Bold claim: Adding odronextamab to CHOP in previously untreated DLBCL can yield deep, potentially durable responses, challenging expectations about standard therapy. But here’s where it gets controversial: does this rituximab-free approach truly outpace established regimens, and at what cost to safety?
Odronextamab, a CD20 × CD3-directed bispecific antibody, paired with CHOP chemotherapy, produced high rates of complete responses (CR) and encouraging signs of durability in adults with previously untreated DLBCL and high-risk features, according to early results from part 1 of the phase 3 OLYMPIA-3 trial (NCT06091865). Initial data from the 2025 ASH Annual Meeting showed objective response rates (ORR) of 78% with weekly odronextamab 80 mg plus CHOP and 100% with weekly odronextamab 160 mg plus CHOP. Complete responses were 44% for 80 mg and 100% for 160 mg. Neither median duration of response (DOR) nor progression-free survival (PFS) had been reached at this early analysis. Based on this efficacy-safety balance, the 160 mg dose was advanced for the randomized portion of the study comparing Odro-CHOP to R-CHOP.
Lead investigator Jean-Marie Michot, MD, highlighted a key finding: in part 1a of OLYMPIA-3, deep and durable responses were observed even without rituximab, suggesting that odronextamab plus CHOP can achieve strong outcomes in untreated DLBCL with high-risk features while maintaining a manageable safety profile compared with prior experiences. No new safety signals emerged relative to earlier reports.
Study design and patient population
OLYMPIA-3 is an open-label study conducted in two parts. Part 1 explored dose escalation and optimization of odronextamab with standard CHOP given on days 1 and 8 of each cycle, with odronextamab starting on day 8. Dosing began with a step-up 0.7/4/20 mg, then progressed to fixed-dose regimens including weekly 80 mg or 160 mg, and every-2-weeks 160 mg or 320 mg. Data reported thus far pertain to the weekly dosing cohorts. Part 2 will randomize patients to receive Odro-CHOP versus R-CHOP.
Across all participants in part 1, median age was 66 years (range 24–81), with 32% aged 75 or older. Performance status was 0 in 40%, 1 in 45%, and 2 in 14%. The primary cell of origin was non-GCB in 59% of cases, and all patients had de novo DLBCL. International Prognostic Index (IPI) scores were 3 in 36% and 4–5 in 27%. Lugano staging was III–IV in 95% of patients.
Dose-specific adherence and early outcomes varied slightly. In the 80 mg group (n = 9), 7 patients completed cycles 1–6; 2 discontinued early by physician decision. In the 160 mg group (n = 13), all completed cycle 1 and 84.6% completed cycle 6; 2 discontinued early by physician decision. Relative dose intensity was 87% for 80 mg and 77% for 160 mg. Overall, most patients completed six cycles, with few odronextamab dose reductions and no permanent discontinuations due to odronextamab-related adverse events. No clinically meaningful safety differences emerged between dose levels.
Efficacy signals beyond the primary endpoints
Follow-up at a median of roughly 9 months for the 80 mg cohort and 7.8 months for the 160 mg cohort showed that most responses persisted, with CRs appearing durable. Biomarker analyses revealed rapid B cell count declines after starting therapy: CHOP initiated an initial B cell drop, followed by complete B cell clearance upon odronextamab initiation. T cell margination occurred transiently after therapy initiation and was similar across dose levels.
Safety profile and adverse events
Grade 3 or higher treatment-emergent adverse events (TEAEs) affected all patients in both dose groups. Serious TEAEs occurred in 77.8% of the 80 mg cohort and 92.3% of the 160 mg cohort. TEAEs caused treatment interruptions or delays in 66.7% (80 mg) and 84.6% (160 mg). Dose reductions of odronextamab were rare, with no reductions in the 80 mg group and one reduction in the 160 mg group. Dose reductions or CHOP modifications due to TEAEs occurred in one and five patients, respectively, across the two groups. There was one death attributed to a TEAE in the 160 mg arm. Importantly, no dose-limiting toxicities were reported.
Common TEAEs across both doses included neutropenia (81.8%), cytokine release syndrome (CRS; 54.5%), anemia (45.5%), and nausea (40.9%). The most frequent treatment-related adverse events mirrored these frequencies. CRS cases were limited to grade 1/2, with 40.9% experiencing grade 1 and 13.6% grade 2. CRS management included tocilizumab in 27.3% of patients and steroids in 18.2%. Median CRS duration was 3.8 months, with a median onset of 9 hours, typically during the initial step-up at 0.7 mg; subsequent CRS rates declined. No cases of immune effector cell–associated neurotoxicity syndrome (ICANS) or tumor lysis syndrome occurred.
Infection risk was high, with 81.8% experiencing at least one infection, and 31.8% being grade 3 infections (9.1% grade 4). Opportunistic infections occurred in 50% of patients, though only one case reached grade 3 or higher. Most common events included CMV infection or reinfection (27%), and COVID-19 and oral candidiasis (18% each).
Regulatory history and ongoing research
In August 2025, the FDA issued a complete response letter regarding odronextamab for relapsed/refractory follicular lymphoma after two or more systemic therapies, citing inspection-related concerns at a manufacturing site. Earlier CRLs in March 2024 addressed DLBCL and follicular lymphoma, based on phase 2 data. Odronextamab is studied across multiple settings, including phase 3 OLYMPIA-2 (follicular lymphoma) and OLYMPIA-5 (odronextamab plus lenalidomide for follicular lymphoma).
References and further reading
- Michot J-M, Yagci M, Kargus K, et al. Odronextamab plus chemotherapy in patients with previously untreated DLBCL: first results from part 1 of the phase 3 Olympia-3 study. Blood. 2025;146(Suppl 1):abstract 65. doi:10.1182/blood-2025-65
- Regeneron financial results and regulatory updates (2025–2024).
Newsletter
Stay up to date on the most recent and practice-changing oncology data.
