Ten-year survival after epithelial ovarian cancer is not associated with BRCA mutation status

Highlights

• No residual tumor is the most important predictor of long-term survival.

• BRCA mutation status does not confer a benefit for long-term mortality.

Abstract

Objectives

After a diagnosis of ovarian cancer, positive BRCA mutation status confers a transient mortality benefit that diminishes with time. The majority of women who survive for 10–12 years are effectively cured of their disease. Thus, it is important to estimate the probability of long-term survival by BRCA mutation status and treatment-related factors.

Methods

We included unselected epithelial ovarian cancers diagnosed in Ontario, Canada from 1995 to 1999 and from 2002 to 2004. Clinical information was obtained from medical records. Survival status was determined by linkage to the Ontario Cancer Registry. We estimated the annual mortality for these patients. We compared women who did and did not survive 10 years for a range of factors including BRCA mutation status and extent of residual disease post-surgery.

Results

Of the 1421 patients, 109 (7.7%) had BRCA1 mutations and 68 (4.8%) had BRCA2 mutations. A status of no residual disease was achieved by 39% of non-carriers and 19% of mutation carriers (P < 0.0001). By 10-years of follow-up, 43% of non-carriers, 57% of BRCA1 mutation carriers and 69% of BRCA2 mutation carriers had died from ovarian cancer. Among women with stage III/IV serous cancers and no residual disease, the 10-year actuarial survival was 42% for non-carriers and 29% for mutation carriers (P = 0.40).

Conclusion

The initial survival advantage among women with BRCA mutations may reflect a higher initial sensitivity of BRCA carriers to chemotherapy, but this response does not predict long-term survival. The strongest predictor of long-term survival is status of no residual disease at resection.

Introduction

We have previously reported that among women with ovarian cancer, a short-term but not a long-term survival advantage is associated with carrying inherited BRCA1 or BRCA2 mutations [1]. During the first two years post-diagnosis, the annual mortality rate was less for mutation carriers than for non-carriers. However, this relationship reversed from year three onwards. In this respect, the survival curves comparing carriers and non-carriers are not proportional, and as a consequence, short-term survival rates cannot be used to predict ultimate survival. The great majority of women who survive ovarian cancer for at least 12 years are effectively cured [2]. Thus, the primary goal of treatment should be to maximize the probability of 12-year survival. To this end, the most important predictors of 12-year survival are no residual disease post-resection and the use of intraperitoneal chemotherapy [3]. The majority of patients will respond to intravenous adjuvant chemotherapy after primary debulking surgery in terms of eliminating visible disease, but many of these women have microscopic cancer deposits in the abdomen and will ultimately relapse and die. For these patients, intravenous chemotherapy is associated with very initial high response rates and favorable five-year survival, but will have little impact on long-term survival rates, compared to women who receive no chemotherapy.

Several studies have shown that ovarian cancer in patients with BRCA mutations shows a superior response to platinum-based chemotherapy [4], [5], [6] although whether or not this translates into a long-term survival advantage is unclear. Together, these findings indicate that in order to obtain an accurate assessment of overall survival, studies with long-term follow-up are needed [7]. Factors bearing upon survival include tumor histology and stage, BRCA mutation status, timing and route of chemotherapy as well as the ability to achieve no residual disease following primary debulking surgery [8], [9]. To date, no study has incorporated all of these factors. We therefore sought to explore which factors predict long-term survival (defined as 10 years) in women with and without genetic predisposition in a cohort of 1421 women with unselected epithelial ovarian cancer diagnosed in Ontario, Canada.