P7: Application of a multivariate Bayesian method as part of the Quality by Design approach for the biopharmaceutical process development

Quality by Design (QbD) is a new approach to the development of biopharmaceuticals to promote a better understanding of the product and its manufacturing process. One of the results of QbD implementation of biopharmaceutical products is the definition of the reliable operating ranges of the critical process parameters (CPPs) to meet specific acceptance criteria of the critical quality attributes (CQAs). Various approaches have been applied to optimize the biologic product fermentation process and define the control ranges for product quality. In this presentation, we presented a stepwise case study to optimize a S. cerevisiae fermentation process through the risk assessment analysis, statistical design of experiments (DoE) and multivariate Bayesian predictive approach. The critical process parameters (CPP) were identified through the risk assessment. The surface response to each product critical quality attribute (CQA) was modeled. The interaction of the CPPs was also modeled for each CQA. A multivariate Bayesian predictive approach was used for multiple response surface optimization to indentify the region of process operating conditions where four or all five CQAs of the product met their specifications simultaneously. The calculated probability by this approach provided the figure to define the reliable operating ranges. The approach presented here can be extended to other fermentation process optimization and the definition of the process control space.