Early mortality and prognostic factors in diffuse large B-cell lymphoma presenting with hemophagocytic lymphohistiocytosis

Abstract Background Diffuse large B-cell lymphoma (DLBCL) complicated with hemophagocytic lymphohistiocytosis (HLH) represents a highly aggressive clinical subtype with diagnostic challenges. These patients frequently exhibit profound immune activation and organ dysfunction. Early mortality is common, and long-term outcomes remain poor. Currently, no established risk stratification tools or standardized treatment strategies are available for this population. Methods We retrospectively analyzed 130 patients diagnosed with DLBCL presenting with HLH at the First Affiliated Hospital of Zhejiang University between 2019 and 2024. Baseline characteristics, treatment regimens, and follow-up data were collected. Early death was defined as death within 30 days. Univariate and multivariate Cox regression models were used to identify independent predictors of early mortality. A preliminary risk score model was developed based on these factors. For long-term outcomes, patients were grouped according to initial treatment strategy: standard-intensity therapy (R-CHOP like or polatuzumab vedotin–R-CHP like), reduced-intensity therapy, or no lymphoma-specific therapy. Inverse probability of treatment weighting (IPTW) was applied to adjust for baseline differences. Overall survival (OS) and progression-free survival (PFS) were assessed. Results Among 130 patients, 25 died within 30 days, yielding an early mortality rate of 19.2%. All patients were Ann Arbor stage III–IV, with bone marrow involvement in 88.2% and non-GCB subtype in 89.5%. Multivariate analysis identified age ≥65 years, platelet count <50×10⁹/L (especially <20×10⁹/L), ferritin >3,500 μg/L (especially >10,000 μg/L), and procalcitonin >0.5 ng/mL as independent risk factors for early death. Based on these variables, a preliminary early mortality risk score was constructed to facilitate clinical risk stratification. For long-term outcomes, both OS and PFS were evaluated across different treatment groups. After IPTW adjustment, the standard-intensity group showed significantly improved OS compared to the reduced-intensity and untreated groups. Twelve-month OS rates were 81.8%, 17.6%, and 0%, respectively (log-rank P<0.001), with differences persisting across multiple time points. Notably, the survival benefit of standard therapy remained evident even in elderly patients and those with impaired organ function. Conclusions DLBCL patients presenting with HLH face high early mortality risk and heterogeneous long-term outcomes. Our study proposes a preliminary risk stratification model to aid clinical decision-making. Standard-dose immunochemotherapy, including R-CHOP and Pola-R-CHP, was associated with superior survival after adjustment, supporting its use even in high-risk subgroups.