Comment from Nexus Concordat Inc

Nexus Concordat Inc. submits this comment on Docket FDA-2026-N-4390 in support of the AI-Enabled Optimization of Early-Phase Clinical Trials Pilot Program. I am the inventor on four U.S. provisional patent applications relevant to the architectural questions this RFI raises (Applications 63/939,190; 63/962,385; 63/988,485; 64/034,536). My background is biochemistry, molecular biology, and clinical laboratory practice subject to CLIA standards. This comment makes two proposals to FDA: (1) Architecturally enforced Pre-determined Change Control Plans. FDA's existing PCCP framework for AI/ML-enabled medical devices enforces change boundaries procedurally. A third architectural category is technically feasible: substrate-grounded AI systems where the weights and opcode store update post-deployment only through a compiler-enforced verification gate with version-traceable lineage. The pilot is well-positioned to evaluate the distinction. (2) Mechanistic in-silico trial emulation as a third evidentiary substrate. Randomized controlled trials and claims-based real-world evidence have known structural limitations in therapeutic areas where claims data does not capture the relevant signal. Mechanistic in-silico trial emulation grounded in published pharmacology can complement both, filling structural gaps without replacing either. The substantive comment, including detailed responses to the questions enumerated in Section II.A and Section II.B of the RFI, is filed as the attached document. The methods paper supporting the technical claims is filed as Exhibit A. I respectfully request that FDA consider these proposals in shaping the pilot's evaluation framework, particularly under Section II.B.4 (AI System Performance), Section II.B.5 (Trustworthiness aligned with NIST AI RMF), and Section II.B.6 (Comparative Evaluation). Marjorie Gayle McCubbins, BSc Founder and Chief Executive Officer Nexus Concordat Inc.