Background: Elranatamab (PF-06863135) is a humanized bispecific antibody that targets both BCMA-expressing MM cells and CD3-expressing T cells. MagnetisMM-3 (NCT04649359) is an open-label, multicenter, non-randomized, phase 2 study to evaluate the safety and efficacy of elranatamab monotherapy in pts with R/R MM. Initial safety results are presented. Methods: MagnetisMM-3 enrolled pts who are refractory to at least 1 proteasome inhibitor, 1 immunomodulatory drug, and 1 anti-CD38 antibody. Pts were assigned to 1 of 2 independent, parallel cohorts: those naïve to BCMA-directed therapies (Cohort A) and those with previous exposure to BCMA-directed antibody-drug conjugates or CAR-T cells (Cohort B). Pts received subcutaneous elranatamab 76 mg QW on a 28-d cycle with a 2-step-up priming dose regimen administered during the first week. Dose modifications were permitted for toxicity. Treatment-emergent adverse events (TEAEs) were graded by CTCAE (v5.0), and cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) by ASTCT criteria. Results: As of the data cutoff on Dec 31, 2021, 60 pts in Cohort A had received ≥1 dose of elranatamab; the last pt’s first dose was ̃2 months prior to the cutoff. Median age was 69.0 y (range, 44−89), 48.3% were male, 63.3% were white, 18.3% were Asian and 11.7% were Black/African American. At baseline, 60.0% of pts had an ECOG performance status 1−2 and pts had received a median of 5 (range, 2−12) prior therapies. Median duration of elranatamab treatment was 9.57 wks (range, 0.1−46.1); median relative dose intensity was 87.4% (range, 23.1−101.4). TEAEs were reported in 100% (Grade [G] 3/4, 75.0%) of pts. Most common (≥30%) hematologic TEAEs were neutropenia (36.7% [G3/4, 35.0%]), anemia (36.7% [G3/4, 30.0%]) and thrombocytopenia (30.0% [G3/4, 21.7%]). Among pts who received the 2-step-up priming regimen (n = 56), CRS and ICANS, respectively, were reported in 58.9% (G3/4, 0%) and 3.6% (G3/4: 0%); of those pts, 57.6% (n = 19/33) and 100% (n = 2/2) received tocilizumab and/or steroids. Most common (≥30%) non-hematologic TEAE, other than CRS/ICANS, was fatigue (31.7% [G3/4, 3.3%]). Infections were reported in 46.7% (G3/4: 18.3%) of pts; most frequently reported were upper respiratory tract infections (11.7% [G3/4: 0%]). Discontinuations due to adverse events were reported in 5.0% of pts. No pts permanently discontinued treatment due to CRS or ICANS. There were 10 deaths; causes were MM progression (n = 8), septic shock (n = 1) and unknown (n = 1). Data will be updated at the time of presentation to include ̃90 pts. Conclusions: Preliminary results of MagnetisMM-3 in pts with R/R MM and no prior BCMA-targeted treatment suggest that 76 mg QW elranatamab with a 2-step-up priming regimen is well tolerated, with no G ≥3 CRS or ICANS observed. Clinical trial information: NCT04649359.