(558a) Enabling PAT-Based Process Control for Therapeutic Antibody Production Via Real-Time Chemometric Model Calibration

Therapeutic antibodies are ubiquitous in pharmaceutical industry as “they provide needed level of specificity for a substantially enlarged therapeutic window” [1]. These antibodies are predominantly produced through a train including: cell expansion, N-1, and production bioreactors. Fed-batch production is particularly beneficial. Numerous shifts and alterations/additions can be incorporated into the process to address wide range of complex environmental and cell biology factors that impact overall cell culture performance and behavior. These shifts and alterations/additions include cell-line characteristics, micro- to macro- scale environment, nutrients needed and available, cell age, metabolic activity, etc..

The importance of real-time control for fed-batch environmental factors such as temperature, pH, and dissolved oxygen, has been well documented in the literature [2], [3]. However, whether the presence (or lack) of real-time control of key nutrients results in meaningful process differences have not yet been clearly demonstrated. For example, the question of “does real-time process control of key nutrients and protein/cell building blocks result in increased process yield and/or more consistent product quality?” has been challenging to answer. This is mainly due to Process Analytical Technologies (PAT) models being product and process specific, which oftentimes requires rigorous chemometric model training and validation workflow, e.g., Partial Least Squares, Random Forests, etc., on Raman spectroscopic data [4].

In this work, we propose a novel PAT-based process control platform where chemometric modeling validation framework is based on “the best chemometric model for the next 24 hours” rather than “the best chemometric model given the past data”. Detailed analysis will be shared on mathematical modeling framework, real-time calibration methods developed, and findings on robustness of cell culture process investigated.

References
[1] Paul S., et al., “Cancer therapy with antibodies,” Nature Review Cancer 24, 399-426, 2024.
[2] Hu W.S., “Cell Culture Bioprocess Engineering”, CRC Press, 2020.
[3] Zakrzewski R., Lee K., Lye G.J., “Development of a miniature bioreactor model to study the impact of pH and DOT fluctuations on CHO cell culture performance as a tool to understanding heterogeneity effects at large-scale,” Biotechnology Progress 38 (4), 2022.
[4] C. Rafferty, K. Johnson, J. O’Mahony, B. Burgoyne, R. Rea, and K. M. Balss, “Analysis of chemometric models applied to raman spectroscopy for monitoring key metabolites of cell culture,” Biotechnology progress, vol. 36, no. 4, p. e2977, 2020.