Since the discovery of BRCA1 and BRCA2, investigators have sought to determine whether the presence of a germline mutation independently influences the outcome of a breast cancer diagnosed in a woman with an inherited mutation. The question is highly relevant to an unaffected woman with a mutation, as knowledge of poor prognosis in mutation-associated cancer could influence her decision whether to pursue enhanced breast cancer surveillance—eg, with breast magnetic resonance imaging (MRI)—or risk-reducing mastectomy. As important as the question is, there are a number of obstacles to arriving at a definitive answer.
Prognostic Significance of Mutations
BRCA1/2 mutations are uncommon, so study sample sizes tend to be small, magnifying the potential effect of confounding influences. Women do not routinely undergo genetic testing at the time of diagnosis, so there is always the potential for ascertainment bias if the decision to offer genetic testing is influenced by clinical factors that also influence outcome. Examples of such factors could include a young age at diagnosis or the presence of triple-negative breast cancer. Delays in testing until sometime after diagnosis may also confound the interpretation of studies evaluating the prognostic significance of mutations, as women surviving to be tested may not be entirely representative of the entire population of women with mutation-associated cancer. Treatment in retrospective studies of outcome is usually not uniform, which may influence the association between germline status and survival in unpredictable ways. Lastly, the outcomes of carriers in retrospective studies may not necessarily be reflective of outcomes in women who are undergoing aggressive MRI-based surveillance.
Over the years, a number of investigators have sought to overcome these challenges. Many of these studies (but not all) have relied upon the presence of founder mutations in specific populations to expand the sample size of women with mutations. Early studies employed retrospective cohort designs, identifying mutation carriers through testing of archived material from women of Ashkenazi ancestry.
These small but relatively unbiased studies suggested that BRCA1 mutations are associated with worse outcomes, largely in women who did not receive adjuvant chemotherapy. However, a larger study of incident breast cancer in Israel identified 110 Ashkenazi carriers over a period of 2 years and found no significant effect of germline BRCA status on outcome. Case ascertainment was incomplete, unfortunately, and treatment was not controlled, raising the possibility of a confounded result.
In 2012, investigators reported the survival experience of mutation carriers in the Breast Cancer Family Registry, an international population-based cohort ascertainment enriched for women with a family history of breast cancer. In this study of 165 women with mutations, germline status had no impact on survival after adjustment for clinical prognostic factors such as stage.
Confirmation of Previous Studies
The study by Huzarski and colleagues reviewed in this issue of The ASCO Post confirms and expands upon the results of the previous studies. This study explores the impact of BRCA1 mutations on prognosis in Poland, where most mutation carriers have one of a limited number of founder mutations. Although this was not, strictly speaking, an unselected ascertainment (as women had to be offered and accept testing), the number of women tested was large and the number of BRCA1 carriers identified (n = 233) was substantially greater than in any previous study.
In keeping with the results of previous studies, the survival of BRCA1 carriers was not significantly worse than the outcome of noncarriers in univariate analysis, despite the well-recognized association of mutations with triple-negative disease. Cancers in women with mutations were not more advanced than in noncarriers, and the 10-year survival of BRCA1 carriers with T1N0 tumors was 89.9%. Interestingly, the survival of women with smaller cancers (< 1 cm) was no better than that of women with 1.1- to 2-cm primaries, although the numbers of cases in each group were relatively small and the power to detect a difference limited.
These overall results should be reassuring to women with mutations, but there are some caveats. Most of the carriers received chemotherapy, which may have mitigated an adverse effect of mutation status. Also, in multivariate analysis, the presence of a mutation was associated with a significantly worse overall survival (hazard ratio = 1.81, 95% confidence interval = 1.26–2.61) if it was forced into the proportional hazards model despite being nonsignificant in the univariate analysis.
The authors were not able to determine whether the deaths were clearly associated with the index breast cancer or with metachronous breast or ovarian cancers, which were not uncommon in women who did not undergo oophorectomy. The significant impact of oophorectomy on survival in women with mutations suggests a substantial force of mortality from metachronous cancers, as the procedure would be expected to reduce the risk of these events without having a significant effect on the outcomes of a primary hormone receptor-negative breast cancer.
Taken together, the studies of prognosis in BRCA mutation-associated breast cancer are quite consistent. With the exception of the Breast Cancer Family Registry report, no study has indicated an effect of BRCA2 mutations on survival. Breast cancer-related survival also appears to be similar in BRCA1 mutation carriers and non-carriers, despite the association between BRCA1 mutations and the more dangerous triple-negative subtype. A working hypothesis is that the aggressive use of adjuvant chemotherapy in young women, especially with receptor-negative disease, may be ameliorating the negative prognostic impact of the breast cancer subtype. This hypothesis is difficult to prove with the available study designs, and a randomized biomarker-driven study to address the question is neither feasible nor ethical.
Second malignancies, especially ovarian cancer, pose a significant threat to women with mutations, and appropriate use of risk-reducing salpingo-oophorectomy consistently improves survival both by reducing the risk of ovarian and fallopian tube cancers and probably by reducing the risk of contralateral breast cancers in women who choose not to undergo risk-reducing mastectomies. Although these conclusions are somewhat reassuring, a 10% mortality rate in young women with small, node-negative tumors remains unacceptably high. One hopes that this rate can be reduced by improvements in systemic adjuvant therapies, possibly including the use of platinum-based regimens or inhibitors of poly (ADP-ribose) polymerase. Studies to assess the impact of these approaches are in development and are urgently needed. ■
Mark Robson, MD, is with the Clinical Genetics and Breast Cancer Medicine Services, Memorial Sloan-Kettering Cancer Center, Weill Cornell Medical College, New York, New York.
Disclosure: Dr. Robson reported no potential conflicts of interest.