Long-term results of a randomized trial comparing cisplatin with cisplatin and cyclophosphamide with cisplatin, cyclophosphamide, and adriamycin in advanced ovarian cancer. GICOG (Gruppo Interregionale Cooperativo Oncologico Ginecologia), Italy.

We report the long-term results of a randomized trial comparing cisplatin (P) with cisplatin and cyclophosphamide (CP) with cisplatin, cyclophosphamide, and adriamycin (CAP) in advanced ovarian cancer. Overall, this update confirms previously published data on 529 cases. Median survival times for the three treatments--CAP, CP, and P--are, respectively, 23, 20, and 19 months. The differences among the three arms are still nonsignificant and the estimated percentage survival at 7 years and confidence limits are, respectively, 21.7 (14.9-28.4), 17.0 (11.0-22.9), and 12.2 (6.9-17.4). According to the results of the Cox regression model on prognostic factors, higher grading, a larger residual tumor size, and performance status less than 80 (Karnovsky) all were independently associated with a poorer outcome, while a serous histotype was related to a better prognosis. The other variables (age, stage, center, type of surgery) initially included in the model did not appear to be significantly related to prognosis. The implications of these long-terms results relative to the application of combination chemotherapy with CAP or CP are discussed.