Development of in vivo bioequivalence methodology for dermatologic corticosteroids based on pharmacodynamic modeling

Background: Dermatologic corticosteroid products produce skin blanching tha t is related to clinical potency and dose. (For application of the vasocons trictor assay to bioavailability and bioequivalence assessment, dose is def ined in terms of duration of treatment exposure [dose duration], so the ter ms dose and dose duration have been used interchangeably). The vasoconstric tor assay is the method of choice to assess dermatologic corticosteroid pro ducts bioequivalence if dose-response is validated, This article examines d ose-response validation to meet objectives of US Food and Drug Administrati on (PDA) bioequivalence guidance for dermatologic corticosteroid products, Methods: An exploratory dose-response study was conducted to determine appl icability of the empirical maximum effect (E-max) model to the individual s ubject and population dose-response relationships of six dermatologic corti costeroid product creams that varied from the most to the least potent clas ses. Products were applied to the skin of 10 healthy subjects in each of tw o dosing periods for dose durations of 0.5, 1, 2, and 6 hours. Skin blanchi ng was measured by reflectance colorimeter through 24 hours after applicati on. Area under the effect curve (AUEC) was determined for each dose duratio n, An E-max model was fitted to the AUEC versus dose duration data. A simil ar analysis was conducted for a bioequivalence study on two formulations of a dermatologic corticosteroid product in 40 healthy subjects. Results: In the exploratory study, the number of individual subject data se ts for which the E-max model provided an acceptable fit generally increased with the potency of the dermatologic corticosteroid product. On the basis of population modeling, dose-response data of all products, except the lowe st potency cream, were adequately described by the E-max model. Values for population ED50 (the dose duration required to achieve 50% of the fitted AU EC(max) value) decreased with increase in dermatologic corticosteroid produ ct potency. Conclusions: Acceptable model fits to all individual subject dose-response data were not achieved for any dermatologic corticosteroid product. However , population dose-responses were adequately described by the E-max model. O n the basis of these data, the optimal dose duration used for comparison of multisource dermatologic corticosteroid products is recommended to be equa l to the ED50 based on population modeling of pilot dose-response study dat a.