Artificial Intelligence–Driven Drug Repurposing and Novel Target Discovery in Precision Oncology
Keywords:
Artificial intelligence, machine learning, deep learning, drug repurposing, target discovery, precision oncology, drug–target interaction, graph neural networks, foundation models, generative AI, multi-omics integration, biomarker discovery.Abstract
Cancer continues to impose a substantial global health burden, while conventional oncology drug discovery remains constrained by lengthy development timelines, escalating costs, and high rates of clinical failure. Artificial intelligence (AI), particularly machine learning (ML) and deep learning (DL), is transforming this landscape by enabling rapid identification of therapeutic candidates and novel molecular targets. This narrative review summarizes recent advances (2020–2026) in AI-driven drug repurposing and target discovery for cancer, highlighting computational approaches that integrate multi-omics data, including genomics, transcriptomics, epigenomics, proteomics, metabolomics, radiomics, and real-world clinical information. We discuss state-of-the-art methodologies such as supervised and unsupervised learning, graph neural networks, deep neural networks, transformer-based architectures, foundation models, and generative AI for predicting drug–target interactions, prioritizing biomarkers, and identifying repurposing opportunities for approved therapeutics. Across diverse malignancies, these approaches have demonstrated improved predictive performance, accelerated therapeutic discovery, and enhanced precision medicine strategies by facilitating personalized treatment selection and biomarker-guided interventions. Despite these advances, important challenges remain, including heterogeneous data quality, limited model interpretability, external validation, regulatory considerations, and integration into routine clinical workflows. Emerging developments in explainable AI, federated learning, multimodal foundation models, digital twins, and prospective clinical validation are expected to further strengthen the reliability and clinical applicability of AI-driven oncology drug discovery. Collectively, AI has the potential to fundamentally reshape precision oncology by accelerating the discovery of safe, effective, and personalized cancer therapies while reducing development costs and time to clinical translation.
