Odronextamab treatment for patients with rare subtypes of relapsed/refractory (R/R) aggressive B-cell non-Hodgkin lymphoma (B-NHL): Updated efficacy and safety analyses from a dedicated cohort of the ELM-2 study

Abstract Introduction Aggressive B-NHL includes rare subtypes such as transformed diffuse large B-cell lymphoma (t-DLBCL), high-grade B-cell lymphoma not otherwise specified (HGBL NOS), primary mediastinal B-cell lymphoma (PMBL), and follicular lymphoma (FL) Grade (Gr) 3b, for which prospective clinical studies are lacking. Patients (pts) with these subtypes tend to have poor prognoses in the R/R setting (e.g., real-world data show a 2-year overall survival [OS] rate of 18% in pts with R/R HGBL NOS [Zayac et al. Blood Adv 2023]). The efficacy and safety of odronextamab, a CD20×CD3 bispecific antibody, were reported in pts with heavily pretreated R/R DLBCL in the Phase 2 ELM-2 study (NCT03888105). ELM-2 includes a cohort with other aggressive B-NHL subtypes, which has reported encouraging preliminary efficacy and safety results (Bachy et al. ASH 2024). Here we present updated data from this cohort. Methods Pts aged ≥18 years with ECOG PS 0–1, adequate organ function, and aggressive B-NHL subtypes (excluding DLBCL and mantle cell lymphoma) that were R/R after ≥2 prior lines of systemic therapy were eligible. In Cycle (C)1 (21-day cycles), intravenous (IV) odronextamab was given with steroid premedication and in weekly step-up doses, which were optimized (original step-up regimen: 1/20 mg; modified: 0.7/4/20 mg) to help mitigate cytokine release syndrome (CRS) risk. Pts received odronextamab 160 mg on Days 1, 8, and 15 of C2–4, then maintenance dosing of 320 mg Q2W post-C4 until disease progression or unacceptable toxicity. Pts with a complete response (CR) for ≥9 months switched to 320 mg Q4W. Infection prophylaxis was recommended, including PJP prophylaxis for all pts and IV immunoglobulin supplementation and antivirals if indicated. Primary endpoint was objective response rate (ORR) per Lugano criteria by independent central review (ICR). Secondary endpoints included ORR by local investigator assessment; CR, duration of response (DOR), and progression-free survival (PFS) per Lugano criteria; OS, and safety. Biomarker analyses were exploratory endpoints. Results At data cutoff (May 5, 2025), 61 pts with other R/R aggressive B-NHL had received odronextamab; most frequent subtypes were t-DLBCL (n=10), HGBL NOS (n=10), PMBL (n=9), and FL Gr 3b (n=7). Median follow-up was 20.2 months. Overall, the median age was 59.0 years (range 24–83) and 57.4% of pts were male. Pts had received a median of 3 (range 2–9) prior lines of therapy, with 68.9% refractory to their last therapy, and 62.3% double refractory to an anti-CD20 antibody and an alkylator. Median odronextamab exposure was 18.9 weeks (range 1–158); at data cutoff, 15 pts (24.6%) remained on treatment. Among efficacy-evaluable pts (n=58), ORR and CR rate by local investigator assessment were 63.8% and 43.1%, respectively. Responses were durable, with a median DOR of 35.9 months and median duration of CR not reached (NR). ORR/CR rates were 70.0%/60.0% in pts with t-DLBCL, 60.0%/40.0% in HGBL NOS, 33.3%/11.1% in PMBL, and 100%/85.7% in FL Gr 3b. In the overall efficacy population, median PFS was 11.4 months, with a 24-month PFS rate of 35.6%. Median OS was 47.6 months, with a 24-month OS rate of 56.1%. In pts who achieved a CR, median PFS was 39.4 months and median OS was NR. In all pts (n=61), the most common any-grade treatment-emergent adverse events (TEAEs) were CRS (49.2%), anemia (41.0%), and neutropenia (composite term; 31.1%). Gr ≥3 TEAEs occurred in 82.0% of pts, the most common being neutropenia (26.2%) and anemia (18.0%). Nine pts (14.8%) discontinued treatment due to TEAEs. CRS occurred in 46.8% of pts (22/47) who received 0.7/4/20 mg dosing: Gr 1, 34.0% (n=16); Gr 2, 10.6% (n=5); Gr 3, 2.1% (n=1). Gr ≥3 infections occurred in 41.0% of pts (Gr 3/4, 34.4%; Gr 5, 6.6%); 24.6% had any-grade COVID-19 infection (Gr 5, n=1). A Gr 3 ICANS event was reported in one pt. Conclusions Longer follow-up confirms the efficacy of odronextamab in heavily pretreated pts with rare subtypes of R/R aggressive B-NHL. In the overall efficacy population, 64% of pts achieved a response (CR rate of 43%), and responses were durable with a median DOR of ~3 years. Deep responses were attained across subtypes, and outcomes were better in pts who achieved a CR. Odronextamab retained a generally manageable safety profile with continued treatment, and no new safety signals were recorded over this longer follow-up period. Updated data, including ICR-assessed response and biomarker analyses, will be presented.