Bispecific antibodies (BsAbs) show encouraging efficacy for relapsed/refractory large B-cell lymphoma (r/r LBCL), but careful patient selection is essential for achieving the best outcomes, the largest real-world study of the compounds has shown.

UK data, presented at the European Haematology Association (EHA) congress, showed that response rates were generally higher than seen in pivotal trials and were highly durable, with 87% of complete responses lasting to one year.

However, patients who did not complete a second BsAb cycle had very poor survival and those who had been exposed to bendamustine in the six months prior to treatment, or had undergone CAR-T within the previous three months, also showed significantly worse outcomes.

“These data may facilitate improved patient selection and inform patients and caregivers of potential outcomes and allow optimisation of advance care planning in RR LBCL,” said study author Dr William Townsend, a consultant haematologist at University College London Hospitals, when presenting the data at EHA in Stockholm.

The study analysed data from r/r LBCL patients at 34 UK centres who were treated with glofitamab (n=212) or epcoritamab (n=112) between November 2023 and May 2025.

Patients were heavily pre-treated, with 50% of the overall cohort having undergone prior CAR T, and a greater proportion of the glofitamab group being CAR T-treated (55% vs 37% for epcoritamab, p=0.002).

Patients were also high-risk, with 55% having primary refractory disease and 25% with ECOG ≥2, with the latter again more common in patients who received epcortiamab (40% vs 17% p<0.001).

Regarding safety, 28% of patients experienced cytokine release syndrome (CRS), most of which (91%) was grade 1/2, which was similar to rates seen in trials and “supports delivery in smaller hospitals, closer to people’s homes,” Dr Townsend suggested.

These baseline differences precluded comparisons of efficacy between the compounds, however, overall response rate (ORR) and complete response rate (CRR) across the overall cohort were 43% and 24%, respectively.

For trial-eligible patients, ORR and CRR were 60% and 43%, which Dr Townsend noted was higher than in the pivotal trials.

After a median follow-up of 10 months, median PFS and OS were 3.2 and 7.9 months, respectively, with 44% of patients still alive at one year.

PFS and OS were both superior for trial-eligible vs ineligible patients (5.8 vs 2.3 months and 21.4 vs 4.8 months, respectively).

A high proportion of patients did not start cycle two (16% glofitamab and 47% epcoritamab) and these patients had a very low one-year OS of 2%, highlighting the importance of patient selection, Dr Townsend said.

Factors associated with inferior PFS were:

• Prior line of treatment refractoriness (HR 2.89, p=0.007)
• Elevated LDH (HR 2.62, p=0.001)
• Bendamustine exposure within 6 months (HR 1.62, p=0.007)
• ECOG 1 (HR 2.70 vs 0, p=0.001)
• ECOG 2 (HR 6.49, p<0.001).

In addition, patients who received BsAbs within three months of CAR-T had inferior PFS and OS compared to those who were given CAR-T more than three months prior.

The only factor associated with better PFS was transformed follicular lymphoma (HR 0.40, p=0.001).

Extended phase 3 data confirms superiority of epcoritamab

Meanwhile, phase 3 data presented elsewhere at EHA also backed the superiority of epcoritamab over chemoimmunotherapy for r/r LBCL, showing a 26% reduction in the risk of progression or death.

The trial also found that nearly 60% of patients who achieved a complete response from epcoritamab were still in complete response three years later.

The study “is the largest randomised phase 3 trial in RR LBCL and the first to demonstrate a statistically significant improvement in PFS with a CD3xCD20 BsAb monotherapy,” said Dr Christopher Fox, a consultant haematologist at Nottingham University Hospitals NHS Trust, when presenting the findings.

The trial compared outcomes of RR LBCL patients who were treated with either epcoritamab (n=236) or standard investigator’s choice of chemoimmunotherapy (n=237).

Patients who received epcoritamab were older than those who received chemoimmunotherapy (median age 72 vs 71 years), with a higher proportion over the age of 65 years (77% vs 70%).

Around 75% of the overall cohort had received at least two prior lines of treatment and over half were primary refractory.

Epcoritamab significantly improved PFS compared to chemoimmunotherapy (3.5 vs 3.0 months, HR 0.74, p=0.0059), while the duration of response and complete response were also superior with epcoritamab, as was time to next treatment.

OS was confounded by COVID-19 and was similar for both treatments when unadjusted, but was superior with epcoritamab when adjusted for COVID-19 mortality and effective subsequent treatments.

The most common treatment-emergent adverse events with epcoritamab were CRS (53% of patients) and grade 3-4 infections (30% vs 12% with CIT).