Get Social With Us

Has the time come to embrace continuous processing?

As in years past, the Recovery of Biological Products XVI Meeting in Rostock, Germany (July 27‑31) provided a forum for lively debate on new technologies for downstream processing, new applications of old technology, and glimpses of what the future of our industry might hold. High among the topics of interest at the conference was continuous processing as a replacement for traditional batch processing of monoclonal antibodies and other recombinant proteins. Jon Coffman (Boehringer Ingelheim) began the conference pointing out that the cost of manufacturing per se is not limiting our ability to deliver products at a cost effective level but rather the high failure rate of products in clinical trials and the time required for overall product development were more critical in driving the cost of medicines today. Jon then showed how continuous processing might help accelerate production of material for clinical trials, allowing companies to shorten overall timelines to commercialization. This was followed by a presentation from Charlie Cooney (MIT) in which he reviewed the evolution of continuous processing for traditional pharmaceuticals and argued that the biopharmaceutical industry can also benefit from the efficiencies and flexibility continuous processing can provide.

Later in the program, Veena Warikoo (Genzyme), Oliver Kaltenbrunner (Amgen), and Alex Xenopoulos (Merck Millipore) presented their views on the development and implementation of integrated, fully continuous processes for recombinant proteins and monoclonal antibodies. While there was discussion and some disagreement over implementation of “end to end” continuous processes, these presentations and several posters at the meeting, did make the case for the use of continuous perfusion cell culture and continuous chromatography for initial product capture for recombinant proteins that are relatively unstable (e.g. Factor VIII) and helped define the objectives for implementing continuous processing for more stable products such as monoclonal antibodies. In these cases, continuous processing, whether for a single unit operation such as Protein A capture or a more fully integrated end-to-end process, there may be economic and/or productivity gains to be had through continuous (or semi-continuous) processing. Furthermore, advances in process analytical technology (PAT) may also favor using aspects of continuous downstream processing. With FDA and most industry experts embracing the concept of continuous processing, it is likely that some level of continuous processing will become commonplace in biopharmaceutical industry in the near future. Looking toward this future, Karol Lacki (GE Healthcare) gave an excellent presentation on a unified approach to process development that supports batch, semi-, and fully continuous processing without a need to repeat costly process development and process characterization studies should a switch be made from one operational mode to another later on.

The case for sticking with batch processing was made in comments by Brian Kelley (Genentech). According to Brian, conventional batch processes can be economical and suitable for production of biopharmaceuticals, even if very large columns or multiple cycles per column are required to process increasing levels of product in the harvest from large volume bioreactors. For a company like Genentech, with a large base of installed capacity and manufacturing assets, perhaps this is true. However, as companies look toward designing and building “factories of the future,” smaller facilities incorporating some aspects of continuous (perfusion) cell culture and continuous chromatography may be more cost effective and productive than larger facilities built around traditional batch processing. In conclusion, a company decision to use aspects of continuous processing may well depend on the stability of the protein, and an economic and risk analysis of placing such technology into an existing facility, or planned new construction.


Blog article by: Howard Levine and Frank Riske