Epcoritamab monotherapy provides superior efficacy vs non–anthracycline-containing regimens in newly diagnosed elderly DLBCL patients deemed unsuitable for anthracycline-containing regimens: A match-adjusted comparative efficacy analysis

Abstract

Abstract Introduction: Newly diagnosed diffuse large B-cell (DLBCL) patients (pts) who are older and/or have specific comorbidities may be unsuitable for standard-of-care anthracycline-containing (AC) chemoimmunotherapies (CITs), eg, increased risk of cardiovascular toxicities. These pts are often treated with non-AC regimens. In the EPCORE DLBCL-3 (NCT05660967) trial, epcoritamab (epcor) monotherapy (mono) showed compelling efficacy and manageable safety in pts with newly diagnosed DLBCL deemed unsuitable for AC CIT (Morschhauser F et al. Blood. 2024;144:867). This analysis contextualized epcor mono vs common non-AC regimens in newly diagnosed elderly DLBCL pts unsuitable for AC regimens. Methods: Newly diagnosed DLBCL pts (2007–2017) were identified from the US National Cancer Institute Surveillance, Epidemiology, and End Results (SEER) database and linked to the US Medicare database for treatment details using diagnosis codes (ICD-O-3: 9680/9688). Individual pt data for first-line (1L) epcor mono from the EPCORE DLBCL-3 trial were match-adjusted and compared with pts treated with commonly used 1L non-AC regimens from US clinical practice. Non-AC regimens were categorized into 3 non–mutually exclusive cohorts: (a) non-AC CIT; (b) non-AC CIT + palliative management (PM); and (c) non-AC CIT + PM + untreated pts (UP). Non-AC CITs included rituximab (R)-cyclophosphamide, doxorubicin, vincristine, prednisone [R-CEOP]/R-cyclophosphamide, vincristine, prednisone [R-CVP]/R-cyclophosphamide, etoposide, prednisone, procarbazine [R-CEPP]), bendamustine and rituximab (BR), and other combination regimens. PM included treatment with single agents (steroids, R-mono, and non-AC chemotherapy). Pts with no claims for DLBCL treatment were included in the UP cohort. Eligibility criteria for the EPCORE DLBCL-3 trial were applied to the non-AC regimen cohort from SEER-Medicare to ensure enhanced comparability of pt populations. Inverse probability of treatment weighting (IPTW) was used to balance all treatment cohorts on age, gender, Ann Arbor stage, high-risk cardiovascular (HRCV) status (Lyon AR et al. Eur J Heart Fail 2020;22:1945), and postdiagnosis time-to-treatment. The primary endpoint was overall survival (OS). Weighted Cox proportional-hazard (PH) regression models estimated hazard ratios (HRs); any residual imbalances in pt characteristics after IPTW (ie, standardized difference >0.1) were further adjusted as independent variables in the Cox PH model. Time-to-event Kaplan-Meier curves were compared using log-rank tests. Subgroup analyses included R-CEOP/R-CVP/R-CEPP and BR treatment groups. Results: Of 49,110 newly diagnosed DLBCL pts in the SEER-Medicare cohort, 4345 (8.8%) were not treated with AC regimens. Among these pts, 1083 (24.9%) received non-AC CIT; 708 (16.3%) received PM, and 2544 (58.6%) qualified as UP. After applying the EPCORE DLBCL-3 trial eligibility criteria, 781 SEER-Medicare pts (non-AC CIT, PM, and UP) were eligible for inclusion in the SEER-Medicare comparator cohort vs epcor (N=45).Before adjustment, the epcor mono cohort had fewer pts aged ≥80 years (82.2% vs 84.4%), fewer pts with Ann Arbor stage III/IV (66.7% vs 71.5%), and fewer pts with HRCV status (91.1% vs 95.7%) than non-AC CIT pts. Similar trends in baseline characteristics were seen when PM and UP were included with non-AC CIT. Median follow-up was 9.0 mo for the epcor mono cohort and 2.2 mo for the non-AC regimen cohort. After match adjustment, the cohorts were largely balanced. Adjusted median OS (mOS) was significantly better in epcor mono (mOS not reached) vs non-AC CIT pts (mOS 15.6 mo; HR 0.44 [95% CI 0.21–0.90]; P=0.024). Consistently superior survival was observed for epcor mono vs non-AC CIT + PM (mOS 9.5 mo; HR 0.35 [95% CI 0.18–0.69]; P=0.002) and non-AC CIT + PM + UP (mOS 2.2 mo; HR 0.13 [95% CI 0.07–0.25]; P<0.0001). Subgroup analyses showed consistent findings with superior survival for epcor mono vs specific non-AC CIT regimens; 53.8% received R-CEOP/R-CVP/R-CEPP (mOS 18.3 mo; HR 0.47 [95% CI 0.22–0.98]; P=0.043) and 26.9% received BR (mOS 12.4 mo; HR 0.44 [95% CI 0.20–0.96]; P=0.040). Conclusions: Fixed-duration epcor mono demonstrated significantly superior OS vs non-AC regimens in newly diagnosed elderly and/or frail DLBCL pts deemed unsuitable for anthracyclines. These findings support epcor mono as a potential 1L chemo-free treatment option for newly diagnosed DLBCL pts unsuitable for AC regimens.