(243f) Quality-By-Control of a Novel Unit for Continuous Integrated Filtration-Drying of Drug Substances

Quality-by-Control (QbC) is a recent evolution [1] of the Quality-by-Design [2] initiative, in which active process control represents the core feature of the control strategy. With an active process control system in place, the critical process parameters of the process are automatically manipulated in response to disturbances, so that the target critical quality attributes of the product are ensured. The QbC framework has acquired even more importance within the recent interest in continuous integrated pharmaceutical manufacturing [3]. Continuous manufacturing (CM) guarantees greater product consistency compared to batch processing, as continuous processes present inherently greater process controllability. QbC has emerged as a very useful tool for the transition to CM, which requires a systematic framework for process control.

In this study, we develop a QbC framework for an intensified unit for continuous integrated filtration-drying of crystallization slurries. Continuous and integrated solutions for filtration and drying in upstream manufacturing are not much studied in the literature, although these processing steps represent a main obstacle for the implementation of end-to-end CM. The unit studied in this work is a novel carousel technology [4]. The carousel presents multiple processing stations located within a main cylindrical body. In the first station, the crystallization slurry is loaded. In the subsequent stations, slurry filtration, cake washing, cake deliquoring and thermal drying are carried out. The dry cake is discharged from the last station through the action of a pneumatic piston. All stations operate simultaneously, processing different batches of material: every fixed cycle duration the carousel rotates, enabling pseudo-continuous operation. Filtration, washing and drying operations present high coupling per se, and are even more interacting within the rotation mechanism of the carousel, which fixes the residence time of each processing step.

To aid the development of the control strategy for the unit, we develop a comprehensive digital twin of the carousel. The digital twin is based on detailed first-principles models of the processing steps [5], combined together for simulating real time operation of the carousel. The developed framework represents a unique and effective environment for testing and comparing different control strategies, since: i) time-consuming experiments with different control strategies can be quickly simulated on the digital twin, and ii) the critical quality attributes of the product, such as the solvents and impurities content, are always accessible in the in-silico simulations, while they cannot be measured in real time on the physical unit.

After having developed the digital twin, we validate it with filtration and drying experimental data collected on a physical carousel. Then, we conceive and implement on the digital twin a three-level control system for the carousel, based on a recently proposed approach [1] that follows the ISA-95 Enterprise-Control System Integration Standard. The Level 0 of the control system is made up of the built-in controllers of the carousel and of the accessory equipment, such as the pumping system. End-point control is resorted to at Level 1, for automatically setting the duration of each cycle to ensure that the discharged cake meets the target residual solvents and impurities content. Within the end-point control loop, the available measurements from the process are used to infer in real time the time instant at which the cake can be discharged because the desired quality has been met, and the carousel rotation routine is consequently automatically run. We also implement a model-based Level 2 control system, which includes a state estimator for continuous monitoring of the critical quality attributes and real time optimization. We successfully test the proposed control system in normal operating conditions and in presence of filter mesh fouling, which is known to be the main disturbance occurring during carousel operation.

References

[1] Su, Q., M. Moreno, C. Laird, Z. Nagy and G. Reklaitis (2019). A perspective on Quality-by-Control (QbC) in pharmaceutical continuous manufacturing, Comput. Chem. Eng., 125, 216–231.

[2] Food & Drug Administration, 2004. Guidance for industry, PAT-A Framework for Innovative Pharmaceutical Development, Manufacturing and Quality Assurance. http://www. fda. gov/cder/guidance/published. html. Accessed on 03/29/2021.

[3] Lee, S. L., T. F. O’Connor, X. Yang, C. N. Cruz, S. Chatterjee, R. D. Madurawe, C. M. V. Moore, L. X. Yu and J. Woodcock (2015). Modernizing Pharmaceutical Manufacturing: from Batch to Continuous Production. J. Pharm. Innov., 10, 191–199.

[4] Liu, Y.C., A. Domokos, S. Coleman, P. Firth, Z.K. Nagy (2019). Development of continuous filtration in a novel continuous filtration carousel integrated with continuous crystallization, Org. Proc. Res. & Dev., 23 (12), 2655-2665.

[5] Destro, F., V. Wang, M. Abdi, X. Feng, E. Wood, E., S. Coleman, P. Firth, A. Barton, M. Barolo and Z. K. Nagy, (2021). Continuous integrated filtration, washing and drying of aspirin: digital design of a novel intensified unit. Proc. of the 11th IFAC Symposium on Advanced Control of Chemical Processes, Venice (Italy), June 13-16 2021. In press.