GMAB: EPCORE DLBCL-2 and first-line epcoritamab evidence

October 5, 2026 update: Phase 3 PFS results

The formerly anticipated EPCORE DLBCL-2 readout has occurred. Genmab and AbbVie reported positive randomized Phase 3 progression-free survival results on October 5, 2026. The existing sections below preserve the earlier development expectations and Phase 1/2 evidence as originally recorded; this dated update supersedes the expectation of an unreleased result. The milestone is a clinical data disclosure, not approval of the epcoritamab–R-CHOP combination. The sponsors intend to engage regulators, and no accepted application or new action date was announced. [7]

NCT05578976 is a global, open-label, randomized 2:1 comparison of subcutaneous epcoritamab plus R-CHOP with R-CHOP-based control in previously untreated large B-cell lymphoma with IPI 2–5. The experimental schedule contains six combination cycles followed by two epcoritamab cycles; the comparator contains six R-CHOP cycles followed by two rituximab cycles. The population includes DLBCL not otherwise specified and specified other aggressive large B-cell histologies. The planned enrollment in the preserved original dossier must not be relabeled as the actual analyzed sample; the topline release does not provide actual randomized or treated arm counts.

What the randomized result changes

A hazard ratio of 0.49 represents an estimated 51% reduction in the instantaneous hazard of progression or death under the analysis model. It is not a 51-percentage-point improvement in the number of patients remaining progression-free and cannot be converted into an NNT without absolute risks at a common time. The matching point estimates in the primary and broader populations are not two independent replications: patients with IPI 3–5 also contribute to the IPI 2–5 analysis. The release labels one endpoint primary and the other key secondary but does not reproduce the complete multiplicity hierarchy, censoring rules or sensitivity analyses. [7]

This randomized comparison strengthens the evidence beyond the small uncontrolled EPCORE NHL-2 cohort preserved in the earlier report. It still leaves the absolute PFS gain, response depth, follow-up maturity, overall survival and patient-reported burden unresolved in the public topline. The exact cutoff, progression/death event counts, median follow-up and arm-specific missing assessments were not supplied. Those fields remain explicitly unavailable, rather than reconstructed from the hazard ratio or borrowed from NHL-2.

Safety disclosure and regulatory limits

The sponsors described the observed safety profile as consistent with known component profiles. No arm-level TEAE, SAE, death, discontinuation, CRS, ICANS or infection table accompanied the release. “Consistent” is a qualitative sponsor characterization, not evidence that these outcomes were absent or balanced. The earlier NHL-2 safety counts remain tied to that earlier study and must not be presented as Phase 3 denominators. A full report will be needed to judge whether disease control is accompanied by a manageable increase in infectious, immune-mediated and chemotherapy-related toxicity. [7]

Genmab and AbbVie are partners in the same epcoritamab program, so this release is counted once under the stable Genmab event ID. It is separate from Genmab’s RAINFOL-01 ADC data. The new result changes the original clinical expectation from pending to reported; it does not erase the original evidence, turn first-line combination treatment into an approved regimen or supply an independent estimate of program funding.

CD20-directed T-cell recruitment in curable disease

Epcoritamab binds CD20 on B cells and CD3 on T cells, bringing cytotoxic cells into contact with lymphoma. R-CHOP already combines antibody-mediated and cytotoxic effects. Adding a T-cell engager in newly diagnosed DLBCL therefore tests whether a deeper immune response can prevent relapse beyond an established curative regimen. [1] [3]

Clinical interpretation: first-line therapy has a different benefit–harm threshold from late-line salvage. Preventing an otherwise likely relapse is valuable, but adding serious infection or treatment-related death to patients already curable with chemotherapy can offset benefit. The trial needs time-to-event and safety evidence, not simply a high response rate at the end of treatment.

Step-up dosing and chemotherapy intensity

The earlier arm1 program used subcutaneous epcoritamab48 mg with six21-day R-CHOP cycles, then epcoritamab maintenance for up to one year. The Phase3 plan uses six combination cycles followed by two epcoritamab cycles, compared with six R-CHOP cycles followed by two rituximab cycles. The Phase2 and Phase3 exposure schedules are not identical. [1] [3]

Analytical judgment: step-up dosing and corticosteroid prophylaxis address early cytokine release, while the long treatment course raises different infection and immune-suppression questions. Maintaining anthracycline and cyclophosphamide intensity matters: if added toxicity reduces delivery of the curative backbone, deeper immune engagement may not translate into improved disease control.

Phase1b/2 signal with transparent denominators

The2024 arm1 update included47 treated high-risk patients, with46 response-evaluable and median follow-up27.4 months. All evaluable patients responded and87% achieved CR. These were promising single-arm results, not randomized proof that epcoritamab caused the increment above R-CHOP. The small double/triple-hit subgroup cannot establish subtype-specific superiority. [1] [5]

The later2025 abstract provides a longer follow-up context, but the numerical table here deliberately retains one consistent2024 cutoff. Best response, MRD negativity and survival estimates use different eligible populations. In particular, the30/33 MRD result is not91% of all47 treated patients; specimen availability may correlate with response and follow-up.

The primary population is IPI3–5

EPCORE DLBCL-2 plans approximately900 patients randomized2:1 to epcoritamab–R-CHOP versus R-CHOP-based control. It is open-label with independent response assessment. The registered primary endpoint is progression-free survival in patients with IPI3–5; PFS in the broader population, EFS, CR, OS and MRD are secondary outcomes. [2] [3] [4]

This matters when reading a headline: success in the predefined high-risk population should not be automatically described as identical benefit across all newly diagnosed DLBCL. IPI is a clinical prognostic index rather than a molecular subtype. Cell-of-origin and rearrangement status provide additional, differently defined subgroup questions and may have limited statistical precision.

Relapse endpoints and multiplicity

PFS counts progression or death; EFS also incorporates specified treatment-failure events such as new nonprotocol lymphoma therapy and residual disease decisions. They are not synonyms. The testing hierarchy must determine whether broader-population PFS, EFS or OS claims remain confirmatory after the primary analysis. A nominally favorable secondary result alone cannot redefine the study objective. [2] [3]

Analytical judgment: early PET responses and ctDNA clearance are supportive but may not track long-term cure perfectly. A convincing result needs absolute progression/death rates at clinically relevant landmarks, the hazard ratio and uncertainty, plus treatment completion. Subsequent CAR-T, transplant or other salvage can influence OS and should be documented rather than used to obscure the randomized PFS result.

Complete response after chemotherapy is not synonymous with cure. Residual lymphoma can fall below PET or circulating-DNA detection and later re-emerge, while inflammatory uptake can produce false-positive imaging. Assay sensitivity, sampling time and confirmation methods should accompany MRD and PET results. All planned treatment components belong to the randomized regimen comparison, including additional rituximab in control and additional epcoritamab in the experimental arm. Interpretation should examine whether fewer lymphoma progressions are offset by non-lymphoma deaths, especially during early infection risk. This matters particularly in high-IPI patients, who have both higher relapse risk and greater vulnerability to toxicity. Long-term event-free survival and function add context beyond an early remission percentage. Treatment effect estimates in small molecular subgroups should be presented with uncertainty rather than used to declare that the regimen overcomes a specific high-risk genotype.

CRS, infection and treatment completion

At the2024 arm1 cutoff, CRS occurred in60%, mostly low grade; four patients discontinued epcoritamab due to adverse events. Two grade5 events were reported, COVID-19 and septic shock. ICANS occurred in two patients and resolved. These counts show why cytokine-release management is only one component of tolerability; infection remains clinically consequential. [1] [4] [6]

The sponsors guide a2026 Phase3 readout. Their established antibody-development infrastructure supports trial execution, but no undisclosed outcome is inferred from that schedule. Assessment should include serious infections by exposure period, immunoglobulin replacement, antimicrobial prophylaxis, neutropenia, chemotherapy dose intensity and patient burden. A shorter pivotal epcoritamab course may change safety relative to arm1 and requires its own data.

Sources / References

1. Falchi et al. EPCORE NHL2 arm1 original ASH2024 abstract
2. EPCORE DLBCL2 registry NCT05578976
3. EPCORE DLBCL2 original Phase3 design, ASCO2023
4. Genmab May2026 pipeline presentation
5. EPCORE NHL2 arm1 three-year results, ASH2025
6. Genmab2025 R&D update
7. Genmab and AbbVie · EPCORE DLBCL-2 results · October 5, 2026