AI-Driven Drug Repurposing: A Case-Based Review WithInsights From Traditional Approaches.

Abstract

Drug repurposing is an efficient strategy to identify new uses for existing drugs, reducing the time and cost typically associated with novel drug development. Traditionally, many repurposed drugs were discovered serendipitously or through clinical observation processes that, while valuable, are often slow and limited in scope. In recent years, integrating artificial intelligence (AI) into drug repurposing has transformed the landscape by enabling data- driven predictions and rapid hypothesis generation. This review explores AI-driven drug repurposing through a series of impactful case studies, including Baricitinib for COVID-19, Ketamine for cocaine use disorder, Efavirenz for Parkinson’s disease, and others. These cases highlight how various AI technologies such as machine learning, natural language processing, and electronic health record mining have been used to uncover novel therapeutic indications. To enhance the value of this analysis, the review also briefly contrasts AI-based approaches with traditional repurposing examples such as Sildenafil and Thalidomide, illustrating the evolution from chance discoveries to algorithm-driven insights. By presenting these comparative case studies, this article underscores the potential of AI to revolutionise drug repurposing specifically. It concludes that while AI is not a replacement for clinical expertise, it is a powerful tool that enhances drug repurposing efforts, enabling more targeted, faster, and scalable therapeutic innovations.