The predictive edge: modeling and simulation in drug product development

It is well-known that drug development is challenging and a time- and resource-intensive endeavor. Historically, it has relied heavily on trial-and-error, empirical approaches that yield a low probability of success. Despite continuous efforts to improve efficiency across the development stages the overall success rate from clinical trial initiation to market approval remains low. In response to these challenges, in-silico predictive modeling and simulations are becoming indispensable tools for accelerating and de-risking the drug product development process. These computational methods use simulated and real-world data to guide decision-making across the entire development pipeline. Notably, these tools are now gaining widespread acceptance not only in discovery but also across the delivery and formulation stages of drug development. Advances in artificial intelligence (AI) and machine learning (ML) are proving transformative, enabling rapid analysis of large datasets and the development of predictive models that enhance classification, prediction, and optimization capabilities across the drug product development process. This review provides an overview of the various in-silico predictive modeling and simulation techniques for drug product development, emphasizing the use of AI/ML, and their applications in drug delivery. We highlight their role in improving drug performance, manufacturability, stability, safety, and overall success from clinical development through commercialization.