Purpose: The paclitaxel vehicle Cremophor EL (CrEL) profoundly influences the cellular distribution of paclitaxel in human blood in vitro by a concentration-dependent decrease of the unbound drug fraction. Because CrEL clearance increases by extending the infusion duration from 3 to 24 hours, we hypothesized that exposure to unbound paclitaxel might also be schedule-dependent.
Patients and methods: CrEL and unbound paclitaxel pharmacokinetics were prospectively analyzed in 29 patients with advanced solid tumors treated with paclitaxel 100 mg/m(2) given as a 1-hour (n = 15) or 3-hour (n = 14) intravenous infusion.
Results: The systemic exposure (area under the curve [AUC]) to CrEL was significantly higher with the 1-hour as compared with the 3-hour schedule (80.2 +/- 24.2 v. 48.5 +/- 24.1 microL x h/mL; P =.002). In contrast, the AUC of unbound paclitaxel was substantially reduced after the 1-hour infusion (0.50 +/- 0.10 v. 0.62 +/- 0.12 micromol/L x h; P =.009). Similarly, clearance and volume of distribution were significantly dependent on infusion duration (P <.005). A trend was observed toward more severe hematologic toxicity with the 3-hour schedule (P =.053), consistent with increased exposure to unbound drug.
Conclusion: Overall, these findings explain, at least in part, previous observations that short-infusion schedules of paclitaxel lack significant myelotoxicity, whereas potentially CrEL-related side effects, including peripheral neuropathy, are augmented.