ICH Q5AICH Q5A(R2) — Viral Safety Evaluation of Biotechnology Products
A risk-based, three-part strategy — screen the cell line, test the unprocessed bulk, and prove the purification process clears virus with margin to spare — for every biotechnology product made in cell culture.

On this page · 9 sections
- 1What ICH Q5A(R2) covers
- 2Where viral contamination can enter
- 3Cell bank testing: MCB, WCB and the limit of in vitro cell age
- 4Testing the unprocessed bulk
- 5Viral clearance studies and the choice of model viruses
- 6The action plan for virus-positive cell lines and the platform concept
- 7What R2 changed: genetically engineered viral vectors (Annex 6)
- 8What R2 changed: continuous manufacturing
- 9Records the Q5A dossier depends on
How does ICH Q5A apply to your shop floor?
Pick your industry and scale — Ask V5 rewrites the definition in your context, gives a worked example, and shows what V5 does on day one.
01What ICH Q5A(R2) covers
ICH Q5A(R2), Viral Safety Evaluation of Biotechnology Products Derived from Cell Lines of Human or Animal Origin, was endorsed for public consultation at Step 2 on 29 September 2022 and adopted at Step 4 by the ICH Assembly's regulatory members on 1 November 2023, replacing the original Q5A text from 1997 and its 1999 editorial correction (R1). The final guideline describes how to evaluate the viral safety of biotechnology products made from cell lines of human or animal origin — mammalian, avian or insect — including what viral-safety data to include in a marketing application. FDA announced its adoption as final guidance for industry in January 2024, issued jointly by CDER and CBER, replacing the November 2022 draft.
The guideline's scope covers biotherapeutics and certain biological products made by recombinant DNA technology in cell culture, naming cytokines, monoclonal antibodies and subunit vaccines explicitly. R2 extended coverage to certain genetically engineered viral vectors and viral vector-derived products — for example AAV gene therapy vectors produced by transient or stable transfection or by infection with a recombinant virus, and vector-derived products such as baculovirus-expressed virus-like particles — provided they are amenable to viral clearance without harming the product; specific guidance for these is in the new Annex 6. Inactivated and live-attenuated viral vaccines containing self-replicating agents, and cell therapies, remain out of scope, though the guideline's principles may still apply to their biological starting or raw materials. The word "virus" in the guideline explicitly excludes nonconventional transmissible agents such as those associated with bovine spongiform encephalopathy; TSE issues are to be discussed separately with regulators.
The guideline frames viral contamination as a risk that has not historically been implicated in transmission from these cell-line-derived products, but one that could have serious clinical consequences if it occurred — so the control strategy is built from three complementary, overlapping approaches rather than any single test: selecting and testing cell lines and raw materials for freedom from undesirable infectious viruses, assessing the production process's capacity to clear adventitious and endogenous virus, and testing the product itself at appropriate production steps.
02Where viral contamination can enter
Section 2 of the guideline separates two origins. Contamination of the master cell bank (MCB) itself can arise from derivation of the cell line from an infected animal, use of a virus to establish the line, contaminated biological reagents or raw materials used in cell culturing (such as animal or human serum or porcine trypsin), or contamination during cell handling and banking. Cells also carry endogenous retroviruses passed from one generation to the next — constitutively expressed or latent — and may harbor latent or persistent viruses such as herpesviruses.
Adventitious contamination introduced during production can come from contaminated biological raw materials or reagents used in cell culture (for example animal serum components), a contaminated virus seed, a contaminated raw material or reagent used in downstream purification (such as a mammalian-derived affinity resin), a contaminated excipient added during formulation, or the manufacturing environment itself, including personnel handling non-biological raw materials or cell culture and medium. Monitoring routine cell-culture parameters such as growth and viability is identified as a useful early-warning signal for potential contamination, alongside the dedicated test program described below.
The guideline is explicit that no single test or approach can establish viral safety on its own: detecting low virus concentrations is limited by sample size for statistical reasons, so confidence that a final product is free of infectious virus comes from a combination of cell-line and raw-material screening, process-level clearance demonstration, and targeted testing — not from any one of those alone.
03Cell bank testing: MCB, WCB and the limit of in vitro cell age
Section 3 sets out virus testing recommended once on the Master Cell Bank (MCB), once on each Working Cell Bank (WCB) derived from it, and once on cells expanded to the Limit of In Vitro Cell Age (LIVCA) used for production — cells also referred to as End of Production Cells. Extensive screening for both endogenous and adventitious viruses is expected on the MCB, including broad and specific detection assays; for hetero-hybrid cell lines with a human or non-human-primate partner, testing should specifically look for viruses of human or primate origin because of the particular risk such cells can pose. An alternative in which complete testing is carried out on the WCB rather than the MCB is also acceptable.
The WCB is tested for adventitious viruses as well; where the MCB has already been tested and LIVCA cells derived from this WCB have also been tested, some duplicate WCB testing may be omitted with justification. LIVCA cells are evaluated once for viruses that may not have been detectable earlier, giving assurance that extended culture at production scale has not induced endogenous virus expression, reactivated latent virus, or amplified a low-level or slow-growing contaminant; any virus detected at this stage triggers a careful check back through the WCB and MCB to find the source.
Recommended assay categories include tests for retroviruses (infectivity assays, reverse-transcriptase activity assays such as the product-enhanced reverse transcriptase or PERT assay, and transmission electron microscopy), in vitro cell-culture infectivity assays using multiple indicator cell lines, in vivo assays where applicable, tests for specific named viruses based on the cell line's species of origin, and molecular methods including next-generation sequencing (NGS) for broad, non-targeted detection of known and novel viruses. The guideline notes it is no longer encouraged to manufacture from hybridoma cells grown in vivo as ascites, both for contamination risk and to support animal-use reduction, and that where this method is still used, NGS should replace the in vivo assay component.
The acceptability of a cell line that is found to contain viruses other than endogenous retroviruses is decided case by case by the relevant regulatory authority, weighing the product's benefit and intended use, the nature of the virus and its potential to infect or cause disease in humans, the demonstrated viral-clearance capability of the purification process, and the extent of testing performed on the purified bulk.
04Testing the unprocessed bulk
Section 4 recommends a routine program testing each production batch's unprocessed bulk for adventitious viruses — the pooled harvest(s) of cells and culture media in batch processing, or a pooled intermediate sample drawn from the flow stream for continuous harvest processes. A representative sample taken before further processing is identified as one of the most suitable points to detect adventitious contamination with high probability, because it reflects the material closest to the production reactor itself. The extent of testing is scaled to the cell line's nature, the cell-bank qualification results, the cultivation method, raw-material sourcing, and the demonstrated viral-clearance capacity of the downstream process.
The standard approach uses in vitro assays across several indicator cell lines or non-targeted NGS, observing indicator cultures for 28 days with at least one sub-passage at two weeks; this duration may be shortened to 14 days with a documented risk-based justification covering the cell substrate, production cultivation period, use of animal-derived raw materials, and the process's viral-clearance margin. Nucleic acid amplification testing (NAT) or targeted NGS can supplement this for specific viruses of concern, such as minute virus of mice, and can support faster, real-time release-type decisions where justified.
For continuous or perfusion processes, the guideline recognizes that cells may not be readily accessible — for example behind hollow-fibre or similar microfiltration systems — in which case cell-free product harvest can be tested instead, and the sampling strategy's periodicity and composition should be justified because adventitious- and endogenous-virus levels can vary across the duration of a continuous culture run.
06The action plan for virus-positive cell lines and the platform concept
Section 5 describes the rationale and action plan that applies when a cell line is known to produce endogenous retrovirus-like particles or other viral sequences, or when a specific virus is a known or suspected contaminant: the manufacturer tests the unprocessed bulk, conducts clearance studies that address that specific virus (or an appropriate model, such as a herpesvirus or pseudorabies virus standing in for a specific virus family), and may need to test the purified bulk directly, with the combined evidence forming the basis for the risk-benefit judgment regulators make about whether the cell line and process are acceptable for the intended product.
Section 6.6's platform concept lets a manufacturer rely on prior in-house viral-clearance data from a closely related product made on the same manufacturing platform — the same cell line type, similar process steps, equivalent buffers and resins — to support a conservative reduction factor claim for a new product's clearance step, rather than repeating a full dedicated study for every new molecule. Annex 5 gives illustrative examples of what qualifies as usable prior knowledge, for instance low pH viral inactivation or virus filtration steps that are well characterised across a platform. The guideline frames this as platform validation specifically for viral clearance — distinct from general process validation — and states that if a new product's characteristics or process diverge too far from the established platform, product-specific studies are still required. Over a product's lifecycle, a manufacturing change that could plausibly affect viral clearance (such as a change in an upstream process that alters the amount of virus produced by the cell line, or a change to a clearance-relevant step) should trigger a re-evaluation using either platform knowledge or a dedicated study.
07What R2 changed: genetically engineered viral vectors (Annex 6)
The most substantial scope expansion in R2 is the new Annex 6, covering genetically engineered viral vectors and viral vector-derived products — AAV gene therapy vectors (including those that depend on helper viruses such as herpes simplex virus or adenovirus for their production), other recombinant viral vectors, and vector-derived products such as baculovirus-expressed virus-like particles and nanoparticle-based protein vaccines. The guideline notes explicitly that some viral clearance steps effective for recombinant proteins may not work as well for these products, and calls for additional risk-reduction measures such as raw-material treatment and more extensive broad virus-detection testing. Viruses deliberately used to produce these products (production viruses, master and working virus seeds) are treated as process-related impurities to be characterised and controlled, with their own testing expectations for unprocessed bulk and virus seed introduction described in the annex.
This is a meaningful shift for cell and gene therapy manufacturers: a product category that could previously fall into regulatory ambiguity between Q5A's traditional recombinant-protein focus and cell-therapy exclusions now has an explicit, named framework describing how cell-bank and unprocessed-bulk testing, and viral clearance study design, should be adapted when the "virus" being evaluated is deliberately part of the product itself rather than purely a contaminant to be removed.
08What R2 changed: continuous manufacturing
R2 adds a new Section 7, Points to Consider for Continuous Manufacturing, which did not exist in the 1997/1999 text. It describes continuous manufacturing (CM) as continuous feeding of input materials through a series of linked unit operations producing a continuous output stream, referencing ICH Q13's description of CM modes for therapeutic proteins, and states that CM can apply to some or all unit operations in a process. The guideline is clear that the underlying science- and risk-based principles for controlling viral contamination are the same as for batch manufacturing, but several technical aspects differ: how viruses are detected and removed, material traceability through a continuous stream, process dynamics, and the frequency of monitoring around start-up and shutdown events.
Section 7.1 calls for a CM-specific viral safety strategy built on a risk assessment of contamination sources (including extended cell-culture duration), the process's demonstrated virus-removal capability, and testing capability, with the type and frequency of adventitious-virus testing during the run following from that assessment. Section 7.2 lists design considerations specific to CM: using scientific understanding from batch-process viral clearance data where suitable for integrated or continuous unit operations; assessing the risk at each unit operation and at connections between equipment (for example a surge or mixing tank smoothing differences in flow rate between steps) for any effect on virus-reduction capability; and putting process monitoring in place to detect both process disturbances that could compromise clearance (such as a pressure or flow interruption) and adventitious contamination itself, with a predefined procedure for deciding whether affected material should be diverted from the product stream if real-time decision-making is used.
09Records the Q5A dossier depends on
A Q5A-aligned viral safety package is, in practice, a set of linked records rather than a single document: the cell-bank testing history (MCB, each WCB, LIVCA/EOPC) with assay methods and results; the raw-material and reagent sourcing record for anything of human or animal origin used in cell culture or purification; unprocessed-bulk adventitious-virus test results tied to each production batch; the viral clearance study report for each relevant purification step, including the scale-down model's representativeness data and the calculated reduction factors; and, where the platform concept is invoked, the specific prior studies and products cited as supporting evidence. A marketing application's viral safety summary, which the guideline explicitly calls for, is only as defensible as the traceability between that summary and the underlying study and test records behind it.
Where a manufacturing change affects a clearance step, upstream cell-culture conditions, or raw-material sourcing, that change and its viral-safety impact assessment belong in the site's change-control record, linked to whichever cell-bank, unprocessed-bulk, or clearance-study records it affects, consistent with the lifecycle review described in Section 6.6. Generic records systems that keep raw-material certificates, batch genealogy, and document/change-control records linked with full audit trails — the kind of records infrastructure a platform like V5 Ultimate maintains with human review and approval at each step — support this traceability without being a substitute for the dedicated virological testing and clearance studies the guideline itself requires; the testing and study work is performed by qualified virology staff and reviewed by the quality unit, and system records exist to keep the resulting evidence linked, attributable and retrievable.
Frequently asked questions
Q.What is ICH Q5A(R2) and when did it take effect?+
ICH Q5A(R2), Viral Safety Evaluation of Biotechnology Products Derived from Cell Lines of Human or Animal Origin, was adopted at Step 4 by the ICH Assembly's regulatory members on 1 November 2023, replacing the original 1997 Q5A guideline. FDA issued it as final guidance for industry in January 2024.
Q.Does Q5A(R2) apply to cell and gene therapy products?+
Cell therapies remain out of scope, though the guideline's principles may still be applied to their biological starting or raw materials. Certain genetically engineered viral vectors and viral vector-derived products — including AAV gene therapy vectors and baculovirus-expressed virus-like particles — were newly brought into scope in R2, with dedicated guidance in the new Annex 6, provided they are amenable to viral clearance without harming the product.
Q.What is the difference between a relevant virus and a model virus in a clearance study?+
A relevant virus is the actual virus, or one closely related to it, known or suspected to contaminate the specific cell line or raw materials used. A model virus is chosen to represent a range of physicochemical properties so the process's general robustness against viruses in general can be characterised, even ones never detected in that cell line; the guideline recommends assessing clearance against at least three viruses with differing characteristics.
Q.How is a viral clearance reduction factor calculated?+
Per Annex 3, an individual step's reduction factor is the log10 of the ratio between the virus load (titre × volume) in the material entering a step and the virus load in the material leaving it. The overall process reduction factor is the sum of the logarithms of the individual step reduction factors. Because the scale is logarithmic, residual infectivity is reduced substantially but is never mathematically brought to exactly zero.
Q.What changed for continuous manufacturing in R2?+
R2 added a new Section 7 specifically addressing continuous manufacturing (CM), referencing ICH Q13's description of CM modes. It states the same science- and risk-based principles apply as in batch manufacturing, while calling for a CM-specific risk assessment, justified sampling strategy and frequency for a continuous flow stream, and process monitoring that can detect both process disturbances affecting clearance and adventitious contamination, with a predefined approach for deciding whether to divert affected material.
Q.Can a manufacturer reuse viral clearance data from another product?+
Yes, under the platform concept in Section 6.6, if the new product is made on a sufficiently similar manufacturing platform (same cell line type, comparable process steps, buffers and resins), prior in-house clearance data can support a conservative reduction factor claim without repeating a full dedicated study. If the new product or process diverges materially from that platform, product-specific viral clearance studies are still required.
Primary sources
Further reading
- ICH Q9 — Quality Risk ManagementThe risk-assessment framework Q5A's case-by-case judgments draw on.
- ICH Q10 — Pharmaceutical Quality SystemThe lifecycle quality system that governs how change control and records support a viral safety dossier.
- ICH Q12 — Lifecycle ManagementHow established conditions and change categories apply when a clearance-relevant process step changes.
- Process Validation (FDA 2011 Guidance)The three-stage lifecycle validation model that parallels Q5A's platform and lifecycle re-evaluation concepts.
- Change ControlWhere a manufacturing change affecting viral clearance gets documented and risk-assessed.
- Donor EligibilityA related viral-safety control point for human-origin starting materials outside Q5A's cell-line scope.
Want to see how ICH Q5A could fit into your own records and workflows? Explore the related V5 pages or talk to our team about what applies to your operation.
