FDA AI Guidance and COA Validation: What Sponsors Need to Know

Marcela Roy, MA | Bill Byrom, PhD | Helen Brooker, PhD

If your clinical outcomes assessment (COA) program uses artificial intelligence, FDA’s new guidance provides a clearer framework for establishing the credibility of AI models used to support regulatory decision-making.

The FDA’s first guidance addressing the use of AI to support regulatory decision-making for drugs and biological products has important implications for sponsors applying AI in clinical development. For programs using AI in areas such as missing-data imputation, endpoint adjudication, or sensor-based measure derivation, the guidance outlines a risk-based approach to assessing—and documenting—model credibility.

Signant Health's scientific and regulatory experts explain what the guidance may mean in practice for COA methodology, including:

• How FDA's two-dimensional risk framework applies to eCOA use cases
• Which AI applications may fall inside or outside the guidance's scope
• What "black box" AI can mean for regulatory confidence and submission planning
• How to build a credibility assessment strategy proportionate to your program's risk 
• When and how to engage FDA early when AI informs decision-making

Download the white paper to understand how to evaluate AI-enabled COA approaches, prepare an appropriate credibility evidence package, and plan for regulatory engagement.

Have questions about how the framework may apply to your program? Talk to our COA experts