Abstract
Breast cancer faces significant therapeutic challenges, particularly for triple-negative breast cancer (TNBC), due to limited targeted therapies and drug resistance. Drug repositioning leverages existing safety and pharmacokinetic data to expedite the identification of new indications with cost-effective benefits compared to de novo drug discovery. In this critical narrative review, we examine recent advances in computational repositioning strategies for breast cancer, focusing on network-based methods, computer-aided drug design, artificial intelligence and machine learning, transcriptomic signature matching, and multi-omics integration. We highlight key case studies that have progressed to preclinical validation or clinical evaluation. We assess comparative performance metrics, experimental validation outcomes, and real-world success rates. We also present critical methodological challenges, including data heterogeneity, bias in real-world data, and the need for study reproducibility. Our review emphasizes the importance of window-of-opportunity trials and the need for standardized data sharing and reproducible pipelines. These insights highlight the groundbreaking potential of in silico repositioning in addressing unmet needs in breast cancer therapy.