Real-world outcome and dose intensity of Pola-R-CHP versus R-CHOP/R-THP-COP in newly diagnosed diffuse large B-cell lymphoma: a single-center, retrospective cohort study

Following POLARIX, Pola-R-CHP has been increasingly adopted as frontline therapy for diffuse large B-cell lymphoma (DLBCL). However, real-world evidence comparing Pola-R-CHP with conventional regimens regarding dose-intensity profiles, toxicity, and survival outcome remains limited. We performed a single-center retrospective study with de novo DLBCL treated between 2014 and 2025. Initial dose intensity (IDI) was calculated for each chemotherapeutic agent delivered at the first cycle, and relative dose intensity (RDI) was calculated for each chemotherapeutic agent through the entire treatment period. Average IDI and average RDI were calculated by average values across cyclophosphamide (CPA), doxorubicin or THP (DXR/THP), and the microtubule inhibitor (vincristine [VCR] or polatuzumab vedotin [Pola]). Primary outcome was progression-free survival (PFS). PFS was analyzed using Kaplan–Meier estimates and multivariable Cox hazards models, including both the conventional proportional model and a restricted cubic spline (RCS)–based model. Among the enrolled 112 patients, forty-seven patients were received Pola-R-CHP regimen. Median age was 76 years; 37 patients (33%) were aged ≥ 80 and 27 patients (24.1%) had ECOG PS ≥ 2. For CPA and DXR/THP, mean IDI and RDI were similar between groups; in contrast, Pola maintained higher IDI and RDI than VCR, with density peaks near 100%, indicating more consistent delivery. Grade ≥ 3 adverse events, including febrile neutropenia (FN) and non-hematologic toxicities, and treatment-related mortality were comparable. Any-grade PN leading to dose reduction/discontinuation of microtubule inhibitors occurred in 13.9% with conventional therapy versus 2.1% with Pola-R-CHP (p = 0.043). Over a median follow-up of 17.5 months (range, 0.03–132.7), 36 patients (32.1%) died and 8 (7.2%) relapsed/progressed. In multivariable analysis, International Prognostic Index and average RDI were independent predictors of PFS, whereas treatment regimen was not. An RCS-Cox model showed no significant regimen–RDI interaction (p for interaction = 0.761), indicating that higher average RDI was associated with improved PFS irrespective of regimen. Pola-R-CHP demonstrated comparable efficacy and safety to conventional regimens while maintaining higher dose intensity as a microtubule inhibitor. Sustaining adequate treatment intensity was independently associated with improved PFS, underscoring the clinical relevance of Pola-R-CHP for optimizing real-world DLBCL management, including older or frail patients.