Engaging with the FDA on AI in Clinical Trials: Beyond Traditional Meetings

Jeanette Towles
Jan 17
3 min read

Updated: Jul 1

The U.S. Food and Drug Administration's (FDA) draft guidance, Considerations for the Use of Artificial Intelligence to Support Regulatory Decision-Making for Drug and Biological Products, offers a

comprehensive framework for the integration of artificial intelligence (AI) in drug development. For regulatory professionals, understanding the various engagement options beyond traditional FDA meetings is crucial for effectively navigating AI applications in clinical trials.

Risk-Based Credibility Assessment Framework

The FDA’s draft guidance outlines a seven-step, risk-based credibility assessment framework designed to ensure the reliability of AI models used in regulatory decision-making. This structured approach helps sponsors tailor validation efforts based on model risk and context of use. For a detailed explanation of each step in the framework, refer to the full guidance summary here.

Lifecycle Maintenance of AI Model Credibility

Maintaining the credibility of AI models throughout their lifecycle is essential to ensure consistent performance. The FDA recommends that sponsors implement lifecycle maintenance strategies to monitor AI model performance, detect data drift, and manage updates or modifications. This includes continuous performance evaluation, retraining models as needed, and applying risk-based oversight to sustain model validity in evolving clinical contexts.

Expanding Engagement Options with the FDA

Beyond standard meeting formats (e.g., Pre-IND, End-of-Phase meetings), the FDA has outlined specialized programs designed to support the development and regulatory evaluation of AI-driven models. These programs provide sponsors with opportunities to engage in focused discussions, obtain iterative feedback, and align their AI strategies with regulatory expectations.

Center for Clinical Trial Innovation (C3TI)

Purpose: To discuss the use of AI in clinical trial designs with the Center for Drug Evaluation and Research (CDER) before formally submitting them to an IND.
Contact:CDERclinicaltrialinnovation@fda.hhs.gov

Complex Innovative Trial Design (CID) Meeting Program

Purpose: To support sponsors interested in incorporating AI into novel clinical trial designs.
Contact: CID.Meetings@fda.hhs.gov
More Info: CID Program Overview

Drug Development Tools (DDTs) and ISTAND Programs

Purpose: For the qualification of AI-based drug development tools (e.g., algorithms for patient evaluation, endpoint adjudication, clinical data analysis).
Contacts:
- Biomarker Qualification: CDER-BiomarkerQualificationProgram@fda.hhs.gov
- Clinical Outcome Assessment Qualification: COADDTQualification@fda.hhs.gov
- Animal Model Qualification: CDERAnimalModelQualification@fda.hhs.gov
- ISTAND Pilot Program: ISTAND@fda.hhs.gov

Digital Health Technologies (DHTs) Program

Purpose: To explore the feasibility of using AI-enabled digital health technologies in drug development programs.
Contact:DHTsforDrugDevelopment@hhs.fda.gov

Emerging Drug Safety Technology Program (EDSTP)

Purpose: To assess AI applications in drug safety and pharmacovigilance (PV), particularly for postmarketing activities.
Contact: AIMLforDrugDevelopment@fda.hhs.gov (Subject: “EDSTP”)

Emerging Technology Program (ETP) and Advanced Technologies Team (CATT)

Purpose: To discuss the use of AI in pharmaceutical manufacturing with CDER and CBER.
Contacts:
- CDER-regulated drugs: CDER-ETT@fda.hhs.gov
- CBER-regulated biological products: Industry.Biologics@fda.hhs.gov (Include “CATT” in the subject line)

Model-Informed Drug Development (MIDD) Paired Meeting Program

Purpose: To incorporate AI within model-informed drug development approaches.
Contact: MIDD@fda.hhs.gov (Subject: “MIDD Program Meeting Package for CDER/CBER”)

Real-World Evidence (RWE) Program

Purpose: To integrate AI into studies using real-world data to generate real-world evidence supporting regulatory decisions.
Contact:CDERMedicalPolicy-RealWorldEvidence@fda.hhs.gov

Strategic Recommendations for Regulatory Professionals

Early and Targeted Engagement:

Initiate discussions with the FDA through relevant programs early in the AI model development lifecycle to align on risk assessment and validation strategies.

Leverage Specialized Programs:

Choose the most appropriate engagement pathway based on the AI model’s context of use (e.g., clinical design, pharmacovigilance, manufacturing).

Iterative Feedback:

Utilize programs that allow for iterative dialogue (e.g., CID, C3TI) to refine AI models in response to FDA feedback.

Comprehensive Documentation:

Prepare detailed documentation outlining the model’s development, validation, and risk assessments to facilitate productive discussions with the FDA.

Conclusion

By proactively leveraging these specialized engagement opportunities, sponsors can streamline the integration of AI into clinical development while ensuring compliance with evolving regulatory expectations.

Interested in global AI regulations? Check out our related blog on the EU AI Act to understand how European regulations intersect with AI applications in clinical trials and drug development.