Population Pharmacokinetics and Model-Informed Precision Dosing of Clobazam Based on the Developmental and Genetic Characteristics of Children with Epilepsy

Abstract

Background/Objectives: This study aimed to characterize the pharmacokinetic profiles of clobazam (CLB) and its active metabolite, N-desmethylclobazam (N-CLB), by establishing a population pharmacokinetic (PPK) model in Chinese children with epilepsy to propose individualized dosing regimens that achieve better clinical outcomes. Methods: This study examined plasma samples collected from 103 pediatric patients with refractory epilepsy undergoing CLB treatment. The plasma concentrations of CLB and its active metabolite N-CLB were measured. The developmental characteristics, CYP2C19 genotype, concomitant medications, and liver and kidney function indicators of the children with epilepsy were considered potential factors affecting the pharmacokinetic characteristics of CLB and N-CLB and analyzed using a PPK modeling approach. Results: A total of 156 samples were attained for PPK model development. The pharmacokinetic profiles of CLB and N-CLB were described using a tandem one-compartment model with first-order elimination. Body weight and CYP2C19 genotype showed statistical significance for CLB and/or N-CLB clearance. The N-CLB/CLB metabolic ratios of AUC_24h, C_min, and C_max in a steady state were as follows: normal metabolizers (NMs) < intermediated metabolizers (IMs) < poor metabolizers (PMs). The final model achieved good prediction performance and stability. Monte Carlo simulations demonstrated that the trough concentrations of CLB and N-CLB in children with epilepsy could reach satisfactory target values under varying dose regimens in CYP2C19 NMs and IMs, whereas there was a failure to achieve the desired trough concentrations of CLB and N-CLB simultaneously in CYP2C19 PMs due to the accumulation of N-CLB. Conclusions: Body weight and CYP2C19 genotype had an impact on CLB and/or N-CLB clearance in children with epilepsy. To ensure safe treatment, it is recommended to use the concentration of N-CLB as the target indicator for therapeutic drug monitoring and dose adjustments in CYP2C19 PMs. These results provide evidence for guiding the precise use of CLB.

Keywords: CYP2C19 genotype; body weight; children; clobazam; dosing optimization; population pharmacokinetics.

Figure 1

Goodness-of-fit plots of the final model for CLB (upper panel) and N-CLB (lower panel). (A,E) Observed concentration versus individual-predicted concentration (IPRED); (B,F) observed concentration versus population-predicted concentration (PRED); (C,G) conditional weighted residuals (CWRES) versus PRED; (D,H) CWRES versus time. The black straight line represents the unity line (y = x), and the red curve represents the LOESS line (Locally Weighted Regression Curve) in (A,B,E,F).

Figure 2

Visual predictive check (VPC) of serum concentrations versus time for (A) CLB and (B) N-CLB. Prediction-corrected VPC (pc-VPC) results for (C) CLB and (D) N-CLB. Observation data (circles); red dashed lines, 10th and 90th percentiles of observed concentrations; red solid lines, 50th percentile of observed concentrations; black dashed lines, 10th and 90th percentiles of simulated concentrations; black solid lines, 50th percentile of simulated concentrations; semitransparent field, 90% confidence intervals of the simulated 10th, 50th, and 90th percentiles from 1000 simulations.

Figure 3

Changes in the clearance of CLB and N-CLB with body weight.

Figure 4

The clearance of CLB and N-CLB in CYP2C19 NMs, IMs, and PMs. Each box-and-whisker plot depicts the distribution of CL across CYP2C19 genotype groups by showing medians (horizontal line), interquartile ranges (box, 25th–75th percentiles), and upper and lower whiskers (5th–95th percentiles).

Figure 5

The ratios of N-CLB to CLB in different CYP2C19 genotype groups related to AUC_24h (red), C_max (green), and C_min (blue) in a steady state estimated using the maximum a posteriori Bayesian method. The box displays the interquartile range (25th–75th percentiles), while the whiskers extend to the 5th–95th percentile range. The horizontal line within each box represents the median value.

Figure 6

The simulated concentration–time profiles for CLB and N-CLB when a 20 kg subject was treated with CLB at the dose of 0.2 mg·kg⁻¹ every 12 h for 10 consecutive days without any cotreatment in CYP2C19 (A) NM; (B) IM; (C) PM. The median values of the simulated CLB and N-CLB concentrations are represented by solid black and red lines, respectively, while their 10th–90th percentile ranges are shown with corresponding dashed lines.