Partnerships in Clinical Trials | Episode 8

This video, an episode of the SCDM Podcast series "Partnership in Clinical Trials," features MaryAnne Rizk, Managing Director, Head of Healthcare & Lifescience Clinical Strategy at Amazon Web Services. The discussion centers on how collaboration and innovation, particularly through AI and big data, are fundamentally reshaping the future of clinical research. Rizk emphasizes that the current era, marked by rapid technological advancements, evolving policies, and a unified commitment from global pharma, biotechs, technology companies, and regulatory bodies, presents an unprecedented opportunity for clinical data managers to embrace change and drive smarter, faster, and more connected clinical trials.

The conversation delves into the historical transformations in clinical trials, from Electronic Data Capture (EDC) to Risk-Based Quality Management (RBQM) and Decentralized Clinical Trials (DCTs), highlighting AI as the latest and most profound shift. Rizk predicts that 2025 will be the "year of agentic AI," where AI moves beyond conversational interfaces to become an integral part of business logic, driving go/no-go decisions in areas like protocol design, site startup, patient recruitment, and data management to accelerate regulatory submissions. This agentic layer, by automating activities and ensuring precision, accuracy, and speed, is seen as truly transformational for the industry.

A significant portion of the discussion focuses on the critical role of data. The speakers underscore the immense potential of maximizing data through standardization and unification, moving towards a "fully digitalized ecosystem" where standards link seamlessly to data and protocols. AWS's vision for this transformation includes a strong emphasis on responsible AI, ensuring ethical considerations and unbiased data, alongside the foundational step of cloud migration. Rizk outlines a layered approach to AI implementation, comprising an infrastructure layer for unified data and standards, an orchestration layer for consistent operational models and workflow determination, and a control layer for automating agents based on established workflows and insights. This framework is crucial for organizing and simplifying the explosion of data from decentralized trials, wearables, and biomarkers, enabling effective automation and human-in-the-loop decision-making.

The podcast also highlights the importance of partnerships across the entire ecosystem—patients, sites, sponsors, and regulatory authorities—to move from raw data to actionable decisions with precision, accuracy, and speed. Rizk envisions AI not as a replacement for humans but as a tool to create "superhumans," capable of making smarter decisions, shortening clinical trial cycles, and improving the body of evidence to recruit the right patient for the right trial at the right time. She advocates for a "three E's" approach to navigating change: Educate, Engage, and Evangelize, stressing the need for continuous learning, active participation in pilot programs, and sharing best practices to collectively advance the industry.

Key Takeaways:

• AI as a Catalyst for Clinical Trial Transformation: The current era is defined by technological advancements, policy changes, and cross-industry commitment, making it the most exciting time for AI to revolutionize clinical research and drug development.
• The Rise of Agentic AI: 2025 is anticipated as the "year of agentic AI," where AI agents will integrate into business logic to automate and optimize critical workflows such as protocol design, site startup, patient recruitment, and data management, accelerating regulatory submissions.
• Evolving Role of Data Professionals: The traditional role of a data manager is transforming into that of a data scientist or data orchestrator, requiring new skills to leverage advanced analytics and AI.
• Importance of Data Standardization and Unification: Unifying data in a common, standardized way (e.g., through initiatives like 360i and CDISC) is crucial for creating a fully digitalized ecosystem that can be effectively maximized by AI solutions.
• Responsible AI and Cloud Foundation: AWS emphasizes responsible AI practices, including ethical considerations and ensuring unbiased data, as well as the foundational step of migrating data and operations to the cloud for modernization and operational excellence.
• Layered Approach to AI Implementation: A structured approach involves an infrastructure layer for unified data and standards, an orchestration layer for consistent operational models and workflow determination, and a control layer for automating agents based on insights.
• Addressing Data Volume and Variety: The explosion of data from decentralized trials, wearables, digital biomarkers, and other sources necessitates robust strategies for organizing, simplifying, and standardizing data to enable automation and human-in-the-loop processes.
• Patient and Site-Centricity: Technology should unify all personas, creating intuitive, integrated, and intelligent user interfaces that empower sites and ensure trials are patient-centered, moving efficiently from data collection to impactful decisions.
• AI Augmentation for "Superhumans": AI is not expected to replace people but rather to create "superhumans" by augmenting human capabilities, leading to smarter decisions, shorter clinical trial cycles, and improved patient recruitment.
• Strategy for Navigating Change ("Three E's"): Individuals and organizations should "Educate" themselves on new policies and technologies, "Engage" in practical applications and proofs of concept, and "Evangelize" learned best practices and findings to foster broader adoption.
• Effective Change Management: Successful transformation requires a combination of "rules and tools," meaning new technologies must be accompanied by clear playbooks and an ecosystem of early adopters to test, demonstrate ROI, and mitigate risks.
• Value of Cross-Stakeholder Partnerships: Collaboration among technology providers, sponsors, regulatory authorities, sites, and patients is essential for harmonizing systems and thinking, ensuring that modernization efforts align with quality by design principles and regulatory expectations.

Tools/Resources Mentioned:

• AWS (Amazon Web Services): Mentioned as a platform providing AI solutions and cloud infrastructure for life sciences.
• CDISC (Clinical Data Interchange Standards Consortium): Implied through the discussion of data standards and unification.
• 360i (Chris Decker's work): Referenced for its focus on radically transforming how standards are conducted and unifying data models.

Key Concepts:

• Agentic AI: AI systems that can understand, reason, plan, and execute tasks autonomously, often by breaking down complex goals into smaller steps and interacting with tools and environments. Predicted to be a major force in 2025.
• Generative AI (GenAI): AI models capable of generating new content, such as text, images, or code, often in a conversational manner, making AI more accessible.
• EDC (Electronic Data Capture): A system used in clinical trials to collect data from sites electronically, replacing paper-based methods.
• RBQM (Risk-Based Quality Management): A proactive approach to clinical trial oversight that focuses on identifying, assessing, monitoring, and mitigating risks to data quality and patient safety.
• DCT (Decentralized Clinical Trials): Clinical trials conducted with remote elements, often leveraging technology like wearables, telemedicine, and digital health tools to reduce the need for in-person site visits.
• Data Orchestrator: An evolved role for data managers, focusing on managing and coordinating data flows, standards, and integration across complex systems to ensure data readiness for advanced analytics and AI.
• Human-in-the-loop: An AI approach where human intelligence is integrated into the machine learning process, allowing for human oversight, intervention, and decision-making at critical junctures.
• Quality by Design (QbD): A systematic approach to development that begins with predefined objectives and emphasizes understanding of product and process, and process control, based on sound science and quality risk management.
• Superhumans: A concept suggesting that AI will augment human capabilities, making individuals more efficient and effective rather than replacing them.