(627e) Quality-By-Digital Design Framework for the Combined Cooling Antisolvent Crystallization and Wet-Milling System of an Active Pharmaceutical Ingredient

Crystallization is a fundamental separation and purification process, playing a crucial role in the chemical and pharmaceutical industries. It directly influences the properties of the solid product, such as dissolution and bioavailability, as well as the efficiency of downstream processes, including filtration, drying, and tableting. To ensure both process efficiency and drug efficacy, crystallization is characterized by critical quality attributes (CQAs), which define key quality-related requirements such as yield, purity, crystal size, and polymorphic form.

Wet-milling is a powerful technique for controlling (i.e., reducing) crystal size, whether integrated within the crystallization process or applied externally. It serves as an essential tool when target CQAs cannot be achieved through crystallization alone. However, crystallization processes, especially when integrated with wet-milling, are often complex, with a high number of interdependent parameters, making it challenging to fully explore the design space through experimentation alone.

Process modeling provides an efficient and systematic approach to overcoming this challenge by enabling rapid simulation of various scenarios and generating valuable insights into process behavior. Population balance models (PBMs) are widely used to simulate, explore, and optimize crystallization and wet-milling processes.

In this work, a general workflow is proposed for the model-based digital design of a combined cooling and antisolvent crystallization integrated with wet-milling system for a commercial active pharmaceutical ingredient (API). First, a rigorous model discrimination and kinetic parameter calibration is conducted to identify the optimal model architecture, using the user-friendly process simulation software CrySiV [1]. Then, the effects of process parameters on the products CQAs are systematically analyzed, and the system is optimized using a Quality-by-digital-design (QbD²) approach. The in-silico investigations determine whether the desired CQAs can be achieved through a single crystallization step or if wet-milling is required to meet quality targets. The proposed framework significantly reduces development time and number of experiments needed for optimal process development.

1. Szilagyi, B. et al. Cross-Pharma Collaboration for the Development of a Simulation Tool for the Model-Based Digital Design of Pharmaceutical Crystallization Processes (CrySiV). Cryst Growth Des 21, 6448–6464 (2021).