Population Pharmacokinetic Modeling of Oxcarbazepine and Its Active Metabolite 10-Monohydroxy Derivative to Inform Dosing in Children with Obesity

Abstract

Background: Oxcarbazepine (OXZ) is an antiepileptic drug whose pharmacological effect is primarily mediated by its active metabolite, 10-monohydroxy derivative (MHD). OXZ is approved for use in adults and children older than 2 years with an age- and body weight-tiered dosing recommendation, but dosing guidance for children with obesity is lacking.

Objective: This work aimed to assess the dosing requirements of OXZ in children with obesity to support label extension.

Methods: Two multicenter studies (NCT01431326 and NCT02993861) were conducted in patients receiving standard-of-care OXZ therapy. Participants ≥ 2 years of age with a body mass index ≥ 95th percentile were classified as obese. Plasma concentrations were measured by a validated liquid chromatography tandem mass spectrometry (LC-MS/MS) assay. Nonlinear mixed effects modeling was performed using NONMEM 7.4 to characterize the population pharmacokinetics of OXZ and MHD simultaneously. Simulations were performed to compare MHD systemic exposure in children ≥ 2 years of age with and without obesity.

Results: One hundred study participants with a median (range) age of 9 years (44 days-20.90 years) contributed 425 plasma concentrations of OXZ (n = 212) and MHD (n = 213). Fifty-two percent of the participants had obesity. A one-compartment joint parent-metabolite model with linear input-output and bi-directional transformation between OXZ and MHD best characterized the pharmacokinetics. Body size was the only covariate affecting pharmacokinetics, and a fat-free mass-based metric termed pharmacokinetic weight (PKWT) best characterized that effect allometrically. Simulation results revealed that the current dosing regimen of OXZ can produce comparable exposure of MHD in children ≥ 2 years of age with and without obesity.

Conclusion: A model-informed analysis confirms that the current pediatric dosing regimen of OXZ applies to children in general, regardless of their obesity status.

Fig. 1

Schematic of joint parent-metabolite model of oxcarbazepine (OXZ) and its mono-hydroxy derivative (MHD) metabolite. ${CL}_{oxz}$ and ${CL}_{mhd}$ are the elimination clearances of OXZ and MHD respectively; $V_{oxz}$ and $V_{mhd}$ are the volumes of distribution of OXZ and MHD, respectively; $K_a$ and $K_{bt}$ are the first order absorption rate constant and backward transformation rate constant, respectively.

Fig. 2

Diagnostic plots of the final population pharmacokinetic model. The top and bottom panels represent the observed concentrations and conditional weighted residual (CWRES) vs. predicted concentrations of mono-hydroxy derivative (MHD), and oxcarbazepine (OXZ), respectively. The solid black line, dashed black line, and dashed blue line represent the line of unity, zero line, and smooth line, respectively.

Fig. 3

Reference-corrected visual predictive check (rcVPC) of the final population pharmacokinetic model. The figure represents the rcVPCs of the Pharmacokinetic weight (PKWT)-based model for oxcarbazepine (OXZ) (left) and mono-hydroxy derivative (MHD) (right) using 1000 replicates. The open circles represent the observed concentrations along with the median (solid red line) and 90% interval (dashed red lines). The simulations are represented as the median (solid black line) and the 90% prediction interval (dotted black lines) along with their 95% confidence interval (CI) of prediction as shaded areas. The rcVPCs were normalized using a reference dataset representing children >2 years of age with obesity who received the recommended age- and body weight–tiered dosing regimen per the TRILEPTAL^® prescribing information (Table 1). Patients whose characteristics did not align with the defined tiers were assigned the same dosing and covariate values as patients aged 2 to 4 years. The reference dataset maintained the observed dosing frequency and included representative (i.e., median) covariates within each dosing tier.

Fig. 4

Simulated steady state exposure of mono-hydroxy derivative (MHD) in children from 2–20 years of age following the labelled dosing regimen, i.e., based on age and body weight bands: (a) plasma concentration-time profiles, (b) trough concentration (${\mathit{C}}_{\mathit{t}\mathit{r}\mathit{o}\mathit{u}\mathit{g}\mathit{h},\mathit{s}\mathit{s}})$, (c) peak concentration (${\mathit{C}}_{\mathit{m}\mathit{a}\mathit{x},\mathit{s}\mathit{s}})$, and (d) average concentration (${\mathit{C}}_{\mathit{a}\mathit{v}\mathit{g},\mathit{s}\mathit{s}})$. The dashed lines represent the reference trough concentration range recommended by the International League of Epilepsy Research.

Fig. 5

Simulated steady state trough concentrations of mono-hydroxy derivative (MHD) (${\mathit{C}}_{\mathit{t}\mathit{r}\mathit{o}\mathit{u}\mathit{g}\mathit{h},\mathit{s}\mathit{s}})$ in various age and weight bands following the labelled dosing regimen. The dashed lines represent the reference trough concentration range recommended by the International League of Epilepsy Research.

Fig. 6.

Visualization of the effects of obesity and age on the clearance of mono-hydroxy derivative (MHD) in children. Individual clearance (CL) of MHD were simulated from the final population pharmacokinetic models. (a) Effects of age and obesity: absolute CL increases both with age and obesity in children (b) Effects of age: body size normalized CL (i.e., CL per kg pharmacokinetic weight [PKWT]) is higher in younger children than the older children irrespective of obesity status; Effect of obesity: body size normalized CL is lower in the obese group at any age (c) Accounting for the effects of both age and obesity by allometric scaling to body size (i.e., less-than proportional scaling to PKWT with estimated exponents 0.67). The red and blue solid lines represent the smooth lines for the populations with and without obesity, respectively.

Fig. 7

Individual dosing rate to clearance ratio in children. The dosing rate was determined by the recommended dosing by age and dosing tiers in the TRILEPTAL^® prescribing information, and the individual clearance (CL) of mono-hydroxy derivative (MHD) were simulated from the final population pharmacokinetic models. The y-axis thus represents the average steady state concentration of MHD. The red and blue solid lines represent the smooth lines for the populations with and without obesity, respectively.