Long-term survival and prognostic factors in advanced epithelial ovarian cancer with special emphasis upon the effects of protocol inclusion.

Abstract

The Danish Ovarian Cancer Study Group registered 722 patients in stages III and IV during the period 1981-1986. The material included 85% of all ovarian cancer patients in the catchment area of the group and patients allocated to protocol as well as patients treated outside protocols. Five and 10-year survival were: stage III 17%, and 8%, respectively; and stage IV 4% and 2%. Patients allocated to protocol had a significantly better survival than patients not included in protocols even when only patients younger than 70 years were compared. All non-protocol patients had a poorer prognosis irrespective of the reason for exclusion. Five-year survival for stage III protocol patients was 25% vs. 9%, for non-protocol patients younger than 70 years. The 10-year survival was 11% and 4% for stage III protocol and non-protocol patients, respectively. A multivariate analysis showed that residual tumor, age, stage, and performance status had prognostic value. In non protocol patients histologic grade had an additional marginal prognostic impact. In conclusion the study showed that the statement that long-term survival in advanced ovarian cancer has been increased could not be proven by comparison of survival from randomized studies performed in the early eighties with survival of stage III and IV patients before the introduction of cisplatinum chemotherapy. It is necessary to consider survival of all patients, protocol and non-protocol in a geographically well-defined region for evaluation of survival improvement.