Gangi A, Cass I, Paik D, Barmparas G, Karlan B, Dang C, Li A, Walsh C, Rimel BJ, Amersi FF. Breast Cancer Following Ovarian Cancer in BRCA Mutation Carriers. JAMA Surg. 2014;149(12):1306-1313. doi:10.1001/jamasurg.2014.1081
BRCA mutation carriers are at increased risk of developing breast cancer. However, the incidence of breast cancer after a diagnosis of epithelial ovarian cancer (EOC), one of the tubal/peritoneal cancers collectively referred to as pelvic serous carcinomas, is not well known. Optimal breast cancer surveillance and detection for these patients have also not been well characterized.
To determine the incidence of breast cancer after a diagnosis of EOC and to evaluate the need for breast cancer surveillance for these patients.
Design, Setting, and Participants
A retrospective database review of 364 patients who underwent BRCA mutation testing for EOC (stages I-IV) between 1998 and 2012 at an academic medical center with gynecologic and breast cancer centers.
Main Outcomes and Measures
Incidence of breast cancer and methods of surveillance.
Of 364 patients, 135 (37.1%) were found to carry a germline BRCA1 or BRCA2 mutation. The mean age of patients at diagnosis of EOC was 49.5 years (range, 28-89 years). Of the 135 patients, 12 (8.9%) developed breast cancer. The median time from diagnosis of EOC to diagnosis of breast cancer was 50.5 months. Annual mammography was performed for 80 patients (59.3%), with annual magnetic resonance imaging of the breasts performed for 60 patients (44.4%). Thirteen patients (9.6%) underwent a bilateral prophylactic mastectomy at a median of 23 months following EOC diagnosis. Breast cancer was most commonly diagnosed by mammography for 7 of the 12 patients (58.3%), 3 (25.0%) of whom had a palpable mass and 2 (16.7%) of whom had incidental breast cancer detected during a prophylactic mastectomy. Seven patients with breast cancer (58.3%) underwent a bilateral mastectomy. All patients had early-stage breast cancer (stages 0-II). Four patients (33.3%) received adjuvant chemotherapy. At a median follow-up of 6.3 years, 4 of the 12 patients (33.3%) died of recurrent EOC after a diagnosis of breast cancer. The overall 10-year survival rate for the entire cohort of 135 patients was 17.0%.
Conclusions and Relevance
The risk of metachronous breast cancer is low in patients with known BRCA mutations and EOC. A majority of these cases of breast cancer at an early stage are detected by use of mammography. Despite the small number of patients in our study, these results suggest that optimal breast cancer surveillance for patients with BRCA-associated EOC needs to be reevaluated given the low incidence of breast cancer among these high-risk patients. Confirmation of our findings from larger studies seems to be indicated.
The cancer risk among BRCA1 and BRCA2 mutation carriers for breast and ovarian cancer is higher than the general population.1 Although the exact magnitude of the risk of developing breast and/or ovarian cancer for a BRCA1/2 mutation carrier remains unknown, it remains clear that these patients face a substantially elevated risk that persists throughout their lifetime. The largest series reported by Antoniou et al2 included 22 studies of 8139 patients, with 510 BRCA mutation carriers. They report an average cumulative risk among BRCA1 carriers of 65% (range, 44%-78%) for breast cancer and 39% (range, 18%-54%) for ovarian cancer by 70 years of age.2 Similarly, the average cumulative risk among BRCA2 carriers was 45% (range, 31%-56%) for breast cancer and 11% (range, 2.4%-19%) for ovarian cancer by 70 years of age. Other studies3,4 have reported an estimated risk of ovarian cancer of 36% to 63% for BRCA1 mutation carriers and 10% to 27% for BRCA2 mutation carriers. The risk of breast cancer has been reported to be between 50% and 75% for BRCA1 carriers and between 33% and 54% for BRCA2 carriers.5- 7
Over the past several years, there have been many studies regarding the benefits of various screening and prevention options for BRCA carriers. The effectiveness of prophylactic bilateral salpingo-oopherectomy and bilateral mastectomy for patients with identified BRCA mutations has been well described.8- 10 In addition, data from van der Kolk et al11 suggest the need for continued screening of these high-risk patients, which also suggests that the incidences of breast cancer and ovarian cancer remain high well into older age. Current breast cancer screening guidelines include annual mammography and annual magnetic resonance imaging (MRI) of the breasts; in addition, semiannual clinician-performed breast examinations are recommended. Data from the National Surgical Adjuvant Breast and Bowel Project (NSABP) P-1 Study12 and the NSABP Study of Tamoxifen and Raloxifene P-2 trial13 demonstrated that the use of selective estrogen receptor modulators reduced the incidence of breast cancer among high-risk patients.
A significant concern for BRCA1/2 mutation carriers who received a diagnosis of epithelial ovarian cancer (EOC), one of the tubal/peritoneal cancers collectively referred to as pelvic serous carcinoma, is the risk of developing breast cancer. Studies14,15 have shown that factors that can increase the risk of subsequent breast cancer after ovarian cancer among patients with BRCA1/2 include diagnosis at a younger age and a family history of breast cancer. There are few studies evaluating the risk of breast cancer in BRCA carriers after a diagnosis of EOC. Guidelines regarding the optimal surveillance for this subpopulation of high-risk patients are lacking. The purpose of this study is to determine the incidence of breast cancer after a diagnosis of EOC and to determine whether aggressive breast cancer imaging surveillance for these patients with known EOC is warranted.
Upon signing admission paperwork at the time of surgery, patients consented to participating in retrospective research studies. After approval of the Cedars-Sinai institutional review board, we reviewed our prospectively maintained database to identify all consecutive patients who were evaluated and treated for EOC (stages I-IV) between January 1, 1998, and December 31, 2012. All patients who were found to have either a BRCA1 or BRCA2 mutation with a diagnosis of EOC (which includes ovarian, fallopian tube, and peritoneal cancers) were included. Patients who underwent a risk-reducing mastectomy before their EOC diagnosis, those who had another type of malignant tumor at the time of their EOC diagnosis, patients with a history of breast cancer prior to their EOC diagnosis, and those patients who had insufficient data with regard to patient or tumor characteristics or who had a lack of follow-up were excluded from the study. Medical records, including pathology reports, were then evaluated for information about patient, tumor, and treatment characteristics with regard to the diagnosis of EOC. Information extracted from the medical records included date of EOC diagnosis, dates of recurrence, date of BRCA mutation testing, date of detection of second malignant tumor, date of last follow-up, current vital status (living or deceased), and cause of death. In addition, information about family history of breast cancer, breast imaging surveillance, and consultation with a breast surgeon, as well as stage of cancer, mode of detection, type of treatment, and follow-up data for all patients diagnosed with breast cancer after EOC, was collected. Screening modalities included both film-screen and digital mammography. Some patients also underwent MRI as a screening modality, most frequently in addition to the mammography. We also performed a subanalysis of overall survival and breast cancer–free survival by BRCA mutation status based on date of last follow-up. Follow-up data were obtained by a retrospective review of both inpatient and outpatient medical records.
The primary outcome measures were development of breast cancer and methods of surveillance. Secondary outcome measures were time to recurrence or disease-free survival and overall survival for both ovarian cancer–specific and breast cancer–specific survival. Time to recurrence was calculated from the date of diagnosis to the date of recurrence of pelvic serous cancer. Survival was calculated from the date of diagnosis to the date of death or to the most recent follow-up for nondeceased patients. Univariate survival analysis was performed using the Kaplan-Meier method and a log-rank test. Multivariate survival analysis was performed using the Cox proportional hazards model. All statistical analysis was conducted using SAS statistical software version 9.2 (SAS Institute Inc), with P < .05 considered to be statistically significant.
Over a 14-year period, 364 patients were identified who underwent BRCA testing for EOC (stages I-IV). Of the 364 patients, 135 (37.1%) were found to carry a germline BRCA1 or BRCA2 mutation. Of these 135 patients, 103 (76.3%) had a BRCA1 mutation, and 32 (23.7%) had a BRCA2 mutation. As shown in Table 1, a majority of the patients had ovarian cancer. The mean (SD) age at the time of diagnosis of EOC was 51.9 (11.1) years. A majority of the patients with EOC had stage IIIC ovarian cancer. Almost all patients received chemotherapy for their gynecologic cancer.
In the entire cohort, 12 patients (8.9%) developed breast cancer. The mean (SD) age of patients at the time of diagnosis of breast cancer was 59.5 (10.9) years. The median time from a diagnosis of EOC to a diagnosis of breast cancer was 50.5 months (mean [SD] time, 68.8 [64.5] months). Of the 115 patients with EOC who had available data on family history, 63 (54.8%) had a first-degree relative with breast cancer, and 8 of the 12 patients (66.7%) who developed breast cancer also had a first-degree relative with breast cancer. Annual mammography was used for surveillance for 91 of 117 patients (77.8%); however, the use of mammography was suspended for patients with EOC who had a disease-free interval of less than 12 months. As expected, because patients with breast cancer underwent a more rigorous follow-up at shorter intervals, they had more mammograms (mean [SD] number of mammograms, 5.5 [3.9]) during the study period than those patients who did develop breast cancer (mean [SD] number of mammograms, 3.1 [2.6]) (P = .03). Annual MRI of the breasts was performed for 60 of 115 patients (52.2%) from the entire cohort and for 7 of 12 patients (58.3%) in the group who developed breast cancer. There was no significant difference between number of MRI scans between those patients who developed breast cancer and those who did not. Thirteen of 135 patients (9.6%) underwent a bilateral prophylactic mastectomy at a median of 23 months following EOC diagnosis. The median time of follow-up was 5.3 years. As shown in Table 2, among those patients who developed breast cancer, tumors were most commonly diagnosed with the use of mammography for 7 of the 12 patients (58.3%). Two of the 12 patients (16.7%) had a palpable mass detected during a clinical examination, and 3 of the 12 patients (25.0%) had incidental breast cancer detected on pathologic specimens obtained during a prophylactic mastectomy. Six of the 12 patients with breast cancer (50.0%) underwent a mastectomy with a contralateral prophylactic mastectomy, whereas the other 6 patients (50.0%) underwent a lumpectomy. All patients received a diagnosis of early-stage breast cancer (stages 0-II). Eight of 12 patients (66.7%) had invasive ductal carcinoma, with 5 of 8 patients (62.5%) with estrogen receptor–negative, progesterone receptor–negative, HER-2/neu–negative breast cancer. Four of 12 patients (33.3%) received adjuvant chemotherapy for their breast cancer. At a median follow-up of 63 months (mean [SD] follow-up, 79.1 [60.3]), 55 of 135 patients (40.7%) had died of recurrent EOC. Of the 12 patients who developed breast cancer after EOC, 3 (25.0%) had died of recurrent EOC, and 1 (0.8%) had died of metastatic breast cancer. The mean (SD) overall survival was significantly higher for the breast cancer group (14.6 [6.3] years) than for the EOC group (6.5 [5.2] years) (P < .002) (Figure 1, Table 3, and Table 4). Overall survival was higher for women who had metachronous breast cancer after EOC than for women who had only EOC. This was not statistically significant when compared with the EOC group that did not develop breast cancer (P = .76).
Sixty-eight percent of patients within the study showed evidence of recurrence. Of these, 50.0% developed breast cancer after EOC, whereas 69.9% did not (P = .20) (Figure 2, Table 3, and Table 4). The 10-year overall survival rate was 17.0% for the entire study population, 50.0% for those patients who developed breast cancer, and 13.8% for those patients who did not develop breast cancer (P = .006). The mean disease-free survival was 2.7 years for the entire study population. There was no significant difference in disease-free survival between those patients who developed breast cancer and those who did not.
According to the most recent estimates, the cumulative risk of breast cancer by 70 years of age for BRCA1 and BRCA2 mutation carriers was 57% (95% CI, 47%-66%) and 49% (95% CI, 40%-57%), respectively.16 For these patients, a prophylactic oophorectomy has been shown to reduce the risk of breast cancer by 50%.8,17,18 Rebbek et al9 reported on 241 BRCA mutation–positive patients, 99 of whom underwent a prophylactic bilateral oophorectomy and 142 of whom were matched controls who did not undergo a prophylactic oophorectomy. With a median follow-up of 8 years, BRCA-positive patients who underwent a prophylactic bilateral oophorectomy had a 96% reduced the risk of ovarian cancer and a 53% reduced risk of breast cancer.8 Olopade and Artioli8 advocated the use of a bilateral oophorectomy to reduce the risk of breast cancer and ovarian cancer for BRCA-positive patients.
Few studies have evaluated the risk of breast cancer after a diagnosis of EOC for BRCA-positive patients. In a recent study by Domchek et al,19 of 164 BRCA mutation–positive patients who developed ovarian cancer, only 18 (11.0%) developed metachronous breast cancer. The low overall patient survival rate was mostly due to ovarian cancer deaths and to the fact that the breast cancer risk was fairly low in this group. The authors19 concluded that nonsurgical management of breast cancer in women with BRCA-associated ovarian cancer is an option. Similarly, our study demonstrates that the risk of breast cancer after EOC is low and that survival is dominated by ovarian cancer–related mortality. Moreover, at 6 years of follow-up, the probability that a BRCA-positive patient who was an ovarian cancer survivor would develop breast cancer was less than 10%. In our study, a substantial proportion of women presented with stage IIIC or IV ovarian cancer and subsequently died of ovarian cancer, and the risk that a woman with newly diagnosed BRCA-associated ovarian cancer will develop breast cancer during surveillance follow-up is also low.
In this study population, only 8.9% of patients developed breast cancer after EOC, with a median follow-up of 6.3 years. The patients in this study who had a recurrence of their ovarian cancer in less than 12 months did not undergo surveillance imaging for breast cancer; however, those patients who had no recurrence of EOC underwent surveillance imaging with mammography and breast MRI.
One possible reason is that 99.0% of our patients received first-line (mainly platinum-based) chemotherapy. It has been shown that a platinum-based chemotherapy regimen is highly effective in the treatment of patients with BRCA-associated ovarian cancer or breast cancer.20,21 It is possible that chemotherapy for EOC may substantially reduce the risk of subsequent breast cancer by eradicating microscopic disease in the breast. Indeed, the risk-reductive effect of adjuvant chemotherapy on the risk of contralateral breast cancer in BRCA mutation carriers who had developed breast cancer has been reported by Reding et al.22 Despite this observation, based on time to development of breast cancer in our cohort of patients, it appears that the aggressive nature of EOC is the key limiting aspect with regard to survival and the potential development of subsequent disease. With this in mind, it is important to assess the need for aggressive surveillance in this subpopulation of patients. A majority of patients received a diagnosis of breast cancer based on mammographic abnormalities, although none of the cancers were detected on MRI scans of the breasts. In addition, of those 87 patients who underwent a prophylactic mastectomy, only 3 (3.4%) were found to have breast cancer on final pathology.
Although MRI has been widely used for screening women at increased risk for breast cancer, none of the patients in this study population were diagnosed via this modality. The sensitivity of MRI for detecting breast tumors ranges from 77% to 91%, which is higher than mammography (33%-40%), for women at high familial risk for breast cancer.23,24 Although using MRI as a screening tool enables the earlier detection of breast cancer in women with familial risk for breast cancer or with BRCA mutations, mammography demonstrates an acceptable level of detecting ductal carcinoma in situ, and it has shown a higher specificity than MRI in the general population.24 In our study, a majority of cases of breast cancer after EOC were diagnosed via mammography because this is the primary surveillance modality used by most physicians. Within our study population, 78% of patients underwent mammography, whereas only 50% underwent MRI. Those patients who were found to have breast cancer were then more likely to undergo screening MRI subsequent to their diagnosis. In addition, those patients who were found to have evidence of recurrence within 6 months of an EOC diagnosis were not screened for breast cancer.
The estimated lifetime risk of ovarian cancer in a BRCA mutation carrier varies from 16% to 63%, depending on the gene and the population studied. It is currently thought that BRCA1 confers a lifetime risk of ovarian cancer of 40% to 50%, whereas BRCA2 carries a lower risk of 20% to 30%. Most ovarian cancers associated with germline BRCA mutations reported in the literature are high-grade and advanced-stage serous carcinomas, and as such our study population follows this pattern.25,26 Approximately 80% of patients within this study were diagnosed with stage III or greater EOC. Sixty-eight percent of patients had evidence of ovarian cancer recurrence within approximately 3 years. Forty-one percent of patients died of disease within the study period, and disease-free survival was estimated to be approximately 2.7 years. Although our study showed that only a small percentage of patients were found to have breast cancer after EOC diagnosis, the proportion of patients with ovarian cancer (stages I-II) was significantly smaller. This could suggest that current guidelines for breast cancer screening of patients with a diagnosis of EOC should take into account the stage of ovarian cancer at the time of diagnosis. Our data suggest that the risk of breast cancer after BRCA mutation–associated ovarian cancer is low, that aggressive management with annual breast MRI for surveillance may not be indicated, and that annual mammography may be sufficient for following these high-risk patients. Furthermore, guidelines suggest that, after a disease-free interval of 12 months, routine screening with breast examination and mammography may be warranted for patients with cancers of a lower stage.
Accepted for Publication: May 14, 2014.
Corresponding Author: Farin Amersi, MD, Division of Surgical Oncology, Department of Surgery, Cedars-Sinai Medical Center, 8700 Beverly Blvd, Los Angeles, CA 90048 (firstname.lastname@example.org).
Published Online: November 5, 2014. doi:10.1001/jamasurg.2014.1081.
Author Contributions: Dr Amersi had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Gangi, Cass, Paik, Karlan, Dang, Li, Walsh, Amersi.
Acquisition, analysis, or interpretation of data: Gangi, Cass, Paik, Barmparas, Karlan, Dang, Li, Rimel, Amersi.
Drafting of the manuscript: Gangi, Barmparas, Amersi.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Gangi, Cass, Barmparas, Dang.
Obtained funding: Karlan.
Administrative, technical, or material support: Barmparas, Karlan, Dang, Rimel.
Study supervision: Karlan, Li, Amersi.
Conflict of Interest Disclosures: None reported.
Previous Presentation: This paper was presented at the 85th Annual Meeting of the Pacific Coast Surgical Association; February 15, 2014; Dana Point, California.
Correction: This article was corrected on December 18, 2014, to fix an error in the text.