Abstract / Description
Breast cancer therapy is limited by heterogeneous tumor states and adaptive resistance that may be missed by single-layer biomarkers. This review focuses on the clinically actionable integration of multi-omics in breast cancer, emphasizing therapeutic stratification, response prediction, resistance monitoring, and implementation readiness. We discuss how genomic, epigenomic, transcriptomic, proteomic, metabolomic, spatial, and liquid-biopsy data can be used jointly to infer functional tumor states; how proteogenomics and surfaceomics refine pathway and antibody-drug conjugate (ADC) target interpretation; and how cell-free DNA (cfDNA) methylation, fragmentomics, foundation models, and digital twins could support longitudinal decision-making. We also highlight the key constraints for translation, including assay standardization, external validation, clinical utility, economic feasibility, and the cautious interpretation of fragmentomic and artificial intelligence (AI)-derived predictions.