Measurement of cremophor EL following taxol: plasma levels sufficient to reverse drug exclusion mediated by the multidrug-resistant phenotype.

Abstract

Paclitaxel (Taxol) is the first of a new class of cytotoxic agents with activity against tumors resistant to other drugs. For clinical use, paclitaxel is currently formulated in a vehicle of 50% ethanol and 50% polyethoxylated surfactant Cremophor EL (Cremophor). We have previously shown that Cremophor will block the P-glycoprotein drug efflux pump responsible for the multidrug-resistant phenotype. Overexpression of P-glycoprotein is one mechanism of in vitro resistance to a number of currently used cytotoxic agents including paclitaxel. Our aim was to develop a bioassay to measure plasma levels of Cremophor and to determine whether or not plasma levels of Cremophor achieved during paclitaxel therapy are sufficient to inhibit the activity of the P-glycoprotein. All patients studied had histologically proven, advanced ovarian carcinoma with measurable or evaluable disease and had received at least one prior platinum-containing regimen. The bioassay used flow cytometry to measure the increase in equilibrium intracellular daunorubicin levels in multidrug-resistant human T-cell leukemia cells (CEM/VLB100) in the presence of a series of concentrations of Cremophor. Levels of Cremophor were measured in plasma from 21 patients after a 3-hour infusion of 135 or 175 mg/m2 paclitaxel. Both dose levels were given following premedication with oral dexamethasone, intravenous promethazine hydrochloride, and intravenous cimetidine. The Cremophor bioassay involved incubation of CEM/VLB100 cells (5 x 10(5)) for 1 hour with 2 micrograms/mL daunorubicin in 0.5 mL HL-1 medium plus 0.5 mL plasma prior to flow cytometric analysis. Pretreatment plasma was used to derive a standard curve for the effect of Cremophor on equilibrium daunorubicin levels. All measurements were done in triplicate. In vitro experiments indicated that, for maximal inhibition of P-glycoprotein activity, concentrations of Cremophor of 0.1% (vol/vol) were required. At the end of a 3-hour infusion of paclitaxel, plasma levels of Cremophor in 19 of 21 patients were 0.1% or higher and 0.09% in the remaining two. Concentrations of 5-20 microM paclitaxel dissolved in ethanol without Cremophor did not inhibit P-glycoprotein in this assay. The concentrations of Cremophor measured in plasma drawn from patients after a 3-hour infusion of paclitaxel at 135 or 175 mg/m2 were found to be sufficient to inhibit P-glycoprotein activity in vitro. The efficacy of paclitaxel against some tumors may be aided by its administration in a vehicle solution containing Cremophor in quantities that reach concentrations in the plasma sufficient to reverse multidrug resistance of neoplastic cells.