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

Wave Life Sciences, a biotechnology company, supports the proposed AI-enabled optimization of early-phase clinical trials pilot program as a way to ensure the technology is thoughtfully controlled. They provide specific recommendations on trial scope, participant selection (particularly for smaller sponsors), infrastructure requirements like secure sandboxes, and safety oversight.

Humane Intelligence, a 501(c)(3) nonprofit organization, supports the FDA's pilot program for AI-enabled optimization of early-phase clinical trials as a necessary step toward modernizing clinical systems. They argue that the FDA should move beyond general AI risk frameworks to establish prescriptive, domain-specific evaluation criteria that ensure data integrity, algorithmic rigor, and equitable representation across diverse patient populations.

The Digital Trials & Research Alliance (DTRA), a nonprofit public-private partnership, supports the FDA's AI-enabled clinical trials pilot program. They offer to serve as a neutral facilitation partner to help define use-case-specific KPIs, develop centralized testing datasets, and conduct stakeholder listening sessions to ensure the program's success.

Parexel International, a global clinical development partner, supports the proposed AI-Enabled Optimization of Early-Phase Clinical Trials Pilot Program. They advocate for including Contract Research Organizations (CROs) as direct participants, prioritizing specific use cases like safety signal detection and dose-escalation, and establishing rigorous pre-specified metrics for evaluating AI performance and decision quality.

Aditya Sanjeev, an independent researcher in clinical AI fairness, supports the pilot program by providing specific technical recommendations to improve its safety and oversight. He argues that the FDA should mandate disaggregated subgroup performance reporting, worst-case performance metrics, and multi-seed evaluation protocols to prevent "minority collapse" in AI models.

VivaLink, Inc., a medical-grade wearable biosensor and connected-health platform company, supports the FDA's proposed pilot program for AI-enabled clinical trials. They recommend prioritizing continuous safety monitoring, establishing real-time data infrastructure standards, and ensuring equitable participation through risk- and competence-based selection criteria.

The Autonomic Resilience Collective, an independent nonprofit research organization, supports the pilot program but argues that AI-enabled safety monitoring must be anchored to within-subject baselines rather than population norms. They recommend that the FDA include within-subject statistical foundations in the pilot's scope to improve the interpretability and reliability of safety signals in small early-phase cohorts.

ALS United, a nationwide partnership of nonprofit organizations, supports the FDA's proposed pilot program for AI-enabled optimization of early-phase clinical trials. They argue that while real-time data sharing can accelerate development for rapidly progressive diseases like ALS, the program must include diverse sponsors, invest in biomarker infrastructure, ensure human oversight of AI decisions, and involve patient organizations in governance.

Jennifer Shannon, MD, representing Glacis Technologies, provides specific recommendations for evaluating the trustworthiness of AI in early-phase clinical trials. The commenter argues that trustworthiness should be assessed through a verifiable record of real-time behavior at the input and output boundaries, rather than through internal system access, to protect proprietary information while ensuring safety and auditability.

The commenter, a researcher specializing in AI reliability and clinical documentation, supports the pilot program but emphasizes the need to measure the underlying data environment's integrity alongside AI performance. They argue for disaggregated reporting by site and subgroup, pre-specified fairness analyses, and the inclusion of patient groups and investigators in governance to ensure safety and equity.