Accelerated drug development using a digital formulator and a self-driving tableting data factory

Abstract Advances in drug discovery and clinical research have shifted the bottleneck in medicines development to chemistry, manufacturing, and controls activities, a critically step for regulatory approval. This includes formulation and process development of a new drug product, which traditionally requires extensive resources, often leading to suboptimal outcomes. These development processes must adapt to follow the advances in drug discovery and clinical research and ultimately shorten timelines while ensuring product quality and safety. In this work, we present an integrated platform for tablet formulation and process development that couples a digital formulator, an in-silico optimisation tool using a predictive material-to-tablet model, with a self-driving tableting data factory, which applies Bayesian optimisation within an automated, fully integrated per-tablet manufacturing to testing workflow. The results demonstrate a reduction in the time from material characterisation to in-specification tablets to 6 h and a reduction in API material use by 65% compared to current state-of-the-art methods.