Development and internal validation of an integrated prognostic model incorporating lymphocyte-to-monocyte ratio and circulating EBV-DNA in diffuse large B-cell lymphoma

Background Diffuse large B-cell lymphoma (DLBCL) is a highly heterogeneous malignancy with variable clinical outcomes. Although the International Prognostic Index (IPI) remains the cornerstone of risk stratification, it does not fully reflect host immune-inflammatory status or virus-associated biological features. This study aimed to identify the optimal systemic inflammatory biomarker and develop an integrated prognostic model incorporating the lymphocyte-to-monocyte ratio (LMR) and circulating Epstein–Barr virus (EBV) DNA for improved risk stratification in patients with DLBCL. Methods A retrospective cohort of 485 newly diagnosed patients with DLBCL was analyzed. Four systemic inflammatory biomarkers, including the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), LMR, and systemic immune-inflammation index (SII), were compared to identify the most informative prognostic biomarker. The optimal cutoff value for LMR was determined using maximally selected rank statistics. The prognostic associations of LMR and pretreatment circulating EBV-DNA status were assessed using Kaplan–Meier survival analysis and Cox proportional hazards regression models. An LMR–EBV-DNA risk score was developed and incorporated with the IPI to construct an individualized prognostic model, which was internally validated using bootstrap resampling. Results Among the evaluated inflammatory biomarkers, LMR demonstrated the best prognostic discrimination for overall survival (OS), outperforming NLR, PLR, and SII. Low LMR (cutoff: 2.84) was independently associated with inferior OS after adjustment for clinical variables. In the EBV-DNA-evaluable cohort, pretreatment circulating EBV-DNA positivity was independently associated with adverse OS and progression-free survival (PFS). The combined LMR–EBV-DNA risk score effectively stratified patients into low-, intermediate-, and high-risk groups with significantly different OS and PFS outcomes. Furthermore, incorporation of the LMR–EBV-DNA risk score into an IPI-based nomogram improved model discrimination, calibration, and clinical utility compared with IPI alone. The integrated model demonstrated superior prognostic performance and risk reclassification compared with the IPI alone. Conclusions We developed and internally validated an integrated prognostic model combining LMR as a systemic immune-inflammatory biomarker with circulating EBV-DNA status for patients with DLBCL. The LMR–EBV-DNA risk score provided complementary prognostic information beyond the conventional IPI system, demonstrating a modest incremental value in risk stratification. Although showing moderate discrimination and calibration, this accessible framework may serve as a potential complementary tool for prognostic assessment in DLBCL, warranting further external validation before clinical application.