Introduction: Epcoritamab monotherapy resulted in deep, durable responses with a manageable safety profile in patients with challenging-to-treat relapsed or refractory (R/R) large B-cell lymphoma (LBCL) in the pivotal phase 2 EPCORE® NHL-1 trial (NCT03625037). Here, we report 3-y follow-up results, including long-term efficacy and safety.Methods: Adults with R/R CD20+ LBCL (including diffuse LBCL [DLBCL], high-grade B-cell lymphoma, primary mediastinal large B-cell lymphoma, and follicular lymphoma [FL] grade [G] 3B) and ≥2 prior lines of systemic therapy were enrolled in the expansion phase of the study and treated with subcutaneous epcoritamab (0.16 mg and 0.8 mg step-up doses followed by 48-mg full doses) in 28-d cycles according to the approved dosing schedule: QW, cycles 1-3; Q2W, cycles 4-9; Q4W, cycles ≥10, until progressive disease or unacceptable toxicity. Corticosteroid prophylaxis was given during cycle 1. The primary endpoint was overall response rate (ORR) per Lugano criteria. Efficacy results presented here are based on investigators' assessment. Minimal residual disease (MRD) was assessed based on the clonoSEQ® (Adaptive Biotechnologies) ctDNA assay.Results: A total of 157 patients were treated with epcoritamab monotherapy. Median age was 64 y, 60% of patients were male, median number of prior lines of therapy was 3 (range, 2-11), and 61% had primary refractory disease; additional baseline characteristics have been presented previously. At data cutoff (May 3, 2024), 19 patients (12%) remained on treatment. Primary reasons for treatment discontinuation included progressive disease in 91 patients (58%) and adverse events (AEs) in 26 patients (17%); few patients who remained on epcoritamab ≥1 y progressed: 5/47 patients from >12 to ≤24 mo and 2/29 patients beyond 24 mo. The ORR and complete response (CR) rate per investigators were 59% and 41%; median follow-up was 37.1 mo (range, 0.3+ to 45.5). Median duration of response was 20.8 mo (95% CI, 13.0-32.0). Median duration of CR was 36.1 mo (95% CI, 20.2 to not reached [NR]). Median progression-free survival (PFS) was 4.2 mo (95% CI, 2.8-5.5); among complete responders, it was 37.3 mo (95% CI, 26.0-NR). Median overall survival (OS) was 18.5 mo (95% CI, 11.7-27.7); among complete responders, it was NR (95% CI, 36.4-NR). At 36 mo, an estimated 63% of complete responders remained alive. For patients with FL G3B (n=5) and patients with DLBCL transformed from FL (n=32), ORR/CR rates were 60%/60% and 50%/44%, respectively. Of the 119 overall patients who were MRD evaluable, 54 (45%) were MRD negative. In a cycle 3 day 1 landmark analysis, 36-mo rates of PFS/OS were 52%/55% among MRD-negative patients, and 18%/30% among MRD-positive patients. In an exploratory cycle 13 day 1 evaluation, 98% (40/41) of MRD-evaluable patients were MRD negative. There were 15 patients with CR who had a temporary epcoritamab hold for >6 wk (primarily due to AEs; median, 70.5 d, range, 48-257 d); all maintained their response at their next imaging assessment following treatment reinitiation. Of these patients, 6 had post-reinitiation MRD samples available, all of which were MRD negative. The most common treatment-emergent AEs (TEAEs) remained CRS (51%; 32% G1, 16% G2, 3% G3), fatigue (25%), and pyrexia (25%); CRS rates remained unchanged since prior reports. Fatal TEAEs were reported in 20 patients; 10 patients had G5 COVID-19 (including COVID-19 pneumonia). Rates of G≥3 infections were stable over time with rates within 3%-17% for 12-wk study intervals up to wk 144. Incidence of G≥3 cytopenias was highest (27%) during the first 8 wks of treatment and rates were within 0%-13% in subsequent 12-wk time periods up to wk 144. Immunoglobulin G levels decreased by a median of ~20% after the start of epcoritamab treatment (baseline median, 540.0 mg/dL) and remained stable over time.Conclusions: With a 3-year follow-up, epcoritamab continued to show deep and durable responses, with more than half of complete responders remaining in remission at 3 years: median duration of CR, 36.1 mo. Sustained MRD negativity was observed. Long-term safety was consistent with prior reports. These results with extended follow-up support the value of epcoritamab monotherapy as an effective and potentially curative therapy for patients with challenging-to-treat R/R LBCL.