3.1 Patients

A total of 187 patients (both cohorts A and B) were enrolled in the MagnetisMM-3 trial. The median (range) of baseline body weight in all 187 patients was 71.8 kg (36.5–159.6 kg), with 25th and 75th percentiles of 61.8 kg and 84.4 kg, respectively. The median (range) of baseline body weight in the 183 patients included in the safety analysis was 72 kg (36.5–159.6 kg), with 25th and 75th percentiles of 62 kg and 84.5 kg, respectively. In cohort A patients included in the efficacy analysis (n = 123), the median (range) baseline body weight was 72 kg (40.9–159.6 kg), with 25th and 75th percentiles of 62 kg and 82.1 kg, respectively.

The baseline characteristics and prior treatment summarized by baseline body weight quartile are presented in Table 1. In general, the baseline characteristics and prior treatment were balanced across body weight quartiles. However, the two lower quartiles had a higher proportion of patients who were female (59.6% and 68.8% for Q1 and Q2, respectively, versus 39.5% and 22.2% for Q3 and Q4, respectively). Q1 had a higher proportion of patients of Asian race (19.1% versus 2.2–9.3% for the other quartiles). In addition, Q1 included a higher proportion of patients with creatinine clearance ≤ 60 mL/min (55.3% versus 13.3–39.6% for the other quartiles). Finally, Q1 had a higher proportion of patients with ≥ 1 poor prognostic feature (76.6% versus 62.5–68.9% for the other quartiles).

Table 1 Baseline characteristics and prior treatment by baseline body weight quartiles3.2 Elranatamab Concentrations by Baseline Body Weight Quartile

The PK analyses included data from 187 patients (both cohorts A and B). The unbound (i.e., free) predose elranatamab concentrations at the start of each cycle are presented in Fig. 1. Predose elranatamab concentrations at cycle 2 significantly correlated with average concentration during cycle 1 (Cave) (Supplementary Fig. 1). Overall, elranatamab predose concentrations were overlapping across all quartiles. Despite the overlapping distributions, the median and geometric mean for Q4 was lower than other quartiles. However, the lower average predose concentrations in Q4 did not have an impact on efficacy. The distributions of maximum elranatamab concentration within 24 hours of the first dose were similar across body weight quartiles (Supplementary Fig. 2).

Fig. 1

Elranatamab predose concentrations by baseline body weight quartile. Boxplot showing the median and 25%/75% quartiles with whiskers to the last point within 1.5 times interquartile range. C, cycle; Ctrough, trough concentration; D, day

3.3 Safety of Elranatamab by Baseline Body Weight Quartile

The safety analysis included data from 183 patients (both cohorts A and B) who received the two-step-up-dose priming regimen (12 mg on day 1 and 32 mg on day 4). An overview of TEAEs by body weight quartiles in the MagnetisMM-3 trial is presented in Table 2. There were no clinically relevant differences in the safety of elranatamab when patients were separated into body weight quartiles, including overall incidence of AEs, Grade 3/4 AEs, serious AEs, and discontinuations owing to AEs. The overall incidence of AEs was identical across quartiles (100%). The incidence of Grade 3 or 4 AEs was 63.8%, 68.8%, 60.5%, and 84.4% for Q1–Q4, respectively. The incidence of serious AEs was 72.3%, 77.1%, 81.4%, and 71.1% for Q1–Q4, respectively. Discontinuations owing to AEs occurred in 27.7%, 16.7%, 37.2%, 28.9% of patients for Q1–Q4, respectively. The incidence of AEs by System Organ Class and the most common any-grade, Grade 3 or 4, and serious AEs across body weight quartiles were also consistent with those reported in the overall population.

Table 2 Overview of treatment-emergent adverse events by body weight quartiles

The most common hematologic TEAEs for elranatamab were previously reported to be anemia, neutropenia, thrombocytopenia, and lymphopenia [16]. The most common TEAEs reported by preferred term (> 20% of patients) are shown in Table 2. For Q1–Q4, the incidences of Grade 3/4 treatment-emergent anemia (40.4%, 47.9%, 48.8%, and 33.3%, respectively), neutropenia (46.8%, 31.3%, 48.8%, and 51.1%, respectively), thrombocytopenia (27.7%, 22.9%, 34.9%, and 20.0%, respectively), and lymphopenia (21.3%, 33.3%, 39.5%, and 17.8%, respectively) were all similar across quartiles. There was no trend for incidence being higher in the lower body weight quartiles.

Infections and infestations were reported in 70.2%, 68.8%, 74.4%, and 73.3% for Q1–Q4, respectively. Grade 3 or 4 infections and infestations were reported in 31.9%, 50.0%, 51.2%, and 51.1% for Q1–Q4, respectively. CRS was reported in 57.4%, 62.5%, 51.2%, and 60.0% for Q1–Q4, respectively, with only one case of Grade 3 CRS in Q2 (2.1%). ICANS was reported in 6.4% (n = 3), 4.2% (n = 2), and 2.3% (n = 1), and 0% in Q1–Q4, respectively. ICANS for one patient in Q1 (2.1%) and Q3 (2.3%) was reported as Grade 3.

3.4 Efficacy of Elranatamab in Cohort A of MagnetisMM-3 Trial by Baseline Body Weight

The median (range) of follow-up since the initial dose was 17.58 months (0.23–36.83 months) for cohort A. A clinically meaningful ORR benefit was observed across the different body weight quartiles consistent with the primary efficacy analysis which showed confirmed ORR by BICR 61.0% (95% CI 51.8–69.6). The confirmed ORR by BICR for Q1, Q2, Q3, and Q4 was 53.1% (95% CI 34.7–70.9), 59.4% (95% CI 40.6–76.3), 62.1% (95% CI 42.3–79.3), and 70.0% (95% CI 50.6–85.3), respectively (Fig. 2).

Fig. 2

Objective response rate by BICR by body weight quartile (cohort A participants). ORR, objective response rate

The confirmed CR or better rate by BICR was 40.6%, 25.1%, 27.5%, and 56.7% for Q1–Q4, respectively. The confirmed very good partial response (VGPR) or better rate was 51.0%, 59.4%, 48.3%, and 66.7% for Q1–Q4, respectively (Fig. 3). The median PFS for Q1–Q4 was 13.9 months (95% CI 2.2 months–not estimable [NE]), 17.6 months (95% CI 3.7 months–NE), 9.8 months (95% CI 2.6–28 months), and not yet reached (95% CI 10.4 months–NE), respectively. The PFS curves for these quartiles were generally overlapping with no specific trend across body weight quartiles (Fig. 4).

Fig. 3

Responses by BICR to elranatamab in patients with relapsed or refractory multiple myeloma by body weight quartile. BICR, blinded independent central review; BOR, best overall response; CR, complete response; PR, partial response; sCR, stringent complete response; VGPR, very good partial response. Stacked bar graph illustrating the rate of sCR, CR, VGPR, and PR in 123 patients who were treated with elranatamab. Responses were assessed by BICR

Fig. 4

Progression-free survival by BICR—Kaplan–Meier plot in cohort A by body weight quartile. BICR, blinded independent central review; NE, not estimable; Q, quartile

The median DOR for Q1–Q4 was not yet reached (95% CI 25 months–NE), not yet reached (95% CI 12.5 months–NE), 12 months (95% CI 3.7 months–NE), and not yet reached (95% CI 24 months–NE), respectively. The DOR curves for these quartiles were generally overlapping with no specific trend across body weight quartiles (Fig. 5).

Fig. 5

Duration of response by BICR—Kaplan–Meier plot in cohort A by body weight quartile. BICR, blinded independent central review; NE, not estimable; Q, quartile