The pharmacokinetics of a 1-h paclitaxel infusion.

Purpose: To characterize the disposition of paclitaxel (PAC) after a 1-h infusion in humans and define if possible a pharmacodynamic relationship between PAC disposition and the observed toxicity.

Patients And Methods: PAC pharmacokinetics were studied in 43 courses of therapy in 30 patients (30 first course, 13 PK third course). PAC was administered at 150, 175, 200, 225 and 250 mg/m2 by a 1-h infusion to patients with advanced cancer (lung, breast, ovarian, cervix, and head and neck). PAC was quantified by high-performance liquid chromatography (HPLC). Pharmacokinetic parameters were calculated by noncompartmental and model-dependent methods.

Results: Increases in the area under the curve and the peak plasma concentration were not proportional to increases in the dose. However, the deviation from linearity is rather moderate. The dose-limiting toxicity was central neuropathy which was not associated with pharmacokinetic deviations. Owing to the absence of grade 3 or 4 myelotoxicity, no clear correlation between this toxicity and pharmacokinetic parameters could be established.

Conclusion: Within the evaluated dose range of the 1-h infusion there was only a moderate nonlinear disposition of PAC in humans and therefore a dose of 225 mg/m2 is recommended as safe. The observation of central neuropathy could not be directly related to a pharmacokinetic parameter. The complexity of the formulation which included Cremophor EL and ethanol may offer an explanation for the observed central neurotoxicity.