The utility of mixed-effects covariate analysis in rapid selection of doses in pediatric subjects: a case study with fexofenadine hydrochloride.

Fexofenadine hydrochloride is a non-sedating antihistamine that is used in the treatment of symptoms associated with seasonal allergic rhinitis and chronic idiopathic urticaria. A pooled analysis of pharmacokinetic data from children 6 months to 12 years of age and adults was conducted to identify the dose(s) in children that produce exposures comparable to those in adults for the treatment of seasonal allergic rhinitis. The pharmacokinetic parameter database included peak and overall exposure data from 269 treatment exposures from 136 adult subjects, and 90 treatment exposures from 77 pediatric allergic rhinitis patients. The data were pooled and analysed using NONMEM software, version 5.0. A covariate model based on body weight and age and a power function model based on body weight were identified as appropriate models to describe the variability in fexofenadine oral clearance and peak concentration, respectively. Individual oral clearance estimates were on average 44%, 36% and 61% lower in children 6 to 12 years (n=14), 2 to 5 years (n=21), and 6 months to 2 years (n=42), respectively, compared with adults. Trial simulations (n=100) were carried out based on the final pharmacostatistical models and parameter estimates to identify the appropriate dose(s) in children relative to the marketed dose of 60 mg fexofenadine hydrochloride in adults. The trials were designed as crossover studies in 18 subjects comprising various potential dosing regimens with and without weight stratification. Pharmacokinetic parameter variability was assumed to have a log-normal distribution. Individual weights and ages were simulated using mean (SD) estimates derived from the studies used in this analysis and proportional measurement/model mis-specification errors derived from the analysis were incorporated into the simulation. The results indicated that a 30 mg dose of fexofenadine hydrochloride administered to children 1 to 12 years of age and weighing >10.5 kg and a 15 mg dose administered to children 6 months and older and weighing <or=10.5 kg produces exposures similar to those seen with the 60 mg dose in adults.