IMPALA Consortium's Role in Advancing Biopharmaceutical Quality through Data Analytics
Inter-Company quality Analytics Industry (IMPALA) Consortium
Patrick O’Sullivan, Global Quality Medicine Learning Manager, Boehringer Ingelheim
Michael Pelosi, Director Quality Analytics, Astellas
RQA Quality Conversation
27Aug2024
Overview of the Inter-Company quality Analytics Industry group (IMPALA) and its impact on modernizing GCP/PV Quality
Version 1.18 - July 2024 Approved for external use
The Inter coMPany quALity Analytics (IMPALA)
The Inter coMPany quALity Analytics (IMPALA)
Vision
IMPALA aims to engage with Health Authorities inspectors on defining guiding principles for the use of advanced analytics to complement, enhance and accelerate current QA practices
The IMPALA ecosystem (industry, regulators, patients) will contribute to a change in paradigm for QA, i.e., where co-developed advanced analytics and best practices can help detect and mitigate issues faster, reduce the burden of retrospective, time-consuming traditional QA activities and ultimately accelerate approval and patient access to innovative drugs
“IMPALA’s mission is to transform the Biopharmaceutical Clinical Quality Assurance process in the Good Clinical Practice and Good Pharmacovigilance Practice areas, by using advanced analytics and promoting the adoption of this new approach and associated methodologies by key industry stakeholders (e.g. pharma quality professionals, Health Authorities) to assure safe use therefore building patient trust, accelerating approvals and ultimately benefiting patients globally.
The IMPALA Consortium will provide the strategic focus for working across the Biopharmaceutical ecosystem to develop and gain industry-wide consensus for the adoption of improved QA using advanced analytics and best practices to be used across the industry.
To achieve these objectives, key projects have been established and are ongoing in focused Work Products.”
The Inter coMPany quALity Analytics (IMPALA)
Members
Group started on Jul-2019
Established as a non-for-profit consortium on Oct2022
In scope
Knowledge sharing and best practices for GCP/PV quality and analytics
Joint Health Authorities engagement for GCP/PV quality 2.0
Co-development of open source tools: analytics packages, templates, methodologies, etc.
Analytics for GCP/PV quality; quality for AI/ML (e.g. validation); quality briefs; etc.
https://impala-consortium.org/work-products/
Ongoing work products
Audit Site Risk Selection (methodology; open-source package)
Anomaly detection clinical data (methodology; open-source package, templates)
Quality Briefs (methodology; standard templates)
Education (framework, maturity assessment)
Service Provider Analytics Interest Group (methodology)
AI Powered Regulatory Intelligence (methodology)
interest groups
Service Provider Oversight and Analytics
RAPID Audit Methodology
RAPID Product Information Audit
https://impala-consortium.org/work-products/
19 Members, 9 Events, 5 publications, 4 Completed work products
Completed work products and publications
Data Analytics for Quality Assurance in Pharmaceutical Development Framework
Clinical Trial Anomalies open-source package
Clinical Safety Reporting open-source package
IMPALA Podcast Series (available on Spotify)
Publications:
Therapeutic Innovation and Regulatory Science publication –“An Open-Source R Package for
Detection of Adverse Events Under-Reporting in Clinical Trials: Implementation and Validation”
Publication in IEEE Spectrum “Big Data Could Accelerate Drug Approval”
IMPALA white paper in CPT Pharmacometrics & System Pharmacology
Publication in QUASAR “Launch of the IMPALA consortium”
CPT: Pharmacometrics and Systems Pharmacology article –“Cross-company collaboration to leverage analytics for clinical quality and accelerate drug development: The IMPALA industry group”
Achievements to date
https://impala-consortium.org/work-products/
Presentations/Webinars
2024 DIA China Sessions - Applying Data Analytics in Audit Planning and Assessment, Quality
Briefs – Supporting Transparency, Innovation & Products to patients faster
DIA Global 2024 Sessions
IMPALA Webinar (2024) - Identifying Anomalies in Clinical Trial Data with CTAS
Clinical Data Analytics Virtual Event (2024)
Engagement sessions at DIA Global 2023
IMPALA Webinar (2023) - Embedding an Open-Source Analytics Package into Analytics-Based Audits
RQA Session (2023) on “Embedding data analytics capabilities within R&D Quality”
RQA Workshop on Data Analytics Write-Up (2022) RQA Coffee Sessions
APEC Roundtable Dialogue (2022)“Leveraging analytics for remote clinical quality oversight”
Data Analytics for Quality Assurance
In Pharmaceutical Development
First steps to fine tuning
IMPALA Structure & Education Work Product Team
ISTO-IEEE
Board of Directors
Strategic Steering Committee
Education WPT
Work Product Team Work Product Team
Work Product Team
• Supports IMPALA’s mission by through development of a framework the framework, including skills, capabilities and requirements needed to develop an analytics domain within biopharma quality organizations
• Sharing of lessons learned and best practices through papers, conferences, and the IMPALA Podcast
• IMPALA Podcast Series | Podcast on Spotify
Getting Started
6
5 Rules and regulations & Potential barriers
Skill Sets Needed
Basic
• Regulatory Knowledge
• Data Analytics Basics
Hard Skills
Intermediate
• Data Visualization:
• Statistical Analysis
• Pharmaceutical Domain Expertise
• Quality Assurance Domain Expertise
• Communication
• Ethical Considerations
• Influencing
• Critical Thinking
• Collaboration
• Adaptability
• Problem-Solving
• Attention to Detail
• Data Manipulation
• Database Management
• Data Storytelling Advanced
• Machine Learning and Predictive Analytics
• Coding and Programming
Interest in data and statistical analysis
Curiosity about the potential of data analysis and the application of new tools and methods
Data AnalyticsProfessional
QA
• Display a genuine interest in exploring data
• Display a genuine interest in exploring data and a willingness to dig into the “why” behind trends and patterns
• “Storyteller” attitude
• Growth Mindset / Continuous Learning
• Persistence – Data Analysis
• Problem-Solving Orientation
• Adaptability and Agility
Assessing Analytics Maturity
Basic
Reactionary Analysis
Operational Strategic Transformational
Immature Data Infrastructure Opportunistic
Formalized Data Availability
Systematic Strategy and Capabilities
Underdeveloped QAAnalytics Strategy
Access to Internal Data Only Differentiating
Transformational
Cross-functional Data Analytics Collaboration
Executive Leadership Support
Data Informs QA Strategy Robust Data Infrastructure
Areas of Application
Assurance to Senior Leaders Leading Indicators
Improving the Business
Lagging Indicators
Detecting Errors
Identifying and Controlling Risk
Procedural Framework
QA Operations
Quality of QAActivities
Determine Required Analytics (Use Cases)
First, seek the following information:
• What are your department goals or strategic objectives? List the actions and decisions needed to achieve the strategic goals
• What are the key business questions that will drive the department goals or strategic objectives?
• What are the problems that stakeholders are experiencing
• and do these lend themselves to being solved with data?
• What business questions aren't you asking?
• What are your business processes (where do data capabilities fit in)?
Then, probe further
•How will the data / report /analytics request achieve your strategic objectives?
• If we give you data /report/analytics, what decisions or actions will you take to achieve your goals?
• Who is the target audience and their depth of knowledge of the data and analytics?
• What do you and your target audience need to be able to understand and do with the information from the analytics?
• What do you think you or your target audience would do differently than what you /they do now?
• If you didn’t have analytics, what different actions would you /your target audience do?
Don't drive data as a question. Drive the business questions and resulting actions
Prioritizing Use Cases
Business Value
(Risk to Submission /Approvability, Patient Safety, Data Integrity, Impact to the Business)
(data/complexity/time) How feasible is the use case considering data, complexity and cost?
Identifying Data that Needs to be Analysed
Data obtained should be based on a use case
Method
Data location, availability and accessibility
Data quality and data integrity Types of data needed for use case
Must Have – essential data elements (non-negotiable)
Should Have – high priority data elements; included whenever possible
Could Have – desired data elements; may be left out due to resource constraints
Won’t Have – least critical data elements; will be left out
Data Analysis Tools
Business Intelligence Tools
Computing and Programming
Office Suite
Consider the organizati on’s IT Infrastruct ure
Identify the goals of your QA analytics strategy
Data Storytelling
Combining data analytics with memorable and cogent storytelling is a key component of conveying meaning and adding impact to presentations, reports, meetings and workshops. Stories are memorable to people and therefore teaching the art of storytelling with data is a vital skill to learn and implement when implementing a data analytics strategy.
Learning Aims and Concepts
One can build a “story” with data education program by focusing on learning the following concepts:
• Understanding data analytics basics
• Understanding your existing data platforms and analysis tools (e.g., Power BI, Tableau, Python etc.)
• Identifying the key message/s you want to convey Training Methods
• Books, other documents, videos and e-Learnings, e.g., for data analytics basic knowledge
• Workshops / focus Sessions (specific data platform training)
• Assessments (learner satisfaction, knowledge and skills attained)
• Try-outs to present your “story” to your peers and receive constructive feedback
• Accolades and certificates for successful completion (also fosters cultural changes)
• Innovative and “wow effect” training methods, for example, employing Virtual Reality / Augmented Reality systems to illustrate how “digital” stories might be told in virtual environments
Lessons Learned
Data Quality is Critical
• Invest in Data Governance
Data Accessibility at Scale
• Establish Enterprise Data Lake & Data Mesh
Data Security & Privacy
• Train in Regulations & Best Practices
Resistance to Change
• Leadership Support
Interdisciplinary Collaboration
• Maintain Regular Communication & Goal Alignment
Regulatory Compliance
• Proactively Engage with Health Authorities
Resource Constraints
• Prioritize Use Cases Based on Return on Investment (ROI)
Culture and Change Management
Changing a QA organization towards data analytics is a transformation rather than an incremental change and so it also require cultural adjustments. Approaches would have similarities to any change management initiative.
Culture
• Leadership
• Communication and Training/Upskilling
• Analytics maturity
• Change Management
• Change Agents / Champions and Analytics Translators
• Stakeholder Management and Collaboration
Also, on LinkedIn - https://www.linkedin.com/company/86955237/
Get a copy of the framework Data Analytics for Quality Assurance In Pharmaceutical Development Framework