Abstracts

Rationale: This study was performed to aid in selection of doses to be used in a Phase III study. The aims of this analysis were to 1) estimate the apparent clearance (CL/F) of CLB and identify covariates that significantly affect CL/F and 2) assess efficacy and safety as a function of CLB exposure. Methods: Data were from a dose-ranging Phase-II randomized, double-blind, efficacy and safety trial in pediatric patients and young adults with Lennox-Gastaut Syndrome (LGS) administered oral CLB. This 11-week study included a 4-week baseline period, 3 weeks of dose titration, and 4 weeks of maintenance dosing (0.25 or 1.0 mg/kg/day). Drug concentrations were obtained from convenience samples at weeks 5 and 7. A non-linear mixed-effect model (NONMEM) incorporating a one-compartment model with first-order absorption and elimination was used to estimate pharmacokinetic parameters and post-hoc individual area-under-the-concentration-time-curves (AUCs). The primary efficacy endpoint in the study was percent reduction in seizure frequency (%RED) in the maintenance period from baseline. Recorded adverse events (AEs) were separated into categories: sedation, seizure, dizziness, behavior, irritable/aggressiveness (I/A), abnormal coordination, increased secretions, nutritional status, and respiratory depression. An AUC breakpoint of 5 mg*hr/L was chosen to define high exposure and low exposure groups and the percent reduction in seizure frequency from baseline was compared by Mann-Whitney test (p<0.05).Results: The final database consisted of 102 observations from 57 subjects. Body size measures were the only significant covariates effecting clearance. Among these, body surface area (BSA) was the most significant. No covariates were found to effect volume of distribution. The final model after backward deletion was: CL/F= 3.19 * BSA. Thus, population mean values of CL/F and V/F, for a typical individual in the population taking into account body surface area (BSA), were 3.19 L/hr/m