Prognostic significance of quantitative EBV biomarkers in extranodal NK/T-cell lymphoma: a meta-analysis of EBV DNA load and EBER-positive cell proportion

Pre-treatment peripheral blood EBV DNA and tumor EBV-encoded small RNA (EBER) are recognized prognostic factors in Extranodal NK/T-Cell Lymphoma (ENKTL). However, their quantitative values and optimal positive thresholds remain inconsistent, hindering clinical application. This meta-analysis systematically assesses their prognostic value in ENKTL. Relevant studies from PubMed, Embase, Web of Science, and Cochrane Library were searched until 24 December, 2025. Eligible studies reported hazard ratios (HRs) for overall survival (OS) based on Pre-treatment EBV DNA levels or tumor EBER positivity. Subsequently, Pooled HRs with 95% confidence intervals (CIs) were calculated via random-effects models. Next, Subgroup analyses examined different blood compartments, EBV DNA cut-offs, and study design. Additionally, Heterogeneity (I²) and publication bias (Egger’s test) were assessed. In total, 17 studies from 15 articles were enrolled in this meta-analysis, including 15 on EBV DNA and 2 on EBER. These studies demonstrated that elevated pre-treatment EBV DNA level (HR = 3.04, 95% CI 1.97–4.69; P < 0.001, I2 = 38%) and high tumor EBER positivity thresholds of ≥ 75% (HR = 5.53, 95%CI2.21-14.38; P < 0.001, I2 = 0%) independently predicted reduced survival. Moreover, Stratification by EBV DNA cut-off (500 copies/ML) showed significantly higher mortality risk in the ≥ 500 group (HR = 2.82, 95% CI 1.84–4.31; P < 0.001) compared to the < 500 group (HR = 2.66, 95% CI 0.56–12.72; P = 0.22). Furthermore, Egger’s test suggested no publication bias(P = 0.111). Finally, Sensitivity analyses suggested the robustness and credibility of our results. High pre-treatment EBV DNA load (cut-off of ≥ 500 copies/mL) and high EBER-Positive Cell Proportion (≥ 75%) predict poor survival in ENKTL. Those quantitative EBV Biomarkers may serve as practical biomarkers for risk stratification and guiding therapeutic decisions in this patient population.