Background : Taxotere, a semisynthetic compound structurally related to taxol, has a broad spectrum of activity in murine transplantable tumors; in the B16 melanoma model, it caused a total log cell kill 2.5 times greater than that caused by taxol at equitoxic doses. Purpose : We conducted a phase I study of Taxotere ( a ) to determine its qualitative and quantitative toxic effects and a starting dose for phase II trials, ( b ) to investigate its clinical pharmacology, and ( c ) to document its antitumor activity. Methods : Taxotere was given as a 1-hour infusion at a starting dose of 1 mg/m 2 per day for 5 consecutive days. The 5-day course of therapy was repeated every 21 days. Thirty-nine cancer patients with advanced disease were entered in the study; at least three patients were entered at each dose level. Initial dose escalations were planned at 100% increments until biologic activity was observed; subsequent escalations were planned at 50% increments until grade 2 toxicity (the National Cancer Institute's Common Toxicity Criteria) occurred and then at 25% increments until the maximum tolerated dose was established. Results : Thirtynine patients were entered in the study. Successive dose levels used were 1, 4, 8, 16, 12, and 14 mg/m 2 per day. The dose-limiting toxic effects were granulocytopenia and concurrent mucositis. Grade 4 granulocytopenia associated with grade 3 mucositis developed in six of 12 patients treated at a dose of 16 mg/m 2 per day, two of 10 treated at 12 mg/m 2 per day, and two of eight treated at 14 mg/m 2 per day. Because these toxic effects occurred concurrently, all patients so affected developed neutropenic fevers and required hospitalization. Neither cardiac nor neurologic toxic effects were noted. Antitumor activity was observed in six patients with ovarian cancer and in one with breast carcinoma. Although pharmacokinetic parameters were consistent between day 1 and day 5 for individual patients, considerable variation existed among those treated at the same dose level. A relationship was observed between the area under the curve for plasma concentration of drug × time (AUC) on day 1 and the percentage decrease in absolute granulocyte counts. Conclusion : Granulocytopenia associated with oral mucositis is the dose-limiting toxicity of this schedule. We recommend a starting dose of 14 mg/m 2 per day for phase II studies of this 5-day schedule. Dose modifications on days 2–5 based on the day-1 AUC may allow individualized dosing. [J Natl Cancer Inst 84:1781–1788, 1992]