ICON 2 and ICON 3 data in previously untreated ovarian cancer: results to date.

The second International Collaborative Ovarian Neoplasm study (ICON 2), was a large, international randomized study of cyclophosphamide/doxorubicin/cisplatin (CAP) versus single-agent carboplatin in patients with previously untreated ovarian cancer. Patients in the CAP arm received 500 mg/m2 cyclophosphamide, 50 mg/m2 doxorubicin, and 50 mg/m2 cisplatin. Carboplatin was dosed to an area under the concentration-time curve of 5. Chemotherapy was given every 3 weeks for a total of 6 months for each regimen. Results of a 1995 interim analysis in 1,377 patients showed that overall, the total dose received in the CAP group was greater than 75% of the planned dose, with 75% of patients receiving greater than 87%, greater than 80%, and greater than 83% of the planned doses of cisplatin, doxorubicin, and cyclophosphamide, respectively. Of carboplatin patients, 75% received greater than 1,450 mg/m2 (of a median planned dose of 1,800 mg/m2). There were significant differences in grades 3/4 toxicity between the two arms: leukopenia occurred in 34% of CAP patients versus 10% of carboplatin patients, alopecia in 70% versus 3%, nausea and vomiting in 21% versus 9%, and mucositis in 21% versus 0%, respectively. However, thrombocytopenia was more frequent in the carboplatin group (16% v 7%). At the 1996 analysis, 1,526 patients had been entered, 1,498 had been randomized, and 740 had progressed or died. The interim conclusions were that although the more toxic CAP regimen may improve progression-free survival by a small amount, there was no evidence of any survival benefits or difference within the subgroups. Given that a sufficient number of patients had been accrued to ICON 2 and that starting accrual for ICON 3 markedly slowed accrual to ICON 2, the trial was closed. The full analysis of ICON 2 should be available in the summer of 1997. ICON 3 was designed to consider the role of paclitaxel (Taxol; Bristol-Myers Squibb Company, Princeton, NJ) in previously untreated ovarian cancer. Patients received paclitaxel at a dose of 175 mg/m2 (3-hour infusion) in combination with carboplatin dosed to an area under the concentration-time curve of 5 (based on chromium ethylenediamine tetraacetic acid) or 6 (based on calculated creatinine clearance). The control arm was either CAP or carboplatin. Patients were randomized 2:1 in favor of the control arm. In all, 1,070 patients have been entered to date. At the first planned interim analysis, in April 1996 in 434 patients, it was too early to provide efficacy data. The plan was to continue accruing to the trial to a maximum of 2,000 patients, with review in mid-1997, when the events for analysis will be virtually doubled and the data can be interpreted in light of the results of the Intergroup study (European Organization for Research and Treatment of Cancer, the National Cancer Institute of Canada, and the Scottish study) and Gynecologic Oncology Group trial 132.