FDA Modernization Act 2.0: What It Actually Changed About Animal Testing

The single most consequential thing to understand about the FDA Modernization Act 2.0 is that it removed a statutory floor, not a regulatory expectation. It deleted words from a law that had required animal tests. It did not tell FDA reviewers to stop asking for animal data, it did not amend the regulation that governs what actually goes into an investigational new drug (IND) application, and it created no right to have any particular non-animal method accepted. Programmes that read the headline — “FDA no longer requires animal testing” — and planned a first-in-human submission around that reading were planning against a statute that was never the binding constraint.

What the Act did was real, and it mattered. It just mattered in a narrower and more procedural way than the coverage suggested, and the four years since have been about FDA slowly building the machinery — guidances, validation expectations, product-class carve-outs — that determines whether a non-animal method will actually be accepted in your submission.

What the Act is, in citation terms

The FDA Modernization Act 2.0 is not a freestanding statute you can look up by that name. It was enacted in December 2022 as Section 3209 of the Consolidated Appropriations Act, 2023 (Pub. L. 117-328). FDA itself describes the operative change in these terms in a 2026 Federal Register notice: “Section 3209(a) of the Consolidated Appropriations Act, 2023 amended section 505(i) of the FD&C Act to clarify that nonclinical tests can be used to support proposed clinical testing.”

Note FDA’s verb: clarify. That framing is the agency’s own, and it is the tell for how the change has been implemented.

The two textual edits

1. The IND authority in the FD&C Act. Section 505(i)(1)(A) of the Federal Food, Drug, and Cosmetic Act — codified at 21 U.S.C. 355(i)(1)(A) — had for decades directed FDA to write regulations exempting drugs from the usual requirements for shipment in interstate commerce for investigational use, conditioned on the sponsor submitting reports of “preclinical tests (including tests on animals)”. The Act struck that parenthetical. The current codified text requires the submission, before clinical testing begins, of “reports … of nonclinical tests of such drug adequate to justify the proposed clinical testing”.

Two words in that sentence do the work, and neither is the one people notice. The first is nonclinical, which replaced preclinical and is method-neutral: an organ chip, a validated computational model and a rat study are all nonclinical tests. The second is “adequate to justify”, which was already there, was not touched, and is where FDA’s discretion lives. Adequacy is a scientific judgement made by a review division against your specific product and your specific proposed clinical protocol. Removing the parenthetical widened the menu of things that can be adequate. It did not lower the adequacy bar, and it did not transfer the adequacy judgement to the sponsor.

2. The biosimilar pathway in the Public Health Service Act. The Act made a parallel edit to Section 351(k) of the PHS Act (42 U.S.C. 262), which sets out what a biosimilar application must contain. The clause that had required “animal studies (including the assessment of toxicity)” was amended so that the word animal no longer appears; the provision now calls for an assessment of toxicity without prescribing the species or system it must be performed in.

What the Act conspicuously did not do

Four things, and each has caught someone out.

• It is permissive, not prohibitive. Nothing in Section 3209 bans, sunsets or phases out animal testing. A 2023 review in Trends in Biotechnology puts it plainly: the statute “does not eliminate animal testing, it clarifies that various nonclinical test methods … as well as computational models, can be used prior to and during the clinical phase”. Animal tests remain a fully valid category of nonclinical test.
• It did not amend FDA’s regulations. This is the gap that catches practitioners, and it is still open. The IND content regulation, 21 CFR 312.23(a)(8), continues to require “adequate information about pharmacological and toxicological studies of the drug involving laboratory animals or in vitro”. Subparagraph (i) asks for the pharmacological effects and mechanism of action “in animals”. Subparagraph (ii) asks for “an integrated summary of the toxicological effects of the drug in animals and in vitro”, including acute, subacute and chronic toxicity testing as appropriate to the drug and the phase. Subparagraph (iii) requires a statement that each nonclinical laboratory study complied with the good laboratory practice regulations in 21 CFR Part 58, or an explanation of why not. The statute above these regulations changed in 2022; the regulations did not.
• It created no obligation on FDA to accept a given method. The Act enlarged the category of admissible evidence. It said nothing about qualification, validation or fitness for purpose — the questions a reviewer will actually ask.
• It changed nothing about animal welfare oversight. The Animal Welfare Act, the PHS Policy, and your institution’s IACUC review obligations are untouched by Section 3209. For animals you do still use, every existing protocol, housing, and reporting duty applies exactly as before.

The regulation gap and the 3.0 bills

The mismatch between a modernised statute and an unmodernised regulation is the explicit motivation for the follow-on legislation. Companion bills styled the FDA Modernization Act 3.0 — S.355 and H.R.2821 in the 119th Congress — were introduced in 2025 to require FDA to update its IND regulations so the regulatory text harmonises with the statutory terminology. Check the current status of both bills on congress.gov before relying on them; introduction is not enactment, and as at the time of writing the 21 CFR 312.23 text quoted above is what an IND is measured against.

Practically, this means the regulation is not the thing that will stop you. FDA has consistently read 312.23 as permitting an appropriately justified package. But you cannot cite Section 3209 as an entitlement, because the operative regulation still speaks the older language.

What FDA has actually done since 2022

The important developments are administrative, not legislative, and they have accelerated markedly since 2025.

The 2025 phase-out announcement

On 10 April 2025 FDA announced a plan to phase out the animal-testing requirement for monoclonal antibodies and other drugs, published alongside a Roadmap to Reducing Animal Testing in Preclinical Safety Studies. The stated ambition, as characterised in a 2026 analysis in Trends in Biotechnology, is to make animal studies “the exception rather than the norm” for preclinical safety and toxicity testing within three to five years, with monoclonal antibodies as the first product class and inclusion of NAM data “encouraged” in the interim rather than mandated.

Monoclonal antibodies: streamlined nonclinical safety studies (December 2025)

FDA published a draft guidance under Docket No. FDA-2025-D-4634 on 3 December 2025 (comment period closed 2 February 2026). It describes when general toxicology studies for monospecific antibodies are not warranted, or may be limited to a short-term study, and where a three-month or longer animal toxicology study may not be needed. The framing is a weight-of-evidence risk assessment, which may draw on new approach methodologies. FDA’s stated aim is to avoid unnecessary use of animals, particularly non-human primates — a target chosen deliberately, because most antibodies are pharmacologically active in NHPs alone, which is precisely why NHPs have historically been used.

This is the single most concrete change for a sponsor. Note its scope: one product class, addressed through the ordinary guidance process, not a general licence.

General considerations for NAMs (March 2026)

FDA published a broader draft guidance, General Considerations for the Use of New Approach Methodologies in Drug Development, under Docket No. FDA-2025-D-6131 on 19 March 2026 (comments closed 18 May 2026). It sets out validation considerations focused on context of use, human biological relevance, technical characterisation, and demonstrating that a method is fit for purpose. FDA frames the goal as facilitating “validated NAMs that accurately predict risks to human health, while also decreasing usage of animal testing”.

Read that ordering carefully. Validation comes first and reduction follows from it. This is the document that tells you what the real gate is, and it is not the statute.

What this means if you are planning a nonclinical package

• Context of use is the unit of acceptance, not the technology. Nobody qualifies “organ chips”. A method is accepted for a defined purpose — a specific endpoint, in a specific product class, at a specific decision point. The same chip can be fit for purpose for one question and not for another.
• Replacement is easier at the margin than at the core. The realistic near-term wins are reducing or shortening a study, replacing a second species, or narrowing an endpoint, rather than eliminating in vivo work outright. The monoclonal antibody guidance is written in exactly those terms.
• Computational evidence is already an established route. Model-informed approaches have a longer regulatory track record than most benchtop NAMs. If your alternative is a model rather than a device, physiologically based pharmacokinetic modelling for regulatory submissions is the more useful precedent to study.
• Engage the review division before you commit. Adequacy is a division-level judgement. A pre-IND meeting where you put the proposed package in front of the reviewers is worth more than any reading of Section 3209.
• GLP still applies where it applies. Removing the animal parenthetical did not remove Part 58 obligations or the 312.23(a)(8)(iii) compliance statement.
• Do not promise it in a grant or a protocol you cannot deliver. Writing “the FDA Modernization Act 2.0 permits us to omit animal studies” into a funding application or an IACUC protocol overstates the law. Describe the specific method, its context of use, and its validation status instead.

Where the disagreement actually is

It would be misleading to present the direction of travel as uncontested. The 2026 Trends in Biotechnology analysis by Gerke, Balamut and Wagner argues the opposite of the popular reading — that FDA “should continue insisting on animal testing data to support the investigational use of new drugs” until specific NAMs demonstrate reliability for particular contexts. That is a stated position rather than a regulatory requirement, but it reflects a genuine split in the regulatory science community about how fast validation evidence is accumulating relative to policy ambition. If you are building a scientific justification, expect it to be read by someone who holds one of these views strongly.

Frequently asked questions

Did the FDA Modernization Act 2.0 ban animal testing?

No. It removed language that required animal tests; it added nothing that prohibits them. Animal studies remain a valid nonclinical test under the amended statute, remain contemplated by 21 CFR 312.23(a)(8), and remain the default for most product classes.

Does FDA still require animal data for an IND?

FDA requires nonclinical evidence “adequate to justify the proposed clinical testing”. In practice, for most products, that expectation is still met with animal data, because that is where the validated precedent sits. Whether a non-animal package is adequate for your product is a case-by-case judgement by the review division, and the answer varies by product class, endpoint and phase.

Which law number is the FDA Modernization Act 2.0?

It was enacted as Section 3209 of the Consolidated Appropriations Act, 2023, Pub. L. 117-328, in December 2022. It amended Section 505(i) of the FD&C Act (21 U.S.C. 355(i)) and Section 351(k) of the PHS Act (42 U.S.C. 262(k)).

What is the difference between “preclinical” and “nonclinical”?

In this statutory context, nonclinical is the deliberately method-neutral term that replaced preclinical tests (including tests on animals). Nonclinical simply means not conducted in human subjects — it carries no implication about species or platform, which was the point of the substitution.

What counts as a new approach methodology?

The term covers cell-based assays and human induced pluripotent stem cell systems, organoids, organ-on-a-chip and microphysiological systems, and computational and AI-based models. See new approach methodologies (NAMs). Which of these is acceptable for a given submission is decided by context of use and validation status, not by the category label.

Does this change anything for my IACUC or animal facility?

No. Section 3209 amended drug-approval statutes, not animal welfare law. Animal Welfare Act coverage, PHS Policy assurances, IACUC review and reporting obligations are unchanged for any animal work you continue to conduct.

Has FDA Modernization Act 3.0 passed?

S.355 and H.R.2821 were introduced in the 119th Congress in 2025 to require FDA to update its IND regulations to match the amended statutory language. Verify current status on congress.gov; introduction is not enactment.

Can I cite the Act to justify dropping a species from my tox package?

Not on its own. Cite the scientific justification, the validation evidence for the method you are substituting, and where applicable the relevant FDA draft guidance for your product class. The Act establishes that a non-animal test is legally admissible; it does not establish that yours is adequate.

References

• 21 U.S.C. 355 (Federal Food, Drug, and Cosmetic Act § 505), 2023 edition — govinfo.gov
• 42 U.S.C. 262 (Public Health Service Act § 351), 2023 edition, including 2022 amendment notes — govinfo.gov
• 21 CFR 312.23, IND content and format — eCFR
• FDA, General Considerations for the Use of New Approach Methodologies in Drug Development; Draft Guidance for Industry, Docket No. FDA-2025-D-6131, 19 March 2026 — Federal Register
• FDA, Monoclonal Antibodies: Streamlined Nonclinical Safety Studies; Draft Guidance for Industry, Docket No. FDA-2025-D-4634, 3 December 2025 — Federal Register
• Gerke S, Balamut J, Wagner JK. The FDA’s plan to phase out animal testing. Trends Biotechnol. 2026;44(8):2178–2181 — PMC12834477
• Yin DE, et al. 3D human tissue models and microphysiological systems for HIV and related comorbidities. Trends Biotechnol. 2024;42(5):526–543 — PMC11065605
• FDA Modernization Act 2.0: transitioning beyond animal models with human cells, organoids, and AI/ML-based approaches. J Clin Invest. 2023 — PMC10617761