OBJECTIVE The feasibility of administering vinorelbine (Navelbine, Burroughs Wellcome Co, Research Triangle Park, NC), a semisynthetic vinca alkaloid with broad activity, as a liquid-filled gelatin capsule was evaluated in a bioavailability (F) and pharmacokinetic study. METHODS Each of 17 cancer patients had pharmacokinetic studies performed after receiving vinorelbine 30 mg/m2 intravenously (IV), which is the maximum-tolerated dose (MTD) for weekly IV administration, and twice after receiving the oral formulation at a nominal dose of 100 mg/m2. Subsequently, these patients and 10 other subjects received the oral formulation at a dose of 100 mg/m2/wk to evaluate the feasibility of chronic oral administration. RESULTS Plasma drug disposition was well described by a triphasic model. Mean central volume of distribution and steady-state volume of distribution (Vss) were large (0.66 +/- 0.46 L/kg and 20.02 +/- 8.55 L/kg, respectively); the mean harmonic terminal half-life (t1/2) was long (18 hours); and the high mean clearance (CI) rate (0.80 +/- 0.68 L/h/kg) approached hepatic blood flow. F was low (0.27 +/- 12), and absorption was rapid (mean time of maximum plasma concentration [Tmax], 0.91 +/- 0.22 hours). Absorption parameters after the first and second oral doses were similar, with mean F values of 0.27 +/- 0.14 and 0.25 +/- 0.11, respectively. Coefficients of variability (CVs) for F, maximum plasma concentration (Cmax), and Tmax were 32%, 42%, and 78%, respectively, indicating moderate intraindividual variability. The pharmacologic profile of this oral formulation indicates that there is a large first-pass effect. Neutropenia was the principal toxicity of oral vinorelbine. Grade 3 or 4 neutropenia occurred in 63% of patients, but only 11% developed neutropenia and infection. Nausea, vomiting, and diarrhea were also common with oral administration, but these effects were rarely severe and could be ameliorated by using a divided-dose schedule and/or prophylactic antiemetic and antidiarrheal agents. The mean nominal oral dose was 82 mg/m2, and the mean percentage of intended dose that was received was 92%. Although dose escalations were permitted for negligible toxicity, doses were not escalated to greater than 100 mg/m2/wk in any patient. Vinorelbine given as a liquid-filled gelatin capsule at 100 mg/m2 provided equivalent pharmacologic exposure as 30 mg/m2 IV. CONCLUSIONS The oral administration of vinorelbine, specifically as a liquid-filled, soft gelatin capsule, is a feasible route of administration. Weekly oral dosing at 100 mg/m2 induces a consistent degree of myelosuppression, but the high frequency of grade 3 or 4 neutropenia, albeit brief and uncomplicated, warrants the recommendation of a slightly lower starting dose, ie, 80 mg/m2/d, for subsequent phase II evaluations, especially in heavily pretreated patients.