A randomized phase II trial in patients with carcinoma of an unknown primary site

BACKGROUND. Current therapy for patients with carcinoma of an unknown primary site (CUP) is inadequate. To develop less toxic and more effective therapies for patients with CUP, a multicenter, randomized, Phase II study was conducted. Patients with CUP received either carboplatin and etoposide (CE) or a combination of paclitaxel, 5-fluorouracil, and leucovorin (TFL). METHODS. Patients randomized to Arm A received paclitaxel, 175 mg/m², intravenously over 3 hours on Day 1 followed by leucovorin, 300 mg, over 30-60 minutes and 5-fluorouracil, 350 mg/m², both intravenously on Days 1-3. Patients randomized to Arm B received etoposide, 100 mg/m², intravenously on Days 1-3 and carboplatin at an area under the curve of 6 on Day 1 only. The cycles in both treatment arms were repeated every 28 days. Patients were followed for tumor response, survival, and toxicity. RESULTS. Thirty-four patients were enrolled, 32 of whom were evaluable for response. An identical overall response rate of 19% (95% confidence interval, 4-45%) was noted in each treatment arm. The median survival for the entire study population was 194 days. The median survivals observed in Arm A and Arm B were 251 days and 194 days, respectively (P = 0.91 [difference not significant]). Hematologic toxicity on Arm B was considerable with 29% of the patients developing neutropenia and fever. Toxicity on Arm A was modest. CONCLUSIONS. In this randomized Phase II trial, CE and TFL appeared to have modest activity in CUP patients, with response rates similar to those reported with previously described chemotherapy regimens. Toxicity with CE was more severe than expected, although TFL was found to be well tolerated.