The results from a quantitative structure-property relationship (QSPR) model was integrated into a fifth-generation pharmacokinetic/pharmacodynamic (PK/PD) model of corticosteroid receptor/gene-mediated effects. The proposed model was developed using previously reported tyrosine aminotransferase (TAT) activity data following a 50 mg/kg intravenous dose of methylprednisolone in male adrenalectomized (ADX) rats. Induced TAT activity is a classical measure of corticosteroid genomic effects and the typical time course shows an initial lag-time, a slow rise to peak response, and a gradual return toward baseline values. The TAT activity profiles were subsequently predicted for two additional steroids (dexamethasone and hydrocortisone), which were confirmed experimentally. Two groups of male ADX Wistar rats (n = 18 each) were given either 0.1 mg/kg dexamethasone or 50 mg/kg hydrocortisone by penile vein injections. Plasma drug concentrations and liver TAT activity were measured at various time points. Baseline TAT activity was significantly lower in this study as compared to previous reports. Model simulations well captured the pharmacodynamic data once initial conditions were corrected for observed baseline values. Additional TAT profiles reported in the literature for prednisolone were also reasonably predicted using the final model. This study serves as a demonstration of how in vitro pharmacologic data and QSPR modeling results may be incorporated into existing mechanistic PK/PD models to anticipate the effects of other chemically related compounds.
Copyright 2003 Wiley-Liss, Inc. and the American Pharmaceutical Association J Pharm Sci 92:881-889, 2003