PD-1-based combination therapy versus monotherapy in relapsed or refractory classical Hodgkin lymphoma: a single-center retrospective cohort study

Background Programmed cell death 1 (PD-1) inhibitors are active in relapsed or refractory classical Hodgkin lymphoma (R/R cHL), but evidence directly comparing combination therapy with monotherapy in routine practice is limited. Methods This single-center retrospective cohort included 123 consecutive patients treated from 2019 through 2025 with PD-1-based combination therapy (n=98) or PD-1 monotherapy (n=25). Responses were assessed using Lugano 2014 criteria. Progression-free survival (PFS) counted progression or death as events. Kaplan-Meier, baseline-only Cox regression, and overlap-weighted propensity score analyses were used. Response endpoints were additionally analyzed with multivariable and overlap-weighted logistic regression, and the robustness of the PFS association was examined in a sensitivity series and with E-values. Results Treatment groups differed substantially at baseline; monotherapy recipients were older and had worse performance status. Unadjusted complete response (CR) was higher with combination therapy (65/98, 66.3% vs 9/25, 36.0%; P = 0.011), but this difference was attenuated and no longer significant after adjustment (overlap-weighted odds ratio [OR] 1.91; 95% CI, 0.67-5.45). The objective response advantage persisted (87/98, 88.8% vs 15/25, 60.0%; overlap-weighted OR 4.57; 95% CI, 1.36-15.43). With 45 PFS events, the unadjusted HR for combination therapy versus monotherapy was 0.27 (95% CI, 0.15-0.49); the baseline-adjusted HR was 0.44 (95% CI, 0.21-0.89). The overlap-weighted HR was 0.45 (95% CI, 0.23-0.91). The PFS association was stable across an extended sensitivity series (HR range, 0.40-0.48) and among the 95 patients who did not undergo consolidative autologous stem cell transplantation (adjusted HR 0.42; 95% CI, 0.20-0.89). Grade 3–4 neutropenia occurred in 27/98 (27.6%) combination-treated patients and 1/25 (4.0%) monotherapy-treated patients. Regimen subgroup analyses did not establish efficacy equivalence. Across eight prespecified maximum-grade safety fields, at least one recorded grade >=3 event occurred in 20/35 (57.1%) with PD-1+GVD, 5/32 (15.6%) with PD-1+brentuximab vedotin and 8/31 (25.8%) with PD-1+bendamustine (P = 0.001). Conclusions PD-1-based combination therapy was associated with a higher objective response rate and longer PFS than monotherapy, whereas the complete response difference did not persist after adjustment. The retrospective design, small monotherapy group, and potential unmeasured confounding mean that further research is needed to confirm this conclusion.