(649d) Development of a Digital Twin for the End-to-End Continuous Manufacturing of the Cancer Drug Lomustine

In recent years, the pharmaceutical manufacturing sector has entered a period of rapid advancement. Traditionally, the industry has relied on manual operation and batch processing, but it is now striving to adopt advanced manufacturing technologies, such as digitalization, automation, and artificial intelligence to modernize its production systems and align with Industry 4.0 standards (Arden et al., 2021). Important initiatives towards achieving this goal have been the development of process models, sensing tools, and digital twins for pharmaceutical process operations, and their application for purposes such as process simulation, real time quality assurance, design space analysis, and plant-wide control and optimization (Chen et al., 2020). In order to provide a consolidated platform for creating these systems, Casas-Orozco and coworkers have developed an open source tool for modeling pharmaceutical manufacturing processes, called PharmaPy, and have published a series of studies showcasing its capabilities (Casas-Orozco et al., 2021; Laky, Casas-Orozco, Destro, et al., 2022; Laky, Casas-Orozco, Laird, et al., 2022). This study aims to illustrate the use of PharmaPy in building and using a process digital twin. To accomplish this, an end-to-end continuous manufacturing process for the production of lomustine, a cancer medication, is utilized as a case study. The process employs two telescoped flow reactors to generate an intermediate compound that is further converted into an active pharmaceutical ingredient (API). The latter is then transferred into a different solvent using a distillation-based solvent switch unit, and further crystallized using a cascade of mixed suspension mixed product removal (MSMPR) crystallizers. The generated API crystals are then transferred from the crystallization solvent into a biocompatible carrier fluid using a three-phase settler system, and then processed into drug product using a drop-on-demand (DoD) 3D printing platform (Sundarkumar et al., 2022a).

Several enhancements have been introduced into the PharmaPy framework to facilitate the creation of the digital twin for this process. First, an extended multiple-curve resolution (MCR) framework has been incorporated in PharmaPy. This enables estimation of the kinetic parameters and their uncertainty regions for first-principles reaction models using full spectral data, and for crystallization models, by integrating both spectroscopic and crystal size measurements within the MCR framework. Furthermore, new unit operation models have also been added to PharmaPy, including traditional continuous distillation models to describe solvent switch operations (frequently encountered in pharmaceutical manufacturing), and novel models like three-phase settling, which describes transfer of API crystals to a biocompatible liquid carrier that is immiscible with the process stream containing the product in crystal form. A previously developed neural network model is used for simulating the operation of the DoD printer (Sundarkumar et al., 2022b). The resulting hybrid digital twin is then used to build a probabilistic design space that maps the likelihood of on-spec drug products and feasible operation for a variety of process operating conditions and scales. This design space can be used to identify suitable manufacturing conditions for different production scales and dosage strengths, thus, enhancing operational flexibility.

References

Arden, N. S., Fisher, A. C., Tyner, K., Yu, L. X., Lee, S. L., & Kopcha, M. (2021). Industry 4.0 for pharmaceutical manufacturing: Preparing for the smart factories of the future. In International Journal of Pharmaceutics (Vol. 602). https://doi.org/10.1016/j.ijpharm.2021.120554

Casas-Orozco, D., Laky, D., Wang, V., Abdi, M., Feng, X., Wood, E., Laird, C., Reklaitis, G. V., & Nagy, Z. K. (2021). PharmaPy: An object-oriented tool for the development of hybrid pharmaceutical flowsheets. Computers and Chemical Engineering, 153. https://doi.org/10.1016/j.compchemeng.2021.107408

Chen, Y., Yang, O., Sampat, C., Bhalode, P., Ramachandran, R., & Ierapetritou, M. (2020). Digital twins in pharmaceutical and biopharmaceutical manufacturing: A literature review. In Processes (Vol. 8, Issue 9). https://doi.org/10.3390/pr8091088

Laky, D. J., Casas-Orozco, D., Destro, F., Barolo, M., Reklaitis, G. V., & Nagy, Z. K. (2022). Integrated Synthesis, Crystallization, Filtration, and Drying of Active Pharmaceutical Ingredients: A Model-Based Digital Design Framework for Process Optimization and Control (pp. 253–287). https://doi.org/10.1007/978-3-030-90924-6_10

Laky, D. J., Casas-Orozco, D., Laird, C. D., Reklaitis, G. V., & Nagy, Z. K. (2022). Simulation–Optimization Framework for the Digital Design of Pharmaceutical Processes Using Pyomo and PharmaPy. Industrial & Engineering Chemistry Research. https://doi.org/10.1021/acs.iecr.2c01636

Sundarkumar, V., Nagy, Z. K., & Reklaitis, G. V. (2022a). Small-Scale Continuous Drug Product Manufacturing using Dropwise Additive Manufacturing and Three Phase Settling for Integration with Upstream Drug Substance Production. Journal of Pharmaceutical Sciences. https://doi.org/10.1016/j.xphs.2022.03.009

Sundarkumar, V., Nagy, Z. K., & Reklaitis, G. V. (2022b). Machine learning enabled integrated formulation and process design framework for a pharmaceutical 3D printing platform. AIChE Journal. https://doi.org/10.1002/aic.17990