Continuous Bioprocessing: why biopharma is moving beyond the batch
Integrated continuous bioprocessing is no longer a lab curiosity. It is a production strategy with its own architecture, its own failure modes, and its own regulatory pathway. The key is understanding where the real design decisions sit.
The pitch for continuous bioprocessing sounds straightforward: smaller footprint, higher volumetric productivity, and fewer vessels sitting idle between steps. The operational reality is more interesting. Continuous processing is not just a faster batch process. It is a different way of connecting upstream, downstream, and quality control into one coordinated system.
Replace multiple large stainless vessels for comparable output campaigns.
Sustains productivity that a fed-batch endpoint cannot match.
Surge tanks replace the hours-to-days pools typical of batch processing.
Purification trains run in motion instead of waiting for the next discrete batch.
The Fundamentals
An integrated continuous biomanufacturing platform pairs a perfusion cell culture with a downstream train that stays connected and intensified, rather than stopping and starting between isolated unit operations. In practice, teams usually choose between two architectures based on campaign length and how much of the downstream train stays continuous.
| Architecture | Typical Duration | Bioreactor Scale | Downstream Character |
|---|---|---|---|
| Hybrid continuous Early-stage |
12-15 days | 500 L or 1,000 L single-use | Continuous capture and polishing, often with one polishing step sufficient |
| Fully end-to-end continuous Late-stage |
25-30 days | 500 L or 1,000 L single-use | Continuous capture, polishing, viral filtration, and inline concentration/diafiltration |
Both architectures lean heavily on single-use technologies and are designed to flex across monoclonal antibodies, bispecifics, biosimilars, and next-generation biologics without a platform redesign every time a new molecule enters development.
The Comparison
The clearest way to see the shift is to compare the same downstream stages in batch and continuous mode.
| Stage | Batch | Continuous |
|---|---|---|
| Bioreactor | Fed-batch with a single harvest endpoint | Perfusion with steady-state or dynamic continuous harvest |
| In-process pools | 3-5 pools held for hours to days | Minimal pooling, with surge tanks measured in minutes |
| Capture cycling | Usually fewer than 10 cycles | More than 10 cycles, typically on multi-column skids |
| Polishing chromatography | 1-2 steps with fewer cycles | Two polishing steps with alternating loads and repeated cycling |
The Design Choice
Even after a team chooses “continuous,” one core decision still shapes the plant design: whether to place a discrete pool tank immediately after low-pH virus inactivation.
Process Design
Upstream still starts from a familiar seed train, but the transition to continuous typically begins at the N-1 or N stage. N-1 may run batch, fed-batch, or perfusion depending on facility fit; N-stage perfusion then runs at dynamic or steady-state conditions depending on the process. Across that train, DoE and PAT support development, monitoring, and control.
Upstream
- CSPR is optimized against a fixed VVD to maintain consistent growth.
- Cell retention typically runs through ATF or TFF systems.
- High-mannose glycosylation variability must be handled proactively, often through media optimization or pooling strategy.
Downstream
- Multi-column capture is sized around binding capacity, flow, and cycle time.
- Inline sensors detect breakthrough and support loading control.
- Continuous UF/DF can run as cycling TFF or single-pass UF/DF depending on maturity and scale-up confidence.
The Trade-Off
Continuous processing does not eliminate complexity. It relocates it into system integration, campaign control, and deviation handling. Four challenges show up almost every time:
Sterility and contamination
A contamination event can propagate across the full production run instead of staying isolated within one discrete batch.
Batch definition
Without a natural start and stop, batch definition becomes a deliberate control-strategy choice rather than an administrative detail.
Process robustness
Upsets in one unit operation can travel downstream quickly because the train has fewer natural buffers.
Upstream/downstream integration
The process now runs in lockstep, so slowdowns on one side immediately affect the other.
Managing Deviations
Because disruptions can propagate, continuous processes need a more structured deviation framework than batch operations. Critical process parameter excursions detected on-line, in-line, or at-line should move through a defined sequence before disposition is decided.
An excursion is identified against established limits.
On-line, in-line, or at-line systems confirm the event.
Root cause and product impact are evaluated together.
Continue, segregate an affected portion, or divert to waste with CAPA.
Example: Bioreactor pH excursion
If pH drops to 6.9 against a 7.0-7.2 limit for several hours, glycosylation may shift through cell stress. The decision then rests on actual glycosylation testing, not just the fact that the alarm occurred.
Example: Protein A overload
If an overloaded capture cycle raises concern for HCP or HMWS, the affected portion can be tested and segregated rather than automatically rejecting the full campaign.
The Regulatory Angle
Continuous manufacturing is not a gray zone anymore. ICH Q13 gives a harmonized framework for continuous manufacturing of drug substances and drug products, including expectations for batch or campaign definition, control strategy, and the role of PAT.
- Batch definition needs to be designed intentionally, not inherited by default from batch thinking.
- CPP deviation handling should be built into the filing strategy, not added later as an operational patch.
- PAT becomes part of the defensible control strategy, not just a nice-to-have monitoring layer.
How Graphtal Helps
Teams evaluating continuous platforms usually do not struggle with the headline benefits. They struggle with the practical decisions underneath them: hybrid versus fully end-to-end architecture, Option 1 versus Option 2 downstream design, PAT-integrated control strategy, batch definition, and CPP deviation logic that remains defensible under ICH Q13.
Evaluating continuous for your next program?
Graphtal helps map architecture, scale, PAT strategy, and quality-system readiness before capital is committed to the wrong process shape.
Tell Us About Your Project
Our team of experienced professionals at Graphtal is ready to transform your project from idea to reality, ensuring alignment with your organisation goals through advanced data analytics and predictive modelling.
Let's simplify your work