Glofitamab and Epcoritamab in the Real World: A UK Multicentre Retrospective Analysis of Efficacy, Tolerability and Practical Implications

Background

Glofitamab (Glofit) and Epcoritamab (Epco), CD3:CD20 bispecific monoclonal antibodies (BsABs), have demonstrated remarkable monotherapy activity in relapsed/refractory large B-cell lymphoma with complete response rates (CRR) of 39% in their registration trials and durable remissions in many responding patients. Data from these trials indicate similar outcomes independent of prior CAR-T exposure. These data led to the licencing of both drugs internationally in 2023; in the UK access was granted in 2023 initially via compassionate access prior to full license and reimbursement. Real world data is limited and urgently needed to better understand efficacy and tolerability. To our knowledge, this is the largest real world data set of both drugs.

Aims

To provide insights into efficacy, tolerability, and practical implications of delivery of BsABs in the real world setting.

Methods

A multicentre, retrospective analysis of anonymised data from patients treated on UK compassionate/early access schemes or licensed approval.

Results

142 patients (pts.) were approved for Glofit at 26 UK centres: median age 68 years (range 25-90), 63% male and 99% performance status 0-2. Histological diagnoses included DLBCL (NOS) (61%), transformed follicular lymphoma (18%) and other NHL (20%). IPI score ≥3 in 54% (n=74), bulky disease in 23% (n=33) and a median of 3 lines of prior treatment with 15% (n=22) having received ≥4 lines. 60% (n=80) had prior CAR-T and 84% (n=117) were refractory to their most recent treatment. The majority (74%, n=86) wouldn’t have met registration trial eligibility, reasons included cytopenias (n=63), performance status (n=24), co-morbidity (n=7) and CNS involvement, prior or current (n=4).

6 pts didn’t receive Glofit due to PD prior to starting. At data cut-off, treatment was ongoing in 30 pts. A median of 3 (1-12) cycles were delivered with treatment delays in 36% (n=36). 33% (n=44) experienced cytokine release syndrome (CRS) (maximum grade 3 (4%, n=5), grade 2 (12%, n=16)) with 23 pts (17%) receiving tocilizumab. Immune effector cell-associated neurotoxicity syndrome (ICANS) was seen in 4% (n=5) (maximum grade 3 (1%, n=1), grade 2 (1%, n=1)). 22% (n=29) experienced grade ≥3 neutropenia and 45% (n=61) experienced infection during treatment. 7 pts (5%) required intensive care unit (ICU) admission, with no treatment-related deaths.

With a median follow-up of 6.1 months (IQR 2.5-9.1), complete metabolic response (CMR) was achieved in 27% (95% CI 19-36) with an overall response rate (ORR) 61% (95% CI 37-56). 6-month duration of CR (DoCR) was 86% (95% CI 53-96) with only 2/30 pts who achieved CR relapsing to date with short follow-up. Intention to treat (ITT) 6 month overall survival was 53% (95% CI 42-62) and 6 month progression free survival 44% (95% CI 34-53). Pts with prior Bendamustine (Benda) exposure (n=84) had inferior responses compared to Benda naïve pts (n=44) CRR 20% (95% CI 11-31) vs. 36% (95% CI 22-52), p=0.05, but this was restricted to those who had received Benda <6months prior to Glofit, CRR 9% (95% CI 2-21) vs. CRR 38% (95% CI 19-59), p=0.01.

Pts previously treated with CAR-T had a CRR 24% (95% CI 14-35), while CAR-T naïve patients had CRR 32% (95% CI 18-47), p=0.4. CRRs were higher in pts receiving Glofit >3 months post CAR-T, CRR 28% (95% CI 16-42) vs. 8% (95% CI 0.2-36), p=0.16.

38 pts were approved for Epco across 16 UK centres with a higher proportion of pts with ECOG ≥2 than the Glofit cohort. 1 pt progressed prior to starting treatment. 33% (n=9) experienced CRS (maximum grade 2, 10% (n=3)) and 2 pts (6%) received tocilizumab. 10% (n=3) experienced ICANS (maximum grade 3 (3% (n=1)), grade 2 (3% (n=1)) with 1 pt requiring ICU admission and no treatment-related deaths. At a median follow-up of 6.5 months (IQR 1.5-8.5), CMR was seen in 26% (95% CI 10-48) and 6-month DoCR was 100% (95% CI N/A).

Summary

In an ITT, real-world analysis of heavily pre-treated pts, BsABs achieved a CMR of 27%. CRS and ICANS occurred at relatively low rates, were mainly low-grade and ICU admission was rare. Initial data suggests inferior outcomes with Glofit in pts with prior Benda exposure, especially if within 6 months. CAR-T within 3 months was also associated with inferior outcome. These data indicate that encouraging outcomes with low toxicity can be achieved in the real world setting. Further analyses are needed to assess in which pts BsABs are an appropriate alternative to CAR-T and how best to sequence these treatments.