Continuous processing, for decades an essential characteristic of advanced large-scale manufacturing, has been adopted slowly for mammalian cell-based processes, and even more haltingly for microbial fermentations. In a recent paper in Frontiers in Bioengineering and Biotechnology, lead author Christoph Herwig, PhD, at the Vienna Institute of Technology, described the reasons for the snail-paced uptake:

* Controlling metabolic load for sustained recombinant production, for example, through assessment of cellular population stability and the analysis of phenotypic and genotypic instabilities

* Changing the product location to circumvent the internal physical limit of the cell

* Achieving prolonged, time-invariant processing for stable productivity and product quality

* Integrating upstream and downstream processing, and connecting or linking individual unit operations

Which raises several questions. For example, will adoption of continuous processing in microbial processes follow the apparent trend in mammalian systems, i.e., slow adoption, eventually reaching a steady state of some low-percentage number of commercial processes?

“I see a lot of progress in turning single unit operations, both upstream and downstream, to continuous processing,” Herwig says. “This is basically solved for most of unit operations in common biopharm processes. However, I do not see enough emphasis on linking or connecting unit operations. Currently, many applications use classical transitions between unit ops, which are triggered by time or volume, as in classical production recipes. But this is insufficient and may explain the slow progress and the small number of success stories for continuous bioprocessing.

“What we need is to compensate for variabilities in the preceding step, which is accomplished by quantifying that variability through Process Analytical Technology (PAT), and through process understanding that informs us on how to compensate for variability.”

Scale plays a significant role

The much larger volumes for microbial systems, compared with mammalian cell culture, affect every aspect of continuous processing. Herwig explains that scale plays a significant role in general, as process economy is mainly governed by the equilibration of the occupation time of all unit operations of a process chain.

“For example if the bioreactor volume is large and the occupation time is long, but the downstream units can process the culture fluid rapidly, then downstream capacity utilization falls. In other words, these assets are not used effectively,” he explains.

Don’t expect any shortcuts here, other than yeast, being more “predictive” in their behavior in culture.

“I am always impressed by the fact that yeast hosts such as Pichia pastoris are incredibly predictive in their metabolic behavior,” says Herwig. “I can feed dual substrates and steer between energy supply and production. Those organisms can even recover completely after recombinant protein production, while E. coli, once induced, is difficult to control due to their heterogeneous population. The low hanging and only real enabling ‘fruit,’ the path forward to a completely continuous microbial bioprocess, is process understanding, driven by smart experimental design and advanced data science, and catalyzed by digitalization and captured by digital twins.”

Previous articleCRISPR Immunity Evolves with Bacteriostatic Antibiotics
Next articleData Key to Modeling Drug Production Process Sustainability