Researchers outline a path to finding faster, cheaper, and more targeted treatment options for complex breast cancer subtypes by combining genomics, explainable AI, and drug repurposing strategies.
Research: Synergy of AI and genomics for drug repurposing in breast cancer: an interpretability-driven framework. Image credit: ProStockStudio / Shutterstock
In a recent review published in a magazine npj genome medicinea group of authors explored how artificial intelligence (AI)-powered genomic analysis could support drug repurposing strategies for precision treatment of breast cancer.
background
Breast cancer remains one of the most common and deadly cancers, with an estimated 2.3 million new cases and 670,000 deaths worldwide in 2022. Developing new breast cancer drugs typically takes more than 10 years and billions of dollars, making it a challenge to quickly bring new treatments to patients. AI and genomic technologies can help reveal how genes, drugs, and pathways are associated with breast cancer. These will help researchers identify existing drugs that can be used to treat more aggressive breast cancer subtypes, providing patients with faster and cheaper treatment options. However, further research is needed to confirm that these predictions are accurate, clinically interpretable, and applicable to diverse patient populations.
Understanding the complexity of the breast cancer genome
Breast cancer is a complex disease characterized by large variations between different tumors in different patients and even between tumors in the same patient. The main molecular subtypes of breast cancer are luminal A and B, human epidermal growth factor receptor 2 (HER2)-enriched breast cancer, and triple-negative breast cancer (TNBC). Advances in genomics have enabled scientists to study multiple layers of biological information simultaneously, including deoxyribonucleic acid (DNA) variation, gene expression, protein expression, and epigenetic regulation.
AI in breast cancer research
AI has emerged as a powerful tool for managing the large amounts of genomic and clinical data generated in cancer research. Machine learning and deep learning models can identify patterns in large datasets, allowing researchers to group tumors, predict how they will respond to treatments, and identify new targets for therapeutic drug development.
Another application of machine learning in cancer research is determining whether genetic mutations are pathogenic. Machine learning models developed using large genomic databases are being developed to help classify variants of uncertain significance (VUS) and link pathogenic variants to potential treatment options.
Researchers can now use explainable AI (XAI) techniques such as SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explains (LIME) to better understand which genomic variables most influence AI model predictions. However, these post hoc tools do not prove causality and their output still requires biological validation.
Drug repurposing as a faster treatment strategy
Drug repurposing involves finding new uses for existing drugs. Approved drugs already have an established safety profile and can move into clinical trials more quickly than entirely new compounds. This strategy is particularly attractive in breast cancer, where treatment resistance remains a major challenge.
Anticancer activity has been found in many frequently prescribed drugs used to treat conditions other than cancer. Metformin, prescribed to treat diabetes, activates AMP-activated protein kinase (AMPK) and may inhibit PI3K/AKT/mTOR signaling in breast cancer models. Statins, prescription drugs used to alter lipid levels, inhibit HMG-CoA reductase and modulate the mevalonate pathway, which is associated with cancer cell proliferation and metastasis.
AI improves drug reuse by integrating genomic, pharmacological, and clinical data. Computational methods can be used to compare oncogene expression patterns and drug-induced molecular signatures to identify compounds that can reverse cancer-related pathways.
Graph neural networks and transformer-based AI systems may predict that existing drugs may be able to effectively target specific genomic abnormalities. Importantly, interpretability techniques help explain the biological mechanisms behind these predictions, increasing confidence in their potential clinical relevance.
Integrating AI, genomics, and precision oncology
Researchers propose an integrated framework that incorporates AI, genomics, and drug repurposing into a continuous precision oncology pipeline. In this model, patient-specific molecular data is analyzed using advanced AI systems to identify potential therapeutic targets and rank drug candidates based on biological plausibility and subtype relevance.
This framework focuses on mechanistic validation rather than relying solely on statistical associations. In the proposed framework, drug candidates will undergo pathway analysis, molecular docking, and experimental validation before clinical trials.
An important feature of the proposed framework is the feedback loop. Data from experimental and clinical research is reintegrated into AI models to enhance future predictions and support adaptive learning. The result is an iterative process in which new biological and clinical data are used to refine AI models to better predict cancer biology and patient response to treatment.
Challenges limiting clinical translation
Despite promising advances, several barriers continue to limit clinical implementation. Many genomic datasets lack diversity and are heavily biased toward populations of European descent, reducing the reliability of predictions for underrepresented groups. The authors warn that these biases could reduce model performance in Africa, Asia, Latin America, and other underrepresented populations, worsening inequalities in breast cancer treatment.
Validation is another key challenge for AI drug prediction. Most of them do not proceed from calculation to laboratory testing. Therefore, experimental testing using cell cultures, patient-derived organoids, and animal models remains essential to establish biological activity and pharmacological safety. Some computer-predicted drugs that have shown great promise, such as metformin and statins, have consistently failed to produce clear clinical effects in randomized or clinical studies.
conclusion
This review concludes that integrating AI and genomic analysis provides a promising strategy to promote drug repurposing in breast cancer. By combining multi-omics data with interpretable machine learning techniques, researchers can prioritize repurposing drug candidates and therapeutic hypotheses for testing. This advancement has the potential to shorten drug development cycles, reduce costs, and enable more efficient and personalized treatments. However, before these technologies can be used in clinical practice, it is essential to address challenges regarding data diversity, reproducibility, experimental validation, and ethical governance.
Click here to download your PDF copy.
Reference magazines:
- Ahmad, R. M., Abrouz, S., Ali, B. R., and Aldhaheri, N. (2026). Synergy of AI and genomics for drug repurposing in breast cancer: an interpretability-driven framework. npj genomic medicine. Doi: 10.1038/s41525-026-00578-9, https://www.nature.com/articles/s41525-026-00578-9
