Getting to Grips With the New Generation

There are a number of different ways of describing next generation bioprocessing, including continuous manufacturing and intensified processing. Next generation bioprocessing covers a variety of different techniques, from combining process steps, to moving from batch to perfusion bioreactors, to new cycling operations – but all are ultimately designed to deliver significant improvements in manufacturing costs, speed, efficiency or flexibility. Some descriptions of particular technologies are laid out in the BioPhorum Operations Group’s (BPOG) Biomanufacturing Technology Roadmap (1). Next generation processing technologies and techniques do not just offer incremental improvements. If we have a 14-inch and a 12-inch filter, the development of an 8-inch filter would not be considered “next generation,” for example. A membrane adsorber though that can be run at shorter residence times and cycle rapidly to reduce a column size from tens of liters to just one liter would hit the mark. The reduction in size would also make the use of single-use technologies viable, and with single use, you can also start to think about fully closed systems for better bioburden control, among other benefits.

As next generation processing is new for biopharma, a perfectly human response is to be wary. One of the key comments we hear all of the time is that “regulators will never allow this”. That is not true. If you listen to regulatory presentations, they often talk about next generation technology to improve the quality of medicines and to make medicines more available (2). Regulators are very open to discussing technologies with companies as a means of encouraging adoption. The FDA has an Emerging Technology Program, the EMA an Innovation Task Force and in Japan the PDMA Innovative Manufacturing Technology Working Group allows manufacturers to obtain early regulatory input on different types of technologies and how to validate them (3).

Risk assessment is a useful tool to identify, prioritize, and start to address uncertainties. Gap assessment is another tool to identify needed data or analysis to reduce uncertainty. One of the most effective ways of managing risks with new technology is gradual adoption, such as rolling out hybrid processes that combine batch and intensified processing. Gradual adoption allows manufacturers to trial and select the processes that have the best benefits for the molecule or the facility in question.

The move to continuous is a natural evolution for a manufacturing industry. Processes typically start out as a batch process because in a batch process it is easy to manage risk. The process steps are uncoupled from each other so if one step goes down, the remaining steps can still proceed. There is slack time and intermediate inventory built in to the batch process to allow response to upsets. As experience is built and process steps become more reliable, you start to think about connecting steps and reducing costly inventories and poor process utilization. Many other industries have now done this. In the early days of biopharma, contamination in the bioreactor was common, but now we can go years without seeing an incident. Why? Because we have more experience. And I believe that we now not only have the experience, but also new technologies like single use, to maintain sterility and allow us to connect steps without significantly increasing overall process downtime.

Process planning

The potential benefits of implementing next generation technologies lie in reduced capital costs, manufacturing flexibility, plant efficiency and speed to market, but there have been many questions raised in the industry about how process development will also be affected and what problems manufacturers might encounter. It’s important to remember that the quality attributes associated with the molecule will be the same regardless of how the drug is produced. However, there will be some changes in process attributes and their relative importance. There is more emphasis on risk for a next generation approach because it is continuous. The stopping points in a batch process provide ample opportunity for recovery from an upset, but if there is a problem in a continuous line then you need to divert the intermediate product to waste, which is expensive. Process monitoring and control is critical to identify upsets and respond quickly to avoid compromising final product quality. Once a process is up and running, however, automation makes the process easier-to-run, with fewer deviations and risk, than batch processes.

Broadly speaking, next generation bioprocessing is not fundamentally very different to traditional batch processing. You’ll be running some of the filtration steps at constant flow instead of constant pressure, cycling faster with shorter residence times, and be running in flow-through mode rather than bind-elute mode, but for the most part the physics underlying how each step works are the same. For example, adsorption is based on charge and is the same whether done with a membrane or a resin – and still the same if you cycle faster in a next gen approach. The shorter residence times, however, mean that the target operating point moves in your design space.

Of course, there are some steps that will be very different, such as perfusion, which has the new operating parameter of turnover in the bioreactor. If you have a 200 liter perfusion bioreactor, you need to consider how many liters equivalent you are then feeding and withdrawing every day. If you have 200 liters of media that you’re adding and 200 liters of product that you’re removing, you’ll have a turnover of one volume per day. With perfusion, there is also a perceived risk by some in the industry that the quality of the drug product in perfusion will be worse than in fed batch. Cells will be producing in perfusion runs that may last for weeks – and I’ve heard concerns raised that the quality of the drug product may degrade over time. Bear in mind, however, that drugs most sensitive to degradation, e.g., Factor VIII, cannot be processed in fed batch and are produced in perfusion because perfusion has higher yields with less product degradation!

Changing from a batch low pH virus inactivation process to an inline continuous process involves replacing multiple mixing and holding tanks with pipe inactivation chambers.  This reduces footprint, the complexity of managing multiple Protein A elutions in multiple tanks, and improves the uniformity of the inactivation solution. Residence times in the inactivation chambers are potentially shorter than with conventional batch operation. The product protein will sit in acid for much less time, meaning lower protein degradation, higher quality and better yields. We’re also able to control the flow and so forth in this incubation chamber much better than in a 2000 liter tank, so it’s much better for monitoring and controlling the process. Informal discussions with health authorities and biomanufacturers show a lot of interest in this work.

One significant difference between batch and next gen is the processing time. In batch, a product spends around two weeks in the bioreactor and then moves downstream, where it is processed by successive steps for a few hours each day. In a next generation process, product quickly moves out of the bioreactor and through the downstream steps in hours. Less residence time means less potential for degradation. This process can continue for perfusion production campaigns that can run 14-90 days. The campaign is typically broken up into smaller lots lasting just a few days, and while this increases the lot release testing costs, it also reduces the risk of an upset, causing one to have to dispose of the entire campaign.  Within a lot, many operations will be performed in a cycling manner to reduce their size and cost. It is possible to size steps so that, at the end of a lot, adsorbers may have cycled through their active life and can be disposed of. The entire wetted path can be replaced with no carryover between lots.

Next generation processing is new, so of course there will be many questions and in time we will resolve these. As one example, I’ve heard many discussions about batch definition, but remember that there is actually flexibility with how a lot is defined so a manufacturer can choose how a lot is defined in continuous – perhaps by certain volumes, or volume turnover of the bioreactor, or days, or mass. It just has to be clearly defined prior to manufacture.  Despite the questions and the industry’s fear of change, a lot of people are very excited by next generation processing.

The topics raised in this article were discussed in more detail in a recent webinar we gave. You can watch the recorded webinar at your convenience at: www.emdmillipore.com/nextgenseries

Herb Lutz is Global Principal Consultant, MSAT, at MilliporeSigma and Brian Hubbard is CEO of CMC Bioprocess Consulting LLC. The life science business of Merck KGaA, Darmstadt, Germany operates as MilliporeSigma in the US and Canada.