Prognosis and conditional disease-free survival among patients with ovarian cancer

Abstract

Purpose: Traditional disease-free survival (DFS) does not reflect changes in prognosis over time. Conditional DFS accounts for elapsed time since achieving remission and may provide more relevant prognostic information for patients and clinicians. This study aimed to estimate conditional DFS among patients with ovarian cancer and to evaluate the impact of patient characteristics.

Patients and methods: Patients were recruited as part of the Hormones and Ovarian Cancer Prediction case-control study and were included in the current study if they had achieved remission after a diagnosis of cancer of the ovary, fallopian tube, or peritoneum (N = 404). Demographic and lifestyle information was collected at enrollment; disease, treatment, and outcome information was abstracted from medical records. DFS was calculated using the Kaplan-Meier method. Conditional DFS estimates were computed using cumulative DFS estimates.

Results: Median DFS was 2.54 years (range, 0.03-9.96 years) and 3-year DFS was 48.2%. The probability of surviving an additional 3 years without recurrence, conditioned on having already survived 1, 2, 3, 4, and 5 years after remission, was 63.8%, 80.5%, 90.4%, 97.0%, and 97.7%, respectively. Initial differences in 3-year DFS at time of remission between age, stage, histology, and grade groups decreased over time.

Conclusion: DFS estimates for patients with ovarian cancer improved dramatically over time, in particular among those with poorer initial prognoses. Conditional DFS is a more relevant measure of prognosis for patients with ovarian cancer who have already achieved a period of remission, and time elapsed since remission should be taken into account when making follow-up care decisions.

Fig 1.

Three–year conditional disease-free survival estimates. Number of patients still in remission at particular time point.

Fig 2.

Three-year conditional disease-free estimates stratified by age at remission (A), stage (B), histology (C), and grade (D). Number of patients still in remission at particular time point. diff., differentiated.

Fig A1.

Traditional disease-free survival curves (Kaplan-Meier curves), stratified by age at remission (A; log-rank test, P = .01) and stage (B; log-rank test, P < .01).