Paper Details

ABSTRACT

INTRODUCTION:

Despite recent therapeutic advancements, patients (pts) with R/R DLBCL ineligible for/unable to receive autologous stem-cell transplantation (ASCT) or chimeric antigen receptor T-cell therapy (CAR T) due to comorbidities or logistics have limited treatment options. Epcoritamab (E), a subcutaneous (SC) CD3×CD20 bispecific antibody, is approved as monotherapy in R/R DLBCL after ≥2 lines of systemic therapy (LOT). In the phase 1b/2 EPCORE NHL-5 trial (NCT05283720), E plus lenalidomide (Len), an oral immunomodulatory drug, showed encouraging clinical activity with manageable safety. Here, we report the primary progression free survival (PFS) analysis of the randomized phase 3 study EPCORE DLBCL-4 (NCT06508658) evaluating the efficacy and safety of E-Len vs R-GemOx in pts with R/R DLBCL after ≥1 LOT.

METHODS:

Adults with confirmed R/R CD20-positive DLBCL who had failed or were ineligible for ASCT and ineligible for or unable to receive CAR T were randomized 1:1 to E-Len or standard R-GemOx stratified by region, primary refractory status, and number of prior (p)LOT. E-Len consisted of up to twelve 28-day cycles of SC E (two C1 step-up doses, then 48-mg full doses [C1–3, QW; C4–12, Q4W]) and Len (25 mg/day po D1–D21), and R-GemOx was administered for up to 4 Q2W 28-day cycles. Hospitalization was not required. The primary endpoint was PFS as assessed by independent review committee (IRC) per Lugano criteria; secondary endpoints included complete response rate (CRR) by IRC, overall survival (OS), and safety. This primary analysis was conducted at 100% PFS events.

RESULTS:

At the data cutoff date of 31Mar2026, median follow-up was 10.1 mo (range, 0–17.5). A total of 339 pts were randomized to E-Len (n=173) or R-GemOx (n=166) with balanced baseline characteristics. Of all pts, 62.2% had one pLOT and 59.3% had primary refractory disease. Median number of cycles was 6 (range, 1–12) with E, 5 (range, 1–12) with Len, and 3 (range, 1–4) with R-GemOx. E-Len delivered a statistically significant and clinically meaningful PFS improvement vs R-GemOx with an HR of 0.40 (95% CI 0.30–0.55; P<0.0001). Median PFS was 16.5 mo (95% CI 5.7–NE) for E-Len vs 3.4 mo (95% CI 2.1–3.7) for R-GemOx. The PFS benefit was consistent across prespecified subgroups, including ≥65 yrs, one pLOT, and primary refractory disease. CRR was significantly higher with E-Len vs R-GemOx (51.4% vs 28.3%; P<0.0001). Overall response rate was 69.4% vs 46.4% for E-Len and R-GemOx, respectively. At data cutoff, when 49% (99/201) of OS events had occurred, OS data remained immature (HR 0.998, P=0.495). 

In safety-evaluable pts (E-Len, n=173; R-GemOx, n=160), grade (G) 3–4 treatment-emergent adverse events (TEAEs) were observed in 89.0% of pts with E-Len vs 74.4% of pts with R-GemOx. The most frequent G3–4 TEAEs for E-Len vs R-GemOx, respectively, were neutropenia (66.5% vs 41.3%), infections (28.9% vs 9.4%), thrombocytopenia (20.2% vs 36.9%), and anemia (15.6% vs 18.1%). G5 TEAEs, excluding progressive disease (PD) reported as an AE, occurred in 6.9% of pts for E-Len vs 2.5% of pts for R-GemOx. TEAEs leading to treatment discontinuation occurred in 13.9% of pts with E-Len (E, 6.9%; Len, 13.9%) vs 7.5% of pts with R-GemOx. Exposure-adjusted AE rates per 100 pt years (E/100PY) account for different treatment duration with E-Len vs R-GemOx: TEAE (3230 vs 3163), G3–4 TEAE (804 vs 939), G5 TEAE excluding PD reported as an AE (12 vs 8), neutropenia (445 vs 421), infections (250 vs 101), thrombocytopenia (146 vs 406), and anemia (168 vs 232). CRS events occurred in 54.3% of pts with E-Len (G3–4, 4.6%) with median onset on C1D16 and median resolution of 2 days. Tocilizumab was used in 22.0% of pts. ICANS was reported in 1.7% (G1–2 only); all resolved.

CONCLUSION: