Use of the Weight of Evidence approach as an alternative to nonhuman primate developmental and reproductive toxicity testing for biopharmaceuticals — PMDA Early Consideration Summary

In Plain Language

This document considers how to assess developmental and reproductive toxicity (DART)—potential harm to reproduction, pregnancy, or development of the embryo and fetus—when a biopharmaceutical can be meaningfully tested only in nonhuman primates (NHPs). NHP studies can have important limits, including small groups, substantial variation between groups, and high background pregnancy loss. PMDA describes assessing the whole body of relevant evidence, called a Weight of Evidence (WOE) approach, as one possible alternative when the scientific justification is appropriate. That evidence may include how the drug works, what is known about its biological target, findings in genetically modified animals, information about similar drugs, and the drug’s transfer across the placenta. If that evidence is not sufficient, sponsors may consider testing with a homologous protein or genetically modified animals, while addressing important differences and uncertainties. The document’s purpose is to support careful risk assessment while taking account of the limitations of NHP testing and the 3Rs of animal experiments: reduce, refine, and replace.

What This Guidance Covers

The document is an Early Consideration titled “Use of the Weight of Evidence approach as an alternative to nonhuman primate developmental and reproductive toxicity testing for biopharmaceuticals.” It presents PMDA’s current view on DART risk assessment for biopharmaceuticals when the NHP is the only relevant species for evaluating safety in humans. The document describes biopharmaceuticals as biotechnology-derived pharmaceuticals and focuses on products with high target and species specificity: in other words, products whose biological activity is specific to a target and may be limited to particular species.

The discussion addresses two broad options: assessment using WOE, and alternative testing using homologous proteins (for example, surrogate antibodies) or genetically modified animals that express human target proteins. It does not present a single testing plan that fits every product. Instead, it explains circumstances in which available evidence may inform a risk assessment, circumstances in which testing may still need consideration, and factors relevant to the appropriateness and interpretation of alternative studies.

The document also makes a distinction that matters for interpreting its scope. If a biopharmaceutical is pharmacologically active only in NHPs, DART testing should be conducted in NHPs; however, the alternative scheme described in the document can be used in place of NHPs if appropriate scientific justification is provided. The focus is therefore on assessing when an alternative approach may be justified, not on treating WOE as an automatic replacement for NHP testing.

Why It Matters

For some biopharmaceuticals, an NHP may be the only laboratory-animal species in which binding to the target and pharmacological effects similar to those in humans can be expected. The document notes that NHPs are also often selected for DART testing because of similarities in evolutionary and physiological characteristics and the placental transfer of antibodies. At the same time, NHP DART studies may be limited to hazard assessment because group sizes are small, results can vary widely across groups, and background pregnancy loss is high. In plain language, these features can make it difficult to draw a clear conclusion about the level of risk from the study alone.

The document frames the question in light of both these limitations and the 3Rs—reduce, refine, and replace—in animal experiments. It says that the need for NHP DART testing should be determined more carefully. The WOE approach is presented as a way to consider relevant existing information together, rather than relying automatically on a single type of animal study. The document also recognises that scientific knowledge and information are not yet fully accumulated and that its views may change as knowledge and scientific advances develop.

Key Recommendations

1. Consider whether a WOE assessment can address the DART risk. Biopharmaceuticals are generally highly specific to their targets and less likely than chemically synthesized drugs to have off-target DART, according to the document. Approaches based on placental transfer, mechanism of action, or biological characteristics of the target may therefore be considered. A comprehensive review of publicly available information may provide sufficient information for DART risk assessment.

2. When there are serious concerns about embryo/fetal development, communicate the concern and generally avoid administration during pregnancy. The document gives fetal malformation, embryo/fetal lethality, and abortions as examples of serious concerns that may arise from the intended pharmacological effect at a clinical dose. Such concerns may be informed by nonclinical efficacy studies, the mechanism of action, literature and database reviews, and assessment of placental transfer. In this situation, the document says information on embryo/fetal developmental toxicity should be provided to healthcare professionals and patients without conducting animal studies, and administration to pregnant women should generally be avoided.

3. Assess whether existing information is sufficient, while examining its quality and consistency. The document identifies possible situations in which an overall assessment may support a judgement that human DART risk is low. These include a biopharmaceutical targeting an endogenous protein with well-characterized biological properties, or a product with several similar products already approved and substantial DART information available for pregnant women treated with them. The reliability, appropriateness, and consistency of the available data should be considered, and the DART evaluation should be conducted carefully.

4. Consider systemic exposure and anticipated biological effect for locally administered products. The document gives intravitreal administration as an example of local administration that may result in very low systemic exposure. If systemic exposure in humans is less than the minimum anticipated biological effect level, DART associated with the pharmacological effect is not expected to occur. In such cases, the document considers DART risk low and says DART testing is not considered necessary.

5. If WOE does not provide enough information, consider alternative testing approaches. When sufficient DART information cannot be obtained through WOE, it is necessary to consider conducting DART testing. In that situation, assessment using a homologous protein, such as a surrogate antibody, or genetically modified animals expressing human target proteins may be considered instead of NHP testing. The document calls for an explanation of the approach’s appropriateness, addressing the target specificity of the homologous protein; the target molecule’s distribution in genetically modified animals; species-specific differences in the test substance’s pharmacological action; and effects on physiological function.

6. Recognise the limitations of alternative tests and consider the context of the intended population. Risk assessment using homologous proteins or genetically modified animals may be difficult because the quality characteristics of the homologous protein and clinical candidate are not identical, and because the expression levels and functions of human target molecules in genetically modified animals are not necessarily equivalent to those in humans. The document also says the need for DART testing may vary with factors affecting fertility, such as age and disease; the seriousness of the target disease; and treatment options in Japan.

7. Discuss the approach with PMDA. The document recommends that these approaches be agreed upon with PMDA through active use of consultation services, to avoid any negative impact on the application for marketing approval.

Practical Implications

Interpretation: For a sponsor, the document points toward a product-specific justification rather than a default choice between “test in NHPs” and “do not test.” A WOE assessment would bring together relevant available evidence and explain how it informs the DART risk assessment. The evidence sources identified by the document include mechanism of action, detailed information on the target’s biological properties, reproductive and developmental findings in existing genetically modified animals, and reproductive and developmental information on humans treated with similar drugs.

For toxicologists and statisticians, the stated limitations of NHP studies are relevant when judging what a study can establish: a small group size, variability across groups, and high background pregnancy loss may restrict the assessment to identifying hazards. The document does not provide a statistical method, sample-size rule, or numerical decision threshold for resolving those limitations. Its emphasis is on weighing the available information and considering the reliability, appropriateness, and consistency of the data.

For clinical, regulatory, and safety teams, the situations described have practical consequences. Where serious embryo/fetal concerns are identified, the document points to communicating toxicity information to healthcare professionals and patients and generally avoiding administration to pregnant women. Where a locally administered product has very low systemic exposure, the comparison with the minimum anticipated biological effect level is relevant to whether pharmacology-associated DART is expected. Where alternative animal models are proposed, teams would need to address the specific sources of uncertainty described by PMDA, rather than assuming that the surrogate protein or animal model is equivalent to the clinical candidate or to humans.

The document also makes the intended patient population and treatment context relevant to the decision. For example, age, disease-related effects on fertility, seriousness of the disease, and treatment options in Japan may affect the need for DART testing. Its recommendation to use PMDA consultation services means sponsors should consider discussing the proposed approach with PMDA rather than relying on this general document alone to settle a product-specific plan.

Technical Details

The decision logic described in the document can be followed in stages:

Two qualifications are especially important. First, a conclusion that available information is sufficient is not based on one evidence source alone: the document describes an overall assessment and asks that the reliability, appropriateness, and consistency of the data be considered. Second, an alternative animal model does not remove uncertainty automatically. A homologous protein may differ in quality characteristics from the clinical candidate, while a genetically modified animal may not have human-equivalent target expression levels or functions. These limitations are part of the risk assessment, not incidental details.

The document does not set out equations or quantitative decision rules. Its technical content is a qualitative scheme for considering the evidence, deciding whether further DART testing needs consideration, and assessing the appropriateness and limitations of alternative models.

Who Should Read It

This document is relevant to sponsors developing biopharmaceuticals for which NHPs may be the only species with relevant pharmacological activity, and to nonclinical safety and toxicology teams planning or interpreting DART assessments. It may also be useful to statisticians considering the interpretive limits of NHP study data, regulatory affairs staff preparing the rationale for a proposed assessment, and clinical or safety teams who need to understand the implications of embryo/fetal risk information. Investigators and other professionals involved in reproductive safety assessment may find it useful when considering available evidence, alternative testing approaches, or discussion with PMDA.

Official PMDA Source

• Official title: Use of the Weight of Evidence approach as an alternative to nonhuman primate developmental and reproductive toxicity testing for biopharmaceuticals (Early Consideration)
• Issuing body: Pharmaceuticals and Medical Devices Agency, Center for Product Evaluation
• Date on the document: October 24, 2025
• PMDA PDF: Official PMDA document (PDF)
• PMDA list page: PMDA Early Consideration documents

Source: Pharmaceuticals and Medical Devices Agency (PMDA) website, PMDA Early Consideration documents. This page was created by editing and summarising the PMDA document.

Please consult the official document; this summary does not replace it. PMDA provides the English version for reference, and the Japanese original prevails if the English version and Japanese original differ. PMDA describes Early Consideration documents as reference information reflecting the Agency’s views at the time of publication; those views may change as scientific knowledge develops.

Related Clinical Biostats Resources

• All Clinical Biostats tutorials — Browse other plain-language tutorials on clinical and regulatory topics.
• Statistical Calculators — Access calculators that may support statistical work related to clinical research.