher marker scores were revealed in the superficial compartment than in the deep compartment. An ISEL index of less than 1% was revealed in 63% (45/71) of superficial compartments and 86% (61/71) of deep compartments. Isolated NF1 alterations were observed mainly in superficial compartments, whereas isolated TP53 abnormalities were present in deep compartments. Lower proliferation and down-regulation of apoptosis define kinetically the deep compartment of muscleinvasive TCC of the bladder and correlate with the topographic heterogeneity, NF1-defective in superficial compartments and TP53-defective in deep compartments.

Transitional cell carcinomas (TCCs) of the bladder show variable biologic potential and kinetic features (proliferation and apoptosis). Several proliferation markers (mitotic count, silver-stained nucleolar organizer regions, immunohistochemical staining with Ki-67, proliferating cell nuclear antigen, and bromodeoxyuridine labeling index) have been tested in TCCs with strong correlations among them and with tumor grade, but not with stage.1 The significance of constitutive apoptosis in the development and progression of TCC has not been investigated fully, and several markers have been proposed to investigate this aspect, including in situ end labeling (ISEL) or DNA cytometric analysis.2-4 The apoptotic index is normally higher in TCCs than in normal transitional epithelium and increases with increasing grades, although with no statistically significant differences between tumor groups.5 The findings reported in TCCs have revealed no statistically significant correlation between the apoptotic index and pathologic stage, but have revealed a strong direct correlation between an increased proliferation rate and bladder cancer progression.5 The prognostic usefulness of the staging system in bladder tumors has been improved after recognizing the relevance of muscularis mucosa (MM).6 The data show that the extent of lamina propria invasion is a clinically relevant prognostic factor for progression of pT1 TCC of the bladder.7,8 Tumors extending beyond the MM behave in a way similar to muscle-invasive TCC, especially if they are high-grade, reveal associated carcinoma in situ, or express nuclear TP53.7-9 TCC heterogeneity is well documented, and it is assumed that the molecular evolution and progression of neoplasms can be estimated from the number of genetic abnormalities, advanced neoplasms accumulating more Am J Clin Pathol 2002;118:93-100

Blanes et al / TOPOGRAPHIC KINETICS IN TRANSITIONAL CELL CARCINOMAS

abnormalities.10-13 Heterogeneity studies, however, have overlooked the cell topography, especially for combined genetic and kinetic features. The presence of abnormal and unrelated clones results in genetic heterogeneity, which correlates with the tumor cell topography, above or below the MM.12 Tumor cells located above the MM mainly showed NF1 microsatellite (MS) alterations, whereas cells located deeper to the MM revealed TP53 MS abnormalities.12 This intratumor genetic heterogeneity might be reflected on kinetic features, but no systematic evaluation by topographic compartments of both genetic and kinetic features is available in muscle-invasive TCC to date. The aim of the present study was to characterize the topographic compartments of muscle-invasive TCC using proliferation and apoptosis markers, as well as NF1 and TP53 MS patterns. Tumor cell heterogeneity was assessed by the independent evaluation of topographic compartments (above and below the MM).

Materials and Methods Case Selection and Sampling We reviewed all muscle-invasive TCCs of the urinary bladder (pT2 and pT3) diagnosed during a period of 5 years from 3 reference hospitals. For 72 cases, properly preserved archival material was available for further analyses. All surgical specimens were completely embedded for histopathologic diagnosis, and 2 topographic compartments were analyzed in each TCC regarding their relationship with the MM, revealed by its ectatic vascular plexus ❚Image 1❚.6 Tumor cells above the MM were labeled superficial, while neoplastic cells located below the MM were considered deep. The same areas in consecutive sections were used in each study, and their cellular composition was confirmed in adjacent H&E-stained sections. Tumor Grading and Mitotic Figure Counting A standardized protocol was developed for reviewing each tumor.14,15 Tumors were evaluated histologically and graded by 3 independent observers (A.B., J.R., and S.J.D.C.). In case of grading disagreement, the lesions were discussed during simultaneous inspection before final categorization. Reproducibility data were not recorded. Mitotic figures (MFs) were counted on H&E-stained sections by the same observers as part of the tumor grading. MFs were identified using reported criteria and screened in 50 high-power fields (HPFs) in each tumor compartment (1 HPF = 0.1428 mm2),16 beginning in the most cellular area. When the tumor compartments were smaller than 50 HPF (3 superficial compartments and 6 deep compartments), the 94

Am J Clin Pathol 2002;118:93-100

whole compartment was screened (at least 28 HPF). Both the number of positive nuclei per HPF and the number of neoplastic cells intercepted by the microscope field diameter were registered. The last score was used to estimate the number of neoplastic cells per HPF using the formula: N = (nπ/4)2 where N is the number of estimated cells per HPF and n the number of cells intercepted by the microscope field diameter.4,17,18 The number of positive nuclei was always expressed per HPF and per 1,000 proliferating cells. Both the mean and the SD values were calculated as representative scores per compartment and case. Immunohistochemical Detection of Ki-67 Antigen The sections were mounted on positively charged microscope slides (Superfrost Plus, Fisher Scientific, Fair Lawn, NJ) and baked at 60°C for 2 hours. The slides were routinely dewaxed and rehydrated. The endogenous peroxidase activity was then quenched with 0.5% hydrogen peroxide in methanol for 10 minutes. A microwave antigen retrieval method (20 minutes in a 10-mmol/L concentration of citrate buffer, pH 6.0, at 600 W) was used, followed by incubation with polyclonal horse serum (20 minutes, 1:100 dilution; DAKO, Glostrup, Denmark) and with monoclonal MIB-1 antibody (overnight, 4°C), at 2 µg/mL (Calbiochem, Cambridge, MA). Then sections were serially incubated with biotinylated antimouse antibody (30 minutes, 1:200 dilution; DAKO) and peroxidase-labeled avidin-biotin complex (60 minutes, 1:100 dilution; DAKO). All incubations were performed in a moist chamber at room temperature unless otherwise specified. The reaction was developed under microscopic control, using 3,3'-diaminobenzidine tetrahydrochloride with 0.3% hydrogen peroxide as chromogen (Sigma Chemical, St Louis, MO), and the sections were counterstained with hematoxylin. Positive (reactive lymph node) and negative (omitting the primary antibody) controls were run simultaneously. The threshold of positivity was established experimentally at the positive control in each staining batch. Only nuclei with staining features similar to those of their corresponding positive control were considered positive. The immunostaining was quantified using the Cell Analysis System model 200 and Quantitative Proliferation Index software (Becton Dickinson, Franklin Lakes, NJ). This analysis provided the percentage of positive tumor nuclei and the percentage of positive nuclear area from each compartment. Both nuclear and positivity thresholds were optimized experimentally in the corresponding positive control. Slide Cytometric Analysis of Nuclear DNA Content Feulgen-stained sections were used for DNA quantification,19 using the Cell Analysis System model 200 and Quantitative DNA Analysis software (Becton-Dickinson).20 At least © American Society for Clinical Pathology

Anatomic Pathology / ORIGINAL ARTICLE

❚Image 1❚ Histologic appearance of tumor compartments in muscle-invasive transitional cell carcinoma. Superficial compartments (A and C) showed higher Ki-67 labeling than deep compartments (B and D) (magnification ×200).

300 nuclei were measured in each TCC compartment and the results recorded separately. Only complete, nonoverlapping, and focused nuclei were interactively selected, beginning in the most cellular area until completion in consecutive microscope HPFs (×400). Several 5-µm sections were used for this analysis, according to previously published protocols that have proven valid in such material.4,21,22 From the same slide, both lymphocytes and histologically normal urothelial cells were used as diploid controls. External diploid controls (rat hepatocytes, Becton Dickinson) were included in each staining batch to normalize results (1 slide per staining holder). They were used for setting the diploid G0/G1 limits and calculating the DNA index of each G0/G1 cell population (10% or more of measured cells and evidence of G 2 + M cells). 23 The © American Society for Clinical Pathology

histogram of nuclear optical density was used to evaluate the DNA index (referred to their corresponding diploid controls), the proliferation rate (PR = S + G2 + M/G1 + S + G2 + M, expressed as a percentage), and the ratio between the nuclear area and the DNA content of the cells in each cell cycle phase. The last variable also was referred to the corresponding values in the histologically normal urothelial cells to normalize the results. Both mean nuclear area and nuclear area/DNA index ratio of G0/G1 cells were recorded. The latter represents a morphometric parameter of apoptosis when coupled with ISEL.24 Age- and sex-matched histologically normal bladders from 10 surgical samples were selected for nuclear area and DNA index analysis. At least 1,000 urothelial cells were evaluated as controls for this purpose. Am J Clin Pathol 2002;118:93-100

Blanes et al / TOPOGRAPHIC KINETICS IN TRANSITIONAL CELL CARCINOMAS

ISEL of Fragmented DNA Since extensive DNA fragmentation is an important characteristic of apoptosis, visualization of DNA breaks has proved useful in the identification of apoptotic cells.3 This extensive DNA fragmentation results in a high density of 5'protruding ends, which can be detected using the Klenow fragment of DNA polymerase I with a mixture of labeled nucleotides.25,26 Briefly, the sections were routinely deparaffinized and hydrated. After incubation in 2× standard saline citrate buffer (80°C, 20 minutes) and protein digestion [500 µg/mL Proteinase K in a 10-mmol/L concentration of tris(hydroxymethyl)aminomethane (Tris) hydrochloride, pH 7.5, a 10-mmol/L concentration of EDTA, 0.5% sodium dodecyl sulfate, at room temperature for 25 minutes], the sections were incubated with the Klenow fragment of Escherichia coli DNA polymerase I under appropriate conditions (20 U/mL in a 50-mmol/L concentration of Tris hydrochloride, pH 7.5; a 10-mmol/L concentration of magnesium chloride; a 1-mmol/L concentration of dithiothreitol; 250 µg/mL of bovine serum albumin with a 100µmol/L concentration of each deoxynucleoside triphosphate, maintaining a proportion of 11-digoxigenin-deoxyuridine triphosphate/deoxythymidine triphosphate of 0.35/0.65; 2 hours at 37°C). The digoxigenin-labeled DNA fragments were immunoenzymatically detected using an antidigoxigenin polyclonal Fab fragment labeled with alkaline phosphatase (1:100 dilution, Boehringer-Mannheim, Mannheim, Germany); the enzymatic reaction was developed under microscopic control with nitroblue-tetrazolium and X-phosphate.27 The sections were counterstained with diluted hematoxylin (25%), dehydrated, and mounted. Both positive (reactive lymph node) and negative (omitting DNA polymerase in the enzymatic incubation) controls were run simultaneously. The threshold of positivity was established experimentally at the positive control, and the ISEL index was expressed as the percentage of positive nuclei referred to the total number of neoplastic cells present in the same HPF as reported.4,17,18,28 The whole lesion or at least 50 consecutive HPFs were screened, beginning in the most cellular area. MS Analysis of Tumor Suppressor Genes DNA was extracted from two 20-µm unstained paraffin sections per tumor compartment and control samples, including at least 100 cells (approximately 0.4 mm2) per sample. Appropriate controls were included for each test (histologically normal urothelium, stroma from the lamina propria, and smooth muscle). DNA was extracted using a modified phenol-chloroform protocol.29 DNA was precipitated with ice-cold absolute ethanol in the presence of a 0.3-mol/L concentration of sodium acetate, pH 5.2, and resuspended in 10 µL of polymerase chain 96

Am J Clin Pathol 2002;118:93-100

reaction (PCR) buffer (10-mmol/L concentration of Tris hydrochloride, pH 8.4; 50-mmol/L concentration of potassium chloride; 1.5-mmol/L concentration of magnesium chloride; and 100 µg/mL of bovine serum albumin). DNA then was used for PCR amplification of intron MS of tumor suppressor gene (TSG; NF1 and TP53), as described.12 The tests were run in duplicate in a Perkin-Elmer Thermal Cycler, model 480 (Perkin-Elmer, Norwalk, CT). The whole PCR volume (10 µL) was subjected to electrophoresis in 8% denaturing gradient polyacrylamide gels (0.75 mm, 20%-80% denaturing conditions from top to bottom). The gels were run at 5 V/cm until the xylene cyanol band was within the bottom gel inch. The gels then were fixed with 7% acetic acid (5 minutes), dried under vacuum (40 minutes, 80°C), and put inside a developing cassette containing 1 intensifying screen and preflashed films (Kodak XAR) facing the intensifying screen (16-48 hours, –70°C). The autoradiograms were developed using an automated processor, Kodak-Omat 100 (Kodak, Rochester, NY). Only informative cases (2 different alleles in control samples) were included in the final analysis and were interpreted as described.2,12,13,30 Allelic imbalance was densitometrically evaluated (EC model 910 optical densitometer, EC Apparatus, St Petersburg, FL). Only allele ratios of 4:1 or more in any TSG were considered evidence of loss of heterozygosity (LOH); otherwise retention of heterozygosity was assigned. Additional allele bands in the tumor samples were considered positive evidence of single nucleotide polymorphism (SNP) in denaturing gradient gels if they were not present in the corresponding control. Statistical Analysis The correlation between different proliferation markers themselves and proliferation and apoptosis markers was studied by regression analyses to calculate the correlation coefficient and the corresponding statistical significance. The data from both tumor compartments were used to calculate a representative value per case for each variable. Student t tests (if normal distribution was confirmed) or nonparametric analysis of variance (if the distribution was not normal) was applied to assess the differences by tumor compartment of mean and SD values of every quantitative variable. Each variable distribution was tested previously for normality using the Kolmogorov-Smirnoff test. Standard values of Ki-67 and ISEL indices of morphologically normal transitional cells obtained from 10 surgical samples were calculated. The upper limits of 95% confidence intervals were Ki-67 index 20% and ISEL index 1% (data not shown). Ki-67 index (20% threshold) and ISEL index (1% threshold) also were tested using the Fisher exact test. Differences were considered statistically significant if P was less than .05 in 2-tailed distributions. © American Society for Clinical Pathology

Anatomic Pathology / ORIGINAL ARTICLE

Results This series of muscle-invasive TCCs revealed highgrade tumors in 62 cases (86%) and aneuploid DNA content in 60 cases (83%) with a close relationship between tumor grade and DNA ploidy. The general kinetic features for the series are given by tumor compartment in ❚Table 1❚. Regression analyses showed a strong positive linear correlation between nuclear counting (percentage) and nuclear area (percentage) for the Ki-67 index (R = 0.992; P < .01) and a moderate positive linear correlation between the Ki-67 index (nuclear area, percentage) and the proliferation rate (percentage) obtained by slide cytometry (R = 0.612; P < .01). In contrast, the MF counting (per thousand) revealed a low positive correlation with the Ki67 index (nuclear area, percentage) (R = 0.287; P < .05) and the proliferation rate (percentage) obtained by slide cytometry (R = 0.258; P < .05). The strong correlation between both Ki-67 indices results in duplicated data, and, therefore, only 1 Ki-67 variable was included for further analyses. The Ki-67 index based on the nuclear area was selected because only this Ki-67 variable is truly continuous and has real biologic meaning in terms of nuclear fractions, eventually resulting in more precise evaluation of Ki-67 expression. ISEL indices showed only moderate positive linear correlation with MF counting (R = 0.512; P < .05), whereas the correlation was weak and nonsignificant (R = 0.207 or less; P > .05) with both Ki-67 index (area, percentage) and proliferation rate obtained by slide cytometry.

❚Table 1❚ Kinetic Features of Muscle-Invasive Transitional Cell Carcinomas by Tumor Compartments* Superficial Compartment Mitotic figure counting (‰) 7.0 Ki-67 index (%) 27.28 Proliferation rate (%) 31.43 In situ end labeling index (‰) 12.1 *

± 3.7 ± 6.15 ± 8.38 ± 13.9

Deep Compartment 2.9 12.37 17.94 12.9

± 2.1 ± 4.73 ± 6.18 ± 17.1

P .0003 .0001 .0018 .06

Data are given as mean ± SD.

The kinetic profile revealed significantly higher proliferation indices in superficial compartments than in deep compartments (Table 1, Image 1), regardless of the nuclear grade ❚Table 2❚ and nuclear DNA content ❚Table 3❚. In each study, the highest proliferation score was obtained by slide cytometry, followed by the Ki-67 index and MF counting (Tables 1-3). ISEL indices revealed variable values as demonstrated by their high SDs (greater than the corresponding means) and similar general scores in both tumor compartments (Table 1). However, tumor stratification by DNA ploidy and grade resulted in opposite apoptosis patterns. Only low-grade TCC showed ISEL indices higher in the deep compartment than in the superficial compartment (Table 2), whereas high-grade TCC and both diploid and aneuploid TCC had higher apoptosis scores in the superficial tumor compartment (Tables 2 and 3). These ISEL indices were not significantly different by tumor compartments and tumor grade, although they were significantly lower in diploid TCC (P = .0464).

❚Table 2❚ Kinetic Features of Muscle-Invasive Transitional Cell Carcinomas by Tumor Compartments and Tumor Grade* Superficial Compartment Low-Grade Mitotic figure counting (‰) Ki-67 index (%) Proliferation rate (%) In situ end labeling index (‰) *

4.1 20.35 25.20 13.5

± 2.0 ± 6.67 ± 5.85 ± 10.3

High-Grade 8.6 28.48 32.32 42.0

± 4.9 ± 5.67 ± 8.40 ± 36.4

Deep Compartment Low-Grade 1.8 9.44 13.77 15.7

± 1.5 ±5.08 ± 7.89 ± 11.7

High-Grade 3.7 13.15 18.65 35.6

± 3.0 ± 5.09 ± 5.71 ± 50.2

P .17 .009 .012 .22

Data are given as mean ± SD.

❚Table 3❚ Kinetic Features of Muscle-Invasive Transitional Cell Carcinomas by Tumor Compartments and DNA Ploidy Superficial Compartment Diploid Mitotic figure counting (‰) Ki-67 index (%) Proliferation rate (%) In situ end labeling index (‰) *

5.2 23.25 26.59 9.8

± 2.3 ± 6.39 ± 7.69 ± 3.0

Aneuploid 8.4 27.95 32.25 19.1

± 5.1 ± 5.90 ± 8.27 ± 21.2

Deep Compartment Diploid 1.7 9.02 14.86 7.7

± 1.6 ± 3.84 ± 3.87 ± 5.3

Aneuploid 3.8 12.96 18.50 18.5

± 3.0 ± 4.66 ± 6.38 ± 21.2

P .08 <.001 .002 .25

Data are given as mean ± SD.

Am J Clin Pathol 2002;118:93-100

Blanes et al / TOPOGRAPHIC KINETICS IN TRANSITIONAL CELL CARCINOMAS

Case stratification by kinetic features (Ki-67 threshold, 20%; ISEL threshold, 1%) revealed statistically significant differences between superficial and deep tumor compartments for both proliferation (Ki-67) and apoptosis (ISEL) indices. In each tumor compartment, no significant differences were observed when the cases were stratified by kinetic features (Ki-67 and ISEL indices), confirming the absence of correlation between Ki-67 and ISEL. The MS analysis of TSG revealed NF1 LOH/SNP in 31 (63%) of 49 informative cases, TP53 LOH/SNP in 44 (63%, screening 2 introns) of 70 informative cases, and no MS alterations of TSG introns in 23 TCCs (32%). Concordant MS patterns of TSG in both tumor compartments were observed in 49 muscle-invasive TCCs (68%) and topographically related genetic heterogeneity (LOH/SNP) in 23 cases (32%). The group of muscle-invasive TCCs with concordant MS patterns of TSG showed a subset of tumors with demonstrable genetic alterations in at least 1 TSG locus (26 cases [36%]) and another subset with no demonstrable alteration in TSG loci (23 cases [32%]). The group showing topographic TSG MS heterogeneity ❚Figure 1❚ comprised 16 TCCs (22%) expressing more genetic changes in the deep compartment (TP53 in all cases, coexistent with NF1 in 2) and 7 TCCs (10%) with more MS abnormalities at the superficial compartment (5 involving the NF1 locus and 2 at TP53).

70 60

Percent

50 40 30 20 10 0

NF1 LOH/SNP

TP53 LOH/SNP

NF1 LOH/SNP

TP53 LOH/SNP

❚Figure 1❚ Actual (white bars) and expected (black bars) frequency of loss of heterozygosity (LOH)/single nucleotide polymorphism (SNP) involving NF1 and TP53 loci revealed in muscle-invasive transitional cell carcinoma with heterogeneous microsatellite profile by topographic compartments. Superficial compartments show mainly NF1 abnormalities, whereas deep compartments reveal TP53 alterations.

Am J Clin Pathol 2002;118:93-100

Discussion This study presents for the first time the correlation between kinetic and genetic features by topographic compartments. Muscle-invasive TCCs of the bladder reveal kinetically distinctive topographic compartments; tumor cells below the MM show down-regulation of both proliferation and apoptosis that matches with the differential TSG MS profile. Early steps involving TSG target NF1 in the high cellular turnover superficial compartment and TP53 in the low cellular turnover deep compartment. The molecular progression, as defined by accumulation of TSG MS abnormalities, also goes along with this kinetic profile. The deep compartment of muscle-invasive TCC shows down-regulation of both proliferation and apoptosis. Studies in other systems have demonstrated a direct correlation between proliferation and apoptosis in hyperplastic conditions and an inverse correlation in both low-grade neoplasms and intraepithelial neoplasias. 4,26 Although apoptosis increases as the tumor progresses,3,13 any focal apoptosis down-regulation contributes to the accumulation of genetic abnormalities13 and confirms and extends our findings in tumor cells located below the MM.12 This process also contributes to the selection of abnormal and unrelated clones assumed to be the result of tumor cell heterogeneity. Genetic abnormalities result in intratumor heterogeneity by topographic cell selection,12 which also is expressed in a distinctive kinetic profile. After malignant transformation, tumor cells can grow independently with variable subsequent genetic alterations in each tumor compartment, explaining the heterogeneity. Down-regulation of apoptosis in lowproliferating cells contributes to the accumulation of genetic abnormalities reported associated with molecular progression13,31,32 and correlates with the topography of tumor cells above or below the MM. Two processes of tumor cell selection seem to involve different TSGs and to be responsible for that topographic heterogeneity.12 The NF1 was preferentially altered in superficial compartments that show high cellular turnover. Genetic (DNA) and gene expression (messenger RNA, protein) analyses have proposed an important role for NF1 in the malignant pathway of muscleinvasive and high-grade TCC.12,33 The NF1 gene product has an inhibitory effect on the RAS protein, which is expressed with the highest levels in immature and proliferating cells.34 The absence of the NF1 inhibitory effect will favor increased cell proliferation, as found in the superficial tumor compartments of muscle-invasive TCC in the present series. In contrast, TP53 abnormalities tend to concentrate in the deep tumor compartment, correlating with the late event of the loss of chromosome 17p reported in tumor progression of superficial TCC.12,35 TP53 abnormalities have been reported in about 65% of invasive TCCs,12,35 normally associated with © American Society for Clinical Pathology

Anatomic Pathology / ORIGINAL ARTICLE

From the Departments of Pathology, 1University Hospital, Malaga, Spain; 2Hospital Clinic “San Carlos,” Madrid, Spain; 3Tufts University–New England Medical Center, Boston, MA; and 4Barts and The London Queen Mary’s School of Medicine, University of London, London, England. †

Deceased. Presented in part in the 88th Annual Meeting of the United States and Canadian Academy of Pathology, San Francisco, CA, March 20-26, 1999. © American Society for Clinical Pathology

T C

TP53

71.4

28.6

TP53

92.3

100

T C

92.3

100

NF1

TP53

Superficial

NF1

NF1-TP53 progression

12.5

100

NF1

TP53

Deep

17p allelic deletions, which leave cells with only mutant forms of the TP53 gene product. The abnormal TP53 gene product will not be able to induce apoptosis of G1-arrested cells that carry genetic abnormalities.36 Progressing in the cell cycle, these cells can accumulate genetic abnormalities, as reported in the TCC deep compartment.12 The accumulation of genetic abnormalities in both compartments of tumors with concordant TSG MS profiles suggests that this homogeneous topographic profile results from progressive cell selection ❚Figure 2❚.12,36-38 For these cases, the accumulation of genetic abnormalities also can be used as a clonality marker.13 Kinetic compartments are assumed in malignant tumors and explain the variable response to radiotherapy and chemotherapy in these conditions.13,32 The present study shows proliferation rates by slide cytometry higher than the corresponding Ki-67 indices (Table 1), suggesting that slide cytometry counts cells not expressing Ki-67 antigen as proliferating cells. Postsynthetic (G2 + M) apoptotic cells would lose part of their fragmented DNA through the nuclear pores, resulting in DNA content within the range of S-phase cells,4,24,26 explaining higher proliferation scores using slide cytometry (DNA content) than using Ki-67 labeling. These findings also stress the importance of evaluating proliferation and apoptosis together to obtain a meaningful conclusion on tumor kinetics potentially applicable for treatment design.2,4,39 To our knowledge, this is the first report to reveal that the kinetic compartments of muscle-invasive TCC keep a close correlation with depth of invasion of tumor cells in the bladder wall. Similar kinetic profiles have been described in other neoplasms, such as skin tumors with ductal differentiation40 and malignant melanomas.41 The presence of this superficial expansive compartment would contribute to the lateral extension and superficial spreading observed in all of these neoplasms. The deep tumor compartments of muscle-invasive TCCs reveal consistent down-regulation of proliferation and apoptosis, which would contribute to the accumulation of genetic abnormalities characteristic of muscle-invasive TCC. This pattern fits with the genetic profile of muscle-invasive TCC, NF1-defective in the superficial compartment and TP53-defective in the deep compartment. The heterogeneous MS profile represents an early molecular step, which progresses to a homogeneous TSG MS profile by accumulation of genetic alterations (Figure 2).

No NF1-TP53 heterogeneity T C

T C

❚Figure 2❚ Molecular progression expressed by accumulation of both NF1 and TP53 microsatellite abnormalities results in topographically homogeneous genetic profile for both loci in muscle-invasive transitional cell carcinoma of the bladder. Arrows point to allele loss (densitometrically proven); the arrowhead shows extra bands as expression of single nucleotide polymorphism. The numbers in the bars refer to the percentage of tumor compartments revealing a given genetic abnormality. C, control; T, tumor.

Address reprint requests to Dr Diaz-Cano: Dept of Histopathology and Morbid Anatomy, The Royal London Hospital, Whitechapel, London E1 1BB, England.

References 1. Cohen MB, Waldman FM, Carroll PR, et al. Comparison of five histopathologic methods to assess cellular proliferation in transitional cell carcinoma of the urinary bladder. Hum Pathol. 1993;24:772-778. 2. Diaz-Cano SJ, de Miguel M, Blanes A, et al. Clonality as expression of distinctive cell kinetic patterns in nodular hyperplasias and adenomas of the adrenal cortex. Am J Pathol. 2000;156:311-319. Am J Clin Pathol 2002;118:93-100

Blanes et al / TOPOGRAPHIC KINETICS IN TRANSITIONAL CELL CARCINOMAS

3. Salomon RN, Diaz-Cano S. Introduction to apoptosis. Diagn Mol Pathol. 1995;4:235-238. 4. Koch M, de Miguel M, Höfler H, et al. Kinetic profiles of intraepithelial and invasive prostatic neoplasias: the key role of downregulated apoptosis in tumor progression. Virchows Arch. 2000;436:413-420. 5. King ED, Matteson J, Jacobs SC, et al. Incidence of apoptosis, cell proliferation and bcl-2 expression in transitional cell carcinoma of the bladder: association with tumor progression. J Urol. 1996;155:316-320. 6. Ro JY, Ayala AG, el-Naggar A. Muscularis mucosa of urinary bladder: importance for staging and treatment. Am J Surg Pathol. 1987;11:668-673. 7. Smits G, Schaafsma E, Kiemeney L, et al. Microstaging of pT1 transitional cell carcinoma of the bladder: identification of subgroups with distinct risks of progression. Urology. 1998;52:1009-1014. 8. Younes M, Sussman J, True LD. The usefulness of the level of the muscularis mucosae in the staging of invasive transitional cell carcinoma of the urinary bladder. Cancer. 1990;66:543-548. 9. Hermann GG, Horn T, Steven K. The influence of the level of lamina propria invasion and the prevalence of p53 nuclear accumulation on survival in stage T1 transitional cell bladder cancer. J Urol. 1998;159:91-94. 10. Brown JL, Russell PJ, Philips J, et al. Clonal analysis of a bladder cancer cell line: an experimental model of tumour heterogeneity. Br J Cancer. 1990;61:369-376. 11. Voorter C, Joos S, Bringuier PP, et al. Detection of chromosomal imbalances in transitional cell carcinoma of the bladder by comparative genomic hybridization. Am J Pathol. 1995;146:1341-1354. 12. Diaz-Cano SJ, Blanes A, Rubio J, et al. Molecular evolution and intratumor heterogeneity by topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder. Lab Invest. 2000;80:279-289. 13. Diaz-Cano SJ, Blanes A, Wolfe HJ. PCR techniques for clonality assays. Diagn Mol Pathol. 2001;10:24-33. 14. Epstein JI, Amin MB, Reuter VR, et al, for the Bladder Consensus Conference Committee. The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Am J Surg Pathol. 1998;22:1435-1448. 15. Hierro Martin I, Alvarez Perez M, Blanes Berenguel A, et al. Contribucion a la categorizacion de los CCT de vejiga, sistematizacion de la gradacion y utilidad de la ploidia de ADN [in Spanish]. Actas Urol Esp. 1999;23:489-496. 16. van Diest PJ, Baak JP, Matze-Cok P, et al. Reproducibility of mitosis counting in 2,469 breast cancer specimens: results from the Multicenter Morphometric Mammary Carcinoma Project. Hum Pathol. 1992;23:603-607. 17. Diaz-Cano SJ, Leon MM, de Miguel M, et al. Mitotic index quantification: different approaches and their value in adrenocortical proliferative lesions [abstract]. Lab Invest. 1996;74:170A. 18. Harjacek M, Diaz-Cano S, Alman BA, et al. Prominent expression of mRNA for proinflammatory cytokines in synovium in patients with juvenile rheumatoid arthritis or chronic Lyme arthritis. J Rheumatol. 2000;27:497–503. 19. Bibbo M, Bartels PH, Dytch HE, et al. Cell image analysis. In: Bibbo M, ed. Comprehensive Cytopathology. Philadelphia, PA: Saunders; 1991:965-983. 20. Bacus JW, Grace LJ. Optical microscope system for standardized cell measurements and analysis. Appl Optics. 1987;26:3280-3294.

100

Am J Clin Pathol 2002;118:93-100

21. Marchevsky AM, Truong H, Tolmachoff T. A rule-based expert system for the automatic classification of DNA “ploidy” histograms measured by the CAS 200 image analysis system. Cytometry. 1997;30:39-46. 22. Farnsworth WV, DeRose PB, Cohen C. DNA image cytometric analysis of paraffin-embedded sections of small renal cortical neoplasms. Cytometry. 1994;18:223-227. 23. Dressler LG. Controls, standards, and histogram interpretation in DNA flow cytometry. Methods Cell Biol. 1990;33:157-171. 24. Sherwood SW, Schimke RT. Cell cycle analysis of apoptosis using flow cytometry. Methods Cell Biol. 1995;46:77-97. 25. Wijsman JH, Jonker RR, Keijzer R, et al. A new method to detect apoptosis in paraffin sections: in situ end-labeling of fragmented DNA. J Histochem Cytochem. 1993;41:7-12. 26. Diaz-Cano SJ, de Miguel M, Blanes A, et al. Clonality patterns are expression of kinetic features in adrenal cortical nodular hyperplasias and adenomas. Lab Invest. 1999;79:174A. 27. Diaz-Cano SJ, Garcia-Moliner M, Carney W, et al. Bcl-2 expression and DNA fragmentation in breast carcinoma, pathologic and steroid hormone receptors correlates. Diagn Mol Pathol. 1997;6:199-208. 28. Simpson JF, Dutt PL, Page DL. Expression of mitoses per thousand cells and cell density in breast carcinomas: a proposal. Hum Pathol. 1992;23:608-611. 29. Diaz-Cano SJ, Brady SP. DNA extraction from formalin-fixed, paraffin-embedded tissues: protein digestion as a limiting step for retrieval of high-quality DNA. Diagn Mol Pathol. 1997;6:342-346. 30. Diaz-Cano SJ. Designing a molecular analysis of clonality in tumours. J Pathol. 2000;191:343-344. 31. Li ZH, Aaltonen LA, Shu Q, et al. Effects of mutation and growth rates on patterns of microsatellite instability. Am J Pathol. 1996;148:1757-1761. 32. Diaz-Cano SJ. Clonality studies in the analysis of adrenal medullary proliferations: application principles and limitations. Endocr Pathol. 1998;9:301-316. 33. Aaltonen V, Bostrom PJ, Soderstrom KO, et al. Urinary bladder transitional cell carcinogenesis is associated with down-regulation of NF1 tumor suppressor gene in vivo and in vitro. Am J Pathol. 1999;154:755-765. 34. Furth ME, Aldrich TH, Cordon-Cardo C. Expression of ras proto-oncogene proteins in normal human tissues. Oncogene. 1987;1:47-58. 35. Sidransky D, Frost P, Von Eschenbach A, et al. Clonal origin bladder cancer. N Engl J Med. 1992;326:737-740. 36. Baithun SI, Naase M, Blanes A, et al. Molecular and kinetic features of transitional cell carcinomas of the bladder: biological and clinical implications. Virchows Arch. 2001;438:289-297. 37. Alexander P. Do cancers arise from a single transformed cell or is monoclonality of tumours a late event in carcinogenesis? Br J Cancer. 1985;51:453-457. 38. Woodruff MF, Ansell JD, Forbes GM, et al. Clonal interaction in tumours. Nature. 1982;299:822-824. 39. Diaz-Cano SJ, de Miguel M, Blanes A, et al. Clonal patterns in phaechromocytomas and MEN-2A adrenal medullary hyperplasias: histologic and kinetic correlates. J Pathol. 2000;192:221-228. 40. Pozo L, Camacho F, Rios-Martin JJ, et al. Cell proliferation in skin tumors with ductal differentiation: patterns and diagnostic applications. J Cutan Pathol. 2000;27:292-297. 41. Blanes A, Marquez A, Sanchez-Carrillo JJ, et al. Proliferative features by tumor cell compartments in skin malignant melanomas [abstract]. J Pathol. 1999;187:45A.

Histopathology 2007, 51, 458–467. DOI: 10.1111/j.1365-2559.2007.02795.x

Microsatellite abnormalities and somatic down-regulation of mismatch repair characterize nodular-trabecular muscle-invasive urothelial carcinoma of the bladder J Rubio,1 A Blanes,1 J J Sanchez-Carrillo1 & S J Diaz-Cano1,2 Departments of Pathology, 1University Hospital of Malaga, Malaga, Spain and 2King’s College Hospital and King’s College London School of Medicine, London, UK Date of submission 12 October 2006 Accepted for publication 20 February 2007

Rubio J, Blanes A, Sanchez-Carrillo J J & Diaz-Cano S J (2007) Histopathology 51, 458–467

Microsatellite abnormalities and somatic down-regulation of mismatch repair characterize nodular-trabecular muscle-invasive urothelial carcinoma of the bladder Aims: To correlate histological infiltration patterns with genetic and mismatch repair (MMR) profiles in muscle-invasive bladder urothelial carcinomas (UroC). Methods and results: Infiltration patterns were assessed in the deep compartment of muscle-invasive UroC (nodular-trabecular, 45 cases; infiltrative, 27 cases). Tumour compartment (superficial and deep to muscularis mucosa) analysis included: microsatellite pattern of TP53, RB1, WT1 and NF1 by polymerase chain reaction ⁄ denaturing gradient gel electrophoresis; mitotic, Ki67, in situ end labelling (ISEL) indices and DNA ploidy. MMR was assessed by MLH1 and MSH2 sequencing and immunohistochemistry in UroC with two or more abnormal microsatellite loci. Statistical differences were tested using anova and Fisher’s exact tests. Infiltrative UroC showed lower Ki67 index

14.94 ± 4.28, ISEL index 14.1 ± 10.0 and shorter median survival (20 months) than nodular-trabecular UroC (Ki67 index 20.65 ± 4.94, ISEL 20.2 ± 22.7, 37-month survival, respectively). The genetic profile was significantly different for RB1 (P ¼ 0.0003) and NF1 (P ¼ 0.0023) only, being more frequently abnormal in nodular-trabecular UroC. A significant decrease in MLH1 or MSH2 protein expression with no gene mutations was identified in UroC with microsatellite abnormalities and a nodular-trabecular growth pattern. Conclusions: Somatic down-regulation of MMR proteins in nodular-trabecular muscle-invasive UroC results in RB1 ⁄ NF1 microsatellite abnormalities, correlating with higher cellular turnover and longer survival.

Keywords: bladder, infiltration pattern, kinetic, microsatellite, prognosis, tumour suppressor gene, urothelial carcinoma Abbreviations: BSA, bovine serum albumin; CIS, carcinoma in situ; HPF, high-power field; ISEL, in situ end labelling; LGUD, low-grade urothelial dysplasia; LOH, loss of heterozygosity; MF, mitotic figure; MMR, mismatch repair; NF1, neurofibromatosis 1; RB1, retinoblastoma; ROH, retention of heterozygosity; SNP, single nucleotide polymorphism; TP53, tumour protein p53; TSG, tumour suppressor gene; UroC, urothelial carcinoma; WT1, Wilm’s tumour 1

Introduction The invasive capacity of tumours partially determines the infiltration pattern and correlates with Address for correspondence: Salvador J Diaz-Cano, MD, PhD, FRCPath, King’s College Hospital, Department of Histopathology, Denmark Hill, London SE5 9RS, UK. e-mail: salvador.diaz-cano@kcl.ac.uk 2007 The Authors. Journal compilation 2007 Blackwell Publishing Limited.

tumour prognosis, which, for bladder urothelial carcinomas (UroC), mainly depends on the distinction between superficial and invasive. These patterns correlate with tumour grade and stage, nuclear DNA content and proliferation in various groups of UroC,1,2 but rarely in muscle-invasive UroC. Pathological T stage and lymph node status remain the most powerful predictors of progression in

MMR and UroC infiltration patterns

muscle-invasive UroC. In this group of patients, an infiltrative growth pattern may be associated with a more dismal prognosis,3 the biological reason for which remains unknown. Substaging of pT1 UroC has improved the prediction of progression:4,5 tumours extending beyond the muscularis mucosa behave like muscle-invasive UroC, especially those high-grade UroC expressing TP53 and revealing associated carcinoma in situ.4–6 The level of muscularis mucosa has also been useful in assessing topographical histological and molecular heterogeneity in bladder UroC.7,8 This has resulted in distinctive microsatellite and clonal profiles leading to topographical segregation of proliferative and invasive tumour cells.9–13 Therefore, topographical analysis of genetic and kinetic features will result in a better understanding of the molecular evolution of neoplasms.8 Genetic and kinetic profiles by topographical compartments have not been analysed in muscle-invasive UroC with respect to their patterns of infiltration (nodular-trabecular versus infiltrative). Th aim of this study was to analyse tumour suppressor gene (TSG) microsatellite patterns, mismatch repair (MMR) profiles, proliferation and apoptosis in muscle-invasive UroC using microdissected samples from the superficial and deep compartments to assess TSGs controlling G1–S transition (TP53, RB1), RAS pathway (NF1) and development (WT1). Tumours were stratified according to their deep compartment infiltration pattern (nodular-trabecular versus infiltrative) and data were correlated with cancer-specific survival.

Materials and methods c a s e s el ec t i on an d s a m p l in g Initial biopsy specimens of all muscle-invasive (pT2a ⁄ b only) lymph node-negative (pN1) UroC of the urinary bladder treated with cystectomy and lymphadenectomy only (72 cases) from three reference hospitals (1990–1992, median follow-up 60 months) were reviewed; all cases had properly preserved archival material for both tumour and control tissues (see below). Topographical compartments were defined as superficial and deep to the muscularis mucosa,7,14 a limit that has been demonstrated to be prognostically useful in high-grade pT1 UroC.5 This protocol was approved by the Hospital Research Board and Ethics Committee and complied with their requirements.

459

t u m o ur i n f i l t r a t i o n p a t t e r n , g r a di n g a n d m i t o t i c f i g u re co un t i n g The pattern of infiltration was evaluated in deep compartments, classifying the tumour by the predominant pattern (> 50%) of nodular-trabecular or infiltrative (Figure 1). Histological grading evaluated architectural features, nuclear grade and mitotic figure (MF) counting.2 MFs were screened in 50 high-power fields (HPF) per compartment (7.140 mm2) or the whole tumour if smaller (three superficial and six deep compartments),15 beginning in the most cellular area. Both the number of positive nuclei per HPF and the number of neoplastic cells intercepted by the microscope field diameter (n) were recorded, the latter to estimate the number of neoplastic cells ⁄ HPF [N ¼ (np ⁄ 4)2];13,16 results were expressed per 1000 cells, calculating average and standard deviation (SD) per compartment and patient. Tumours were graded by three independent observers (J.R., A.B. and S.J.D-C.); in cases of disagreement, tumours were reviewed simultaneously to achieve a consensus. Reproducibility data were not recorded. Dysplastic lesions were classified according to the World Health Organization ⁄ International Society of Urological Pathology system as low-grade dysplasia (LGUD) and carcinoma in situ (CIS).17 t s g m i c r os a t el l it e a n a l y si s DNA was extracted from the most cellular areas of superficial and deep compartments,7,11,18–20 after microdissecting at least 100 cells ( 0.4 mm2, laser capture; Arturus, Mountain View, CA, USA) from two 20-lm unstained paraffin sections ⁄ compartment (Figure 2). Appropriate controls (histologically normal urothelium, stroma from the lamina propria and smooth muscle) and quality assurance (sensitivity, specificity, positive and negative) were run for each test.21–23 DNA was extracted using a modified phenol–chloroform protocol, precipitated with ice-cold absolute ethanol and resuspended in 10 ll of Tris–HCl buffer at pH 8.4.23 DNA was then used for polymerase chain reaction (PCR) amplification of TSG intron microsatellites (Table 1).7,24 The tests were run in a Perkin-Elmer thermal cycler model 480 (Perkin-Elmer, Norwalk, CT, USA). The whole 10-ll PCR volume was electrophoresed onto 8% denaturing gradient polyacrylamide gels; dried gels were put inside developing cassettes containing one intensifying screen and preflashed films (Kodak XAR; Kodak Co., Rochester, NY, USA).7,24,25 The radiographs were developed using an automated processor Kodak-Omat 100 (Kodak Co.).

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

460

J Rubio et al.

Figure 1. A,B, Nodular-trabecular urothelial carcinoma (UroC) shows sheets of neoplastic cells with minimal stromal reaction replacing the muscularis propria (haematoxylin and eosin). C,D, Infiltrative UroC shows small tumour nests ⁄ thin cords embedded in a prominent desmoplastic reaction, dissecting the smooth muscle fibres (haematoxylin and eosin).

Interpretation and inclusion criteria in each sample were achieved as follows:7,11,21,22,26,27 (i) allelic imbalance was densitometrically evaluated (EC model 910 optical densitometer; EC Apparatus Corp., St Petersburg, FL, USA). For evidence of loss of heterozygosity (LOH) only allele ratios ‡ 4 : 1 in any TSG were considered; otherwise retention of heterozygosity (ROH) was assigned.7,11 This ratio represents 80% of clonal cells in the sample and was used to increase the detection specificity;22,26,28 (ii) additional allele bands present in tumour samples but not in the corresponding controls were considered evidence of somatic single nucleotide polymorphism (SNP) by PCR ⁄ denaturing gradient gel electrophoresis.7,19,22,29 dna s equencing All microsatellite marker extra bands were cut from gels and DNA was purified using a QIA quick gel extraction kit (Qiagen, Valencia, CA, USA). The amplified product was diluted 20-fold in Tris–ethylenediamine tetraaceticacid buffer and 1 ll of the diluted

reaction product was subjected to a second round of PCR amplification using the appropriate primers for 30 cycles under the above conditions. Normal and extra bands from tumour-derived samples were PCR amplified along with the corresponding controls using a high-fidelity polymerase, Platinum PFX (Life Technologies, Gaithersburg, MD, USA). PCR products were directly sequenced after purification (QIAquick PCR purification kit; Qiagen). All sequencing was performed on an ABI Prism 3700 automated DNA Analyser and the sequence data analysed using the program Sequencher (Gene Codes Corp., Ann Arbor, MI, USA), which reverses and complements the antisense strand. All mutations were confirmed by sequencing in both directions and indicated by an ‘N’ in the sequencing chromatogram. MLH1 ⁄ MSH2 exons were completely sequenced in cases with microsatellite abnormalities in at least 40% of loci (high microsatellite instability) and ⁄ or complete loss of mlh1 ⁄ msh2 immunoreactivity,19,30 as well as in a representative sample from mlh1 ⁄ msh2 immunoreactive cases (20 UroC) used as controls.

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

MMR and UroC infiltration patterns

461

Microsatellite patterns mlh1 Superficial

Control

Deep

msh2

TSG microsatellite stable pathway

Infiltrative pT2 UroC

TSG microsatellite profile Infiltrative

Nod-trab

Urothelium

Normal MLH1 sequence

Normal MSH2 sequence 0% NAD

20%

40%

1 TSG Locus

60%

2 TSG loci

80%

100%

≥3 TSG loci

P = 0.01184

TSG microsatellite abnormal pathway

Microsatellite patterns Superficial Control

Deep

mlh1

msh2

Superficial Deep

Nodular-trabecular pT2 UroC Homogeneous

Heterogeneous

Somatic MMR protein down-regulation

Figure 2. Microsatellite pathways in urothelial carcinomas (UroC). Microsatellite patterns. Muscle-invasive transitional cell carcinomas (UroC) with infiltrative infiltration pattern. Gels show no microsatellite abnormalities of tumour suppressor genes (TSG) in either superficial or deep compartments. Muscle-invasive UroC with nodular-trabecular infiltration pattern. Gels show concordant (homogeneous) and discordant (heterogeneous) microsatellite patterns of tumour suppressor genes in the superficial and deep compartments. TP53, Tumour protein p53 (two loci); RB1, retinoblastoma; WT1, Wilm’s tumour 1; NF1, neurofibromatosis 1. TSG microsatellite profile. Number of TSG abnormalities detected in transitional cell carcinomas by infiltration patterns: infiltrative UroCs reveal two or more TSG loci with microsatellite abnormalities in 5 ⁄ 27 cases (18.5%), whereas nodular-trabecular UroCs show lower incidence of microsatellite lesions. Mismatch protein expression. Nuclear mlh1 and msh2 expression is demonstrated in UroC in the microsatellite stable pathway and at least one of these proteins is absent (in particular mlh1) in UroC showing microsatellite instability. MLH1 and MSH2 exon sequencing. Normal sequence is demonstrated for these genes, regardless of the microsatellite pattern.

i m m u n o h i s t o c h e m i c a l de t e c t i o n o f k i 6 7 , m lh 1 and msh 2 Sections were mounted on positively charged slides (Superfrost Plus; Fisher Scientific, Fair Lawn, NJ, USA), baked at 60 C for 2 h and processed as described.7,20,25 After routine dewaxing and rehydration, endogenous peroxidase quenching and antigen heat retrieval, the slides were transferred to a moist chamber. Non-specific binding was blocked with polyclonal horse serum and sections incubated with monoclonal primary antibodies (overnight, 4 C): 2 lg ⁄ ml MIB-1 (Calbiochem, Cambridge, MA, USA), hMLH1 clones G168 728 and G168-15 (BD PharMingen, San Jose, CA, USA) and hMSH2 clone FE11 (Oncogene Research,

La Jolla, CA, USA). Sections were then serially incubated with biotinylated antimouse antibody and peroxidase-labelled avidin–biotin complex. The reaction was developed under microscopic control, using 3,3¢diaminobenzidine tetrahydrochloride with 0.3% H2O2 as chromogen (Sigma Co., St Louis, MO, USA) and the sections counterstained with haematoxylin. Positive (reactive lymph node) and negative (omitting the primary antibody) controls were run simultaneously.

in situ en d l a b e l li n g o f f ra g me n t ed d n a Extensive DNA fragmentation associated with apoptosis was detected by in situ end labelling (ISEL), as reported.10,13 After routine dewaxing and hydration,

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

462

J Rubio et al.

Primers

Primer sequences

Tandem repeat ⁄ PCR product

TP53(1)-a*

5¢-AGG GAT ACT ATT CAG CCC-3¢

CA repeat ⁄

TP53(1)-b*

5¢-ACT GCC ACT CCT TGC CCC ATT C-3¢

103–135 bp

TP53(2)-a*

5¢-GAA TCC GGG AGG AGG TTG-3¢

AAAAT repeat ⁄

TP53(2)-b*

5¢-AAC AGC TCC TTT AAT GGC AG-3¢

140–175 bp

RB1-a*

5¢-CTC CTC CCC TAC TTA CTT GT-3¢

CTTT(T) repeat ⁄

RB1-b*

5¢-AAT TAA CAA GGT GTG GTG GTA CAC G-3¢

266–306 bp

WT1-a*

5¢-AAT GAG ACT TAC TGG GTG AGG-3¢

CA repeat ⁄

WT1-b*

5¢-TTA CAC AGT AAT TTC AAG CAA CGG-3¢

144 bp

NF1-a*

5¢-CAG AGC AAG ACC CTG TCT-3¢

CA repeat ⁄

NF1-b*

5¢-CTC CTA ACA TTT ATT AAC CTT A-3¢

171–187 bp

Table 1. Primer sequences and polymerase chain reaction (PCR) cycling conditions for the amplification of polymorphic DNA regions

All reactions were run in duplicate using 1.5 mM of MgCl2 and 1 ll of template. A long denaturation (4 min) and expansion (90 s) were used in the first three cycles for each set of primers. *The polymorphic regions of tumour suppressor gene were amplified using 0.25 lM of each primer, 50 lM of each dNTP (Boehringer-Mannheim, Indianapolis, IN, USA) and internally labelled with 0.3 lCi a32P-dCTP (3000 Ci ⁄ mmol, 10 mCi ⁄ mL) (New England Nucleotide, Boston MA, USA). The annealing temperature was 55 C for all primer sets (except NF1, for which it was 52 C) and the number of cycles was experimentally optimized to 26.

the sections were incubated in 2· standard saline citrate (20 min at 80 C) and digested with pronase (500 lg ⁄ ml, 25 min, room temperature) in a moist chamber. DNA fragments were labelled on 5¢-protuding termini by incubating the sections with the Klenow fragment of Escherichia coli DNA polymerase I [20 U ⁄ ml in 50 mmol ⁄ l Tris–HCl, pH 7.5, 10 mmol ⁄ l MgCl2, 1 mmol ⁄ l dithiothreitol, 250 lg ⁄ ml bovine serum albumin (BSA), 5 lm of each dATP, dCTP, dGTP, as well as 3.25 lmol ⁄ l dTTP and 1.75 lmol ⁄ l 11-digoxigenin-dUTP], at 37 C in a moist chamber. The incorporated digoxigenin–dUMPs were immunoenzymatically detected using antidigoxigenin Fab fragments labelled with alkaline phosphatase (7.5 U ⁄ ml, in 100 mmol ⁄ l Tris–HCl, pH 7.6, 150 mmol ⁄ l NaCl, 1% BSA) for 4 h at room temperature. The reactions were developed with the mixture nitroblue tetrazolium-X phosphate in 100 mmol ⁄ l Tris–HCl (pH 9.5), 100 mmol ⁄ l NaCl, 50 mmol ⁄ l MgCl2 under microscopic control. Appropriate controls were simultaneously run, including positive (reactive lymph node), negative (same conditions omitting DNA polymerase I) and enzymatic (DNase I digestion before the end labelling) controls. The enzymatic controls were used to establish the positivity threshold reliably in each sample.

n u c le a r d na q u an t i fi c a t i o n by s l id e c y t om e t r y Feulgen-stained sections were used for DNA quantification.31 Densitometric evaluation was performed with the cell analysis system model 200 and quantitative DNA analysis software package (Becton Dickinson, San Jose, CA, USA). At least 300 complete, non-overlapping and focused nuclei (or the whole lesion if smaller) were measured in every case, beginning in the most cellular area until completion in consecutive HPFs. External staining calibration was carried out with complete rat hepatocytes (Becton Dickinson; one slide per staining holder) to normalize the internal controls (lymphocytes and histologically normal urothelial cells present in the same tissue section), used for setting the G0 ⁄ G1 cell limits and calculating the DNA index of each G0 ⁄ G1 peak (> 10% of measured cells with evidence of G2+ M cells).32 Proliferation rate (PR ¼ S + G2 + M-phases fraction) was calculated from the DNA histogram by subtracting the number of cells within G0 ⁄ G1 limits from the total number of measured cells and expressed as a percentage.31,32 The scatter analysis of nuclear area and DNA content allowed apoptotic cell identification in each cell cycle phase (low nuclear area for a given DNA

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

MMR and UroC infiltration patterns

q u a n t i f i c at i o n o f p o s i t iv e n u c l e i a n d sta ti st ic al a na ly s i s At least 50 HPF (7.6 mm2) were screened in each pathological group, beginning in the most cellular area. The number of positive nuclei was expressed per HPF and per 1000 tumour cells, and the average and SD calculated in each pathological condition and patient as described.11,13,16 The positivity threshold was experimentally established from the positive control in each staining batch. Only nuclei with staining features similar to those of their corresponding positive control were considered positive for any marker. All variables were compared by infiltration pattern (nodular-trabecular versus infiltrative) in deep compartments and the cancer-specific survival assessed (Kaplan–Meier analysis). Qualitative variables were statistically tested using Fisher’s exact tests, whereas quantitative variables were compared using Student t-tests and analysis of variance. Differences were considered significant if P < 0.05 in two-tailed distributions.

Results Patients aged 59–78 years (64.3 ± 8.7) complained of painless haematuria (72 cases, 100%), irritative symptoms (15 cases, 20.8%), with no differences between infiltration patterns. Tumours revealed a nodulartrabecular infiltration pattern in 45 cases (63%, Figure 2) and infiltrative in 27 cases (37%, Figure 2), aneuploid nuclear DNA content in 59 cases (82%, always concordant in superficial and deep compartments) and high-grade in 62 cases (86%). LGUD was found in 18 cases and CIS in 12, both lesions being coexistent in five patients. LGUD was mainly associated with infiltrative UroC, whereas CIS was associated with nodular-trabecular UroC only (P ¼ 0.0053) (Table 2). No dysplastic changes were found in 47 UroCs (65%). Nodular-trabecular UroCs were more frequently aneuploid and high-grade than infiltrative UroCs (Table 2). The number of diploid (10 cases) and lowgrade (eight cases) UroCs precluded any statistical comparisons of these features. The median survival was significantly longer for nodular-trabecular than for infiltrative UroCs (Figure 3, P ¼ 0.0445). Nodular-

Table 2. Muscle-invasive urothelial carcinomas by infiltration patterns: morphological and DNA cytometry features Nodulartrabecular Infiltrative pT2a ⁄ Significance, pT2a ⁄ b UroC P b UroC Tumour grade Low

0.0329

High Nuclear grade Low

0.0329

HIgh Urothelial dysplasia No dysplasia Low-grade CIS DNA ploidy Diploid

0.0004

< 0.0001

Aneuploid CIS, Carcinoma in situ.

Cancer-specific survival

content)33 and was coupled with ISEL to identify apoptotic DNA fragmentation (see above). External diploid controls were used to determine DNA indices (lymphocytes from reactive lymph nodes) and to standardize the nuclear area ⁄ DNA content analysis (normal transitional cells).33

463

Median survival

0.9

Nodular-trabecular UroC – 37 months Infiltrative UroC – 20 months

0.8 0.7 0.6

P = 0.0445

0.5 0.4 0.3 0.2 0.1 0 6

12 18 24 30 36 42 48 54 60 66 70 Time (months)

Figure 3. Cancer-specific survival in patients with muscle-invasive transitional cell carcinomas (UroC) by infiltration pattern (nodulartrabecular versus infiltrative).

trabecular UroCs and superficial compartments showed significantly higher values for both proliferation and apoptosis markers (Table 3). Nodular-trabecular UroCs revealed more abnormal loci than infiltrative UroCs (Figure 2, P ¼ 0.0001). Discordant genetic patterns by tumour compartments were observed in only three infiltrative UroCs, precluding any statistical assessment, but all showed nuclear

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

464

J Rubio et al.

Table 3. Kinetic features of muscle-invasive urothelial carcinomas (UroC) by infiltration patterns (nodular-trabecular versus infiltrative) and tumour compartments (superficial versus deep)

MF counting

Nodular-trabecular pT2a ⁄ b UroC

Infiltrative pT2a ⁄ b UroC

Superficial

Deep

Significance, P

9.0 ± 5.1

4.1 ± 3.1

5.9 ± 3.5

2.0 ± 1.8

0.012

Ki67 index

28.25 ± 6.01

13.09 ± 4.89

25.12 ± 6.22

10.76 ± 4.28

0.005

Proliferation rate

33.74 ± 7.60

19.69 ± 6.15

26.46 ± 8.36

14.57 ± 4.58

0.009

21.1 ± 22.5

19.3 ± 22.9

11.0 ± 4.6

17.2 ± 15.4

0.048

ISEL index

MF, Mitotic figure; ISEL, in situ end labelling.

TP53 expression and more LOH ⁄ SNP(s) in the deep compartment (WT1 LOH ⁄ SNP in two and NF1 LOH ⁄ SNP in one). The distribution of LOH ⁄ SNP(s) was also significantly different according to the infiltration pattern: nodular-trabecular UroC revealed a higher proportion of abnormal RB1 (P ¼ 0.0003) and NF1 (P ¼ 0.0023) loci (Figure 4). Considering all genetic lesions equally important, the normal tissue ROH probability was PROH ¼ 1 ) PLOH ¼ 1 ) 0.2 ¼ 0.8 for a given marker,34–36 0.85 for five markers and (0.85)2 for concordant superficial and deep compartments results of those five markers.7,20 It will result in [(0.85)2]12 for 12 UroC and [(0.85)2]12 · 0.3812 ¼ 2.85 · 1017 if they are infiltrative UroC (38% of muscle-invasive UroC). At least one MMR protein (always including mlh1) was not expressed in nodular-trabecular UroC with two or more abnormal TSG loci, but both MMR proteins were present in tumours with less than two abnormal TSG loci. MLH1 and MSH2 exon sequencing revealed no mutations, regardless of the number of abnormal TSG loci (Figure 2).

Discussion Somatic down-regulation of MMR proteins in nodulartrabecular muscle-invasive UroC results in microsatellite abnormalities characterized by deletion ⁄ SNP(s) in RB1 and NF1, which correlates with higher cellular turnover and longer survival for these patients. Infiltrative UroC showed a low incidence of TSG LOH ⁄ SNP, whereas nodular-trabecular UroC accumulated TSG alterations (Figure 2). Technical reasons were excluded. The sensitivity threshold of our optimized protocol was 1% for positive detection,7,11,22,26 which applied to 100+ cell samples would result in false-negative results for DNA samples smaller than one cell equivalent. This is probably clinically irrelevant and frequently related to contamination. Repea-

ted microdissection under microscopic control with the same results and multiple sampling excluded any significant contamination with normal tissue.11,20,22,26,28 Although TSG LOH ⁄ SNP(s) can be present outside of the screened introns, the importance of these results is still supported by two facts: the low probability of it as a random finding and its significant association with infiltrative UroC and shorter survival. MMR protein down-regulation and abnormal TSG microsatellites characterized nodular-trabecular UroC with CIS (12 cases, P ¼ 0.0053) and deep UroC compartments,25,37,38 correlating with lack of mlh1 ⁄ msh2 immunoexpression and normal gene sequences. MMR proteins normally identify and correct mismatched DNA sequences that can occur during DNA replication.30 MMR protein down-regulation in deep compartments and nodular-trabecular UroC would contribute to: (i) lower DNA indices and decreased prevalence of aneuploid cell lines detected in neoplasms with microsatellite abnormalities,39,40 which frequently show diploid DNA content,25,37,41 and loss of the physiological cell kinetic correlations in deep compartments;37 and (ii) tumour cell heterogeneity, genetic instability and biological progression, which must be studied with several samples of sufficient size from each tumour.18,21,22,26 Because of intratumoral heterogeneity, at least two samples from each tumour should be screened, preferably from superficial and deep compartments to allow for topographical heterogeneity.18,21,22,26,42,43 MMR gene inactivation (by either mutation or protein down-regulation) leading to mutation accumulation (as proven in this series in TSG) and molecular progression,19,22,44,45 not necessarily independent of chromosomal instability, may coexist in a given neoplasm and show a significant degree of overlap.46 Nodular-trabecular UroC revealed significantly higher proliferation and apoptosis than infiltrative UroC, which would contribute to a relative sensitivity to

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

MMR and UroC infiltration patterns

465

MS profile in pT2 UroC by infiltration pattern

A NF1 (Infilt)

NF1 (Nod-trab)

WT1 (Infilt)

WT1 (Nod-trab)

RB1 (Infilt) 0

RB1 (Nod-trab)

TP53 (Infilt)

TP53 (Nod-trab)

40%

20%

60%

LOH/SNP Concordant

UroC with topographically homogeneous MS pattern

ROH

32%

UroC with topographically heterogeneous MS pattern

NF1, D<S

NF1 WT1

RB1

WT1, D<S

RB1, D<S

WT1, D>S

TP53

13 5

RB1, D>S 0

TP53, D>S

50% LOH/SNP

100%

TP53, D<S

NF1, D>S

100%

Discordant

68%

80%

20%

ROH

40%

60%

LOH/SNP

80%

100%

ROH

Figure 4. Distribution of microsatellite abnormalities in each tumour suppressor gene (TSG) in urothelial carcinoma (UroC) by infiltration patterns (A, solid-nodular versus infiltrative). NF1 loss of heterozygosity (LOH) ⁄ single nucleotide polymorphism(s) (SNP) was more frequently found in superficial compartments and TP53 LOH ⁄ SNP(s) in deep compartments, although only TP53 showed significant differences by tumour compartments (P ¼ 0.0213). UroC showed topographically concordant microsatellite (MS) patterns in 49 cases (68%, B) and discordant in 23 cases (32%, C), TSG LOH ⁄ SNP average being higher in the former; the variability, however, was higher in the latter group. Significant differences were demonstrated only for the comparisons between UroCs with more MS abnormalities in the deep compartment (15 cases) and (A) UroCs with no topographical heterogeneity (49 cases; TP53, P ¼ 0.0322; RB1, P ¼ 0.0001; and NF1, P ¼ 0.0017), and (B) UroCs with more LOH ⁄ SNP(s) in the superficial compartment (eight cases; TP53, P ¼ 0.0363; and NF1, P ¼ 0.0450) (c). Comparison between UroCs with no topographical heterogeneity and UroCs with more LOH ⁄ SNP(s) in the superficial compartment revealed no statistically significant differences. TP53, Tumour protein p53; RB1, retinoblastoma; WT1, Wilm’s tumour 1; NF1, neurofibromatosis 1; ROH, retention of heterozygosity.

conventional treatment and eventually longer survival (Figure 3), as reported for neoplasms with TSG microsatellite abnormalities.30 Nodular-trabecular UroC

more frequently revealed high tumour ⁄ nuclear grade, aneuploid DNA content (Table 2) and higher proliferation and apoptotic indices. Lack of MMR protein

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

466

J Rubio et al.

expression in these neoplasms would lead to accumulation of genetic alterations, which, reaching lethal limits, results in increased apoptosis and, eventually, a better response to therapy.19,30 The aneuploid DNA content and proliferation rate are directly related to tumour ⁄ nuclear grade,7,20,25 both predominating in nodular-trabecular UroC; however, the up-regulated apoptosis in this subgroup of muscle-invasive UroC would account for the paradoxically better survival of these patients. Nodular-trabecular UroCs were significantly associated with topographical heterogeneity, RB1 and NF1 LOH ⁄ SNP(s) and higher cellular turnover.7,8,25 It has been postulated that aberrant pRB1 expression deregulates G1 cell cycle checkpoint and provides tumour cells with increased proliferation and a reduced response to programmed cell death.47 However, high levels of pRB1 expression may reflect a dysfunctional RB1 pathway and do not necessarily reflect the tumour suppressor effects of the protein.48 Nevertheless, the absence of an inhibitory effect of functional pRB1 leads to increased proliferation in nodular-trabecular UroC. The presence of NF1 LOH ⁄ SNP was especially documented in the superficial compartment of nodulartrabecular UroC. The NF1 gene product has an inhibitory effect on RAS, whose protein is highly expressed in immature and proliferating cells, and the lack of its inhibitory effect will favour increased cell proliferation,8,25,49 also confirmed by decreased NF1 mRNA and protein levels in high-grade UroC, suggesting an NF1 role in bladder carcinogenesis.7,50 These findings are an expression of the disturbed tumour kinetics,10,11,13 which result in high cellular turnover and the preferentially expansive nature of nodulartrabecular UroC. In conclusion, nodular-trabecular muscle-invasive UroC reveals greater proliferation and a higher incidence of RB1 and NF1 LOH ⁄ SNP(s) than infiltrative UroC, together with longer survival. A significantly low incidence of TSG LOH ⁄ SNP(s) suggests a microsatellite stable pathway for infiltrative UroC.

10.

11.

12. 13.

14.

15.

Acknowledgements Presented in part as abstract at the Meeting of the Pathological Society of Great Britain and Ireland, London, UK, 2000 and USCAP Meetings in New Orleans, Louisiana, 2000 and in San Antonio, Texas, 2005.

16.

17.

References 1. Cohen MB, Waldman FM, Carroll PR, Kerschmann R, Chew K, Mayall BH. Comparison of five histopathologic methods to assess

cellular proliferation in transitional cell carcinoma of the urinary bladder. Hum. Pathol. 1993; 24; 772–778. Hierro Martin I, Alvarez Perez M, Blanes Berenguel A et al. Contribucion a la categorizacion de los CCT de vejiga, sistematizacion de la gradacion y utilidad de la ploidia de ADN. Actas Urol. Esp. 1999; 23; 489–496. Jimenez RE, Gheiler E, Oskanian P et al. Grading the invasive component of urothelial carcinoma of the bladder and its relationship with progression-free survival. Am. J. Surg. Pathol. 2000; 24; 980–987. Smits G, Schaafsma E, Kiemeney L, Caris C, Debruyne F, Witjes JA. Microstaging of pT1 transitional cell carcinoma of the bladder: identification of subgroups with distinct risks of progression. Urology 1998; 52; 1009–1013; discussion 1013– 1004. Younes M, Sussman J, True LD. The usefulness of the level of the muscularis mucosae in the staging of invasive transitional cell carcinoma of the urinary bladder. Cancer 1990; 66; 543–548. Hermann GG, Horn T, Steven K. The influence of the level of lamina propria invasion and the prevalence of p53 nuclear accumulation on survival in stage T1 transitional cell bladder cancer. J. Urol. 1998; 159; 91–94. Diaz-Cano SJ, Blanes A, Rubio J, Matilla A, Wolfe HJ. Molecular evolution and intratumor heterogeneity by topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder. Lab. Invest. 2000; 80; 279–289. Baithun SI, Naase M, Blanes A, Diaz-Cano SJ. Molecular and kinetic features of transitional cell carcinomas of the bladder. Biological and clinical implications. Virchows Arch. 2001; 438; 289–297. Pozo L, Camacho F, Rios-Martin JJ, Diaz-Cano SJ. Cell proliferation in skin tumors with ductal differentiation: patterns and diagnostic applications. J. Cutan. Pathol. 2000; 27; 292–297. Diaz-Cano SJ, Garcia-Moliner M, Carney W, Wolfe HJ. Bcl-2 expression and DNA fragmentation in breast carcinoma, pathologic and steroid hormone receptors correlates. Diagn. Mol. Pathol. 1997; 6; 199–208. Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ. Clonality as expression of distinctive cell kinetic patterns in nodular hyperplasias and adenomas of the adrenal cortex. Am. J. Pathol. 2000; 156; 311–319. Salomon RN, Diaz-Cano S. Introduction to apoptosis. Diagn. Mol. Pathol. 1995; 4; 235–238. Koch M, de Miguel M, Ho¨fler H, Diaz-Cano SJ. Kinetic profiles of intraepithelial and invasive prostatic neoplasias: the key role of downregulated apoptosis in tumor progression. Virchows Arch. 2000; 436; 413–420. Ro JY, Ayala AG, el-Naggar A. Muscularis mucosa of urinary bladder. Importance for staging and treatment. Am. J. Surg. Pathol. 1987; 11; 668–673. van Diest PJ, Baak JP, Matze-Cok P et al. Reproducibility of mitosis counting in 2,469 breast cancer specimens: results from the Multicenter Morphometric Mammary Carcinoma Project. Hum. Pathol. 1992; 23; 603–607. Harjacek M, Diaz-Cano S, Alman BA et al. Prominent expression of mRNA for proinflammatory cytokines in synovium in patients with juvenile rheumatoid arthritis or chronic lyme arthritis. J. Rheumatol. 2000; 27; 497–503. Epstein JI, Amin MB, Reuter VR, Mostofi FK. The World Health Organization ⁄ International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference Committee. Am. J. Surg. Pathol. 1998; 22; 1435–1448.

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

MMR and UroC infiltration patterns

18. Blanes A, Diaz-Cano SJ. DNA and kinetic heterogeneity during the clonal evolution of adrenocortical proliferative lesions. Hum. Pathol. 2006; 37; 1295–1303. 19. Blanes A, Sanchez-Carrillo JJ, Diaz-Cano SJ. Topographic molecular profile of pheochromocytomas: role of somatic downregulation of mismatch repair. J. Clin. Endocrinol. Metab. 2006; 91; 1150–1158. 20. Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ. Clonal patterns in phaechromocytomas and MEN-2A adrenal medullary hyperplasias: histologic and kinetic correlates. J. Pathol. 2000; 192; 221–228. 21. Diaz-Cano SJ. Are PCR artifacts in microdissected samples preventable? Hum. Pathol. 2001; 32; 1415. 22. Diaz-Cano SJ, Blanes A, Wolfe HJ. PCR techniques for clonality assays. Diagn. Mol. Pathol. 2001; 10; 24–33. 23. Diaz-Cano SJ, Brady SP. DNA extraction from formalin-fixed, paraffin-embedded tissues: protein digestion as a limiting step for retrieval of high-quality DNA. Diagn. Mol. Pathol. 1997; 6; 342–346. 24. Cawkwell L, Lewis FA, Quirke P. Frequency of allele loss of DCC, p53, RBI, WT1, NF1, NM23 and APC ⁄ MCC in colorectal cancer assayed by fluorescent multiplex polymerase chain reaction. Br. J. Cancer 1994; 70; 813–818. 25. Blanes A, Rubio J, Martinez A, Wolfe HJ, Diaz-Cano SJ. Kinetic profiles by topographic compartments in muscle-invasive transitional cell carcinomas of the bladder: role of TP53 and NF1 genes. Am. J. Clin. Pathol. 2002; 118; 93–100. 26. Diaz-Cano SJ. Designing a molecular analysis of clonality in tumours. J. Pathol. 2000; 191; 343–344. 27. Mutter GL, Boynton KA. X chromosome inactivation in the normal female genital tract: implications for identification of neoplasia. Cancer Res. 1995; 55; 5080–5084. 28. Diaz-Cano SJ. Clonality studies in the analysis of adrenal medullary proliferations. Application principles and limitations. Endocr. Pathol. 1998; 9; 301–316. 29. Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Wolfe HJ. Germline RET 634 mutation positive MEN 2A-related C-cell hyperplasias have genetic features consistent with intraepithelial neoplasia. J. Clin. Endocrinol. Metab. 2001; 86; 3948–3957. 30. Blanes A, Diaz-Cano SJ. Complementary analysis of microsatellite tumor profile and mismatch repair defects in colorectal carcinomas. World J. Gastroenterol. 2006; 12; 5932–5940. 31. Bibbo M, Bartels PH, Dytch HE, Wied GL. Cell image analysis. In Bibbo M ed. Comprehensive cytopathology. Philadelphia: Saunders Co., 1991; 965–983. 32. Dressler LG. Controls, standards, and histogram interpretation in DNA flow cytometry. Meth. Cell. Biol. 1990; 33; 157–171. 33. Sherwood SW, Schimke RT. Cell cycle analysis of apoptosis using flow cytometry. Meth. Cell Biol. 1995; 46; 77–97. 34. Chen LC, Kurisu W, Ljung BM, Goldman ES, Moore DI, Smith HS. Heterogeneity for allelic loss in human breast cancer. J. Natl Cancer Inst. 1992; 84; 506–510. 35. Deng G, Lu Y, Zlotnikov G, Thor AD, Smith HS. Loss of heterozygosity in normal tissue adjacent to breast carcinomas. Science 1996; 274; 2057–2059.

467

36. Diaz-Cano SJ, Wolfe HJ. Clonality in Kaposi’s sarcoma [letter; comment]. N. Engl. J. Med. 1997; 337; 571–572. 37. Blanes A, Jimenez-Martin JJ, Miranda MT, Diaz-Cano SJ. Absence of the physiologic cellular kinetic balance and down-regulation of mlh1 ⁄ msh2 characterize deep topographic compartments of colo-rectal adenocarcinomas. Lab. Invest. 2003; 83; 114A–115A (Abstract). 38. Jimenez JJ, Blanes A, Diaz-Cano SJ. Microsatellite instability in colon cancer. N. Engl. J. Med. 2003; 349; 1774–1776; author reply 1774–1776. 39. Liang JT, Chang KJ, Chen JC et al. Clinicopathologic and carcinogenetic appraisal of DNA replication error in sporadic T3N0M0 stage colorectal cancer after curative resection. Hepatogastroenterology 1999; 46; 883–890. 40. Thibodeau SN, French AJ, Cunningham JM et al. Microsatellite instability in colorectal cancer: different mutator phenotypes and the principal involvement of hMLH1. Cancer Res. 1998; 58; 1713–1718. 41. Jimenez-Martin JJ, Miranda MT, Blanes A, Diaz-Cano SJ. Expression of mlh1 ⁄ msh2 in colo-rectal adenocarcinomas: SAGE ⁄ microarray profile and tumor phenotype. Lab. Invest. 2003; 83; 296A (Abstract). 42. Chapusot C, Martin L, Bouvier AM et al. Microsatellite instability and intratumoural heterogeneity in 100 rightsided sporadic colon carcinomas. Br. J. Cancer 2002; 87; 400–404. 43. Chapusot C, Martin L, Mungra N et al. Sporadic colorectal cancers with defective mismatch repair display a number of specific morphological characteristics: relationship between the expression of hMLH1 and hMSH2 proteins and clinicopathological features of 273 adenocarcinomas. Histopathology 2003; 43; 40–47. 44. Duval A, Hamelin R. Mutations at coding repeat sequences in mismatch repair-deficient human cancers: toward a new concept of target genes for instability. Cancer Res. 2002; 62; 2447–2454. 45. Peltomaki P. DNA mismatch repair and cancer. Mutat. Res. 2001; 488; 77–85. 46. Goel A, Arnold CN, Niedzwiecki D et al. Characterization of sporadic colon cancer by patterns of genomic instability. Cancer Res. 2003; 63; 1608–1614. 47. Cordon-Cardo C, Zhang ZF, Dalbagni G et al. Cooperative effects of p53 and pRB alterations in primary superficial bladder tumors. Cancer Res. 1997; 57; 1217–1221. 48. Cote RJ, Dunn MD, Chatterjee SJ et al. Elevated and absent pRb expression is associated with bladder cancer progression and has cooperative effects with p53. Cancer Res. 1998; 58; 1090– 1094. 49. Furth ME, Aldrich TH, Cordon-Cardo C. Expression of ras protooncogene proteins in normal human tissues. Oncogene 1987; 1; 47–58. 50. Aaltonen V, Bostrom PJ, Soderstrom KO et al. Urinary bladder transitional cell carcinogenesis is associated with down-regulation of NF1 tumor suppressor gene in vivo and in vitro. Am. J. Pathol. 1999; 154; 755–765.

2007 The Authors. Journal compilation 2007 Blackwell Publishing Ltd, Histopathology, 51, 458–467.

0021-972X/06/$15.00/0 Printed in U.S.A.

Topographic Molecular Profile of Pheochromocytomas: Role of Somatic Down-Regulation of Mismatch Repair Alfredo Blanes, Juan J. Sanchez-Carrillo, and Salvador J. Diaz-Cano Department of Pathology, University of Malaga School of Medicine (A.B., J.J.S.-C., S.J.D.-C.), Malaga E29010, Spain; and Department of Pathology, King’s College Hospital and King’s College School of Medicine (S.J.D.-C.), London SE5 9RS, United Kingdom Context and Objective: Despite extensive molecular investigation of adrenal pheochromocytomas, no information is available on their molecular and mismatch repair (MMR) profiles by topographic compartments. Design and Setting: Microdissected samples from the peripheral and internal zones of 143 pheochromocytomas from a referral hospital (95 sporadic and 48 associated with multiple endocrine neoplasia type 2A) were selected for loss of heterozygosity and single nucleotide polymorphism analyses. Five polymorphic DNA regions from TP53, RB1, WT1, and NF1 were systematically studied by PCR-denaturing gradient gel electrophoresis. Patients, Outcome Measures, and Interventions: Pheochromocytomas were classified as malignant (16 sporadic tumors with distant metastases), locally invasive (30 sporadic tumors showing retroperitoneal infiltration only), and benign (all remaining tumors). Statistical differences were evaluated using Fisher’s exact test. MMR was assessed by MLH1/MSH2 sequencing and immunostaining in pheochromocytomas with two or more abnormal microsatellites. No interventions were performed in this study.

HEOCHROMOCYTOMAS (PCC) ARE heterogeneous tumors (1– 4) that show clonal expansions due to loss of heterozygosity (LOH) of several tumor suppressor genes (TSG) located on chromosomes 1p, 3p, 17p, and 22q (5–9). Those markers have revealed significant association with clinicopathological parameters, such as tumor volume (5), or distinctive transformation pathways (7), although their relative incidence is quite variable because of the limited number of cases analyzed. A monoclonal tumor origin is supported by concordant TSG abnormalities, such as point mutations or single nucleotide polymorphisms (SNPs) (10, 11), which have been found associated with LOH of certain loci (12). The coexistence of several genetic abnormalities and intratumor heterogeneity would be the expression of either tumor cell selection or a simple passive byproduct of genetic instability (10, 13, 14). However, the association of multiple genetic alterations

First Published Online January 4, 2006 Abbreviations: DGGE, Denaturing gradient gel electrophoresis; ISEL, in situ end labeling; LOH, loss of heterozygosity; MEN 2A, multiple endocrine neoplasia type 2A; MMR, mismatch repair; NF1, neurofibromatosis 1; PCC, pheochromocytoma; RB1, retinoblastoma; ROH, retention of heterozygosity; SNP, single nucleotide polymorphism; TP53, tumor protein p53; TSG, tumor suppressor gene; WT1, Wilms tumor 1. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

Results: Loss of heterozygosity/single nucleotide polymorphism involved TP53 in 40 of 134 informative cases (29.9%), RB1 in 22 of 106 informative cases (20.8%), WT1 in 32 of 120 informative cases (26.7%), and NF1 in 32 of 80 informative cases (40.0%). More genetic abnormalities involving the peripheral compartment were revealed in 34 pheochromocytomas (23.8%): 12 of 16 malignant, 10 of 30 locally invasive, and 12 of 97 benign. Multiple and coexistent genetic abnormalities characterized malignant pheochromocytomas (P ⬍ 0.001), whereas locally invasive pheochromocytomas showed a significantly higher incidence of NF1 alterations (P ⬍ 0.001). No mutations were identified in MLH1/MSH2, but MMR proteins significantly decreased in peripheral compartments. Conclusions: Multiple microsatellite alterations and topographic intratumor heterogeneity characterize malignant pheochromocytomas, suggesting a multistep tumorigenesis through somatic topographic down-regulation of MMR proteins. Locally invasive pheochromocytomas reveal topographic heterogeneity and single-locus microsatellite alterations, especially involving NF1. (J Clin Endocrinol Metab 91: 1150 –1158, 2006)

would become statistically less probable as the number of molecular markers increases (10, 11, 13, 15) and is useful to test clonal expansions in tumors (10, 11). Although genetic abnormalities are probably asymmetrically acquired (16), there is a correlation with tumor cell topography, as demonstrated in bladder and colon (13); the topography role has not been studied in the adrenal gland. Any potential topographic segregation of tumor cells will influence interpretations of the results and would help in designing more effective therapies to maximize the effect in the most sensitive areas (e.g. zones with higher proliferation). No information is available on the molecular and mismatch repair (MMR) profiles of PCC by topographic compartments. This study investigates the TSG microsatellite pattern in sporadic and multiple endocrine neoplasia (MEN) 2A-associated PCC, using microdissected samples from the peripheral and internal compartments to assess both LOH and somatic SNP of TSG controlling G1-S transition (TP53, RB1), RAS pathway (NF1), and development (WT1). Patients and Methods Case selection Sporadic (n ⫽ 95) and MEN 2A (n ⫽ 48) PCC were included in this study. MEN 2A patients revealed adrenal medullary hyperplasia (17), but only nodules larger than 1 cm were included in this study (3). Standard protocols were followed for PCC sectioning and sampling

1150 Downloaded from jcem.endojournals.org on March 13, 2006

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

(at least one block per centimeter of tumor), being appropriate archival material available in all cases. The same areas in consecutive sections were used in each study, and their cellular composition was confirmed in adjacent hematoxylin-eosin-stained sections. This protocol was approved by the Hospital Research Board and ethical committee and complied with their requirements.

TSG microsatellite analysis DNA was extracted from the most cellular areas of peripheral and internal compartments after microdissecting at least 100 cells (⬃0.4 mm2) from two 20-␮m unstained paraffin sections/compartment (Fig. 1). Appropriate controls were included for each test (adrenal medulla, adrenal cortex, and periadrenal soft tissue). DNA was extracted using a modified phenol-chloroform protocol, precipitated with ice-cold absolute ethanol, and resuspended in 10 ␮l Tris-HCl buffer (pH 8.4) (18). DNA was then used for PCR amplification of TSG intron microsatellites (Table 1) (13, 19). The tests were run in a PerkinElmer thermal cycler model 480 (PerkinElmer, Norwalk, CT). The entire 10-␮l PCR volume was electrophoresed into 8% denaturing gradient polyacrylamide gels; dried gels were put inside developing cassettes containing one intensifying screen and preflashed films (Kodak XAR, Eastman Kodak, Inc., Rochester, NY) (13, 19, 20). The radiographs were developed using an automated processor Kodak O-MAT 100 (Eastman Kodak Co.). Interpretation and inclusion criteria in each sample were previously reported (11, 13, 21–23). Allelic imbalance was densitometrically evaluated (EC model 910 optical densitometer, EC Apparatus Corp., St. Petersburg, FL), considering evidence of only LOH allele ratios of 4:1 or more in any TSG; otherwise, retention of heterozygosity (ROH) was assigned (13, 22). This ratio represents 80% of clonal cells in the sample and was used to increase the detection specificity (10, 11, 23). Additional allele bands present in tumor samples, but not in the corresponding controls, were considered evidence of somatic SNP by PCR/denaturing gradient gel electrophoresis (DGGE) (13).

DNA sequencing All extra bands were cut from gels, and DNA was purified using a QIAquick gel extraction kit (QIAGEN, Valencia, CA). The amplified

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158 1151

product was diluted 20-fold in Tris-EDTA buffer, and 1 ␮l of the diluted reaction product was subjected to a second round of PCR amplification using the appropriate primers for 30 cycles under the above conditions. Normal and extra bands from tumor-derived samples were PCR amplified along with the corresponding controls using a high-fidelity polymerase, Platinum PFX (Invitrogen Life Technologies, Inc., Carlsbad, CA). PCR products were directly sequenced after purification (QIAquick PCR purification kit, QIAGEN). All sequencing was performed on an ABI PRISM 3700 automated DNA analyzer (Applied Biosystems, Foster City, CA), and the sequence data were analyzed using the program Sequencher (Gene Codes Corp., Ann Arbor, MI), which reverses and complements the antisense strand. All mutations were confirmed by sequencing in both directions and are indicated by an N in the sequencing chromatogram. MLH1/MSH2 exons were completely sequenced in cases with microsatellite abnormalities in at least 40% loci and/or complete loss of mlh1/msh2 immunoreactivity as well as in a representative sample from mlh1/msh2-immunoreactive cases (30 PCC), which was used as the control group.

Immunohistochemical expression of mlh1 and msh2 The sections were mounted on positively charged slides (SuperFrost Plus, Fisher Scientific, Fair Lawn, NJ), baked at 60 C for 2 h, and processed as previously described (13, 20, 24). After routine dewaxing and rehydration, endogenous peroxidase quenching, and antigens heat retrieval, the slides were transferred to a moist chamber. Nonspecific binding was blocked with polyclonal horse serum, and sections were incubated with monoclonal primary antibodies (overnight, 4 C): 2 ␮g/ml for hMLH1 (clones G168 728 and G168-15, BD Pharmingen, San Diego, CA) and hMSH2 (clone FE11, Oncogene Research Products, San Diego, CA). Then sections were serially incubated with biotinylated antimouse antibody and peroxidase-labeled avidin-biotin complex. The reaction was developed under microscopic control, using 3,3⬘-diaminobenzidine tetrahydrochloride with 0.3% H2O2 as chromogen (SigmaAldrich Corp., St. Louis, MO), and the sections counterstained with hematoxylin. Both positive (reactive lymph node) and negative (omitting the primary antibody) controls were simultaneously run.

Quantification of positive nuclei At least 50 high-power fields (7.6 mm2) were screened in each pathological group, beginning in the most cellular area. The number of positive nuclei was expressed per high-power field and per 1000 tumor cells, and the average and sd were calculated in each pathological condition and patient as previously described (25, 26). The positivity threshold was experimentally established at the positive control in each staining batch. Only nuclei with staining features similar to those of their corresponding positive control were considered positive for any marker.

Statistical analysis The results were compared by tumor compartment (peripheral vs. internal) in PCC classified by the presence or absence of genetic heterogeneity, the genetic background (sporadic vs. MEN 2A), and the biological behavior (nonmetastatic vs. metastatic). Qualitative variables were compared using Fisher’s exact tests, and quantitative variables were compared by Student’s t tests and ANOVA. Differences were considered significant at P ⬍ 0.05 in two-tail distributions.

Results

FIG. 1. PCC were analyzed by topographic compartments. The peripheral compartment comprised the 2.5-mm width tissue next to the transition between the tumor and the surrounding gland, whereas the internal compartment was the remaining tumor tissue inner to the peripheral rim. The same areas from these compartments were separately evaluated regarding microsatellites profile, sequencing, and immunoexpression of mlh1 and msh2. The brown dots in the inset represent the microscopic fields examined for the mlh-1 and msh-2 immunoexpression evaluations.

Malignant PCC (16 sporadic PCC) had histologically confirmed liver metastases and elevated catecholamine levels during follow-up (Table 2), whereas locally invasive PCC (30 sporadic PCC) showed retroperitoneal soft tissue infiltration. Benign PCC (49 sporadic and 48 MEN 2A neoplasms) had no evidence of extraadrenal tumor growth and normal catecholamine levels during follow-up (⬍100 ␮g/24 h). The tumors were found in 76 males (51 sporadic and 25 MEN 2A) and 67 females (44 sporadic and 23 MEN 2A), and all MEN

Downloaded from jcem.endojournals.org on March 13, 2006

1152

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

TABLE 1. Primer sequences and PCR cycling conditions for the amplification of polymorphic DNA regions of TSGs Primers

Primer sequences (5⬘–3⬘)

TP53(1)-a TP53(1)-b TP53(2)-a TP53(2)-b RB1-a RB1-b WT1-a WT1-b NF1-a NF1-b

AGG ACT GAA AAC CTC AAT AAT TTA CAG CTC

GAT GCC TCC AGC CTC TAA GAG CAC AGC CTA

ACT ACT GGG TCC CCC CAA ACT AGT AAG ACA

ATT CCT AGG TTT TAC GGT TAC AAT ACC TTT

CAG TGC AGG AAT TTA GTG TGG TTC CTG ATT

CCC CCC TTG GGC CTT GTG GTG AAG TCT AAC

Tandem repeat/ PCR product (bp)

CA repeat/103–135 ATT C AAAAT repeat/140 –175 AG GT GTA CAC G AGG CAA CGG

CTTT(T) repeat/266 –306 CA repeat/⬃144 CA repeat/171–187

CTT A

All reactions were run with 1.5 mM of MgCl2, using 0.25 ␮M of each primer and 50 ␮M of each dNTP for the polymorphic regions of TSG tested. The PCR products were internally labeled with 0.3 ␮Ci ␣关32P兴dCTP (3000 Ci/mmol, 10 mCi/ml, New England Nucleotide, Boston MA). All reactions were run in duplicate using 1 ␮l of template. A long denaturation (4 min) was used in the first three cycles for each set of primers. The annealing temperature was 55 C for all primer sets (except for NF1, for which it was 52 C). The number of cycles was experimentally optimized to 26 for the polymorphic regions of TSG.

2A patients had either medullary thyroid carcinoma (36 cases) or C cell hyperplasia (12 cases). Microsatellite alterations were observed in 80 PCC (55.9%; 45 sporadic and 35 MEN 2A). TSG microsatellite analysis revealed TP53 alterations in 39 of 131 informative cases (29.8%, screening two introns), RB1 abnormalities in 22 of 104 informative cases (21.2%), WT1gene lesions in 31 of 118 informative cases (26.3%), and NF1 alterations in 31 of 78 informative cases (39.7%). No TSG microsatellite abnormalities were revealed in 63 PCC (44.1%, only two metastatic). The same microsatellite locus was involved in both compartments of 22 PCC (15.4%): TP53 in 10 benign cases (7.0%) and WT1 in 12 cases (8.4%, only two metastatic). Microsatellite abnormality in one TSG in the peripheral compartment only was observed involving NF1 locus in 14 nonmetastatic PCC (9.8%), TP53 in four benign PCC (2.8%), and WT1 locus in two benign PCC (1.4%). The remaining PCC had two TSG loci altered in 20 cases (14.0%, eight malignant and 12 benign PCC) and three TSG loci in 20 cases (14.0%, four of them metastatic). Mutations were confirmed by sequencing of the corresponding introns of all cases with microsatellite abnormalities (Fig. 2), but were not detected in control tissues, confirming their somatic nature. Concordant TSG microsatellite patterns in both tumor

compartments were observed in 110 PCC (76.9%; Fig. 2), and topographic genetic heterogeneity (LOH and/or somatic SNP) were found in 33 cases (23.1%; Fig. 2 and Table 3): 12 benign, nine locally invasive, and 12 malignant. PCC with discordant microsatellite pattern revealed more TSG loci involved in the peripheral compartment: NF1 locus in 16 nonmetastatic PCC (11.2%), TP53 in six benign PCC (4.2%), and variable combinations of TP53-RB1-WT1-NF1 in 12 malignant PCC (8.4%; Table 3). PCC with concordant microsatellite pattern of TSG showed demonstrable genetic alterations in at least one TSG locus in 47 cases (32.9%, only two of them malignant). Finally, a subset of PCC (63 cases, 44.1%, two malignant only) revealed no alterations in the TSG loci analyzed. PCC with topographic genetic heterogeneity showed significantly higher incidence of TP53 (P ⫽ 0.00933) and NF1 (P ⬍ 0.00001) alterations at the peripheral compartment than topographically homogeneous PCC (Table 4). TP53 topographic heterogeneity (Tables 3 and 4) was found associated with other genetic alterations in malignant PCC, but as isolated genetic alterations in MEN 2A-related PCC (Fig. 3). NF1 microsatellite alterations were shown in 73.3% informative PCC (Table 4 and Figs. 2 and 3) and were the only TSG genetic alterations with significant differences between spo-

TABLE 2. Clinicopathological features of malignant PCC Case

PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC a

M1 M2 M3 M4 M5 M6 M7 M8 M9 M10 M11 M12 M13 M14 M15 M16

Topographic molecular profilea

Urine catecholamines (␮g/24 h)

Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Heterogeneous Homogeneous Homogeneous Homogeneous Homogeneous

257 198 211 208 312 300 290 266 260 245 289 300 241 200 198 168

Metastasis Liver

Lymph node

⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹ ⫹

⫹ ⫺ ⫹ ⫺ ⫹ ⫹ ⫹ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ⫺ ⫹ ⫹

See Figs. 2 and 3 for details.

Downloaded from jcem.endojournals.org on March 13, 2006

Soft tissue infiltration

⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫹ ⫺ ⫺ ⫹ ⫺ ⫺ ⫹ ⫺ ⫹ ⫺

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158 1153

FIG. 2. Microsatellite pathways in PCC. Microsatellite patterns; PCCs with concordant microsatellite pattern in internal and peripheral compartments. This homogeneous microsatellite pattern reveals a significant decrease in the intensity of the smaller allele of lanes 2 from both internal and peripheral PCC compartments (arrows). The allele pattern is the opposite of control (adrenal medulla) that shows higher intensity in the larger allele. The other tumor suppressor gene loci show no abnormalities. PCCs with intratumor heterogeneity and disconcordant microsatellite pattern in internal and peripheral compartments. Heterogeneous microsatellite patterns reveal a decreased intensity of the larger allele in lane 1 and additional allele bands in lane 3 (arrows) in the peripheral compartment. 1, TP53(1); 2, TP53(2); 3, RB1; 4, WT1; and 5, NF1. NF1 locus microsatellite patterns. Lanes 1 (adrenal cortex) and 2 (adrenal medulla) correspond to two representative controls (from the same case shown in lanes 3 and 4), whereas lanes 3–12 show the microsatellite pattern in both peripheral (P) and internal (I) PCC compartments of five cases. Lanes 3 and 4 show concordant loss of the larger allele in both tumor compartments (compared with their matched controls in lanes 1 and 2). Lanes 5–12 show a discordant allele pattern in peripheral and internal compartments. The TSG microsatellite profile is shown. Malignant PCC (12 of 16, 75%) reveal two or more TSG loci with microsatellite abnormalities in the peripheral compartment, whereas all locally invasive and benign PCC show a lower incidence of microsatellite lesions. Mismatch protein expression is shown. Nuclear mlh1 and msh2 expression is demonstrated in PCC in the microsatellite-stable pathway, and at least one of these proteins was absent (in particular, mlh1) in PCC, showing microsatellite instability. MLH1 and MSH2 exon sequencing is shown. A normal sequence is demonstrated for these genes regardless of the microsatellite pattern. We are currently performing promoter methylation assays for MLH1.

radic and MEN 2A PCC (Table 5). NF1 abnormalities were also detected in tumors with topographic genetic heterogeneity in the case of sporadic PCC (eight of nine, 89%), whereas NF1 abnormalities were equally demonstrated in MEN 2A PCC regardless of the presence of topographic heterogeneity. No significant differences were observed for RB1 and WT1 loci. The comparison between sporadic and MEN 2A PCC revealed a significantly higher proportion of NF1 locus abnormalities (LOH-somatic SNP), so that the other TSG microsatellites differences were nonsignificant (Table 5). We found concordant TSG microsatellite pattern by compartments in 47 PCC (32.9%, 24 sporadic and 23 MEN 2A). Because the normal tissue LOH probability was 0.2 (10, 11,

27–31), the probability of getting the same n loci involved in both peripheral and internal samples (two) would be (0.22)n. Under these circumstances, the probability of randomly finding concordant results in both PCC compartments (one TSG locus in 21 patients, two TSG loci in 10, and three TSG loci in 16) would be [(0.22)1]21[(0.22) 2]10[(0.22)3]16. Likewise, because the normal tissue ROH probability, PROH ⫽ 1 ⫺ PLOH⫽1– 0.2 ⫽ 0.8 for a given marker, 0.85 for 5 markers, and (0.85)2 for concordant peripheral and internal compartments, it will result in [(0.85)2]32 ⫽ 9.75 ⫻ 10⫺32 for 32 PCC. Immunostaining for mlh1/msh2 revealed statistically significant reduction of at least one of the proteins in the peripheral compartment of PCC with two or more TSG microsatellite abnormalities regardless of the histological

Downloaded from jcem.endojournals.org on March 13, 2006

1154

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

TABLE 3. Pheochromocytomas with topographic heterogeneity in microsatellite pattern of TSGsa Caseb

PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC PCC

1-P 1-I 2-P 2-I 3-P 3-I 4-P 4-I 5-P 5-I 6-P 6-I 7-P 7-I 8-P 8-I 9-P 9-I 10-P 10-I 11-P 11-I 12-P 12-I 13-P 13-I 14-P 14-I 15-P 15-I 16-P 16-I 17-P 17-I 18-P 18-I 19-P 19-I 20-P 20-I 21-P 21-I 22-P 22-I 23-P 23-I 24-P 24-I 25-P 25-I 26-P 26-I 27-P 27-I 28-P 28-I 29-P 29-I 30-P 30-I 31-P 31-I 32-P 32-I 33-P 33-I

TP53–1c

TP53–2c

RB1c

WT1c

NF1c

ROH ROH ROH ROH ROH ROH LOH-S ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH LOH-S ROH ROH ROH ROH ROH NI NI ROH ROH ROH ROH NI NI ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH NI NI LOH-L ROH LOH-S ROH LOH-L ROH LOH-S ROH ROH ROH NI NI LOH-L ROH ROH ROH LOH-L ROH LOH-L ROH LOH-L ROH

LOH-S ROH ROH ROH ROH ROH ROH ROH ROH ROH NI NI ROH ROH ROH ROH LOH-L ROH ROH ROH ROH ROH NI NI ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH LOH-L ROH NI NI ROH ROH ROH ROH NI NI LOH-S ROH LOH-S ROH LOH-L LOH-L LOH-L ROH ROH ROH ROH ROH LOH-L LOH-L

NI NI LOH-L ROH ROH ROH ROH ROH LOH-L ROH ROH ROH ROH ROH ROH ROH ROH ROH NI NI ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH NI NI ROH ROH ROH ROH ROH/SNP-SL ROH LOH-L ROH NI NI LOH-L ROH NI NI NI NI NI NI ROH ROH ROH/SNP-SL ROH ROH ROH ROH ROH ROH ROH

ROH ROH ROH ROH LOH-S ROH ROH ROH ROH ROH LOH-L ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH ROH NI NI ROH ROH ROH ROH NI NI ROH ROH NI NI ROH ROH ROH ROH NI NI ROH ROH ROH ROH ROH ROH LOH-S ROH ROH ROH LOH-S ROH LOH-L ROH LOH-L ROH

ROH/SNP-S ROH/SNP-S ROH/SNP-S ROH/SNP-S NI NI NI NI ROH/SNP-S ROH/SNP-S NI NI LOH-S ROH LOH-L SNP-L ROH/SNP-L ROH/SNP-L NI NI LOH-L SNP-L LOH-L ROH ROH/SNP-L ROH LOH-S ROH ROH/SNP-S ROH LOH-S ROH LOH-L ROH LOH-L ROH LOH-L ROH LOH-S ROH LOH-S ROH ROH ROH ROH ROH ROH/SNP-SL ROH ROH ROH ROH/SNP-SL ROH LOH-L/ SNP-S ROH LOH-L/ SNP-S ROH ROH/SNP-SL ROH/SNP-L ROH ROH ROH/SNP-L ROH ROH/SNP-SL ROH/SNP-L ROH/SNP-L ROH

PCC typed

Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Benign, MEN-2A Locally invasive, sporadic Locally invasive, sporadic Locally invasive, sporadic Locally invasive, sporadic Locally invasive, sporadic Locally invasive, sporadic Locally invasive, sporadic Locally invasive, sporadic Locally invasive, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic Malignant, sporadic

Genes showing different microsatellite pattern in the peripheral and internal compartments are highlighted in italics. Two samples were systematically evaluated from the peripheral (P) and internal (I) PCC compartments. Gene intron results were categorized as ROH, LOH, and SNP involving either the larger allele (L) or the smaller allele (S). Samples with only one allele band were considered noninformative (NI). d PCC types, Associated to MEN-2A (benign) and sporadic (benign, locally invasive, and malignant). b c

Downloaded from jcem.endojournals.org on March 13, 2006

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158 1155

TABLE 4. Distribution of genetic abnormalities in tumor suppressor genes by topographic compartments in pheochromocytomas Genes, tumor groupa

ROHb number (%)c

LOH–SNPb number (%)c

Statistical significance

TP53, MS heterogeneity TP53, MS homogeneity TP53, total RB1, MS heterogeneity RB1, MS homogeneity RB1, total WT1, MS heterogeneity WT1, MS homogeneity WT1, total NF1, MS heterogeneity NF1, MS homogeneity NF1, total

17 (52.9) 75 (76.0) 92 (70.1) 19 (76.9) 63 (80.0) 82 (79.2) 24 (80.0) 63 (71.1) 87 (73.3) 8 (26.7) 39 (80.0) 47 (60.0)

15 (47.1) 24 (24.0) 39 (29.9) 6 (23.1) 16 (20.0) 22 (20.8) 6 (20.0) 25 (28.9) 31 (26.7) 21 (73.3) 10 (20.0) 31 (40.0)

P ⫽ 0.00933 Not significant Not significant P ⬍ 0.00001

a Pheochromocytomas were classified according to the microsatellite (MS) pattern of tumor suppressor genes in tumors revealing peripheral heterogeneity (more genetic alterations in the peripheral compartment) and tumors with identical microsatellite pattern in both compartments. b Tumor suppressor gene results were categorized in ROH, and LOH–SNP. c Table shows row percentage.

diagnosis (Fig. 2). No significant difference was observed in the mlh1/msh2 immunoexpression in PCC with less than two TSG genetic abnormalities (MS stable or MS instable low), but revealed a deficient MMR system at the peripheral compartment. Normal MLH1/MSH2 exons sequences were observed in all PCC analyzed regardless of immunoexpression and microsatellite status (Fig. 2). Discussion

PCCs revealed two types of TSG microsatellite patterns: 1) PCC with no evidences of TSG microsatellite abnormalities

FIG. 3. Malignant PCC frequently show topographic accumulation of TSG abnormalities at the peripheral tumor compartment (A) and simultaneous alterations of multiple TSG (B), especially involving TP53.

(44.1%), and 2) PCC with microsatellite alterations (55.9%), probably resulting from multistep tumorigenesis and topographic clonal selection at the peripheral compartment through somatic topographic down-regulation of MMR proteins. Both accumulation of microsatellite lesion accumulation and intratumor heterogeneity characterize malignant PCC, in contrast to locally invasive PCC, which frequently reveals single locus alterations, especially involving NF1. A PCC subgroup showed no demonstrable TSG alteration in tested introns (32 PCC, 43.8%). We previously optimized the DGGE protocol with appropriate controls, including positive, negative, and sensitivity (data not shown); the progressive dilution of a known positive case in a background of germline DNA gave us a sensitive threshold of 1% for a positive detection. We systematically microdissected at least 100 cells from each tumor compartment, a technique that would only miss the positive results from the DNA equivalent to less than one cell, which is probably clinically irrelevant. Normal tissue contamination could dilute the mutated DNA below the detection threshold, but this possibility was excluded by consistently negative results after careful microdissections. Although these TSG can show alterations in other introns, the low probability of finding no TSG alterations suggests that nonmalignant and sporadic PCC evolve through alternative molecular pathways (5, 7, 32, 33). The topographic intratumor heterogeneity suggests a differential selection of tumor cells by compartments, but can also be the expression of either selective clonal evolution or a simple passive byproduct of genetic instability (10, 13, 14, 34). The differential kinetic profiles of topographic tumor compartments revealed lower cell turnover and apoptosis down-regulation in deep/peripheral compartments, resulting in accumulation of genetic alterations and segregation of tumor cells with differential genetic backgrounds, as demonstrated in colon and bladder (13, 20, 35). In these organs (as in the present study), the process has been linked with MMR protein down-regulation, and it is unlikely to be related to hypoxia, which is more pronounced in central compartments. However, the coexistence of genetic alterations in malignant PCC supports a key role in tumorigenesis, the topographic heterogeneity resulting from the accumulation of genetic damage in TP53 and NF1 loci.

Downloaded from jcem.endojournals.org on March 13, 2006

1156

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

TABLE 5. Distribution of genetic alterations in tumor suppressor genes in sporadic and MEN 2A pheochromocytomas

a b

Genes resultsa

Sporadic number (%)b

MEN 2A number (%)b

Statistical significance

TP53, ROH TP53, LOH–SNP TP53, total RB1, ROH RB1, LOH–SNP RB1, total WT1, ROH WT1, LOH–SNP WT1, total NF1, ROH NF1, LOH–SNP NF1, total

63 (68.1) 25 (65.0) 88 (67.2) 56 (69.0) 13 (63.6) 69 (67.9) 57 (65.9) 21 (68.8) 78 (66.7) 35 (75.0) 17 (56.3) 52 (67.5)

29 (31.9) 14 (35.0) 43 (32.8) 26 (31.0) 9 (36.4) 35 (32.1) 30 (34.1) 10 (31.2) 39 (33.3) 12 (25.0) 14 (43.7) 26 (32.5)

Not significant Not significant Not significant P ⫽ 0.00005

Tumor suppressor gene results were categorized in ROH and LOH–SNP. Table shows row percentage.

TP53 abnormalities were more frequently observed in topographically heterogeneous and malignant PCC. Krijger et al. (36) reported p53 immunoexpression between 10 –50% in most malignant PCC, supporting TP53 heterogeneity regardless of the genetic background (sporadic or familial) (5, 6). The lower sensitivity of immunohistochemistry and the heterogeneous tissue distribution would explain the absence of significant immunoexpression differences. We found similar topographic heterogeneity in muscle-invasive transitional cell carcinoma of the urinary bladder, where TP53 abnormalities tended to concentrate in the deep tumor compartment, suggesting that it can represent the consequence of tumor cell selection (10, 13, 20). NF1 microsatellite alterations were significantly more frequent in PCC with topographic heterogeneity and were the only markers differentiating sporadic from MEN 2A tumors. Reduced or absent NF1 gene expression (both mRNA and protein) has been previously documented in one of four sporadic PCC, three of 10 tumors from MEN 2A patients, and two of four tumors from patients with MEN2B, most of them expressing predominantly the type 1 NF1 isoform (37). These findings together support Knudson’s two-mutation theory and the importance of the NF1 gene in PCC tumorigenesis in patients without neurofibromatosis (9, 38 – 40), especially those with MEN syndromes [seven of 13 (53.8%) of our informative MEN 2A patients showed NF1 locus abnormalities]. The NF1 gene product has an effect on RAS inhibition; this protein is expressed at the highest levels in immature and proliferating cells (41). The absence of NF1 inhibitory effects will favor increased cell proliferation, frequent presence of tetraploid cells, and, eventually, tumorigenesis (24, 42). The accumulation of TSG genetic lesions supports a monoclonal origin of tumors and expresses molecular tumor progression (10, 11), contributing to the relatively high incidence of NF1 locus abnormalities found in MEN 2A patients (seven of 13 informative cases, 53.8%). Those results strongly support a homogeneous cell selection and a clonal origin of this PCC subset through a pathway that targets the evaluated TSG. This convergent tumor cell selection ends in dominant clones sharing genetic abnormalities, explaining the utility of coexistent LOH as a clonality marker (11). Therefore, clonality would be both the origin and the byproduct of neoplastic transformation (10, 14, 43). Tumor initiation and/or progres-

sion in PCC might involve multiple genes apart from the RET oncogene (44). The MEN 2A kindred of this series carried RET germline mutation in codon 634 (45) and had coexistent thyroid lesions (18 medullary carcinoma and six C cell hyperplasia). Recently, RET mutations in codon 768 were found to segregate with medullary thyroid carcinomas or C cell hyperplasia only (46), and variable mechanisms are responsible for RET allelic imbalance (homozygosity, hemizygosity, and poly/monosomy), suggesting that multiple factors contribute to tumor development (47, 48). X-Chromosome inactivation assays in patients from this kindred revealed the same X-chromosome inactivated in C cell hyperplasia foci from each lobe and in different nodules from adrenal medullary hyperplasias in a given patient (24, 49). Those findings would be explained by early clonal expansions of both C cell and adrenal medullary precursors, which result in neoplasms when other genes are targeted and genetic alterations accumulate (18 of 24 MEN 2A PCC, 75.0%, had at least one TSG microsatellite abnormality). The accumulation of TSG microsatellite abnormalities in the peripheral compartment correlates with the lack of mlh1/ msh2 immunoexpression and normal gene sequences (50). Replication error-positive samples have been reported in 30.8% of heterozygous PCC, the instability being the result of impaired cellular MMR (51), which leads to mutation accumulation in every cellular division. These findings express tumor progression with decreased aneuploid cell prevalence and loss of physiological cell kinetic correlations. The data suggest that molecular mechanisms of genomic instability are not necessarily independent and may not be fully defined by either microsatellite or chromosomal instability pathways; a subgroup of tumors showed no evidence of alterations in either of these two pathways of genomic instability (13, 52). MMR proteins normally identify and correct mismatched DNA sequences that can occur during DNA replication (11, 53, 54). Tumor progression in peripheral compartments may be the result of MMR protein down-regulation, which would contribute to: 1) lower DNA indices and the decreased prevalence of aneuploid cell lines detected in microsatellite-unstable neoplasms (55, 56), and deep compartments of sporadic colorectal carcinomas (2, 57, 58); and 2) tumor cell heterogeneity, genetic instability, and biological progression. Because of intratumoral heterogeneity, at least two samples

Downloaded from jcem.endojournals.org on March 13, 2006

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

from each tumor should be screened (11), preferably from internal and peripheral compartments to cover the topographic tumor heterogeneity described in sporadic colorectal carcinomas (59) and muscle-invasive transitional cell carcinomas of the bladder (13). In conclusion, the accumulation of TSG microsatellite alterations supports a convergent genetic selection and multistep tumorigenesis in sporadic and MEN 2A PCC. This selection process also expresses intratumoral heterogeneity with accumulation of microsatellite abnormalities at the PCC peripheral compartment, multiple loci, and TP53 for malignant PCC and NF1 for MEN 2A and locally invasive PCC. A PCC subset (nonmalignant and sporadic) with no TSG microsatellite abnormalities evolves through an alternative pathway. Acknowledgments This work was performed at the Department of Pathology, Tufts University-New England Medical Center (Boston, MA) and Barts and The London National Health Service Trust (London, UK).

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158 1157

14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

Received July 22, 2005. Accepted December 27, 2005. Address all correspondence and requests for reprints to: Dr. Salvador J. Diaz-Cano, Department of Histopathology, King’s College Hospital, Denmark Hill, London SE5 9RS, United Kingdom. E-mail: salvador.diazcano@kcl.ac.uk. This paper was presented in part as abstracts at the Annual Meetings of United States and Canadian Academies of Pathology in Orlando, FL (March 1997), and San Antonio, TX (March 2005); the XXII International Congress of the International Academy of Pathology in Nice, France (October 1998); and the Pathological Society of Great Britain and Ireland, Maastricht 2001.

25. 26. 27. 28. 29.

References

30.

1. Neumann HP, Berger DP, Sigmund G, Blum U, Schmidt D, Parmer RJ, Volk B, Kirste G 1993 Pheochromocytomas, multiple endocrine neoplasia type 2, and von Hippel-Lindau disease. N Engl J Med [Erratum (1994) 331:1535] 329:1531–1538 2. Lack EE 1997 Tumors of the adrenal gland and extra-adrenal paraganglia. Washington DC: Armed Forces Institute of Pathology 3. Carney JA, Sizemore GW, Sheps SG 1976 Adrenal medullary disease in multiple endocrine neoplasia, type 1. Pheochromocytoma and its presursors. Am J Clin Pathol 66:279 –290 4. Koch CA, Pacak K, Chrousos GP 2002 The molecular pathogenesis of hereditary and sporadic adrenocortical and adrenomedullary tumors. J Clin Endocrinol Metab 87:5367–5384 5. Khosla S, Patel VM, Hay ID, Schaid DJ, Grant CS, van Heerden JA, Thibodeau SN 1991 Loss of heterozygosity suggests multiple genetic alterations in pheochromocytomas and medullary thyroid carcinomas. J Clin Invest 87: 1691–1699 6. Shin E, Fujita S, Takami K, Kurahashi H, Kurita Y, Kobayashi T, Mori T, Nishisho I, Takai S 1993 Deletion mapping of chromosome 1p and 22q in pheochromocytoma. Jpn J Cancer Res 84:402– 408 7. Vargas MP, Zhuang Z, Wang C, Vortmeyer A, Linehan WM, Merino MJ 1997 Loss of heterozygosity on the short arm of chromosomes 1 and 3 in sporadic pheochromocytoma and extra-adrenal paraganglioma. Hum Pathol 28:411– 415 8. Yokogoshi Y, Yoshimoto K, Saito S 1990 Loss of heterozygosity on chromosomes 1 and 11 in sporadic pheochromocytomas. Jpn J Cancer Res 81:632– 638 9. Koch CA, Huang SC, Zhuang Z, Stolle C, Azumi N, Chrousos GP, Vortmeyer AO, Pacak K 2002 Somatic VHL gene deletion and point mutation in MEN 2A-associated pheochromocytoma. Oncogene 21:479 – 482 10. Diaz-Cano SJ 1998 Clonality studies in the analysis of the adrenal medullary proliferations: application principles and limitations. Endocr Pathol 9:301–316 11. Diaz-Cano SJ, Blanes A, Wolfe HJ 2001 PCR techniques for clonality assays. Diagn Mol Pathol 10:24 –33 12. Sternlicht M, Mirell C, Safarians S, Barsky SH 1994 A novel strategy for the investigation of clonality in precancerous disease states and early stages of tumor progression. Biochem Biophys Res Commun 199:511–518 13. Diaz-Cano SJ, Blanes A, Rubio J, Matilla A, Wolfe HJ 2000 Molecular evolution and intratumor heterogeneity by topographic compartments in muscle-

31. 32. 33.

34. 35. 36.

37. 38. 39. 40.

41. 42.

invasive transitional cell carcinoma of the urinary bladder. Lab Invest 80:279 – 289 Diaz-Cano SJ, Blanes A 1999 Influence of intratumor heterogeneity in the interpretation of marker results in pheochromocytomas. J Pathol 189:627– 628 Smith HS 1990 Stochastic model for interpreting the data on loss of heterozygosity in breast cancer. J Natl Cancer Inst 82:793–794 Kinzler KW, Vogelstein B 1996 Lessons from hereditary colorectal cancer. Cell 87:159 –170 DeLellis RA, Wolfe HJ, Gagel RF, Feldman ZT, Miller HH, Gang DL, Reichlin S 1976 Adrenal medullary hyperplasia. A morphometric analysis in patients with familial medullary thyroid carcinoma. Am J Pathol 83:177–196 Diaz-Cano SJ, Brady SP 1997 DNA extraction from formalin-fixed paraffinembedded tissues: Protein digestion as a limiting step for retrieval of high quality DNA. Diagn Mol Pathol 6:342–346 Cawkwell L, Lewis FA, Quirke P 1994 Frequency of allele loss of DCC, p53, RBI, WT1, NF1, NM23 and APC/MCC in colorectal cancer assayed by fluorescent multiplex polymerase chain reaction. Br J Cancer 70:813– 818 Blanes A, Rubio J, Martinez A, Wolfe HJ, Diaz-Cano SJ 2002 Kinetic profiles by topographic compartments in muscle-invasive transitional cell carcinomas of the bladder: role of TP53 and NF1 genes. Am J Clin Pathol 118:93–100 Mutter GL, Boynton KA 1995 X chromosome inactivation in the normal female genital tract: Implications for identification of neoplasia. Cancer Res 55:5080 – 5084 Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ 2000 Clonality as expression of distinctive cell kinetic patterns in nodular hyperplasias and adenomas of the adrenal cortex. Am J Pathol 156:311–319 Diaz-Cano SJ 2000 Designing a molecular analysis of clonality in tumours. J Pathol 191:343–344 Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Wolfe HJ 2001 Germline RET 634 mutation positive MEN 2A-related C-cell hyperplasias have genetic features consistent with intraepithelial neoplasia. J Clin Endocrinol Metab 86:3948 –3957 Simpson JF, Dutt PL, Page DL 1992 Expression of mitosis per thousand cells and cell density in breast carcinomas. A proposal. Hum Pathol 23:608 – 611 Diaz-Cano SJ, Leon MM, de Miguel M, Galera-Davidson H, Wolfe HJ 1996 Mitotic index quantification: Different approaches and their value in adrenal cortical proliferative lesions. Lab Invest 74:170A (Abstract) Chen LC, Kurisu W, Ljung BM, Goldman ES, Moore DI, Smith HS 1992 Heterogeneity for allelic loss in human breast cancer. J Natl Cancer Inst 84: 506 –510 Deng G, Lu Y, Zlotnikov G, Thor AD, Smith HS 1996 Loss of heterozygosity in normal tissue adjacent to breast carcinomas. Science 274:2057–2059 Wolman SR, Heppner GH 1992 Genetic heterogeneity in breast cancer. J Natl Cancer Inst 84:469 – 470 Sager R 1989 Tumor suppressor genes: The puzzle and the promise. Science 246:1406 –1412 Diaz-Cano SJ, Wolfe HJ 1997 Clonality in Kaposi’s sarcoma. N Engl J Med 337:571–572 Woodward ER, Eng C, McMahon R, Voutilainen R, Affara NA, Ponder BA, Maher ER 1997 Genetic predisposition to phaeochromocytoma: analysis of candidate genes GDNF, RET and VHL. Hum Mol Genet 6:1051–1056 Moley JF, Brother MB, Fong CT, White PS, Baylin SB, Nelkin B, Wells SA, Brodeur GM 1992 Consistent association of 1p loss of heterozygosity with pheochromocytomas from patients with multiple endocrine neoplasia type 2 syndromes. Cancer Res 52:770 –774 Pohlink C, Tannapfe A, Eichfeld U, Schmidt F, Fuhrer D, Paschke R, Koch CA 2004 Does tumor heterogeneity limit the use of the Weiss criteria in the evaluation of adrenocortical tumors? J Endocrinol Invest 27:565–569 Jimenez JJ, Blanes A, Diaz-Cano SJ 2003 Microsatellite instability in colon cancer. N Engl J Med 349:1774 –1776; author reply 1774 –1776 Krijger RR, van der Harst E, van der Ham F, Stijnen T, Dinjen WNM, Koper JW, Bruining HA, Lamberts SWJ, Bosman FT 1999 Prognostic value of p53, bcl-2, and c-erbB-2 protein expression in phaeochromocytomas. J Pathol 188: 51–55 Gutmann DH, Geist RT, Rose K, Wallin G, Moley JF 1995 Loss of neurofibromatosis type I (NF1) gene expression in pheochromocytomas from patients without NF1. Genes Chromosomes Cancer 13:104 –109 Knudson Jr AG 1971 Mutation and cancer: statistical study of retinoblastoma. Proc Natl Acad Sci USA 68:820 – 823 Knudson Jr AG, Strong LC 1972 Mutation and cancer: neuroblastoma and pheochromocytoma. Am J Hum Genet 24:514 –532 Huang SC, Koch CA, Vortmeyer AO, Pack SD, Lichtenauer UD, Mannan P, Lubensky IA, Chrousos GP, Gagel RF, Pacak K, Zhuang Z 2000 Duplication of the mutant RET allele in trisomy 10 or loss of the wild-type allele in multiple endocrine neoplasia type 2-associated pheochromocytomas. Cancer Res 60: 6223– 6226 Furth ME, Aldrich TH, Cordon-Cardo C 1987 Expression of ras proto-oncogene proteins in normal human tissues. Oncogene 1:47–58 Gonzalez-Campora R, Diaz Cano S, Lerma-Puertas E, Rios Martin JJ, Salguero Villadiego M, Villar Rodriguez JL, Bibbo M, Davidson HG 1993 Paragangliomas. Static cytometric studies of nuclear DNA patterns. Cancer 71:820 – 824

Downloaded from jcem.endojournals.org on March 13, 2006

1158

J Clin Endocrinol Metab, March 2006, 91(3):1150 –1158

43. Alexander P 1985 Do cancers arise from a single transformed cell or is monoclonality of tumours a late event in carcinogenesis? Br J Cancer 51:453– 457 44. Koch CA, Huang SC, Moley JF, Azumi N, Chrousos GP, Gagel RF, Zhuang Z, Pacak K, Vortmeyer AO 2001 Allelic imbalance of the mutant and wild-type RET allele in MEN 2A-associated medullary thyroid carcinoma. Oncogene 20:7809 –7811 45. Gagel RF, Cote GJ, Martins Bugalho MJ, Boyd III AE, Cummings T, Goepfert H, Evans DB, Cangir A, Khorana S, Schultz PN 1995 Clinical use of molecular information in the management of multiple endocrine neoplasia type 2A. J Intern Med 238:333–341 46. Boccia LM, Green JS, Joyce C, Eng C, Taylor SA, Mulligan LM 1997 Mutation of RET codon 768 is associated with the FMTC phenotype. Clin Genet 51:81– 85 47. Marx SJ 2005 Molecular genetics of multiple endocrine neoplasia types 1 and 2. Nat Rev Cancer 5:367–375 48. Spiegel AM 2004 Focus on hereditary endocrine neoplasia. Cancer Cell 6:327– 332 49. Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ 2000 Clonal patterns in phaechromocytomas and MEN-2A adrenal medullary hyperplasias: histologic and kinetic correlates. J Pathol 192:221–228 50. Blanes A, Diaz-Cano SJ, Sporadic colorectal carcinomas: are microsatellite tumor profile and mismatch repair defects equivalent? World J Surg, in press 51. Pecina-Slaus N, Nikuseva-Martic T, Gall-Troselj K, Radic K, Hrascan R 2005 Replication error-positive samples found in pheochromocytomas. In Vivo 19:359 –365 52. Goel A, Arnold CN, Niedzwiecki D, Chang DK, Ricciardiello L, Carethers JM, Dowell JM, Wasserman L, Compton C, Mayer RJ, Bertagnolli MM,

Blanes et al. • Topographic Molecular Profile of Pheochromocytomas

53.

54. 55.

56.

57.

58.

59.

Boland CR 2003 Characterization of sporadic colon cancer by patterns of genomic instability. Cancer Res 63:1608 –1614 Duval A, Hamelin R 2002 Mutations at coding repeat sequences in mismatch repair-deficient human cancers: toward a new concept of target genes for instability. Cancer Res 62:2447–2454 Peltomaki P 2001 DNA mismatch repair and cancer. Mutat Res 488:77– 85 Thibodeau SN, French AJ, Cunningham JM, Tester D, Burgart LJ, Roche PC, McDonnell SK, Schaid DJ, Vockley CW, Michels VV, Farr Jr GH, O’Connell MJ 1998 Microsatellite instability in colorectal cancer: different mutator phenotypes and the principal involvement of hMLH1. Cancer Res 58:1713–1718 Liang JT, Chang KJ, Chen JC, Lee CC, Cheng YM, Hsu HC, Chien CT, Wang SM 1999 Clinicopathologic and carcinogenetic appraisal of DNA replication error in sporadic T3N0M0 stage colorectal cancer after curative resection. Hepatogastroenterology 46:883– 890 Jimenez-Martin JJ, Miranda MT, Blanes A, Diaz-Cano SJ 2003 Expression of mlh1/msh2 in colo-rectal adenocarcinomas: SAGE/microarray profile and tumor phenotype. Lab Invest 83:296A (Abstract) Blanes A, Jimenez-Martin JJ, Miranda MT, Diaz-Cano SJ 2003 Abscence of the physiologic cellular kinetic balance and down-regulation of mlh1/msh2 characterize deep topographic compartments of colo-rectal adenocarcinomas. Lab Invest 83:114A–115A (Abstract) Chapusot C, Martin L, Bouvier AM, Bonithon-Kopp C, Ecarnot-Laubriet A, Rageot D, Ponnelle T, Laurent Puig P, Faivre J, Piard F 2002 Microsatellite instability and intratumoural heterogeneity in 100 right-sided sporadic colon carcinomas. Br J Cancer 87:400 – 404

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

Downloaded from jcem.endojournals.org on March 13, 2006

PO Box 2345, Beijing 100023, China www.wjgnet.com wjg@wjgnet.com

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

REVIEW

Complementary analysis of microsatellite tumor profile and mismatch repair defects in colorectal carcinomas Alfredo Blanes, Salvador J Diaz-Cano Alfredo Blanes, Department of Pathology, University of Malaga School of Medicine, 29010-Malaga, Spain Salvador J Diaz-Cano, Department of Histopathology, King’s College Hospital and King’s College London School of Medicine, London SE5 9RS, United Kingdom Correspondence to: Salvador J Diaz-Cano, MD, PhD, FRCPath King’s College Hospital, Department of Histopathology, Denmark Hill, London, SE5 9RS, United Kingdom. salvador.diaz-cano@kcl.ac.uk Telephone: +44-20-7346-3041 Fax: +44-20-7346-3670 Received: 2005-07-04 Accepted: 2005-08-26

Abstract Microsatellite instability (MSI) is a prognostic factor and a marker of deficient mismatch repair (MMR) in colorectal adenocarcinomas (CRC). However, a proper application of this marker requires understanding the following: (1) The MSI concept: The PCR approach must amplify the correct locus and accurately identify the microsatellite pattern in the patient’s normal tissue. MSI is demonstrated when the length of DNA sequences in a tumor differs from that of nontumor tissue. Any anomalous expansion or reduction of tandem repeats results in extra-bands normally located in the expected size range (100 bp, above or below the expected product), differ from the germline pattern by some multiple of the repeating unit, and must show appropriate stutter. (2) MSI mechanisms: MMR gene inactivation (by either mutation or protein down-regulation as frequently present in deep CRC compartments) leads to mutation accumulation in a cell with every cellular division, resulting in malignant transformation. These mechanisms can express tumor progression and result in a decreased prevalence of aneuploid cells and loss of the physiologic cell kinetic correlations in the deep CRC compartments. MSI molecular mechanisms are not necessarily independent from chromosomal instability and may coexist in a given CRC. (3) Because of intratumoural heterogeneity, at least two samples from each CRC should be screened, preferably from the superficial (tumor cells above the muscularis propria) and deep (tumor cells infiltrating the muscularis propria) CRC compartments to cover the topographic tumor heterogeneity. (4) Pathologists play a critical role in identifying microsatellite-unstable CRC, such as occur in young patients with synchronous or metachronous tumors or with tumors showing classic histologic features. In these cases, MSI testing and/or MMR immunohistochemistry are advisable, along with gene sequencing and genetic

www.wjgnet.com

Key words: Colon carcinoma; Microsatellites; Mismatch Repair; Hereditary non-polyposis colon cancer Blanes A, Diaz-Cano SJ. Complementary analysis of microsatellite tumor profile and mismatch repair defects in colorectal carcinomas. World J Gastroenterol 2006; 12(37): 5932-5940

http://www.wjgnet.com/1007-9327/12/5932.asp

INTRODUCTION Colorectal carcinoma (CRC) is generally classified into three categories, based on increasing hereditary inuence and cancer risk [1,2]: sporadic CRC (approximately 60% of cases and comprises patients with no notable family history and, by definition, with no identifiable inherited gene mutation that accelerates cancer development), familial CRC (approximately 30% of cases and refers to patients who have at least one blood relative with CRC or an adenoma, but with no specific germline mutation or clear pattern of inheritance), and hereditary CRC syndromes (approximately 10% of cases, which result from inheritance of a single gene mutation in highly penetrant cancer susceptibility genes). Although the last group has the lowest frequency, it has elucidated molecular mechanisms of carcinogenesis applicable to sporadic CRC[3]. The microsatellite profile of sporadic CRC is a prognostically useful marker [4-7] . Microsatellites are repeating DNA sequences of unknown function that are found throughout the genome[8]. Microsatellite instability (MSI) is demonstrated when the length of DNA sequences in tumor and nontumor tissues is different and MSI has been identied in a wide variety of human tumors, due to defects in one of the DNA mismatch repair (MMR) genes, especially MLH1 or MSH2[1,9]. However, MSI presence alone does not establish a diagnosis of hereditary nonpolyposis colon cancer (HNPCC) because MSI has also been identified in 10%-30% of sporadic CRC. Certain histological features also correlate with the presence of

Blanes A et al. Microsatellites in Sporadic colon carcinomas

5933

MSI CRC

Feature

MSS CRC

94%

Right Side

34%

59%

Size > 6 cm

29%

53%

Poorly differentiated

35%

Mucinous

47%

Lymph Inďƒžltrates

7% 7% 10%

Figure 1 Microsatellite unstable CRCs are normally located proximal to the splenic flexure and reveal non-polypoid pattern. MSI can be demonstrated by fluorescence insitu hybridization (FISH) or molecular techniques. Histopathologically, these tumors show solid growth, mucinous differentiation, prominent lymphocytic infiltrate, and no dirty necrosis.

MSI in sporadic CRC (Figure 1)[5,10-13], which can be key elements in the design of more effective therapeutic protocols[12,13]. Both basic and clinical implications of MSI and MMR defects need to be considered in an appropriate context, which requires clarifying the definition of MSI, the biological consequences of tumor MSI, interference of intratumor heterogeneity on MSI detection, differences in clinical testing for MSI and for MMR defects, and MSI prognostic and therapeutic implications.

MSI DEFINITION AND CLINICAL TESTING FOR MSI Any useful application of prognostic factors requires a reliable definition of the factor. Microsatellites belong to the family of highly polymorphic and repetitive noncoding DNA sequences that, although widely distributed in the human genome, are not uniformly spaced (underrepresented in subtelomeric chromosome regions). Microsatellites are useful molecular markers due to their www.wjgnet.com

5934

ISSN 1007-9327

CN 14-1219/R

World J Gastroenterol

ubiquity, PCR typability (except for (dA)n multimers, whose size polymorphisms are difcult to type), Mendelian co-dominant inheritance, and extreme polymorphism[8], but their origin and function are not clear[14]. They have been demonstrated to be very useful in cell lineage delineation, positional cloning, and several applications in forensic medicine[15,16]. Microsatellite instability (MSI) is demonstrated when the lengths of DNA sequences in a tumor differ from those of nontumor tissue. MSI has been identied in a wide variety of human tumors. Currently, tumor MSI analyses require molecular tests and the application of strict criteria. MSI can be dened as a change in any DNA sequence length due to either insertion or deletion of repeating units in a microsatellite within a tumor when compared to normal tissue[17,18]. The tests must be run with appropriate controls (known positive and negative controls along with the patient’s normal tissue)[8,19], which are extremely important due to the nonexceptional presence of extra-bands. The PCR approach must amplify the correct locus and accurately identify the microsatellite pattern in the patient’s normal tissue. Any anomalous expansion or reduction of tandem repeats due to MSI results in extra-bands. True extra-bands expressing tumor MSI are normally located in the expected size range (usually about 100 bp), are above or below the expected PCR product, differ from the germline pattern by some multiple of the repeating unit (e.g. delta 6 bases for dinucleotides), must show appropriate stutter (e.g. -2, -4 for dinucleotides), and are not present in the normal control. These tests should be carefully analyzed considering the following: (1) Sample homogeneity/heterogeneity can vary. Very small samples (even single cells) have been used in genetic analyses to avoid normal cell contamination. However, the lower the number of cells the higher the probability of technically-related abnormal results[20], which can be partially resolved with appropriate methods. The high incidence of PCR artifacts using microdissected samples is related to the small concentration of target DNA, xation induced changes of DNA, and conditions in the amplication of repetitive sequences (especially for those CG-rich sequences) favouring misannealing and hairpin formation. Appropriate modications to avoid the above conditions will signicantly improve the reproducibility of LOH and MSI tests in microdissected samples[21]. (2) Appropriate controls are necessary for every step of the molecular tests to avoid false results. Sufcient levels of amplication with all markers should be obtained to detect low amounts of shifted microsatellites. (3) PCR bias against one allele (especially the larger one in a pair) can result in preferential amplication of the other allele (usually the smaller in a pair), which is the so-called artifactual allele dropout[22,23]. An appropriate extraction method, providing DNA of quality[24], and PCR designs including both long denaturation and extension in the rst three cycles and 7-deazadGTP in the amplication mixture to improve the amplication of CG-rich DNA regions, will be reasonably helpful in avoiding that bias[8,19,21,23,25]. (4) The number of polymorphic DNA regions agreed to at the NCI consensus conference includes a primary panel of at least 2 mononucleotide

www.wjgnet.com

October 7, 2006

Volume 12

Number 37

Table 1 False negative in antigen positive neoplasms,comparative features of microsatellite unstable sporadic adenocarcinoma and hnpcc colon carcinomas MSI-H Sporadic Adenocarcinomas

HNPCC Adenocarcinomas

Patient age

Older

Younger

Number of tumors

Single

Single/Multiple (synchronic/ metachronic)

Colonic distribution Right colon

Right colon

Histological clues

Poorly differentiated, Poorly differentiated, medullary type medullary type Crohn-like inammation Crohn-like inammation

Mechanism of MMR deciency

MLH1 promoter hypermethylation

Inactivating germline mutation of MMR proteins

Tumor prognosis

Better than MSI-L/ MSS sporadic adenoca

and 3 dinucleotide microsatellites, along with 19 alternate loci (both mono- and dinucleotides)[26]. The choice of microsatellite markers is important in MSI testing, with the examination of mononucleotide repeats being sufficient for detection of MMR decient tumors, whereas instability only in dinucleotides is characteristic of MSI-L/MMRpositive tumors[27]. Depending on the number of abnormal loci from the total analyzed, the cases are classied into MSI-high (! 30%-40% of abnormal loci), MSI-low (< 30%-40% of abnormal loci), and MSS (no abnormal loci). Which patients should be tested? The neoplasm histological features closely correlate with MSI and should be the key elements used to select sporadic CRC for MSI investigation[5,10-13]. The sets of criteria for the clinical diagnosis of HNPCC appear under Clinical Testing for MMR defects. The implications of these analyses in sporadic and HNPCC carcinomas are compared in Table 1.

BIOLOGICAL CONSEQUENCES OF TUMOR MSI Microsatellite-unstable CRC are biologically different and have a better sur vival rate than sporadic CRC when matched for cancer stage[28-30]. The development of proximal and distal CRC involves partly different mechanisms associated with the MSI and the chromosomal instability (CIN) pathways[31]. These two pathways are not always independent and some CRCs show a signicant degree of overlap between these two mechanisms[32]. In one study, 35% of CRC were microsatellite-unstable (21% MSI-low and 14% were MSIhigh) and 51% of CRC had at least one LOH event, with the most frequent chromosomal losses observed on 18q (72.5%)[32]. A signicant degree of overlap between MSI and CIN pathways has been reported in that series: 6.5% of CRC with LOH were also MSI-high, and 23.3% of MSI-high CRC also had one or more LOH events. These data suggest that molecular mechanisms of genomic instability are not necessarily independent and may not

Blanes A et al. Microsatellites in Sporadic colon carcinomas

be fully defined by either the MSI or CIN pathways. In addition, a subgroup of CRCs showed no evidence of alterations in either of these two pathways of genomic instability (37.8% of microsatellite-stable CRCs had no LOH events identified)[32], a situation similar to that reported in muscle-invasive transitional cell carcinomas of the bladder[25]. MMR proteins normally identify and correct mismatched DNA sequences that can occur during DNA replication. An inactivating mutation in any of these genes leads to mutation accumulation in a cell with every cellular division, resulting in malignant transformation [6,8,33-36]. Tumor progression in the deep compartments may be the result of MMR protein down-regulation, which would contribute to the following: (1) There is a decreased prevalence of aneuploid cell lines and K-RAS and B-RAF mutations detected in microsatellite-unstable CRC and in the deep compartments of sporadic CRC [7,29,37,38] . Microsatellite-unstable CRCs tend to be diploid[37,39], and to have lower DNA indices[39]. Supporting these ndings, the MMR protein down-regulation observed in the deep compartments of sporadic CRC has shown correlation with increased frequency of diploid DNA content[40,41]. (2) Differential cell kinetics (proliferation and apoptosis) has been identied in supercial and deep compartments (above muscularis propria vs. muscularis propria) of sporadic CRC, which has demonstrated a close correlation with MMR protein expression (Figure 2)[41,42]. Physiologic correlations between MMR protein expression and kinetic variables (mitotic figures, Ki-67 expression, ISEL index) were preserved in the superficial compartment only. In addition, G2 + M phase fraction correlated with hMLH1 expression only in supercial compartments and hMSH2 expression only in deep compartments. Both the high cellular turnover and the maintained cell kinetic balance suggest that superficial compartments of sporadic CRC are expansile. In the deep compartments, the expression of MMR proteins is inefcient (not correlated with G2 + M phase fraction) and is dissociated (only one gene product correlates with G2 + M), which would eventually result in mutation accumulation and progression[41].

INTRATUMOR HETEROGENEITY AND MICROSATELLITE ANALYSIS Tumor cell heterogeneity is linked to genetic instability and biologic progression. This problem must be studied by including several tumor samples of sufcient size from each tumor. The sample size is an important parameter. Microdissection techniques allow selectively picking up very small samples, which can show false cellular homogeneity, based on the loss of heterozygosity or allelic imbalance. If the tumor cell populations selected for molecular analysis are taken before they become a biologically prominent component (with kinetic or invasive advantages), the results might be confusing and clinically non-relevant. This would be a case of tumor microheterogeneity, which tends to give disparate results with meanings essentially unknown. Except for intraepithelial proliferation, all

5935

microdissected cell samples provide target cell-rich samples with varying degrees of host cell contamination (including stromal, inflammatory, and endothelial cells). Therefore, multiple samples from the same case should always be studied and assays performed in duplicate before accepting the results as relevant. The intratumor heterogeneity can result in discordant results for a given marker depending on the sample origin. The comparison of MMR protein expression and PCRbased MSI studies has revealed discordant results in 8% of right-sided sporadic CRC and complete concordance after performing further analyses on other tumor areas[43]. Because of this intratumor heterogeneity, at least two samples from each CRC should be screened, although no systematic approach has been used to address this topic in sporadic CRC. Microsatellite analysis in muscle-invasive transitional cell carcinomas of the bladder have revealed topographic heterogeneity in 32% of cases, showing that the deep compartment had more microsatellite abnor malities (20%) [25] . We have found significant differences between superficial (tumor cells above the muscularis propria) and the deep (tumor cells inltrating the muscularis propria) compartments of sporadic CRC, the deep compartments showing MMR protein downregulation and increased MSI[41,44]. At least one-third of unstable tumors in deep compartments can be expected to be stable in superficial compartments. These differences can eventually result in the classication of a given tumor as MMS or MSI depending on the sample origin (supercial or deep).

CLINICAL TESTING FOR MMR DEFECTS MSI results from the dysfunction of MMR proteins, which can be detected at genetic or protein levels. It is recommended that a CRC should be tested for MSI prior to gene testing, since this test is inexpensive and will help predict whether or not an individual has a germline MMR gene mutation[45,46]. Since up to 5% of HNPCC tumors do not have MSI, negative MSI tests cannot completely r ule out HNPCC. Conversely, a positive MSI test is not diagnostic of HNPCC because 15%-30% of unselected CRC have MSI (due to MLH1 promoter methylation), whereas only 1%-6% of all CRC are associated with detectable HNPCC mutations. If the tumor is MSI-positive, further analyses for MMR defects are recommended. Genetic testing for MMR defects MMR defects are due to either inactivating point mutations spread throughout the genes, therefore needing full-length sequencing, or promoter hypermethylation (especially MLH1 in sporadic CRC). HNPCC is an autosomal dominant disorder caused by ger mline MMR gene mutations, in particular in MLH1, MSH2, MSH6, and PMS2. No strong genotype-phenotype correlations have been observed to date, but MSH2 mutations do appear to be associated with more extracolonic manifestations than MLH1 mutations. MSH6 mutations are more common in endometrial tumors and PMS2 mutations are especially

www.wjgnet.com

5936

ISSN 1007-9327

CN 14-1219/R

World J Gastroenterol

GTP

Ras

GDP

GOP P

Ras

GTP

October 7, 2006

Volume 12

Number 37

hMLH1/hMSH2

PIP2

PI3P

PI3K PIP3 ERK pathway

AKT AKT pathway

Rac-GEF

IKK Bad

Rac

NF!B

p27

AFX

Raf1

MEK

ERK

Caspase 9

Bcl-xL Expression of antiapoptotic genes

ELK-1

Fas ligand expression

Caspase Cascade Pathway

M G2 Cell cycle DNA Fragmentation

G1 R

TNF

ISEL Apoptosis Inflammatory

Response

Ki-67 Proliferation

Figure 2 Molecular pathways contributing to the phenotype of microsatellite stable (MSS) CRC. MMR protein expression results in enhanced RAS signaling through ERK pathway (increased proliferation) and down-regulation of PI3P phosphatase, caspase 9 (apoptosis blockade), as well as TNF (decreased inflammation).

associated with Turcotâ€™s syndrome[47]. The original HNPCC diagnostic criteria were established by the International Collaborative Group on Hereditar y Nonpolyposis Colorectal Cancer (ICG-HNPCC) and are known as the Amsterdam criteria [30], but only 50%-70% of HNPCC

www.wjgnet.com

families meeting these criteria have been found to have germline MSH2 or MLH1 mutations.[48] The Amsterdam criteria were revised by the ICG-HNPCC in 1999 to include extracolonic cancers. The least stringent criteria are the Bethesda guidelines (more sensitive but less specific

Blanes A et al. Microsatellites in Sporadic colon carcinomas

than either the Amsterdam I or Amsterdam II criteria in identifying HNPCC families with pathogenic mutations), which aim to determine which patients should have MSI testing[48]. These criteria propose MSI testing for: Individuals with cancer in families that meet the Amsterdam criteria. Individuals with two hereditary nonpolyposis colon cancer syndrome (HNPCC)-related cancers, including synchronous and metachronous colorectal cancers or associated extracolonic cancers (endometrial, ovarian, gastric, hepatobiliary, or small bowel cancer or transitional cell carcinoma of the renal pelvis or ureter). Individuals with colorectal cancer and a first-degree relative with colorectal cancer and/or HNPCC-related extracolonic cancer and/or a colorectal adenoma: one of the cancers diagnosed by age 45, and the adenoma diagnosed by age 40. Individuals with colorectal cancer or endometrial cancer diagnosed by age 45. Individuals with right-sided colorectal cancer with an undifferentiated pattern (solid/cribriform) on histopathology diagnosed by age 45. Individuals with signet-ring-cell-type colorectal cancer diagnosed by age 45. Individuals with adenomas diagnosed by age 40. The American Gastroenterological Association recommends genetic testing for HNPCC for individuals from families meeting Amsterdam criteria, as well as for individuals with two HNPCC-related cancers (for instance, colorectal and endometrial cancer) and individuals with colorectal cancer who have a rst degree relative with an HNPCC-related cancer (or colorectal adenoma) where at least one was diagnosed before age 50[45,46,49]. Ideally, testing should rst be offered to a family member with colorectal or endometrial cancer.[26,45,48,49]. In some individuals, genetic analysis may be offered after prescreening for MSI in an HNPCC-related tumor specimen. Such prescreening should be offered where an HNPCC-related cancer is present in two individuals related by rst-degree regardless of age of onset, or individuals with early-onset CRC regardless of family history. Genetic testing is indicated if MSI is present. The majority (90%) of mutation-positive HNPCC cases are caused by mutations in MLH1 or MSH2[1,9]. For this reason, the mutation analysis is generally performed for these two genes, MS H6 being included in the analysis more recently. Although several methods can be used to detect these mutations, direct exon-by-exon gene sequencing is considered the gold standard. The sequencing should analyze each of the protein-coding regions of the MLH1 and MSH2 genes in their entirety, with all positive results being repeated for conrmation. Once a specic mutation that has been found in a relative by previous genetic testing, a test examining only the specic portion of the gene containing the known familial mutation can be offered to all family members. There are some benefits and limitations of genetic testing for HNPCC. Relying solely on family history can underestimate the risk of developing cancer in mutation carriers and over-estimate risk in those who do not

5937

inherit the mutation. When an individual has a personal or family history that suggests the possibility of HNPCC, an important step is to determine whether the person is interested in genetic testing. Genetic testing for HNPCC can have important benefits for members of high-risk families who choose to be tested[50]. Those who are found to carry deleterious mutations can take steps to reduce their cancer risk, especially through earlier and more intensive surveillance or consideration of prophylactic surgery. Individuals with HNPCC-related CRC can undergo surgical management designed to address the increased risk of a second cancer. In families in whom a deleterious mutation has been found, those who are mutation-negative can be spared the need for more intensive surveillance and intervention.[50] However, these individuals remain at risk for sporadic CRC and should be encouraged to adhere to age-appropriate general population screening guidelines. Before consenting to genetic analysis, patients should also consider the limitations of testing. Currently, genetic testing cannot detect unusual mutations responsible for HNPCC, such as those occurring in MMR genes other than MLH1 and MSH2. Therefore, a negative result in an individual who does not have a family member with a documented mutation must be interpreted cautiously. The test may also detect a variant of uncertain significance whose effect on cancer risk has not yet been established. In such situations, testing other family members for the specic variant to determine if it is associated with cancer may provide clarication of the signicance[50]. Immunohistochemical testing for MMR defects At the protein level, hMLH1/hMSH2 immunohistochemistry has a role in detecting MMR defects [51-53] , with data suggesting that the effectiveness of immunohistochemical screening of the MMR proteins would be similar to that of the more complex strateg y of microsatellite genotyping [54]. This technique can guide which gene to sequence and can help differentiating sporadic from hereditary mutations: hMSH2 loss is likely to be HNPCC, whereas hMLH1 loss could be HNPCC or sporadic CRC (MLH1 promoter methylation). MMR proteins heterodimerize to function; the hMSH2 loss almost always accompanies hMSH6 loss and when hMLH1 is lost, generally so is hPMS2[55-57]. In addition, immunohistochemistry can miss functional loss; i.e. presence of the protein with antigen positivity in the absence of function. Several antibodies have been used for these analyses, but the most widely used are hMSH2 (clone FE11, Oncogene Research), hMLH1 (clones G168 728 and G168-15, BD Pharmingen), hMSH6 (clone 44, BD Transduction Laboratories), and hPMS2 (clone A16-4, BD Pharmingen, and polyclonal C terminus, Santa Cruz Biotechnology). MMR immunohistochemical studies are based on a complete absence of at least one MMR protein[5,12,37,51-53,58-61]. But these studies do not consider the immunostaining topographic heterogeneity[41]. Since the MMR proteins function as heterodimers, it could be advocated to validate the immunohistochemical results of hMSH2/hMSH6 and

www.wjgnet.com

5938

ISSN 1007-9327

CN 14-1219/R

World J Gastroenterol

hMLH1/hPMS2. More studies are required to clarify the inuence of this predictable tumor heterogeneity to select the appropriate sample for immunohistochemical and/or MSI analyses.

PROGNOSTIC AND THERAPEUTIC IMPLICATIONS OF MSI The CRC microsatellite prole provides useful prognostic information[6,26,39], showing the patients with microsatelliteunstable neoplasms have a better overall survival rate and a modified response to conventional chemotherapy[62-67]. MSI also helps in predicting the treatment response of CRC[63,64,68], and could modify the chemotherapy protocols offered to the patients in the future[64], but these results should be applied with caution before this predictive tool is veried[64]. Molecular markers as predictive factors in treatment decisions have been developed in the last few years. The initial studies in sporadic CRC showed that the retention of heterozygosity at one or more 17p or 18q alleles in microsatellite-stable CRCs and mutation of the gene for the type Ⅱ receptor for TGF-"1 in CRCs with high levels of microsatellite instability correlated with a favorable outcome after adjuvant chemotherapy with fluorouracilbased regimens, especially for stage Ⅲ CRC[63,68]. However, most recent studies have revealed that uorouracil-based adjuvant chemotherapy beneted patients with stage Ⅱ or stage Ⅲ CRC with MSS tumors or tumors exhibiting lowfrequency MSI but not those with CRCs exhibiting highfrequency MSI[64]. The reasons for these responses must be related to the distinctive cell kinetics associated with MMR down-regulation (significantly increased apoptosis and decreased proliferation), which can certainly contribute to tumor cell resistance to conventional chemotherapy[40,41]. The topographic heterogeneity of sporadic CRC is a key element to explain the discrepant results reported[41]. This point has not been systematically addressed yet, but a homogeneous selection of the samples from the same topography must be considered in the molecular test design[25].

www.wjgnet.com

Volume 12

Number 37

entirety, although this technique will miss MLH1 gene inactivation by promoter methylation. Finally, the selection of samples for molecular tests must be carefully designed considering predictable heterogeneity, such as topographic heterogeneity, to avoid misinterpretations.

REFERENCES 1

3 4 5

7 8 9

CONCLUSIONS Many CRC show MSI, for which confirmatory analyses are warranted because of prognostic and therapeutic implications. Pathologists play a critical role in identifying microsatellite-unstable CRC, such as occur in young patients with synchronous or metachronous tumors and tumors with classic histologic features. In these cases, MSI testing and/or MMR immunohistochemistry are advisable, along with sequencing and genetic counseling if appropriate. Microsatellite analysis is an excellent functional and prognostic test, whereas MMR immunohistochemistry can guide gene sequencing but can result in false negatives (false negative in antigen positive neoplasms, especially cases with MLH1 promoter methylation). Direct exonby-exon gene sequencing is considered the gold standard and should be used to analyze each of the proteincoding regions of the MLH1 and MSH2 genes in their

October 7, 2006

14 15 16 17 18

19 20

Liu B, Parsons R, Papadopoulos N, Nicolaides NC, Lynch HT, Watson P, Jass JR, Dunlop M, Wyllie A, Peltomaki P, de la Chapelle A, Hamilton SR, Vogelstein B, Kinzler KW. Analysis of mismatch repair genes in hereditary non-polyposis colorectal cancer patients. Nat Med 1996; 2: 169-174 Ivanovich JL, Read TE, Ciske DJ, Kodner IJ, Whelan AJ. A practical approach to familial and hereditary colorectal cancer. Am J Med 1999; 107: 68-77 Kinzler KW, Vogelstein B. Lessons from hereditary colorectal cancer. Cell 1996; 87: 159-170 Arzimanoglou II, Gilbert F, Barber HR. Microsatellite instability in human solid tumors. Cancer 1998; 82: 1808-1820 Alexander J, Watanabe T, Wu TT, Rashid A, Li S, Hamilton SR. Histopathological identification of colon cancer with microsatellite instability. Am J Pathol 2001; 158: 527-535 Lawes DA, SenGupta S, Boulos PB. The clinical importance and prognostic implications of microsatellite instability in sporadic cancer. Eur J Surg Oncol 2003; 29: 201-212 Shackney SE, Shankey TV. Common patterns of genetic evolution in human solid tumors. Cytometry 1997; 29: 1-27 Diaz-Cano SJ, Blanes A, Wolfe HJ. PCR techniques for clonality assays. Diagn Mol Pathol 2001; 10: 24-33 Marra G, Boland CR. Hereditary nonpolyposis colorectal cancer: the syndrome, the genes, and historical perspectives. J Natl Cancer Inst 1995; 87: 1114-1125 Kim H, Jen J, Vogelstein B, Hamilton SR. Clinical and pathological characteristics of sporadic colorectal carcinomas with DNA replication errors in microsatellite sequences. Am J Pathol 1994; 145: 148-156 Mecklin JP, Sipponen P, Jarvinen HJ. Histopathology of colorectal carcinomas and adenomas in cancer family syndrome. Dis Colon Rectum 1986; 29: 849-853 Chapusot C, Martin L, Mungra N, Rageot D, Bouvier AM, Bonithon Kopp C, Ponnelle T, Faivre J, Piard F. Sporadic colorectal cancers with defective mismatch repair display a number of specic morphological characteristics: relationship between the expression of hMLH1 and hMSH2 proteins and clinicopathological features of 273 adenocarcinomas. Histopathology 2003; 43: 40-47 Greenson JK, Bonner JD, Ben-Yzhak O, Cohen HI, Miselevich I, Resnick MB, Trougouboff P, Tomsho LD, Kim E, Low M, Almog R, Rennert G, Gruber SB. Phenotype of microsatellite unstable colorectal carcinomas: Well-differentiated and focally mucinous tumors and the absence of dirty necrosis correlate with microsatellite instability. Am J Surg Pathol 2003; 27: 563-570 Koreth J, O‘Leary JJ, O‘D McGee J. Microsatellites and PCR genomic analysis. J Pathol 1996; 178: 239-248 Boland CR. Setting microsatellites free. Nat Med 1996; 2: 972-974 Brentnall TA. Microsatellite instability. Shifting concepts in tumorigenesis. Am J Pathol 1995; 147: 561-563 Thibodeau SN, Bren G, Schaid D. Microsatellite instability in cancer of the proximal colon. Science 1993; 260: 816-819 Parsons R, Li GM, Longley MJ, Fang WH, Papadopoulos N, Jen J, de la Chapelle A, Kinzler KW, Vogelstein B, Modrich P. Hypermutability and mismatch repair deciency in RER+ tumor cells. Cell 1993; 75: 1227-1236 Diaz-Cano SJ. Designing a molecular analysis of clonality in tumours. J Pathol 2000; 191: 343-344 Sieben NL, ter Haar NT, Cornelisse CJ, Fleuren GJ, CletonJansen AM. PCR artifacts in LOH and MSI analysis of

Blanes A et al. Microsatellites in Sporadic colon carcinomas

21 22

29 30

35 36

microdissected tumor cells. Hum Pathol 2000; 31: 1414-1419 Diaz-Cano SJ. Are PCR artifacts in microdissected samples preventable? Hum Pathol 2001; 32: 1415-1416 Findlay I, Matthews P, Quirke P. Multiple genetic diagnoses from single cells using multiplex PCR: reliability and allele dropout. Prenat Diagn 1998; 18: 1413-1421 Ray PF, Handyside AH. Increasing the denaturation temperature during the first cycles of amplification reduces allele dropout from single cells for preimplantation genetic diagnosis. Mol Hum Reprod 1996; 2: 213-218 Diaz-Cano SJ, Brady SP. DNA extraction from formalin-xed, parafn-embedded tissues: protein digestion as a limiting step for retrieval of high-quality DNA. Diagn Mol Pathol 1997; 6: 342-346 Diaz-Cano SJ, Blanes A, Rubio J, Matilla A, Wolfe HJ. Molecular evolution and intratumor heterogeneity by topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder. Lab Invest 2000; 80: 279-289 Boland CR, Thibodeau SN, Hamilton SR, Sidransky D, Eshleman JR, Burt RW, Meltzer SJ, Rodriguez-Bigas MA, Fodde R, Ranzani GN, Srivastava S. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res 1998; 58: 5248-5257 Hatch SB, Lightfoot HM Jr, Garwacki CP, Moore DT, Calvo BF, Woosley JT, Sciarrotta J, Funkhouser WK, Farber RA. Microsatellite instability testing in colorectal carcinoma: choice of markers affects sensitivity of detection of mismatch repairdecient tumors. Clin Cancer Res 2005; 11: 2180-2187 Jarvinen HJ, Aarnio M, Mustonen H, Aktan-Collan K, Aaltonen LA, Peltomaki P, De La Chapelle A, Mecklin JP. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology 2000; 118: 829-834 Lynch HT, de la Chapelle A. Genetic susceptibility to nonpolyposis colorectal cancer. J Med Genet 1999; 36: 801-818 Vasen HF, Mecklin JP, Khan PM, Lynch HT. The International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer (ICG-HNPCC). Dis Colon Rectum 1991; 34: 424-425 Stewenius Y, Gorunova L, Jonson T, Larsson N, Hoglund M, Mandahl N, Mertens F, Mitelman F, Gisselsson D. Structural and numerical chromosome changes in colon cancer develop through telomere-mediated anaphase bridges, not through mitotic multipolarity. Proc Natl Acad Sci USA 2005; 102: 5541-5546 Goel A, Arnold CN, Niedzwiecki D, Chang DK, Ricciardiello L, Carethers JM, Dowell JM, Wasserman L, Compton C, Mayer RJ, Bertagnolli MM, Boland CR. Characterization of sporadic colon cancer by patterns of genomic instability. Cancer Res 2003; 63: 1608-1614 Duval A, Hamelin R. Mutations at coding repeat sequences in mismatch repair-deficient human cancers: toward a new concept of target genes for instability. Cancer Res 2002; 62: 2447-2454 Peltomaki P. Deficient DNA mismatch repair: a common etiologic factor for colon cancer. Hum Mol Genet 2001; 10: 735-740 Peltomaki P. DNA mismatch repair and cancer. Mutat Res 2001; 488: 77-85 Woerner SM, Kloor M, Mueller A, Rueschoff J, Friedrichs N, Buettner R, Buzello M, Kienle P, Knaebel HP, Kunstmann E, Pagenstecher C, Schackert HK, Moslein G, Vogelsang H, von Knebel Doeberitz M, Gebert JF. Microsatellite instability of selective target genes in HNPCC-associated colon adenomas. Oncogene 2005; 24: 2525-2535 Thibodeau SN, French AJ, Cunningham JM, Tester D, Burgart LJ, Roche PC, McDonnell SK, Schaid DJ, Vockley CW, Michels VV, Farr GH Jr, O‘Connell MJ. Microsatellite instability in colorectal cancer: different mutator phenotypes and the principal involvement of hMLH1. Cancer Res 1998; 58: 1713-1718

5939 38

47 48

50 51

Miyaki M, Iijima T, Yamaguchi T, Kadofuku T, Funata N, Mori T. Both BRAF and KRAS mutations are rare in colorectal carcinomas from patients with hereditary nonpolyposis colorectal cancer. Cancer Lett 2004; 211: 105-109 Liang JT, Chang KJ, Chen JC, Lee CC, Cheng YM, Hsu HC, Chien CT, Wang SM. Clinicopathologic and carcinogenetic appraisal of DNA replication error in sporadic T3N0M0 stage colorectal cancer after curative resection. Hepatogastroenterology 1999; 46: 883-890 Jimenez-Martin JJ, Miranda MT, Blanes A, Diaz-Cano SJ. Expression of mlh1/msh2 in colo-rectal adenocarcinomas: SAGE/microarray profile and tumor phenotype. Lab Invest 2003; 83: 296A (Abstract) Blanes A, Jimenez-Martin JJ, Miranda MT, Diaz-Cano SJ. Abscence of the physiologic cellular kinetic balance and down-regulation of mlh1/msh2 characterize deep topographic compartments of colo-rectal adenocarcinomas. Lab Invest 2003; 83: 114A-115A (Abstract) Yamamoto H, Sawai H, Weber TK, Rodriguez-Bigas MA, Perucho M. Somatic frameshift mutations in DNA mismatch repair and proapoptosis genes in hereditary nonpolyposis colorectal cancer. Cancer Res 1998; 58: 997-1003 Chapusot C, Martin L, Bouvier AM, Bonithon-Kopp C, Ecarnot-Laubriet A, Rageot D, Ponnelle T, Laurent Puig P, Faivre J, Piard F. Microsatellite instability and intratumoural heterogeneity in 100 right-sided sporadic colon carcinomas. Br J Cancer 2002; 87: 400-404 Jimenez JJ, Blanes A, Diaz-Cano SJ. Microsatellite instability in colon cancer. N Engl J Med 2003; 349: 1774-1776; author reply 1774-1776 Salovaara R, Loukola A, Kristo P, Kaariainen H, Ahtola H, Eskelinen M, Harkonen N, Julkunen R, Kangas E, Ojala S, Tulikoura J, Valkamo E, Jarvinen H, Mecklin JP, Aaltonen LA, de la Chapelle A. Population-based molecular detection of hereditary nonpolyposis colorectal cancer. J Clin Oncol 2000; 18: 2193-200 Aaltonen LA, Salovaara R, Kristo P, Canzian F, Hemminki A, Peltomaki P, Chadwick RB, Kaariainen H, Eskelinen M, Jarvinen H, Mecklin JP, de la Chapelle A. Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease. N Engl J Med 1998; 338: 1481-1487 Eng C, Hampel H, de la Chapelle A. Genetic testing for cancer predisposition. Annu Rev Med 2001; 52: 371-400 Syngal S, Fox EA, Eng C, Kolodner RD, Garber JE. Sensitivity and specicity of clinical criteria for hereditary non-polyposis colorectal cancer associated mutations in MSH2 and MLH1. J Med Genet 2000; 37: 641-645 American Gastroenterological Association medical position statement: hereditary colorectal cancer and genetic testing. Gastroenterology 2001; 121: 195-197 Syngal S. Hereditary nonpolyposis colorectal cancer: a call for attention. J Clin Oncol 2000; 18: 2189-2192 Mathiak M, Rutten A, Mangold E, Fischer HP, Ruzicka T, Friedl W, Propping P, Kruse R. Loss of DNA mismatch repair proteins in skin tumors from patients with MuirTorre syndrome and MSH2 or MLH1 germline mutations: establishment of immunohistochemical analysis as a screening test. Am J Surg Pathol 2002; 26: 338-343 Southey MC, Young MA, Whitty J, Mifsud S, Keilar M, Mead L, Trute L, Aittomaki K, McLachlan SA, Debinski H, Venter DJ, Armes JE. Molecular pathologic analysis enhances the diagnosis and management of Muir-Torre syndrome and gives insight into its underlying molecular pathogenesis. Am J Surg Pathol 2001; 25: 936-941 Marcus VA, Madlensky L, Gryfe R, Kim H, So K, Millar A, Temple LK, Hsieh E, Hiruki T, Narod S, Bapat BV, Gallinger S, Redston M. Immunohistochemistry for hMLH1 and hMSH2: a practical test for DNA mismatch repair-decient tumors. Am J Surg Pathol 1999; 23: 1248-1255 Hampel H, Frankel WL, Martin E, Arnold M, Khanduja K, Kuebler P, Nakagawa H, Sotamaa K, Prior TW, Westman J,

www.wjgnet.com

5940

56 57

ISSN 1007-9327

CN 14-1219/R

World J Gastroenterol

Panescu J, Fix D, Lockman J, Comeras I, de la Chapelle A. Screening for the Lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med 2005; 352: 1851-1860 Drummond JT, Li GM, Longley MJ, Modrich P. Isolation of an hMSH2-p160 heterodimer that restores DNA mismatch repair to tumor cells. Science 1995; 268: 1909-1912 Lahue RS, Au KG, Modrich P. DNA mismatch correction in a dened system. Science 1989; 245: 160-164 Prolla TA, Pang Q, Alani E, Kolodner RD, Liskay RM. MLH1, PMS1, and MSH2 interactions during the initiation of DNA mismatch repair in yeast. Science 1994; 265: 1091-1093 Hawkins NJ, Ward RL. Sporadic colorectal cancers with microsatellite instability and their possible origin in hyperplastic polyps and serrated adenomas. J Natl Cancer Inst 2001; 93: 1307-1313 Parc YR, Halling KC, Wang L, Christensen ER, Cunningham JM, French AJ, Burgart LJ, Price-Troska TL, Roche PC, Thibodeau SN. HMSH6 alterations in patients with microsatellite instability-low colorectal cancer. Cancer Res 2000; 60: 2225-2231 Torlakovic E, Skovlund E, Snover DC, Torlakovic G, Nesland JM. Morphologic reappraisal of serrated colorectal polyps. Am J Surg Pathol 2003; 27: 65-81 Wahlberg SS, Schmeits J, Thomas G, Loda M, Garber J, Syngal S, Kolodner RD, Fox E. Evaluation of microsatellite instability and immunohistochemistry for the prediction of germ-line MSH2 and MLH1 mutations in hereditary nonpolyposis colon cancer families. Cancer Res 2002; 62: 3485-3492 Elsaleh H, Powell B, Soontrapornchai P, Joseph D, Goria F, Spry N, Iacopetta B. p53 gene mutation, microsatellite

October 7, 2006

Number 37

instability and adjuvant chemotherapy: impact on survival of 388 patients with Dukes‘ C colon carcinoma. Oncology 2000; 58: 52-59 Barratt PL, Seymour MT, Stenning SP, Georgiades I, Walker C, Birbeck K, Quirke P. DNA markers predicting benefit from adjuvant fluorouracil in patients with colon cancer: a molecular study. Lancet 2002; 360: 1381-1391 Ribic CM, Sargent DJ, Moore MJ, Thibodeau SN, French AJ, Goldberg RM, Hamilton SR, Laurent-Puig P, Gryfe R, Shepherd LE, Tu D, Redston M, Gallinger S. Tumor microsatellite-instability status as a predictor of benet from uorouracil-based adjuvant chemotherapy for colon cancer. N Engl J Med 2003; 349: 247-257 Elsaleh H, Iacopetta B. Microsatellite instability is a predictive marker for survival benet from adjuvant chemotherapy in a population-based series of stage III colorectal carcinoma. Clin Colorectal Cancer 2001; 1: 104-109 Liang JT, Huang KC, Lai HS, Lee PH, Cheng YM, Hsu HC, Cheng AL, Hsu CH, Yeh KH, Wang SM, Tang C, Chang KJ. High-frequency microsatellite instability predicts better chemosensitivity to high-dose 5-uorouracil plus leucovorin chemotherapy for stage IV sporadic colorectal cancer after palliative bowel resection. Int J Cancer 2002; 101: 519-525 Chen X, Lai MD, Huang Q. Increased sensitivity of colorectal cancer cell lines with microsatellite instability to 5-uorouracil in vitro. Chin Med J (Engl) 2002; 115: 1048-1052 Watanabe T, Wu TT, Catalano PJ, Ueki T, Satriano R, Haller DG, Benson AB 3rd, Hamilton SR. Molecular predictors of survival after adjuvant chemotherapy for colon cancer. N Engl J Med 2001; 344: 1196-1206 S- Editor Wang J

www.wjgnet.com

Volume 12

L- Editor Lutze M E- Editor Ma WH

JEADV ISSN 1468-3083

ORIGINAL ARTICLE

Blackwell Publishing Ltd

Differential kinetic features by tumour topography in cutaneous small-cell neuroendocrine (Merkel cell) carcinomas L Pozo,† JJ Sanchez-Carrillo,‡ A Martinez,§ A Blanes,‡ SJ Diaz-Cano*¶ † Department of Dermatology, Homerton University Hospital, London, UK ‡ Department of Pathology, University Hospital of Malaga, Malaga, Spain § Department of Pathology, University Hospital ‘San Carlos’, Madrid, Spain ¶ Department of Pathology, King’s College Hospital, King’s College London School of Medicine, University of London, London, UK

Keywords Apoptosis, DNA ploidy, Merkel cell carcinoma, proliferation, topographic heterogeneity *Corresponding author, King’s College Hospital, Department of Histopathology, Denmark Hill, London SE5 9RS, UK. tel. +44 20 7346 3041; fax +44 20 7346 3670; E-mail: salvador.diaz-cano@kcl.ac.uk Received: 24 November 2006, accepted 12 January 2007 DOI: 10.1111/j.1468-3083.2007.02236.x

Abstract Background/Objectives Merkel cell carcinomas (MCC) reveal epithelial and neuroendocrine differentiation, but its topographic cell kinetics remains unknown. This study analyses proliferation, apoptosis, and DNA ploidy by topography, features that can help planning therapeutic protocols. This study topographically analyses proliferation, apoptosis, and DNA ploidy. Methods We selected 27 small-cell MCCs (expressing one epithelial and two neural markers, with consistent ultrastructural findings) to evaluate mitotic figure counting, Ki-67 index, apoptosis index based on the in situ end labelling of fragmented DNA (using Escherichia coli DNA polymerase I, Klenow fragment), DNA ploidy, and BCL2 and TP53 immuno-expression. At least 50 high-power fields were screened per topographic compartment (superficial or papillary dermis, and deep or reticular dermis), recording average and standard deviation for each variable. Variables were statistically compared in each tumour compartment using analysis of variance and Student’s t-test (significant if P < 0.05). Results MCCs revealed superficial aneuploid DNA content, and no topographic differences for proliferation markers. Apoptosis showed significantly lower values in the deep compartment (average, P = 0.0050, and standard deviation, P = 0.0074), correlating with increased BCL2 and TP53 immuno-expressions. Conclusions High homogeneously distributed proliferation and superficial aneuploid DNA content defines MCCs. Apoptosis follows proliferation in superficial compartments, being less variable and proliferation independent in deep compartments, where it is inversely correlated with BCL2/TP53 expression.

Introduction Cutaneous Merkel cell carcinomas (MCC) are unusual neoplasms that reveal both epithelial and neuroendocrine differentiation. MCCs usually present as a rapidly growing painless nodule in sun-exposed areas of the head and neck of elderly patients.1 They are highly aggressive tumours with propensity toward locoregional recurrence and distant metastasis. 1220

Three cytological patterns have been described in MCCs (small-cell variant, intermediate and large cell), which have been extensively studied by immunohistochemical and ultrastructural methods.2–12 The small-cell variant is the most frequent subtype and has been related with the poorest prognosis.11 High-grade neoplasms tend to be heterogeneous and tumour heterogeneity is assumed to correlate with the molecular progression as estimated from the number of genetic abnormalities, advanced

Pozo et al.

Topographic kinetics in Merkel cell carcinomas

ﬁg. 1 Small-cell-type Merkel cell carcinomas (MCC) showed solid-trabecular growth pattern, high nuclear-cytoplasmic ratio, and inconspicuous nucleolus (H&E, ×400), along with neursecretory granules in electron microscopy analysis. Immunohistochemically, epithelial (cytokeratin 20, ABC-peroxidase ×100) and neural (neuroﬁlament, ABC-peroxidase ×400) markers were expressed in all cases.

neoplasms accumulating more alterations. Heterogeneity studies, however, have frequently overlooked the cell topography. The presence of abnormal and unrelated clones results in genetic heterogeneity, which should correlate with the tumour cell topography, as demonstrated in muscle-invasive transitional cell carcinomas of the bladder.13,14 In addition, unique kinetic profiles have been described in neoplasms, such as the inverse correlation between proliferation and apoptosis segregating with adrenal cortical adenomas in contrast to nodular hyperplasias; this profile closely correlates with the underlying genetic abnormalities.15,16 Topographic tumour heterogeneity has also been correlated with the kinetic profile in skin tumours with ductal differentiation, bladder transitional cell carcinomas, and endocrine neoplasms.13,14,17,18 This intratumour topographic heterogeneity might be reflected on kinetic features, but no detailed analysis of cell kinetics (proliferation and apoptosis) by topographic compartments of kinetic features is available in MCC to date. The knowledge of cell kinetic in these neoplasms can also help in the selection of the therapeutic options. The aim of this study is to characterize the kinetic features of a homogeneous population of small-cell-type MCCs by a detailed combined evaluation of proliferation, apoptosis, DNA ploidy, and BCL2/TP53 expression by topographic compartments, considering the correlation between proliferation and apoptosis in both superficial and deep compartments to assess any intratumour kinetic heterogeneity.

Materials and methods Case and sample selection A total of 27 cases of small-cell-type MCC (1985–1999) were retrieved from the pathology files of three reference centres (University Hospital and School of Medicine, Malaga, Spain; and ‘San Carlos’ University Hospital, Madrid, Spain). Information taken from the records included age, sex, and site of the tumour. Metastases from internal malignancies were excluded by imaging techniques in all cases. This protocol was approved by the Hospital Research Board and Ethical Committee and complied with their requirements. Cell size was based primarily on appearance in haematoxylin-and-eosin-stained sections. Tumour cells had a diameter of 14 µm or less (approximately twice the size of mature lymphocytes), scant cytoplasm, and tended to form sheets and trabeculae with little organoid growth (fig. 1). Tumours composed of cells larger than 14 µm with more discernible and generally grey cytoplasms, frequently revealing a more organoid growth pattern, were excluded from this study (six cases). All MCCs were required to express at least one epithelial and two neural markers from a panel including cytokeratin cocktail AE1– AE3, cytokeratin 7/20, synaptophysin, chromogranin A, neurofilament protein, and neurone-specific enolase, along with consistent ultrastructural findings (fig. 1). All tumours were also TTF-1 negative.

1221

Topographic kinetics in Merkel cell carcinomas

Pozo et al.

A topographic analysis was performed in all cases, considering superficial tumour cells located above the transition papillary-reticular dermis (Clark level III) and deep tumour cells those below this level.

analyse the percentage of positive tumour nuclei and the percentage of positive nuclear area from each compartment. Both nuclear and positivity thresholds were experimentally optimized in the corresponding positive control.

Mitotic ﬁgure counting

Slide cytometric analysis of DNA content

Mitotic figures were screened in the same 50 high-power fields (HPF) used for DNA cytometry and DNA extraction in each compartment (7.140 mm2) as reported,19 beginning in the most cellular area. Both the number of positive nuclei per HPF and the number of neoplastic cells intercepted by the microscope field diameter (n) were registered. The last score was used to estimate the number of neoplastic cells/HPF (N) using the formula N = (nπ/4)2 as reported,16,20,21 and expressed per 1000 tumour cells. Both the average and the standard deviation (SD) were calculated as representative scores per compartment and patient.

Feulgen-stained sections were used for DNA quantification,22 according to validated protocols for such material.15,16,22 CAS 200 and Quantitative DNA Analysis software (Becton Dickinson) were used for the densitometric evaluation. At least 200 complete, non-overlapping and focused nuclei (or the whole lesion if smaller) were measured in every case, beginning in the most cellular area until completion in consecutive HPFs. External diploid controls were used to determine DNA indices (complete rat hepatocytes and lymphocytes from reactive lymph nodes) and to standardize the nuclear area/DNA content analysis (basal keratinocytes from normal areas).23,24 The internal controls were used for setting the G0/G1 cell limits and calculating the DNA index of each G0/G1 peak (> 10% of measured cells with evidence of G2 + M cells).25–27 DNA histograms were used to calculate proliferation rate (PR = S + G2 + M/G0 + G1 + S + G2 + M), 5c exceeding rate (5cER or percentage of non-octaploid cells with DNA content exceeding 5c), and percentage of tetraploid cells (DNA index 1.85–2.25).16,25–27

Immunohistochemical detection of Ki-67, BCL2 and TP53 The sections were mounted on positively charged microscope slides (Superfrost Plus; Fisher Scientific, Fair Lawn, NJ, USA) and baked at 60 °C for 2 h. The slides were routinely dewaxed and rehydrated. The endogenous peroxidase activity was then quenched with 0.5% H2O2 in methanol, 10 min). A microwave antigen retrieval method (20 min in 10 mM citrate buffer, pH 6.0, at 600 watts) was used, followed by incubation with polyclonal horse serum (20 min, 1 : 100 dilution: Dako, Glostrup, Denmark) and with monoclonal primary antibody (overnight, 4 °C), at 2 µg/mL (MIB1 for Ki-67 and BCL2 Calbiochem, Cambridge, MA, USA; and DO7 for TP53, Dako). Then sections were serially incubated with biotinylated antimouse antibody (30 min, 1 : 200 dilution; Dako), and peroxidase-labelled avidin-biotin complex (60 min, 1 : 100 dilution; Dako). All incubations were performed in moist chamber at room temperature unless otherwise specified. The reaction was developed under microscopic control, using 3,3′diaminobenzidine tetrahydrochloride with 0.3% H2O2 as chromogen (Sigma Chemical, St Louis, MO, USA), and the sections counterstained with haematoxylin. Both positive (reactive lymph node) and negative (omitting the primary antibody) controls were simultaneously run. The threshold of positivity was experimentally established at the positive control in each staining batch. Only those nuclei with staining features similar to those of their corresponding positive control were considered positive. The immunostaining was quantified using the Cell Analysis System (CAS) model 200 and Quantitative Proliferation Index software (Becton Dickinson, San Jose, CA, USA) to 1222

Flow cytometric analysis of nuclear DNA Serial 50-µm-thick sections were microdissected, and nuclear preparations were stained with propidium iodine after RNase A digestion to study DNA ploidy.28 DNA quantification parameters included DNA indices and PRs as described,15,25,26 and the scatter analysis of nuclear area and DNA content to identify apoptotic cells in each cell cycle phase (low nuclear area for a given DNA content in each cell cycle phase), when coupled with in situ end labelling (ISEL, see below).15,23 External diploid controls from paraffin-embedded tissues (lymphocyte from reactive lymph nodes and histologically normal keratinocytes) were used to determine DNA indices and to standardize the nuclear area/DNA content analysis (considering only keratinocytes for the last purpose).15 PR was calculated as described for slide cytometry, using the rectangular model for cell cycle histogram evaluation.25

ISEL of fragmented DNA Extensive DNA fragmentation associated with apoptosis was detected by ISEL as reported.13–16,18,21 After routine dewaxing and hydration, the sections were incubated in 2× standard saline citrate (20 min at 80 °C) and digested

Pozo et al.

Topographic kinetics in Merkel cell carcinomas

fig. 2 Assessment of heterogeneity by highpower field (F) in each compartment. At least 50 fields were screened for each marker to calculate the average and standard deviation (σ). The average of standard deviation represents the intralesional heterogeneity per compartment and the standard deviation of the average per compartment and case corresponds to the interlesional heterogeneity.

with pronase (500 µg/mL, 25 min, room temperature) in a moist chamber. DNA fragments were labelled on 5′-protuding termini by incubating the sections with the Klenow fragment of Escherichia coli DNA polymerase I (20 U/mL in 50 mmol/L Tris-HCl, pH 7.5, 10 mmol/L MgCl2, 1 mmol/L DTT, 250 µg/mL bovine serum albumin, 5 µM of each dATP, dCTP, dGTP, as well as 3.25 µmol/L dTTP, and 1.75 µmol/ L 11-digoxigenin-dUTP), at 37 °C in a moist chamber. The incorporated digoxigenin-dUMPs were immunoenzymatically detected by using antidigoxigenin Fab fragments labelled with alkaline phosphatase (7.5 U/mL, in 100 mmol/L Tris-HCl, pH 7.6, 150 mmol/L NaCl, 1% bovine serum albumin) for 4 h at room temperature. The reactions were developed with the mixture nitroblue tetrazolium-X phosphate in 100 mmol/L Tris-HCl (pH 9.5), 100 mmol/L NaCl, 50 mmol/L MgCl2 under microscopic control. Appropriate controls were simultaneously run, including positive (reactive lymph node), negative (same conditions omitting DNA polymerase I), and enzymatic (DNase I digestion before the end labelling). The enzymatic controls were used to reliably establish the positivity threshold in each sample.

Quantiﬁcation of positive nuclei and statistical analysis 2

At least 50 HPF (7.6 mm ) were screened in each compartment, beginning in the most cellular area. The number of positive nuclei was expressed per HPF and per 1000 tumour cells, and the average and standard deviation (SD) calculated in each pathological condition and patient as described.13–16,18,21 The positivity threshold was experi-

mentally established at the positive control in each staining batch. Only nuclei with staining features similar to those of their corresponding positive control were considered positive for any marker. The data from both tumour compartments were used to calculate a representative value per patient for each variable (average and standard deviation). Case SD is a non-markovian texture that informs on intratumour heterogeneity in a given tumour and their SDs (considering SDs as variables) inform on series variability of intratumour heterogeneity (SD of variable-SD),29–31 that is intertumour heterogeneity (fig. 2).17 Student’s t-tests (if normal distribution was confirmed) or non-parametric analysis of variance (ANOVA, if the distribution was not normal) were applied to assess the differences by tumour compartment of average and SD values of every quantitative variable. Each variable distribution was previously tested for normality using the Kolmogorov–Smirnov test and for variance equality using the Snedecor’s F-test. Differences were considered statistically significant if P < 0.05 in two-tailed distributions. The correlation between proliferation and apoptosis markers was studied by regression analyses to calculate the correlation coefficient and the corresponding statistical significance.

Results MCC were found in 15 men and 12 women aged 72 years (± 3.3 SD) that showed the clinical features presented in Table 1. MCC revealed high cellular turnover (mitotic figure counting, Ki-67 index, and ISEL index), concordant with the undifferentiated appearances of these neoplasms (fig. 1). MCCs revealed high cell density (over 425 cell/

1223

Topographic kinetics in Merkel cell carcinomas

Pozo et al.

Table 1 Clinical features of Merkel cell carcinomas Case

Location

Local recurrence

Lymph nodes

Metastasis

Treatment

Coexistent malignancy

MCC-1 MCC-2 MCC-3 MCC-4 MCC-5 MCC-6 MCC-7 MCC-8 MCC-9 MCC-10 MCC-11 MCC-12 MCC-13 MCC-14 MCC-15 MCC-16 MCC-17 MCC-18 MCC-19 MCC-20 MCC-21 MCC-22 MCC-23 MCC-24 MCC-25 MCC-26 MCC-27

Head-neck Extremity Head-neck Head-neck Head-neck Extremity Head-neck Head-neck Extremity Extremity Trunk Head-neck Extremity Head-neck Head-neck Extremity Extremity Head-neck Trunk Head-neck Extremity Head-neck Extremity Unknown Extremity Extremity Head-neck

+ – + + + + – – + – – – + – – – – – – – + – – + – – –

+ + + + – – + – + + + + + – – + + – + + + – – + + + –

+ + – + – + + + – – + – + – – – – – – – + + + + – – –

S + RT + ChT S + RT + ChT S + RT S + RT + ChT S + RT S + ChT S + RT + ChT S + ChT S S + RT S + RT + ChT S + RT S + RT + ChT S + RT S + RT S + RT S + RT S + RT S + RT S + RT S + RT + ChT S + RT + ChT S + RT + ChT S + RT + ChT S + RT S + RT S + RT

+ – + + + – – – + + – + + + + – + – + – – – – – + – –

S, surgery; RT, radiotherapy; ChT, chemotherapy.

Table 2 Kinetic features by topographic compartments in Merkel cell carcinomas

Mitotic Figure counting Ki-67 index ISEL index

Ave. SD Ave. SD Ave. SD

Superﬁcial compartment (Ave. ± SD)

Deep compartment (Ave. ± SD)

Statistical signiﬁcance

1.01 ± 0.16 0.36 ± 0.10 47.61 ± 2.44 16.78 ± 0.83 1.84 ± 0.90 1.44 ± 0.39

1.09 ± 0.24 0.30 ± 0.07 46.33 ± 6.98 15.43 ± 2.33 1.16 ± 0.69 0.90 ± 0.31

NS NS NS NS P = 0.0050 P = 0.0074

Ave., average; SD, standard deviation; ISEL, in situ end labelling. NS, not signiﬁcant.

HPF) and no statistical differences for the proliferation markers by topographic compartments (superficial 10.10 ± 1.99%, deep 10.90 ± 2.53%). Although the scores were generally revealed higher in the superficial compartments than in the deep compartments, the differences reached statistical significance for the ISEL index only (Table 2). Apoptosis showed significantly lower values in the deep compartment (superficial 1.84 ± 0.80‰, deep 1.16 ± 0.62‰, P = 0.0050; fig. 3), with a variability level (ISEL index SD, 1224

Table 2) reached by 3 apoptotic cells/HPF in 17 HPF or 11 apoptotic cells in a single HPF for superficial compartments and 2 apoptotic cells/HPF in 14 fields or 7 apoptotic cells in a single HPF for deep compartments. These ISEL results demonstrated statistically significant differences for apoptosis intralesional heterogeneity (ISEL index, SD, P = 0.0074), the higher values obtained in the superficial compartments. In contrast, the interlesional heterogeneity (ISEL index, average SD) was not significantly different

Pozo et al.

Topographic kinetics in Merkel cell carcinomas

fig. 3 In situ end labelling (ISEL) in MCC. Significantly higher ISEL index was observed in superficial compartments (Sup) than in deep compartments (alkaline phosphatase – NBT-BCIP ×400), resulting in a significantly increased kinetic index (Ki-67/ISEL).

ﬁg. 4 Small-cell-type Merkel cell carcinomas (MCC) showed solid-trabecular growth pattern (H&E, ×200), and immunoexpression of TP53 (nuclear, ABC-peroxidase ×100) and BCL2 (cytoplasmic, ABCperoxidase ×400) in the deep compartments.

(P > 0.05 for Snedecor’s F-test). Nine cases showed strong and extensive staining for TP53, with strong BCL2 staining in 18 patients (fig. 4). In 14 cases the expression of TP53 and BCL2 was mutually exclusive in the superficial compartments only. The expression of TP53 and BCL2 correlated inversely with the ISEL index in deep compartments only (R2 = 0.8824, P = 0.014 for TP53/ISEL index; R2 = 0.9024, P = 0.025 for BCL2/ISEL index). Proliferation and apoptosis were statistically correlated in the superficial compartment only (R2 = 0.8909, P = 0.003), following a polynomic function y = –0.1992x5 + 1.8046x 4 – 5.4049x3 + 5.4435x2 + 0.2602x – 0.8562. The DNA ploidy analysis revealed aneuploid cell lines in the superficial compartments of all MCC (average DNA index aneuploid lines 1.53), while the deep compartments showed aneuploid cell lines in 14/27 cases (average DNA index 1.50); the remaining deep compartments displayed diploid DNA content (DNA index between 0.95

and 1.10) (fig. 5). The PR was not significantly different from the Ki-67 index obtained for both compartments and the 5cER was higher for those cases with hyperdiploid DNA content (all superficial compartments and 14/27 deep compartments) than for cases with diploid DNA content (13/27 deep compartments): 9.12 ± 1.25% vs. 2.89 ± 0.10%.

Discussion MCCs show topographic kinetic heterogeneity and are characterized by high cellular turnover and homogeneously distributed high proliferation. Apoptosis follows the distribution pattern of proliferation in superficial compartments, whereas in deep compartments apoptosis is down-regulated, is less variable and is independent from proliferation. These features should be considered when designing any effective treatment.

1225

Topographic kinetics in Merkel cell carcinomas

Pozo et al.

fig. 5 DNA ploidy histograms from superficial and deep compartments in MCC. Hyperdipoid cell populations (1.10 = DNA index = 1.85) were observed in superficial compartments, while dipoid/peridiploid cell populations (0.93 = DNA index = 1.10) were present in the deep compartments. Aneuploid DNA content correlated with higher cellular turnover (both proliferation and apoptosis assessed by Ki-67 immunostaining and ISEL, respectively) in MCC superficial compartments.

Apoptosis down-regulation in MCC deep compartments can contribute to genetic abnormalities accumulation in cycling cells. The topographic intratumour heterogeneity suggests a differential selection of tumour cells by compartments, but can also be the expression of either selective clonal evolution or a simple passive by-product of genetic instability.13,15,32,33 The differential kinetic profile of topographic tumour compartments has revealed lower cell turnover and apoptosis down-regulation in deep/ peripheral compartments, resulting in accumulation of genetic alterations and segregation of tumour cells with differential genetic backgrounds as demonstrated previously in colon, bladder or adrenal medullary neoplasms.13,14,18,34,35 In these organs the process is unlikely to be related with hypoxia, which is more pronounced in central compartments. The association of multiple genetic alterations would become statistically less probable as the number of molecular markers increases,13,32 and become useful to test clonal expansions in tumours. Although genetic abnormalities are probably asymmetrically acquired,36 our results show a correlation between tumour cell topography, kinetic profile, and DNA aneuploidy. Although similar results have been demonstrated in bladder, colon, or adrenal medulla,13,14,18,34 this study represents the first time the role of topography on cell kinetic segregation is assessed in skin MCC. The topographic segregation of tumour cells will influence any marker result interpretation and would help selecting better therapeutic options maximizing the effect in the most sensitive areas (e.g. zones with higher proliferation). This heterogeneous distribution would explain the result variability reported on DNA ploidy,9 or the lack of correlation between apoptosis and regression.37 No correlation between apoptosis and prognosis/therapeutic response has been reported in MCC to date; however, these analyses are based on short series and evaluate samples taken 1226

with no topographic consideration.8,12,37 A more careful selection of the samples to analyse might demonstrate such correlation and help selecting the therapeutic options. The tumour response to chemo- and radiotherapy depends in great extend on the apoptosis up-regulation induced by these therapies.38,39 Therefore, our results support a better chemo-radiotherapy response in the superficial compartments due to their higher proliferation and apoptosis, while the deep compartments would be expected to show some degree of resistance due to down-regulated apoptosis, due to up-regulation of BCL2 expression in MCC.3,6,7,10,40 In this scenario, chemo- or radiotherapy would be less effective for the treatment of MCCs with big deep compartments (e.g. tumours with extensive tumour burden in reticular dermis or deeper). For these neoplasms, alternative therapies need to be investigated. The heterogeneous distribution and clustering of apoptotic cells (significantly higher ISEL-index SD) in the superficial compartments also support a sort of clonal origin for these apoptotic cells, most likely due to the accumulation of genetic abnormalities reaching cytologically lethal levels.32,41 Likewise, the coexistence of genetic alterations in MCC supports a key role in tumorigenesis, the topographic heterogeneity resulting from the accumulation of genetic damage, partially explained by TP53 overexpression in these neoplasms.6,37,42 This TP53 overexpression frequently correlates with mutated TP53 that partially blocks apoptosis and allows accumulation of mutations.13,18 The presence of DNA-aneuploid cells support the coexistence of multiple genetic alterations and expresses predominant chromosomal instability in MCC. In this scenario, decreased aneuploid cells prevalence and loss of the physiological cell kinetic correlations (dissociated proliferation-apoptosis) in deep compartments suggest additional and collaborative molecular mechanisms of genomic instability (not necessarily independent) in this location.13,18,43

Pozo et al.

In conclusion, MCC are characterized by aneuploid DNA content (chromosomal instability), homogeneously distributed high proliferation, and apoptotic index following the proliferation distribution pattern in superficial compartments. Deep tumour compartments show downregulated apoptosis, which is less variable and independent from proliferation, and up-regulated BCL2 and TP53 expression that would make these compartments more resistant to non-surgical conventional therapy.

References 1 Akhtar S, Oza KK, Wright J. Merkel cell carcinoma: report of 10 cases and review of the literature. J Am Acad Dermatol 2000; 43: 755–767. 2 Alvarez-Gago T, Bullon MM, Rivera F, Velasco A, Mayo A. Intermediate filament aggregates in mitoses of primary cutaneous neuroendocrine (Merkel cell) carcinoma. Histopathology 1996; 28: 349–355. 3 Feinmesser M, Halpern M, Fenig E et al. Expression of the apoptosis-related oncogenes bcl-2, bax, and p53 in Merkel cell carcinoma: can they predict treatment response and clinical outcome? Hum Pathol 1999; 30: 1367–1372. 4 Hierro I, Blanes A, Matilla A, Munoz S, Vicioso L, Nogales FF. Merkel cell (neuroendocrine) carcinoma of the vulva. A case report with immunohistochemical and ultrastructural findings and review of the literature. Pathol Res Pract 2000; 196: 503–509. 5 Jemec B, Chana J, Grover R, Grobbelaar AO. The Merkel cell carcinoma: survival and oncogene markers. J Eur Acad Dermatol Venereol 2000; 14: 400–404. 6 Kennedy MM, Blessing K, King G, Kerr KM. Expression of bcl-2 and p53 in Merkel cell carcinoma. An immunohistochemical study. Am J Dermatopathol 1996; 18: 273–277. 7 Moll I, Gillardon F, Waltering S, Schmelz M, Moll R. Differences of bcl-2 protein expression between Merkel cells and Merkel cell carcinomas. J Cutan Pathol 1996; 23: 109– 117. 8 Mori Y, Hashimoto K, Tanaka K, Cui CY, Mehregan DR, Stiff MA. A study of apoptosis in Merkel cell carcinoma: an immunohistochemical, ultrastructural, DNA ladder, and TUNEL labeling study. Am J Dermatopathol 2001; 23: 16–23. 9 Parrado C, Bjornhagen V, Eusebi V et al. Prognosticating tools in primary neuroendocrine (Merkel-cell) carcinomas of the skin: histopathological subdivision, DNA cytometry, cell proliferation analyses (Ki-67-immunoreactivity) and NCAM immunohistochemistry. A clinicopathological study in 25 patients. Pathol Res Pract 1998; 194: 11–23. 10 Plettenberg A, Pammer J, Tschachler E. Merkel cells and Merkel cell carcinoma express the BCL-2 proto-oncogene. (Corrected and republished article originally printed in Exp Dermatol 1996; 5: 102–107.) Exp Dermatol 1996; 5: 183–188.

Topographic kinetics in Merkel cell carcinomas

11 Skelton HG, Smith KJ, Hitchcock CL, McCarthy WF, Lupton GP, Graham JH. Merkel cell carcinoma: analysis of clinical, histologic, and immunohistologic features of 132 cases with relation to survival. J Am Acad Dermatol 1997; 37: 734–739. 12 Takenaka H, Kishimoto S, Shibagaki R, Nagata M, Yasuno H. Merkel cell carcinoma with partial spontaneous regression: an immunohistochemical, ultrastructural, and TUNEL labeling study. Am J Dermatopathol 1997; 19: 614–618. 13 Diaz-Cano SJ, Blanes A, Rubio J, Matilla A, Wolfe HJ. Molecular evolution and intratumor heterogeneity by topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder. Lab Invest 2000; 80: 279–289. 14 Blanes A, Rubio J, Martinez A, Wolfe HJ, Diaz-Cano SJ. Kinetic profiles by topographic compartments in muscleinvasive transitional cell carcinomas of the bladder: role of TP53 and NF1 genes. Am J Clin Pathol 2002; 118: 93–100. 15 Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ. Clonality as expression of distinctive cell kinetic patterns in nodular hyperplasias and adenomas of the adrenal cortex. Am J Pathol 2000; 156: 311–319. 16 Koch M, de Miguel M, Höfler H, Diaz-Cano SJ. Kinetic profiles of intraepithelial and invasive prostatic neoplasias: the key role of downregulated apoptosis in tumor progression. Virchows Arch 2000; 436: 413–420. 17 Pozo L, Camacho F, Rios-Martin JJ, Diaz-Cano SJ. Cell proliferation in skin tumors with ductal differentiation: patterns and diagnostic applications. J Cutan Pathol 2000; 27: 292–297. 18 Blanes A, Sanchez-Carrillo JJ, Diaz-Cano SJ. Topographic molecular profile of pheochromocytomas: role of somatic down-regulation of mismatch repair. J Clin Endocrinol Metab 2006; 91: 1150–1158. 19 van Diest PJ, Baak JP, Matze-Cok P et al. Reproducibility of mitosis counting in 2469 breast cancer specimens: results from the Multicenter Morphometric Mammary Carcinoma Project (see comments). Hum Pathol 1992; 23: 603–607. 20 Harjacek M, Diaz-Cano S, Alman BA et al. Prominent expression of mRNA for proinflammatory cytokines in synovium in patients with juvenile rheumatoid arthritis or chronic lyme arthritis. J Rheumatol 2000; 27: 497–503. 21 Harjacek M, Diaz-Cano S, De Miguel M, Wolfe H, Maldonado CA, Rabinovich GA. Expression of galectins-1 and -3 correlates with defective mononuclear cell apoptosis in patients with juvenile idiopathic arthritis. J Rheumatol 2001; 28: 1914–1922. 22 Bibbo M, Bartels PH, Dytch HE, Wied GL. Cell image analysis. In: Bibbo M, ed. Comprehensive Cytopathology. W.B. Saunders Co., Philadelphia, 1991: 965–983. 23 Sherwood SW, Schimke RT. Cell cycle analysis of apoptosis using flow cytometry. Methods Cell Biol 1995; 46: 77–97. 24 Gonzalez-Campora R, Diaz Cano S, Lerma-Puertas E et al. Paragangliomas. Static cytometric studies of nuclear DNA patterns. Cancer 1993; 71: 820–824. 25 Dressler LG. Controls, standards, and histogram

1227

Topographic kinetics in Merkel cell carcinomas

30 31

32 33

Pozo et al.

interpretation in DNA flow cytometry. Methods Cell Biol 1990; 33: 157–171. Dressler LG, Bartow SA. DNA flow cytometry in solid tumors: practical aspects and clinical applications. Semin Diagn Pathol 1989; 6: 55–82. Diaz-Cano S, Gonzalez-Campora R, Rios-Martin JJ et al. Nuclear DNA patterns in adrenal cortex proliferative lesions. Virchows Arch A Pathol Anat Histopathol 1993; 423: 323–328. Hedley DW, Friedlander ML, Taylor IW, Rugg CA, Musgrove EA. Method for analysis of cellular DNA content of paraffinembedded pathological material using flow cytometry. J Histochem Cytochem 1983; 31: 1333–1335. Dawson AE, Cibas ES, Bacus JW, Weinberg DS. Chromatin texture measurement by Markovian analysis. Use of nuclear models to define and select texture features. Anal Quant Cytol Histol 1993; 15: 227–235. Pressman NJ. Markovian analysis of cervical cell images. J Histochem Cytochem 1976; 24: 138–144. Veltri RW, Partin AW, Epstein JE et al. Quantitative nuclear morphometry, Markovian texture descriptors, and DNA content captured on a CAS-200 Image analysis system, combined with PCNA and HER-2/neu immunohistochemistry for prediction of prostate cancer progression. J Cell Biochem Suppl 1994; 19: 249–258. Diaz-Cano SJ, Blanes A, Wolfe HJ. PCR techniques for clonality assays. Diagn Mol Pathol 2001; 10: 24–33. Diaz-Cano SJ, Blanes A. Influence of intratumor heterogeneity in the interpretation of marker results in pheochromocytomas. J Pathol 1999; 189: 627–628. Jimenez JJ, Blanes A, Diaz-Cano SJ. Microsatellite instability in colon cancer. N Engl J Med 2003; 349: 1774– 1776; author reply 1774–1776.

1228

35 Blanes A, Diaz-Cano SJ. Complementary analysis of microsatellite tumor profile and mismatch repair defects in colorectal carcinomas. World J Gastroenterol 2006; 12: 5932– 5940. 36 Kinzler KW, Vogelstein B. Lessons from hereditary colorectal cancer. Cell 1996; 87: 159–170. 37 Inoue T, Yoneda K, Manabe M, Demitsu T. Spontaneous regression of merkel cell carcinoma: a comparative study of TUNEL index and tumor-infiltrating lymphocytes between spontaneous regression and non-regression group. J Dermatol Sci 2000; 24: 203–211. 38 Orlandi A, Bianchi L, Costanzo A et al. Evidence of increased apoptosis and reduced proliferation in basal cell carcinomas treated with tazarotene. J Invest Dermatol 2004; 122: 1037– 1041. 39 Singh RP, Tyagi AK, Zhao J, Agarwal R. Silymarin inhibits growth and causes regression of established skin tumors in SENCAR mice via modulation of mitogen-activated protein kinases and induction of apoptosis. Carcinogenesis 2002; 23: 499–510. 40 Schlagbauer-Wadl H, Klosner G, Heere-Ress E et al. Bcl-2 antisense oligonucleotides (G3139) inhibit Merkel cell carcinoma growth in SCID mice. J Invest Dermatol 2000; 114: 725–730. 41 Diaz-Cano SJ. Designing a molecular analysis of clonality in tumours. J Pathol 2000; 191: 343–344. 42 Carson HJ, Reddy V, Taxy JB. Proliferation markers and prognosis in Merkel cell carcinoma. J Cutan Pathol 1998; 25: 16–19. 43 Pecina-Slaus N, Nikuseva-Martic T, Gall-Troselj K, Radic K, Hrascan R. Replication error-positive samples found in pheochromocytomas. In Vivo 2005; 19: 359–365.

Reduced E-cadherin expression correlates with disease progression in Paget’s disease of the vulva but not Paget’s disease of the breast Patricia E Ellis1, Salvador Diaz Cano2, Mark Fear2, David P Kelsell2, Lucy Ghali2, Julie C Crow1, Christopher W Perrett1 and Allan B MacLean1 1

Department of Obstetrics and Gynaecology, Royal Free and University College Medical School (Hampstead Campus), University College London, London, UK and 2Centre for Cutaneous Research, Institute of Cell and Molecular Sciences, Barts and The London Queen Mary’s School of Medicine and Dentistry, London, UK

The growth and metastasis of many cancers is due in part to loss of cell–cell adhesion. E-cadherin, plakoglobin and b-catenin are important in cell adhesion. Our aim was to examine the presence of these molecules in Paget’s disease of the vulva and Paget’s disease of the breast, and to correlate any differences in their expression with the presence of invasive disease or an underlying carcinoma. Sixty-three archival cases of Paget’s disease of the vulva, including eight associated with invasive disease, and 23 archival cases of Paget’s disease of breast, which included 10 cases with ductal carcinoma in situ alone, four cases with both ductal carcinoma in situ and invasive carcinoma, and five cases with underlying invasive carcinoma alone, were analysed immunohistochemically for expression of E-cadherin, plakoglobin and b-catenin proteins. The respective mRNAs were also detected by in situ hybridisation using digoxigenin-labelled cRNA probes. Seventy-six percent (41/54) of Paget’s disease of vulva cases had 450% of Paget cells expressing the Ecadherin protein, compared with 28 % (2/7) of Paget’s disease vulva with invasive disease. This result was significant, with a P-value of 0.039. Twenty-five percent (14/55) of the intraepidermal Paget’s disease of the vulva cases had 450% of Paget cells expressing the plakoglobin protein, compared with 12% (1/8) of cases of Paget’s disease of vulva with invasive disease, and for b-catenin, 9% (5/55) of the non-invasive Paget’s disease of the vulva had 450% of Paget cells expressing b-catenin, compared with 12% (1/8) of Paget’s disease of the vulva cases with invasive disease. Sixty-five percent (15/23) of the Paget’s disease of the breast had 450% of Paget cells expressing E-cadherin, and for plakoglobin and b-catenin it was 17% (4/23) and 28% (6/21), respectively. The results were not significant. The results suggest that reduced expression of E-cadherin may have a role to play in the pathogenesis of invasive Paget’s disease of the vulva. Abnormal plakoglobin expression may be involved in the formation of some cases of Paget’s of the vulva and the breast. Modern Pathology advance online publication, 9 May 2008; doi:10.1038/modpathol.2008.50 Keywords: E-cadherin; Paget’s disease; vulva

Paget’s disease of the vulva was described 100 years ago.1 It is rare and there still is a limited understanding about its association with cancer. Most cases of Paget’s disease of the vulva are adenocarcinoma in situ disease; however, in some cases there is an associated invasive adenocarcinoma present. Previous studies2,3 have suggested that the frequency of occurrence of an associated underlying carcinoma is of the order of 10–30%. Some theories

Correspondence: Dr PE Ellis, MBBS, MD, c/o Professor Allan MacLean, Department of Obstetrics and Gynaecology, Royal Free and University College Medical School (Hampstead Campus), University College London, Rowland Hill Street, London NW3 2PF,UK. E-mail: peellis@hotmail.com Received 01 October 2007; revised and accepted 01 February 2008; published online 9 May 2008

suggest that intraepidermal adenocarcinoma cells have metastasised to the overlying surface from the underlying carcinoma, whereas others have suggested that Paget cells migrate downwards to develop invasive disease.4 In contrast to Paget’s disease of the vulva, the general consensus for Paget’s disease of the breast, which was originally described by Sir James Paget in 1874,5 is that almost all cases are associated with a ductal carcinoma in situ (DCIS) or an invasive ductal carcinoma.6 Cells within the epidermis are held together by a number of adhesion complexes and their associated proteins. The growth of many cancer cells has been linked to the loss of negative regulation of cell proliferation conferred by cell–cell adhesion. The reduced expression or lack of cell adhesion molecules has therefore been implicated in the invasion and

E-cadherin expression in Paget’s disease of the vulva and breast PE Ellis et al 2

metastatic process.7 Loss or decreased cell adhesiveness has been demonstrated in cancers and thought to represent a greater risk of tumour dissemination.8 E-cadherin is a member of a multifunctional family of calcium-dependent, transmembrane glycoproteins, which promote and maintain cell adhesion.9 In vitro experiments have demonstrated E-cadherin to have invasion-suppressing properties.10,11 Downregulation of E-cadherin expression has been documented in breast12 and oesophageal carcinomas,7 squamous cell carcinomas of the skin13 and bladder carcinomas.14 Plakoglobin and b-catenin are closely related proteins that have a key role in cell adhesion (adherens junctions),15 activation of transcriptional factors, for example, T-cell factor (Tcf)16 and in the Wnt wingless-signalling pathways,17 an important pathway in the development of the cell and cancer. The objective of this study was to determine whether expression of E-cadherin, plakoglobin and b-catenin correlated with disease progression in Paget’s disease of the vulva and Paget’s disease of the breast, in order to investigate the role of the proteins as potential diagnostic markers. This study complements our other work investigating the role of the cell cycle and angiogenesis in Paget’s disease of the vulva and Paget’s disease of the breast.18,19

Materials and methods Tissue Specimens

Ethical approval has been granted by the Royal Free Hospital NHS Trust. Sixty-three cases of Paget’s disease of the vulva, including eight associated with invasive disease, and 23 cases of Paget’s disease of the breast, which included 10 cases with DCIS alone, four cases with both DCIS and invasive carcinoma, five with an underlying invasive carcinoma, and four with Paget’s disease of the breast alone, were analysed for expression of E-cadherin, plakoglobin and b-catenin. These cases were retrieved from the Histopathology Department at the Royal Free Hampstead NHS Trust, and from collaborators as listed in the acknowledgement section. The cases were diagnosed and treated between 1984 and 2000.

The archival cases had all been formalin-fixed and paraffin wax-embedded. Drs Diaz-Cano (SDC) and Julie C Crow (JCC; collaborating Histopathologists) reviewed all haematoxylin and eosin-stained slides of the cases to confirm diagnosis of Paget’s disease.

Immunohistochemistry

The streptavidin–biotin–peroxidase detection system was employed for immunohistochemistry. Paraffin wax-embedded sections of Paget’s tissue, 5-mm thick, were cut and mounted on aminopropyltriethoxysilane-coated glass slides. The sections were deparaffinised in xylene (2 5 min) and rehydrated in graded solutions of ethanol (100, 90 and 70%, 3 min each). To block endogenous peroxidase, sections were immersed in 3% hydrogen peroxide in methanol for 10 min. The sections were then left in running water (2 min). Conditions for antigen retrieval, incubation times and the primary antibodies used are described in Table 1. Sections were incubated with the secondary antibody, biotinylated rabbit anti-mouse IgG (1:50 dilution in phosphate buffered-saline, PBS; Dako) for 30 min and washed in PBS (3 5 min). Addition of streptavidin–biotin– horseradish peroxidase complex (Dako) to the sections (1:500 dilution in Tris-buffered saline) for 30 min was followed by a further wash in PBS (3 min). Antibody binding was visualised with a solution containing the chromogen 3,30 -diaminobenzidine (Sigma-Aldrich, Poole, UK). The sections were counterstained with Mayer’s haematoxylin (Merck, Lutterworth, UK). Finally, slides were rehydrated in graded ethanol rinses, cleared in xylene and mounted in DPX. Normal vulval skin was used as positive control for cases of Paget’s disease of the vulva, and normal breast skin was for cases of Paget’s disease of the breast. Apocrine and eccrine glands were used as internal controls. For negative controls, the primary antibodies were replaced by PBS. Sections were stained on two separate occasions and scored separately by two individuals (SDC, PEE) to ensure reproducibility. There was o5% variation between sections and observers.

Table 1 Antibodies used, antigen retrieval and incubation times used for immunhistochemistry. Forward and reverse primer sequences used for PCR Antigen

Source

Clone

Working dilutiona

Antigen retrieval

Incubation time

Primer

E-cadherin

Zymed

1:250

1½ h

Plakoglobin

Zymed

1:100

b-Catenin

Zymed

1:50

Microwave Pressure cooking Microwave Pressure cooking None

TTAGGTTAGAGGGTTATCGCGT TAACTAAAAATTCACCTACCGAC GCCTGCCTTCTTCTTGTGTC CTGAAGCTTTAGTGGCCAGG GGAGGTCTCCTTGGGACTC ACTAGTCGTGGAATGGCACC

Diluted in PBS.

Modern Pathology (2008), 1–8

1½ h 1½ h

E-cadherin expression in Paget’s disease of the vulva and breast PE Ellis et al 3

The location of the pattern of staining of the protein was recorded as one of the following: membranous, diffuse cytoplasmic, paranuclear and nuclear. To score, initially four quarters as a percentage of Paget cells expressing each protein were used. However, since the numbers were small in each group, 50% positivity was used as threshold. The intensity of staining was also recorded as either nil, mild or strong.

In Situ Hybridisation

Paraffin-wax-embedded sections (5-mm thick) were cut on Superfrost slides (Merk). Sections were deparaffinised, microwaved in pre-warmed 10% citrate buffer (10 min) and pre-hybridised at 421C for 4 h. Following this, sections were hybridised with 40 ng/ml of digoxigenin-labelled cRNA probe in pre-hybridisation buffer containing 20% dextran sulphate for overnight at 421C, using a Hybaid Omnislide machine. Following stringency washes to 0.1 SCC/50% formamide, sections were stained with anti-digoxigenin-alkaline phosphatase Fab fragments. After washing, sections were developed with nitroblue tetrazolium chloride/bromo-chloroidoly-phosophate (Sigma) chromogen solutions. Stained sections were scored for intensity of the mRNA signal at the intracellular junctions. Nuclear and cytoplasmic staining were recorded separately. Sections were stained on two separate occasions and the results analysed by two independent observers (PEE and Lucy Ghali (LG)). There was o5% variation between sections and observers. The surrounding apocrine glands and epidermal cells were used as positive controls. The sense (mRNA) probe was used as the negative control. Primer sequences for E-cadherin, plakoglobin and b-catenin are described in Table 1.

Statistical Analysis

Statistical analysis was performed by w2 and Fisher’s exact tests, as appropriate. A P-value of o0.05 was considered significant.

Results Evaluation of E-cadherin, Plakoglobin and b-Catenin in Paget’s Disease of the Vulva Without Invasive Disease

Forty-one of the 54 (76%) cases of Paget’s disease of the vulva without invasive disease had 450% of Paget cells expressing E-cadherin (Figure 1). The staining pattern was membranous and the intensity of staining was mostly strong. Six cases were suitable for scoring E-cadherin mRNA. Decreased mRNA signal in the Paget cells (Figure 2) was noted in all cases, as compared with

Figure 1 E-cadherin protein expression in non-invasive Paget’s disease of the vulva ( 200) with a score of o50%.

Figure 2 E-cadherin mRNA signal in Paget cells in non-invasive Paget’s disease of the vulva ( 200).

the surrounding apocrine glands and epidermal cells. Fourteen of 55 (25%) cases had 450% of Paget cells expressing plakoglobin compared with 74% with a score of o50% (Figure 3). In most cases the staining pattern was membranous and the staining intensity mild. Ten Paget’s disease of the vulva cases were analysed for plakoglobin mRNA, but only eight were suitable for scoring. All cases had decreased mRNA signal in the Paget cells, compared with in the apocrine and epidermal cells. Only 5 of 55 (9%) cases had 450% of Paget cells expressing the b-catenin protein, compared with 50 of 55 (91%) of cases with o50% (Figure 4). Again, staining was membranous and intensity was mild. Fifteen cases were analysed for b-catenin mRNA, but only nine were suitable for scoring. Eight cases revealed decreased mRNA signal in the Paget cells, compared with the surrounding apocrine glands and epidermal cells. Modern Pathology (2008), 1–8

E-cadherin expression in Paget’s disease of the vulva and breast PE Ellis et al 4

Paget’s Disease of the Vulva with Invasive Disease

Two of seven (28%) cases had 450% of Paget cells expressing E-cadherin. This result was significant, P-value ¼ 0.039, when compared with 41/54 (76%) of Paget’s disease of the vulva cases without invasive disease.

One of eight (12%) cases had 450% of Paget cells expressing the plakoglobin and b-catenin protein, respectively. The results were not significant when comparing with Paget’s disease of the vulva cases without invasive disease. The staining pattern was found to be membranous, cytoplasmic and nuclear, and staining intensity was mild. One case of invasive Paget’s disease of the vulva was suitable for scoring E-cadherin and plakoglobin mRNA. Decreased mRNA signal in the Paget cells was noted in both cases as compared with the surrounding apocrine glands and epidermal cells. The results are presented in Table 2. Evaluation of E-cadherin, Plakoglobin and b-Catenin Expression in Paget’s Disease of the Breast

Figure 3 Plakoglobin protein expression in non-invasive Paget’s disease of the vulva, with a score of o50%.

Figure 4 b-Catenin protein expression in non-invasive Paget’s disease of the vulva ( 200) with a score of o50%.

In Paget’s disease of the breast with DCIS alone (10 cases), 5 cases (50%) had 450% of Paget cells expressing E-cadherin. Only one case (10%) had 450% of Paget cells expressing plakoglobin, and 2 cases (20%) expressed b-catenin. Of the four cases of Paget’s disease of the breast with DCIS and invasive disease, all (100%) had 450% of Paget cells expressing E-cadherin, and 2 (50%) had 450% of Paget cells expressing plakoglobin and b-catenin. There were five cases of Paget’s disease of the breast with invasive disease, of which four (80%) had 450% of Paget cells expressing E-cadherin. One case (20%) had 450% of Paget cells expressing plakoglobin and two cases (40%) had 450% of Paget cells expressing b-catenin. Of the four cases of Paget’s disease of the breast alone, two (50%) had 450% of Paget cells expressing E-cadherin (Figure 5). Plakoglobin and b-catenin were not expressed in 450% of Paget cells (Figure 6). There was no significant difference between the subsets of Paget’s disease of the breast expressing E-cadherin, plakoglobin and b-catenin proteins. Table 2 shows the subsets’ combined expression of E-cadherin, plakoglobin and b-catenin proteins. Nine of the 14 Paget’s disease of breast cases, which included 4 with associated DCIS, 4 with invasive carcinoma and one case of Paget’s disease of the breast only, analysed were suitable for E-cadherin mRNA scoring. In all cases, the mRNA signal was found to be reduced in the Paget cells as compared with the surrounding apocrine glands and epidermal cells.

Table 2 E-cadherin, plakoglobin and b-catenin expression in Paget’s disease of the vulva and Paget’s disease of the breast Protein expression

Paget’s disease of the vulva with invasive disease (threshold of positivity)

Paget’s disease of the vulva without invasive disease (threshold of positivity)

P-value

Paget’s disease of the breast (threshold of positivity)

2/7 ( 450%)

41/54 (450%)

15/23 (450%)

Plakoglobin

1/8 (450%)

14/55 (450%)

P ¼ 0.039 significant P ¼ 0.77

b-Catenin

1/8 (450%)

5/55 (450%)

P ¼ 0.96

6/21 (450%)

E-cadherin

P-value for Paget’s disease of the vulva with invasive disease compared with Paget’s disease without invasive disease. Modern Pathology (2008), 1–8

4/23 (450%)

E-cadherin expression in Paget’s disease of the vulva and breast PE Ellis et al 5

Figure 5 E-cadherin protein expression in Paget’s disease of the breast alone ( 200) with a score of 450%.

Figure 7 b-Catenin mRNA signal in Paget’s disease of the breast alone ( 100).

Discussion E-cadherin Expression in Paget’s Disease of the Vulva

Figure 6 b-Catenin protein expression in Paget’s disease of the breast alone ( 200) with a score of o50%.

Eight of 11 cases of Paget’s disease of the breast, which included 5 with associated DCIS, 4 with invasive carcinoma and 2 with Paget’s disease of the breast only, were suitable for scoring mRNA plakoglobin. The mRNA signal was noted to be decreased or absent in the Paget cells, as compared with the surrounding apocrine glands and epidermal cells. Thirteen cases of Paget’s disease of the breast were analysed for b-catenin mRNA, which included 6 with associated DCIS, 4 with invasive carcinoma and 2 with Paget’s disease of breast only. In the 12 cases that were suitable for scoring, mRNA signal was noted to be decreased in 10 cases as compared with the surrounding apocrine glands and epidermal cells (Figure 7). The signal was found to be similar in intensity in both the antisense and sense slides in two cases.

This is the largest study of E-cadherin, plakoglobin and b-catenin expression in Paget’s disease of the vulva and Paget’s disease of the breast of which we are aware. Since loss of expression of E-cadherin has been suggested as one of the mechanisms contributing to development of invasion and metastasis of cancer cells, its expression has been investigated in several cancers with a view to correlating lack of expression with invasive disease. Only two smaller studies have examined this role in Paget’s disease of the vulva. The Shirahama et al study20 found that E-cadherin was not expressed in their three cases of Paget’s disease of the vulva, and Tada and coworkers21 showed decreased expression of E-cadherin in their invasive Paget’s disease of the vulva (four cases) compared with the in situ cases (four cases). In the current study we found similar results. E-cadherin expression was significantly reduced (P ¼ 0.039) in the intraepidermal compartment of the Paget’s disease of the vulva cases with invasive disease when compared with the intraepidermal compartment of Paget’s disease of the vulva cases without invasive disease. Loss of E-cadherin expression is likely to lead to reduced cell adhesiveness and therefore detachment of Paget cells from the intraepidermal lesion, the result being invasive disease. There appeared to be no significant difference in the cellular localisation of the proteins between non-invasive and invasive Paget’s disease of the vulva. There was a decrease in mRNA signal intensity when compared with the surrounding apocrine glands and epidermal cells, in both non-invasive Paget’s disease of the vulva, and in those cases associated with invasive disease, suggesting decreased transcriptional regulation of the protein Modern Pathology (2008), 1–8

E-cadherin expression in Paget’s disease of the vulva and breast PE Ellis et al 6

rather than factors affecting its stability and degradation.

may suggest involvement of aberrant activation of the Wnt-signalling pathway in some cases of Paget’s disease of the vulva.

Plakoglobin Expression in Paget’s Disease of the Vulva

The reduced expression of cell adhesion molecules has been implicated in promoting invasion and metastasis.22,23 There is a paucity of information regarding expression of plakoglobin, E-cadherin and b-catenin in Paget’s disease of the vulva and, as such, the role of these molecules has not been clearly established in Paget’s disease of the vulva.20,21,24 To our knowledge, only one study has examined the expression of plakoglobin in Paget’s disease of the vulva. Tada et al24 examined the expression of desmoglein I and plakoglobin in skin carcinomas, which included 11 cases of extramammary Paget’s disease; only one case was from a female of which the location was documented as the pudendum. This case was negative for plakoglobin. In the current study, fewer Paget cells expressed plakoglobin in the intraepidermal compartment of the Paget cases with invasive disease when compared with the intraepidermal compartment of Paget’s disease cases without invasive disease, although this did not reach statistical significance. Plakoglobin mRNA signal was also decreased compared with the surrounding apocrine glands. These findings may suggest a possible dysfunction in the regulation of plakoglobin in Paget’s disease of the vulva with invasive disease, as compared with those cases without invasion. Nuclear and diffuse cytoplasmic staining of the plakoglobin protein was noted in 13 of the cases, with only one case being associated with invasive disease.

E-cadherin Expression in Paget’s Disease of the Breast

The role of E-cadherin in breast cancers is debatable. Reduced expression of E-cadherin has been reported to be associated with poor outcome,27 whereas other investigators have demonstrated no independent prognostic value for E-cadherin in breast carcinomas.28,29 To our knowledge, only one study21 has examined the expression of E-cadherin in Paget’s disease of the breast, and both cases were negative for the protein. In contrast, in our larger study, 65% of the Paget’s disease of the breast cases in our study had 450% of Paget cells expressing E-cadherin. It is therefore unlikely that reduced E-cadherinmediated cell–cell adhesion is an important factor in the pathogenesis of Paget’s disease of the breast. Plakoglobin Expression in Paget’s Disease of the Breast

Plakoglobin has not been examined in Paget’s disease of the breast, although it has been examined in primary breast carcinomas.30–33 Loss of heterozygosity and reduced expression of plakoglobin associated with disease progression have been demonstrated in breast cancer. In the current study, only 17% of the cases of Paget disease of the breast had 450% of Paget cells expressing plakoglobin. This implicates a possible role for plakoglobin in the formation of Paget’s disease of the breast. There was good correlation between plakoglobin mRNA and protein expression.

b-Catenin Expression in Paget’s Disease of the Vulva

The transcriptional activity of b-catenin and its degradation are regulated by the Wnt pathway. In normal epithelial cells, b-catenin is localised at the cell membrane. The unbound b-catenin is degraded by the ubiquitin–proteasome system, which involves the GSK-3b. Stabilisation of cytoplasmic b-catenin by aberrant activation of Wnt signalling leads to its accumulation, complex with lymphoid enhancer factor/Tcf (LEF/Tcf) transcription factors and transactivation of LEF/Tcf target genes. Activation of these genes can lead to cell proliferation or inhibition of apoptosis. Nuclear accumulation of b-catenin can be the result of gene mutations.25 In colon cancers, disruption of the Wnt-signalling pathway by mutations of either the adenomatous polyposis coli or the b-catenin gene plays a crucial part in the early stage of tumorigenesis.26 In the current study, diffuse cytoplasmic and paranuclear staining of the b-catenin protein was found in 11 and four cases of Paget’s disease of the vulva, respectively. Three of these cases were of Paget’s disease of the vulva with invasion. These results Modern Pathology (2008), 1–8

b-Catenin Expression in Paget’s Disease of the Breast

Few studies have sought to examine the role of b-catenin in breast carcinomas,30,31 and none have examined b-catenin in Paget’s disease of the breast. Forty-eight percent of our cases of Paget’s disease of the breast had nuclear and paranuclear staining of b-catenin protein. The extramembranous accumulation could be due to mutation in the b-catenin gene or its lack of degradation by the APC/GSK3b/ proteasome system. Except in one case, there was good correlation between mRNA signal and protein expression. The adhesive function of cadherins is dependent on their interaction with catenins. Some reports have revealed reduced expression of both cadherins and catenins in certain tumours, for example, oral squamous cell carcinoma.34 In the current study, there appeared to be no correlation between reduced expression of E-cadherin and b-catenin. In summary, reduced expression of E-cadherin may a have a role in the pathogenesis of Paget’s disease of the vulva with invasive disease, unlike in

E-cadherin expression in Paget’s disease of the vulva and breast PE Ellis et al

Paget’s disease of the breast, where in the majority of cases, normal expression of the protein was demonstrated. Abnormal plakoglobin expression may be involved in some cases of Paget’s disease of the vulva and Paget’s disease of the breast. These results suggest that there may be different mechanisms underlying the aetiology of these two diseases, and loss of cell–cell adhesion may be an important factor in their development.

Acknowledgement We thank the following for their assistance in obtaining cases of Paget’s disease of the vulva and Paget’s disease of the breast: Dr C Andrews (The General Infirmary, Leeds), Dr S Andrews (Hope Hospital, Manchester), Dr L Brown (Leicester Royal Infirmary, Leicester), Dr E Courtauld (Farrer-Brown Laboratory, London), Dr P Cross (Queen Elizabeth Hospital, Gateshead), Dr A Desai (Whittington Hospital, London), Dr R Dino (Queen Charlotte’s and Chelsea Hospital, London), Dr A Flanagan (St Mary’s Hospital, London), Dr J Johnson (Nottingham City Hospital, Nottingham), Professor T Krauz (Hammersmith Hospital, London), Dr S Lakhani (University College London), Professor D Lowe (St Bartholomew’s Hospital, London), Dr P Millard (The John Radcliffe Hospital, Oxford), Dr N Nasseri (The Royal Marsden Hospital, London), Dr J Smith (The Northern General Hospital, Sheffield), Dr P Trott (The London Clinic, London), Professor M Wells (The Royal Hallamshire Hospital, Sheffield), Dr G Wilson (Manchester Royal Infirmary, Manchester), Dr M Young (St George’s Hospital, London). Partial support for P Ellis came from the South Essex Medical Education and Research Trust, UK.

Duality of interest There was no duality of interest in this study.

References 1 Dubreuilh W. Paget’s disease of the vulva. Br J Dermatol 1901;13:407–413. 2 Chanda JJ. Extramammary Paget’s disease: prognosis and relationship to internal malignancy. J. Am Acad Dermatol 1985;113:1009–10013. 3 Bohem F, Morris JMCL. Paget’s disease and apocrine gland carcinoma. Obstet Gynaecol 1971;38:185–192. 4 Fox H, Wells M. Recent advances in the pathology of the vulva. Histopathology 2003;42:209–216. 5 Paget J. On the disease of the mammary areola preceding cancer of the mammary gland. St Bartholomew Hospital Reports 1874;10:87–89. 6 Fu W, Lobocki C, Silberberg B, et al. Molecular markers in Paget’s disease of the breast. J Surg Oncol 2001;77:171–178.

7 Shiozaki H, Tahara H, Oka H, et al. Expression of immunoreactive E-cadherin adhesion molecules in human cancers. Am J Pathol 1991;139:17–23. 8 Conman CR. Adhesiveness and stickiness: two independent properties of the cell surface. Cancer Res 1961;21:1436–1438. 9 Takeichi M. Cadherin cell adhesion receptors as a morphogenetic regulator. Science 1991;251:1451– 1455. 10 Behrens J, Mareel MM, Van Roy FM, et al. Dissecting tumour cell invasion: epithelial cells acquire invasion properties after the loss of uvomorulin-mediated cell– cell adhesion. J Cell Biol 1989;108:2435–2447. 11 Vleminck K, Vakaet Jr L, Mareel M, et al. Genetic manipulation of E-cadherin expression by epithelial tumour cells reveals invasion suppressor role. Cell 1991;66:107–119. 12 Charpin C, Garcia S, Bonnier P, et al. Reduced E-cadherin immunohistochemical expression in node-negative breast carcinomas. Am J Clin Pathol 1997;109:431–438. 13 Fuller LC, Allen MH, Montesu M, et al. Expression of E-cadherin in human epidermal non-melanoma cutaneous tumours. Br J Dermatol 1996;134:28–32. 14 Bringuier PP, Umbas R, Schaafsma E. Decreased E-cadherin immunoreactivity correlates with poor survival in patients with bladder tumours. Cancer Res 1993;52:5104–5109. 15 Ben Zee’ev A, Geiger B. Differential molecular interactions of beta-catenin and plakoglobin in adhesion, signalling and cancer. Curr Opin Cell Biol 1998;10:629–639. 16 Barker N, Morin PJ, Clevers H. The Yin-Yang of TCF/ beta-catenin signalling. Adv Cancer Res 2000;77: 1–24. 17 Willert K, Nusse R. Beta-catenin: a key mediator of Wnt signalling. Curr Opin Genet Dev 1998;8:95–102. 18 Ellis PE, Wong Te Fong LF, Rolfe KJ, et al. The role of p53 and Ki67 in Paget’s disease of the vulva and breast. Gynecol Oncol 2002;86:150–156. 19 Ellis PE, Wong Te Fong LF, Rolfe KJ, et al. The role of vascular endothelial growth factor-A (VEGF-A) and platelet-derived endothelial cell growth factor/thymidine phosphorylase (PD-ECGF/TP) in Paget’s disease of the vulva and breast. Anticancer Res 2002;22:857– 861. 20 Tada H, Hatoko M, Muramatsu T, et al. Expression of E-cadherin in skin carcinomas. J Dermatol 1996;23: 104–110. 21 Shirahama S, Furukawa F, Wakita H, et al. E- and P-cadherin expression in tumour tissues and soluble E-cadherin levels in sera of patients with skin cancer. J Dermatol Sci 1996;13:30–36. 22 Oka H, Shiozaki H, Kobayashi K, et al. Expression of E-cadherin cell adhesion molecules in human breast cancer tissues and its relationship to metastasis. Cancer Res 1993;53:1696–1701. 23 Miyata M, Shiozaki H, Iihara K, et al. Relationship between E-cadherin expression and lymph node metastasis in human oesophageal cancer. Int J Oncol 1994;4:61–65. 24 Tada H, Hatoko M, Tanaka A, et al. Expression of desmoglein I and plakoglobin in skin carcinomas. qJ Cutan Pathol 2000;27:24–29. 25 Sparks PJ, Korinek V, Barker N, et al. Activation of b-catenin–Tcf signalling in colon cancer by mutations b-catenin or APC. Science 1994;275:1787–1790.

Modern Pathology (2008), 1–8

E-cadherin expression in Paget’s disease of the vulva and breast PE Ellis et al 8

26 Morin P, Sparks A, Korinek V, et al. Activation of b-catenin–Tcf signalling in colon cancer by mutations in b-catenin or APC. Science 1997;275:1787–1790. 27 Oka H, Shiozaki H, Kobayashi K, et al. Expression of E-cadherin cell adhesion molecules in human breast cancer tissues and its relationship to metastasis. Cancer Res 1993;53:1696–1701. 28 Lipponen P, Saarelainen E, Ji H, et al. Expression of E-cadherin (E-CD) as related to other prognostic factors and survival in breast cancer. J Pathol 1998;174:101– 109. 29 Soler AP, Knudsen KA, Salazar H, et al. P-cadherin expression in breast carcinomas indicate poor survival. Cancer 1999;86:1263–1272. 30 Bukholm IK, Nesland JM, Karesen R, et al. E-cadherin and a-, b- and g-catenin protein expression in relation

Modern Pathology (2008), 1–8

31 32

to metastasis in human breast carcinoma. J Pathol 1995;185:262–266. Zschiesche W, Schonborn I, Behrens J, et al. Expression of E-cadherin and catenins in invasive mammary carcinomas. Anticancer Res 1997;17:561–567. Sommers CL, Gelmann EP, Kemler R, et al. Alterations in beta-catenin phosphorylation and plakoglobin expression in human breast cancer cells. Cancer Res 1994;54:3544–3552. Aberle H, Bierkamp C, Torchard D, et al. The human plakoglobin gene localizes on chromosome 17q21 and is subjected to loss of heterozygosity in breast and ovarian cancers. Proc Natl Acad Sci USA 1995;92:6384–6388. Muzio LO, Staibano S, Pannore G, et al. Beta and gamma catenin expression in oral squamous cell carcinoma. Anticancer Res 1999;19:3817–3826.

Human Pathology (2009) 40, 988–997

www.elsevier.com/locate/humpath

Original contribution

Coexistent intraurothelial carcinoma and muscle-invasive urothelial carcinoma of the bladder: clonality and somatic down-regulation of DNA mismatch repair Alfredo Blanes MD, PhD a , Javier Rubio MD, PhD a , Juan J. Sanchez-Carrillo MD, PhD a , Salvador J. Diaz-Cano MD, PhD, FRCPath a,b,⁎ a

Department of Pathology, University Hospital of Malaga, 29071 Malaga, Spain Department of Pathology, King's College Hospital and King's College London School of Medicine, SE5 9RS London, United Kingdom b

Received 29 July 2008; revised 11 December 2008; accepted 19 December 2008

Keywords: Bladder; Carcinoma in situ; Low-grade dysplasia; Urothelial carcinoma; Heterogeneity; Tumor suppressor genes; DNA mismatch repair

Summary Muscle-invasive urothelial carcinomas are heterogeneous neoplasms for which the clonal relationship with low-grade urothelial dysplasia and carcinomas in situ remains unknown, and both monoclonal and field change models have been proposed. Low-grade dysplasia (18) and carcinoma in situ (12) associated with muscle-invasive urothelial carcinoma were microdissected and topographically analyzed (intraepithelial and invasive superficial and deep to muscularis mucosa) for methylation pattern of androgen receptor alleles, TP53, RB1, WT1, and NF1 microsatellite analysis to assess clonal identity; MLH1 and MSH2 sequencing/immunostaining. Appropriate controls were run. Carcinoma in situ (100%) and invasive urothelial carcinoma (100%) revealed monoclonal patterns, whereas low-grade dysplasia was preferentially polyclonal (80%). Carcinoma in situ showed aneuploid DNA content and more abnormal microsatellites than the corresponding invasive compartments, opposite to low-grade dysplasia. Absent MLH1 protein expression with no gene mutations were identified in carcinoma in situ and nodular-trabecular urothelial carcinoma with high microsatellite abnormalities. Somatic mismatch repair protein down-regulation and the accumulation of tumor suppressor gene microsatellite abnormalities contribute to a molecular evolution for monoclonal carcinoma in situ divergent from coexistent muscle-invasive urothelial carcinoma. Low-grade dysplasia is however unlikely connected with this molecular progression. © 2009 Elsevier Inc. All rights reserved.

1. Introduction Presented in part as abstracts in the meetings of the United States and Canadian Academy of Pathology, Atlanta, GA, 2001; San Antonio, TX, 2005; and Denver, CO, 2008; and Pathological Society of Great Britain and Ireland, Maastricht, Netherland, 2001. ⁎ Corresponding author. Department of Histopathology, King's College Hospital, SE5 9RS London, United Kingdom. E-mail address: salvador.diaz-cano@kcl.ac.uk (S. J. Diaz-Cano). 0046-8177/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.humpath.2008.12.009

Urothelial dysplasia is assumed to be the putative precursor of urothelial carcinoma (UCC) and confers a significant risk for the development of carcinoma in situ (CIS) and invasive UCC, as reported for intraepithelial breast or melanocytic lesions [1]. Although the accumulated molecular data indicate that most recurrent and multiple tumors are monoclonal, the controversial

MMR and Microsatellites in VCC definitions of flat lesions with atypia, that is, reactive atypia, atypia of unknown significance, low-grade urothelial dysplasia (LGUD), and high-grade urothelial dysplasia-CIS [2], have contributed to create confusion for coexistent lesions [3]. In addition, the concept of tumor progression is not consistently used in bladder pathologic examination [4], being a cytologic progression documented in urothelial dysplasias but with a different topography for LGUD and CIS. This supports the multifocal distribution of low-grade UCC without proving a clonal identity of LGUD and CIS to sustain the sequence LGUD 窶年 CIS. The clonality status of multifocal bladder tumors is still controversially discussed with experimental evidence for both monoclonality and field cancerization. Early stage urothelial neoplasms have shown chromosome 9 deletions and FGFR3 mutations [5,6], the same genetic alteration being observed in coexistent low-grade papillary superficial UCCs and histologically normal urothelium, whereas 17p13 hemizygosity was observed in a minority of urothelial hyperplasias and papillary tumors but not in normal urothelium. This genetic profile suggests that the earliest molecular alterations in the pathogenesis of low-grade UCC involve p16/CDKI2 but not TP53 even in histologically normal areas, but this study does not analyze urothelial dysplasia and high-grade muscle-invasive UCC [1,7]. Clusters of discontinuous deleted segments of tumor suppressor gene loci on chromosomes 13q14 and 17p13 have been associated with clonal expansion of in situ bladder preneoplasia using single nucleotide polymorphisms [8]. The clonal relationships between LGUD and CIS associated with muscle-invasive UCC have not been addressed. Although bladder UCC infrequently reveals microsatellite instability [9-11], the reduced expression of mismatch repair (MMR) proteins contributes to the development of a subset of UCC [12,13]. The relationship (linear versus divergent) between intraurothelial and invasive compartments (superficial and deep) has not been topographically analyzed in muscle-invasive UCC showing these 3 compartments. This study evaluates in each topographic compartment as follows: clonality, MMR protein expression/sequencing, and loss of heterozygosity (LOH)microsatellite profile of tumor suppressor genes (TSG) controlling G1-S transition (TP53, RB1), RAS pathway (NF1), and development (WT1). All these analyses have not been performed simultaneously in a series of muscle-invasive and intraurothelial lesions and will inform on both monoclonal/field change models in muscle-invasive UCC and tumor progression from the perspective of the accumulation of genetic alterations in tumor suppressor genes.

2. Materials and methods 2.1. Case selection and sampling We reviewed the initial transurethral resection biopsy of pT2a/b UCC with no special differentiation of the bladder

989 diagnosed in women (44 cases), treated with cystectomy and lymphadenectomy from 3 reference hospitals (1990-1992, median follow-up 60 months). Transurethral resection biopsies were selected because they provided better cellular preservation, and the results for molecular tests are more reliable as they show lower frequency of artifacts [14,15]. Intraurothelial neoplastic lesions were classified according to the World Health Organization system in low-grade dysplasia (LGUD, 11 patients) and carcinoma in situ (CIS, 7 patients) [3], both coexisting in 3 patients. To add power to the clonality analysis, a combined approach of X chromosome inactivation and tumor suppressor gene microsatellite profile was selected. The X chromosome inactivation requires samples from female that represent the initial patient selection; to extend the analysis, cases from male were selected (28, which revealed LGUD in 7 patients and CIS in 5 patients, coexistent lesions in 2), the selection criteria then included matching cases by age and conventional histologic features of the muscle-invasive component to avoid any biases from those aspects. The cases were classified independently by 2 pathologists (AB, SDC). In case of grading disagreement, the lesions were discussed during simultaneous inspection before final categorization. Reproducibility data were not recorded. All surgical specimens were completely embedded for histopathologic diagnosis. The topographic compartment limit was the muscularis mucosa (superficial and deep to the muscularis mucosa), being the same areas analyzed in each study [16-18]. The muscularis mucosae was selected as limit because tumors invasive to this level had shown a much better 5-year survival than tumors invasive through the level of the muscularis mucosae, which showed survival comparable with patients with tumors invasive of the muscularis propria [19]. This protocol was approved by the Hospital Research Board and Ethical Committee and complied with their requirements.

2.2. Clonality assay and TSG microsatellite analysis DNA was extracted from the most cellular areas of intraurothelial, superficial, and deep compartments, after microdissecting at least 100 cells (approximately 0.4 mm2, laser capture; Arturus, Sunnyvale, CA) from two 20-ﾎｼm unstained paraffin sections/compartment. Appropriate tissue controls (histologically normal urothelium, stroma from the lamina propria, and smooth muscle) and quality-assurance controls (sensitivity, specificity, positive, and negative) were run for each test [14,15,20]. DNA was extracted using a modified phenol-chloroform protocol, precipitated with ice-cold absolute ethanol, and resuspended in 10 ﾎｼL of Tris-HCl buffer pH 8.4 [15]. DNA was then used for polymerase chain reaction (PCR) amplification of TSG intron microsatellites and the hypervariable CAG repeat in the first exon of the human androgen receptor (see Table 1 for primer sequences and cycling conditions), using

990 Table 1

A. Blanes et al. Primer sequences and PCR cycling conditions

Primers

Primer sequences

Repeats/PCR product

PCR cycling conditions

AR-F⁎ AR-R⁎

5′-CCG AGG AGC TTT CCA GAA TC-3′ 5′-TAC GAT GGG CTT GGG GAG AA-3′

CAG/215-300 bp

⁎AR alleles were amplified using “hot start” protocol, 0.3 μmol/L of each primer, and 200 μmol/L of each dNTP (including 7-deaza-dGTP instead of dGTP) (Boehringer-Mannheim, Indianapolis, IN), completing 28 cycles with an annealing temperature of 55°C. The amplicon was internally labeled with 0.3 μCi α[32P]-dTTP (800 Ci/mmol, 10 mCi/mL; New England Nucleotide, Boston, MA). ‡ TSG polymorphic regions were amplified using 0.25 μmol/L of each primer, 50 μmol/L of each dNTP (Boehringer-Mannheim, Indianapolis, IN), and internally labeled with 0.3 μCi α[32P]-dCTP (3000 Ci/mmol, 10 mCi/mL; New England Nucleotide, Boston, MA). The annealing temperature was 55°C (except for NF1, 52°C), and the number of cycles was experimentally optimized to 26.

TP53(1)-F‡ 5′-AGG GAT ACT ATT CAG CCC-3′ TP53(1)-R‡ 5′-ACT GCC ACT CCT TGC CCC ATT C-3′

TP53(2)-F‡ TP53(2)-R‡ RB1-F‡ RB1-R‡ WT1-F‡ WT1-R‡ NF1-F‡ NF1-R‡

CA/103-135 bp

5′-GAA TCC GGG AGG AGG TTG-3′ AAAAT/140-175 bp 5′-AAC AGC TCC TTT AAT GGC AG-3′ 5′-CTC CTC CCC TAC TTA CTT GT-3′ CTTT(T)/266-306 bp 5′-AAT TAA CAA GGT GTG GTG GTA CAC G-3′ 5′-AAT GAG ACT TAC TGG GTG AGG-3′ CA/approximately 144 bp 5′-TTA CAC AGT AAT TTC AAG CAA CGG-3′ 5′-CAG AGC AAG ACC CTG TCT-3′ CA/171-187 bp 5′-CTC CTA ACA TTT ATT AAC CTT A-3′

Abbreviations: F, forward; R, reverse. All reactions were run in duplicate using 1.5 mmol/L of MgCl2 and 1 μL of template, as well as long denaturation (4 minutes) and expansion (90 seconds) in the first 3 cycles. The appropriate PCR primers were designed using Genrunner software (version 3.02; Hastings Software Inc., Hudson, NY).

HhaI-undigested and digested samples for the X chromosome inactivation assay that contained XhoI-linearized φX174-RII phage (Gibco-BRL, Gaithersburg, MD) as mimicker of digestion completion checked by gel electrophoresis (Table 1) [16,17,20-22]. The tests were run in a Perkin-Elmer thermal cycler model 480 (Perkin-Elmer, Norwalk, CT). The whole 10μL PCR volume was electrophoresed into 8% denaturing gradient polyacrylamide gels; dried gels were put inside developing cassettes containing one intensifying screen and preflashed films (Kodak XAR) [17,23]. The radiographs were developed using an automated processor Kodak-Omat 100 (Kodak Co, Rochester, NY). Interpretation and inclusion criteria included [14,17,20-22] the following: (a) allelic imbalance was densitometrically evaluated (EC model 910 optical densitometer, EC Apparatus Corporation, St Petersburg, FL), considering evidence of LOH only allele ratios 4:1 or greater in any TSG; otherwise, retention of heterozygosity was assigned [17,22]. This ratio would represent 80% of clonal cells in the sample and was used to increase the detection specificity [20,21,24]. (b) Additional allele bands present in tumor samples but not in

the corresponding controls were considered evidence of somatic microsatellite abnormality by PCR/denaturing gradient gel electrophoresis [17].

2.3. DNA sequencing All microsatellite extrabands were cut from gels, and DNA was purified using a QIA quick gel extraction kit (Qiagen, West Sussex, UK). The amplified product was diluted 20-fold in TE buffer, and 1-μL of the diluted reaction product was subjected to a second round of PCR amplification using the appropriate primers for 30 cycles under the above conditions. Normal and extrabands from tumorderived samples were PCR amplified along with the corresponding controls using a high-fidelity polymerase, Platinum PFX (Life Technologies). PCR products were directly sequenced after purification (QIAquick PCR purification kit, Qiagen). All sequencing was performed on an ABI Prism 3700 automated DNA analyzer, and the sequence data were analyzed using the program Sequencher

MMR and Microsatellites in VCC (Gene Codes Corporation, Ann Arbor, MI), which reverses and complements the antisense strand. All mutations were confirmed by sequencing in both directions and indicated by an “N” in the sequencing chromatogram. MLH1/MSH2 exons were completely sequenced in cases with microsatellite abnormalities in at least 40% loci and/or complete loss of mlh1/msh2 immunoreactivity, as well as in a sample of mlh1/msh2 immunoreactive UCC [20] as controls.

2.4. Immunohistochemical detection of TP53, MLH1, and MSH2 The sections were mounted on positively charged slides (Superfrost Plus, Fisher Scientific, Fair Lawn, NJ), baked at 60°C for 2 hours, and processed as described [16,17,25]. After routine dewaxing and rehydration, endogenous peroxidase quenching, and antigen heat retrieval (pressure cooker, citrate buffer [10 mmol/L], for all antibodies), the slides were transferred to a moist chamber. Nonspecific binding was blocked with polyclonal horse serum, and sections were incubated with monoclonal primary antibodies (overnight, 4°C) as follows: 2 μg/mL of p53 DO-7, Calbiochem, Cambridge, MA; hMLH1 clones G168 728 and G168-15,

991 BD Pharmingen Biosciences, Oxford, UK; hMSH2 clone FE11, Oncogene Research (Merck Chemicals Ltd., Nottingham, UK). Then sections were serially incubated with biotinylated antimouse antibody and peroxidase-labeled avidin-biotin complex. The reaction was developed under microscopic control, using 3,3′-diaminobenzidine tetrahydrochloride with 0.3% H2O2 as chromogen (Sigma Co, St Louis, MO), and the sections were counterstained with hematoxylin. Both positive (reactive lymph node) and negative (omitting the primary antibody) controls were simultaneously run. Basal cells of the unaffected urothelial mucosa were used as internal positive controls for mlh1 and msh2.

2.5. Nuclear DNA quantification by slide cytometry Feulgen-stained sections were used for DNA quantification [26]. The densitometric evaluation was performed with the cell analysis system model 200 and the quantitative DNA analysis software package (Becton Dickinson, Oxford, UK). At least 300 complete, nonoverlapping, and focused nuclei (or the whole lesion if smaller) were measured in every case, beginning in the most cellular area until completion in consecutive high-power fields (HPFs).

Fig. 1 Coexistent urothelial CIS was associated with nodular-trabecular UCC (sheets of neoplastic cells with minimal stromal reaction, replacing the muscularis propria) but not with infiltrative UCC (small tumor nests/thin cords embedded in a prominent desmoplastic reaction, dissecting the smooth muscle fibers) (hematoxylin-eosin, original magnification ×400).

992

A. Blanes et al.

Fig. 2 Clonality and microsatellite profile in bladder intraurothelial lesions associated with muscle-invasive UCC. Frequency of tumor suppressor gene microsatellite (TSG MS) abnormalities in LGUD and CIS. Abbreviations: ROH, retention of heterozygosity (green cell); LOH/SNP, loss of heterozygosity/single nucleotide polymorphism (red cell); NI, noninformative (gray cell); TP53, tumor protein p53; RB1, retinoblastoma; WT1, Wilms tumor 1; NF1, neurofibromatosis 1.

External staining calibration was carried out with complete rat hepatocytes (Becton-Dickinson; one slide per staining holder) to normalize the internal controls (lymphocytes and histologically normal urothelial cells present in the same tissue section), used for setting the G0/G1 cell limits and calculating the DNA index of each G0/G1 peak (N10% of measured cells with evidence of G2 + M cells) [27]. Proliferation rate (PR = S + G2 + M-phases fraction) was calculated from the DNA histogram by subtracting the number of cells within G0/G1 limits from the total number of measured cells and expressed as percentage [26,27]. External diploid controls were used to determine DNA indices (lymphocytes from reactive lymph nodes) and to standardize the nuclear area/DNA content analysis (normal transitional cells) [28].

2.6. Tumor infiltration pattern, grading, and mitotic figure counting The infiltration pattern was evaluated in deep compartments, classifying the tumor by the predominant pattern (N50%) in nodular-trabecular and infiltrative [29]. The

histologic grading evaluated architectural features, nuclear grade, and mitotic figure (MF) counting [13]. MFs were screened in 50 HPF/compartment (7.140 mm2) or the whole tumor if smaller (3 superficial and 6 deep compartments), beginning in the most cellular area [16]. Both the number of positive nuclei/HPF and the number of neoplastic cells intercepted by the microscope field diameter (n) were registered, the latter to estimate the number of neoplastic cells/HPF (N = [nĎ€/4]2) [30]; results were expressed per 1000 cells, calculating average and SD per compartment and patient. Tumors were graded by 3 independent observers (JR, AB, and SJD-C), being the tumor discussed during simultaneous inspection before final categorization in case of disagreement. Reproducibility data were not recorded.

2.7. Quantification of positive nuclei and statistical analysis At least 50 HPF (7.6 mm2) were screened in each pathologic group, beginning in the most cellular area. The number of positive nuclei was expressed per HPF and per 1000 tumor cells, calculating average and SD for each pathologic condition

MMR and Microsatellites in VCC and patient [22,30]. The positivity threshold was experimentally established at the positive control in each staining batch. Only nuclei with staining features similar to those of their corresponding positive control were considered positive for any marker. Categorized variables were tested using Fisher exact tests and quantitative variables using Student t tests (if normally distributed) and nonparametric tests (Mann-Whitney for 2group comparisons and Kruskal-Wallis for N2-group comparisons). Differences were considered significant if P b .05 in 2-tailed distributions.

3. Results CIS was identified in patients with nodular-trabecular UCC (Fig. 1) and revealed more abnormal TSG loci than the corresponding invasive compartment (10/12 patients), TP53 loci being involved in all patients (Fig. 2) with expression of abnormal p53 protein. CIS showed either an additional TSG

993 locus involved (8 patients; TP53 in 5, RB1 in 2, and WT1 in 1) or a combined pattern of superficial and deep compartments (2 patients). The other 2 cases showed TP53 abnormality only or different microsatellite abnormalities at the same loci in intraurothelial and invasive compartments. In contrast, LGUD revealed LOH in 2 patients, one at RB1 (monoclonal methylation of androgen receptor alleles) and the second at WT1-NF1 loci (polyclonal pattern), respectively (Fig. 2). CIS (6; 100%), invasive UCC (13; 100%), and LGUD (2; 20%) from informative females revealed unbalanced methylation pattern of androgen receptor alleles, whereas polyclonal patterns were observed in LGUD only (8; 80%; Figs. 2 and 3). Discordant pattern of AR allele was observed in one case, the larger allele being methylated in LGUD and the smaller allele in CIS-invasive UCC. The UCC microsatellite profile of superficial and deep compartments was proven statistically different from CIS profile at TP53 locus only (P = 0.042; Fig. 4), showing similar topographic heterogeneity in UCC invasive compartments, regardless of the presence or absence of CIS.

Fig. 3 TSG microsatellite pattern in carcinoma in situ (CIS, all monoclonal), LGUD (mainly polyclonal), and muscle-invasive UCC (from superficial [sup] and deep compartments) (panel A). Representative gels of the methylation allele pattern of androgen receptor from CIS (monoclonal, TCC1-CIS) and LGUD (monoclonal, TCC7-LGUD, and polyclonal, TCC2-LGUD) (panel B). Mismatch protein expression. Nuclear mlh1 and msh2 expression is demonstrated in UCC with no microsatellite abnormalities and at least one of these proteins was absent (in particular mlh1) in UCC with microsatellite abnormalities. MLH1 and MSH2 exon sequencing. Normal sequence is demonstrated for these genes, regardless of the microsatellite pattern.

994

A. Blanes et al.

Fig. 4 Clonality and microsatellite profile of muscle-invasive UCC by topographic compartments. The genetic pattern was statistically different in UCC with dysplasia compared with UCC without dysplasia. Microsatellite abnormalities were more frequent in superficial compartments (RB1, P b .001; NF1, P = .004) and deep compartments (RB1, P = .003) of UCC without dysplasia. Different patterns were noted in the superficial and deep compartments from a given patient in 5 patients with urothelial dysplasia (20%). This topographic genetic heterogeneity was due to the presence of additional abnormalities in superficial compartments (NF1, 1 patient) or deep compartments (TP53, 2 patients; RB1, 1 patient) or discordant pattern (1 patient). Frequency of TSG microsatellite abnormalities in superficial and deep compartments of muscle-invasive UCC. Abbreviations: ROH, retention of heterozygosity (green cell); LOH/SNP, loss of heterozygosity/single nucleotide polymorphism (red cell); NI, noninformative (gray cell); TP53, tumor protein p53; RB1, retinoblastoma; WT1, Wilms tumor 1; NF1, neurofibromatosis 1; S, superficial; D, deep.

Nodular-trabecular UCCs were more frequently aneuploid (27/28; 96%) and high grade (26/28; 93%) than infiltrative UCCs (9/16, 56%, and 12/16, 75%, respectively). Nodulartrabecular UCCs and superficial compartments showed significantly higher values for both mitotic figure counting and proliferation rate (Table 2). The number of diploid (6 cases) and low-grade (5 cases) UCCs precluded any statistical comparisons of these features. Nodular-trabecular UCCs revealed more abnormal loci than infiltrative UCCs (Fig. 2; P = .0001). Discordant genetic patterns by tumor compartments were observed in only 2 infiltrative UCCs precluding any statistical assessment, but all showed nuclear TP53 expression and more LOH/single nucleotide polymorphisms (SNPs) in the deep compartment (WT1 LOH/SNP in 1 and NF1 LOH/SNP in 1). Immunostaining for mlh1/msh2 revealed statistically significant reduction of at least one of the proteins (especially mlh1) in CIS and the deep compartment of trabecular UCC

with 2 or more TSG microsatellite abnormalities (Fig. 3). No significant difference was observed in the mlh1/msh2 immunoexpression in UCC with less than 2 TSG genetic abnormalities (MS stable or MS instable-low) but revealed deficient MMR system at the deep compartment. Normal MLH1/MSH2 exons sequences were observed in all UCC analyzed, regardless of immunoexpression and microsatellite status (Fig. 3).

4. Discussion Microsatellite analysis of TSG supports that coexistent CIS and muscle-invasive UCC evolves independently, contributing to intratumoral heterogeneity, despite having a common progenitor (monoclonal proliferation). The somatic MMR protein down-regulation contributes to the

MMR and Microsatellites in VCC

995

Table 2 Kinetic features of muscle-invasive UCC by infiltration patterns (nodular-trabecular versus infiltrative) and tumor compartments (superficial versus deep)

MF counting Proliferation rate

Nodular-trabecular pT2a/b UCC

Infiltrative pT2a/b UCC

Superficial

Deep

Superficial

Deep

9.0 ± 5.1 33.74 ± 7.60

4.1 ± 3.1 19.69 ± 6.15

5.9 ± 3.5 26.46 ± 8.36

2.0 ± 1.8 14.57 ± 4.58

accumulation of genetic alterations, which suggests an independent evolution rather than a precancerous lesion/ early neoplasm. LGUD is mainly a polyclonal lesion revealing low incidence of TSG microsatellite abnormalities, suggestive of a nonneoplastic condition. A linear progression model for UCC would expect to find progressive accumulation of genetic alterations in the transition intraurothelial-superficial invasive-deep invasive components of UCC. However, this gradient is not found. CIS displays microsatellite alterations similar to UCC deep compartments [1,16,17] but with more genetic abnormalities in CIS (Figs. 2,4,5). This and the differential TSG microsatellite pattern of UCC superficial compartments (NF1-defective) [16,17] do not support the sequence

Significance

P = .012 P = .009

coexistent CIS—Nsuperficial UCC—Ndeep UCC [4]. Several intraepithelial foci show more alterations than matched invasive foci, suggesting a more extensive genetic evolution for the former and supporting multifocality and independent clonal evolution of these coexistent carcinomas [30,31]. The accumulation of genetic abnormalities in coexistent CIS is consistent with an independent progression of bladder carcinoma (Fig. 2) [32-37]. Genetic alterations centered around RB1 may represent an incipient event in bladder neoplasia. However, the inactivation of RB1 occurred later and was associated with the onset of severe dysplasia/carcinoma in situ [17]. There is also evidence for the presence of critical alternative candidate genes mapping to the 13q14 region that are involved in

Fig. 5 Muscle-invasive UCC can evolve through pathways with and without microsatellite abnormalities that correlate with the invasive growth pattern (trabecular-nodular with coexistent CIS versus infiltrative) and TSG (NF1/RB1/TP53) regulation of the G1-S transition (interstitial DNA loss versus no DNA loss).

996 clonal expansion of neoplasia within the bladder antecedent to the inactivation of the RB1 gene [38]. Finally, we performed high-resolution mapping using single nucleotide polymorphism markers within one region on chromosome 13q14, containing the model tumor suppressor gene RB1, and defined a minimal deleted region associated with clonal expansion of in situ neoplasia. These analyses provided new insights on the involvement of several noncoding sequences mapping to the region and identified novel target genes, termed forerunner (FR) genes, involved in early phases of cancer development [39]. In addition, the invasive compartment microsatellite pattern of UCC with intraurothelial lesions revealed a significant decrease of RB1 and NF1 abnormalities (Fig. 4), which correlated with a nodular-trabecular pattern and high cellular turnover [13], proving the topographic genetic heterogeneity of muscle-invasive UCC [16,17]. LGUD shows much lower incidence of genetic abnormalities, making the direct connection with the linear progression unlikely. Although a clonal relationship has been suggested by LOH and/or comparative genomic hybridization analyses, supporting the hypothesis that flat urothelial hyperplasias can display many genetic alterations commonly found in bladder cancer [40], the kinetic features of LGUD makes unlikely this progression. The combined genetic-kinetic studies are needed for a full assessment [7,16]. LGUD showed low incidence of TSG microsatellite abnormalities, no TP53 alterations, and polyclonal patterns. These results support the existence of 2 transformation pathways for bladder UCC, TP53 alterations in high-grade UCC (frequently muscle-invasive) and p16 in low-grade UCC (often pT1) [7]. Our results also disprove the clonal identity of coexistent LGUD and CIS, and their dissimilar genetic profiles and topography question the sequence LGUD 窶年co-existent CIS. The hypotheses of tumor evolution and oligoclonality as derived from LOH data need to be supported by deletion-independent clonality studies as X chromosomal inactivation analysis [20,21,41]. Multifocality and recurrence are clinically important features of urothelial carcinomas of the urinary bladder. Combination of molecular data with histopathologic bladder mapping suggested a monoclonal development of the multifocal lesions mostly via intraurothelial migration. Recent molecular genetic studies have suggested that multifocal urothelial carcinomas are monoclonally derived from an identical transformed progenitor cell [31,42]. MMR protein down-regulation, normal MLH1/MSH2 sequences, and microsatellite abnormalities characterized CIS (12 cases, P = .0053) and nodular-trabecular UCC. Microsatellite profiles of bladder UCC have shown infrequent instability [9-11], which can be an independent prognostic marker for assessing risk of recurrence in superficial tumors irrespective of the grade [43]. Likewise, reduced expression of the MMR proteins may have an important contribution in the development of a subset of UCC and is a potentially useful prognostic marker [12,13],

A. Blanes et al. in particular for upper urinary tract tumors [44]. However, the pattern of microsatellite instability depends on the location and elevated microsatellite alterations at select tetranucleotides being reported more frequently in bladder [44]. Our results add other type of microsatellite abnormalities (extrabands due to single nucleotide substitutions) in muscle-invasive UCC revealing solid-trabecular growth pattern and coexistent CIS. MMR gene inactivation (by either mutation or protein down-regulation) leads to mutation accumulation and molecular progression not necessarily independent from chromosomal instability [20]. In conclusion, TSG microsatellite patterns support a nonlinear and independent genetic evolution of coexistent CIS and invasive UCC (Fig. 5). The combined assessment of clonal identity (LOH of tumor suppressor genes and Xchromosome inactivation) and molecular progression (defined as accumulation of genetic abnormalities) is comprehensive and incorporates some kinetic features that allow the persistence and transmission of genetic alterations to descendant cells. Somatic MMR protein down-regulation, the accumulation of TSG microsatellite abnormalities, and monoclonal pattern suggest a divergent molecular evolution for CIS and coexistent muscleinvasive UCC (in particular those with trabecular growth pattern). UCC with infiltrative pattern should follow alternative pathways and LGUD are most likely unrelated with the mechanisms of somatic down-regulation of mismatch repair system. In contrast, LGUD shows polyclonal pattern and low incidence of TSG microsatellite abnormalities, suggestive of a nonneoplastic condition.

References [1] Rosin MP, Cairns P, Epstein JI, Schoenberg MP, Sidransky D. Partial allelotype of carcinoma in situ of the human bladder. Cancer Res 1995; 55:5213-6. [2] McKenney JK, Gomez JA, Desai S, Lee MW, Amin MB. Morphologic expressions of urothelial carcinoma in situ: a detailed evaluation of its histologic patterns with emphasis on carcinoma in situ with microinvasion. Am J Surg Pathol 2001;25:356-62. [3] Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization classification of tumours: pathology and genetics of tumours of the urinary system and male genital organs. Lyon: IARC Press; 2004. [4] Diaz-Cano SJ. General morphological and biological features of neoplasms: integration of molecular findings. Histopathology 2008;53: 1-19. [5] Lindgren D, Liedberg F, Andersson A, Chebil G, Gudjonsson S, Borg A, et al. Molecular characterization of early-stage bladder carcinomas by expression profiles, FGFR3 mutation status, and loss of 9q. Oncogene 2006;25:2685-96. [6] Eiber M, van Oers JMM, Zwarthoff EC, van der Kwast TH, Ulrich O, Helpap B, et al. Low frequency of molecular changes and tumor recurrence in inverted papillomas of the urinary tract. Am J Surg Pathol 2007;31:938-46. [7] Baithun SI, Naase M, Blanes A, Diaz-Cano SJ. Molecular and kinetic features of transitional cell carcinomas of the bladder: biological and clinical implications. Virchows Arch 2001;438:289-97. [8] Tuziak T, Jeong J, Majewski T, Kim MS, Steinberg J, Wang Z, et al. High-resolution whole-organ mapping with SNPs and its significance to early events of carcinogenesis. Lab Invest 2005;85:689-701.

MMR and Microsatellites in VCC [9] Bonnal C, Ravery V, Toublanc M, Bertrand G, Boccon-Gibod L, Henin D, et al. Absence of microsatellite instability in transitional cell carcinoma of the bladder. Urology 2000;55:287-91. [10] Ericson KM, Isinger AP, Isfoss BL, Nilbert MC. Low frequency of defective mismatch repair in a population-based series of upper urothelial carcinoma. BMC Cancer 2005;5:23. [11] Mongiat-Artus P, Miquel C, van der Aa M, Buhard O, Hamelin R, Bangma C, et al. Infrequent microsatellite instability in urothelial cell carcinoma of the bladder in young patients. Eur Urol 2006;49:685-90. [12] Catto JW, Xinarianos G, Burton JL, Meuth M, Hamdy FC. Differential expression of hMLH1 and hMSH2 is related to bladder cancer grade, stage and prognosis but not microsatellite instability. Int J Cancer 2003;105:484-90. [13] Rubio J, Blanes A, Sanchez-Carrillo JJ, Diaz-Cano SJ. Microsatellite abnormalities and somatic down-regulation of mismatch repair characterize nodular-trabecular muscle-invasive urothelial carcinoma of the bladder. Histopathology 2007;51:458-67. [14] Diaz-Cano SJ. Are PCR artifacts in microdissected samples preventable? HUM PATHOL 2001;32:1415. [15] Diaz-Cano SJ, Brady SP. DNA extraction from formalin-fixed, paraffin-embedded tissues: protein digestion as a limiting step for retrieval of high-quality DNA. Diagn Mol Pathol 1997;6:342-6. [16] Blanes A, Rubio J, Martinez A, Wolfe HJ, Diaz-Cano SJ. Kinetic profiles by topographic compartments in muscle-invasive transitional cell carcinomas of the bladder: role of TP53 and NF1 genes. Am J Clin Pathol 2002;118:93-100. [17] Diaz-Cano SJ, Blanes A, Rubio J, Matilla A, Wolfe HJ. Molecular evolution and intratumor heterogeneity by topographic compartments in muscle-invasive transitional cell carcinoma of the urinary bladder. Lab Invest 2000;80:279-89. [18] Ro JY, Ayala AG, el-Naggar A. Muscularis mucosa of urinary bladder. Importance for staging and treatment. Am J Surg Pathol 1987;11: 668-73. [19] Younes M, Sussman J, True LD. The usefulness of the level of the muscularis mucosae in the staging of invasive transitional cell carcinoma of the urinary bladder. Cancer 1990;66:543-8. [20] Diaz-Cano SJ, Blanes A, Wolfe HJ. PCR techniques for clonality assays. Diagn Mol Pathol 2001;10:24-33. [21] Diaz-Cano SJ. Designing a molecular analysis of clonality in tumours. J Pathol 2000;191:343-4. [22] Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ. Clonality as expression of distinctive cell kinetics patterns in nodular hyperplasias and adenomas of the adrenal cortex. Am J Pathol 2000;156:311-9. [23] Blanes A, Sanchez-Carrillo JJ, Diaz-Cano SJ. Topographic molecular profile of pheochromocytomas: role of somatic down-regulation of mismatch repair. J Clin Endocrinol Metab 2006;91:1150-8. [24] Diaz-Cano SJ. Clonality studies in the analysis of adrenal medullary proliferations: application principles and limitations. Endocr Pathol 1998;9:301-16. [25] Diaz-Cano SJ, de Miguel M, Blanes A, Tashjian R, Galera H, Wolfe HJ. Clonal patterns in phaeochromocytomas and MEN-2A adrenal medullary hyperplasias: histological and kinetic correlates. J Pathol 2000;192:221-8. [26] Bibbo M, Bartels PH, Dytch HE, Wied GL. Cell image analysis. In: Bibbo M, editor. Comprehensive cytopathology. Philadelphia: Saunders Co; 1991. p. 965-83. [27] Dressler LG. Controls, standards, and histogram interpretation in DNA flow cytometry. Methods Cell Biol 1990;33:157-71.

997 [28] Sherwood SW, Schimke RT. Cell cycle analysis of apoptosis using flow cytometry. Methods Cell Biol 1995;46:77-97. [29] Jimenez RE, Gheiler E, Oskanian P, Tiguert R, Sakr W, Wood Jr DP, et al. Grading the invasive component of urothelial carcinoma of the bladder and its relationship with progression-free survival. Am J Surg Pathol 2000;24:980-7. [30] Koch M, de Miguel M, Hofler H, Diaz-Cano SJ. Kinetic profiles of intraepithelial and invasive prostatic neoplasias: the key role of downregulated apoptosis in tumor progression. Virchows Arch 2000;436: 413-20. [31] Denzinger S, Mohren K, Knuechel R, Wild PJ, Burger M, Wieland WF, et al. Improved clonality analysis of multifocal bladder tumors by combination of histopathologic organ mapping, loss of heterozygosity, fluorescence in situ hybridization, and p53 analyses. 2006;37:143-51. [32] Cairns P, Proctor AJ, Knowles MA. Loss of heterozygosity at the RB locus is frequent and correlates with muscle invasion in bladder carcinoma. Oncogene 1991;6:2305-9. [33] Cappellen D, Gil Diez de Medina S, Chopin D, Thiery JP, Radvanyi F. Frequent loss of heterozygosity on chromosome 10q in muscleinvasive transitional cell carcinomas of the bladder. Oncogene 1997; 14:3059-66. [34] Habuchi T, Devlin J, Elder PA, Knowles MA. Detailed deletion mapping of chromosome 9q in bladder cancer: evidence for two tumour suppressor loci. Oncogene 1995;11:1671-4. [35] Habuchi T, Ogawa O, Kakehi Y, Ogura K, Koshiba M, Hamazaki S, et al. Accumulated allelic losses in the development of invasive urothelial cancer. Int J Cancer 1993;53:579-84. [36] Habuchi T, Ogawa O, Kakehi Y, Ogura K, Koshiba M, Sugiyama T, et al. Allelic loss of chromosome 17p in urothelial cancer: strong association with invasive phenotype. J Urol 1992;148:1595-9. [37] Habuchi T, Yoshida O, Knowles MA. A novel candidate tumour suppressor locus at 9q32-33 in bladder cancer: localization of the candidate region within a single 840 kb YAC. Hum Mol Genet 1997;6: 913-9. [38] Kim MS, Jeong J, Majewski T, Kram A, Yoon DS, Zhang RD, et al. Evidence for alternative candidate genes near RB1 involved in clonal expansion of in situ urothelial neoplasia. Lab Invest 2006; 86:175-90. [39] Majewski T, Lee S, Jeong J, Yoon DS, Kram A, Kim MS, et al. Understanding the development of human bladder cancer by using a whole-organ genomic mapping strategy. Lab Invest 2008;88:694-721. [40] Obermann EC, Junker K, Stoehr R, Dietmaier W, Zaak D, Schubert J, et al. Frequent genetic alterations in flat urothelial hyperplasias and concomitant papillary bladder cancer as detected by CGH, LOH, and FISH analyses. J Pathol 2003;199:50-7. [41] Stoehr R, Hartmann A, Hiendlmeyer E, Murle K, Wieland W, Knuechel R. Oligoclonality of early lesions of the urothelium as determined by microdissection-supported genetic analysis. Pathobiology 2000;68:165-72. [42] Hartmann A, Rosner U, Schlake G, Dietmaier W, Zaak D, Hofstaedter F, et al. Clonality and genetic divergence in multifocal low-grade superficial urothelial carcinoma as determined by chromosome 9 and p53 deletion analysis. Lab Invest 2000;80:709-18. [43] Vaish M, Mandhani A, Mittal RD, Mittal B. Microsatellite instability as prognostic marker in bladder tumors: a clinical significance. BMC Urol 2005;5:2. [44] Catto JW, Azzouzi AR, Amira N, Rehman I, Feeley KM, Cross SS, et al. Distinct patterns of microsatellite instability are seen in tumours of the urinary tract. Oncogene 2003;22:8699-706.