As of July 2, 2020, 75 patients received AMG 701. Patients had a median age of 63 years, a median time since diagnosis of 5.9 years, and a median (range) of 6 (1-25) prior lines of therapy; 27% of patients had extramedullary disease, 83% prior SCT, and 93% prior anti-CD38 Ab; 68% were triple refractory to a PI, an IMiD, and an anti‑CD38 Ab. Median (Q1, Q3) treatment duration was 6.1 (3.1, 15.3) weeks and median follow-up on treatment was 1.7 (1.0, 3.7) months. Patients discontinued drug for PD (n=47), AEs (adverse events, n=4, 3 CRS, 1 CMV / PCP pneumonia), withdrew consent (4), other therapy (1), investigator discretion (1), and CNS disease (1); 17 patients remain on AMG 701.

The most common hematological AEs were anemia (43%), neutropenia (23%), and thrombocytopenia (20%). The most common non-hematological AEs were CRS (61%), diarrhea (31%), fatigue (25%), and fever (25%). CRS was mostly grade 1 (n=19) or 2 (n=21) per Lee Blood 2014 criteria. All grade 3 CRS (n=5, 7%) were assessed as dose-limiting toxicities (DLTs); all were reversible with corticosteroids and tocilizumab, with median duration of 2 days. CRS grade 3 drivers included transient LFT increases in 3 patients and hypoxia in 2 patients. Other DLTs were 1 case each of transient grade 3 atrial fibrillation, transient grade 3 acidosis, and grade 4 thrombocytopenia. Serious AEs (n=29, 39%) included infections (13), CRS (7), and asymptomatic pancreatic enzyme rise (2, no imaging changes, 1 treatment related). There were 4 deaths from AEs, none related to AMG 701 (2 cases of sepsis, 1 of retroperitoneal bleeding, and 1 of subdural hematoma). Reversible treatment-related neurotoxicity was seen in 6 patients, with median duration of 1 day, all grade 1-2, and associated with CRS in 4 patients.

The response rate was 36% (16/45) at doses of 3-12 mg; at ≤1.6 mg (n=27), there was 1 response at 0.8 mg in a patient with low baseline soluble BCMA (sBCMA). With earlier dose escalation with 9 mg, the response rate was 83% (5/6, 3 PRs, 2 VGPRs), with 4/5 responders being triple refractory and 1 DLT of grade 3 CRS in this group. Across the study, responses included 4 stringent CRs (3 MRD-negative, 1 not yet tested), 1 MRD-negative CR, 6 VGPRs, and 6 PRs (Table). Median (Q1, Q3) time to response was 1.0 (1.0, 1.9) month, time to best response was 2.8 (1.0, 3.7) months, and response duration was 3.8 (1.9, 7.4) months, with maximum duration of 23 months; responses were ongoing at last assessment in 14/17 patients (Figure). MRD was tested in 4 patients (3 sCR, 1 CR) and all were negative (3 by NGS, 1 by flow); MRD negativity was ongoing at last observations up to 20 months later. AMG 701 exhibited a favorable PK profile in its target patient population of RR MM, with AMG 701 exposures increasing in a dose‑related manner. Patient baseline sBCMA levels were identified as a determinant of AMG 701 free drug exposures; at higher doses, encouraging preliminary responses were seen even at the higher end of baseline sBCMA values.