Design-oriented regression models for H2O2 decontamination processes in sterile drug product manufacturing considering rapidity and sterility

Abstract

We developed regression models for designing rapid and effective H2O2 decontamination processes in the manufacturing of sterile drug products such as injectables. Decontamination, which is typically performed by using H2O2, is a critical changeover process used to establish a sterile environment for filling products. In the process, there is a trade-off relationship between the duration of the process and the level of sterility assurance that needs to be considered in the design. Our model defines these two items as objective functions and the parameters describing the profile of H2O2 injection and the initial humidity as design variables. Our model also considers aeration, i.e., removal of H2O2 from the environment, as a part of the entire process. This design-oriented modeling considering the entire process is the novelty of the work. Experiments were performed using an industrial isolator to develop and validate a set of regression models that describe the relationship between the design variables and the objective functions. In the application of the model, Pareto-optimal conditions could be indicated given the target H2O2 concentration in aeration, which is useful for the makers of drugs and/or isolators for designing decontamination processes.