Modeling and Simulation Analysis of Aprepitant Pharmacokinetics in Pediatric Patients With Postoperative or Chemotherapy-Induced Nausea and Vomiting

Abstract

Objectives: Aprepitant is effective for the prevention of chemotherapy-induced or postoperative nausea and vomiting (CINV/PONV). The aim of this study was to develop a population pharmacokinetic (PK) model of aprepitant in pediatric patients and to support dosing recommendations for oral aprepitant in pediatric patients at risk of CINV.

Methods: A population PK model was constructed based on data from 3 clinical studies in which children (6 months to 12 years) and adolescents (12-19 years) were treated with a 3-day regimen of oral aprepitant (capsules or suspension), with or without intravenous fosaprepitant on day 1 (CINV), or a single dose of oral aprepitant (capsules or suspension; PONV). Nonlinear mixed-effects modeling was used for model development, and a stepwise covariate search determined factors influencing PK parameters. Simulations were performed to guide final dosing strategies of aprepitant in pediatric patients.

Results: The analysis included 1326 aprepitant plasma concentrations from 147 patients. Aprepitant PK was described by a 2-compartment model with linear elimination and first-order absorption, with allometric scaling for central and peripheral clearance and volume using body weight, and a cytochrome P450 3A4 maturation component for the effect of ontogeny on systemic clearance. Simulations established that application of a weight-based (for those <12 years) and fixed-dose (for those 12-17 years) dosing regimen results in comparable exposures to those observed in adults.

Conclusions: The developed population PK model adequately described aprepitant PK across a broad pediatric population, justifying fixed (adult) dosing for adolescents and weight-based dosing of oral aprepitant for children.

Keywords: CINV; PK; aprepitant; chemotherapy-induced nausea and vomiting; modeling and simulations; pediatric; pharmacokinetics.

Conflict of interest statement

Disclosure Anne Chain and Rebecca Wrishko are employees of Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA, and hold stock in the company. Grygoriy Vasilinin is an employee of Certara Inc, Montreal, Quebec, and a paid consultant for Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. Samer Mouksassi is an employee of Certara Inc, Montreal, Quebec, and a paid consultant for Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ, USA. The authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. ClinicalTrials.gov identifiers: NCT00080444, NCT00818259, NCT00819039

Copyright Pediatric Pharmacy Association. All rights reserved. For permissions, email: mhelms@pediatricpharmacy.org 2020.

Figures

Figure 1.

Goodness-of-fit plots. Top: Dependent variable or modeled observed concentrations (DV) versus population predictions (PRED) and individual predictions (IPRED) on a log-log scale with a locally estimated smoothing (loess) line in gray. The loess line is consistent and close to the identity (black line), indicating a good model fit. Middle: Conditional weighted residuals (CWRES) versus time after first dose (TIME) and versus time after last dose (TAD). The gray loess trend line on top of the zero horizontal line shows symmetrical and well-distributed residuals around zero, indicating model adequacy. Bottom: CWRES versus PRED; the gray loess trend line on top of the zero horizontal line shows symmetrical and well-distributed residuals around zero, indicating model adequacy. Absolute individual weighted residuals (|IWRES|) versus IPRED; the flat gray loess line shows a constant variance, indicating a good error model.

Figure 2.

Prediction-corrected visual predictive check (N = 1000) for the final population model of aprepitant in pediatric patients (log10-transformed data). The median, 10th, and 90th percentiles of prediction-corrected concentrations versus time after last dose by age group were computed on observed and simulated data for each age group. Pred Corr = prediction correction normalized data for multiple doses, accumulation, and other factors. Dark gray areas represent the simulated median, whereas light gray represents the 10th and 90th percentiles; the upper and lower edges of each delimit a 95% confidence interval. White solid lines represent the observed data median, and white dashed lines represent the 10th and 90th percentiles.

Figure 3.

Simulated versus observed concentrations of aprepitant following administration (A) with oral suspension (3, 2, and 2 mg/kg on days 1, 2, and 3) in pediatric patients (0.5 to

Figure 4.

Box-and-whisker plots showing simulated (pediatric patients in study 1, oral suspension) versus observed (adolescent and adult patients, oral capsules) pharmacokinetic parameters of aprepitant following oral administration, stratified by age group. The center of the box represents the median, lower and upper hinges represent the first and third quartiles, and whiskers extend to the most extreme data points.