Simulation Framework to Investigate Efficacy and Ocular Safety of Belantamab Mafodotin Combinations in Relapsed/Refractory Multiple Myeloma

Authors:
Open Access
Publisher website
Google Scholar
Abstract
Quantitative modeling can guide drug development and regulatory decisions by providing insights into pharmacokinetics, pharmacodynamics, efficacy, and safety. Integrating models allows simulations that support dosing considerations. A simulation framework for belantamab mafodotin (B) in relapsed/refractory multiple myeloma was developed using data from monotherapy studies and combination studies with bortezomib/dexamethasone (BVd) and pomalidomide/dexamethasone (BPd). Individual changes from baseline in serum M‐protein concentrations were modeled using a tumor growth inhibition model. Changes in ophthalmic exam findings (OEFs) grades per the Keratopathy and Visual Acuity (KVA) scale were modeled using a continuous‐time Markov model. Belantamab mafodotin concentrations were modeled with a population pharmacokinetic model. All final models adequately described the clinical data. The M‐protein/KVA/pharmacokinetic models were integrated in a simulation framework, and simulations of treatment outcomes were performed. Simulated regimens included dose modifications (holds/reductions) for OEFs. Simulations of protocol dosing and dose modifications from the phase 3 DREAMM‐7 (BVd) and DREAMM‐8 (BPd) trials predicted similar results to the respective trials (for BVd simulations/DREAMM‐7 data: very good partial response rates [VGPR+] 62.0%/62.0%, median progression‐free survival [mPFS] 39.4/36.6 months, and grade ≥ 3 OEFs 87.5%/76.5%; for BPd simulations/DREAMM‐8 data: VGPR+ rates 60.2%/68.0%, mPFS not reached/not reached, and grade ≥ 3 OEFs 77.2%/82.5%). Simulations of alternative BVd regimens indicated that a 2.5 mg/kg starting dose produces the highest efficacy, and schedule extensions after the first BVd dose can improve tolerability while continuing to demonstrate efficacy. This framework adequately predicted outcomes from DREAMM‐7/DREAMM‐8, supporting its use to predict benefit–risk profiles of alternate belantamab mafodotin dosing strategies.

This publication has 25 references indexed in Scilit: