Hyeon Soo Park(1,2), Yun Seob Jung (1,2), Dongwoo Chae (1), Choon Ok Kim (3), Kyungsoo Park (1)
(1) Department of Pharmacology, Yonsei University College of Medicine, Seoul, Korea, (2) Brain Korea 21 Plus Project for Medical Science, Yonsei University, Seoul, Korea, (3) Clinical Trial Center, Severance Hospital, Yonsei University College of Medicine, Seoul, Korea
Objectives: Glimepiride is an oral hypoglycemic agent used to treat type II diabetes mellitus (T2DM). Glimepiride directly stimulates insulin secretion and induces insulin receptor expression[1]. Hypoglycemia is a commonly reported side effect, and elucidation of a dose-response relationship would contribute to its prevention. In a previous study, H.-Y. Yun et al, constructed a PK-PD model based on the relationship between glimepiride and blood glucose.[2] In this study, glimepiride concentration, insulin concentration and blood sugar were measured in 6 healthy adult korean subjects and PK-PD model was constructed based on these results. In this regard, we have initiated a study to develop a population PK-PD model of glimepiride to investigate PK-PD relationship of the drug and the associated influencing factors. As part of the final PK-PD model, this presentation was to report on a PK model developed so far.
Methods:24 healthy male volunteers between the age of 19 and 55 with body weight over 55 kg were recruited. The subjects were given multiple doses of glimepiride 4 mg, QD for 7 days. Blood samples were collected for 48 hours after the last dose from which glimepiride concentrations were analyzed. Blood sugar was measured up to 24 hours after dose. Based on the collected drug concentration data, nonlinear mixed effect modeling was carried out using NONMEM software version 7.3. Covariate search was carried out using age, weight, alcohol uptake, smoking history, and caffeine intake as covariate candidates. To do so, stepwise covariate model building was implemented with likelihood ratio test at significance levels of p < 0.01 for forward selection and p < 0.001 for subsequent backward deletion. Inter-occasion variability(IOV) as well as inter-individual variability (IIV) was incorporated and correlations among random effects implemented using $OMEGA BLOCK. Exploratory data analysis was carried out using R software version 3.3.3.
Results:A two-compartment disposition model parameterized by central volume of distribution (Vc), peripheral volume of distribution (Vp), clearance (CL), and inter-compartmental clearance (Q), with zero-order absorption[2] was selected as the base structural model. Covariate search yielded age as a significant covariate of Vp. The estimates of Vc, Vp, CL, and Q were 9.45 L, 16.03 L, 4.429 L/h, and 1.41 L/h, respectively. The variance estimates of IIV of Vp, CL, and Q in the coefficient of variation (CV) scale were 54.18%, 26.62%, and 85.5%, respectively. IOV of Vc and CL were 53.86% and 11.93%, respectively. Residual error variability was best described by a proportional error model, yielding residual error variance of 32.94% (CV). Relative standard errors of all parameters were less than 30%, indicating the reliability of model parameters. Our model well described the observed concentration-time profile.
Conclusions: This preliminary work demonstrates that only the age-Vp relationship has significant influence on glimepiride PK while substantial variability (> 50% CV) still remains unexplained in IIV of Vp and Q and IOV of Vc.. Such unexplained high variability was conjectured to be related with a small number of subjects used here. Thus, in the future, to develop a model that better characterizes glimepiride PK, more analyses in a larger patient population possibly with a more diverse covariate distribution will be needed.
References:
[1] FDA, Amaryl label, revised Oct 2013
[2] H.-Y. Yun, H.-C. Park, W.Kang and K.-I. Kwon. Pharmacokinetic and pharmacodynamics modeling of the effects of glimepiride on insulin secretion and glucose lowering in healthy humans. Journal of Clinical Pharmacy and Therapeutics(2006) 31, 469-476
Reference: PAGE 28 (2019) Abstr 8932 [www.page-meeting.org/?abstract=8932]
Poster: Drug/Disease Modelling - Endocrine