Quality-Related Considerations for Human Cell-Derived Extracellular Vesicle (EV) Products at the Time of Initial Clinical Trial Notification — PMDA Early Consideration Summary

In Plain Language

The document addresses quality considerations for human cell-derived extracellular vesicle (EV) products being prepared for an initial clinical trial notification. “EVs are membrane-enclosed structures released by cells”, and the document notes that biological raw materials used to make them, as well as difficulties separating EVs from viruses, create particular safety-control challenges. It asks developers to build a suitable strategy for preventing and testing for unwanted infectious agents, to assess impurities and establish relevant tests, and to evaluate possible risks from particles and materials carried through manufacturing. It also discusses how to assess quality attributes linked to a product’s intended effects when its active components or mechanism of action may not yet be clear. The overall purpose is to support safety at the start of clinical investigation while encouraging important product and cell-substrate decisions to be made early.

What This Guidance Covers

The document applies to products containing purified EVs derived from human cells as the active substance and intended for marketing authorization as therapeutic drugs. It also says that its considerations may be applied to certain engineered EV products. These include products where producer cells have been modified by gene transfer or other means, or where EVs have been processed—for example, by peptide conjugation or encapsulation of specific molecules—to control biodistribution or confer pharmacological activity.

Its focus is quality-related considerations at the time of an initial clinical trial notification. It covers basic quality principles, control of adventitious agents, other safety-related impurity considerations, and considerations beyond safety. It does not set out a complete product-development or clinical-trial design framework; its scope is the quality of the EV product and related manufacturing considerations at this early notification stage.

Why It Matters

The document highlights two connected challenges. First, EV products are manufactured using biological raw materials, including cell substrates. At the same time, EVs can have physicochemical properties similar to viruses, such as size and charge, which may make it difficult to establish a purification process that can clear viruses. PMDA therefore emphasises establishing an appropriate control strategy for adventitious agents—the unwanted infectious agents that may enter through raw materials or the manufacturing process.

Second, the document explains that EV products can contain or carry materials other than the intended EVs. These may include process-related impurities, non-vesicular extracellular particles, other cell-substrate-derived particles, or extrinsic particles introduced during manufacturing. Some EV-derived proteins may be difficult to distinguish from host cell proteins. These features make it important to understand what is present, consider its potential safety impact, and establish appropriate tests where needed.

The document also recognises that, at the initial clinical trial notification stage, the mechanism of action and the components responsible for biological activity may not be fully understood. It therefore recommends gathering broad information on product quality attributes related to efficacy, rather than assuming that a single characteristic fully describes the product’s activity.

Key Recommendations

1. Provide information that supports safety at the initial notification. The document says that sufficient information to support the safety of the investigational medicinal product should be provided, taking account of the matters specified in Q&A No. 27 of the “Revision of Questions and Answers (Q&A) regarding the Notification of Clinical Trial Plans and Implementation of Clinical Trials for Drugs.” If there are concerns about the safety of a product under development, applicants are recommended to consult PMDA through a face-to-face consultation meeting before submitting the initial clinical trial notification.

2. Build a control strategy for adventitious agents. The document points out that effective virus removal and inactivation may be difficult during EV-product manufacturing. It therefore calls particular attention to choosing and testing raw materials for the absence of viruses and conducting virus testing at appropriate manufacturing steps, in accordance with the principles of “Viral Safety Evaluation of Biotechnology Products Derived from Cell Lines of Human or Animal Origin” (ICH Q5A(R2)). A control strategy should be established with reference to those principles and the document’s three areas of consideration below.

   • Prevent adventitious agents from entering through raw materials. Biological raw materials used in manufacturing must comply with the Standards for Biological Raw Materials. Where a cell substrate is prepared from donor tissue, the appropriateness of the donor interview, medical examinations, tests and other procedures should be confirmed. When establishing a master cell bank or working cell bank, cell-substrate virus testing should be conducted in accordance with ICH Q5A(R2). Testing should confirm that the cell substrate is not contaminated with at least hepatitis B virus (HBV), hepatitis C virus (HCV) and human immunodeficiency virus (HIV). Other viruses should also be tested for according to infection risk, taking the tissue of origin into account. Biological raw materials used in manufacturing—including culture-medium components and additives—should also be confirmed not to be contaminated with bacteria, fungi, mycoplasma and viruses.
   • Test manufacturing material at suitable process steps. In general, testing for bacteria, fungi, mycoplasmas and viruses should be conducted at the step that provides the highest sensitivity, and the process fluid should be confirmed not to be contaminated. The document notes that unprocessed or unpurified bulk provides the highest sensitivity for virus testing of therapeutic products consisting of recombinant proteins. For EV products, however, viruses may become concentrated during ultrafiltration or other processes. Developers should therefore also consider whether testing a purification-process intermediate, such as a post-filtration intermediate, could provide the highest sensitivity.
   • Evaluate virus clearance during process development. Purification methods capable of separating viruses from EVs are being developed, so virus removal may be expected during EV purification. The document says virus-clearance studies should be conducted during manufacturing-process development to evaluate the process’s virus-clearance capability. If no purification step is expected to remove viruses, modification of the process or addition of a virus-removal step should be considered.

3. Characterise other impurities and consider safety-related tests. The removal of process-related impurities—including host cell proteins (HCPs), host cell DNA, medium additives and column packing materials used in purification—should be characterised. Because it can be difficult to distinguish HCPs from EV-derived proteins in the drug substance or drug product, monitoring the protein-to-particle ratio is described as one possible approach to indicate impurity levels. Non-vesicular extracellular particles, other particles derived from the cell substrate, and extrinsic particles introduced during manufacturing should be identified through characterisation of the process and drug substance, and their impact on safety should be evaluated. As necessary, impurity tests should be established as part of the specifications and analytical procedures for the drug substance or drug product.

   Depending on the dosage form, tests such as sub-visible particle testing, inspection for foreign insoluble matter, sterility testing and bacterial endotoxins testing should be established in accordance with the General Rules for Preparations of the Japanese Pharmacopoeia.

4. Assess impurities associated with immortalizing genes. When immortalized primary cultured cells are used as the cell substrate, nucleic acids and proteins derived from the immortalizing gene may contaminate the EV product and potentially cause it to exhibit carcinogenicity. The contents of these nucleic acids and proteins in the product should be evaluated, and these impurities should be removed to the greatest extent possible.

5. Assess efficacy-related quality attributes broadly. At the initial clinical trial notification stage, the mechanism of action and the EVs’ biological activities or active components—such as proteins and microRNAs responsible for biological activity—may not yet be fully identified. The document says it is desirable to assess efficacy-related quality attributes from multiple dimensions using multiple analytical methods, and to collect broad information on the mechanism of action and active components.

6. Select the cell substrate early. If the cell substrate changes, such as through immortalization of host cells or a change of donor, comparability before and after the change is highly likely not to be sufficiently demonstrable. The document therefore says the cell substrate should be selected at an early stage of development, including before non-clinical studies are conducted.

Practical Implications

Interpretation: The recommendations point toward treating product quality and safety controls as an integrated manufacturing question, rather than relying on a single final-product test. A sponsor’s development work will need to consider what enters through donor tissue and other biological raw materials, where testing is most informative in the process, and whether the purification steps have been evaluated for virus clearance.

For manufacturing and quality teams, this means documenting the rationale for testing locations and impurity assessments, while characterising both the intended EV material and other particles that may be present. For analytical teams, the document identifies the protein-to-particle ratio as one possible indicator of impurity levels, not as a complete substitute for characterisation or as a universal specification. The tests included in specifications and analytical procedures depend on the product and, for certain dosage-form tests, on the dosage form.

For development and regulatory teams, the document draws attention to decisions that may be difficult to reverse later. Choosing a donor or changing the cell substrate can affect whether comparability can be sufficiently demonstrated, so the stated preference is to select the substrate early, including before non-clinical studies. If safety concerns remain, the document recommends considering a face-to-face consultation with PMDA before the initial notification is submitted. These are practical implications of the document’s points; they do not replace the official document or product-specific regulatory discussions.

Technical Details

The technical logic can be understood as a sequence of linked assessments:

The document’s key distinctions matter when translating these steps into a development plan. Testing at the most sensitive process step is a general principle, but the document specifically cautions that the best step for EV-product virus testing may differ from the unprocessed or unpurified bulk approach described for recombinant-protein products. Similarly, the protein-to-particle ratio is presented only as one possible way to monitor impurity levels where HCPs and EV-derived proteins are difficult to distinguish.

For products made using immortalized primary cultured cells, the assessment includes the contents of nucleic acids and proteins derived from the immortalizing gene, because the document identifies potential carcinogenicity as a concern. More generally, impurity removal should be characterised, and removal of these immortalizing-gene-derived impurities should be pursued to the greatest extent possible.

The document also separates safety-related quality work from efficacy-related quality assessment. It describes specific potential contaminants and safety tests, while acknowledging that active components and mechanisms may remain unclear at the notification stage. For efficacy-related attributes, its stated approach is multidimensional assessment with multiple analytical methods and broad information gathering—not reliance on a single identified active component.

Who Should Read It

This document is relevant to sponsors and developers preparing an initial clinical trial notification for a therapeutic product containing purified human cell-derived EVs. It will also be useful to regulatory affairs staff, quality and manufacturing professionals, analytical scientists, microbiology and viral-safety specialists, and investigators involved in early development. The considerations may also be relevant to developers of the engineered EV products described in the scope, including those modified at the producer-cell level or processed to affect biodistribution or pharmacological activity.

Official PMDA Source

• Official title: “Quality-Related Considerations for Human Cell-Derived Extracellular Vesicle (EV) Products at the Time of Initial Clinical Trial Notification (Early Consideration)”
• Issuing body: Pharmaceuticals and Medical Devices Agency, Office of Cellular and Tissue-based Products
• Date printed on the cover: June 16, 2026
• Official PMDA PDF: Read the PMDA document
• PMDA Early Consideration list page: View the PMDA list page

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

Consult the official document; this summary does not replace it. PMDA provides the English version for reference purposes, and the Japanese original prevails if the Japanese original and English version 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 summaries, including material on clinical trial notifications and product development.
• Statistical Calculators — Access statistical calculators that may support quantitative work in clinical research.