Leveraging Generative AI to Prioritize Drug Repurposing Candidates: Validating Identified Candidates for Alzheimer's Disease in Real-World Clinical Datasets

Abstract

Drug repurposing represents an attractive alternative to the costly and time-consuming process of new drug development, particularly for serious, widespread conditions with limited effective treatments, such as Alzheimer's disease (AD). Emerging generative artificial intelligence (GAI) technologies like ChatGPT offer the promise of expediting the review and summary of scientific knowledge. To examine the feasibility of using GAI for identifying drug repurposing candidates, we iteratively tasked ChatGPT with proposing the twenty most promising drugs for repurposing in AD, and tested the top ten for risk of incident AD in exposed and unexposed individuals over age 65 in two large clinical datasets: 1) Vanderbilt University Medical Center and 2) the All of Us Research Program. Among the candidates suggested by ChatGPT, metformin, simvastatin, and losartan were associated with lower AD risk in meta-analysis. These findings suggest GAI technologies can assimilate scientific insights from an extensive Internet-based search space, helping to prioritize drug repurposing candidates and facilitate the treatment of diseases.

Figure 1. An illustration of the study design.

a, Employing iterative queries of ChatGPT to recommend twenty drugs for AD repurposing. b, Evaluating the potential efficacy of the ten most frequently suggested drugs using electronic health records (EHR) data from two large clinical databases.

Figure 2. Associations between exposure to ChatGPT-suggested drug repurposing candidates and AD risk.

Hazard ratios (HR) and 95% confidence intervals (CI) are shown for VUMC (blue squares), the NIH All of Us Research Program (red squares), and the combined meta-afinalysis (gray squares). ** indicates drugs associated with significantly reduced AD risk using VUMC data (p<0.05); * indicates drugs associated with significantly reduced AD risk in the meta-afinalysis (p<0.05). To ensure adequate statistical power, we did not report drugs with fewer than five AD cases in the study cohort (i.e., bexarotene and nilotinib in both VUMC and All of Us; minocycline, candesartan, rapamycin, and lithium in All of Us).