Use of Bayesian Modeling for Risk Assessment and Robustness Evaluation

The definition of a robust control strategy is critical to ensuring quality in the manufacture of active pharmaceutical ingredients (APIs). This publication provides a case study wherein Bayesian probabilistic modeling was used to evaluate the process risk of exceeding the permitted level for a critical impurity in the deucravacitinib drug substance (DS). Formation and purging of this impurity were studied via a Design of Experiments (DoE) and univariate investigation of the reaction and crystallization parameters. Bayesian probabilistic models were generated to predict process performance and assess predicted failure rates at the target parameter set points as well as across the multivariate parameter space. This analysis confirmed the designation of a critical process parameter (CPP) and provided process knowledge on high-risk multivariate parameter conditions. The probabilistic models also enabled estimation of the impact of common cause plant variability on the anticipated failure rate for a given specification level of the critical impurity.