Considerations for the General Toxicity Evaluation of Monoclonal Antibodies — PMDA Early Consideration Summary

In Plain Language

Monoclonal antibodies often act on a specific target, and the animal species in which they behave sufficiently like they do in humans may be limited, often to monkeys. PMDA sets out factors to weigh when deciding whether a six-month repeated-dose toxicity study in monkeys is likely to add useful information after shorter-term studies and other evidence are considered. It also describes approaches for products that cannot be assessed in an animal species with relevant pharmacological activity, including the possible use of transgenic animals, homologous proteins, New Approach Methodologies, and other relevant data. The aim is to evaluate potential risks while avoiding animal studies that may add limited information, consistent with promoting the 3Rs: reduce, refine, and replace animal experiments. The document emphasizes a case-by-case, weight-of-evidence assessment rather than a single approach for every antibody.

What This Guidance Covers

The document presents PMDA’s current thinking on general toxicity evaluation using a Weight of Evidence (WOE) approach for conventional monoclonal antibodies, identified in the document as “-tug, unmodified immunoglobulins.” Its focus is nonclinical safety evaluation: assessing toxicity information from animal studies and other sources to inform development and clinical-trial participant safety.

It first addresses whether a six-month repeated-dose toxicity study in monkeys is needed when monkeys are the only relevant species. It then describes general considerations for cases in which no appropriate animal species exists, with separate discussion of targets that are exogenous factors, such as bacteria or viruses, and targets that exist only in humans or do not show sufficient pharmacological activity in animals.

The document also refers to other elements of antibody-therapeutic safety evaluation, including tissue cross-reactivity studies, risk assessment of cytokine release mediated by immune activation when appropriate, and assessment of potential secondary effects associated with off-target binding. These points appear as part of the wider evaluation; the detailed considerations that follow focus on general toxicity evaluation.

Its stated scope is conventional monoclonal antibodies. It does not set out a general approach for every type of biotechnology-derived product, nor does it give a universal study-duration rule applicable to all products. Instead, it describes factors and examples to consider in the situations it discusses.

Why It Matters

PMDA describes a challenge specific to many monoclonal antibodies: high target specificity and species specificity can mean that only a limited number of animal species bind the target sufficiently and show pharmacological effects comparable to those expected in humans. In many cases, the relevant species may be non-human primates, typically monkeys. This can make conventional animal toxicity evaluation difficult to design and interpret.

The document also notes that most toxicities associated with monoclonal antibodies are recognized as related to their intended mechanism of action, while the risk of off-target toxicity is low. Retrospective studies discussed in the document found that short- and long-term monkey studies often observed predictable findings, such as exaggerated pharmacological effects and immunogenicity. The cited data suggest that a three-month general toxicity study can provide sufficient information to support clinical development in many cases.

Against this background, PMDA considers whether longer studies consistently add useful information, and how evidence from other sources can contribute. The document links this evaluation to the 3Rs and to recent advances in New Approach Methodologies (NAMs). Its concern is to retain attention to human safety while considering whether toxicity testing can be rationalized on a scientific basis.

Key Recommendations

1. Begin by assessing whether an appropriate animal species exists. In general toxicity evaluation of biotechnology-derived pharmaceuticals, the document says it is important first to assess whether a species exists in which the product’s pharmacological activity is exhibited. The fundamental approach differs depending on whether such a species is available.

2. When monkeys are the only relevant species, consider the need for a six-month study using WOE. The document says the factors below should be comprehensively taken into account when considering whether a six-month repeated-dose toxicity study is necessary. They include results from toxicity studies to date, including three-month monkey studies and toxicokinetic data; the mechanism of action and pharmacological characteristics; literature and database information on potential toxicities associated with the target molecule; other nonclinical data, including NAM data; and available clinical safety and pharmacokinetic data. Information on other monoclonal antibodies targeting the same antigen is also relevant, including their nonclinical study results and human safety information.

3. Recognize circumstances in which a six-month study is likely to be required. The document gives examples: toxicity findings that cannot be explained by pharmacological effects or class effects and have an unknown mechanism; findings that cannot be appropriately monitored in humans; target expression and distribution across multiple organs and tissues where diverse toxic effects may occur; or limited precedent for the pharmacological target, making risk prediction from existing knowledge difficult. These are examples for consideration within the WOE assessment.

4. Consider whether the additional information from a six-month study would be limited. The document says that if none of the listed circumstances applies, or if immunogenicity has a substantial impact on repeated-dose monkey studies and makes toxicity interpretation difficult, the additional value of a six-month study may be limited. The scientific merit of conducting the study should therefore be considered case by case, based on risk assessment using WOE.

5. When no relevant species exists, consider what forms of evidence can address the gap. The document refers to consideration of relevant transgenic animals expressing the human receptor or homologous proteins. If neither can be used, it describes the referenced approach of a repeated-dose toxicity study of up to 14 days in a single animal species, including evaluation of critical functional endpoints such as the cardiovascular and respiratory systems.

6. Distinguish between types of targets when no appropriate species exists. For an exogenous target, such as a bacterium or virus, the primary pharmacological action is generally not expected to cause toxicity, so the main concerns are off-target effects. In this situation, the document considers the relevance of a stand-alone repeated-dose study in a species without demonstrated pharmacological activity to be limited; it considers it sufficient to confirm that there are no particular concerns regarding critical functional endpoints in animal studies conducted for efficacy or pharmacokinetics. For targets that exist only in humans, or where an animal ortholog does not show sufficient pharmacological activity, transgenic animals or homologous proteins may be appropriate to assess hazards related to pharmacological activity.

7. Use all relevant evidence when direct animal assessment is not possible. For human-only targets or insufficiently active animal orthologs, the document describes Target Safety Assessment information—such as target physiology, involved pathways, protein and gene expression, genetically modified animal phenotypes, and human gene annotations—as a potentially useful reference. NAMs focused on specific endpoints may also be used in combination to assess potential risks anticipated from the pharmacological activity. If neither transgenic-animal studies nor homologous-protein studies can be conducted, information for general toxicity evaluation is limited; the document stresses maximizing relevant available information, including nonclinical and clinical safety data for related products, and evaluating toxicity using WOE. It also says to confirm in any available animal study that there are no particular concerns regarding critical functional endpoints.

8. Discuss the approach with PMDA as appropriate. In its conclusion, the document encourages sponsors to proactively use consultation opportunities to discuss the need for long-term repeated-dose toxicity studies in monkeys and the application of a WOE approach, including use of NAM data. It states that PMDA intends to promote the 3Rs on the premise of ensuring human safety and to advance case-by-case consideration of rationalized evaluation.

Practical Implications

Interpretation: Sponsors can organize the evidence around the decision being considered: whether a longer monkey study is likely to resolve a safety uncertainty, or whether other evidence is more informative. The document does not describe a checklist that automatically determines the answer. It calls for a comprehensive review of product-specific findings, target biology, related antibodies, and clinical and nonclinical information.

For study planning, this means the rationale for the duration should be linked to the uncertainties identified. A three-month study may provide sufficient information in many cases, according to the data described, but findings that are unexplained, difficult to monitor in humans, potentially broad because of target distribution, or hard to predict from limited precedent are examples that may support considering a six-month study. Conversely, substantial immunogenicity that impairs interpretation is a reason the added value of a longer study may be limited.

For statisticians and toxicologists, interpretation of results depends on understanding the product’s pharmacology and the quality and relevance of the evidence being combined. Toxicokinetic results, for example, are included among the information from existing studies to consider. A study’s duration alone does not answer whether it provides interpretable or decision-relevant evidence.

For teams developing products without a relevant animal species, the document points to a different planning question: which available methods can address hazards associated with the pharmacological activity, and which critical functional endpoints can still be checked in available animal studies? Depending on the target and available methods, the evidence may involve transgenic animals, homologous proteins, NAMs, target biology, and safety information from related products. The document’s approach is to combine relevant information rather than treat a single unavailable model as the end of the evaluation.

For regulatory teams, the conclusion makes consultation relevant to case-specific discussion of study duration and WOE, including NAM information. This is particularly pertinent where the available animal model has limitations or where the evidence does not point clearly to the value of a longer study.

Technical Details

The practical sequence described by the document can be understood as follows:

Where no appropriate species exists, the document describes different evidence options depending on the target. For exogenous factors, it focuses on off-target concerns and checking critical functional endpoints in animal studies that are being conducted for efficacy or pharmacokinetics. For human-only targets or targets with insufficient activity in animals, the assessment may use transgenic animals or homologous proteins, target-safety information, NAMs, and other relevant data. If the first two study options are unavailable, the document emphasizes using all relevant available information and checking critical functional endpoints in any available animal study.

The document does not give a numerical decision rule for choosing a study duration. It describes WOE factors, examples of circumstances relevant to the decision, and the need for case-by-case scientific consideration. It names cardiovascular and respiratory systems as examples of critical functional endpoints in the discussion of studies when no relevant species exists.

Who Should Read It

This document is relevant to nonclinical toxicologists and pharmacologists planning or interpreting general toxicity studies for conventional monoclonal antibodies, as well as biostatisticians who contribute to study design or interpretation. It may also help regulatory affairs staff, clinical research teams, and project leaders understand how nonclinical findings and clinical safety information can inform decisions about study duration.

It is especially pertinent when monkeys are the only relevant species and a six-month repeated-dose study is under consideration, or when no animal species shows appropriate pharmacological activity. Researchers working with NAMs, transgenic models, homologous proteins, or target-safety information may also find its evidence framework relevant.

Official PMDA Source

• Official title: Considerations for the General Toxicity Evaluation of Monoclonal Antibodies (Early Consideration)
• Issuing body: Pharmaceuticals and Medical Devices Agency, Center for Product Evaluation
• Date on the document: May 13, 2026
• Document: PMDA PDF
• PMDA Early Consideration list: PMDA list page

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

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

Related Clinical Biostats Resources

• All Clinical Biostats tutorials — Browse other educational tutorials across clinical research and related topics.
• Statistical Calculators — Access calculation tools that may support quantitative work in clinical research.