There is evidence for a causal role of Epstein-Barr virus (EBV), a known carcinogen, in an increasing number of lymphoma subtypes, but a systematic evaluation of EBV attributable fraction is lacking. We conducted a systematic review and meta-analysis of EBV prevalence in lymphoma subtypes from 1990 to 2024, stratified by HIV status and geographical region. For each subtype, we calculated pooled prevalence and corresponding 95% confidence intervals (CIs) of EBV, measured by EBV-encoded RNA in situ hybridisation. 307 eligible studies, including > 21,140 lymphoma cases, were included. In HIV-negative persons, EBV pooled prevalence was 10.8% (95% CI 8.6%-13.5%) in Diffuse large B-cell lymphoma (DLBCL), 45.8% (35.2%-56.8%) in Burkitt lymphoma (BL), 53.0% (46.9%-58.9%) in Hodgkin lymphoma, 52.5% (29.6%-74.4%) in plasmablastic lymphoma, 92.4% (83.3%-96.7%) in extranodal NK/T-cell lymphoma (ENKTL), and 5.1% (2.7%-9.4%) in follicular lymphoma. EBV prevalence was higher among tumors diagnosed among persons living with HIV (PLHIV), for DLBCL (43.6% (95% CI 31.5%-56.5%)), BL (43.3% (27.9%-60.2%)), and PBL (79.4% (65.6%-88.6%)), compared to HIV-negative tumors. BL was the only subtype with evidence of heterogeneity in EBV positivity by region, with near-universal EBV positivity in East African BL. This provides the first systematic and comprehensive evidence on EBV's substantial contribution to lymphoma, highlighting its particular importance in lymphomas arising in PLHIV. These data can contribute to estimating the complete burden of EBV-related disease and inform public health and clinical strategies, including vaccines, early diagnosis, or targeted therapies.
Hirabayashi et al. (2026) studied this question.