FDA Backs In Silico Shift
The FDA issued draft guidance encouraging a shift from animal testing toward AI‑ and human‑based computer models to accelerate drug and diagnostic development, creating regulatory room for in silico validation approaches. The change could affect how AI diagnostics are validated and accepted. (medicaldialogues.in)
The FDA published a draft guidance titled “General Considerations for the Use of New Approach Methodologies in Drug Development” in the Federal Register (Docket No. FDA‑2025‑D‑6131) with a notice dated March 19, 2026. (public-inspection.federalregister.gov)) The agency framed the document as a CDER-led validation framework and, in a March 18, 2026 press release, explicitly listed NAM examples including organoids, organs‑on‑chips, complex in‑vitro assays, chemical reactivity studies, and computer simulations (in‑silico). (fda.gov)) The draft organizes validation expectations around four core features—context of use, human biological relevance, technical characterization, and fit‑for‑purpose—that sponsors should address when submitting nonclinical NAMs data. (pharmtech.com)) FDA’s notice instructs interested parties to submit comments within 60 days of Federal Register publication (publication 03/19/2026), which corresponds to a comment‑deadline of May 18, 2026. (public-inspection.federalregister.gov)) Regulatory precedent for in‑silico evidence already exists: the UVA/Padova Type 1 Diabetes simulator has been accepted by FDA reviewers to support regulatory clearance activities for continuous glucose monitoring devices. (mondaq.com)) Industry and legal analyses interpret the March 2026 draft as creating explicit regulatory space for in‑silico and human‑relevant evidence that could be incorporated into submissions beyond traditional animal data, with commentators highlighting potential cross‑cutting implications for AI‑enabled diagnostics and SaMD evaluation pathways. (jdsupra.com))
