Integration of Mechanistic and Data-Driven Models to Support the Transition to Continuous Biomanufacturing

The vast majority of pharmaceutical processes operate in both batch or fed-batch mode, as it allows them a high degree of flexibility and a large number of manufacturing routes. Production processes involving recombinant cell lines are no exception and are operated in fed-batch mode as well. Fed-batch operation allows nutrients to be fed throughout the cycle, thus preventing the culture from nutrient depletion before production goals are met. Nevertheless, the pharmaceutical industry has dedicated great effort into transitioning their operations from batch to continuous. The reason for this is that operating in continuous mode offers several advantages such as high-volumetric productivity, reduced equipment size and low-cycle times. Additionally, some of the problems associated with operating in batch, such as product variability between batches, can be minimized as well.
Chinese hamster ovary (CHO) cell systems are one of the most used host cell lines for therapeutic protein production. These systems enable the application of certain post-translational modifications in order to obtain specific structures of protein products that are compatible with humans. Thus, CHO cell systems are of great importance, being used for 70% of all recombinant biotherapeutics on the market. Though CHO cell systems are typically operated in fed-batch mode, they would greatly benefit from continuous operation, since it would ensure the production of a stable and consistent product, which is particularly relevant within the highly regulated pharmaceutical industry.
It is important to understand how fed-batch CHO cell systems can be adapted to a continuous mode: what are the critical quality attributes (CQA) and if they remain the same, what are the new ranges of operability that guarantee product quality; what are the critical process parameters and how will they impact CQAs. In this project, we propose a framework, both computational and experimental, that allows the transition from fed-batch CHO cell culture to continuous. We will initially focus on adapting a genome scale model (GeM) of a CHO cell system to a continuous setting. Having a model compatible with continuous operation is of great importance as it it enables the prediction of CQAs given a set of critical process parameters, thus reducing experimental expense.