(614e) Mechanistic Modelling, Uncertainty Analysis, and Advanced Process Control of Particle Size Distribution in Continuous Tubular Protein Precipitation

Modern biopharmaceutical manufacturing pipelines for monoclonal antibodies (mAbs) are dominated by batch Protein A chromatography for primary capture. As process intensification successes have brought increasingly higher product titers in the upstream bioreactor, column-free continuous protein precipitation is becoming an economically competitive alternative [1,2]. Meanwhile, there has been increasing interest in the integration of process analytical technology (PAT), first-principles models, and model-based feedback control in biopharma processes [3]. The current literature on model-based feedback control (aka model predictive control) for continuous protein precipitation, however, is nonexistent. No computationally efficient software has been reported even for the simulation of first-principles models for continuous protein precipitation, which limits the incorporation of such models into model predictive control (MPC) algorithms.

To address this need, we formulate first-principles models for the evolution of precipitate particle size distribution (PSD) in a dynamic, non-isothermal, and continuously operated tubular reactor with spatial dosing of precipitating agents. The population balance model (PBM) features precipitate nucleation, growth, and aggregation dynamics formulated using integral terms, with the PBM coupled to mass and energy balances for solute depletion and temperature changes due to jacketed cooling. The most closely related models [4,5] are shown to be special cases of our model. The distributed parameter system is solved efficiently using a high-resolution finite volume method scheme. Open- and closed-loop system operation is analyzed using the method of moments and numerical bifurcation tools to provide stability and robustness guarantees for the control of the system. Finally, a fast adaptive MPC strategy is implemented for controlling the mean and variance of the PSD to desired CQA targets. Overall, the results of this work demonstrate the potential for on-line model adaptation and control of continuous precipitation for capture of monoclonal antibodies. Financial support is acknowledged by Contract 75F40121C00111 from the U.S. Food and Drug Administration. Further financial support is acknowledged by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, Department of Energy Computational Science Graduate Fellowship under Award Number DE-SC0022158.

References:

[1] Martinez, M., Spitali, M., Norrant, E. L., Bracewell, D. G. (2019). Precipitation as an enabling technology for the intensification of biopharmaceutical manufacture. Trends in Biotechnology, 37, 237–241.

[2] Minervini, M., Mergy, M., Zhu, Y. C., Gutierrez Diaz, M. A., Pointer, C., Shinkazh, O., Oppenheim, S. F., Cramer, S. M., Przybycien, T. M., Zydney, A. L. (2024). Continuous precipitation-filtration process for initial capture of a monoclonal antibody product using a four-stage countercurrent hollow fiber membrane washing step. Biotechnology & Bioengineering, in press.

[3] Hong, M. S., Severson, K. A., Jiang, M., Lu, A. E., Love, C., Braatz, R. D. (2018). Challenges and opportunities in biopharmaceutical manufacturing control. Computers & Chemical Engineering, 110, 106–114.

[4] Mozdzierz, N. J., Lee, Y., Hong, M. S., Benisch, M. H. P., Rasche, M. L., Tropp, U. E., Jiang, M., Myerson, A. S., Braatz, R. D. (2021). Mathematical modeling and experimental validation of continuous slug-flow tubular crystallization with ultrasonication-induced nucleation and spatially varying temperature. Chemical Engineering Research and Design, 169, 275–287.

[5] Zhang, W., Przybycien, T., Schmölder, J., Leweke, S., von Lieres, E. (2024). Solving crystallization/precipitation population balance models in CADET, Part I: Nucleation growth and growth rate dispersion in batch and continuous modes on nonuniform grids. Computers & Chemical Engineering, 183, 108612.