病毒清除研究优化与工艺相关杂质检测表征的实践考量

Porcine parvoviruses, depicted here, are among the model viruses that companies can use to demonstrate viral-clearance capability. (https://stock.adobe.com)

Viral clearance (VC) studies remain a cornerstone of ensuring viral safety during the development and manufacture of biologics, vaccines, and advanced therapies. As modalities such as adenoassociated virus (AAV) gene therapies and increasingly complex biopharmaceuticals continue to evolve, so too do expectations for demonstrating robust and reliable VC. Although regulatory frameworks such as the International Council on Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) Q5A and European Medicines Agency (EMA) virus-safety guidance documents provide essential frameworks, successful execution of such studies depends heavily on how well theoretical principles translate into real-world practice (1, 2).

In reality, even well-designed VC studies can fall short — not because of gaps in scientific understanding, but because study design, process maturity, analytical limitations, and execution are not always fully aligned. Increasingly, success depends on experience-driven decision-making, careful planning, and integration of complementary analytical approaches to support comprehensive detection and characterization of process-related impurities.

Bridging the Gap Between Study Design and Execution: Lessons Learned

VC studies are intended to demonstrate a manufacturing process’s ability to remove or inactivate potential viral contaminants. Translating that objective into scientifically sound and regulatorily defensible data is not always straightforward. In practice, studies most often fall short when design decisions inadequately reflect process maturity, operational realities, and analytical strategy. Timing, study scope, material readiness, and assay selection all shape whether a study produces reliable, decision-ready data or creates delays, ambiguity, and the need for repeat work. Several practical lessons consistently emerge across VC programs.

Design studies around the process. Although standard frameworks provide a useful starting point, VC studies must be designed for a specific modality, process, and development stage. Each study should start with a comprehensive risk assessment; otherwise critical variables may be missed and lead to data that are not fully representative of a manufacturing process.

One especially important consideration is defining the appropriate worst-case scenario. Downstream purification teams do not always evaluate process conditions at their extremes, but VC studies are expected to demonstrate robustness under such conditions. A well-designed study defines its purpose clearly and ensures that tested parameters will support defensible conclusions, including potential process excursions.

Timing and preparation are critical to success. Study timing directly affects data relevance and development timelines. Studies conducted too early might rely on processes that are not yet representative of the final manufacturing workflow, resulting in data that are difficult to interpret or ultimately need to be confirmed with additional experiments. Conducted too late, VC studies can create timeline pressures and reduce flexibility if issues arise.

VC assessment also requires significant preparation, often taking several months to complete depending on a process’s development stage and regulatory requirements. Early planning for materials, virus stocks, controls, and testing strategy is essential to prevent delays and support successful execution.

Input materials and controls can influence outcomes significantly. VC-data reliability depends heavily on the quality and characterization of study inputs. Variability in virus stocks used in spiking studies can skew measured log reduction values (LRVs) and complicate interpretation.

Preliminary assessments, including toxicity and interference testing, are essential. In some cases, process materials might be unexpectedly toxic to indicator cell lines used in infectivity assays. Often, a spiking strategy can be modified to increase the total viral load within the starting material. Increased viral load with large-volume testing might overcome lost sensitivity from an unexpectedly high starting dilution. However, process materials still might require higher starting dilutions, reducing assay sensitivity. Addressing such issues early can help to preserve the achievable LRV and support meaningful results.

Assay selection influences conclusions. Analytical strategy is central to how VC performance is understood. Infectivity-based assays remain foundational, but they do not capture the full spectrum of particles present, particularly noninfectious particles and other process-related impurities. If a process step, such as protein A column chromatography, elutes at a low pH, then molecular methods are typically used for enveloped viruses. Doing so ensures differentiation of virus removal from the chromatography step from removal and additional inactivation that may be detected in infectivity assays.

Relying on a single method therefore can leave important gaps in interpretation. Integrating complementary analytical approaches provides a more complete understanding of process performance and helps to strengthen confidence in study conclusions.

Data context is as important as data generation. Even well-executed studies can be difficult to interpret without the right scientific context. Historical experience, platform knowledge, and prior studies all contribute to understanding whether results are expected, anomalous, or indicative of a broader issue.

Data do not exist in isolation, and the ability to interpret them appropriately can determine whether a study moves a program forward or triggers additional investigation. Experienced scientific input throughout the study life cycle — from initial design through final analysis — can help to identify potential issues early and reduce the likelihood of repeat testing and extended follow-up.

Common Pitfalls and Their Impacts

When the fundamentals of study design and execution do not align, the consequences tend to show up in a few ways. One of the most common pitfalls is misalignment between scaled-down study models and actual manufacturing processes. If parameters such as residence time, pH, or filtration conditions are not truly representative, then the resulting data might not accurately reflect process performance. That can raise regulatory questions and, in some cases, require additional studies to support the original conclusions.

Assay-related limitations also can affect interpretation. For example, in filtration steps that take place over extended periods, viral infectivity can decline due to process length and not solely because of nanofilter virus removal. If that loss of infectivity is not accounted for, then measured clearance might be overstated and reflect assay behavior rather than actual process performance.

I once witnessed a case in which a nanofiltration step conducted over several days raised concerns that tested viruses might not remain infectious throughout the study duration. Some viruses are not stable at room temperature and may lose infectivity due to the timing rather than the nanofiltration step alone. By incorporating additional controls and evaluating fractions collected throughout the process, it was possible to calculate accurate LRVs for each filtration stage rather than relying on a single endpoint control-sample measurement.

Another common pitfall involves misunderstanding which worst-case parameters to evaluate. Worst-case parameters for manufacturing excursions might not represent a worst-case scenario for viral contamination. For example, low-pH inactivation in VC studies typically are performed at relatively low temperatures to reduce additional sources of viral lipid envelope disruption beyond the pH condition alone.

Other challenges, including toxicity or assay interference, also can reduce confidence in generated data if not addressed proactively. Taken together, such issues can disrupt development programs, contributing to delays, repeat work, increased costs, and added regulatory scrutiny.

Enhancing Impurity Detection with Complementary Approaches

As VC strategies evolve, there is growing recognition that traditional assays might need to be supplemented with additional analytical methods to broaden impurity characterization across modalities.

Certain process-related impurities and particles are not captured fully by infectivity-based assays but still can influence viral-safety strategy and study design. In monoclonal antibody (mAb) manufacturing, such impurities include retrovirus-like particles (RVLPs). In virally delivered gene therapies, targets for additional characterization targets include empty capsids.

Transmission electron microscopy (TEM) can play an important role in characterization. By enabling direct visualization of particles and structural features, TEM provides an orthogonal approach for evaluating impurities that may not be understood fully through molecular- or infectivity-based methods alone.

Importantly, TEM generally is performed before VC studies when used for bulk harvest testing. The information generated is essential to informing overall viral-safety strategy, including assessment of potential impurities and calculation of total LRV requirements. Such applications are consistent with considerations outlined in regulatory guidance documents, including ICH Q5A(R2), European Pharmacopoeia Chapter 5.1.7, and EMA guidance related to gene-therapy medicinal products (1, 3, 4).

TEM’s “seeing is believing” capability can be particularly valuable when additional confirmation is needed regarding the presence or absence of specific particle types. When used alongside other analytical approaches, TEM can strengthen process understanding and support comprehensive impurity characterization strategies.

The Role of Data in Strengthening Confidence

Beyond individual assays, data strategy plays a critical role in VC studies. Well-curated datasets and thoughtful analytical frameworks can improve both study design and interpretation of results.

Historical data can help to establish expected performance ranges, identify sources of variability, and inform decision-making. When combined with study-specific results, that information enables teams to make proactive adjustments and better understand whether observed outcomes reflect true process behavior or analytical limitations. As recognized in ICH Q5A(R2) and related guidance documents, historical and in-house data also can support calculation and interpretation of LRVs when those data are well understood and appropriately assessed for comparability across products (1–3).

Integrating data from multiple analytical approaches also allows for cross-validation. When results from different methods align, confidence increases. When they diverge, further investigation can help to uncover the underlying cause. Experienced scientific oversight throughout the study life cycle can strengthen confidence by ensuring that data are interpreted appropriately and that potential issues are addressed early.

Regulatory Considerations and Expectations

Regulatory agencies expect VC studies to be scientifically rigorous, well controlled, and representative of a manufacturing process. Key areas of focus include model-virus selection, study-design relevance, analytical robustness, and justification of scaled-down models.

Among the most common areas of regulatory scrutiny is whether study conditions accurately reflect manufacturing parameters, including residence time, filtration throughput, and processing conditions. Thoroughly understanding a manufacturing process, including its expected variability, is essential to demonstrating that VC has been evaluated under meaningful conditions.

Knowing your process is particularly important when defining worst-case scenarios. For example, understanding the full operating range of a viral-inactivation step, including its maximum pH and duration, is critical to ensuring that studies adequately demonstrate clearance under the most challenging conditions. Similar worst-case considerations might apply to parameters such as filtration pressure, pauses during filtration runs, and conductivity conditions in ion-exchange chromatography steps.

Recent updates to ICH Q5A(R2) (1), along with evolving EMA guidance related to virus safety and advanced therapies (4), place increased emphasis on comprehensive risk assessments to accompany VC studies. Whereas a study provides evidence of viral safety, risk assessment supports the rationale behind study design decisions, including selection of process steps, controls, virus panel, and conditions. The revised guidance also recognizes the use of prior knowledge and platform approaches when supported by sufficient in-house data, process understanding, and appropriate comparability assessments. Clear documentation and scientific justification remain essential to meeting regulatory expectations and minimizing challenges during review.

Looking Ahead

As biologics and advanced therapies continue to evolve, VC studies must adapt to increasing complexity. Future approaches are likely to place greater emphasis on integrated analytical strategies and evolving regulatory expectations for impurity characterization across modalities.

Rather than relying on single-method approaches, studies increasingly will incorporate complementary techniques to enhance understanding of VC performance. At the same time, the importance of risk assessment, process understanding, and cross-functional expertise will continue to expand.

Ultimately, successful VC studies will balance scientific rigor with practical execution, leveraging experience, thoughtful design, and robust analytical strategies to deliver reliable and defensible results.

References

1 ICH Q5A(R2). Viral Safety Evaluation of Biotechnology Products Derived from Cell Lines of Human or Animal Origin. International Council on Harmonisation of Technical Requirements for Pharmaceuticals for Human Use: Geneva, Switzerland, 1 November 2023; https://database.ich.org/sites/default/files/ICH_Q5A(R2)_Guideline_2023_1101.pdf.

2 EMEA/CHMP/BWP/398498/2005. Guideline on Virus Safety Evaluation of Biotechnological Investigational Medicinal Products. European Medicines Agency: London, UK, 1 February 2009; https://www.ema.europa.eu/en/documents/scientific-guideline/guideline-virus-safety-evaluation-biotechnological-investigational-medicinal-products_en.pdf.

3 General Chapter 5.1.7. Viral Safety. European Pharmacopeia. European Directorate for the Quality of Medicines & HealthCare: Strasbourg, France, 2008.

4 Guidelines Relevant for Advanced Therapy Medicinal Products. European Medicines Agency: Amsterdam, the Netherlands, 2026; https://www.ema.europa.eu/en/human-regulatory-overview/advanced-therapy-medicinal-products-overview/guidelines-relevant-advanced-therapy-medicinal-products#genetically.

Kathleen Kenney is director of viral clearance at Minaris, 400 Rouse Boulevard, Philadelphia, PA 19112; https://minaris.com. Please email inquiries to the BPI editors.

This article was first published in a May 2026 eBook titled Process-Related Impurities — Making the Most of Purification and Analysis. Find the eBook at https://www.bioprocessintl.com/viral-clearance/ebook-process-related-impurities-making-the-most-of-purification-and-analysis.