(626c) Utilizing Mechanistic Modelling to Assist in the Process Development of Pharmaceutical Drug Substance Processes

The development of a pharmaceutical API/drug product requires a large number of experiments, the design and interpretation of which is often based on statistical modelling tools. This approach achieves the definition of parameter ranges that delivers products of high quality in a robust manner. However, the drawback to this approach is that the data only covers the processing conditions investigated and cannot be extrapolated outside of these conditions. Also, this approach does not guarantee a full understanding of all the phenomena involved in a given process and hence, it may yield a viable but not an optimised processes or sequence of operations. For the above reason, mechanistic modelling approaches would be the preferred choice, although sometimes not the most practical, yielding a higher degree of understanding of the chemical and physical phenomena occurring in a given pharmaceutical process. This approach can be applied to the development of both single unit and multi-stage operations, for both batch and continuous processes. Coupling models across multiple unit operations would lead to the optimisation not only of a single operation but the entire process, thus leading to reduced costs and maximising product yields.

In this context, the work presented models batch unit operations, including reactions, separations, and links them together following a common pharmaceutical process structure. In this case, gPROMS FormulatedProducts software was applied to meet this mechanistic modelling need. This tool includes a wide array of mechanistic models for drug substance applications. The kinetic parameters for the reaction mechanisms considered were validated using a small number of data rich experiments, allowing for reaction model and mechanism discrimination to achieve the most descriptive mechanistic model for the batch reaction step. In the future, the validated process models will be utilized to describe the integrated drug substance processing steps. This end-to-end mechanistic modelling approach will help to both optimize a full pharmaceutical manufacturing process and determine its overall robustness against disturbances.