Preclinical pharmacokinetics and oral bioavailability of BMS-310705, a novel epothilone B analog.

Purpose: BMS-310705, a novel semisynthetic derivative of epothilone B, is a tubulin-polymerization agent currently in phase I clinical trials for anticancer therapy. The in vitro and in vivo pharmacokinetics and oral bioavailability of BMS-310705 were investigated in mice, rats, and dogs. In addition, comparison of the pharmacokinetics of BMS-310705 using various formulations was conducted in rats.

Methods: The permeability of BMS-310705 was evaluated in Caco-2 cells, an in vitro model of the human intestinal epithelium. Human liver microsomes were used to determine the cytochrome P450 enzymes involved in the metabolism of BMS-310705. Plasma protein binding of BMS-310705 was determined in mouse, rat, dog, and humans. BMS-310705 was administered to female nude mice as single doses of 5 mg/kg intravenously or 15 mg/kg orally. Male Sprague-Dawley rats were treated with single doses of BMS-310705 either intraarterially (2 mg/kg) or orally (8 mg/kg). The effect of Cremophor on the pharmacokinetics of BMS-310705 was evaluated in rats using various formulations with and without Cremophor. Male dogs were treated with 0.5 mg/kg intravenously or 1 mg/kg orally in a crossover study design.

Results: Systemic clearance of BMS-310705 was high in mice (152 ml/min/kg), rats (39 ml/min/kg), and dogs (25.7 ml/min/kg). The volume of distribution (Vss) in mice, rats, and dogs was 38, 54, and 4.7 l/kg, respectively, and greater than total body water. BMS-310705 showed moderate binding to plasma proteins in all four species tested. The clearance in humans may be intermediate to high based on both allometric scaling using parameters obtained from three species, and in vitro human liver microsomal stability data. In rats, the presence of Cremophor in the formulation resulted in a significant increase in exposure compared to buffered vehicles not containing Cremophor. Inhibition of p-glycoprotein and/or CYP3A4 by Cremophor may be responsible for this phenomenon, and studies in Caco-2 cells and human liver microsomes suggested that BMS-310705 may be a substrate for both p-glycoprotein and CYP3A4. The oral bioavailability of BMS-310705 in pH buffered formulations was 21% in mice, 34% in rats and 40% in dogs.

Conclusion: In summary, BMS-310705 is cleared rapidly and distributes extensively in mice, rats, and dogs. The presence of Cremophor in the formulation could significantly increase exposure in rats, possibly due to interactions with p-glycoprotein and/or CYP3A4. Oral bioavailability using formulations not containing Cremophor were found to be adequate, suggesting potential for development of BMS-310705 as an oral anticancer drug.