This report presents final survival survey results from a previously reported study using oregovomab immunotherapy in patients with advanced ovarian epithelial cancer. Follow-up surveys to 5 years from randomization were collected for the cohort of stage III/IV ovarian cancer patients achieving initial remission who received subsequent maintenance immunotherapy with oregovomab or placebo. The relationship of time-to-relapse, survival postrelapse, and overall survival was analyzed. One hundred forty-five patients in the intent-to-treat population and the hypothesis generating subset of 67 patients (debulked to < or =2 cm, CA125 < or =65 U/mL before cycle 3, normal CA125 and no evidence of disease postchemotherapy) previously reported were evaluated for long-term outcomes. Patterns of relapse and survival were consistent in both groups for the intent-to-treat population. Median survival time was 57.5 months for oregovomab and 48.6 months for placebo with an adjusted hazard ratio of 0.72 (95% confidence interval, 0.41-1.25). Median survival has not been reached in the hypothesis generating subset of patients receiving oregovomab. Cox multivariate regression analysis identified velocity of CA125 rise at relapse to be a highly statistically significant predictor of postrelapse outcome (P = 0.006). Although time-to-relapse may be a useful surrogate of survival in ovarian cancer immunotherapy studies, 5 years of follow-up has proved insufficient to permit a definitive survival analysis and it has been extended in ongoing phase III studies of oregovomab. Velocity of CA125 rise at relapse is a highly significant predictor of survival after relapse.