Sr. Research Investigator, CPP Lead Clinical Pharmacology and Pharmacometrics, Incyte, United States
Disclosure(s):
Yunlan Fang, PhD: No relevant disclosure to display
Objectives: INCAGN02390 is a fully human, immunoglobulin G1κ (IgG1κ) monoclonal antibody that selectively binds to the extracellular domain of T-cell immunoglobulin and mucin domain containing protein-3 (TIM-3), and potentially elicits an antitumor response by enhancing T-cell responsiveness. INCAGN02390 is being developed for the treatment of advanced malignancies. The objective of this study was to develop a population pharmacokinetic (PK)/pharmacodynamic (PD) model to characterize INCAGN02390 exposures, peripheral TIM-3 receptor occupancy (RO), and interpatient variability in patients with advanced malignancies.
Methods: The model was developed using INCAGN02390 PK and TIM-3 RO profiles from the first-in-human study of INCAGN02390 (NCT03652077); a total of 348 serum concentration records from 37 patients enrolled in the study were used for this analysis. Patients received intravenous INCAGN02390 doses ranging from 10 to 1,600 mg given every two weeks (q2w). Flow cytometry was performed to monitor total TIM-3 and unbound TIM-3 on circulating monocytes. TIM-3 RO was subsequently calculated as the percentage of INCAGN02390 bound TIM-3 in total TIM-3. PK/PD analyses were performed using a nonlinear mixed effects model within the Monolix® software (version 2021R2). Patient demographic and baseline characteristic data, including body weight, age, sex, baseline serum albumin and alkaline phosphatase concentrations, were evaluated during a covariate search using a stepwise forward-addition process. The predictive performance of the model was assessed by goodness of fit (GOF) plots and visual predictive checks (VPCs).
Results: INCAGN02390 PK and TIM-3 RO profiles were described using a two-compartment PK model with linear elimination coupled with a sigmoid Emax PD model, where Emax is the maximal drug effect. All parameters and their variabilities were estimated within an acceptable degree of precision and were physiologically plausible for an antibody. The mean (% relative standard error of estimate) estimates of linear clearance (CL) and central volume of distribution (V) were 0.011 L/h (6.4%) and 2.96 L (5.4%), respectively. Patient body weight was identified as a significant covariate for both CL and V. Baseline serum albumin concentration was also found to influence CL. The GOF plots and VPCs demonstrated that the final model provided adequate description of the observed data.
Conclusions: A population PK/PD model was successfully developed to describe the INCAGN02390 PK and peripheral TIM-3 RO in patients with advanced malignancies. The model could be leveraged as a valuable prediction tool to support future dose selection for clinical development.
