2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
The FDA accepted a BLA for 2 belantamab mafodotin–based combinations in relapsed/refractory multiple myeloma.
The FDA has accepted a biologics license application (BLA) seeking the approval of 2 belantamab mafodotin (Blenrep)–based combinations; one in combination with bortezomib (Velcade) and dexamethasone (BVd), and the other with pomalidomide (Pomalyst) and dexamethasone (BPd), for the treatment of patients with multiple myeloma who have received at least 1 prior line of therapy.1
The BLA is supported by data from the phase 3 DREAMM-7 (NCT04246047) and DREAMM-8 (NCT04484623) trials, which evaluated BVd and BPd, respectively. Both studies met their primary end points with the belantamab mafodotin–based combinations generating a statistically significant and clinically meaningful improvement in progression-free survival (PFS) vs standard-of-care triplet regimens in patients with relapsed/refractory multiple myeloma.
The FDA has set a target action date for the BLA of July 23, 2025, under the Prescription Drug User Fee Act.
“Relapsed/refractory multiple myeloma treatment could be transformed by additional, efficacious treatment options with manageable [adverse] effects [AEs] and community-based administration,” Hesham Abdullah, senior vice president, Global Head Oncology, R&D, at GSK, stated in a news release. “The evidence from DREAMM-7 and DREAMM-8 supporting our [belantamab mafodotin] combinations submission has been further strengthened by the statistically significant overall survival [OS] results from the DREAMM-7 trial. We look forward to working with the FDA on this review.”
Findings from the open-label, randomized DREAMM-7 study published in the New England Journal of Medicine demonstrated that at a median follow-up of 28.2 months (range, 0.1-40.0), patients treated with BVd (n = 243) achieved a median PFS of 36.6 months (95% CI, 28.4-not reached [NR]) compared with 13.4 months (95% CI, 11.1-17.5) for those treated with daratumumab (Darzalex) plus bortezomib and dexamethasone (DVd; n = 251; HR, 0.41; 95% CI, 0.31-0.53; P < .001).2
At the data cutoff for this analysis, OS data did not reach statistical significance. Patients in the experimental arm experienced an 18-month OS rate of 84% compared with 73% for those in the DVd arm.
Investigators enrolled patients with multiple myeloma who received at least 1 prior line of therapy and experienced disease progression during or after their most recent line of treatment. Prior treatment with a BCMA-directed therapy was not permitted, and patients could not have disease that was refractory to anti-CD38 therapy.
Patients were randomly assigned 1:1 to receive BVd or DVd. In both arms, patients received bortezomib subcutaneously at 1.3 mg/m2 on days 1, 4, 8, and 11 of each 21-day cycle and dexamethasone at 20 mg on the day of and the day after bortezomib for the first eight cycles. In the experimental arm, belantamab mafodotin was given at 2.5 mg/kg once every 3 weeks, and those in the control arm received daratumumab at 16 mg/kg once per week in cycles 1 to 3, every 3 weeks in cycles 4 to 8, and every 4 weeks thereafter. Treatment with belantamab mafodotin and daratumumab continued until disease progression, unacceptable toxicity, death, or patient withdrawal.
PFS served as the trial’s primary end point; key secondary end points included OS, duration of response (DOR), minimal residual disease (MRD)–negative status, and safety.
Regarding safety, treatment-related AEs (TRAEs) led to discontinuation of any trial drug in 26% of evaluable patients in the BVd group (n = 242) and 15% of patients in the DVd group (n = 246). Serious AEs led to death in 10% of patients in the BVd group vs 8% of patients in the DVd group; serious AEs leading to death that were deemed related to treatment occurred in 3% and 1% of patients, respectively.
Grade 3 or higher AEs were reported in 95% of the patients in the BVd group vs 78% in the DVd group; the respective rates of serious AEs were 50% and 37%. Ocular AEs occurred in 79% of patients in the experimental arm vs 29% in the DVd arm. These AEs were managed with dose modifications, and most instances of worsened visual acuity resolved.
Data from the randomized, open-label study presented at the 2024 ASCO Annual Meeting showed that at a median follow-up of 21.8 months (range, 0.03-39.23), patients treated with BPd (n = 155) experienced a median PFS that was NR (95% CI, 20.6-NR) compared with 12.7 months (95% CI, 9.1-18.5) for those given bortezomib plus pomalidomide and dexamethasone (PVd; n = 147; HR, 0.52; 95% CI, 0.37-0.73; P < .001).3
A trend favoring the BPd regimen was also observed (HR, 0.77; 95% CI, 0.53-1.14).
The study included adult patients with multiple myeloma who received at least 1 prior line of therapy that included lenalidomide (Revlimid). They were required to have documented disease progression during or after their most recent line of therapy. Patients could not have prior treatment with an anti-BCMA therapy or pomalidomide; disease that was refractory or intolerant to bortezomib was not allowed.
Patients were randomly assigned 1:1 to receive BPd or PVd. In the experimental arm, treatment comprised belantamab mafodotin once every 4 weeks at 2.5 mg/kg in cycle 1 and 1.9 mg/kg in subsequent cycles; pomalidomide at 4 mg per day on days 1 to 21 of each 28-day cycle; and dexamethasone at 40 mg on days 1, 8, 15, and 22 of each cycle. Those in the control arm were given 1.3 mg/m2 of bortezomib on days 1, 4, 8, and 11 during the first 8 21-day cycles, then on days 1 and 8 of subsequent cycles; pomalidomide at 4 mg per day on days 1 to 14 of each cycle; and dexamethasone at 20 mg the day of and the day after bortezomib.
PFS was the trial’s primary end point, and key secondary end points included OS, MRD negativity, and DOR.
Regarding safety, any-grade AEs occurred in more than 99% of patients in the BPd arm (n = 150) vs 96% of patients in the PVd arm (n = 145). The rates of grade 3/4 AEs were 91% and 73%, respectively. In the experimental arm, AEs led to treatment interruptions, dose reductions, and permanent discontinuation of any study treatment in 91%, 61%, and 15% of patients, respectively. These respective rates were 75%, 61%, and 12% in the control arm. Serious AEs were reported in 63% of patients in the BPd arm and 45% of patients in the PVd arm. Serious AEs led to death in 11% of patients in both arms.
Ocular AEs of any grade occurred in 89% of patients in the BPd arm vs 30% of patients in the control arm. In the BPd arm, ocular AEs were managed with dose holds (83%) and reduced doses (59%); these AEs led to treatment discontinuation in 9% of patients in the experimental arm.