Relationship between systemic corticosteroid exposureand growth velocity: development and validation of a pharmacokinetic/pharmacodynamic model

Abstract

Background: Use of high-dose oral corticosteroids (CSs) can reduce growth velocity (GV) in children, whereas use of low-dose topical CSs has either no effect or transient effects on short-term growth and no effect on final adult height Despite the large body of literature on this topic, some fundamental questions remain concerning the relationship between CS exposure and growth effects. Objectives: The aims of this study were to determine the relationship between CS exposure and GV in children receiving CS therapy for asthma or rhinitis, and to examine whether there is likely to be a link between GV and cortisol suppression. Methods: Data from 32 published studies of the effect on growth of inhaled, intranasal, and oral CSs, includingdelivery by dry powder inhaler, metered-dose inhaler, and aqueous nasal spray, were consolidated by expressing CS exposure in cortisol equivalents using a physiologically based pharmacokinetic/pharmacodynamic approach. The relationship between change in GV and CS exposure in cortisol equivalents was described using a nonlinear sigmoid maximum-effect (E max) model with the following parameters: E max = −5.9 cm/y; steady-state unbound AUC for 50% reduction in GV, in cortisol equivalents = 20,000 ng-h/L; Hill constant= 1.2; and change in GV at zero systemic exposure = 0.06 cm/y. Validation was achieved by comparing the model's predictions with data from 5 studies that were not included in the model development The model was also used to predict the potential of various CS regimens to reduce GV Results: Exploratory data analysis established that change in GV was highly correlated with exposure in cortisol equivalents ( P < 0.001). CSs with high systemic bioavailability by the intranasal route were predicted to have short-term growth effects exceeding the clinical equivalence limit for change in GV (±0.8 cm/y), whereas those with lower bioavailability were predicted to produce systemic exposures below the threshold for significant effects on GV The findings were similar for inhaled CSs and for regimens combining delivery by the intranasal and inhaled routes. In descending order, the model predicted the following ranking of the potential of the various intranasal, inhaled, and oral regimens to reduce GV, expressed as fractions or multiples of the pediatric dose (in μg/d): oral prednisolone 5000 μg/d, 0.14; inhaled beclomethasone dipropionate metered-dose inhaler 400 μg/d, 0.54; inhaled budesonide dry powder inhaler 400 μg/d, 0.66; intranasal triamcinolone acetonide aqueous nasal spray 220 μg/d, 0.74; inhaled triamcinolone acetonide metered-dose inhaler 400 μg/d, 0.75; intranasal beclomethasone dipropionate aqueous nasal spray 336 pg/d, 0.89; inhaled mometasone furoate dry powder inhaler 200 μg/d, 2.4; intranasal budesonide aqueous nasal spray 128 μg/d, 2.5; inhaled fluticasone propionate dry powder inhaler 200 μg/d, 2.6; intranasal mometasone furoate aqueous nasal spray 100 μg/d, 120; and intranasal fluticasone propionate aqueous nasal spray 100 pg/d, 150. Values >1 are predictive of no significant effect on GV The model predicted that a 10% to 15% reduction in plasma cortisol concentration should be detectable at the lower equivalence limit for growth reduction (−0.8 cm/y). The validation procedure showed that the model was capable of predicting the results of the 5 comparative growth studies not included in model development with a correlation coefficient of 0.98. Conclusions: Growth effects appear to be nonlinearly related to CS exposure; therefore, no-effect exposure should be possible for CSs with low systemic exposure. Growth inhibition appears unlikely to occur in the absence of detectable reductions in cortisol concentrations.