Population Pharmacokinetics (PopPK) and Exposure-Response (E-R) Analyses for Belantamab Mafodotin Monotherapy in Patients with Relapsed/Refractory Multiple Myeloma (RRMM) Enrolled in DREAMM-2 and DREAMM-3

Abstract

Background: Belantamab mafodotin (belamaf) is a B-cell maturation antigen (BCMA)-targeting antibody-drug conjugate (ADC) with a humanized afucosylated anti-BCMA monoclonal antibody (mAb) conjugated to a cytotoxic payload (microtubule-disrupting agent monomethyl auristatin F [MMAF]) by a protease-resistant maleimidocaproyl (mc) linker. This ADC binds to BCMA and induces cell death by direct cell killing and immune-mediated mechanisms. The DREAMM-2 trial (NCT03525678) showed rapid, deep, and durable responses to single-agent belamaf in patients with RRMM. The Phase 3 DREAMM-3 trial (NCT04162210) evaluated single-agent belamaf vs. pomalidomide plus low-dose dexamethasone in patients with RRMM at second relapse or later (Weisel et al., J Clin Oncol 41, 2023 [suppl 16; abstr 8007]); DREAMM-3 did not meet its primary endpoint on superiority for progression free survival (PFS). The present analysis evaluated the PopPK of the ADC and cytotoxin cysteine-mcMMAF (cys-mcMMAF) as well as the E-R relationships for key efficacy and safety endpoints to contribute to further understanding of belamaf. Methods: Patients with RRMM were administered intravenous (IV) doses of belamaf 2.5 or 3.4 mg/kg Q3W in DREAMM-2 (n=218) and 2.5 mg/kg Q3W in DREAMM-3 (n=217). Previously built PopPK models of ADC and cys-mcMMAF were updated and used to predict individual Cycle 1 exposure measures for the E-R analyses performed on the DREAMM-3 belamaf arm only. Exposure-efficacy analyses evaluated the probability of response (PoR), PFS, and time to response (TTR). Exposure-safety analyses evaluated the probability of occurrence of adverse events of special interest such as changes in best corrected visual acuity (BCVA), grade ≥2 or ≥3 corneal events, grade ≥3 thrombocytopenia and infusion-related reactions (IRR). Logistic regression models were used to assess event probabilities, while Cox proportional-hazard models were used to assess time to event endpoints. Stepwise covariate analyses were performed. Results: The PopPK data were described well by two-compartment models with linear elimination. The ADC model included time-varying clearance and the cys-mcMMAF model included decreasing MMAF:mAb ratio after each dose. ADC clearance was reduced on average by 30% over the first 5 months on treatment. The PopPK of ADC and cys-mcMMAF were both influenced by disease-related factors (baseline concentrations of soluble BCMA [sBCMA], immunoglobulin G, and albumin), as well as body weight and to a lower extent body mass index. Mild to severe renal impairment, mild hepatic impairment, age, race (Black or Asian), or prior therapies were not found to impact the PopPK of belamaf. Cycle 1 ADC average concentrations (C avg) were inversely correlated with disease burden factors and highly correlated (r=0.97) with ADC trough concentration (C tau). In DREAMM-3, PoR and PFS were higher and longer with higher ADC exposure when evaluated independently of other covariates. In the final model, PoR and PFS were respectively lower and shorter with higher baseline sBCMA or prior anti-CD38 mAb treatment; PoR was also related to presence of extra medullary disease. TTR was not related to ADC exposure or other factors. Safety endpoints were associated with exposure. Higher ADC C avg was associated with higher probability of developing grade ≥2 or ≥3 corneal events and inversely associated with time to their onset. ADC C avg was similarly associated with changes in BCVA but with a lower intercept for BCVA-related endpoints. Baseline sBCMA was inversely associated with probability of grade ≥2 corneal events. Lower baseline platelet count (but not ADC exposure) was associated with increased probability of grade ≥3 thrombocytopenia. ADC C avg was inversely associated with probability of IRR. In addition, the observed incidence of post-treatment anti-belamaf antibodies was low (<1%) with no observed impact on PK, efficacy, or safety. Conclusion: PopPK and E-R analyses for DREAMM-3 demonstrated the impact of disease factors and patient characteristics on PK, efficacy, and safety endpoints following single-agent belamaf, consistent with prior results for later-line populations.