FDA: Next Step to Reduce Animal Based Testing Methods

FDA and EMA Move Toward Fewer Animal Tests in Drug Development

We have previously reported on approaches in Europe, the United States and other countries to implement the 3Rs strategy—replacing animal testing, reducing the number of animal tests, or reducing the number of animals used in pharmaceutical development and manufacturing. On September 21, the FDA published “FDA Updates Regulations to Advance Innovative Alternatives to Animal Testing.”

Key message

The texts and the underlying publications, guidance documents and regulations point to a clear shift in preclinical drug development: animal testing is not to be abolished across the board, but reduced or replaced where modern, scientifically robust and human-relevant methods can provide equivalent or better evidence. The focus is on so-called New Approach Methodologies (NAMs). These include, in particular:

• complex in vitro models;
• organoids and organ-on-chip systems;
• microphysiological systems;
• computer-based and AI-based models;
• in silico toxicology;
• virtual control groups; and
• combinations of several methods as part of a “weight-of-evidence” approach.

The key paradigm shift is that animal testing is no longer regarded by regulators as the only permissible source of nonclinical safety data. This does not mean a blanket move away from animal testing. Instead, it means a more flexible, scientifically justified assessment tailored to the product and the specific question being addressed.

1. Regulatory developments in the United States

On September 21, 2026, the FDA published a Direct Final Rule. It clarifies that appropriate non-animal methods may be used to assess the safety of drugs and biological products before they are tested in humans. In the affected regulations, the rule replaces terms such as “animal tests” and “animal studies” with “nonclinical tests” and “nonclinical studies.” This aligns the terminology with the Food and Drug Omnibus Reform Act of 2022 (FDORA).

The rule:

• recognizes both traditional animal testing and modern alternatives as possible bases for regulatory decisions;
• removes wording that could suggest animal testing is the only permissible approach;
• does not change existing standards of evidence;
• does not prohibit animal testing; and
• according to the FDA, does not impose additional requirements or costs on sponsors.

The FDA explicitly emphasizes that animal testing may still be necessary when it is better suited to the specific scientific question. The determining factor, therefore, is not the method itself, but whether it is suitable for answering a particular regulatory question. (FDA: FDA Updates Regulations to Advance Innovative Alternatives to Animal Testing).

To illustrate the approach in practice, the FDA has also established a database initially containing 25 specific examples of NAM applications drawn from publicly available review documents. These examples are intended to show sponsors the contexts in which alternative data have already been assessed by regulators.

2. Implementation of the FDA roadmap since 2025

The current developments build on the FDA Roadmap to Reducing Animal Testing in Preclinical Safety Studies, published in April 2025. One year later, the FDA reported significant progress in implementing the roadmap.

The measures cited include:

• draft guidance on reducing or avoiding longer-term primate studies for certain monoclonal antibodies;
• promotion of recombinant alternatives to reagents derived from horseshoe crabs for endotoxin testing;
• greater use of “weight-of-evidence” approaches;
• broader regulatory acceptance of in vitro, computer-based and AI-based methods;
• qualification of a first AI-based drug development tool;
• development of a searchable database of acceptable alternative methods; and
• closer collaboration with international authorities

According to the FDA, moving away from horseshoe-crab-based endotoxin reagents alone could save more than one million animals each year. PharmTech identifies the AI tool qualified by the FDA as AIM-NASH. (FDA: FDA Achieves Year 1 Goals in Reducing Animal Testing in Drug Development; PharmTech: Using Virtual Control Groups to Reduce Animal Testing)

The FDA also justifies this shift by pointing to the limited translatability of much animal data to humans. It notes that more than 90% of drugs that successfully pass animal testing do not later receive FDA approval—often because of safety or efficacy issues in clinical trials. This figure should be understood as a general FDA rationale for the strategic shift, not as a uniform failure rate for all drug classes or development programs.

3. FDA draft guidance on the validation of NAMs

A key practical step was the draft guidance General Considerations for the Use of New Approach Methodologies in Drug Development, published in March 2026. It is intended to help sponsors demonstrate the scientific suitability of NAMs for regulatory purposes.The FDA identifies four main assessment criteria:

1. Context of Use: It must be clearly defined which specific regulatory question the method is intended to address and at what stage of development it will be used.

2. Human Biological Relevance: The method must be demonstrably linked to human biology, relevant disease mechanisms or human toxicological processes.

3. Technical Characterization: The method must be adequately described and shown to be reliable, reproducible and scientifically robust.

4. Fit-for-Purpose: It must be demonstrated that the method can contribute to the regulatory decision for which it is intended.

The guidance does not require every NAM to undergo comprehensive formal validation first. Instead, its suitability is to be assessed in the context of its intended application. At the same time, the FDA notes that early consultation with the relevant FDA division is advisable, particularly for organ-, disease-, indication- and endpoint-specific applications. (FDA: FDA Releases Draft Guidance on Alternatives to Animal Testing in Drug Development)

5. The EMA’s approach: The EMA follows a similar approach, but one more firmly embedded in the European 3Rs framework:

• Replacement — replacing animal testing;
• Reduction — reducing the number of animals used; and
• Refinement — improving procedures and reducing the burden on animals.

The EMA considers NAMs to be methods capable of generating data of comparable or greater scientific value than established methods. It supports their regulatory acceptance for both human and veterinary medicines.

Developers have several ways to engage with the EMA:

• early, informal discussions through the Innovation Task Force;
• scientific advice;
• formal qualification procedures; and
• voluntary submission of data outside a specific marketing authorization procedure.

The EMA also places the Context of Use at the centre of its approach. In particular, it expects:

• a clearly described methodology and defined endpoints;
• a well-founded biological and regulatory rationale;
• evidence of reliability and robustness.

Particularly high standards apply to safety questions. Formal qualification is not mandatory in every case; depending on the question, acceptance may also be considered as part of a specific marketing authorization procedure. The EMA nevertheless recommends engaging with regulators as early as possible. (EMA: Regulatory Acceptance of New Approach Methodologies)

5. Virtual control groups as a practical example

Virtual control groups show that reducing animal testing does not have to involve fully replacing individual studies. This approach uses sufficiently characterized historical control data to create a virtual comparison group. Under certain conditions, this can eliminate the need for a concurrent control group.

Potential benefits include:

• fewer animals used;
• shorter study durations;
• greater statistical power through the use of larger datasets; and
• more efficient use of existing study data.

However, the approach depends heavily on the quality, completeness and comparability of the historical data. Biases, data gaps, or differences in animal strain, housing, study methods and laboratory conditions can affect the reliability of the results. Conventional control groups may still be needed for new biological mechanisms, rare endpoints or poorly characterized species.

Overall assessment and practical implications

Developments at the FDA and EMA mark a transition from a preclinical model centred on animal testing to an integrated, human-relevant and risk-based approach to generating evidence. Animal testing is not ruled out; rather, it is being considered as one option within a broader range of methods.

For pharmaceutical companies, this has several practical implications:

• NAMs should be considered when planning a development program.
• The intended regulatory use must be defined early and precisely.
• Relevance to human biology is just as important as technical validity.
• Individual NAMs may be used as replacement methods, as complementary methods or as part of a weight-of-evidence approach.
• Data quality, standardization, reproducibility and traceability will be critical to regulatory acceptance.
• Early scientific advice from the FDA or EMA can reduce the risk that alternative data will not be adequately considered later in the authorization process.
• International harmonization is becoming increasingly important, as an approach should ideally be accepted by multiple authorities.

In summary, the FDA and EMA are creating increasingly concrete regulatory pathways for the use of NAMs. The current shift does not mean “no more animal testing”; it means making a scientifically justified choice of the most appropriate method for each purpose. The approaches most likely to succeed will be those with a clearly defined context of use, demonstrated relevance to human biology, robust technical characteristics and a clear contribution to regulatory decision-making.