Virus clearance validation study: Consistency strategy for NMPA/FDA/EMA triple declaration

NMPA Regulatory System

After more than 20 years of development, NMPA's virus safety regulatory system has formed a relatively complete regulatory framework. The core documents include: the 2002 "Technical Methods and Validation Guidelines for Removal/Inactivation of Viruses from Blood Products" (Sinopharm [2002] No. 160) as the earliest special guidance document; the 2005 "General Principles for the Technical Review of Viral Safety Evaluation of Biological Tissue Extraction Products and Eukaryotic Cell Expression Products" as the basic framework for technical review; the 2024 "Technical Guidelines for the Validation of Viral Clearance Process Platforms for Clinical Trials of Therapeutic Recombinant Protein Products (Trial Implementation)" as an innovative document, introducing the platform validation concept for the first time; and the General Provisions 0238 "Safety Control of Biological Products Viruses" in the 2025 edition of the Chinese Pharmacopoeia. It is worth noting that the NMPA has clearly stated that it will gradually align with ICH Q5A (R2), and the platform verification guideline for 2024 is a reflection of this trend.

FDA Regulatory System

FDA's viral clearance verification requirements rely primarily on the following documents: ICH Q5A (R2) Guidelines, which FDA adopted as the Guidance for Industry in January 2024; USP Section < 1050.1 >, which provides general-level technical guidance on the design, evaluation, and characterization of viral clearance procedures; and the FDA's biennial Viral Clearance Symposium (most recently in 2023), which provides a platform for industry to interact face-to-face with regulators. A notable feature of the FDA's approach to viral clearance validation is that its Points to Consider documentation and the outcomes of the PDA workshops have a profound impact on industry practice.

EMA Regulatory System

The EMA's regulatory framework for viral safety includes the CPMP/BWP/268/95 Guidelines for the Design of Viral Validation Studies (1996), which has been harmonized and updated with ich Q5A; the EMEA/CHMP/BWP/398498/2005 Guidelines for the Viral Safety Evaluation of Biotechnological Drugs for Clinical Trials, which clarifies the viral safety data requirements for different clinical trial stages; and the European Pharmacopoeia (Ph. Eur.) (Chapter 5.1.7 "Viral Safety"). A notable feature of the EMA is its detailed requirements for the statistical treatment of viral validation studies, and the statistical evaluation of viral titer and the calculation of reduction factors are specifically discussed in the appendix to CPMP/BWP/268/95.

Indicating Consistency of Virus Selection Requirements

Table 1 Comparison of virus selection requirements indicated by three places

As can be seen from Table 1, the three locations are highly consistent in the core principles that indicate virus selection, and all require a risk-based assessment to cover the viruses with different physical and chemical characteristics such as lipid/non-lipid envelopes, DNA/RNA, and large/small. This high degree of consistency provides a solid basis for designing a unified indicative virus program that meets the requirements of the three places.

Narrowing the consistency of model validation requirements

All three places require that the reduced model can represent the virus removal performance of the production process and be consistent in key process parameters. This is elaborated in ICH Q5A (R2) Section 6.2.2. There are the following subtle differences in the specific requirements of the three regulatory authorities: the FDA and EMA place more emphasis on the scientific demonstration of linear/nonlinear amplification of the reduced model; the NMPA puts forward more explicit product representation evaluation requirements for the verification of the reduced model in the platform verification guidelines; and the EMA CPMP/BWP/268/95 provides specific statistical evaluation methods in the appendix.

Consistency of Inactivation Mechanics with Worst Condition Requirements

All three places require that inactivation verification research should include inactivation mechanics research. The NMPA "Technical Methods and Validation Guidelines for Removal/Inactivation of Viruses from Blood Products" clearly requires that "the dynamics of viral inactivation, including the rate of viral inactivation and the inactivation curve, should be studied". The FDA also highlighted the importance of inactivation kinetic curves in assessing the effectiveness of inactivation processes at the 2023 Viral Clearance Symposium. All three places require verification under worst-case conditions, but the NMPA lists the worst-case parameters in more detail in the platform verification guidelines.

Consistency of LRV statistical calculation methods

Table 2 Comparison of LRV statistical calculation methods in three places

The three places are completely consistent in the LRV calculation method, and all use the log10 (N0/N) formula. The main difference is that the EMA has more stringent requirements for confidence interval assessment (Appendices I and II provide detailed statistical methods), while the NMPA, while recommending the use of confidence intervals, does not mandate them

Unified solution design based on ICH Q5A (R2)

Given that ICH Q5A (R2) has been adopted by the regulatory authorities of the three places, designing a unified verification scheme based on the core requirements of Q5A (R2) is the most effective way to achieve the coordination of the three places. Here are some suggestions:

First, a uniform combination of indicative viruses is used. It is recommended to contain at least the following viruses: MVM or PPV (non-lipid enveloped small virus), PRV or HSV (lipid enveloped large virus), X-MuLV (retroviral model), BVDV or VSV (lipid enveloped medium virus), and Reo-3 (non-lipid enveloped medium virus). This combination is able to fully cover ICH Q5A (R2) and the directive virus requirements of the three regulatory authorities.

Second, use a unified miniaturized model design. Ensure that the reduced model represents a commercial production process on key process parameters and is fully validated as required by Q5A (R2). It is recommended to include the comparison of product quality attributes, key process parameters, and process variability in the miniaturized model validation.

Third, adopt a unified statistical analysis method. It is recommended that the 95% confidence interval for LRV be calculated in all validation studies to meet the mandatory requirements of the EMA and the recommendations of the NMPA/FDA. At the same time, comprehensive evaluation methods are used in effectiveness evaluation, rather than relying on a single threshold.

Modular Document Design Strategy

In order to achieve the goal of adapting a report to three regional declarations, it is recommended to adopt a modular document design strategy: core modules (common parts) - including risk assessment reports, narrowed model description and validation data, virus removal verification results for each process step, statistical analysis reports, and overall virus safety evaluation. NMPA Supplemental Module - Includes a declaration form that meets the specific format requirements of the NMPA, a description of the comparison with the requirements of the Chinese Pharmacopoeia, and a description of the response to the NMPA-specific guidelines. FDA Supplemental Module - Includes FDA Guidance Statement of Suitability, Controlled Analysis of USP < 1050.1 >. EMA Supplemental Module - Includes supplementary materials for statistical analysis as required in the CPMP/BWP/268/95 Appendix, instructions for step-by-step declaration during the clinical trial phase.

The suitability of the three places verified by the platform

The "Guidelines for the Validation of Viral Clearance Process Platforms for Clinical Trials of Therapeutic Recombinant Protein Products (Trial)" issued by NMPA in 2024 is an important milestone in the coordination of the three places. The guidelines clearly state that the platform validation is applicable to clinical trial applications for therapeutic recombinant protein products produced using mammalian cells, and the scope of application includes three processes: S/D inactivation, low pH inactivation, and virus interception and filtration. Although the guideline currently only applies to NMPA jurisdictions, its technical requirements are highly consistent with ICH Q5A (R2), so harmonized technical standards can be considered when designing platform verification schemes.

For gene therapy products, although it has not yet been included in the scope of application of platform verification, the concept of platform verification "based on sufficient internal prior knowledge, evaluation of virus clearance performance of other similar products" also has reference value. With the full adoption of ICH Q5A (R2) by NMPA, the scope of platform validation is expected to be further expanded.

Differentiated coping strategies

While pursuing consistency, the following differentiation requirements require special attention:

Confidence Interval Requirements: The EMA mandates a 95% confidence interval for LRV, which is recommended to be included by default in the Unified Protocol to avoid the need for complementary studies. Although the NMPA threshold of ≥ 4 logs is mainly for blood products, it should be designed to ensure that the clearance effect of key process steps is not lower than this level.

Inactivation Mechanics Study: The NMPA explicitly requires the study of viral inactivation mechanics, including inactivation rate and inactivation curve. It is recommended that multi-point sampling designs be included in the validation of all inactivation steps to meet the specific requirements of NMPA. The NMPA requirements for the third generation of blind transmission should also be considered in the testing protocol.

Forward Communication

In view of the complexity of virus clearance verification and the slight differences in the three local regulations, it is strongly recommended to communicate fully with the various regulatory agencies before the verification plan is determined: communication with the NMPA can be conducted through the Pre-IND meeting or the general technical consultation channels of the CDE, and it is especially recommended to communicate with the CDE before adopting the platform verification. Communication with the FDA can be through a Pre-IND or Pre-BLA meeting. Communication with the EMA can be done through the Scientific Advice procedure. In the communication, it is suggested to focus on the rationality of indicating virus selection, the representativeness of narrowing the model design, the basis for setting the worst conditions, and special considerations for specific products.