Integrated Digitalized Production of Protein Therapeutics: Linking Molecular Attributes to Process Optimization via Digital Twins

Context: Eight of the ten top-selling pharmaceutical products in 2016 were biopharmaceuticals. Their production through cell culture involves extensive clone selection and bioprocess optimization. In this context, not only productivity and viability of the host cells are pivotal, but also the molecular composition and heterogeneity of the biopharmaceutical are critical to ensure therapeutic safety and efficacy of the drug. More specifically, quality attributes of biopharmaceuticals are profoundly affected by cell culture conditions, therefore necessitating tight process control strategies. This is also emphasized in recent initiatives by the Food and Drug Administration to enhance Quality by Design in the production of biotherapeutics through moving away from empiricism to science‐based method development and process understanding (ICH Q12).
Hypothesis: Simultaneous monitoring of quality attributes facilitates instant tuning of critical process parameters (CPPs) to achieve a desired critical quality attribute (CQA) product profile. As the relationship between CQAs and CPPs is highly nonlinear and multidimensional, additional metabolome and proteome data have to be merged with well-controlled CPPs and enhanced CQAs for a mechanistic understanding. This data integration strategy paves the way for enhanced modelling approaches, for capturing the process understanding and deploying it in digital twins for robust control strategies.
Approach: Advanced computational modeling is based on the unique combination of comprehensive process data, omics-data, and multiple molecular attributes of the therapeutic proteins, specifically biosimilar bevacizumab and anti-Interleukin-8 monoclonal antibody as representative model products.
Innovation and added value: This initiative elaborates a better understanding and design of bioprocesses in order to yield biotherapeutics of predictable quality, safety, and efficacy. The developed methodologies will provide novel computational algorithms enabling a directed and sound data-based prediction of target quality attribute profiles, which will contribute both to a better understanding of bioprocesses implemented in the production of biopharmaceuticals as well as to a more cost- and labor effective rational design of bioprocesses for the manufacture of biopharmaceutical drugs. The methodological novelty lies in the combination of omics and process data sources with advanced hybrid models. This project will also promote new modality Advanced Therapy Medicinal Products (ATMPs) by providing tools to predict control strategies for highly varying and tailored drugs of the future.
Researchers: The three involved workgroups at the University of Salzburg and the Technische Universität Wien have acquired broad experience with the characterization of molecular attributes of biopharmaceuticals, host cell omics, as well as with the design and deployment of computational modeling of bioprocesses for the production of biotherapeutics.