In 1981, approximately 17,000 new cases of ovarian carcinoma will be diagnosed and 11,200 of these women will die (approximately 6% of all female cancer deaths).’ Roughly one woman in 70 will develop this disease during her lifetime and only 25% will be diagnosed at a time when the disease is confined to the ovary. Furthermore, even in those patients with localized disease who have apparent complete resections or are rendered disease-free by pelvic irradiation, over one-third will eventually recur. As a result of the failure of local modalities to control disease in the vast majority of patients with ovarian adenocarcinoma, systemic chemotherapy has assumed a major role in disease management. Chemotherapy was originally used as second line treatment for radiation therapy failures, but is now commonly used as initial treatment for patients with advanced disease as well as an adjuvant to surgery for early stage disease (reviewed elsewhere in this series). In recent years, single alkylating agent chemotherapy has been frequently used for patients with bulky residual disease. Response rates of 33% to 65% can be anticipated with this form of therapy. Median survivals for all treated patients range from 1 O-l 4 months with median survivals of 17-22 months in those responding to treatment. Overall five year survivals range from O-9% with 16-22% of responders alive at five years.” Melphalan has been studied most thoroughly and is the standard single alkylating agent to which new drugs and combinations should be compared. Several non-alkylating agents also have significant activity in ovarian cancer. Adriamycin, hexamethylmelamine, and cis-platinum are of particular interest.9 In one prospective randomized trial, adriamycin, hexamethylmelamine and L-PAM achieved response rates of 29%, 33%, and 30%, respectively. In addition, hexamethylmelamine and cis-platinum are the only drugs thus far that have been useful as second line agents in patients failing initial therapy with alkylating agents. A summary of some of the active single agents is provided in Table 1. The relatively high response rates to various single agents has provided the impetus for today’s interest in combination chemotherapy. Combination chemotherapy has produced overall response rates which vary from 22-91% with complete remission rates of lo-69%. Unfortunately, many published reports are single arm studies of previously treated patients. Few combinations have been studied in randomized trials where differences in patient age, performance status, histologic grade, and amount of post-operative residual disease have been considered in assessing response rates. In many studies, it is likely that patient characteristics have a more important influence on survival than the chemotherapy regimen being tested. In addition, complete clinical responses are not always surgically confirmed and long term follow-up is often limited. Despite the problems in interpreting much of the published literature, there are several useful studies which show promise. Several combination chemotherapy studies appear to significantly increase response rates over those achieved with single agents. Successful prospective comparisons with single alkylating agents have employed the HexaCAF regimen (hexamethylmelamine, cyclophosphamide, methotrexate and 5fluorouracil);” AC (adriamycincyclophosphamide);6 and CHF (cyclophosphamide, hexamethylmelamine and 5-fluorouracil),4 and subsequently in a COG prospective trial randomizing non-optimal Stage III patients to receive AC, alkeran-h.examethylmelamine, or alkeran alone.8 Data from the Hexa-CAF study show a higher overall response rate (76% versus 54%), more complete responses (33% versus 16%), and a longer overall median survival (29 months versus 17 months). Recently cis-platinum has been included in a number of combination chemotherapy programs. These combinations have not been in use long enough to fully evaluate their effect on survival in comparison with currently established combination treatments or single agents. The