(W-128) Population pharmacokinetics and exposure-safety of DS-6157a in patients with advanced gastrointestinal stromal tumor (GIST)

Maura Fallon, MD – Executive Director, Clinical Safety & Pharmacovigilance, Daiichi Sankyo, Inc.; Kristin Follman, PhD – Clinical Pharmacologist, Certara; Sergio Iadevaia, PhD – Sr Director, Pharmacometrics, Certara; Madan Kundu, PhD – Director, Biostatistics & Data Management, Daiichi Sankyo, Inc.; Abderrahmane Laadem, MD – Global Head, Clinical Development, Daiichi Sankyo, Inc.; Yvonne Lau, PhD – Executive Director, Oncology, Quantitative Clinical Pharmacology, Daiichi Sankyo, Inc.; Satoshi Nishioka, MS, PhD – Clinical Scientist, Daiichi Sankyo, Inc; Julia Shi, PhD – Biostatistition, Sarah Cannon Development Innovations; Xinyuan Zhang, PhD – Director, Quantitative Clinical Pharmacology, Daiichi Sankyo, Inc.

Objectives: DS-6157a is a first-in-class antibody–drug conjugate (ADC) that targets G protein-coupled receptor 20 (GPR20) and comprises a humanized anti-GPR20 monoclonal antibody that is covalently conjugated to an enzyme-cleavable linker, and a topoisomerase I inhibitor payload (DXd, also known as MAAA-1181a) at a drug-to-antibody ratio of ~8. We aimed to characterize the pharmacokinetics (PK) of DS-6157a (intact ADC) and DXd (released payload) through the development of a population PK (PopPK) model and to evaluate exposure–response (ER) relationships for select safety endpoints.

Methods: PopPK and ER models were developed using data from a first-in-human dose escalation study of DS-6157a (n=34) in patients with GIST, with DS-6157a dosages ranging from 1.6–12.8 mg/kg Q3W (once every three weeks) intravenous (IV) administration.[1] Nonlinear mixed effects modeling was used to develop the PopPK model. First, a sequential DS-6157a – payload model was developed. A final simultaneous DS-6157a and payload model was then developed. Exposure-safety analyses for select safety endpoints were conducted using multivariate logistic regression. Exposure metrices of DS-6157a and DXd as predicted by the PopPK model were evaluated in the exposure-safety analyses.

Results: DS-6157a PK was well described with a 2-compartment PK model with linear elimination, and baseline body weight on the volume of the central compartment was identified as a significant covariate. DXd PK was well described with a 1-compartment PK model with time-dependent payload release rate and parallel linear and nonlinear elimination. Baseline albumin on DXd clearance was found to be a significant covariate in PopPK. A univariate logistic regression model comprised of an intercept term and a slope coefficient for DXd steady-state minimum concentration best predicted the occurrence of any CTCAE Grade and Grade ≥3 neutropenia and thrombocytopenia events. No covariates were found to be significant in the exposure-safety analysis.

Conclusions: PK of DS-6157a and DXd were well characterized by the PopPK model. Although there were limited efficacy signals,[1] positive ER relationships for safety endpoints (neutropenia and thrombocytopenia) were observed.

Citations: [1] George S, et al. Clin Cancer Res. 2023;29(18):3659–3667.