This cohort study evaluates the association between bilateral oophorectomy and all-cause mortality among women with BRCA1 or BRCA2 sequence variations.
Key Points
Question
Is bilateral oophorectomy associated with a decreased risk of death among women with a BRCA1 or BRCA2 sequence variation?
Findings
In this cohort study of 4332 women, after a mean follow-up of 9.0 years, 901 incident cancers were noted in 851 women, and 228 women had died (57 of ovarian or fallopian tube cancer, 58 of breast cancer, 16 of peritoneal cancer, and 97 of other causes). The age-adjusted hazard ratio was 0.32 for all-cause mortality associated with oophorectomy, 0.28 for women with BRCA1 sequence variations, and 0.43 for women with BRCA2 sequence variations.
Meaning
Results suggest that among women with a BRCA1 or BRCA2 sequence variation, oophorectomy is associated with a significant reduction in all-cause mortality.
Abstract
Importance
Preventive bilateral salpingo-oophorectomy is offered to women at high risk of ovarian cancer who carry a pathogenic variant in BRCA1 or BRCA2; however, the association of oophorectomy with all-cause mortality has not been clearly defined.
Objective
To evaluate the association between bilateral oophorectomy and all-cause mortality among women with a BRCA1 or BRCA2 sequence variation.
Design, Setting, and Participants
In this international, longitudinal cohort study of women with BRCA sequence variations, information on bilateral oophorectomy was obtained via biennial questionnaire. Participants were women with a BRCA1 or BRCA2 sequence variation, no prior history of cancer, and at least 1 follow-up questionnaire completed. Women were followed up from age 35 to 75 years for incident cancers and deaths. Cox proportional hazards regression was used to estimate the hazard ratios (HRs) and 95% CIs for all-cause mortality associated with a bilateral oophorectomy (time dependent). Data analysis was performed from January 1 to June 1, 2023.
Exposures
Self-reported bilateral oophorectomy (with or without salpingectomy).
Main Outcomes and Measures
All-cause mortality, breast cancer–specific mortality, and ovarian cancer–specific mortality.
Results
There were 4332 women (mean age, 42.6 years) enrolled in the cohort, of whom 2932 (67.8%) chose to undergo a preventive oophorectomy at a mean (range) age of 45.4 (23.0-77.0) years. After a mean follow-up of 9.0 years, 851 women had developed cancer and 228 had died; 57 died of ovarian or fallopian tube cancer, 58 died of breast cancer, 16 died of peritoneal cancer, and 97 died of other causes. The age-adjusted HR for all-cause mortality associated with oophorectomy was 0.32 (95% CI, 0.24-0.42; P < .001). The age-adjusted HR was 0.28 (95% CI, 0.20-0.38; P < .001) and 0.43 (95% CI, 0.22-0.90; P = .03) for women with BRCA1 and BRCA2 sequence variations, respectively. For women with BRCA1 sequence variations, the estimated cumulative all-cause mortality to age 75 years for women who had an oophorectomy at age 35 years was 25%, compared to 62% for women who did not have an oophorectomy. For women with BRCA2 sequence variations, the estimated cumulative all-cause mortality to age 75 years was 14% for women who had an oophorectomy at age 35 years compared to 28% for women who did not have an oophorectomy.
Conclusions and Relevance
In this cohort study among women with a BRCA1 or BRCA2 sequence variation, oophorectomy was associated with a significant reduction in all-cause mortality.
Introduction
Women with a germline pathogenic (or likely pathogenic) variant in the BRCA1 or BRCA2 gene are encouraged to undergo a bilateral salpingo-oophorectomy (ie, oophorectomy) before the ages of 40 and 45 years, respectively, to reduce their risk of developing ovarian and fallopian tube cancer. The surgery is effective; however, there is a small risk of developing peritoneal cancer postoophorectomy. It is not clear if this represents a primary cancer or a metastatic spread of an occult tubal cancer or precancerous lesion.
Early surgical menopause has unintended consequences, including a decline in cardiovascular and bone health, and has negative impacts on fertility, sexual function, and quality of life. In women without sequence variations, premenopausal oophorectomy has been associated with an increase in all-cause mortality, predominantly due to noncancer deaths. It is important to accurately measure the risks and benefits of oophorectomy in women found to carry a BRCA sequence variation. Some women with a BRCA sequence variation choose to postpone surgery until childbearing is complete, while others forgo oophorectomy altogether.
In 2014, we reported a 70% reduction in all-cause mortality among women with a BRCA1 or BRCA2 sequence variation following an oophorectomy. The current study is an update of our previous report; we have expanded the size of our international cohort and extended the follow-up period. Further, we excluded women diagnosed with cancer prior to study enrollment to focus solely on the impact of oophorectomy for women without a cancer diagnosis who become aware of their BRCA sequence variation status and face the decision of preventive surgery.
Methods
Study Population
Eligible study participants were identified from a longitudinal study of 17 947 women with a pathogenic or likely pathogenic germline variant in the BRCA1 or BRCA2 gene, which was initiated in 1995 and includes 83 participating centers from 16 countries. The women underwent genetic testing over an extended period of time (1994-2019). These women sought genetic testing because of a personal or family history of breast and/or ovarian cancer. The criteria for genetic testing were not standardized and varied from center to center. Sequence variation detection was performed using a range of techniques, and all abnormal nucleotide sequences were confirmed by direct DNA sequencing. All participants provided written informed consent, and the institutional ethics committees of all the participating centers approved the study.
Data Collection
Participants completed a baseline questionnaire at the time of enrollment and a follow-up questionnaire every 2 years thereafter to update exposures and ascertain incident cancers and deaths. Questionnaires were mailed to the study participant or were administered over the telephone by a genetic counselor or research assistant. The questionnaires requested information regarding surgery (eg, oophorectomy, hysterectomy), and exogenous hormone use (eg, hormone replacement therapy [HRT], oral contraceptives [OCs]). For the current study, oophorectomy was defined as a bilateral oophorectomy, with or without a concomitant hysterectomy or salpingectomy. Prior to 2012, the questionnaires did not distinguish between oophorectomy and salpingo-oophorectomy. An oophorectomy performed for the treatment of clinically or screen-detected ovarian cancer was not considered as an exposure for this analysis (patients were considered unexposed). In contrast, an oophorectomy performed with preventive intent, but for which an invasive cancer was detected in the fallopian tubes or ovaries at the time of surgery (occult), was considered as a preventive oophorectomy (exposed). Women who underwent a unilateral oophorectomy were considered unexposed (no oophorectomy); however, they were reclassified to the exposed group if the second ovary was removed later. In the subanalysis of breast cancer–specific mortality, a woman who had an oophorectomy following the diagnosis of breast cancer was included as exposed.
Incident Cancer Diagnoses, Vital Status, and Cause of Death
Pathology reports and medical records were requested for all women who reported incident breast, ovarian, or endometrial cancer, and information regarding histologic type, stage, site of origin, and other pathologic features was abstracted. Serous peritoneal cancer diagnosed 6 or more months after a preventive oophorectomy was considered a primary peritoneal cancer, whereas ovarian, fallopian tube, or peritoneal cancer diagnosed in women with 2 ovaries intact was classified as ovarian or fallopian tube cancer.
For women who died, the cause and date of death were obtained from the collaborating investigator at each site. This was determined by review of patient records, by correspondence with the treating physician, or from next of kin. Furthermore, for women from Ontario, Canada, cause and date of death were determined by record linkage to the Ontario Cancer Registry (Ontario Health). For the participants from Poland, vital status and date of death were determined by record linkage to the Vital Statistics Database of the Polish Ministry of Administration and Internal Affairs. In some cases, women were diagnosed with cancer but died before completion of a follow-up questionnaire. These diagnoses were reported by the study center or by next of kin and were confirmed through pathology or medical record review. In some cases, the cause of death was listed as cancer, but no incident cancer was reported prior to the death. For these cases, we considered the date of diagnosis to be 1 year prior to the date of death.
Inclusions and Exclusions
Participants were excluded if they had been diagnosed with any cancer (except for thyroid or nonmelanoma skin cancer) prior to completion of the baseline questionnaire or if they did not complete at least 1 follow-up questionnaire. Women were also excluded if they (1) completed the most recent follow-up questionnaire before age 35 years; (2) completed the baseline questionnaire after age 74 years; (3) had follow-up time less than 1 year; or (4) were missing data on key variables (eg, date of birth). After these exclusions, there were 4332 women eligible for the analysis. See eFigure 1 in Supplement 1 for a detailed summary of the exclusions.
Statistical Analysis
In the primary analysis, we estimated the extent of risk reduction (hazard ratio [HR]) for all-cause mortality associated with bilateral oophorectomy using a Cox proportional hazards model. We included oophorectomy as a time-dependent variable (women who underwent bilateral oophorectomy during the follow-up were transferred from unexposed to exposed at that time). Women who had an oophorectomy prior to the baseline questionnaire were considered exposed from the date of completion of the baseline questionnaire. Participants were followed up from the date of the baseline questionnaire or age 35 years (whichever was later) until either death, age 75 years, or date of completion of the last follow-up questionnaire. All HRs were adjusted for age at study entry and country of residence (Canada, US, Poland, other). We conducted an additional multivariate analysis and included smoking (ever/never), OC use (ever/never), parity, HRT (yes/no), and bilateral preventive mastectomy (yes/no; time dependent) as covariates. A second analysis was conducted with death from ovarian, fallopian tube, or peritoneal cancer as the composite end point. A third analysis was conducted with death from breast cancer as the end point; here, we censored the participants at the time of preventive mastectomy, and we considered oophorectomies that occurred both before and after the diagnosis of breast cancer as exposed.
Annual mortality rates were calculated by age, gene sequence variation, and oophorectomy status. These rates were based on the ratio of the number of events (deaths) and the number of person-years of risk accumulated in that specific age interval. We used the annual mortality rates to construct cumulative mortality curves (from age 35 to 75 years) for 4 theoretical cohorts of women who (1) had an oophorectomy at age 35 years and (2) did not have an oophorectomy before age 75 years, and subdivided by gene.
The SAS statistical package, version 9.4 (SAS Institute), was used to conduct the analyses. P values were based on 2-sided tests and were considered statistically significant if P < .05.
Results
In this prospective study, we observed 4332 women (mean age, 42.6 years; 3177 with a BRCA1 sequence variation and 1155 with a BRCA2 sequence variation) for up to 24 years (mean, 9.0 years) for incident cancers and death from all causes. The characteristics of the study participants are presented in Table 1. There were 901 incident cancers diagnosed (among 851 women) in the cohort: 582 breast (496 invasive and 86 ductal carcinoma in situ), 140 ovarian or fallopian tube (94 clinically detected through symptoms or screening and 46 occult diagnosed at oophorectomy), and 35 primary peritoneal cancers diagnosed after oophorectomy, as well as 24 melanomas; 17 pancreatic, 16 endometrial, 13 colorectal, and 12 lung cancers; and 62 cancers of other sites (Table 2). Table 3 summarizes the 228 deaths reported in the cohort. Of these, 182 (80.0%) were from cancer, 28 (12.3%) were from other causes, and for 18 deaths (7.9%), the cause of death was unknown.
Table 1. Characteristics of the 4332 Women With BRCA1 and BRCA2 Sequence Variations, Overall and According to Preventive Oophorectomy Status (Ever/Never).
| Variable | Participants, No. (%) | P valuea | |
|---|---|---|---|
| No oophorectomy (n = 1400) | Oophorectomy (n = 2932) | ||
| Year of birth, mean (range) | 1968 (1922-1986) | 1962 (1924-1985) | <.001 |
| Year of baseline, mean (range) | 2006 (1992-2019) | 2006 (1993-2019) | .32 |
| Age at baseline, mean, y | 37.8 | 44.9 | <.001 |
| Years of follow-up from baseline, mean (range) | 8.2 (1.0-24.3) | 9.4 (1.0-24.5) | <.001 |
| Years of follow-up from oophorectomy, for those who had surgery after baseline, mean | NA | 6.7 | NA |
| BRCA sequence variation | |||
| BRCA1 | 1071 (76.5) | 2106 (71.8) | .001 |
| BRCA2 | 329 (23.5) | 826 (28.2) | |
| Age at oophorectomy, mean (range), y | NA | 45.4 (23.0-77.0) | NA |
| Bilateral mastectomy | |||
| No | 1141 (85.5) | 1182 (66.6) | <.001 |
| Yes | 193 (14.5) | 945 (33.4) | |
| Missing | 66 | 105 | |
| Hysterectomy | |||
| No | 1305 (94.0) | 1240 (45.4) | <.001 |
| Yes | 82 (5.9) | 1590 (54.6) | |
| Missing | 13 | 92 | |
| Parity | |||
| Mean (range) | 1.5 (0-10) | 1.9 (0-8) | <.001 |
| Nulliparous | 425 (31.3) | 428 (14.9) | <.001 |
| Parous | 933 (68.7) | 2454 (85.2) | |
| Missing | 42 | 50 | |
| Breastfeedingb | |||
| Mean (range), mo | 13.0 (0-96) | 12.6 (0-147) | .50 |
| Never | 124 (15.4) | 341 (16.0) | .72 |
| Ever | 680 (84.6) | 1796 (84.0) | |
| Missing | 129 | 317 | |
| Smoking | |||
| Never | 769 (57.8) | 1600 (57.3) | .73 |
| Ever | 561 (42.2) | 1195 (42.8) | |
| Missing | 70 | 137 | |
| Oral contraceptive use | |||
| Never | 439 (31.3) | 750 (26.0) | <.001 |
| Ever | 939 (68.7) | 2130 (74.0) | |
| Missing | 22 | 52 | |
| HRT use | |||
| Never | 1273 (90.9) | 1392 (47.5) | <.001 |
| Ever | 127 (9.1) | 1540 (52.5) | |
| Country of residenceb | |||
| Canada | 277 (24) | 882 (76) | .01 |
| Poland | 625 (44) | 786 (56) | |
| US | 252 (24) | 806 (76) | |
| Other | 246 (35) | 458 (65) | |
Abbreviations: HRT, hormone replacement therapy; NA, not applicable.
P value comparing women who did and did not undergo bilateral oophorectomy; t test for continuous variables and χ2 test for categorical variables.
Among parous women only.
Table 2. Incident Cancers Diagnosed in the Follow-Up Period, by Oophorectomy Status.
| Cancer type | Cancers, No. | ||
|---|---|---|---|
| Total (N = 4332) | Oophorectomy (n = 2932) | No oophorectomy (n = 1400) | |
| Breast (invasive) | 496 | 384 | 112 |
| Ovarian | 121 | 32 | 89 |
| DCIS | 86 | 67 | 19 |
| Peritoneal | 35 | 34 | 1 |
| Melanoma | 24 | 20 | 4 |
| Fallopian tube | 19 | 18 | 1 |
| Pancreatic | 17 | 12 | 5 |
| Endometrial | 16 | 12 | 4 |
| Colorectal | 13 | 10 | 3 |
| Lung | 12 | 8 | 4 |
| Lymphoma | 7 | 7 | 0 |
| Brain | 6 | 4 | 2 |
| Kidney | 5 | 5 | 0 |
| Liver | 4 | 2 | 2 |
| Parotid gland | 3 | 3 | 0 |
| Stomach | 3 | 3 | 0 |
| Angiosarcoma | 2 | 1 | 1 |
| Bladder | 2 | 2 | 0 |
| Bone | 2 | 2 | 0 |
| Cervical | 2 | 2 | 0 |
| Leukemia | 2 | 2 | 0 |
| Vulvar | 1 | 1 | 0 |
| Multiple myeloma | 1 | 1 | 0 |
| Sarcoma | 1 | 1 | 0 |
| Other cancer | 18 | 15 | 3 |
| Unknown | 3 | 2 | 0 |
| Totala | 901 | 650 | 251 |
Abbreviation: DCIS, ductal carcinoma in situ.
There was a total of 901 incident cancers among 851 participants.
Table 3. Causes of Death in the Cohort, by Oophorectomy Status.
| Cause of death | Participants, No. (%) | ||
|---|---|---|---|
| Total (N = 4332) | Oophorectomy (n = 2932) | No oophorectomy (n = 1400) | |
| Breast cancer | 58 (1.34) | 30 (1.02) | 28 (2.00) |
| Ovarian cancer | 55 (1.27) | 8 (0.27) | 47 (3.36) |
| Pancreatic cancer | 16 (0.37) | 10 (0.34) | 6 (0.43) |
| Peritoneal cancer | 16 (0.37) | 16 (0.55) | 0 |
| Cardiovascular diseasea | 9 (0.21) | 5 (0.17) | 4 (0.29) |
| Lung cancer | 8 (0.18) | 5 (0.17) | 3 (0.21) |
| Cancer (other) | 7 (0.16) | 6 (0.20) | 1 (0.07) |
| Brain tumor | 4 (0.09) | 2 (0.07) | 2 (0.14) |
| Unintentional injury | 3 (0.07) | 1 (0.03) | 2 (0.14) |
| Colorectal cancer | 3 (0.07) | 2 (0.07) | 1 (0.07) |
| Liver cancer | 3 (0.07) | 2 (0.07) | 1 (0.07) |
| Endometrial cancer | 2 (0.05) | 0 | 2 (0.14) |
| Fallopian tube cancer | 2 (0.05) | 1 (0.03) | 1 (0.07) |
| Leukemia | 2 (0.05) | 2 (0.07) | 0 |
| ALS | 2 (0.05) | 2 (0.07) | 0 |
| Thrombosis | 2 (0.05) | 1 (0.03) | 1 (0.07) |
| Amyloidosis | 1 (0.02) | 1 (0.03) | 0 |
| Brain aneurysm | 1 (0.02) | 0 | 1 (0.07) |
| COPD | 1 (0.02) | 1 (0.03) | 0 |
| Cervical cancer | 1 (0.02) | 0 | 1 (0.07) |
| Diabetes | 1 (0.02) | 0 | 1 (0.07) |
| Kidney cancer | 1 (0.02) | 0 | 1 (0.07) |
| Lymphoma | 1 (0.02) | 1 (0.03) | 0 |
| Mesothelioma | 1 (0.02) | 1 (0.03) | 0 |
| Myelodysplastic syndrome | 1 (0.02) | 1 (0.03) | 0 |
| Sepsis | 1 (0.02) | 1 (0.03) | 0 |
| Complications of surgery | 1 (0.02) | 1 (0.03) | 0 |
| Suicide | 1 (0.02) | 0 | 1 (0.07) |
| Thyroid cancer | 1 (0.02) | 0 | 1 (0.07) |
| Vulvar cancer | 1 (0.02) | 0 | 1 (0.07) |
| Kidney failure | 1 (0.02) | 0 | 1 (0.07) |
| Other cause | 3 (0.07) | 3 (0.10) | 0 |
| Death cause unknown | 18 (0.42) | 9 (0.31) | 9 (0.64) |
| Total | 228 (5.26) | 112 (3.82) | 116 (8.29) |
Abbreviations: ALS, amyotrophic lateral sclerosis; COPD, chronic obstructive pulmonary disease.
Cardiovascular disease includes myocardial infarction, stroke, and circulatory failure.
A total of 2932 women underwent a preventive oophorectomy: 2106 of the 3177 women with BRCA1 sequence variations (66.3%) and 826 of the 1155 women with BRCA2 sequence variations (71.5%). We included women who had an oophorectomy prior to or following study enrollment in the latter group; the mean time from baseline to surgery was 4.3 years. Among the 2932 women who had an oophorectomy, 112 died (3.8%). Among the 1400 women who did not have an oophorectomy, 116 died (8.3%). The age-adjusted HR for all-cause mortality associated with oophorectomy (time dependent) was 0.32 (95% CI, 0.24-0.42; P < .001) (Table 4). The age-adjusted HR was 0.28 (95% CI, 0.20-0.38; P < .001) for women with BRCA1 sequence variations and 0.43 (95% CI, 0.22-0.90; P = .03) for women with BRCA2 sequence variations. After adjustment, the HR for all-cause mortality associated with oophorectomy was 0.31 (95% CI, 0.23-0.44; P < .001).
Table 4. Hazard Ratios (HRs) for Oophorectomy and All-Cause Mortality, by Timing of Oophorectomy and Age at Oophorectomy and by BRCA Sequence Variation.
| Variable | Alive/dead, No. | Basic model, HR (95% CI)a | P value | Multivariate HR (95% CI)b | P value |
|---|---|---|---|---|---|
| Women with BRCA1 and BRCA2 sequence variations | |||||
| Oophorectomy | |||||
| No | 1284/116 | 1 [Reference] | NA | 1 [Reference] | NA |
| Yes | 2820/112 | 0.32 (0.24-0.42) | <.001 | 0.36 (0.27-0.49) | <.001 |
| Timing of oophorectomy | |||||
| Before baseline | 808/43 | 0.38 (0.21-0.56) | <.001 | 0.47 (0.31-0.70) | <.001 |
| Within 1 y of baseline | 1089/38 | 0.27 (0.19-0.40) | <.001 | 0.31 (0.21-0.46) | <.001 |
| >1 y After baseline | 923/31 | 0.33 (0.22-0.49) | <.001 | 0.36 (0.24-0.54) | <.001 |
| Age at oophorectomy, y | |||||
| <40 | 704/12 | 0.26 (0.14-0.47) | <.001 | 0.33 (0.18-0.62) | <.001 |
| 40-50 | 1405/58 | 0.36 (0.26-0.50) | <.001 | 0.42 (0.30-0.59) | <.001 |
| >50 | 743/42 | 0.29 (0.19-0.43) | <.001 | 0.30 (0.20-0.45) | <.001 |
| Women with BRCA1 sequence variation | |||||
| Oophorectomy | |||||
| No | 967/103 | 1 [Reference] | NA | 1 [Reference] | NA |
| Yes | 2104/92 | 0.28 (0.20-0.38) | <.001 | 0.32 (0.23-0.44) | <.001 |
| Timing of oophorectomy | |||||
| Before baseline | 552/36 | 0.31 (0.20-0.48) | <.001 | 0.40 (0.25-0.64) | <.001 |
| Within 1 y of baseline | 756/29 | 0.22 (0.14-0.34) | <.001 | 0.26 (0.17-0.40) | <.001 |
| >1 y After baseline | 706/27 | 0.32 (0.21-0.49) | <.001 | 0.34 (0.22-0.53) | <.001 |
| Age at oophorectomy, y | |||||
| <40 | 574/11 | 0.22 (0.11-0.41) | <.001 | 0.30 (0.16-0.59) | <.001 |
| 40-50 | 982/48 | 0.32 (0.22-0.45) | <.001 | 0.38 (0.26-0.55) | <.001 |
| >50 | 458/33 | 0.25 (0.16-0.39) | <.001 | 0.26 (0.16-0.40) | <.001 |
| Women with BRCA2 sequence variation | |||||
| Oophorectomy | |||||
| No | 316/13 | 1 [Reference] | NA | 1 [Reference] | NA |
| Yes | 806/20 | 0.43 (0.21-0.90) | .03 | 0.44 (0.20-0.96) | .04 |
| Timing of oophorectomy | |||||
| Before baseline | 256/7 | 0.46 (0.17-1.22) | .12 | 0.48 (0.17-1.37) | .17 |
| Within 1 y of baseline | 333/9 | 0.47 (0.20-1.14) | .10 | 0.49 (0.20-1.23) | .13 |
| >1 y After baseline | 217/4 | 0.35 (0.11-1.10) | .07 | 0.34 (0.11-1.10) | .07 |
| Age at oophorectomy, y | |||||
| <40 | 125/1 | 0.27 (0.04-2.13) | .22 | 0.25 (0.03-2.04) | .20 |
| 40-50 | 403/10 | 0.51 (0.22-1.78) | .11 | 0.51 (0.21-1.23) | .13 |
| >50 | 278/9 | 0.39 (0.15-0.99) | .05 | 0.41 (0.15-1.10) | .08 |
Abbreviation: NA, not applicable.
Basic model adjusted for age at baseline and country of residence.
Multivariable model adjusted for age at baseline and country of residence, smoking, oral contraceptives, bilateral mastectomy, hormone replacement therapy, and parity. Oophorectomy included bilateral surgery only (with or without salpingectomy) and was included as a time-dependent exposure.
Among women who had an oophorectomy prior to baseline, the age-adjusted HR for all-cause mortality was 0.38 (95% CI, 0.21-0.56; P < .001). Among women who had an oophorectomy the same year as the baseline or thereafter, the HR for all-cause mortality to age 75 years was 0.33 (95% CI, 0.22-0.49; P < .001). The magnitude of the risk reduction was similar irrespective of age at surgery (Table 4).
The age-adjusted HR associated with bilateral oophorectomy was 0.19 (95% CI, 0.12-0.33; P < .001) for ovarian, fallopian tube, or peritoneal cancer mortality. The age-adjusted HR for death from any cancer to age 75 years associated with oophorectomy was 0.31 (95% CI, 0.22-0.42; P < .001).
There were 582 incident breast cancers in the follow-up period. The HR for breast cancer incidence associated with oophorectomy was 0.72 (95% CI, 0.60-0.88; P < .001). The HR was 0.79 (95% CI, 0.63-0.99; P = .04) for women with BRCA1 sequence variations and 0.55 (95% CI, 0.38-0.81; P = .003) for women with BRCA2 sequence variations. Among these patients with breast cancer, 58 (10%) died in the follow-up. The overall HR for death from breast cancer associated with oophorectomy was 0.44 (95% CI, 0.25-0.78; P = .005) and was 0.50 (95% CI, 0.27-0.95; P = .03) for women with BRCA1 sequence variations and 0.22 (95% CI, 0.06-0.84; P = .03) for women with BRCA2 sequence variations. Among women with a BRCA1 sequence variation, the 15-year cumulative mortality from breast cancer postoophorectomy was 2.6% (there were too few deaths among women with BRCA2 sequence variations to estimate this). There were 1138 women who opted to have a bilateral preventive mastectomy; none of these women died of breast cancer.
Only 918 of the 2932 women who had an oophorectomy in the current cohort (31.3%) had an oophorectomy by the recommended age (age 40 years for BRCA1 and age 45 years for BRCA2). Among the 585 women with BRCA1 sequence variations who had an oophorectomy prior to age 40 years, there were 3 occult cancers (0.5%) diagnosed (compared with 41 occult cancers [2.7%] diagnosed in the 1521 women who had an oophorectomy after age 40 years), and 11 deaths occurred: 4 from breast cancer, 1 from an (occult) ovarian cancer, 2 from primary peritoneal cancer, 1 from leukemia, 1 from lung cancer, 1 from chronic obstructive pulmonary disease, and 1 with an unknown cause of death. Among the 333 women with BRCA2 sequence variations who had an oophorectomy prior to age 45 years, there were no occult cancers diagnosed (compared with 2 occult cancers [0.4%] diagnosed in the 493 women who had an oophorectomy after age 45 years), and there were 7 deaths: 2 from breast cancer, 2 from primary peritoneal cancer, 1 from lung cancer, 1 from surgical complications, and 1 with an unknown cause of death. Among the women with BRCA1 sequence variations, 1538 (48.4%) received their test result after age 40 years, and among women with BRCA2 sequence variations, 546 (47.3%) received their test result after age 45 years. Among the 2699 women who did not have an oophorectomy by age 45 years, 268 (9.3%) had a child after age 35 years.
The annual all-cause mortality rates by age (in 10-year intervals) according to oophorectomy status are presented in Table 5. Based on the age-specific rates, we estimated the all-cause cumulative mortality to age 75 years for women with a BRCA1 sequence variation and who had oophorectomy at age 35 years to be 25%, vs 62% for women who never had an oophorectomy (eFigure 2 in Supplement 1). We estimated the all-cause cumulative mortality to age 75 years for women with a BRCA2 sequence variation and who had oophorectomy at age 35 years to be 14%, vs 28% for women who never had an oophorectomy (eFigure 3 in Supplement 1).
Table 5. Annual Mortality Rates of Women With BRCA1 and BRCA2 Sequence Variations (All-Cause) by Oophorectomy Status.
| Age group, y | No oophorectomy (BRCA1, n = 1071; BRCA2, n = 329) | Oophorectomy (BRCA1, n = 2106; BRCA2, n = 826) | P valuea | |||||
|---|---|---|---|---|---|---|---|---|
| Person-years | Deaths, No. | Annual rate, %b | Person-years | Deaths, No. | Annual rate, %b | |||
| BRCA1 | ||||||||
| 35-44.9 | 6370.3 | 25 | 0.39 | 3768.5 | 7 | 0.19 | .07 | |
| 45-54.9 | 2248.8 | 40 | 1.78 | 6694.3 | 34 | 0.51 | <.001 | |
| 55-64.9 | 791.0 | 28 | 3.54 | 4045.1 | 35 | 0.87 | <.001 | |
| 65-74.9 | 276.0 | 10 | 3.62 | 1249.0 | 16 | 1.28 | .07 | |
| All | 9686.2 | 103 | 1.06 | 15 756.9 | 92 | 0.58 | NA | |
| BRCA2 | ||||||||
| 35-44.9 | 1669.4 | 3 | 0.18 | 887.7 | 0 | 0.00 | .56 | |
| 45-54.9 | 906.2 | 1 | 0.11 | 2253.6 | 7 | 0.31 | .45 | |
| 55-64.9 | 441.5 | 6 | 1.36 | 1746.3 | 8 | 0.46 | .05 | |
| 65-74.9 | 203.7 | 3 | 1.47 | 723.9 | 5 | 0.69 | .38 | |
| All | 3220.8 | 13 | 0.40 | 5610.6 | 20 | 0.36 | NA | |
Abbreviation: NA, not applicable.
Fisher 2-sided test.
Annual rate = ratio of the number of events (deaths) and the number of person-years of risk accumulated in each age interval expressed as percentage.
Discussion
In this prospective study of 4332 women with a BRCA1 or BRCA2 sequence variation, we calculated age-specific annual mortality rates for women with and without intact ovaries. From these annual rates, we estimated that, among women with BRCA1 sequence variations, cumulative mortality from all causes to age 75 years fell from 62% to 25% for women who had their ovaries removed at age 35 years, compared to those who retained their ovaries. The reduction for women with a BRCA2 sequence variation was smaller but substantial (28% to 14%). These findings are consistent with our earlier report and those of others. The current study extends our prior study by including an additional 827 women with BRCA1 sequence variations and 709 women with BRCA2 sequence variations and by extending the follow-up time by 2 years. Importantly, we excluded women who had cancer diagnosed prior to baseline, allowing us to focus on women who were healthy at the time they discovered they carry a BRCA sequence variation.
In a prospective study of over 350 000 individuals in the UK who underwent whole-exome sequencing, protein-truncating variants in BRCA1 were associated with a significant decrease in a woman’s life span. For those who know their sequence variation status, the opportunity to have an oophorectomy can help mitigate this effect. The impact on mortality was primarily attributed to a reduction in deaths from ovarian and fallopian tube cancer; however, there was also a contribution from reducing deaths from breast cancer.
Our findings contrast with those for women in the general population. For women at average cancer risk, premenopausal oophorectomy is associated with an increased risk of death from cardiovascular disease as well as a decline in quality of life. We also see a decline in quality of life in women with BRCA sequence variations with oophorectomy prior to menopause, but this is offset by the prolongation of life expectancy. The members of the current cohort were relatively young (mean age at end of follow-up, 51.7 years), and thus we had limited power to estimate with accuracy the association of oophorectomy with other chronic conditions. Nevertheless, for women between the ages of 55 and 75 years, the all-cause mortality rate was lower in women without ovaries than in those with ovaries (Table 5), suggesting that the benefit of oophorectomy is retained in older women.
Only 31.3% of women who had an oophorectomy in the current cohort had an oophorectomy by the recommended age (age 40 years for BRCA1 and 45 years for BRCA2). It is not clear why some women in this study chose not to have an oophorectomy, and we do not have information in this regard. All were aware of their sequence variation status; however, many women did not receive their genetic test result until after the recommended age for oophorectomy. Among the women with BRCA1 sequence variations, 48.4% received their test result after age 40 years, and among women with BRCA2 sequence variations, 47.3% received their test result after age 45 years.
Women with an oophorectomy were also more likely to undergo a risk-reducing mastectomy and to take HRT, but a secondary analysis adjusting for these exposures gave similar results. One cannot rule out the possibility that unmeasured confounders, such as alcohol use and exercise, might contribute to differences in life expectancy.
The women were from 16 different countries and had genetic testing over an extended period of time (1994-2019). It is possible that not all these women were counseled about the risks and benefits of oophorectomy or the safety of HRT postsurgery to alleviate menopausal symptoms. There may be some who were reliant on screening with CA 125 testing and transvaginal ultrasonography; however, screening has not been shown to be effective in terms of reducing mortality. Some may have avoided oophorectomy to maintain their fertility; however, among the 2699 women who did not have an oophorectomy by age 45 years, only 268 (9.3%) had a child after age 35 years. Misclassification exposure is unlikely, as self-reported oophorectomy has previously been shown to be highly reproducible and valid. We were missing cause of death for 18 participants (7.9%) who died, and we did not have information on the presence of premalignant lesions in the ovaries of women who underwent preventive surgery.
The prevalence of occult ovarian and fallopian tube cancers identified at the time of oophorectomy varied with age at surgery. Among women with BRCA1 sequence variations, there were 3 occult cancers diagnosed in 585 women who had an oophorectomy before age 40 years (0.5%) vs 41 occult cancers diagnosed in 1521 women who had an oophorectomy after age 40 years (2.7%). Among the women with BRCA2 sequence variations, there were no occult cancers diagnosed in the 333 women who had an oophorectomy before age 45 years, compared to 2 occult cancers diagnosed in the 493 women who had an oophorectomy after age 45 years (0.4%). The HRs for all-cause mortality did not vary substantially by age at oophorectomy. The benefit of oophorectomy was present for women older than 50 years as well as for premenopausal women. This suggests that factors associated with oophorectomy other than estrogen and progesterone may affect mortality, but to date, these factors remain unknown and will be the subject of future studies. In an earlier study, we showed that those women who took estrogen-only hormonal therapy postoophorectomy did not experience any increase in the rate of breast cancer. In the current study, we saw a modest but significant effect on the incidence of breast cancer associated with oophorectomy in women with BRCA1 sequence variations (HR, 0.79; 95% CI, 0.63-0.99; P = .04). This result contrasts with that of our recent nested case-control study of oophorectomy and breast cancer incidence in women with BRCA1 sequence variations (odds ratio, 1.21; 95% CI, 0.87-1.70; P = .26), but the present study includes more incident cases (n = 448) than in the earlier study (n = 330). This will be addressed in future studies.
Limitations
This study has limitations. Causes of death were reported by the study center or the next of kin and were not all confirmed by medical records or death certificates. This was an observational study, and the choice of oophorectomy was not randomized but was determined by patient choice. We did not have information on the presence of precancerous lesions of the fallopian tube. The mean duration of follow-up was approximately 9 years; ideally, all participants would be followed up until age 75 years. This will be the topic of a future study.
Conclusions
To our knowledge, this cohort study represents the largest prospective cohort of oophorectomy and all-cause mortality in women without a cancer diagnosis with a BRCA sequence variation. The data on relative cancer rates, the estimated HRs, and the prevalence of occult cancers by age at oophorectomy support the current National Comprehensive Cancer Network guidelines for oophorectomy between ages 35 and 40 years for women with BRCA1 sequence variations and before age 45 years for women with BRCA2 sequence variations. We hope that the findings in this study will reassure women with a positive genetic test result who face high risks of breast and ovarian cancer.
eFigure 1. Overview of inclusions and exclusions
eFigure 2. All-cause cumulative mortality to age 75 for women with a BRCA1 mutation and who had oophorectomy at age 35 vs. oophorectomy
eFigure 3. All-cause cumulative mortality to age 75 for women with a BRCA2 mutation and who had oophorectomy at age 35 vs. oophorectomy
Hereditary Breast Cancer Clinical Study Group Nonauthor Collaborators
Data Sharing Statement
References
- 1.Daly MB, Pal T, Berry MP, et al. ; CGC; CGC; LCGC; CGC; CGC . Genetic/familial high-risk assessment: breast, ovarian, and pancreatic, version 2.2021, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2021;19(1):77-102. doi: 10.6004/jnccn.2021.0001 [DOI] [PubMed] [Google Scholar]
- 2.Finch AP, Lubinski J, Møller P, et al. Impact of oophorectomy on cancer incidence and mortality in women with a BRCA1 or BRCA2 mutation. J Clin Oncol. 2014;32(15):1547-1553. doi: 10.1200/JCO.2013.53.2820 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Giannakeas V, Sopik V, Shestopaloff K, et al. A model for estimating ovarian cancer risk: application for preventive oophorectomy. Gynecol Oncol. 2015;139(2):242-247. doi: 10.1016/j.ygyno.2015.08.020 [DOI] [PubMed] [Google Scholar]
- 4.Domchek SM, Friebel TM, Singer CF, et al. Association of risk-reducing surgery in BRCA1 or BRCA2 mutation carriers with cancer risk and mortality. JAMA. 2010;304(9):967-975. doi: 10.1001/jama.2010.1237 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Steenbeek MP, van Bommel MHD, Bulten J, et al. Risk of peritoneal carcinomatosis after risk-reducing salpingo-oophorectomy: a systematic review and individual patient data meta-analysis. J Clin Oncol. 2022;40(17):1879-1891. doi: 10.1200/JCO.21.02016 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Stanciu PI, Ind TEJ, Barton DPJ, et al. Development of peritoneal carcinoma in women diagnosed with serous tubal intraepithelial carcinoma (STIC) following risk-reducing salpingo-oophorectomy (RRSO). J Ovarian Res. 2019;12(1):50. doi: 10.1186/s13048-019-0525-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Soong TR, Howitt BE, Horowitz N, Nucci MR, Crum CP. The fallopian tube, “precursor escape” and narrowing the knowledge gap to the origins of high-grade serous carcinoma. Gynecol Oncol. 2019;152(2):426-433. doi: 10.1016/j.ygyno.2018.11.033 [DOI] [PubMed] [Google Scholar]
- 8.Rocca WA, Mielke MM, Gazzuola Rocca L, Stewart EA. Premature or early bilateral oophorectomy: a 2021 update. Climacteric. 2021;24(5):466-473. doi: 10.1080/13697137.2021.1893686 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hall E, Finch A, Jacobson M, et al. Effects of bilateral salpingo-oophorectomy on menopausal symptoms and sexual functioning among women with a BRCA1 or BRCA2 mutation. Gynecol Oncol. 2019;152(1):145-150. doi: 10.1016/j.ygyno.2018.10.040 [DOI] [PubMed] [Google Scholar]
- 10.Kotsopoulos J, Gronwald J, Lubinski J, et al. ; Hereditary Breast Cancer Clinical Study Group . Does preventive oophorectomy increase the risk of depression in BRCA mutation carriers? Menopause. 2020;27(2):156-161. doi: 10.1097/GME.0000000000001437 [DOI] [PubMed] [Google Scholar]
- 11.Cusimano MC, Chiu M, Ferguson SE, et al. Association of bilateral salpingo-oophorectomy with all cause and cause specific mortality: population based cohort study. BMJ. 2021;375:e067528. doi: 10.1136/bmj-2021-067528 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Rocca WA, Gazzuola-Rocca L, Smith CY, et al. Accelerated accumulation of multimorbidity after bilateral oophorectomy: a population-based cohort study. Mayo Clin Proc. 2016;91(11):1577-1589. doi: 10.1016/j.mayocp.2016.08.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kotsopoulos J, Lubinski J, Gronwald J, et al. Bilateral oophorectomy and the risk of breast cancer in BRCA1 mutation carriers: a reappraisal. Cancer Epidemiol Biomarkers Prev. 2022;31(7):1351-1358. doi: 10.1158/1055-9965.EPI-21-1196 [DOI] [PubMed] [Google Scholar]
- 14.Mejia-Gomez J, Gronwald J, Senter L, et al. ; and the Hereditary Breast Cancer Clinical Study Group . Factors associated with use of hormone therapy after preventive oophorectomy in BRCA mutation carriers. Menopause. 2020;27(12):1396-1402. doi: 10.1097/GME.0000000000001629 [DOI] [PubMed] [Google Scholar]
- 15.Kotsopoulos J, Lubinski J, Lynch HT, et al. ; Hereditary Breast Cancer Clinical Study Group . Oophorectomy and risk of contralateral breast cancer among BRCA1 and BRCA2 mutation carriers. Breast Cancer Res Treat. 2019;175(2):443-449. doi: 10.1007/s10549-019-05162-7 [DOI] [PubMed] [Google Scholar]
- 16.Kotsopoulos J, Gronwald J, Karlan B, et al. ; Hereditary Ovarian Cancer Clinical Study Group . Age-specific ovarian cancer risks among women with a BRCA1 or BRCA2 mutation. Gynecol Oncol. 2018;150(1):85-91. doi: 10.1016/j.ygyno.2018.05.011 [DOI] [PubMed] [Google Scholar]
- 17.Domchek SM, Friebel TM, Neuhausen SL, et al. Mortality after bilateral salpingo-oophorectomy in BRCA1 and BRCA2 mutation carriers: a prospective cohort study. Lancet Oncol. 2006;7(3):223-229. doi: 10.1016/S1470-2045(06)70585-X [DOI] [PubMed] [Google Scholar]
- 18.Ingham SL, Sperrin M, Baildam A, et al. Risk-reducing surgery increases survival in BRCA1/2 mutation carriers unaffected at time of family referral. Breast Cancer Res Treat. 2013;142(3):611-618. doi: 10.1007/s10549-013-2765-x [DOI] [PubMed] [Google Scholar]
- 19.Eleje GU, Eke AC, Ezebialu IU, Ikechebelu JI, Ugwu EO, Okonkwo OO. Risk-reducing bilateral salpingo-oophorectomy in women with BRCA1 or BRCA2 mutations. Cochrane Database Syst Rev. 2018;8(8):CD012464. doi: 10.1002/14651858.CD012464.pub2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Liu JZCCY, Chen CY, Tsai EA, et al. The burden of rare protein-truncating genetic variants on human lifespan. Nat Aging. 2022;2(4):289-294. doi: 10.1038/s43587-022-00182-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Gazzuola Rocca L, Smith CY, Bobo WV, et al. Mental health conditions diagnosed before bilateral oophorectomy: a population-based case-control study. Menopause. 2019;26(12):1395-1404. doi: 10.1097/GME.0000000000001413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Rocca WA, Grossardt BR, Geda YE, et al. Long-term risk of depressive and anxiety symptoms after early bilateral oophorectomy. Menopause. 2018;25(11):1275-1285. doi: 10.1097/GME.0000000000001229 [DOI] [PubMed] [Google Scholar]
- 23.Rocca WA, Gazzuola Rocca L, Smith CY, et al. Bilateral oophorectomy and accelerated aging: cause or effect? J Gerontol A Biol Sci Med Sci. 2017;72(9):1213-1217. doi: 10.1093/gerona/glx026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Menon U, Gentry-Maharaj A, Burnell M, et al. Ovarian cancer population screening and mortality after long-term follow-up in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS): a randomised controlled trial. Lancet. 2021;397(10290):2182-2193. doi: 10.1016/S0140-6736(21)00731-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gronwald J, Lubinski J, Huzarski T, et al. A comparison of ovarian cancer mortality in women with BRCA1 mutations undergoing annual ultrasound screening or preventive oophorectomy. Gynecol Oncol. 2019;155(2):270-274. doi: 10.1016/j.ygyno.2019.08.034 [DOI] [PubMed] [Google Scholar]
- 26.Colditz GA, Stampfer MJ, Willett WC, et al. Reproducibility and validity of self-reported menopausal status in a prospective cohort study. Am J Epidemiol. 1987;126(2):319-325. doi: 10.1093/aje/126.2.319 [DOI] [PubMed] [Google Scholar]
- 27.Kotsopoulos J, Gronwald J, Karlan BY, et al. ; Hereditary Breast Cancer Clinical Study Group . Hormone replacement therapy after oophorectomy and breast cancer risk among BRCA1 mutation carriers. JAMA Oncol. 2018;4(8):1059-1065. doi: 10.1001/jamaoncol.2018.0211 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure 1. Overview of inclusions and exclusions
eFigure 2. All-cause cumulative mortality to age 75 for women with a BRCA1 mutation and who had oophorectomy at age 35 vs. oophorectomy
eFigure 3. All-cause cumulative mortality to age 75 for women with a BRCA2 mutation and who had oophorectomy at age 35 vs. oophorectomy
Hereditary Breast Cancer Clinical Study Group Nonauthor Collaborators
Data Sharing Statement