Background: The prognostic significance of histological transformation (HT) in treatment-naive diffuse large B-cell lymphoma (DLBCL) remains controversial. This study aimed to evaluate the clinical outcomes and failure patterns of treatment-naive transformed DLBCL (trDLBCL) compared with de novo DLBCL using a large-scale cohort. Methods: We retrospectively analyzed 1735 consecutively enrolled treatment-naive DLBCL patients (118 trDLBCL and 1617 de novo). Propensity score matching (PSM) was performed to balance baseline characteristics. Survival outcomes were assessed using Kaplan–Meier and Cox proportional hazards models. Subgroups were defined by pathology (t-FL vs. t-MZL) and pattern: concurrent (synchronous indolent lymphoma and DLBCL components at diagnosis)vs. pure transformation (DLBCL occurring as the sole histology in patients with a prior history of untreated indolent lymphoma). Results: In the overall cohort, trDLBCL was associated with significantly inferior progression-free survival (PFS) compared with de novo disease and remained an independent adverse prognostic factor in multivariable analysis (HR 1.754, p < 0.001). These findings were confirmed in a 1:1 propensity score-matched cohort (108 pairs), where trDLBCL continued to show worse PFS (p < 0.01), while overall survival (OS) was comparable (p = 0.99). Within trDLBCL patients, the underlying indolent subtype (t-FL vs. t-MZL) did not significantly affect survival (PFS p = 0.17, OS p = 0.35), whereas “pure transformation” was associated with markedly inferior PFS (p = 0.005) and OS (HR 2.56, p = 0.02) compared with concurrent transformation. Failure pattern analysis revealed a higher risk of early progression in trDLBCL (POD24: 30.56% vs. 18.52%; OR 1.94, 95% CI: 1.05–3.56), whereas central nervous system (CNS) involvement was low and comparable between groups (2.78% vs. 0.93%, p = 0.62). Conclusions: Treatment-naive trDLBCL is associated with inferior PFS driven by early progression, whereas OS is comparable due to effective salvage therapies. Pure transformation appeared to define a higher-risk subgroup with inferior disease control, supporting the need for future prospective studies to evaluate risk-adapted frontline, consolidation, or maintenance strategies.
Chai et al. (Tue,) studied this question.