Twelve-year follow-up of a randomized trial comparing cisplatin and cyclophosphamide with cisplatin, doxorubicin and cyclophosphamide in patients with advanced epithelial ovarian cancer

From 1982 to 1984, 131 patients with FIGO stage Ic-IV epithelial ovarian cancer were included in a randomized clinical trial comparing cisplatin 50 mg m(-2) plus cyclophosphamide 600 mg m(-2) (PC regimen) with PC plus doxorubicin 45 mg m(-2) (PAC regimen). Chemotherapy was repeated every 4 weeks for six cycles. The criteria for entry, the characteristics of the elegible patients, the response rates and the toxicities have been previously reported. The study was updated in August 1994 with a median follow-up of 10.5 years (range 10-12 years). In the whole series, the median time to progression is 13 months and the 12-year progression-free survival (PFS) is 18%, whereas the median time to survival is 21 months and the 12-year survival is 21%. By log-rank test survival is significantly related to residual disease after first surgery (P<0.0001), ECOG performance status (PS) (P<0.0001), FIGO stage (P=0.0001) and histologic grade (P=0.04), but not to type of chemotherapy and age. By Cox proportional hazard model residual disease (P=0.0004), histologic grade (P=0.01) and ECOG performance status (P=0.049), but not FIGO stage, treatment arm and age, are independent prognostic variables for survival. The survival curves are superimposable in the two treatment arms among patients with residual disease <2 cm, whereas there is a trend in favor of the PAC regimen among patients with larger residual disease. By log-rank test PFS is not significantly related to chemotherapy arm. However, it is worth noting that among patients with residual disease >2 cm 12-year PFS is 12.5% for PAC arm, while all patients of PC arm progressed by the sixth year. Conversely, the PFS curves are superimposable in the two treatment arms among patients with residual disease <2 cm.