Abstract
Belantamab mafodotin was investigated for RRMM in combination with lenalidomide + dexamethasone in Arm A of the phase I/II DREAMM-6 (NCT03544281) study, in combination with bortezomib + dexamethasone (BVd) in Arm B of DREAMM-6 and in the phase III DREAMM-7 (NCT04246047) study, and in combination with pomalidomide + dexamethasone (BPd) in the phase III DREAMM-8 (NCT04484623) study. In DREAMM-7 (Hungria V, et al. Lancet Oncol 2025) and DREAMM-8 (Dimopoulos MA, et al. NEJM 2024), BVd and BPd demonstrated significant progression-free survival (PFS) benefits, with >20-month benefits in median PFS vs comparators; a significant overall survival (OS) benefit was also observed in DREAMM-7, and in DREAMM-8 there was a positive trend for OS favoring BPd, with follow-up ongoing. An integrated exposure–response (ER) analysis was conducted across these trials to characterize the benefit–risk profile of belantamab mafodotin, with a focus on comparing efficacy and ocular safety relationships and model predicted outcomes at 2.5 mg/kg vs 1.9 mg/kg starting doses.
Belantamab mafodotin was administered IV at 1.9 mg/kg or 2.5 mg/kg with multiple schedules in DREAMM-6, at 2.5 mg/kg every 3 weeks in DREAMM-7, and at 2.5 mg/kg in Cycle 1 (C1) followed by 1.9 mg/kg every 4 weeks in DREAMM-8. Individual C1 exposures were derived using a population pharmacokinetic (PK) model (Papathanasiou T, et al. Clin Pharmacokinet 2025) with actual patient (pt) dosing, demographics, and PK post-hoc parameters that were generated from individual PK data. ER relationships for efficacy (e.g., very good partial response or better [≥VGPR]) and ocular safety (e.g., occurrence/timing of first grade [Gr] ≥3 ocular adverse events [oAEs] per Common Terminology Criteria for Adverse Events [CTCAE], bilateral best-corrected visual acuity [BCVA] worsening to 20/50 or worse, and Gr≥2 ophthalmic exam findings [OEFs] per the Keratopathy and Visual Acuity scale comprising BCVA changes and slit lamp findings) during the course of treatment were assessed using logistic regression or Cox proportional hazards models. Graphical and statistical analyses included multivariate models incorporating baseline characteristics and disease-related covariates.
Data from 516 pts were included (n=124 DREAMM-6; n=242 DREAMM-7; n=150 DREAMM-8). Higher C1 belantamab mafodotin exposure was associated with greater probability of achieving a VGPR or better response. Across the C1 exposure range, there was no meaningful difference in Gr≥3 oAE risk per CTCAE or probability of bilateral BCVA worsening to 20/50 or worse, however the probability of Gr≥2 OEFs was associated with exposure. In multivariate model predictions (accounting for disease factors and baseline characteristics), probability of ≥VGPR was 70% at a starting dose of 2.5 mg/kg vs 54% at 1.9 mg/kg. The probability of Gr≥3 oAEs was 52% at both doses, probability of bilateral BCVA worsening to 20/50 or worse was 33% at 2.5 mg/kg vs 28% at 1.9 mg/kg, and probability of Gr≥2 OEFs was 90% at 2.5 mg/kg and 79% at 1.9 mg/kg. The exposure–efficacy curve for ≥VGPR had a steeper slope than the exposure–safety curves for Gr≥3 oAEs and bilateral BCVA worsening to 20/50 or worse; the exposure–safety curve for Gr≥2 OEFs had a similar slope to the exposure–efficacy curve for ≥VGPR, suggesting that lower C1 exposures/starting doses reduce the risk of Gr≥2 OEFs, but with a reduction in efficacy.
This analysis demonstrates that in RRMM, higher C1 belantamab mafodotin exposure is associated with deeper response (≥VGPR), which is critical for improved long-term, durable clinical outcomes such as PFS/OS. While lower C1 exposure is associated with improvements in Gr≥2 OEFs, only modest benefits in Gr≥3 oAEs and bilateral BCVA worsening to 20/50 or worse were observed with lower exposure, and there was a demonstrable tradeoff in efficacy. This was further validated by multivariate model predictions. As schedule extensions in subsequent doses allow for better management of ocular events without loss in efficacy (Mateos MV, et al. Blood Adv 2025:in press), a 2.5 mg/kg starting dose followed by dose reductions and schedule extensions to manage ocular events is a rational belantamab mafodotin dosing strategy for RRMM treatment. This strategy can induce deep responses and subsequently provide a more tolerable dosing schedule while maintaining response and led to robustly positive efficacy outcomes in DREAMM-7 and -8.
