(160f) Model-Based Process Optimization of Ion-Exchange Chromatography for Protein Bioseparation

The separation and purification of proteins with ion-exchange chromatography (IEC) in the biopharmaceutical industry have progressed over recent decades. Process development and optimization are critical for IEC. In order to improve process efficiency, many researchers prefer to develop mechanistic models. However, the time-consuming and cumbersome model development and calibration hinder the model-based process optimization.

The equilibrium dispersive model and the steric mass action (SMA) model, proposed by Prof. Cramer in 1992, were combined to create a model. Regarding model calibration, a new method called "parameter-by-parameter (PbP)" was proposed with mechanistic derivations of the SMA model. Compared to the traditional calibration strategy, the SMA parameters could be determined one-by-one in sequence with in-depth physical understanding and high efficiency, resulting in a reduction of the number of unknown parameters per species from four to one. The PbP method has been successfully used in the separation of model proteins (laboratory-level) and monoclonal antibodies (industrial-level). Regarding model-based process optimization, a new heuristic method was implemented for solving the optimization problem. For the industrial-level separation of charge variants of monoclonal antibodies, the yield of the main product can be improved significantly from 55% to 87%. The mechanistic models have also been used in the continuous mode of IEC for better process design, such as multi-column counter-current solvent gradient purification.

In general, compared to the design of experiments (DoE), model-based process optimization can provide the comprehensive process understanding of IEC, leading to lower cost and higher efficiency of process development. It is a promising technique for the digitalization of biopharmaceutical processes.