FDA Updates Regulations to Advance Alternatives to Animal Testing

By Kevin Fernando · Oct 5, 2026 · 8 min read

The FDA updated its regulations in September 2026 to recognize that nonclinical tests may include cell-based assays, organs-on-chips, and computer models alongside animal testing. Companies may now use these alternatives where appropriate to demonstrate drug safety before human trials, though some endpoints still require animal studies.

The FDA has shifted how companies must demonstrate drug safety before human trials. Rather than requiring animal testing as the default, the agency now permits alternatives, cell-based assays, engineered tissue systems, and computational models, where they provide adequate evidence of safety and efficacy.

This is not a symbolic change. The regulatory update, formalized in September 2026, rewrites which methods satisfy preclinical requirements and which do not. It specifies which testing phases alternatives can address, which endpoints still require animal studies, and what validation data companies must submit to use a new method. Understanding the scope and limits of this shift matters for researchers, drug developers, and policy analysts tracking how preclinical testing actually works.

What the FDA's new regulations on alternatives to animal testing actually require

The FDA issued a direct final rule in September 2026 that updates terminology in regulations governing human drug and biological product development. The rule recognizes that nonclinical tests may include new approach methodologies (NAMs) such as cell-based assays, organs-on-chips, and computer models, as well as nonhuman in vivo testing.

This builds on the FDA Modernization Act 2.0, signed into law on December 29, 2022, which amended Section 505 of the Federal Food, Drug, and Cosmetic Act. That amendment removed the longstanding requirement that drugs must be tested in animals before entering human clinical trials and explicitly named cell-based assays, organ chips, microphysiological systems, and computer modeling as qualifying alternatives for nonclinical testing.

The FDA also issued a draft guidance intended to help drug developers validate NAMs to be used instead of animal testing in drug development. This guidance describes recommendations for validating NAMs when nonclinical NAMs data are provided in support of a drug application or regarding an order issued under section 505G of the FD&C Act for an over-the-counter monograph. The baseline before this shift was the Animal Rule, published on May 31, 2002, which established the regulatory framework for animal testing in drug development.

Which testing methods now qualify as alternatives under the new rules

The FDA recognizes three broad categories of alternatives to animal testing. NAMs encompass in vitro human-based systems, in silico modeling, and other innovative platforms that can collectively evaluate immunogenicity, toxicity, and pharmacodynamics in humans. Each category addresses different questions about how a drug will behave in the human body.

In vitro and cell-based methods

In vitro testing uses human cells or tissues grown outside the body. Static models are traditional in vitro cell culture systems that lack dynamic environmental factors found in living organisms, such as fluid flow or mechanical forces. These include 2D cell cultures, such as transwells and co-cultures, and 3D models, such as spheroids or organoids.

Cell-based assays measure how a drug interacts with human cells, whether it causes cell death, triggers inflammation, or alters cell function. They can screen for toxicity across multiple cell types and tissues. Because they use human cells rather than animal models, they directly measure effects on human biology. The FDA now accepts these assays for specific endpoints where the science demonstrates they provide adequate predictive value.

Computational and modeling approaches

Computational models, also called in silico methods, use computer simulations to predict how a drug will behave based on its chemical structure and known properties of similar compounds. These models can simulate drug absorption, distribution, metabolism, and excretion in the human body without running physical experiments.

Computational approaches reduce the number of experiments needed by identifying which compounds are most likely to succeed before any testing begins. They can flag potential safety concerns early, such as whether a drug structure resembles known toxins or whether it might interact dangerously with common medications. The FDA now accepts these models as supporting evidence for safety and efficacy claims, particularly when combined with experimental data from other methods.

Organ-on-chip and tissue engineering systems

Organ-on-chip (OoC) refers to a microfluidic device containing living cells arranged to model organ-level function. Microphysiological system (MPS) is the broader regulatory term encompassing OoC and other 3D micro-engineered tissue platforms. These systems recreate the structure and behavior of human organs in miniature.

Unlike static cell cultures, organ-on-chip systems incorporate dynamic environmental factors such as fluid flow and mechanical forces that exist in living organisms. They can contain multiple cell types arranged to mimic how different tissues interact within an organ. A liver-on-chip, for example, can model how a drug is metabolized and whether it damages liver cells. A heart-on-chip can measure whether a drug causes abnormal electrical activity. Because these systems use human cells and replicate organ-level complexity, they predict human responses more accurately than animal models for many drug classes.

How a company uses alternatives in a drug development pathway

Consider a pharmaceutical company developing a new small-molecule drug for inflammation. Before any human testing, the company must gather preclinical safety data.

The company begins with computational screening. Using in silico models, researchers identify which molecular variants are most likely to be safe and effective based on their structure and similarity to known compounds. This narrows the candidate list from hundreds to a handful of promising compounds.

Next, the company tests the leading candidates in cell-based assays. Human liver cells, immune cells, and other relevant tissues are exposed to each drug candidate. Researchers measure whether the drug causes cell death, triggers inflammatory responses, or damages DNA. This phase typically takes weeks and costs far less than animal studies. If a candidate shows toxicity signals, it is dropped.

For candidates that pass cell-based screening, the company may use an organ-on-chip system to model how the drug is metabolized in the liver and whether it damages hepatocytes. A heart-on-chip test can check whether the drug prolongs the QT interval, a cardiac electrical abnormality that can cause dangerous arrhythmias.

Computational models run in parallel, predicting how the drug will be absorbed and distributed in the human body based on its chemical properties.

If all these tests support safety, the company submits an Investigational New Drug (IND) application to the FDA with the cell-based and organ-on-chip data, along with computational predictions. The FDA may accept this package and allow the company to begin human trials. Some drug classes still require animal toxicity studies at this stage, but for many drugs, alternatives now suffice.

During human clinical trials, additional animal studies may still be required for specific endpoints, such as long-term toxicity or reproductive effects. But the shift to alternatives in early preclinical screening reduces the total number of animals used and accelerates the timeline to human testing.

What companies still cannot avoid testing on animals

Despite the regulatory shift, animal testing remains mandatory or is the only accepted method for certain endpoints. After IND approval, GLP experiments for the evaluation of chronic toxicity, reproductive and developmental toxicity, carcinogenicity, and genotoxicity are carried out during the clinical phase of development. These endpoints measure long-term and multigenerational effects that cannot yet be reliably predicted by in vitro or computational methods.

Chronic toxicity studies require dosing animals repeatedly over weeks or months to detect effects that emerge only with prolonged exposure. Reproductive and developmental toxicity studies require intact organisms to assess whether a drug harms a developing fetus or affects fertility. Carcinogenicity studies track whether a drug causes cancer over an animal's lifetime. Genotoxicity studies measure whether a drug damages DNA in ways that might lead to cancer or genetic damage.

For some drug classes, particularly those targeting the central nervous system or those with novel mechanisms of action, animal studies remain the regulatory expectation even in early preclinical phases. Monoclonal antibodies, which are large protein molecules, can now often be tested using alternatives, but traditional small-molecule drugs with unknown mechanisms may still require animal data before the FDA permits human trials.

The regulatory change permits alternatives where the science supports them, but it does not eliminate animal testing. It shifts the burden of proof: companies must now justify why animal testing is necessary, rather than justify why alternatives are acceptable.

Why the FDA made this regulatory shift now

The scientific case for alternatives has strengthened over decades. Over 90% of drugs safe in animals fail in humans, which demonstrates that animal models predict human safety poorly. This failure rate has driven research into human-relevant alternatives.

Technological advances have made alternatives feasible. Organ-on-chip technology, organoid culture, and computational modeling have matured to the point where they can answer specific safety questions reliably. The FDA Commissioner noted that technological advances are allowing the agency to move beyond animal testing in drug development, which has a poor track record of predicting safety and efficacy in humans.

International harmonization has also pushed the shift. The European Medicines Agency supports the regulatory acceptance of NAMs across the European Union to ensure NAMs are scientifically sound and can be used in regulatory decision-making. When major regulators align on accepting alternatives, companies can use the same preclinical data across markets, reducing duplication and cost.

NAMs are intended to generate data that provide similar or improved value compared with current test methods and potentially replace, reduce, or refine animal use, in line with the 3Rs principles. The 3Rs, replacement, reduction, and refinement, are principles that guide the development of alternatives to animal testing. The FDA's regulatory update operationalizes this principle by formally recognizing alternatives.

How this change affects the drug approval timeline and cost

The practical impact of this shift depends on which drug type a company is developing. For drugs where alternatives provide adequate evidence, such as monoclonal antibodies and many small-molecule drugs with well-understood mechanisms, the timeline to IND submission can shorten significantly. Cell-based assays and organ-on-chip studies take weeks, while animal toxicity studies take months. Computational screening takes days.

NAMs also have enormous cost-saving potential. Animal facilities, veterinary staff, and the long duration of animal studies make preclinical testing expensive. In vitro and computational methods require smaller teams and shorter timelines, reducing overhead.

However, the timeline benefit applies only to early preclinical phases. If a drug still requires animal toxicity studies during clinical development, which remains true for many drug classes, the overall approval timeline may not change substantially. The shift accelerates the path from compound identification to IND submission, but not necessarily the path from IND to market approval.

For drug classes where alternatives are not yet accepted, such as some novel therapies or drugs with poorly understood mechanisms, companies still face the full preclinical timeline. The regulatory change is permissive, not prescriptive. It allows alternatives but does not require them, and it does not eliminate animal testing where the science does not yet support alternatives.

What other regulators are doing with alternatives to animal testing

The FDA's shift aligns with international regulatory trends. The European Medicines Agency supports the regulatory acceptance of NAMs across the European Union to ensure NAMs are scientifically sound and can be used in regulatory decision-making. This means companies developing drugs for both US and EU markets can use the same preclinical data package, reducing the need to run parallel studies under different regulatory standards.

The International Council for Harmonisation (ICH), which coordinates regulatory standards across the US, EU, Japan, and other major markets, has also moved toward accepting alternatives. When ICH guidance evolves, all member regulators typically align their requirements, creating a global shift rather than a patchwork of regional rules.

This convergence matters because it allows companies to design a single preclinical program that satisfies multiple regulators. Before the FDA's update, a company might have needed to run additional animal studies for the US market even if the EMA accepted alternatives. Now, as regulators harmonize, companies can use the same NAMs data across jurisdictions, reducing total preclinical cost and timeline.

Do this automatically

A pharmaceutical company that previously conducted animal toxicity studies on ten drug candidates per year, each requiring six to nine months and substantial per-animal costs, can now substitute in vitro cell cultures, computational models, and organ-on-chip systems that compress timelines to weeks while reducing expenses per candidate. These nonclinical alternatives now satisfy FDA requirements for human drug development pathways, making the shift from animal testing economically and regulatory feasible. Healthcare research reads ClinicalTrials.gov records, regulatory filings, and published literature directly, and cites each claim to the record it came from, across as many subjects as you need, with Mimir attaching a citation to every figure. It will not make the decision for you. It gathers and cites the evidence; the judgment stays yours, and deliberately so. Every plan includes it, including the free one.

Start your first report and run it on something you already know well.

This article is general information and research commentary. It is not professional advice, and it does not account for your particular circumstances.