FDA’s Nonclinical Testing Rule: What It Means for CROs Offering NAMs

In September 2026, the US Food and Drug Administration (FDA) published a direct final rule titled “Nonclinical Testing Terminology” (91 FR 59988; Docket No. FDA-2026-N-5347). The rule replaces references to “animal” tests and studies with “nonclinical” tests and studies across FDA’s drug and biologics regulations.

For years, some sponsors hesitated to rely on nonanimal data in investigational new drug (IND) applications. Regulatory text that named animal studies explicitly fed that caution. The new wording removes the ambiguity. The FDA says the rule aims to remove an emphasis on animal testing as the only way to assess drug safety in the nonclinical setting.

The agency also stresses what the rule does not do: Classifies the amendments as noncontroversial terminology changes that are not expected to affect industry practice and add no new requirements. For CROs that offer new approach methodologies (NAMs), the opportunity lies in clearer regulatory language and growing sponsor interest, not in a new approval pathway.

The rule is effective February 4, 2027. Comments on the rule and its companion proposed rule are due December 7, 2026. If FDA receives significant adverse comments, it will withdraw the direct final rule and proceed through standard notice-and-comment rulemaking.

Key Takeaways

• The FDA published a direct final rule on September 22, 2026, that replaces “animal,” “preclinical,” and “in vitro” with “nonclinical” in 21 CFR parts 312, 314, 315, 361, and 601.
• The rule takes effect February 4, 2027, unless FDA receives significant adverse comments by December 7, 2026.
• The FDA describes the changes as terminology updates that add no new requirements. Nonanimal data were already permitted.
• FDA’s NAMs use-case database lists 25 illustrative examples, none of which FDA treats as method approvals.
• Validation expectations sit in FDA’s March 2026 draft guidance, which is non-binding.

On September 22, 2026, the US Food and Drug Administration (FDA) published a direct final rule titled “Nonclinical Testing Terminology” (91 FR 59988; Docket No. FDA-2026-N-5347). The rule replaces references to “animal” tests and studies with “nonclinical” tests and studies across FDA’s drug and biologics regulations.

For years, some sponsors hesitated to rely on nonanimal data in investigational new drug (IND) applications. Regulatory text that named animal studies explicitly fed that caution. The new wording removes the ambiguity. FDA says the rule aims to remove an emphasis on animal testing as the only way to assess drug safety in the nonclinical setting.

The agency also stresses what the rule does not do. FDA classifies the amendments as noncontroversial terminology changes that are not expected to affect industry practice and that add no new requirements. For contract research organizations (CROs) that offer new approach methodologies (NAMs), the opportunity lies in clearer regulatory language and growing sponsor interest, not in a new approval pathway.

The rule is effective February 4, 2027. Comments on the rule and its companion proposed rule are due December 7, 2026. If FDA receives significant adverse comments, it will withdraw the direct final rule and proceed through standard notice-and-comment rulemaking.

What the Direct Final Rule Changes

The rule aligns FDA’s regulations with the Food and Drug Omnibus Reform Act of 2022 (FDORA), enacted within the Consolidated Appropriations Act, 2023. Section 3209 of FDORA, widely known as the FDA Modernization Act 2.0, amended section 505(i) of the Federal Food, Drug, and Cosmetic Act. It replaced “preclinical tests (including tests on animals)” with “nonclinical tests” and added a statutory definition of that term.

The rule adapts that definition for 21 CFR 312.3(b). A nonclinical test or study means a test or study conducted in vitro, in silico, or in chemico, or a nonhuman in vivo test or study. It may include:

• Cell-based assays
• Organ chips and microphysiological systems
• Computer modeling
• Other nonhuman or human biology-based test methods, such as bioprinting
• Animal tests or studies

The definition keeps animal studies in scope. The rule broadens the regulatory vocabulary rather than excluding animal data. It also adds a definition of “nonclinical study” to parts 314, 315, and 601. The table below lists selected amendments most relevant to CROs supporting IND and new drug applications (NDA) programs. The full list appears in the Federal Register notice.

Section	What it covers	Change
21 CFR 312.3(b)	Part 312 definitions	Adds definitions of “nonclinical test” and “nonclinical study”
21 CFR 312.23(a)(5)(ii)	Investigator’s brochure	“in animals” becomes “nonclinical studies”
21 CFR 312.23(a)(8)	IND pharmacology and toxicology	“studies of the drug involving laboratory animals or in vitro” becomes “nonclinical pharmacological and toxicological studies”
21 CFR 312.23(a)(8)(ii)(a)	Integrated toxicology summary	“in animals and in vitro” becomes “based on nonclinical studies”
21 CFR 312.32(c)(1)(iii)	IND safety reporting	“animal or in vitro testing” becomes “nonclinical testing”
21 CFR 312.82(a)	Early meetings for serious illnesses	“animal studies” becomes “nonclinical studies”
21 CFR 314.50(d)(2)	NDA nonclinical section	“animal and in vitro studies” becomes “nonclinical studies”

The FDA notes that it has long read paired terms such as “animal or in vitro” to cover all nonclinical testing outside humans. Sponsors have already submitted in silico and in chemico data under the current wording, and the agency has accepted such data when appropriate.

Are you a CRO offering in vitro or digital pathology services?
Pharma sponsors are actively searching for labs with “fit-for-purpose” NAM capabilities. Expand your client base and access hundreds of pre-approved leads.

Validation Still Decides Acceptance

The rule widens the vocabulary, but it does not approve any specific assay. The FDA states that it will examine data from nonanimal studies to determine whether they adequately support the proposed clinical investigation. Validation evidence, therefore, remains the deciding factor.

The FDA’s March 2026 draft guidance offers drug developers a validation framework and general recommendations for using NAMs. The guidance is a draft that contains non-binding recommendations and does not address specific NAMs or their use in drug discovery. It encourages NAMs in regulatory submissions, though, especially where they improve the predictivity of nonclinical studies.

CROs can help sponsors meet these expectations in three ways:

• Context of use. Define the specific regulatory question each assay answers and the endpoints it supports.
• Technical characterization. Supply reproducibility, reference-compound, and performance data generated under documented quality systems.
• Early FDA engagement. Prepare sponsors for Type D meetings with CDER or CBER, or for qualification through FDA’s Innovative Science and Technology Approaches for New Drugs (ISTAND) program.

NAM Technologies CROs Can Offer

The statutory definition names several method families. A 2024 report to FDA’s Science Board, summarized in the rule, found that many newer methods are already available to produce scientifically valid data for regulatory needs. These include systems biology, engineered tissues, in silico methods, alternative organisms such as zebrafish and C. elegans, and microphysiological systems.

• Microphysiological systems (MPS). Organ chips culture human cells under controlled perfusion to model tissue-level responses. They can assess human-specific endpoints, such as liver or cardiac toxicity, that animal models may not capture.
• Complex in vitro models (CIVMs). 3D organoids, spheroids, and precision-cut tissue slices add structural and cellular complexity beyond 2D culture.
• Established in vitro and in chemico assays. OECD-aligned methods for skin sensitization, skin irritation, and eye irritation already appear in approved drug applications, as the FDA database shows.
• In silico models. Computational toxicology and quantitative structure-activity relationship (QSAR) models fall within the definition as “computer modeling.”

The FDA notes that animal testing remains important in some settings. For a product that inhibits a novel molecular target, animal studies may reveal toxicities driven by complex interactions between organ systems, such as hormone, neurotransmitter, or cytokine signaling. CROs that offer both NAMs and targeted in vivo work can help sponsors design integrated strategies that use each where it fits best.

FDA has also explored virtual control groups to reduce the number of animals in studies. These approaches lower animal use within in vivo programs rather than replacing them.

What FDA’s NAMs Use-Case Database Shows

The FDA’s Center for Drug Evaluation and Research (CDER) recently launched a NAMs use-case examples database. It lists 25 examples drawn from public FDA review documents for approved drugs.

Every entry involves an established local-tolerance or sensitization assay:

• Eye irritation: Bovine Corneal Opacity and Permeability (BCOP)
• Skin irritation: EpiSkin and EpiDerm skin irritation tests
• Skin corrosion: Corrositex
• Skin sensitization: DPRA, KeratinoSens, h-CLAT, and U-SENS

The examples come from NDAs for products such as Veklury, Cobenfy, and Journavx. The database links each method to its record in the NIH Collection of Alternative Methods for Regulatory Application (CAMERA).

The FDA attaches a clear limit to the list: The database is not exhaustive, does not constitute guidance or approval of any method, and does not predict acceptance in future submissions. CROs should not describe their assays as “FDA-approved” or “FDA-cleared” based on these entries.

Used accurately, the database is still valuable. CROs that run these OECD-aligned assays can show sponsors documented precedent in approved applications, with page references to the original reviews. CROs offering organ chips, organoids, or in silico tools have no comparable precedent in the database yet. Those labs should focus on context-of-use documentation and early FDA engagement.

Key Dates and Next Steps for CROs

Date	Milestone
March 2026	FDA issues draft NAM guidance (Docket FDA-2025-D-6131)
September 18, 2026	CDER launches the NAMs use-case database
September 22, 2026	Direct final rule and companion proposed rule published
December 7, 2026	Comment period closes
By January 6, 2027	FDA confirms the effective date or withdraws the rule (within 30 days of comment close)
February 4, 2027	Rule takes effect if not withdrawn

CROs can use the months before the effective date to prepare:

• Track the docket. Monitor Docket FDA-2026-N-5347 for significant adverse comments or a withdrawal notice.
• Map services to the definition. Align each assay with the statutory categories, such as cell-based assays, organ chips, or computer modeling.
• Document context of use. Prepare validation packages that state the regulatory question, endpoints, reference compounds, and reproducibility data.
• Cite precedent accurately. Reference database entries and review pages where relevant, without implying FDA approval.
• Update quality agreements. Define escalation timelines for nonclinical findings that may trigger IND safety reports.
• Offer integrated strategies. Pair NAMs with targeted in vivo studies where FDA still expects animal data.

Are you a CRO offering in vitro, organ-chip, or in silico services? Sponsors are searching for labs with documented NAM capabilities.
Register your laboratory to reach them.

Need a NAM-capable testing partner?
Submit a lab request to connect with CROs offering toxicology and pharmacology and drug development services.

This article was created with the assistance of Generative AI and has undergone editorial review before publishing.

Sources and Further Reading

• Nonclinical Testing Terminology, direct final rule, 91 FR 59988
• FDA NAMs Database of Use Case Examples
• General Considerations for the Use of New Approach Methodologies in Drug Development (draft guidance, March 2026)

Frequently Asked Questions (FAQs)