Yoonjin Kim1, Eunsol Yang1, JaeJin Nah2, Jae Min Cho2, Yoonhye Jeong2, SeungHwan Lee1
1Department of Clinical Pharmacology and Therapeutics, Seoul National University College of Medicine and Hospital, 2Daewoong Pharmaceutical Co., Ltd.
Objectives: Enavogliflozin is a novel sodium-glucose cotransporter 2 inhibitor developed for the treatment of type 2 diabetes mellitus (T2DM). This study aimed to develop a population pharmacokinetic (PK) model to characterize its PK in target patients, including those with impaired organ function. Additionally, it sought to identify factors contributing to its PK variability. Methods: A prior-base model was developed using log-transformed plasma concentration-time data from seven clinical studies with full PK data (five in healthy volunteers, one each in patients with hepatic or renal impairment). A base model was then refined by incorporating trough PK data from four studies in T2DM patients. The final model incorporated covariates that affect the pharmacokinetics of enavogliflozin. Continuous covariate candidates included age, height, weight, body mass index, estimated glomerular filtration ratio, aspartate transaminase, alanine transaminase, albumin, total bilirubin, and total protein, while categorical covariates included sex, T2DM status (T2DM patient/healthy), hepatic function (normal/Child-Pugh class A/Child-Pugh class B), and renal function (normal/mild/moderate/severe). Covariates were screened through visual inspection and tested using stepwise forward selection (p < 0.05) and backward elimination (p < 0.01) based on the criteria of the objective function value (OFV). Model evaluation included parameter plausibility, OFV analysis, goodness-of-fit (GOF) plots, and visual predictive checks (VPCs). The model development was performed with NONMEM version 7.4 (ICON PLC, MD, USA), Python version 3.9 (Python Software Foundation, DE, USA) was used for data exploration and generating diagnostic plots. Results: The pharmacokinetics of enavogliflozin was best described with a three-compartment model with zero-order absorption and first-order elimination. An exponential variance model was used to describe interindividual variability for apparent clearance (CL/F), apparent volume of distribution for three compartments (V1/F, V2/F, V3/F), and absorption lag-time (ALAG1). An additive residual error model was used to describe residual variability. Age was a significant covariate for CL/F. The GOF plots for the final PK model showed that the model-predicted concentrations were generally in agreement with the observed concentrations. Additionally, the VPC plots showed that the median, 5th, and 95th percentiles of the observed concentrations were well captured by the 95% confidence intervals for the corresponding percentiles of the simulated concentrations. Conclusion: A three-compartment model with zero-order absorption and first-order elimination adequately described the pharmacokinetics of enavogliflozin in target patients, including those with impaired hepatic or renal function, with age identified as a significant covariate for CL/F. Model evaluation confirmed the reliability of the final PK model, as indicated by good agreement between observed and predicted concentrations. These findings provide a foundation for further investigations into dose optimization in various patient populations.
Reference: PAGE 33 (2025) Abstr 11506 [www.page-meeting.org/?abstract=11506]
Poster: Drug/Disease Modelling - Endocrine