In continuous bioprocessing, pilot scale-up is a key transitional step. The project has moved past the discovery and characterization phases, and the scale-up and workflow for producing the target protein of interest is being scrutinized and optimized. In this article we will address the tools scientists use as they transition from generating material in research quantities (milligrams) to clinical trial and commercial scales (grams to kilograms).
Bioreactor scale-up
A successful scale-up maintains productivity and quality of a process. A larger-volume process should be at least as robust as a smaller volume process. This can be evaluated through measures such as cell density, viability, and expression rates. In addition, quality of the resulting model can be determined by assessing features like glycosylation.
Surendra Balekai, senior global product manager, single-use technologies, Thermo Fisher Scientific, says that expectations should be consistent, whether the process is batch, fed batch, or continuous for suspended or adherent cultures. “Equipment design must help in establishing a robust process, and validation of the process needs to be scalable. Any features of the equipment like geometry, agitation, sparging, heat and mass transfer, exhaust management, process control, and data logging are primary aspects of scale-up,” Balekai says.
Thermo Fisher Scientific prioritizes scale up aspects of its single-use systems, including BioProcess Containers, and its controllers and software are identical for lab-scale and large-scale cGMP production.
Biocompare’s Cell Culture Systems Search Tool
Find, compare and review cell culture
systems from different suppliers Search
Barbara Paldus, vice president and general manager, Finesse Solutions, says that it’s important to choose the right process platform at the research level for future scale-up. “Starting from the wrong process platform will limit the ability to successfully scale-up. Worse, scaling-up without these criteria may lead to an unusable or poorly yielding process in manufacturing, resulting in financial losses, overbuilding of production capacity, or an inability to meet market needs,” says Paldus.
Beyond platform choice, an array of specific parameters must be chosen. Some of the product or process features for mammalian- or microbial-derived products include:
- Power-to-volume ratios
- Height-to-diameter ratios
- Equipment geometry across scale (ratio of impeller diameter to vessel diameter, impeller blade length, width, and angle)
- Fluid dynamics (type of agitation, number of impellers, tip speed)
- Heat and mass transfer properties (oxygen transfer coefficient, oxygen transfer rate, oxygen uptake rate)
- Sparger design
- Exhaust management
Missteps are most often taken at the preclinical or Phase I clinical manufacturing stage. “Any mistakes made at this stage will have significant impact at later stages,” says Balekai. For example, Balekai points out that there is often not much CO2 accumulation in the culture at bench scale. “When the same process is at the 50 L scale or above, CO2 accumulation becomes an issue. Here the selection of a sparger that is effective in CO2 stripping and O2 delivery to the cells becomes important.
It is important also to verify the equipment design with reference to scale-up parameters across all sizes required. The performance aspects of various spargers, for example, change quite a bit with changes in volume.
Other potential pitfalls include not accounting for changes in bioreactor performance as a function of volume, mixing control systems where control strategies are different and components differ in performance, not making scale-down models from a large volume reactor, and believing that one must “start over” with the a larger volume system because bioreactors are so dissimilar.
Downstream scale-up
Downstream chromatography steps also require careful attention to technical parameters for successful scale up. For example, smaller diameter chromatography columns have wall effects that are absent when moving to a larger diameter column. You can also expect a high pressure drop at a lower flow rate when scaled up. According to Hana Kim, senior global product manager, process chromatography at Bio-Rad, “You might have a pressure drop of 2 bars with 1 centimeter ID column at 1,000 centimeters per hour, but you might reach a pressure drop of 2 bars with a 20 cm ID column at a 300 centimeter per hour flow rate.”
Another factor contributing to results in downstream scale up is the resin—its quality, life cycle, lot-to-lot variability, and availability in large quantity. Some resins used at lab scale are not available, or prohibitively expensive, in clinical- or commercial-scale quantities.
There are some limits to the scalability of certain chromatography processes. “You might reach some technical and physical limitations,” says Stefan Schmidt, vice president, process science and production at Rentschler Biotechnologie. For example, lab-scale flow rates might not be manageable on a larger scale column, and some steps might not be scalable at all. “Usually size exclusion is not scaled, and would be replaced with ultrafiltration,” says Schmidt, who also noted that, very flat gradients can be difficult to work with at larger scales, resulting in an unmanageably large elution volume.
Planning for scale-up
Scaling down to scale-up is a crucial planning step for any process that will eventually be produced in large quantities. Parameters such as operational windows for buffers, feed, resin life cycle, and resin lot-to-lot variations can be tested thoroughly on smaller versions of commercial bioprocessing equipment, such as Bio-Rad’s NGC system, according to Kim.
Scaling down to scale-up is a crucial planning step for any process that will eventually be produced in large quantities.
Scaling down comes with some of its own caveats. In some instances, scaled-down performance of commercial equipment will not produce the same results. “If you miniaturize a column too much, and you don’t have instruments that operate at low flow rates, the gradients will not be formed properly,” says Schmidt. Furthermore, bioreactors may not have equivalent performance when scaled-down, “Other than that, you should have a proper scale-down model.”
Other ways to plan for scaling up include choosing resins with an appropriate particle size and availability in large quantity, testing the effect of variables on process performance, confirming reproducibility of the process, and evaluating the limitations of pilot or plant equipment such as tank size, pump flow rates, etc. It is also helpful to know the stability of the product at each stage of purification.
Image: Shutterstock Images