AI-Enabled Drug Repurposing: Accelerating the Discovery of New Therapeutic Uses

Abstract

ABSTRACT

Objective To critically examine how artificial intelligence (AI) is being used to identify new therapeutic uses for existing or investigational medicines, with emphasis on validation, reproducibility, clinical translation and regulatory considerations.

Methods A structured narrative literature search was conducted on 18 October 2026 using PubMed/MEDLINE, publisher platforms for peer-reviewed biomedical literature, and U.S. Food and Drug Administration (FDA) regulatory sources. Searches covered drug repurposing/repositioning combined with artificial intelligence, machine learning, deep learning, natural language processing, knowledge graphs, graph neural networks, generative AI, validation and clinical translation. Priority was given to primary studies, comparative benchmarking studies, authoritative reviews and current regulatory documents. Title/abstract screening was followed by full-text assessment for relevance; reference-list citation chasing was used to identify additional key studies. This was a narrative review and was not designed as a PRISMA systematic review or meta-analysis.

Findings: AI can integrate heterogeneous molecular, omics, literature and clinical data and can prioritize drug–disease associations. However, reported performance is highly dependent on dataset construction and evaluation design. TxGNN, a graph foundation model, improved zero-shot indication and contraindication prediction relative to benchmark methods, whereas recent benchmarking studies show that random or transductive splits can inflate performance because of similarity, redundancy and information leakage. AI-assisted hypotheses have reached experimental and clinical validation in selected examples, notably baricitinib for COVID-19 and generative-AI prioritization of Alzheimer disease candidates using real-world electronic health records; these examples demonstrate different levels of evidence and should not be interpreted as proof that computational predictions alone establish efficacy.

Conclusion AI-enabled drug repurposing is best viewed as a hypothesis-generation and candidate-prioritization approach. Its translational value will depend on leakage-resistant external validation, mechanistic and pharmacokinetic feasibility, prospective clinical evaluation, transparent reporting, regulatory engagement and multidisciplinary review involving pharmacists, pharmacologists and clinicians.

Keywords

Authors & affiliations

Cite this article

Shubham Joshi, Prince Dadhania, Rushil Barad, Hetvi Baraiya. AI-Enabled Drug Repurposing: Accelerating the Discovery of New Therapeutic Uses. ClinSphere Journals. 2026;1(1). https://doi.org/10.5281/zenodo.23558080

1. Ashburn TT, Thor KB. Drug repositioning: identifying and developing new uses for existing drugs. Nat Rev Drug Discov. 2004;3(8):673-683. doi:10.1038/nrd1468.
2. Pushpakom S, Iorio F, Eyers PA, Escott KJ, Hopper S, Wells A, et al. Drug repurposing: progress, challenges and recommendations. Nat Rev Drug Discov. 2019;18(1):41-58. doi:10.1038/nrd.2018.168.
3. Vamathevan J, Clark D, Czodrowski P, Dunham I, Ferran E, Lee G, et al. Applications of machine learning in drug discovery and development. Nat Rev Drug Discov. 2019;18(6):463-477. doi:10.1038/s41573-019-0024-5.
4. Ghandikota SK, Jegga AG. Application of artificial intelligence and machine learning in drug repurposing. Prog Mol Biol Transl Sci. 2024;205:171-211. doi:10.1016/bs.pmbts.2024.03.030.
5. Wan Z, Sun X, Li Y, Chu T, Hao X, Cao Y, et al. Applications of artificial intelligence in drug repurposing. Adv Sci (Weinh). 2025;12(14):e2411325. doi:10.1002/advs.202411325.