The risk of primary and contralateral breast cancer after ovarian cancer in BRCA1/BRCA2 mutation carriers

Implications for counseling

Authors

  • Peggy M. L. H. Vencken MD,

    1. Department of Gynecological Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Mieke Kriege PhD,

    1. Department of Medical Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Maartje Hooning PhD,

    1. Department of Medical Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Marian B. Menke-Pluymers MD, PhD,

    1. Department of Surgical Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Bernadette A. M. Heemskerk-Gerritsen MSc,

    1. Department of Medical Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Lena C. van Doorn MD, PhD,

    1. Department of Gynecological Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Margriet M. Collée MD, PhD,

    1. Department of Clinical Genetics, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Agnes Jager MD, PhD,

    1. Department of Medical Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Cees van Montfort PhD,

    1. Department of Statistics, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Curt W. Burger MD, PhD,

    1. Department of Gynecological Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    Search for more papers by this author
  • Caroline Seynaeve MD, PhD

    Corresponding author
    1. Department of Medical Oncology, Family Cancer Clinic, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Rotterdam, the Netherlands
    • Department of Medical Oncology, Erasmus University Medical Center-Daniel den Hoed Cancer Center, Groene Hilledijk 301, 3075 EA Rotterdam, the Netherlands
    Search for more papers by this author
    • Fax: (011) 31-10-7041003


Abstract

BACKGROUND:

The objective of this study was to assess the incidence of primary breast cancer (PBC) and contralateral breast cancer (CBC) in patients who had BRCA1/BRCA2-associated epithelial ovarian cancer (OC).

METHODS:

From the database of the Rotterdam Family Cancer Clinic, patients who had BRCA-associated OC without a history of unilateral breast cancer (BC) (at risk of PBC; n = 79) or with a history of unilateral BC (at risk of CBC; n = 37) were selected. The control groups consisted of unaffected BRCA mutation carriers (n = 351) or mutation carriers who had a previous unilateral BC (n = 294), respectively. The risks of PBC and CBC were calculated using the Kaplan-Meier survival method with death considered as a competing risk event.

RESULTS:

Women with BRCA-associated OC had lower 2-year, 5-year, and 10-year risks of PBC (3%, 6%, and 11%, respectively) compared with unaffected mutation carriers (6%, 16%, and 28%, respectively; P = .03), although they had a considerably higher mortality rate at similar time points (13%, 33%, and 61%, respectively, vs 1%, 2%, and 2%, respectively; P < .001). In BRCA mutation carriers with a previous unilateral BC, the 2-year, 5-year, and 10-year risks of CBC were nonsignificantly lower in patients with OC than in those without OC (0%, 7%, and 7%, respectively, vs 6%, 16%, and 34%, respectively; P = .06), whereas the mortality rate was higher in patients with OC (19%, 34%, and 55%, respectively, vs 4%, 11%, and 21%, respectively; P < .001).

CONCLUSIONS:

Patients with BRCA-associated OC had a lower risk of developing a subsequent PBC or CBC than mutation carriers without OC, whereas the risk of dying from OC was greater than the risk of developing BC. These data may facilitate more tailored counseling for this patient subgroup, although confirmative studies are warranted. Cancer 2013. © 2012 American Cancer Society.

INTRODUCTION

Women who have a germline mutation in the breast cancer susceptibility genes BRCA1 or BRCA2 have substantially elevated risks of developing both breast cancer (BC) and ovarian cancer (OC). It has been estimated that the cumulative 10-year risk of primary BC (PBC) ranges between 6% and 23% for BRCA1/BRCA2 mutation carriers, and the highest risk is observed between ages 40 and 50 years.1-6 After unilateral BC, the estimated 10-year risk of contralateral BC (CBC) is 29% to 39% for patients with BRCA1-associated BC and 23% to 35% for patients with BRCA2-associated BC.1,7-10 Factors that potentially affect the latter include age at first BC diagnosis (especially when diagnosed at ages <50 years),7,8,11adjuvant systemic therapy,12 and risk-reducing salpingo-oophorectomy (RRSO).13 In view of these increased BC risks, BRCA1 and BRCA2 mutation carriers are informed about various risk-reducing strategies, consisting of intensive BC surveillance (including magnetic resonance imaging [MRI]),14,15 chemoprevention,12,16,17 risk-reducing mastectomy (RRM),18-22 and/or RRSO.13,23-28

Although several publications have reported about the PBC risk in unaffected mutation carriers and the CBC risk in patients with BRCA-associated BC, to our knowledge, no data are available about the risks of BC after BRCA-associated OC. For this subgroup, these risks may be reduced by the treatment women receive for OC, which consists mainly of surgery (including salpingo-oophorectomy) and platinum-based chemotherapy. First, salpingo-oophorectomy in mutation carriers has been associated with a reduced BC risk, which is most pronounced if undergone at a premenopausal age.24 Furthermore, it has been reported that BRCA mutation carriers are highly sensitive to platinum-based chemotherapy, which leads to improved survival for BRCA-associated OC patients compared with those who have sporadic OC.29,30 However, despite the improved outcome in patients with BRCA-associated OC, the 5-year and 10-year survival rates remain poor (63% and 35%, respectively),31 which has to be taken into account at counseling. Therefore, precise data concerning the risks of PBC or CBC after BRCA-associated OC are warranted to enable optimization of and more patient-tailored counseling regarding the subsequent strategies for this patient subgroup. In the current study, we assessed the incidences of PBC and CBC in BRCA-associated OC patients without and with a history of unilateral BC compared with mutation carriers who did not have OC.

MATERIALS AND METHODS

Patient Selection

From the database of the Rotterdam Family Cancer Clinic (FCC) at the Erasmus University Medical Center-Daniel den Hoed Cancer Center, we selected all patients who had BRCA1-associated or BRCA2-associated epithelial OC diagnosed between January 1, 1980 and January 1, 2009. DNA testing was performed, as previously described.32 We excluded patients who underwent RRM before their OC diagnosis, patients who had another malignancy (other than unilateral BC) before their OC diagnosis, BC patients who had recurrent disease, and patients of whom inadequate data concerning tumor and treatment characteristics and follow-up were available. In total, 79 women who had BRCA-associated OC without a history of unilateral BC (at risk of PBC) and 37 women who had BRCA-associated OC with a previous unilateral BC (at risk of CBC) were included in the current analyses.

The control group for the patients with BRCA-associated OC (without BC) consisted of all unaffected BRCA1/BRCA2 mutation carriers aged ≥35 years who were registered at the FCC (n = 351; at risk of PBC). For the OC group being at risk of CBC, the control group consisted of BRCA1/BRCA2 mutation carriers with a previous unilateral BC (without OC), as assessed at their first visit to the FCC (n = 294). RRSO was not an exclusion criterion, because this more accurately represents the group of BRCA mutation carriers who attend FCCs.

For all eligible patients, information regarding patient, tumor, and treatment characteristics; preventive surgery; and follow-up data was retrieved from medical files. The study was approved by the Institutional Review Board.

Statistical Analyses

The start of study follow-up for patients with BRCA-associated OC was the date of their OC diagnosis; for both control groups, the starting date was their first visit to the FCC or their 35th birthday, whichever came last. The study endpoint was BC, either PBC or CBC (both invasive cancer and ductal carcinoma in situ [DCIS]). The cumulative BC incidence was calculated using the Kaplan-Meier method. Because OC is associated with a high mortality rate, and death is an event that precludes the occurrence of BC, death was considered as a competing risk event in the current analyses.33 BRCA1/BRCA2 mutation carriers were censored at the date of RRM, the date they were lost to follow-up, or the end date of the study (June 1, 2009). The log-rank test was used to compare BC incidence and mortality rates between the different groups. To investigate whether age influenced the risk of BC, the regression coefficient of age on the risk of BC was calculated. To assess whether PBC and CBC incidence rates differed between the patients with OC and the control groups, independent of the mortality rate in the OC group, Kaplan-Meier analyses also were performed with death as a censoring event.

A multivariate Cox proportional hazards model was used to adjust the differences in BC risk between the OC and control groups for potential confounders, considering age at study entry, year of study entry, and type of mutation (BRCA1 vs BRCA2) as variables. In the analyses of CBC risk, the period between the first BC and the start of follow-up was considered an additional potential confounder. Furthermore, women were censored at the dates of death, RRM, loss to follow-up, or study end (June 1, 2009). Two-sided P values < .05 were considered statistically significant. Statistical analyses were performed using the SPSS (version 15.0; SPSS, Inc., Chicago, Ill) and STATA (version 11.1; Stata Corporation, College Station, Tex) statistical software packages.

RESULTS

Patient and treatment characteristics are described in Table 1. The OC group that was at risk of PBC consisted of significantly more BRCA1 mutation carriers and was older than the control group at study entry. In the OC group that was at risk of CBC, the PBC occurred approximately 10 years before the OC diagnosis. Because OC was mostly diagnosed at an advanced stage (International Federation of Gynecology and Obstetrics [FIGO] stage III/IV in 76%), most patients received chemotherapy (mainly platinum-based) as part of primary OC treatment. The mean follow-up for women at risk of PBC was 6.7 years and 6.5 years for patients with and without OC, respectively; whereas, in the groups that were at risk of CBC, the mean follow-up was 6.6 years and 8.9 years, respectively.

Table 1. Patient Demographics
CharacteristicBRCA Mutation Carriers at Risk of PBCBRCA Mutation Carriers at Risk of CBC
No. of Patients (%)PNo. of Patients (%)P
Patients with OCUnaffected ControlsPatients with OC and Previous BCControls: History of BC at FCC Registration
  1. Abbreviations: BC, breast cancer; PBC, primary breast cancer; CBC, contralateral breast cancer; FCC, Family Cancer Clinic; y, year; FIGO, International Federation of Gynecology and Obstetrics; OC, ovarian cancer; RRM, risk-reducing mastectomy; RRSO, risk-reducing salpingo-oophorectomy.

Total no.79351 37294 
Mutation status      
 BRCA172 (91)252 (72)<.00130 (81)216 (72).32
 BRCA27 (9) 99 (28) 7 (19) 78 (27) 
Median age at study entry [range], y50.3 [33.4-72.8]40.1 [35.0-69.2]<.00155.0 [39.5-68.3]41.3 [35.0-73.9]<.001
Median age at BC diagnosis [range], y 45.8 [33.0-59.3]40.1 [24.8-69.4]<.001
Year at study entry      
 1980-199016 (20) 6 (2)<.001 8 (22) 39 (13).17
 1990-200043 (54) 99 (28) 16 (43)107 (37) 
 2000-200920 (26)246 (70) 13 (35)148 (50) 
FIGO stage of OC      
 I10 (13)  4 (11)  
 II 8 (11)  5 (14)  
 III47 (63)  22 (61)  
 IV10 (13)  5 (14)  
 Unknown4  1  
Chemotherapy for OC      
 No5 (6) 1 (3) 
 Platinum/taxol32 (41) 16 (43) 
 Platinum without taxol41 (52) 18 (49) 
 Nonplatinum-based1 (1) 2 (5) 
RRSO      
 No181 (52) 172 (59) 
 Yes170 (48) 122 (41) 
 Median age [range], y47.5 [29.2-71.3] 46.7 [34.4—72.1] 
RRM      
 No76 (96)242 (69)<.00135 (95)212 (72).003
 Yes3 (4)109 (31) 2 (5) 82 (28) 
 Median age [range], y49.8 [44.6-52.8]40.9 [35.4-60.4] 50.6 [48.0-53.1]44.5 [35.4-65.1] 

In total, 8 PBCs were detected in the OC group, and 49 PBCs were detected in the control group (Table 2), including 3 PBCs in controls (6.1%) which were found at RRM. In patients who had a previous unilateral BC, 4 CBCs were detected in the OC group, and 70 CBCs were detected in the control group (Table 2), including 6 that were found at RRM (1 in the OC group, 5 in the control group). One OC patient of the PBC risk group had metastatic BC at diagnosis (detected before 2000), and 1 OC patient who was at risk of CBC was diagnosed with a pathologic T4 tumor, whereas lymph node and metastasis classifications did not differ significantly between patients with OC (all diagnosed after 2000) and control patients.

Table 2. Incidence of Primary and Contralateral Breast Cancer and the Risk of Death
Patients at Risk of PBCPatients With OCControls: Unaffected Mutation CarriersP-value
  • Abbreviations: BC, breast cancer; CBC, contralateral breast cancer; CI, confidence interval; FCC, Family Cancer Clinic; y, yeras; HR, hazard ratio; OC, ovarian cancer; PBC, primary breast cancer; Tis, tumor in situ.

  • a

    P values < .05 were considered statistically significant.

  • b

    This analysis was adjusted for year at study entry. Age at study entry and mutation type had <10% impact on the HR.

  • c

    This analysis was adjusted for age at study entry and time between first BC and study entry.

No. of patients79351 
Mean time at risk, y5.84.0 
No. of PBCs, including DCIS849 
Median age at PBC diagnosis [range], y57.8 [49.0–65.1]44.7 [35.3–70.7] 
Tumor classification: No. (%)   
 Tis1 (17)3 (6) 
 T12 (33)38 (78) 
 T23 (50) 7 (14) 
 T31 (2) 
 Tx2.106
Lymph node status: No. (%)   
 Negative3 (43)35 (74) 
 Positive4 (57)12 (26).087
Metastasis status: No. (%)   
 Negative6 (86)46 (100) 
 Positive1 (14).01
No. of deaths424 
PCB risk, %   
 After 2 y36 
 After 5 y616 
 After 10 y1128.03a
Risk of death, %   
 After 2 y131 
 After 5 y332 
 After 10 y612<.001a
HR/95% CI   
 Univariate analysis0.43/0.20−0.95Reference group.04
 Multivariate analysis0.35/0.15−0.82bReference group.02
Patients at Risk of CBCPatients With OC and Previous BCControls: History of BC at FCC Registration 
No. of patients37294 
Mean time at risk, y6.45.6 
No. of CBCs, including DCIS470 
Median age at CBC diagnosis [range], y56.9 [46.1–71.9]46.2 [35.5–77.7] 
Tumor classification: No. (%)   
 Tis 7 (11) 
 T13 (75)45 (68) 
 T213 (20) 
 T41 (25)1 (1) 
 Tx4.036
Lymph node status: No. (%)   
 Negative3 (100)44 (71) 
 Positive18 (29) 
 Unknown11.272
Metastasis status: No. (%)   
 Negative4 (100)63 (100) 
No. of deaths1739 
CBC risk, %   
 After 2 y06 
 After 5 y716 
 After 10 y734.06a
Risk of death, %   
 After 2 y194 
 After 5 y3411 
 After 10 y5521<.001a
HR/95% CI   
 Univariate analysis0.40/0.14–1.09Reference group.07
 Multivariate analysis0.52/0.18–1.53cReference group.24

Table 2 and Figure 1 show that significantly fewer BRCA-associated OC patients developed a PBC compared with unaffected mutation carriers. The 2-year, 5-year, and 10-year risks of PBC in patients with OC were 3%, 6%, and 11%, respectively, compared with 6%, 16%, and 28%, respectively, in unaffected mutation carriers (P = .03). In contrast, at similar time points, the risk of death was significantly higher in patients with OC than in unaffected mutation carriers (Table 2). Age at study entry was not statistically significantly associated with BC risk (regression coefficient, 0.89; 95% confidence interval, 0.72-1.11), indicating that it was not a confounder in our study. Of the 8 women who were diagnosed with PBC after OC, 4 women died (3 from BC, 1 from an unknown reason) within a mean follow-up after PBC of 4.1 years.

Figure 1.

The risk of primary breast cancer is illustrated for patients with BRCA-associated ovarian cancer (OC) (blue line) and for unaffected BRCA mutation carriers (red line). Note that the numbers of BRCA mutation carriers with and without OC were 61 and 212, respectively, at 2 years; 35 and 107, respectively, at 5 years; and 13 and 33, respectively, at 10 years.

In the patient groups of women with a previous unilateral BC, those with OC developed a CBC nonsignificantly less often than controls (Table 2, Fig. 2). The 2-year, 5-year, and 10-year risk of CBC was 0%, 7%, and 7%, respectively, in patients with OC versus 6%, 16%, and 34%, respectively, in controls (P = .06). At similar time points, the risk of death in patients with OC was significantly higher than that in controls (Table 2). Age at study entry was not statistically significantly associated with CBC risk (regression coefficient, 0.81; 95% confidence interval, 0.57-1.11). Of the 4 women who developed a CBC after OC, 1 patient died, but not because of cancer.

Figure 2.

The risk of contralateral breast cancer is illustrated for patients with BRCA-associated ovarian cancer (OC) (blue line) and for BRCA mutation carriers without OC (red line). Note that the numbers of BRCA patients with and without OC were 28 and 208, respectively, at 2 years; 16 and 126, respectively, at 5 years; and 7 and 43, respectively, at 10 years.

In the analyses with death as a censoring event (vs as competing risk event), the risk of PBC at 2 years, 5 years, and 10 years also remained significantly lower in the OC group than in the control group being 3%, 7%, and 19%, respectively, compared with 6%, 21%, and 53%, respectively (P = .03). Furthermore, we observed a trend toward a lower risk of CBC in patients with OC versus controls, with 2-year, 5-year, and 10-year incidences of 0%, 11%, and 11%, respectively, versus 6%, 19%, and 40%, respectively (P = .06; data not shown). Performing these analyses after the exclusion of controls who underwent premenopausal RRSO, the significance increased regarding PBC risk, whereas the risk of CBC in patients with OC became significantly lower.

In multivariate analysis, after adjusting for age at OC diagnosis, period of OC diagnosis, and mutation status, the risk of developing PBC remained significantly lower in women with BRCA-associated OC compared with healthy mutation carriers (hazard ratio, 0.35; 95% confidence interval, 0.15-0.82). The risk of CBC was also lower in patients with OC than in controls without OC, but the confidence interval was wide (hazard ratio, 0.52; 95% confidence interval, 0.18-1.53) (Table 2).

DISCUSSION

To our knowledge, these are the first estimates on the risk of developing a PBC or CBC after BRCA-associated OC. Compared with mutation carriers without OC, patients who had BRCA-associated OC had a significantly lower risk of PBC and a nonsignificantly lower risk of developing a CBC within the first 10 years after OC. Also, the mortality rate for patients with BRCA-associated OC during this period was substantial and was higher than the risk of developing a subsequent BC.

The observation of a lower BC incidence in patients with BRCA-associated OC versus controls without OC, which persisted in the analyses that considered death as a censoring event, suggests that primary therapy for OC may influence the occurrence of BC thereafter. The reductive effect of salpingo-oophorectomy on the risk of BC has been demonstrated repeatedly, and probably is most pronounced in healthy mutation carriers who undergo RRSO at ages <50 years,13,26 and concerns PBC rather than CBC.13 It is not clear whether surgery in our OC groups contributed much to the observed lower BC risk, because approximately 50% of the patients with OC were aged >50 years at diagnosis. Furthermore, 97% of the OC patients received first-line (mainly platinum-based) chemotherapy. In view of the observation that platinum-based chemotherapy is highly effective in patients with BRCA-associated OC,29,31 as also reported for patients with BRCA-associated BC,34,35 we hypothesize that chemotherapy for OC may reduce the risk of subsequent BC by eradicating occult microscopic disease. Indeed, a risk-reductive effect of adjuvant chemotherapy on CBC in BRCA mutation carriers has been reported by Reding et al.36 Because only 6 OC patients in our series did not receive chemotherapy as part of their primary treatment, it was not possible to separately analyze the risk of developing BC after OC in patients who did and did not receive chemotherapy.

We are aware that some limitations to the current study have to be considered. First, the OC and control patients differed significantly with regard to age at and year of study entry, and the majority of OC patients underwent genetic testing after their OC diagnosis. Over time, various BC screening schemes have been used, because BC screening for high-risk women at FCCs was introduced around 1994, when genetic testing for BRCA became available, whereas MRI for high-risk women was introduced from 2000 onward in the Netherlands.15,37 On the other hand, population screening for BC (biennial mammography for women ages 50-75 years) was introduced in the Netherlands in 1990 and is highly attended (>80%). Also, follow-up after BC includes annual mammography. The mentioned factors potentially may have resulted in a higher BC detection rate in unaffected controls because of more intensive screening and the greater sensitivity of MRI, because the majority of these women started study follow-up after 2000 (Table 1). Conversely, lead time (the time between tumor detection and developing symptoms) is very short in BRCA/BRCA2 mutation carriers because of the high tumor growth rate.38 Furthermore, >40% of the controls underwent RRSO at ages <50 years, lowering their BC risk. In addition, increasing the age at study entry in the control groups from 35 years to 40 or 45 years, respectively, resulted in more comparable median ages at study entry, whereas differences in the BC risks became somewhat smaller, but significance did not disappear (data available on request). Furthermore, age at study entry was not significantly correlated with BC risk; and, in multivariate analyses, the conclusions did not alter after adjustment for age at and year of study entry. Comparable studies in further nationwide and international studies that include larger patient cohorts are warranted that then can match OC patients with controls for age at diagnosis and period of diagnosis.

Second, our study groups were too small to distinguish between both BRCA1 and BRCA2 mutation carriers and OC patients with early versus advanced disease. Because most of the OC patients in our series were diagnosed with advanced-stage disease, further investigation should be conducted to determine whether our results also are applicable to patients with FIGO stage I or II OC.

Finally, in all of our study groups, the majority of patients underwent genetic testing after they developed BC and/or OC, and index patients were not excluded from the analyses, suggesting that the risk of BC in the cancer groups may have been overestimated. However, if the risk of BC in the OC groups has been overestimated due to ascertainment bias, then the unbiased risk of BC would be even lower. When we repeated the analyses starting at the date of first DNA diagnosis in the family, the risk of PBC in the OC group remained nonsignificantly lower than that in the control group (hazard ratio, 0.40; 95% confidence interval, 0.16-1.37).

Despite the limitations mentioned above, our data indicating that the risk of BC after advanced-stage, BRCA-associated OC is lower than for a mutation carrier without OC, whereas the risk of OC-related death within the first years remains substantial, provide helpful information for involved clinicians that may allow more tailored counseling for patients with BRCA-associated OC. First, the risk of BC in the first decade after BRCA-associated OC can be communicated more precisely. Second, the risk of BC in view of the mortality rates after advanced-stage OC indicates that the option of RRM may not be justified for this patient group, also because RRM may be associated with negative effects regarding body image, potential surgical complications, and relevant costs.39-41 Further research should try to obtain more data concerning the BC risks after early stage, BRCA-associated OC and for OC patients with a disease-free interval >5 years, because different risks and strategies possibly may apply for these patients.

Furthermore, our data underscore the reality that the counseling of a patient with BRCA-associated OC is complex, because counseling should address not only the subsequent risk of BC but also the consideration of this risk against the OC prognosis. Therefore, we propose that a multidisciplinary team, including a clinical geneticist and a gynecologic, medical, and surgical oncologist, should be involved in decision-making and counseling, which currently is available only at FCCs in an academic center.

In addition, the current findings suggesting a lower BC risk over the first years after BRCA-associated OC raise questions concerning the optimal BC surveillance program for mutation carriers affected with OC. Currently, in most developed countries, the BC surveillance for BRCA mutation carriers, irrespective of a previous cancer, consists of yearly mammography and MRI between ages 30 years and 60 years, which are expensive, time-consuming program that may lead to additional examinations because of the relative low specificity of MRI, although cost-effectiveness has been demonstrated only for healthy mutation carriers.42 Potentially, an alternative regimen of annual imaging deserves consideration for the BRCA-associated OC subgroup.

In conclusion, compared with BRCA mutation carriers without OC, patients with BRCA-associated OC had a lower risk of PBC or CBC, whereas the mortality rate within the first years after OC was higher than the risk of developing a subsequent BC. These findings provide additional data enabling more patient-tailored counseling for this patient subgroup and indicate that RRM within the first years after advanced-stage OC should not be considered. We propose referring patients with BRCA-associated OC for optimal decision-making and counseling to a FCC with an adequately equipped multidisciplinary team. Given the heterogeneity of our study cohorts, we believe that confirmation of our data is warranted in future studies with larger sample sizes.

Acknowledgements

We thank Petra Bos, Demelza Hoogwerf, and Judith Rikken for their useful help with data collection.

FUNDING SOURCES

No specific funding was disclosed.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

Ancillary