Biopharmaceuticals are growing in importance and represent a rapidly expanding segment of pharmaceutical drug products. This group of complex molecules is produced using biological expression systems that are instrumental in their assembly, folding, and post-translational modification.
Process optimization of these systems is challenging, and interactions between critical quality attributes (QA) and process parameter (PP) changes are plentiful. To minimize risk, newly developed production processes leverage platform processes known to be robust, transferred from successfully implemented products, in combination with design of experiment-based optimization studies investigating known critical process parameters. Often, the design space of these optimization studies is limited by access to analytical capabilities of routinely measured analytes. Important process parameters, such as amino acid media composition changes during the culture, are not considered and might result in suboptimal processes. There is a need for analytical methods able to assess additional parameters to make data-driven decisions and increase process understanding with the aim of minimizing potential future risk in later development phases.
In this case study, we present newly developed analytical technologies able to deeply characterize the cell culture process to address a variety of potential needs in process development and increase data availability for understanding the relationship of complex PP/QA relationships in bioprocesses. Firstly, a mass spectrometry-based workflow able to monitor and quantify changes in expressions of host cell proteins over the culture duration and in relation to process parameter changes, allowing for risk-based assessment of changes in expression levels of critical HCP. Secondly, an HPLC-based workflow able to routinely measures amino acids and other cell culture media component profiles, allowing media batch variability assessment, monitoring of depletion and accumulation of critical components for media optimization, as well as data generation for metabolic modelling to gain deeper insights into the behaviour of the biological systems in the production process.