Automated Microbioreactors and the Characterization of Media Dependent Changes in Antibody Product Glycosylation and Aglycosylation

Abstract

The glycosylation state of therapeutic antibodies is a critical quality attribute due to its effect on product efficacy and safety. With the advent of the biosimilar development and the need to match the glycan profile of the originator product, better understanding of how the variables of the bioprocessing procedure control glycosylation is critical. To this effect, we used automated microbioreactors with our in-house model CHO DG 44 cell line and different media types to study differences in antibody outcomes, specifically focusing on N-glycosylation profiles and aglycosylation rates. We observed that different media types resulted in vastly different amounts of high mannose and terminal galactosylation of N-glycans. By measuring the percentage of antibodies that was not N-glycosylated we observed that high mannose outcomes were not correlated to aglycosylation rates. For further analysis, we utilized multivariate data analysis to determine the process variables that best explained our glycan profile findings. Factors linked to glutamine consumption were determined to be the most important in predicting high mannose outcomes, while factors related to the temporal aspects of cell growth rate were linked to terminal galactosylation. Our work discovered in-process parameters in the cell culture process that have significant effects on the glycan profile of an antibody product, further elucidating the link between the biomanufacturing process and product quality outcomes.