AI-Enabled Optimization of Early- Phase Clinical Trials Pilot Program; Request for Information

Panoptive, a clinical trial oversight technology company, supports the FDA's pilot program for AI-enabled clinical trials. They argue that the pilot should focus on the "full oversight loop"—measuring how AI-surfaced signals are converted into accountable human decisions and inspection-ready records—rather than just signal detection speed. They provide specific recommendations on metrics, participant selection, and the need for human-in-the-loop governance to ensure reliability and regulatory compliance.

The Sylvester Comprehensive Cancer Center, an NCI-designated cancer center, supports the proposed AI-enabled pilot program to improve clinical trial operations, safety monitoring, and data quality. They advocate for a focus on high-risk therapeutic areas like oncology, the inclusion of robust data infrastructure and governance requirements, and the use of pre-competitive collaboration models.

PhRMA, representing the pharmaceutical research and manufacturing industry, supports the FDA's proposed pilot program to explore AI-enabled optimization in early-phase clinical trials. They advocate for a risk-based approach to AI validation, the protection of confidential commercial information, and the use of established practices to share learnings from the pilot.

Exponential Technologies Corp. supports the FDA's proposed pilot program to evaluate AI-enabled technologies in early-phase clinical trials. The company recommends prioritizing oncology and first-in-human studies and suggests specific metrics for measuring trial efficiency, decision quality, and participant safety.

Medrio, a software and services vendor for clinical trial technology, supports the FDA's pilot program for AI-enabled optimization of early-phase clinical trials. They advocate for including technology vendors as formal collaborators, establishing clear regulatory guardrails for data privacy and signal response, and ensuring that AI remains a tool to inform human clinical judgment rather than replace it.

The commenter supports the FDA's pilot program for AI in early-phase clinical trials and emphasizes the need for continuous monitoring of AI models throughout their lifecycle. They also recommend prioritizing reproducibility, measuring both operational and clinical outcomes, and ensuring the pilot includes organizations of varying sizes to promote scalability.

Anthropic, an AI safety and research company, supports the FDA's pilot program but recommends a framework that distinguishes between the obligations of the AI tool sponsor and the developer of the underlying general-purpose model. They argue for architecture-neutral technology selection, structured provenance documentation for training data, and a managed lifecycle for model versioning rather than a general requirement to freeze model versions during a trial.

The Duke-Margolis Institute for Health Policy supports the FDA's pilot program to use AI for optimizing early-phase clinical trials, praising the agency's commitment to advancing evidence generation. They provide constructive feedback on prioritizing generalizable results, managing the compressed timeline, and balancing AI-driven speed with rigorous risk-benefit frameworks.

IQVIA, a global provider of clinical research services and advanced analytics, expresses strong support for the FDA's proposed pilot program to evaluate AI-enabled technologies in early-phase clinical trials. The company argues that the pilot should focus on high-decision-density settings, prioritize critical foundational capabilities like safety monitoring and dose-escalation support, and establish a structured, risk-based pathway to generate durable regulatory guidance.

McKesson Corporation supports the FDA's initiative to explore AI-enabled optimization of early-phase clinical trials and commends the agency's focus on addressing bottlenecks in the research ecosystem. The company recommends a risk-based governance framework, a focus on data quality, and the inclusion of operational feasibility and equity metrics in the pilot's evaluation.