Population Pharmacokinetic (popPK) of BMS-986012, a Novel Anti-Fucosyl-GM1 Monoclonal Antibody, in Patients with First-Line and Relapsed/Refractory Extensive-Stage Small Cell Lung Cancer (ES-SCLC)

Objectives: BMS-986012 is a first-in-class, fully human IgG1 mAb designed to target the ganglioside fucosyl-GM1, a unique surface antigen with high expression in SCLC cells [1]. In Phase 1 and 2 trials, BMS-986012 has shown potential therapeutic benefit when administered in combination with nivolumab (nivo) and carboplatin/etoposide (chemo) in ES-SCLC patients [2]. The objectives of this analysis were to characterize BMS-986012 PK when administered alone or in combination with nivo or nivo/chemo, determine the influence of covariates on model parameters and exposures using a popPK approach, and to provide exposures for Exposure-Response analyses of efficacy and safety.Methods: The analysis dataset included data from 2 studies, CA001-030 and CA001-050, in relapsed/refractory and first-line ES-SCLC patients receiving intravenous (IV) infusions of BMS-986012 ranging from 70 to 1000 mg, either alone or in combination with nivo or nivo/chemo. The dataset contained 857 measurable concentrations from 170 adult patients. The popPK model was developed by determining the structural, inter-individual variability, and residual error components. Several covariates were assessed, including body weight, age, sex, hepatic/renal function, tumor size, and performance score. The popPK model was specified in terms of fixed- and random-effect parameters estimated by nonlinear regression using the FOCEI method in NONMEM (v.7.5). Individual BMS-986012 exposures were simulated for different dosing regimens based on empirical Bayes estimates of PK parameters.Results: The PK of BMS-986012, was best described by a 2-compartment zero order IV infusion model with first order elimination. The model includes a proportional residual error model and random effects on clearance (CL), central and peripheral volume (Vc, Vp), and correlation of random effects. The effect of body weight on CL and volume, tumor size on CL, are found to be statistically significant. This led to numerically higher ( < 30% differences) CL and volume in participants with higher baseline body weight. However, the body weight effect on BMS-986012 PK was unlikely to be clinically relevant, given the flat exposure-response relationships for efficacy and safety at the exposure range of tested doses. Additionally, target-mediated drug disposition, which was observed in nonclinical species, was not detected in SCLC patients at current dose range. Through individual/population participants’ exposure simulations, dose linearity across various dosage levels was observed with no drug-drug interaction between BMS-986012 and co-administered nivo or chemo.Conclusions: The popPK model adequately described the observed serum BMS-986012 concentration-time data following IV infusion in SCLC patients and enabled robust exposure prediction.Citations: [1] Ponath, P., et al (2018). Cancer. Clin Cancer Res, 24(20), 5178-5189.[2] Chu, Q., et al (2022). JTO Clin Res Rep, 3(11), 100400.