A Streamlined Multivariate Analysis Platform for Optimizing Mammalian Cell Culture and Biologics Production

Biologics-based therapeutics have witnessed remarkable market growth in the past couple of decades with continuing future potential. Expiring patents on tried-and-true monoclonal antibody (mAb) drugs put tremendous pressure on pharmaceutical manufacturing compa-nies to develop biosimilars as quickly and efficiently as possible. One of the biggest bottle-necks in the scale-up of biopharmaceuticals manufacturing is the production of biomole-cules, and this is largely a consequence of cell culture constraints. Robust mammalian cell lines, such as Chinese hamster ovary (CHO) or non-secreting myeloma (NS0) cells, have been extensively characterized using -omics technologies; however, there is still more re-search to be done. One of the most high-impact changes that a pharmaceutical development company can make is the optimization of their cultures, and one of the best ways to do this is through the optimization of media composition and feeding strategy (for fed-batch cul-tures). Since basal media is often generalized for many cell lines of a particular cell type, supplementations must be made to cater to the specific nutrient requirements for the clone being cultured and biologic being produced. These needs are determined by collecting data on the consumption and production of nutrients and metabolites in the culture media and within the cells. The aim of this project is to culture mAb-producing GS-CHO cells and characterize their metabolic needs and behavior using quantitative methods such as high-performance liquid chromatography (for amino acids, vitamins, and mAb titer), flow cy-tometry (for cell count and viability), fluorescence measurements (for ATP) and BioProfil-ing (for glucose, lactate, and inorganic salts). It can be difficult to draw conclusions from so many data, so a streamlined and automated platform for multivariate analysis using Excel and R programming allows for clearer visualization of the data at a high level using tech-niques like hierarchical clustering and principal component analysis. This analysis helps elucidate relationships and shifts between the variables, which can be used to improve upon future cultures by informing model-based optimization software and providing a better un-derstanding of specific culture needs.M.S