Sejung Hwang (1), Soyoung Lee (1), Jangsoo Yoon (1), Jae-Yong Chung (2)
(1) Department of Clinical Pharmacology and Therapeutics, Seoul National University College of Medicine and Hospital, Seoul, Republic of Korea, (2) Department of Clinical Pharmacology and Therapeutics, Seoul National University College of Medicine and Bundang Hospital, Seongnam, Republic of Korea.
Objectives: Carvedilol is an adrenergic receptor antagonist and used as first-line therapy for cardiovascular disease including hypertension. It is mainly metabolized by cytochrome P450 enzyme (CYP) 2D6 and CYP2C9 and the active metabolites are less than 10% of the parent form. The study aimed to establish the population pharmacokinetic (PK) – pharmacodynamic (PD) model of carvedilol and explored the impact of CYP2D6 genetic polymorphism on PK and PD properties of carvedilol.
Methods: The PK-PD model was developed from a clinical study conducted in healthy subjects with three CYP2D6 phenotype groups. Of 21 subjects, six subjects were CYP2D6 *1/*1 or *1/*2 genotype, seven subjects were CYP2D6*1/*10 or *2/*10 genotype, and eight subjects were CYP2D6*10/*10 genotype. All subjects received oral doses of carvedilol 12.5 mg QD (3 days), sequentially 25 mg QD (5 days), and 12.5 mg (3 days). Isoproterenol sensitivity test was conducted to measure the effect of carvedilol suppressing tachycardia induced by isoproterenol. The PK-PD model was developed sequentially using nonlinear mixed-effect modeling (NONMEM) software (version 7.4). A total of 450 plasma observations and 1003 heart rate observations were used for the construction of the PK and PD model, respectively. Subjects’ age, height, weight, and CYP2D6 phenotype were assessed as potential covariates. The changes in heart rates induced by isoproterenol were explored using a simple Emax or sigmoid Emax model. The PD model was linked with PK by computing individual PK parameters of the final PK model. A direct effect response and inhibitory Emax model was used for developing the PK-PD model of carvedilol. The final model was evaluated by assessing goodness-of-fit (GOF) plots, visual prediction checks (VPC), and bootstrapping.
Results: The PK of carvedilol was well described by a two-compartment model with zero-order absorption, and absorption lag time, and first-order elimination. The population PK parameter estimates (relative standard error, RSE) of clearance (CL) in EM and IM-1, the volume of distribution of central compartment (V1) and peripheral compartment (V2) were 153 L/h (0.7%), 440 L (35%), and 754 L (2.3%), respectively. The estimates (RSE) of duration of zero-order absorption (D1), absorption lag time (Tlag), and inter-compartmental clearance (Q) were 0.38 h (70%), 0.22 h (0.6%), and 41.3 L/h (5.1%), respectively. The clearance of carvedilol decreased 32.8% in the CYP2D6*10/*10 group compared to the other groups. The population PD parameter estimates (residual standard error) of baseline heart rate (E0), the potency of isoproterenol (ED50), and maximal effect of carvedilol (Emax) on heart rate reduction were 60.4 bpm (3.1%), 0.69 μg (30.9%), and 30.7 bpm (21.9%), respectively. The IC50 estimated from the final PK-PD model was 16.5 ng/mL regardless of the CYP2D6 phenotype. The model evaluation by GOF plot, VPC, and bootstrapping results suggested that the final PK-PD model was robust and adequate with good precision.
Conclusions: The PK-PD model identified that the decreased function allele of CYP2D6 affected the clearance of carvedilol and the genotype effect on the PD appears to be minimal.
References:
[1] Ruffolo RR Jr, et al. Eur J Clin Pharmacol (1990) 38, S82-88.
[2] Morgan T. Clin Pharmacokinet (1994) 26, 335-46.
[3] Tenero D, et al. J Clin Pharmacol (2000) 40, 844-53.
[4] Sehrt D, et al. Pharmacogenetics (2011) 12, 783-95.
[5] Cleaveland CR, et al. Arch Intern Med (1972) 130, 47-52.
Reference: PAGE 29 (2021) Abstr 9718 [www.page-meeting.org/?abstract=9718]
Poster: Drug/Disease Modelling - Other Topics