Mathematical models of dissolution testing: Challenges and opportunities toward real-time release testing

Real-time release testing (RTRt) of tablet dissolution can significantly improve manufacturing efficiency along with the adoption of continuous manufacturing in the pharmaceutical industry. To assure product quality without destructive testing, models for RTRt should be sufficiently reliable and robust. Whereas mechanistic models have merits of broader applicability and interpretability, data-driven models have been common approaches due to computational speed. This paper discusses challenges and opportunities in the application of mechanistic models for dissolution testing to enable RTRt of solid dosage. After a comprehensive literature review on mechanistic dissolution models and RTRt, the potential benefits and challenges of mechanistic models are presented. Compared to data-driven models, mechanistic models require less experimental data that can reduce time and cost for RTRt development. However, to enable the implementation of mechanistic models in RTRt, computational time should be short either by using a simple mechanistic model or by applying surrogate models.