Going from Fed-Batch to Perfusion

For fully integrated biomanufacturing, continuous upstream and downstream operations must be coupled together. The current study used bioprocess fluid generated in a fed-batch cell-culture process because that is the most common upstream processing method. 

Continuous cell culture, or perfusion, processes have been implemented to produce marketed biopharmaceuticals. To date, however, they have generally been adopted because traditional batch or fed-batch processes are not suitable-in most cases because the purification of the biologic drug substance is time sensitive (unstable or labile in some way) and undergoes degradation within the bioreactor production process.

Perfusion processes are attractive for these products because once a steady state has been achieved, cell-culture media containing the product is removed from the bioreactor (along with enough cells to account for cell doubling so that a consistent cell count is maintained) and replaced with fresh media on a continual basis. Fed-batch reactions typically last 10-14 days, while perfusion processes run for 30-60 days or longer. 

Perfusion processes also can offer significantly higher productivities in grams/L of bioreactor working volume per day, enabling the use of smaller, single-use bioreactors and reducing capital expenditures. They are particularly attractive for multiproduct facilities designed for flexibility and employing single-use technologies. As with any continuous process, process development is generally accelerated and the need for scale-up and technology transfer is often avoided with perfusion processes.

Most Chinese hamster ovary (CHO) cell lines will grow in both perfusion and fed-batch modes. While perfusion titers are typically less than those observed for fed-batch reactions, this feature is mitigated by the higher cell densities possible in perfusion reactions, which allows for the high productivities. Both perfusion and fed-batch reactions provide, after clarification, harvest cell culture fluids that are suitable for downstream purification.

Perfusion, however, has not yet been widely adopted for bioprocessing. At the present, there is concern about the cost of perfusion due to the high quantities of cell-culture media consumed. For instance, for a 50 L bioreactor, removal of one to two reactor volumes per day for 60 days adds up to 3000–6000 L of media, which carries significant cost. Logistics and supply chain issues must also be addressed; those massive quantities of media must be on hand in real time all the time.

Switching from a commercialized fed-batch process to a perfusion process is also unlikely due to the extensive regulatory requirements for demonstration of comparability. It is much easier for manufacturers to scale up or scale out the existing fed-batch process. 

However, there is growing interest in perfusion because these processes are continuous and often achieve much higher productivities. These benefits have the potential to offset the higher media and other costs associated with perfusion. Initial technical barriers to removal of media with cell retention within the bioreactor were initially addressed with tangential flow filtration. Today several other non-filtration technologies, including acoustic wave separation, are under development and showing real promise. It is still early days with respect to the adoption of perfusion for upstream bioprocessing, but the future looks exciting.

Article Details

BioPharm International
Volume 30, Number 7
July 2017
Page: 32

Citation

When referring to this article, please cite it as X. Gjoka, R. Gantier, and M. Schofield, “Going from Fed-Batch to Perfusion," BioPharm International 30 (7) 2017.