BACKGROUND Elranatamab is a humanized, bispecific antibody that targets B-cell maturation antigen (BCMA) on myeloma cells and CD3 on T cells, with the aim of inducing T-cell-mediated cytolysis of the myeloma cells. In the MagnetisMM-3 (NCT04649359) trial, an open-label, multicenter, non-randomized, phase 2 registrational study of elranatamab monotherapy, patients with RRMM who had not received prior BCMA-directed therapy (ie, BCMA-naïve patients; n=123) achieved objective responses with an overall response rate (ORR) of 61%. Here, we report the long-term efficacy and safety of elranatamab. METHODS Eligible patients had previously received at least 1 proteasome inhibitor, 1 immunomodulatory drug, and 1 anti-CD38 antibody. Patients received step-up doses of 12 and 32 mg elranatamab subcutaneously on days 1 and 4 of cycle 1, respectively, followed by 76 mg elranatamab once-weekly (QW), starting on day 8 of the first 4-week cycle. Patients who received ≥6 months of QW dosing and achieved ≥ partial response lasting at least 2 months were transitioned to a once every 2 weeks (Q2W) dosing schedule and from Q2W to once every 4 weeks after at least 6 Q2W cycles. Treatment with elranatamab continued until disease progression, unacceptable toxicity, or withdrawal of consent. The primary endpoint was ORR, assessed by blinded-central review per International Myeloma Working Group criteria. Minimal residual disease (MRD) status was assessed using next-generation sequencing, with MRD negativity defined as <1 myeloma cell in 10 5 nucleated cells. Adverse events were graded using the National Cancer Institute Common Terminology Criteria for Adverse Events (version 5.0). Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) severity were graded by American Society for Transplantation and Cellular Therapy criteria. RESULTS After a median follow-up of 15.9 months (data cutoff, April 16, 2023), 32.5% of patients remained on treatment. The confirmed ORR was 61.0% (n=123; 95% CI, 51.8-69.6), with 35.8% of patients achieving complete responses (CR) or better. Among evaluable patients (those with ≥CR and evaluable for MRD, n=29), 89.7% achieved MRD negativity. Median duration of response (DOR), progression-free survival (PFS), and overall survival (OS) have not been reached, and the probabilities of maintaining a response, being progression-free, and being alive at 15 months were 70.8% (95% CI, 58.2-80.2), 50.2% (95% CI, 40.2-59.3), and 56.3% (95% CI, 47.0-64.6), respectively. All participants (100%) had ≥1 any grade treatment-emergent adverse events (TEAEs), and 70.7% of patients had grade 3/4 TEAEs. The most common any grade (≥25%) and grade 3/4 (≥10%) adverse events (any grade, grade 3/4) were infections (69.9%, 40.7%), CRS (57.7%, 0%), anemia (48.8%, 37.4%), neutropenia (48.8%, 48.8%), diarrhea (41.5%, 1.6%) thrombocytopenia (31.7%, 23.6%), lymphopenia (26.8%, 25.2%), fatigue (36.6%, 4.1%), nausea (26.8%, 0%), injection site reaction (26.8%, 0%), hypokalemia (26.0%, 10.6%), cough (26.0%, 0%) pyrexia (30.9%, 4.1%), decreased appetite (33.3%, 0.8%), leukopenia (15.4%, 13.0%) and COVID-19 pneumonia (14.6%, 11.4%). CONCLUSIONS Extended follow-up from the ongoing phase 2 MagnetisMM-3 trial of elranatamab in heavily pretreated participants with RRMM demonstrated sustained clinical efficacy, with median DOR, PFS, and OS not reached, and no new safety signals.