Prognostic value of tissue-based DNA methylation biomarkers in non-small cell lung cancer: a systematic review

Abstract

Abstract Background Non-small cell lung cancer remains a leading cause of cancer-related death worldwide, with varied outcomes despite advances in therapy. Epigenetic modifications, especially DNA methylation, are crucial regulators of gene expression and have been linked to NSCLC initiation, progression, and prognosis. Although previous reviews have focused on methylation biomarkers for early detection or risk prediction, tissue-based DNA methylation biomarkers for post-diagnosis prognostication in NSCLC remain lacking. This systematic review synthesizes the prognostic value of tissue-based DNA methylation biomarkers in NSCLC and evaluates their association with overall survival (OS) and disease-free survival (DFS). Methods A systematic search of PubMed, Medline, Embase, Web of Science, Cochrane CENTRAL, and grey literature from database inception to December 2025 was conducted. Studies reporting tissue-based DNA methylation and survival outcomes in NSCLC patients were included. Two reviewers independently screened studies, extracted data, and assessed risk of bias. Due to heterogeneity in biomarkers and outcomes, a narrative synthesis was performed. Results Of 4,235 records, 20 studies conducted between 2003 and 2025 were included, examining 28 methylation biomarkers. Most associations (77.5%, 31/40) indicated that methylation was associated with worse survival outcomes. Key negative prognostic markers included miR-34b/c (HR = 3.34), EGLN2 (HR = 1.71), and TGFBI (HR = 2.88). Inconsistent results were found for DAPK and SPARC with reports of both adverse (HR = 4.10–8.26) and protective (HR = 0.61–0.73) effects. No two studies analyzed the same biomarker outcome pairs, preventing meta-analysis. The risk of bias was mostly moderate, particularly in areas like prognostic factor measurement and study dropout. Discussion The predominance of adverse associations (77.5%) supports epigenetic silencing as a driver of aggressive NSCLC, though inconsistencies for markers like DAPK and SPARC underscore context-dependent effects. Fragmented evidence and methodological heterogeneity precluded meta-analysis, with moderate bias in key domains limiting comparability. Standardization of assays and prospective validation in stratified cohorts are essential to translate these biomarkers into clinical practice. Conclusion Tissue-based DNA methylation is frequently associated with adverse prognosis in NSCLC, but evidence is fragmented and methodologically heterogeneous. Standardization of assays and collaborative validation in prospective cohorts are needed to translate these epigenetic biomarkers into clinically useful tools.