(499b) Model-Based Optimization and Design of Continuous Chromatography for Protein Capture

The petrochemical, chemical and food industries have practiced continuous manufacturing for many years. However, biotechnology is still late to embrace continuous processing, especially for downstream processing. Recently, multi-column periodic counter-current (PCC) chromatography has been proposed as continuous capturing technology to improve the process productivity and resin capacity utilization, as well as reduce buffer consumption and equipment footprint, which is applying for monoclonal antibody capture with Protein A affinity resin. Due to the complexity of continues chromatography, some mathematical framework should be developed to aid the optimization and process design.

To better understand and describe continuous chromatography process, a fully mechanism model was built for fitting and predicting breakthrough curves as well as optimizing the operation conditions. An integrated approach was developed to combine the Equilibrium Dispersive Model of column, General Rate Model of resin, as well as the Equilibrium Adsorption Model for protein adsorption. Firstly, breakthrough curves at different flow rates and loading concentrations are fitted to obtain the model parameters, including pore diffusion coefficient, mass transfer coefficient and so on. Then, multi-column adsorption pattern and continuous separation are evaluated and predicted through these parameters under varying operation conditions. Finally, the working window would be determined with the design strategy, considering the process productivity and capacity utilization. In addition, the performance of continuous chromatography would be compared with traditional batch process, and the advantages and limitation of continuous processing would be discussed. The model-based approach developed were tested with different continuous multi-column modes, including twin-column CaptureSMB, 3/4-column PCC from GE Healthcare and multi-column BioSMB from Pall. The interactive interface was also developed to easily input the data and visually output the results. The results indicated that model-based approach could aid the process design and enhance both the development and application of continuous chromatography.

This work was supported by the International Science & Technology Cooperation Program of China and National Natural Science Foundation of China.