Nonlinear pharmacokinetics and metabolism of paclitaxel and its pharmacokinetic/pharmacodynamic relationships in humans.

Abstract

PURPOSE: To characterize and model the disposition of paclitaxel in humans and define a pharmacodynamic relationships between paclitaxel disposition and its toxicity and efficacy. PATIENTS AND METHODS: Paclitaxel pharmacokinetics were studied in 55 courses of therapy in 30 patients. Paclitaxel was administered at 135 mg/m2 or 175 mg/m2 by either a 3- or a 24-hour infusion schedule to patients with advanced ovarian cancer (n = 15), or at 225 mg/m2 by 3-hour infusion to patients with advanced breast cancer (n = 15). Paclitaxel and 6 alpha-hydroxylpaclitaxel were quantified by high-performance liquid chromatography (HPLC). Pharmacokinetics were assessed by noncompartmental and model-dependent methods. Pharmacodynamic correlations were evaluated statistically and by regression models. RESULTS: Paclitaxel disposition is nonlinear in humans and, on the 3-hour schedule, 6 alpha-hydroxylpaclitaxel was identified in the plasma of all patients treated. The plasma disposition of paclitaxel and 6 alpha-hydroxylpaclitaxel was well described by a model that featured multiple nonlinear processes. Neutropenia was not related to the areas under the curves (AUCs) of paclitaxel or 6 alpha-hydroxylpaclitaxel, or to palitaxel peak concentrations (Cmax). Neutropenia was related to the duration that plasma concentrations were > or = 0.05 mumol/L, a relationship that is well described by a sigmoid maximum response (Emax) model. CONCLUSION: The disposition of paclitaxel in humans is nonlinear. Paclitaxel metabolism to 6 alpha-hydroxylpaclitaxel is likely an important detoxification pathway. Myelosuppression is related to the duration that plasma paclitaxel concentrations are > or = 0.05 mumol/L. Trials of new doses and schedules of paclitaxel should take into account its nonlinear disposition to rule out adverse clinical consequences, especially if the drug is administered by short infusion. Our pharmacokinetic model should prove to be a powerful tool in predicting paclitaxel disposition, regardless of dose and schedule, and should facilitate further pharmacodynamic investigations.