A direct final rule published by the FDA on September 22, 2026 replaces "animal test" and "animal study" — along with "preclinical" and "in vitro" — with the term "nonclinical" across safety and reporting sections of its human drug and biological product regulations. The rule, available in the Federal Register, takes effect February 4, 2027, unless the agency receives significant adverse comments by December 7, 2026, in which case it will be withdrawn and rulemaking will continue through a companion proposed rule published simultaneously.
The amendments span 21 CFR Parts 312, 314, 315, 361, and 601 — the regulatory backbone governing investigational new drug applications (INDs), new drug applications (NDAs), biologics license applications (BLAs), and postmarketing reporting for both drugs and biologics. The amended regulations now codify definitions of “nonclinical test” and “nonclinical study” across key drug and biologics provisions that explicitly encompass cell-based assays, organ chips and microphysiological systems, computer modeling, bioprinting, and other human biology-based methods, with animal studies listed as one option among several rather than the default. The definition is adapted from the Food and Drug Omnibus Reform Act of 2022 (FDORA), which amended the Federal Food, Drug, and Cosmetic Act (FD&C Act) to replace "preclinical tests (including tests on animals)" with "nonclinical tests" — meaning the regulatory text now mirrors the statute. The rule adds no new requirements and, according to FDA, imposes no new costs on industry.
For drug developers, the most operationally significant changes sit in the IND framework under Part 312. Sections governing the investigator's brochure, pharmacology and toxicology summaries, IND safety reporting, and the early meeting process for life-threatening-disease programs (§312.82) have all been updated to accept nonclinical data regardless of whether it originates from animals. The safety reporting provision at §312.32(c)(1)(iii) — which requires sponsors to report findings suggesting significant human risk within 15 calendar days — now explicitly covers signals generated by computer modeling and organ chips, technologies not in common use when the provision was originally written. FDA said sponsors who already submit in silico or in chemico data under current regulations are unaffected in practice; the rule codifies what the agency describes as its longstanding interpretive position. For NDA holders, the postmarketing annual report requirement under §314.81(b)(2)(v) now captures nonclinical toxicological findings from any methodology, not just animal and in vitro studies — a change relevant to lifecycle management programs that increasingly rely on computational safety screening.
Alongside the rule, FDA launched a publicly accessible database of 25 new approach methodology (NAM) use-case examples drawn from existing FDA review materials, giving sponsors an immediate reference point for how the agency has already evaluated non-animal data in reviewed or approved submissions. Together with FDA’s March 2026 draft guidance on NAM validation and its 2025 roadmap for reducing animal testing, the database gives sponsors a growing set of regulatory reference points for incorporating alternative methods into development programs. Sponsors can also request Type D meetings with the Center for Drug Evaluation and Research (CDER) or the Center for Biologics Evaluation and Research (CBER) to discuss specific NAM applications; the CBER Advanced Technologies Team is available for novel manufacturing methods incorporating NAMs across multiple products.