(W-116) Simultaneous Population PKPD Analysis of Belantamab Mafodotin, Soluble BCMA, and Serum M-Protein in Subjects With Relapsed/Refractory Multiple Myeloma (RRMM)

John D. Clements, N/A – Director Clinical Pharmacology Modeling and Simulation, GSK, Clinical Pharmacology Modeling and Simulation, Collegeville, PA, USA; Christine Neumar, N/A – Associate Director Pharmacometrics, Certara Drug Development Solutions, Certara USA, Inc., Radnor, PA USA; Adekemi Taylor, N/A – Vice President, Certara Drug Development Solutions, Certara Drug Development Solutions, Certara USA, Inc., Radnor, PA USA; Mun Sang Yue, N/A – Associate Director Pharmacometrics, Vir Biotechnology Inc., San Francisco, CA, USA; Cindy Chen, N/A – Director Clinical Pharmacology Modeling and Simulation, GSK, Clinical Pharmacology Modeling and Simulation, Collegeville, PA, USA; Geraldine Ferron-Brady, N/A – Senior Director Clinical Pharmacology Modeling and Simulation, GSK, Clinical Pharmacology Modeling and Simulation, Collegeville, PA, USA; Herbert Struemper, N/A – Director, Clinical Pharmacology Modeling and Simulation, GSK, Clinical Pharmacology Modeling and Simulation, Research Triangle Park, NC, USA

Objectives: Belantamab mafodotin, an antibody-drug conjugate (ADC), targets B-cell maturation antigen (BCMA) on multiple myeloma (MM) cells. This analysis aimed to provide insights into belamaf pharmacology related to patients’ soluble BCMA (sBCMA) and M-protein levels.

Methods: Building on previous population pharmacokinetic (PK) and serum M-protein models [1,2], a simultaneous PKPD model with sBCMA for patients with RRMM was developed to elucidate interactions between PK and pharmacodynamics (PD) based on belantamab mafodotin monotherapy data from DREAMM-1 (NCT02064387), DREAMM-2 (NCT03525678), DREAMM-3 (NCT04162210), and DREAMM-5 (NCT04126200). Sensitivity analyses [3] addressed overparameterization. Model development utilized NONMEM (v7.4.3) with Monte Carlo SAEM and importance sampling methods. Selected extrinsic/intrinsic factor impacts were assessed. Simulations performed in a relevant virtual population used model parameters sampled from eta distributions and a dose and schedule range.

Results: Data from 565 patients were included. The 2-compartment ADC base model had a linear clearance component and target-mediated drug disposition assuming quasi-equilibrium binding between ADC and sBCMA. M‑protein reduction is key to efficacy in MM. Free ADC affected M-protein elimination rate via an effect compartment; M‑protein in turn affected sBCMA synthesis rate. sBCMA was assumed to diffuse into the total sBCMA compartment via a reservoir compartment, allowing the model to capture different timescales observed in the data. The model included between-subject variability (BSV) on clearance and central/peripheral volume of distribution (Vd) of total ADC and M-protein growth/elimination rates. Stepwise covariate modeling identified significant effects of: sex, bodyweight, baseline sBCMA, and albumin levels on ADC clearance; sex and bodyweight on central Vd; and baseline sBCMA and extramedullary disease on M-protein elimination rate. All model parameters were well estimated (relative standard error < 30%, and BSV shrinkage < 30%); diagnostics indicated good model fit to observed data. Simulations indicated steady-state concentrations of free sBCMA, ADC-sBCMA complex, and M-protein decrease as free ADC increases with increasing belantamab mafodotin dose in the simulated dose range.

Conclusions: A fully integrated simultaneous PKPD model of total and free ADC, ADC-sBCMA complex, free sBCMA, and M-protein captured the highly dynamic nature of ADC PK and its effect on MM cells and disease progression. ADC free fraction predictions may aid a deeper understanding of the drivers of belantamab mafodotin efficacy.

Citations: 1. Rathi, C, et al. CPT Pharmacometrics Syst Pharmacol 2021;10:851–63.

2. Collins, J, et al. CPT Pharmacometrics Syst Pharmacol 2023;12:1411–24.

3. Sorribes IC, et al. J R Soc Interface 2020;17:20190722.