Epigenetic plasticity and chemoresistance in cancer: mechanisms, biomarkers, and translational opportunities for real-world evidence

Chemoresistance remains a major barrier to durable cancer control and is increasingly understood as a dynamic process shaped not only by genetic selection, but also by epigenetically regulated changes in cellular state. Evidence supports a model in which a subset of tumor cells survives treatment through drug-tolerant persister states characterized by slow cycling, stress tolerance, transcriptional rewiring, and altered interactions with the tumor microenvironment. In this context, chromatin remodeling, histone-state regulation, chromatin accessibility, enhancer reprogramming, and lineage plasticity emerge as central mechanisms enabling adaptive survival under therapeutic pressure and facilitating transition toward more stable resistant phenotypes. These mechanisms also provide a biological rationale for epigenetic therapies as priming, combination, or resensitization strategies. Clinical and translational studies suggest that targeting epigenetic regulators may help restore treatment susceptibility, although current evidence remains heterogeneous, with variable regimens, endpoints, and biomarker sampling strategies. Epigenomic biomarkers may therefore be particularly valuable for identifying adaptive cell states, monitoring target engagement, and tracking resistant trajectories over time. Real-world data and real-world evidence can complement mechanistic and clinical studies by capturing treatment sequencing, heterogeneous populations, and post-approval effectiveness and safety patterns. However, their translational value depends on fit-for-purpose design, analytical validity, transparent provenance, and bias-aware methods, particularly in sequential treatment settings prone to confounding, endpoint misclassification, and non-random molecular testing. In this mini-review, we examine how epigenetic plasticity drives chemoresistance and how epigenomic biomarkers and real-world data may support clinical research and more rigorous evidence generation.