Explore available options to select a path based on molecule and facility needs


Large biotech, emerging biotech, and CMOs alike are increasing investment in novel bioprocessing approaches to achieve success in the rapidly evolving biotherapeutic market. Addressing diverse molecule pipelines and uncertain demand projections depending on clinical outcomes, requires that bioprocessing approaches increase manufacturing supply chain flexibility and utilization. For example, novel buffer management and semi-continuous approaches are becoming commonplace as they represent powerful levers to increase facility utilization—a key driver of cost.

With many bioprocessing approaches to choose from, drug manufacturers select a path based on defined outcomes that are aligned with their specific molecule pipeline and/or current manufacturing capabilities. To illustrate this, consider how decisions will be affected given the following pipelines: complex molecules (bispecific antibodies, enzymes, or viral vectors); high volume demand (e.g. large populations and /or large dose) therapies; or biosimilars (cost pressures driving down CAPEX and, potentially, cost of goods sold (COGS). How would these pipelines affect global supply chain and process considerations? How do their needs for flexibility, speed, and costs differ? This article will attempt to address these questions, because understanding available options and how to best deploy them in the right circumstances will be crucial for successful outcomes.

Increasing process productivity is one of the most effective ways to improve facility utilization, which is largely synonymous with flexibility. For example, high yielding upstream processes can produce the same amount of drug product in shorter periods of time or reduce footprint by being implemented in smaller bioreactors. Additional flexibility, in this instance, occurs because the same pilot plant setup can be implemented at production scale, thereby reducing scale-up activities and enabling a scale-out mentality. Furthermore, it is easier to transfer smaller processes to new geographies, which is necessary to establish and maintain a global manufacturing network—a key for establishing a global presence.

Some approaches, however, can adversely affect flexibility and require greater analysis in context of the overall business strategy. For example, while operating in upstream perfusion mode the process times are prolonged, which can have considerable impact on a facility’s flexibility. There are some preliminary studies indicating that the impact of a perfusion process on the COGS of the product is strongly dependent on both the total yearly production output or plant output and the subsequent downstream process.1, 2, 3 Therefore, a complete business analysis and proposal for a continuous operation up- and downstream needs to be compiled.


So, what are next generation processes? These processes overarch the whole manufacturing chain from cell to molecule, upstream to downstream. They include semi-continuous and continuous approaches, focus on operational excellence, implementation of disruptive technologies up- and downstream or supplier partnership. Here, we will center the discussion around two large concepts:


  1. Continuous manufacturing: Continuous unit operations with high volumetric productivities, see Table 1 below.
  2. Process intensification for eliminating existing bottlenecks: often refers to the shortening of process times or reducing non-value-added process steps. In upstream processing, this can be exemplified by intensified seed train (see example in Figure 1) or intensified production bioreactor operations (see example in Figure 2). In downstream processing, it can be intensified capture, in-line virus inactivation, flow through polishing, in-line virus filtration or single-pass filtration.


While not covered in this article, it is important to note that success in both categories will be amplified by complementary advancements in automation, e.g. digitalization, in-line analytics, sensors, and in-line process control. Other strategies underpinning successful deployment include raw material controls, supply chain security, supplier partnership and single-use technology.

Today’s processes are often a hybrid of continuous and batch processes. Continuous upstream processes have been long established, and therefore represent lower risk for GMP implementation. This is in contrast with downstream operations, where regulatory concerns and understanding of critical process parameters for adequate control remain to be understood. We anticipate more adoption of continuous upstream approaches to take advantage of the opportunity to create a greater impact in short-term development. These approaches will be the focus in the remainder of this article.


The seed train presents several process intensification opportunities. One possibility is to increase the cell density (from 20 to 100 E6 cells per mL) and the volume (from 1 to 500 mL) during cryopreservation, called high density cell banking. Another is to use perfusion in the seed bioreactors to eliminate processing time from subsequent steps. An application of both approaches for seed train intensification is shown in Figure 1.

Several strategies may also be applied to production bioreactors for improving cell culture outcomes. Inoculating at a higher density (through perfusion in the N-1 step, e.g. resulting in inoculation densities of 10 E6 cells/mL instead of traditionally 0.5 E6 cells/mL), can shorten the time in the production bioreactor by about five days.4 Other modes of operation include intensified fed-batch, steady state perfusion, dynamic perfusion, hybrid perfusion/ fed-batch. An illustration of some of these modes of operation is given in Figure 2 and the impact on the volumetric productivity is given in Table 1.

Perfusion mode presents the highest potential to increase volumetric productivities. However, only a small percentage of worldwide bioreactors are operated this way today. One reason for this is that many parameters need to be optimized, e.g. medium composition, steady state cell density, strategies for cell separation, bleed rate, process automation, cell line stability and product quality attributes, resulting in long process development time. Table 2 gives an overview what to consider for these parameters. Taken together, the perfusion parameters with the most impact on process economy and productivities are cell line productivity and the medium composition required to support high productivity and low media consumption.

Furthermore, studies from mAb processes show how the perfusion rate and media composition have the potential to be used as a tuning fork for product quality.5 Through the growth of the biosimilar market a need has emerged for better and tighter control of product quality. Consistent quality (or more stringent quality) has traditionally been demonstrated for the production of unstable proteins like coagulations factors in perfusion mode. Production at steady state has the promise of a more consistent product quality.

Biopharmaceutical companies adding more molecules to a facility or facing increased volumetric demands may choose to increase productivity and throughput using their existing infrastructure. For example, cell culture suites can be modified for concentrated fed-batch (CFB) operations using existing bioreactor platforms. However, for new classes of biologics and when designing new platform processes and creating a global manufacturing network, perfusion may be the better option. In order to determine which approach fits the pipeline and facility, it is important to analyze both current and future bottlenecks in the manufacturing plant, and how these can be mitigated, e.g. medium and buffer handling, in-line buffer solutions and raw materials. Single-use technology is an important facilitator for these intensified processes. To unlock the potential of single-use technology, a partnership between suppliers and users is recommended. Already, today, many companies have engineers and technicians working directly with single-use vendors. This emphasizes the importance of partnership between suppliers and companies. In Table 3 we have summarized potential bottlenecks for the development of next generation processes and outlined and discussed mitigation strategies to navigate in this jungle of possible solutions.


When does it make sense to intensify your process and/or deploy continuous steps? With your molecule pipeline and facility/business strategy understood, we propose a situational analysis with respect to:


  1. The mode of operation that is suitable for your plant (see Table 1)
  2. The bottlenecks and key parameters for your processes, but also the critical product quality attributes
  3. The business drivers for each mode (time, costs, pro/cons)


The choices may appear complicated, but here we have presented considerations that will aide in determining the right process and facility decisions when transitioning into next generation bioprocessing approaches. Productivity and flexibility increases can be achieved when the right tools for single-use technology, hardware, automation, process design, cell line and media development are combined. For many, it is helpful to engage a knowledgeable partner to discuss and explore the range of options. A strategic partner that understands the process application and challenges of successfully implementing next generation bioprocessing, and has a successful track record of customer collaborations, can facilitate finding the optimal solution to fit their situation.

Tables and Figures:

Figure 1: Comparison of a traditional seed train process with an intensified process. The traditional process is starting with a standard cell bank vial containing about 25 E6 cells and multiple seed stages before the 2K production bioreactor. The intensified seed train process is inoculated from a high volume, high density vial containing about 225 E6 cells directly to a WAVE bioreactor. That single seed train stage, is used both to expand the volume and reach high cell densities in perfusion to inoculate directly the 2K production bioreactor.6

Figure 2: Process options for intensification of the production bioreactor: classic perfusion (A), intensified fed-batch (B) and hybrid process (C).

Table 1: Impact on volumetric productivities based on choice of process mode
Mode of operation Cell densities Approximate volumetric productivities (g/L/d)
Traditional fed-batch 20–50 0,3
High seed fed-batch 20–50 0,5
Concentrated fed-batch 100–200 1–2
Hybrid perfusion/ fed-batch 50–100 1–2
Dynamic perfusion 50–100 1–2
Steady state perfusion 30–50 1–3
Table 2: Key points to consider in a perfusion process
Parameter Considerations
Medium composition, cell specific perfusion rate (CSPR) Target a low volumetric perfusion rate to minimize media consumption and improve process economy. This can be achieved by medium optimization based on a DOE setup or spent medium analysis.
Cell specific productivity, qP A cell line with high productivity during the whole process is the key to good process economy and low COGS.
Bleed rate Bleeding means product loss, but is critical for the health of the culture and maintaining a high qP throughout the whole process. An optimized bleeding strategy needs to be investigated.
Steady state cell density High volumetric productivity should be weighed against the impact on media consumption (volumetric perfusion rate), supporting process conditions (bioreactor) and load on the cell separation device.
Cell separation Many robust separation devices are available, mostly filtration based. Performance is dependent on cell density, the viability of the culture, and the rate of cell lysis.
Automation Operating high density processes at steady state demands automation to acquire good process economy, e.g. automatic bleeding based on inline VCD sensor. In a continuous set-up, there is additional need for feed-back and feed-forward control between unit operations.
Cell line stability Due to high cell densities and long process times, perfusion processes require increased cell line stability. Calculations reveal that at least 20 additional stable number of generations is needed compared to a fed/batch process.
Quality attributes There is a need to understand the critical quality attributes for a product and how these are affected through a perfusion process set-up.
Table 3: Decision matrix to address bottlenecks and develop next generation processes
Technology solution Pros Cons Comment
Bottleneck: limited facility foot print
Perfusion High volumetric productivity Operational complexity Needs more process development than fed-batch
Bottleneck: limited CAPEX
Perfusion Smaller bioreactor Operational complexity Needs more process development than fed-batch
Single use Lower investment cost Supplier dependency  
Bottleneck: limited capacity in production bioreactor
N-1 perfusion Time for cell expansion is moved to the N-1, shorter time in N with maintained titer Operational complexity, possible impact on process performance Make sure you have capacity for the shorter turnaround time
Perfusion High volumetric productivity Operational complexity Identify a separation technique that will suit your process
Single use Fast turnaround time Waste management
Supplier partnerships
Bottleneck: QC/QA release
On-line sensors and PAT implementations Reduce the number of release methods through tighter process control Resource demanding to develop E.g. spectral methods in combination with multi variant analysis can offer several process parameters or golden batch approaches
Multi-attribute release methods Drastic reduction of the number of release methods Resource demanding to develop Still in development
Bottleneck: low yield process
Perfusion High volumetric productivity Operational complexity Identify a separation technique that will suit your process
Concentrated fed-batch High volumetric productivity Operational complexity  
Bottleneck: several product and processes
Perfusion Higher level of flexibility Operational complexity Scale out depending on batch volume needed
Single use Control cross contamination Waste management
Supplier partnership


Reproduced by permission of Genetic Engineering News.

Authors: Andreas Castan and Helena Öhrvik are R&D Scientists in BioProcess R&D; Daniel Nelson is Sr. Product Marketing Leader, Upstream, Cytiva.


  1. Pollock J. et al., Fed-batch and perfusion culture processes: economic, environmental, and operational feasibility under uncertainty, Biotechnol. Bioeng. 110, 206–219 (2013)
  2. Pollock J. et al., Optimising the design and operation of semi-continuous affinity chromatography for clinical and commercial manufacture, J. Chromatogr. A, 1284, 17-27 (2013)
  3. Walther J. et al., The business impact of an integrated continuous biomanufacturing platform for recombinant protein production, J. Biotechnol., 213, 3-12 (2015)
  4. Yang W. C. et al. Perfusion seed cultures improve biopharmaceutical fed-batch production capacity and product quality. Biotechnol. Prog. 30(3), 616-625 (2014)
  5. Walther J. et al., Perfusion Cell Culture Decreases Process and Product Heterogeneity in a Head-to-Head Comparison with Fed-Batch, J. Biotechnol., doi.org/10.1002/biot.201700733 (2018)
  6. Cytiva, Application Note, “One-Step Seed Culture Expansion from One Vial of High-Density Cell Bank to 2000 L Production Bioreactor”, 29160932, Edition AB (2016).