The Stepwise Future of Continuous Viral Vector Manufacturing

Key Takeaways

Continuous viral vector manufacturing will likely advance incrementally, not through immediate end-to-end transformation. Near-term progress will center on upstream intensification, more frequent harvest, semi-continuous downstream processing, and tighter links between unit operations.

Lentiviral vector manufacturing highlights the importance of residence time, hold time, and functional recovery. Because LVs are labile, continuous and semi-continuous models must be designed to reduce exposure, processing delays, and recovery losses.

AAV manufacturing presents a different challenge: capsid quality and analytical resolution. Higher productivity only creates value if manufacturers can characterize and control full, empty, and intermediate capsid populations.

Process analytics and CMC strategy will determine how far continuous viral vector manufacturing can advance. Better-defined critical quality attributes, in-process testing, process analytical technology, and comparability strategies will be essential.

For CDMOs, continuous viral vector manufacturing creates a process-architecture opportunity. The strongest partners will help developers decide where intensification adds value, where batch processing remains appropriate, and how each change affects quality, recovery, and regulatory control.

Viral vector manufacturing has reached a point where the central question is no longer only whether vectors can be produced for clinical use but whether they can be produced with the scalability, consistency, efficiency, and control needed to support a maturing gene and cell therapy field. As demand for advanced therapies grows, the limitations of traditional fed-batch viral vector production have become more consequential. Variability, long processing times, and limited scalability continue to complicate efforts to move these therapies from highly specialized development programs toward broader clinical and commercial use.

That pressure has made continuous viral vector manufacturing an increasingly important frontier. However, “continuous” should not be treated as a single, fully realized manufacturing model that the field can simply adopt wholesale. For viral vectors, the more realistic near-term path is a spectrum of intensification and integration: perfusion-enabled upstream production, more frequent or continuous harvest, semi-continuous downstream processing, tighter links between unit operations, and eventually more complete end-to-end continuous systems. The principles described in ICH Q13 provide a useful framework for this transition, defining continuous manufacturing around the continuous feeding of input materials, transformation of in-process material, and concomitant removal of output material, while noting that these principles may also apply to biological and biotechnological entities beyond the guideline’s main focus on chemical entities and therapeutic proteins.¹

For viral vector manufacturing, that distinction matters because the next major advance is unlikely to come only from larger bioreactors or more batch campaigns. It will depend on the gradual integration of upstream production, downstream purification, and analytical control into more closed, responsive, and process-aware manufacturing models. That evolution is technically demanding because viral vectors are fragile, process-sensitive, and analytically complex. Cell health, vector stability, impurity clearance, product quality, and potency cannot be managed as isolated concerns; each one shapes whether a continuous or semi-continuous process can deliver usable product with the consistency required for clinical and commercial manufacturing.

The opportunity is substantial but still developing. The most credible story is one of staged progress: applying continuous-manufacturing logic first where it can address the most urgent constraints, while building the process understanding and control strategies needed for more integrated platforms over time.

Traditional fed-batch production remains a key reference point for viral vector manufacturing, but it brings well-recognized limitations around variability, processing time, and scalability. Those limitations matter more as gene and cell therapies move beyond early clinical proof-of-concept and toward broader development pipelines, more demanding commercial supply expectations, and, in some cases, larger target populations. A process model that may be workable for small, specialized campaigns can become harder to sustain when developers need consistent vector quality, predictable output, and efficient use of facility time.²

The pressure is not simply about making more material. Viral vector manufacturing must also manage biological and process variability that can affect productivity, impurity profiles, and product quality. As production systems scale, small differences in cell culture conditions, transfection or producer-cell performance, harvest timing, and downstream recovery can become more consequential. That makes process monitoring, in-process testing, and well-defined control strategies central to any effort to improve manufacturing reliability. Continuous and semi-continuous approaches are attractive in part because they can create more stable operating conditions, reduce disruptive process transitions, and support more frequent measurement and adjustment, provided the underlying analytics and controls are strong enough.

Viral vector manufacturing also lacks the same degree of platform standardization seen in monoclonal antibody manufacturing. The comparison is useful, but it should not be overstated. Monoclonal antibody (mAb) production benefited from years of platform convergence around expression systems, upstream processes, purification strategies, and analytical expectations. Viral vectors remain more fragmented. Adeno-associated viral vectors (AAVs), lentiviral vectors (LVs), and other vector systems differ in biology, production methods, stability, impurity concerns, and downstream requirements. Production-system choice, downstream processing optimization, and standardized chemistry, manufacturing, and controls (CMC) methods and quality assays remain major challenges for scaling viral vector manufacturing, underscoring that the field still needs more mature and transferable manufacturing logic.³

That platformization gap helps explain why continuous manufacturing is gaining attention. It is not only a way to increase throughput. It is a potential route toward more disciplined process control in a field where each program can otherwise become highly bespoke. Still, the first wave of progress is likely to look less like a universal viral-vector platform and more like selective standardization: reusable approaches to perfusion, harvest, clarification, capture, impurity control, and analytical monitoring that can be adapted to the specific vector, product, and clinical application.

Continuous viral vector manufacturing cannot be created by transferring a conventional continuous biologics template onto a different product class. Viral vectors bring a tightly coupled set of biological, physical, analytical, and regulatory constraints: producer cells must remain viable and productive, vectors must remain functional, impurities must be removed, and quality data must be generated quickly enough to support process control.

ICH Q13 provides a useful framework for understanding that challenge. The guideline focuses on integrated continuous manufacturing systems in which two or more unit operations are directly connected, while also recognizing that continuous principles can apply to individual unit operations, including perfusion cell culture and process chromatography.¹ For viral vectors, that distinction matters because a process may be continuous in one part of the workflow while still relying on batch steps elsewhere. Upstream perfusion can extend production and enable more frequent harvest, but it does not by itself solve downstream clarification, nuclease treatment, concentration, capture, polishing, formulation, or release testing.

The upstream challenge begins with sustaining viable, productive cells over an extended operating window. Perfusion-based approaches present a way to continuously add nutrients and remove waste, allowing cells to remain productive longer than in batch processes, but that advantage depends on maintaining process robustness across the full production period.² At the same time, viral vectors must remain functional as they move through harvest, hold, clarification, purification, and formulation. If downstream processing cannot keep pace, the process may trade one bottleneck for another while increasing the risk of vector degradation or loss of activity.

Downstream integration is therefore central. Continuous upstream production has limited value if vector-containing harvest is pooled, frozen, or held for batch downstream processing in ways that undermine the benefits of a more stable upstream operation. The practical frontier is the connection between production and purification: linking harvest, clarification, nuclease treatment, concentration, capture, and polishing to reduce hold times, protect product quality, and maintain impurity clearance. Purification and formulation must be part of the required integration for continuous viral vector manufacturing, and connected unit operations must be managed within an overall control strategy.^1,2

Analytics may be the hardest constraint to resolve. Continuous manufacturing depends on timely information about whether the process remains in control, but viral vector quality can be difficult to measure quickly. The U.S. Food and Drug Administration’s (FDA) gene therapy CMC guidance emphasizes the need to understand critical quality attributes (CQAs), process controls, process consistency, stability, impurities, and potency assays for gene therapy products.⁴ In a continuous or semi-continuous setting, the process must not only produce vector but also generate enough reliable information to support decisions about material quality, process performance, and product comparability over time.

The most immediate entry point for continuous viral vector manufacturing is upstream intensification. Before the field can achieve fully integrated continuous production, it needs upstream systems that can keep cells productive for longer, generate vector more consistently, and feed downstream operations in a more controlled manner. Perfusion is attractive because nutrients can be supplied continuously, waste can be removed, and cultures can remain productive beyond a conventional batch window.²

For viral vector production, however, a longer operating window is valuable only if it improves both productivity and control. A perfusion process that generates more harvest volume but introduces new variability, product instability, or downstream overload would not solve the underlying manufacturing problem. The goal is to create a more stable upstream environment that supports predictable vector generation, more frequent harvest, and tighter linkage to downstream purification.

For LV manufacturing, stable producer cell lines offer one of the clearest upstream strategies for moving toward continuous production. Unlike transient systems that require repeated introduction of plasmid DNA and transfection reagents, stable producer lines can support longer production periods and may improve batch-to-batch reproducibility. One recent study demonstrated continuous LV manufacturing using a stable producer cell line in an iCELLis Nano fixed-bed bioreactor, highlighting the potential for producer-cell systems to support extended vector generation without repeated transient transfection events.⁵

Recombinant AAV (rAAV) manufacturing may follow a different route. Transient transfection remains important, but recent work shows that it can also be adapted into intensified perfusion formats rather than simply replaced. A recent study described perfusion-based rAAV production using an intensified transient transfection process in which tangential flow filtration allowed rAAV to pass through while retaining cells in the bioreactor.6 Multiple time-separated plasmid doses at high cell density improved productivity and plasmid utilization compared with a single bolus under standard conditions.

That example complicates a simple replacement narrative: transient transfection may not disappear, but it may be re-engineered around perfusion, higher cell densities, and staged plasmid addition. For rAAV, upstream intensification can evolve through multiple routes, depending on the vector system, production platform, and maturity of the cell-line technology.

Upstream gains still need downstream support. A process that keeps cells productive for longer will generate more vector-containing harvest over time, and that material must be clarified, treated, concentrated, captured, and purified without eroding the quality or efficiency gained upstream. Perfusion can create the conditions for continuous viral vector manufacturing, but it also raises the stakes for downstream integration.

For continuous viral vector manufacturing, stability is not only a formulation issue addressed at the end of the process. It shapes the design of the entire manufacturing train. If vectors lose infectivity, potency, or structural integrity during production, harvest, hold, clarification, or purification, a longer or more connected process will not necessarily improve manufacturing performance. Continuous operation must therefore reduce unnecessary exposure, shorten vulnerable hold steps, limit process stress, and move vector-containing material efficiently into downstream conditions that preserve quality.

This issue is especially clear for LVs, which are inherently labile, lose function over time, and are sensitive to environmental factors across bioprocess unit operations. Residence time and hold time are therefore core process-design variables, not merely operational details. Because reducing processing and hold times can help preserve functional vector particles, LV lability provides a strong rationale for tighter integration between upstream production and downstream processing.⁷

The same principle appears in continuous fixed-bed LV production. In continuous viral vector manufacturing, flow rate, harvest frequency, and exposure time are not only productivity levers; they also affect how long the vector remains in conditions that may reduce functional recovery.

AAV manufacturing raises a related but distinct quality challenge. The verified source base supports a strong analytical-control argument rather than a broad claim about AAV lability. For AAV products, product quality depends in part on the ability to identify, quantify, and control full, intermediate, and empty capsid populations. Intermediate and empty capsid impurities should be controlled in clinical preparations, and intermediate capsids can be technically difficult to separate from full capsids.⁸ In an intensified or continuous process, productivity cannot be assessed only by total vector output; the process must also preserve the ability to characterize and control the vector population being produced.

Stability is one reason continuous viral vector manufacturing will likely advance through carefully designed hybrid systems rather than a simple move from batch to fully continuous operation. For LVs, that means minimizing time and conditions that reduce functional titer. For AAV, it means ensuring that intensification does not outpace the ability to resolve and control capsid quality. In both cases, the value of continuous manufacturing depends on whether the process can protect the product while improving productivity.

The next stage of continuous viral vector manufacturing will be defined less by perfusion alone than by the ability to connect upstream production with downstream purification. Longer production windows and more frequent harvest can improve upstream productivity, particularly for labile products, such as LVs, but those gains only matter if vector-containing harvest can move efficiently into downstream operations without new hold steps, freeze–thaw cycles, or recovery losses.

This distinction is important because continuous viral vector manufacturing has often referred mainly to perfusion-based upstream production, while downstream processing remains batch-based. In that model, harvest may be pooled and frozen every 24 hours before sequential downstream processing. For LV manufacturing, where functional activity can decline over time and across unit operations, upstream-only continuity can simply shift the bottleneck rather than remove it.⁷

A more integrated model requires downstream steps that can keep pace with production. One recent study provides a useful proof of concept, connecting perfusion-based upstream LV production with semi-continuous downstream processing, including nuclease treatment, clarification, and capture.⁷ By combining clarification and capture loading while operating those steps in parallel with capture purification, the process reduced processing and hold times, which are especially important for preserving functional LV particles.

The recovery data make the case more concrete. In the semi-continuous process, recoveries of functional vector particles and total vector particles improved by 26% and 18%, respectively, compared with batch-mode processing of the same volume using the same membrane size. Those results show that downstream integration can directly affect how much usable vector is retained from a production run.

Filtration and chromatography are likely to be central enabling operations. Tangential-flow filtration and chromatography may be critical to continuous viral vector manufacturing when adapted for continuous-flow operation. These technologies address core downstream needs, including removal of cellular debris, spent-media components, host-cell impurities, residual nucleic acids, and other contaminants while retaining product quality.² For now, the strongest claim is not that a universal downstream platform has emerged but that filtration and chromatography are becoming the operational bridge between intensified upstream production and more integrated manufacturing.

The most credible near-term model is therefore semi-continuous rather than fully continuous. Upstream perfusion, more frequent harvest, nuclease treatment, clarification, capture, tangential flow filtration, and chromatography can be linked more tightly over time, but each connection must be justified by improvements in recovery, process control, impurity clearance, and product quality.

AAV manufacturing brings a different set of constraints than LV manufacturing. If LV production highlights the importance of vector lability and functional recovery, AAV production highlights product heterogeneity. Higher vector output is valuable only if the process can also support consistent product quality, analytical resolution, and control of clinically relevant capsid populations.

AAV manufacturing continues to face process variability, low yields, and scalability challenges, making intensified production attractive. Producer cell lines may play an important role in addressing those constraints, particularly if they can reduce or eliminate reliance on plasmids, transfection, or infection.9 Still, greater upstream efficiency only matters if the resulting material can be characterized and controlled.

Capsid quality is central to that challenge. AAV preparations can include full, empty, and intermediate capsids, and intermediate and empty capsid impurities must be identified, quantified, and controlled in clinical preparations.8 Intermediate capsids can also be technically difficult to separate from full capsids, creating a substantial analytical burden for any intensified or continuous process.

For that reason, continuous or intensified AAV manufacturing cannot be evaluated only by productivity metrics. Higher output must be paired with analytical methods and control strategies capable of detecting meaningful product-quality differences.

The AAV-specific frontier is therefore defined by control as much as output. Continuous or intensified production will only advance the field if it improves the relationship between upstream productivity, downstream separation, and analytical confidence.

Continuous viral vector manufacturing will depend as much on regulatory-quality process understanding as on engineering innovation. More integrated processes may reduce hold times, improve efficiency, and support more consistent operation, but those advantages only matter if manufacturers can show how the process is controlled and how product quality is maintained. For gene therapy products, CMC expectations define what must be understood, monitored, justified, and documented as development advances.

That begins with CQAs. The FDA’s gene therapy CMC guidance is intended to help sponsors provide sufficient information to assure the safety, identity, quality, purity, and strength, including potency, of investigational gene therapy products. The guidance recommends identifying potential CQAs relevant to product safety and biological activity while recognizing that product characteristics tied to clinical performance can be difficult to define early in development.⁴ For continuous and semi-continuous processes, that uncertainty creates a clear need to understand how upstream conditions, harvest timing, downstream processing, impurity clearance, and formulation affect the attributes that matter.

The guidance also highlights the process information needed to support that understanding. Manufacturing descriptions should include process flow diagrams, detailed narratives, process controls, and in-process testing, with examples including titer, bioburden, viability, and impurities. Those measures map directly onto continuous vector manufacturing: titer tracks productivity, viability reflects production-system health, bioburden supports contamination control, and impurity testing helps demonstrate that process- and product-related contaminants are being removed or controlled.

ICH Q13 adds a continuous-manufacturing control strategy layer, suggesting that process monitoring and control support a state of control during continuous production and enable real-time evaluation of system performance, with PAT well suited to detect disturbances in real time during continuous manufacturing.¹ For viral vectors, this does not mean every critical attribute can already be measured in real time, but it does raise the importance of frequent, meaningful, and connected process information.

Comparability will also be central. Well-established CQAs are generally necessary for demonstrating product comparability by analytical methods when manufacturing changes are made during development.⁴ A move from batch or fed-batch processing to intensified, semi-continuous, or continuous manufacturing would represent a meaningful process evolution, requiring developers to show not only that the new process is more efficient but also that it supports the intended quality, safety, and biological activity profile.

Guidance from the European Medicines Agency (EMA) reinforces the same broader principle: gene therapy medicinal products require attention to quality, vector design, manufacture, characterization, and testing, even when the manufacturing model does not fit neatly into traditional product categories.¹⁰ Continuous viral vector manufacturing will therefore need to pair process intensification with CQA definition, analytical capability, process monitoring, and life cycle control.

What Continuous Will Likely Mean in the Near Term

The most realistic near-term future for continuous viral vector manufacturing is hybrid and application-specific. Continuous manufacturing principles are being applied first where they can address defined constraints: upstream productivity, harvest timing, downstream hold times, product recovery, and process monitoring. That staged model is more consistent with the current evidence than a direct shift from batch or fed-batch production into fully integrated end-to-end continuous platforms.

Upstream intensification will likely remain the most accessible entry point. Perfusion, high-density cultures, extended production windows, and more frequent harvest strategies can help address some limitations of traditional production formats.

The next stage is closer integration between production and downstream processing. The term “continuous” must be used carefully in this field: it has often referred mainly to continuous upstream perfusion, while downstream processing still occurs in batch mode after harvests are pooled and frozen. A semi-continuous downstream approach, linking perfusion upstream production with nuclease treatment, clarification, and capture, represents a practical intermediate stage between upstream-only intensification and fully integrated continuous manufacturing.⁷

Stronger analytical and control systems will determine how far the field can move from that intermediate stage. For viral vectors, progress will depend on better-defined CQAs, more frequent monitoring, process models, and PAT tools capable of supporting decisions during production.

More complete end-to-end continuous platforms remain a longer-term objective. The near-term opportunity is to identify where continuity, semi-continuity, and integration can remove specific bottlenecks without compromising vector quality or regulatory control.

Continuous viral vector manufacturing will not be adopted through equipment upgrades alone. It will require process architectures that connect upstream production, downstream purification, analytics, documentation, and regulatory control strategy. That makes the transition highly relevant for contract development and manufacturing organizations (CDMOs), because many developers will need partners that can help determine which parts of the process should be intensified, which should remain batch-based, and how each change affects productivity, stability, recovery, impurity clearance, and product quality.

These needs align closely with current viral vector manufacturing challenges. Production-system choice, downstream processing optimization, and standardized CMC methods and quality assays remain major hurdles for scaling viral vector production.³ Decisions about adherent versus suspension culture, transient transfection versus producer-cell systems, or perfusion versus fed-batch production can affect downstream load, impurity profiles, analytical strategy, facility fit, and regulatory comparability. As continuous or semi-continuous approaches enter the field, those dependencies become even more important.

For CDMOs, this creates a shift from capacity provider to process architect. The strongest partners will help design manufacturing strategies that balance output with control. A higher-yield upstream process has limited value if it generates harvest volumes that downstream operations cannot process efficiently, increases hold-time risk, or complicates impurity clearance. Similarly, a more integrated downstream train only adds value if it protects functional vector recovery and supports a defensible control strategy.

Analytics may become one of the most important differentiators. Continuous and semi-continuous manufacturing require more frequent, process-relevant information than conventional batch models, and process monitoring, process control, and PAT will be essential to support a state of control during continuous production and detect disturbances more quickly. For viral vectors, partners with strong analytical development, assay qualification, in-process testing, and data interpretation capabilities may be better positioned to support intensified platforms.

Regulatory fluency is closely tied to that analytical capability. The FDA expects innovators to describe manufacturing processes with sufficient detail, including process flow diagrams, narratives, process controls, and in-process testing, while identifying potential CQAs relevant to safety and biological activity.⁴ As processes move from batch to intensified or semi-continuous formats, developers will need to explain how those changes affect product quality and how comparability will be assessed. CDMOs that understand the documentation burden, not just the technical workflow, can help sponsors treat process intensification as both an operational improvement and a regulatory life cycle event.

The platformization opportunity should also be framed carefully. Viral vectors are unlikely to converge around one universal manufacturing platform in the way mAbs did. LV, AAV, and other vector systems differ in production biology, stability, impurity concerns, and analytical requirements. The more realistic opportunity is modality-specific platformization: reusable process logic that can be adapted across related programs without pretending that all vectors can be manufactured the same way.

Continuous viral vector manufacturing is approaching through specific gains in upstream intensification, producer-cell technologies, downstream integration, and process control. These advances make the topic highly relevant for gene therapy developers and their manufacturing partners, but they also show why the transition will be gradual. The field is not moving toward continuous manufacturing because continuity is inherently valuable. It is moving in that direction because current processes need better scalability, shorter vulnerable hold steps, stronger recovery, more consistent quality, and more defensible control strategies.

The central test will be whether continuous and semi-continuous approaches can improve manufacturing performance without compromising product quality. For LVs, that means designing around vector lability and functional recovery. For AAVs, it means ensuring that productivity gains do not outpace the ability to characterize and control capsid quality. Across vector types, continuous manufacturing only creates value if it improves the relationship between production efficiency and product control.

The next phase of progress will likely come from selective integration rather than wholesale process replacement. Upstream perfusion can extend production windows. Semi-continuous downstream processing can reduce hold times and improve recovery. Better-defined CQAs, stronger in-process testing, PAT, and process models can help manufacturers understand whether the system remains in control. Over time, these pieces may converge into more complete continuous platforms, but near-term progress will remain hybrid, staged, and application-specific.

For developers, CDMOs, and technology providers, the competitive advantage will belong not simply to those who can produce more vector, but to those who can design, monitor, and control more integrated vector processes. The future of viral vector manufacturing will likely be continuous by degrees: not a single leap from batch to fully integrated platforms, but a progressive tightening of the links between cell culture, harvest, purification, analytics, and control.

1. Q13 Continuous Manufacturing of Drug Substances and Drug Products: Guidance for Industry. U.S. Food and Drug Administration. Mar. 2023.

2. Dream, Robert, and Jeffery Odum. “Continuous Viral Vector Manufacturing.” ISPE. May/Jun 2025.

3. Capra, Emily, et al. “Viral-vector therapies at scale: Today’s challenges and future opportunities.” McKinsey & Company. 29 Mar. 2022.

4. Chemistry, Manufacturing, and Control Information for Human Gene Therapy Investigational New Drug Applications; Guidance for Industry. U.S. Food and Drug Administration. Jan. 2020.

5. Stibbs, Dale J, et al. “Continuous manufacturing of lentiviral vectors using a stable producer cell line in a fixed-bed bioreactor.” Mol. Ther. Methods Clin. Dev. 32: 101209 (2024).

6. Nguyen, Tam NT, et al. “Perfusion-Based Production of rAAV via an Intensified Transient Transfection Process.” Biotechnol. Bioeng. 122: 1424–1440 (2025).

7. Tran, Michelle Yen, et al. “Integrated Semi-Continuous Manufacturing of Lentiviral Vectors Using a HEK-293 Producer Cell Line.” Processes. 11: 3347 (2023).

8. McColl-Carboni, Aisleen, et al. “Analytical characterization of full, intermediate, and empty AAV capsids.” Gene Therapy. 31: 285–294 (2024).

9. Kowshik, N Charan SS, and Pushpendra Singh. “Advancing AAV vector manufacturing: challenges, innovations, and future directions for gene therapy.” Front. Mol. Med. Sec. Gene and Virotherapy. 5: 1709095 (2025).

10. Quality, preclinical and clinical aspects of gene therapy medicinal products – Scientific guideline. European Medicines Agency. 13 Jul. 2018.