Abstract
Background
To estimate the incidence of primary peritoneal cancer after preventive bilateral oophorectomy in women with a BRCA1 or BRCA2 mutation.
Methods
A total of 6310 women with a BRCA1 or BRCA2 mutation who underwent a preventive bilateral oophorectomy were followed for a mean of 7.8 years from oophorectomy. The 20-year cumulative incidence of peritoneal cancer post-oophorectomy was estimated using the Kaplan-Meier method. A left-truncated Cox proportional hazards analysis was used to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) associated with the age at oophorectomy, year of oophorectomy, and family history of ovarian cancer as well as hormonal and reproductive risk factors.
Results
Fifty-five women developed primary peritoneal cancer (n = 45 in BRCA1, 8 in BRCA2, and 2 in women with a mutation in both genes). Their mean age at oophorectomy was 48.9 years. The annual risk of peritoneal cancer was 0.14% for women with a BRCA1 mutation and 0.06% for women with a BRCA2 mutation. The 20-year cumulative risk of peritoneal cancer from the date of oophorectomy was 2.7% for BRCA1 carriers and 0.9% for BRCA2 mutation carriers. There were no peritoneal cancers in BRCA1 carriers who had the operation before age 35 or in BRCA2 carriers who had the operation before age 45.
Conclusions
For BRCA1 mutation carriers, the annual risk of peritoneal cancer for 20 years post-oophorectomy is 0.14% per year. The risk is lower for BRCA2 carriers (0.06% per year).
Preventive salpingo-oophorectomy is offered to women who carry a deleterious mutation (pathogenic variant) in BRCA1 or BRCA2 with the goal of preventing ovarian (and fallopian tube) cancer. In the absence of preventive surgery, the lifetime risk of ovarian cancer is estimated to be 40% for BRCA1 mutation carriers and 20% for BRCA2 mutation carriers (1,2). The operation is associated with a significant reduction in ovarian cancer risk (3) and all-cause mortality (4). However, after surgery a significant residual risk of primary peritoneal cancer persists late into adulthood. This was first described before the era of genetic testing among women who underwent preventive oophorectomy because of a strong family history of ovarian cancer (5,6).
It has been proposed that high-grade serous ovarian cancers in general and among BRCA1 and BRCA2 carriers originate in the fallopian tubes. In this model, cancer (or precancer) cells may migrate from the tubes to the ovarian and/or peritoneal epithelium and are the source of peritoneal cancers (7,8). For this reason, salpingectomy is now routinely recommended at the time of oophorectomy. An alternate theory is that peritoneal cancers originate in the peritoneal endothelium (possibly in Mullerian ductal remnants) and that malignant transformation occurs there (9). If the tubal origin hypothesis is correct, we would expect that salpingo-oophorectomies done at a young age (ie, before age 45) should offer greater protection against peritoneal cancer than oophorectomies done later in life. Conversely, if primary peritoneal cancers are of peritoneal origin, we might expect the risk of peritoneal cancer to be independent of the age at surgery, and the incidence rate of peritoneal cancer would resemble the incidence pattern of ovarian cancer, peaking late in adulthood.
To provide an accurate estimate of the risk of primary peritoneal cancer, we followed a cohort of 6310 women enrolled in an international prospective study of female BRCA1 or BRCA2 mutation carriers who underwent bilateral oophorectomy for cancer prevention and documented new cases of primary peritoneal cancer post-oophorectomy. We included as endpoints all peritoneal or ovarian cancers that were recorded in the years after the year of oophorectomy, and we estimated the annual risks of peritoneal cancer by year of oophorectomy, age at oophorectomy, attained age, and time since oophorectomy.
Materials and methods
Study population
Eligible study participants included women enrolled in a multicenter, longitudinal study of BRCA1 and BRCA2 mutation carriers from 83 participating centers in 12 countries. These women had sought genetic testing for a BRCA mutation because of a personal or family history of breast and/or ovarian cancer or were enrolled in a research study. Mutation detection was performed using a range of techniques, but all nucleotide sequences were confirmed by direct DNA sequencing. The current study included those women with a deleterious mutation in BRCA1 or BRCA2 who had a bilateral oophorectomy before study enrollment or during the follow-up period. The study was approved by the institutional ethics review boards of the respective host institutions, and all study participants provided written informed consent. The procedures were followed in accordance with the ethical standards of the Helsinki Declaration.
Exposure data
Each study participant completed a baseline questionnaire at the time of a clinic appointment or entry in the research study or at their home at a later date. Follow-up questionnaires were completed every 2 years thereafter to update information. These were either mailed to the participant to complete and return or administered over the phone by a genetic counselor or research assistant. The questionnaires requested information on family and personal history of cancer, as well reproductive and medical histories, including preventive oophorectomy (removal of both ovaries), tubal ligation, and hysterectomy. Salpingectomy without oophorectomy was recorded since 2005. Information regarding exogenous hormone use (ie, hormone replacement therapy [HRT], oral contraceptives) was also collected. We only considered oral contraceptive use before oophorectomy and HRT use after oophorectomy (yes/no). If the woman had an oophorectomy and a hysterectomy, we assumed that her fallopian tubes were also removed.
Inclusions/exclusions
In total, there were 18 343 participants in the BRCA mutation carrier database. Participants who had an oophorectomy before the baseline questionnaire or in the follow-up period and who completed at least 1 follow-up questionnaire were eligible. Participants were not eligible for inclusion if they had a diagnosis of ovarian, fallopian, endometrial, or peritoneal cancer before or in the same year as their oophorectomy or before the completion of their baseline questionnaire (ie, at study enrollment). We excluded 93 participants who had an (occult) carcinoma of the ovary or tubes diagnosed at the time of preventive oophorectomy. If the peritoneal cancer was diagnosed in the same year as the year of oophorectomy, the participant was excluded. (These exclusions were done because we were hesitant to diagnose a peritoneal cancer in women with intact ovaries.) We excluded women who had an oophorectomy before age 25 (n = 28) or after age 70 (n = 88) and women who had an oophorectomy before 1980 (n = 127). After these exclusions, 6310 participants who underwent oophorectomy were available for the analysis, including 4440 women with a BRCA1 mutation, 1870 women with a BRCA2 mutation, and 23 women with a mutation in both genes; the latter women were included with the BRCA1 mutation carriers.
Endpoints
For the present analysis, cases of incident peritoneal cancer that occurred a minimum of 6 months after the date of the oophorectomy were included. We counted all cases of “peritoneal” or “ovarian” or “omental” cancer that occurred in the years after the year of oophorectomy as peritoneal cancer. Three women died of peritoneal cancer but had no previous record of a diagnosis of peritoneal cancer. These were also included, and the date of death was taken as the date of diagnosis.
Statistical analysis
Participants were followed from the date of oophorectomy or the date of baseline questionnaire (whichever came last) until either the date of peritoneal cancer, the date of death, or the date of completion of their last follow-up questionnaire. The cumulative incidence of peritoneal cancer was estimated using Kaplan-Meier survival analysis. Participants for whom the oophorectomy was done before the baseline were left-censored to the date of baseline questionnaire (ie, study enrollment). We used Cox proportional hazards regression to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for peritoneal cancer associated with oral contraceptive use (yes/no), breastfeeding (none, <1 year, 1 year), HRT use (yes/no), hysterectomy (yes/no), tubal ligation (yes/no), and the number of first- or second-degree relatives with ovarian cancer. The multivariate model was adjusted for age at oophorectomy, year of oophorectomy (<2005; 2005), breast cancer (time-dependent covariate), and country of residence (Poland, Canada, United States, other).
All analyses were conducted using the SAS statistical package, version 9.4 software (SAS Institute, Cary, NC, USA). All P values were 2-sided and were considered statistically significant if the P value was .05 or less.
Results
A total of 6310 BRCA1 or BRCA2 mutation carriers who had a preventive bilateral oophorectomy were included in the study. Of these, 61% had the surgery before baseline and 39% had surgery in the follow-up period. Participant characteristics are summarized in Table 1. Fifty-five incident primary peritoneal cancers were diagnosed in the follow-up period (47 in BRCA1 carriers and 8 in BRCA2 carriers) (Table 2). The mean time elapsed from oophorectomy to peritoneal cancer was 7.6 years, but the range was wide (1-29 years). The mean annual risk of peritoneal cancer between age 35 and 75 was 0.14% for BRCA1 and 0.05% for BRCA2 mutation carriers. The 20-year cumulative risk of peritoneal cancer from the date of oophorectomy was 2.7% for BRCA1 and 0.9% for BRCA2 mutation carriers. There were 23 women with a mutation in both genes, and 2 of them developed peritoneal cancer.
Table 1.
Characteristics of the 6310 BRCA mutation carriers included in the study
| Variables | Total (N = 6310) n (%)/mean (range) |
BRCA1 (n = 4440)a | BRCA2 (n = 1870) |
|---|---|---|---|
| Year of birth | 1959.2 (1920-1989) | 1959.7 (1920-1989) | 1959.3 (1920-1989) |
| Years of follow-upb | 7.8 (0.1-25.8) | 8.0 (0.1-25.8) | 7.5 (0.2-24.8) |
| Oophorectomy | |||
| At baseline | 3854 (61.1) | 2663 (60.0) | 1191 (63.7) |
| At follow-up | 2456 (38.9) | 1777 (40.0) | 679 (36.3) |
| Age at oophorectomy | |||
| ≤40 years | 1560 (24.7) | 1224 (27.6) | 336 (18.0) |
| 41-50 | 2839 (45.0) | 2010 (45.3) | 829 (44.3) |
| 51+ | 1911 (30.3) | 1206 (27.2) | 705 (37.7) |
| Mean age (range) | 46.7 (25-69) | 46.0 (25-69) | 48.4 (26-69) |
| Year of oophorectomy | |||
| 1980-2000 | 1424 (22.6) | 1078 (24.3) | 346 (18.5) |
| 2001-2005 | 1598 (25.3) | 1143 (25.7) | 455 (24.3) |
| 2006-2010 | 1710 (27.1) | 1148 (25.9) | 562 (30.1) |
| 2011-2022 | 1578 (25.0) | 1071 (24.1) | 507 (27.1) |
| Mean year (range) | 2005.4 (1980-2022) | 2005.1 (1980-2022) | 2005.5 (1980-2022) |
| Breast cancer | |||
| No | 2748 (43.6) | 1904 (42.9) | 844 (45.1) |
| Yes | 3562 (56.5) | 2536 (57.1) | 1026 (54.8) |
| At baseline | 3070 | 2153 | 917 |
| At follow-up | 492 | 383 | 109 |
| Oral contraceptive use | |||
| Never | 1946 (31.3) | 1611 (36.7) | 335 (18.2) |
| Ever | 4278 (68.7) | 2774 (63.3) | 1504 (81.8) |
| Missing | 86 | 55 | 31 |
| HRT use | |||
| Never | 3967 (62.9) | 2782 (62.7) | 1185 (63.5) |
| Ever | 2339 (37.1) | 1656 (37.3) | 683 (36.5) |
| Missing | 4 | 2 | 2 |
| Breastfeeding | |||
| Never | 1769 (32.1) | 1195 (31.0) | 574 (34.8) |
| Yes, <1 year | 2276 (41.3) | 1662 (43.1) | 614 (37.2) |
| Yes, >1 year | 1465 (26.0) | 1001 (26.0) | 464 (28.1) |
| Missing | 800 | 582 | 218 |
| Hysterectomy | |||
| No | 2440 (40.4) | 1617 (38.3) | 823 (45.3) |
| Yes | 3597 (59.6) | 2603 (61.7) | 994 (54.7) |
| Missing | 273 | 220 | 53 |
| Tubal ligation | |||
| No | 4901 (84.0) | 3549 (86.5) | 1352 (78.2) |
| Yes | 931 (16.0) | 554 (13.5) | 377 (21.8) |
| Missing | 478 | 337 | 141 |
| Number of first-degree relatives with ovarian cancer | |||
| 0 | 3158 (72.2) | 2132 (68.3) | 1026 (81.6) |
| 1 | 1078 (24.6) | 861 (27.6) | 217 (17.2) |
| 2+ | 141 (3.2) | 127 (4.1) | 14 (1.1) |
| Missing | 1933 | 1320 | 613 |
| Country of residence | |||
| USA | 1983 (31.4) | 1200 (27.0) | 783 (41.9) |
| Canada | 1765 (28.0) | 948 (21.4) | 817 (43.7) |
| Poland | 1668 (26.4) | 1632 (36.8) | 36 (1.9) |
| Norway | 364 (5.8) | 304 (6.9) | 60 (3.2) |
| Israel | 138 (2.2) | 93 (2.1) | 45 (2.4) |
| Italy | 124 (2.0) | 66 (1.5) | 58 (3.1) |
| Austria | 114 (1.8) | 91 (2.1) | 23 (1.2) |
| Netherlands | 50 (0.8) | 41 (0.9) | 9 (0.5) |
| France | 32 (0.5) | 25 (0.6) | 7 (0.4) |
| Spain | 26 (0.4) | 10 (0.2) | 16 (0.9) |
| Bahamas | 23 (0.4) | 22 (0.5) | 1 (0.1) |
| China | 23 (0.4) | 8 (0.2) | 15 (0.8) |
Includes 23 women who had a mutation in both genes.
Follow-up started at the year of oophorectomy, and oophorectomy was left truncated at the year of baseline if the surgery is earlier.
Table 2.
Characteristics of the 55 cases of incident peritoneal cancer among BRCA mutation carriers
| Variables | Mean (range)/frequency (%) (N = 55) |
|---|---|
| Age of oophorectomy | 48.9 (37-67) |
| Year of oophorectomy | 2001.2 (1984-2014) |
| Age of peritoneal cancer | 56.0 (38-78) |
| Year of diagnosis of peritoneal cancer | 2008.8 (1996.9-2020.6) |
| Time from oophorectomy to peritoneal cancer (years) | 7.6 (1-29) |
| Ovarian cancer in first and second-degree relatives | |
| None | 26 (47.3%) |
| One | 19 (34.6%) |
| Two or more | 4 (7.3%) |
| Missing | 6 (10.9%) |
| Hysterectomy | |
| No | 16 (29.1%) |
| Yes, before dx | 38 (69.1%) |
| Missing | 1 (1.8%) |
| Salpingectomy | |
| Yes | 48 (87.3%) |
| No | 3 (5.5%) |
| Missing | 4 (7.3%) |
| Tubal ligationa | |
| No | 44 (80.4%) |
| Yes | 6 (10.9%) |
| Missing | 5 (9.1%) |
| BRCA mutations | |
| BRCA1 | 45 (81.8%) |
| BRCA2 | 8 (14.6%) |
| Both mutations | 2 (3.6%) |
| Country of residence | |
| Canada | 17 (30.9%) |
| USA | 17 (30.9%) |
| Poland | 12 (21.8%) |
| Norway | 5 (9.1%) |
| Austria | 2 (3.6%) |
| France | 1 (1.8%) |
| Israel | 1 (1.8%) |
Tubal ligation before salpingectomy.
The annual risks of peritoneal cancer based on current age, age at surgery, year of surgery, and calendar year at diagnosis are presented in Table 3. The annual risk of peritoneal cancer increased with age and decreased with recent year of oophorectomy (after 2010) and recent calendar year (after 2009). The risk of peritoneal cancer in BRCA1 mutation carriers was approximately twice that of BRCA2 mutation carriers, although there were very few cases among women with a BRCA2 mutation (n = 8) and the difference was not statistically significant.
Table 3.
Annual rates of peritoneal cancer according to current age, age at oophorectomy, years after oophorectomy, year of oophorectomy, and year of diagnosis (follow-up started at oophorectomy and was left truncated at year of baseline if it is earlier)
|
BRCA1
|
BRCA2
|
||||||||
|---|---|---|---|---|---|---|---|---|---|
| No. of peritoneal cancers | Total participants | Person-years | Annual rate | No. of peritoneal cancers | Total participants | Person-years | Annual rate | ||
| Current age | Current age | ||||||||
| 35-40 | 1 | – | 1800 | 0.06% | 35-40 | 0 | – | 467 | 0 |
| 41-50 | 11 | – | 10 536 | 0.10% | 41-50 | 1 | – | 3294 | 0.03% |
| 51-60 | 22 | – | 12 618 | 0.17% | 51-60 | 4 | – | 5035 | 0.08% |
| 61-75 | 12 | – | 7536 | 0.16% | 61-75 | 2 | – | 3944 | 0.05% |
| 35-75 | 46 | – | 32 490 | 0.14% | 35-75 | 7 | – | 12737 | 0.05% |
| Age at oophorectomy | Age at oophorectomy | ||||||||
| 25-35 | 0 | 357 | 2580 | 0.00% | 25-35 | 0 | 85 | 544 | 0.00% |
| 36-40 | 6 | 867 | 6142 | 0.10% | 36-40 | 0 | 251 | 1619 | 0.00% |
| 41-50 | 26 | 2010 | 15 588 | 0.17% | 41-50 | 4 | 829 | 5290 | 0.07% |
| 51-60 | 13 | 967 | 7293 | 0.18% | 51-60 | 3 | 529 | 3784 | 0.08% |
| 61-69 | 2 | 239 | 1657 | 0.12% | 61-69 | 1 | 176 | 1236 | 0.08% |
| Total (25-69) | 47 | 4440 | 33 253 | 0.14% | Total (25-69) | 8 | 1870 | 13092 | 0.06% |
| (46-69) | 30 | 2151 | 16 667 | 0.18% | (46-69) | 8 | 1149 | 8267 | 0.10% |
| Years after oophorectomy | Years after oophorectomy | ||||||||
| 0-4 | 16 | – | 14 463 | 0.11% | 0-4 | 4 | – | 5975 | 0.07% |
| 5-9 | 18 | – | 10 133 | 0.18% | 5-9 | 3 | – | 3957 | 0.05% |
| 10-14 | 8 | – | 5007 | 0.16% | 10-14 | 1 | – | 1873 | 0.08% |
| 15-19 | 3 | – | 2058 | 0.15% | 15-19 | 0 | – | 733 | 0.00% |
| Total (0-19) | 45 | – | 31 661 | 0.14% | Total (0-19) | 8 | – | 12538 | 0.06% |
| Year of oophorectomy | Year of oophorectomy | ||||||||
| ≤2005 | 36 | 2221 | 20 954 | 0.17% | ≤2005 | 6 | 801 | 7077 | 0.08% |
| 2006-2010 | 10 | 1148 | 7976 | 0.13% | 2006-2010 | 2 | 562 | 3918 | 0.05% |
| 2011-2022 | 1 | 1071 | 4323 | 0.02% | 2010-2022 | 0 | 507 | 2097 | 0.00% |
| Total | 47 | 4440 | 33 253 | 0.14% | Total | 8 | 1870 | 13092 | 0.06% |
| Year of diagnosis of peritoneal cancer | Year of diagnosis of peritoneal cancer | ||||||||
| 1995-2009 | 28 | – | 14 956 | 0.19% | 1995-2009 | 5 | – | 4801 | 0.10% |
| 2010-2014 | 9 | – | 10 393 | 0.09% | 2010-2014 | 2 | – | 4608 | 0.04% |
| After 2014 | 10 | – | 7904 | 0.13% | After 2015 | 1 | – | 3683 | 0.03% |
| Total | 47 | – | 33 253 | 0.14% | Total | 8 | – | 13092 | 0.06% |
| 1995-2009 | 28 | – | 14 956 | 0.19% | 1995-2009 | 5 | – | 4801 | 0.10% |
| After 2010 | 19 | – | 18 297 | 0.10% | After 2010 | 3 | – | 8291 | 0.04% |
The annual risk of peritoneal cancer for BRCA1 mutation carriers who had their oophorectomy at or before 2005 was 0.17%, compared with 0.13% for those who underwent surgery from 2006 to 2010 and 0.02% for women with surgery performed after 2010. Of the 3369 BRCA1 carriers who had an oophorectomy at or before 2010, the average age at the end of follow-up was 57.1 years. Of the 1071 BRCA1 carriers who had an oophorectomy after 2010, the average age at the end of follow-up was 49.9 years. There were too few women with BRCA2 mutations to do a robust analysis.
The univariate and multivariate HRs and 95% CIs for peritoneal cancer after oophorectomy according to age at oophorectomy and other risk factors are summarized in Table 4. A history of breast cancer was reported by 3562 of the participants (56.5%) and was associated with a lower risk of peritoneal cancer (HR = 0.47, 95% CI = 0.32 to 1.02; P = .01). The reduced risk was observed only for those patients who were previously treated with chemotherapy (HR = 0.41, 95% CI = 0.23 to 0.94; P = .01).
Table 4.
Hazard ratios for peritoneal cancer for all subjects, for selected variables
| Variables | Cases/total | Univariate RRc (95% CId) | P-value | Multivariatea RR (95% CI) | P-value |
|---|---|---|---|---|---|
| Age at oophorectomy | |||||
| ≤40 | 6/1560 | 1 | 1 | ||
| 41-50 | 30/2839 | 2.50 (1.03 to 6.06) | .04 | 2.83 (1.16 to 6.84) | .02 |
| 51+ | 19/1911 | 2.44 (0.96 to 6.18) | .06 | 3.18 (1.22 to 8.30) | .02 |
| Year of oophorectomy | |||||
| ≤2005 | 42/3022 | 1 | 1 | ||
| >2005 | 13/3288 | 0.48 (0.22 to 0.81) | .01 | 0.48 (0.25 to 0.94) | .03 |
| Breast cancerb | |||||
| No | 35/2748 | 1 | 1 | ||
| Yes | 20/3562 | 0.51 (0.30 to 0.89) | .02 | 0.47 (0.27 to 0.83) | .01 |
| Yes, no chemo | 7/717 | 0.82 (0.36 to 1.87) | .64 | 0.92 (0.31 to 1.66) | .44 |
| Yes, had chemo | 12/2491 | 0.44 (0.23 to 0.85) | .01 | 0.41 (0.21 to 0.80) | .01 |
| Yes, chemo missing | 1/354 | ||||
| Oral contraceptive use | |||||
| No | 18/1946 | 1 | 1 | ||
| Yes | 36/4278 | 0.87 (0.49 to 1.52) | .62 | 1.09 (0.60 to 1.98) | .78 |
| Missing | 1/86 | ||||
| Breastfeeding | |||||
| No | 17/1769 | 1 | 1 | ||
| ≤1 year | 21/2276 | 0.95 (0.50 to 1.80) | .87 | 0.93 (0.49 to 1.78) | .82 |
| >1 year | 10/1465 | 0.74 (0.34 to 1.62) | .45 | 0.75 (0.34 to 1.66) | .48 |
| Missing | 7/800 | ||||
| No. relatives with ovarian cancer | |||||
| 0 | 26/3158 | 1 | 1 | ||
| 1 | 19/1078 | 1.98 (1.10 to 3.58) | .02 | 1.55 (0.85 to 2.82) | .16 |
| 2+ | 4/141 | 3.02 (1.04 to 8.73) | .04 | 1.83 (0.62 to 5.38) | .28 |
| Missing | 6/1933 | ||||
| Hysterectomy | |||||
| No | 16/2440 | 1 | 1 | ||
| Yes | 38/3597 | 1.18 (0.65 to 2.13) | .59 | 1.11 (0.59 to 2.10) | .74 |
| Missing | 1/273 | ||||
| Tubal ligation | |||||
| No | 44/4901 | 1 | 1 | ||
| Yes | 6/931 | 0.60 (0.26 to 1.41) | .24 | 0.57 (0.24 to 1.36) | .20 |
| Missing | 5/478 | ||||
| Mutation | |||||
| BRCA1 | 47/4440 | 1 | 1 | ||
| BRCA2 | 8/1870 | 0.43 (0.20 to 0.95) | .03 | 0.45 (0.21 to 0.99) | .05 |
follow-up started at oophorectomy and left truncated at baseline.
breast cancer is a time-dependent variable.
RR: Relative Risk.
CI: Confidence Interval.
Past oral contraceptive use, breastfeeding, and tubal ligation were not associated with the risk of peritoneal cancer. There was a suggestion that tubal ligation was protective, but this was not statistically significant (HR = 0.57, 95% CI = 0.24 to 1.36; P = .20). In the crude analysis, women with a family history of ovarian cancer were at a higher risk of peritoneal cancer than those with no affected relative (unadjusted HR = 2.10, 95% CI = 1.20 to 3.69); however, this was not statistically significant after adjusting for other factors (HR = 1.59, 95% CI = 0.89 to 2.82; P = .12). The 20-year risk was 5.1% for BRCA1 mutation carriers with a family history of ovarian cancer in one or more first-degree relatives.
There were 3597 women who reported a hysterectomy at the time of oophorectomy (60%). Of the 55 women who developed peritoneal cancer, 38 had a prior hysterectomy and 16 had no prior hysterectomy. Of the 16 women with no prior hysterectomy, 9 women had their fallopian tubes removed before peritoneal cancer, 3 had fallopian tubes left in situ, and for 4 women this status was unknown. The risk of peritoneal cancer was not reduced with hysterectomy (HR = 1.11, 95% CI = 0.59 to 2.10; P = .74).
Discussion
In this prospective study of genetically predisposed women, we estimated the risk of peritoneal cancer after oophorectomy for women with a BRCA1 or BRCA2 mutation. The annual risk of peritoneal cancer was relatively constant with age and time elapsed since oophorectomy with no apparent early peak period of risk or decline with time since surgery. The risks were substantially lower when oophorectomy was performed at a relatively young age (<45 years) and when it was performed after 2005, after which time salpingectomy was universally recommended. The 20-year risk was 0.9% for BRCA2 mutation carriers, but only 8 incident cases of peritoneal cancer were detected among 1870 BRCA2 carriers, and the estimate for these women is less precise. The 20-year risk was 5.1% for BRCA1 mutation carriers with a family history of ovarian cancer in 1 or more first-degree relatives. Twenty-three of the 55 cases of peritoneal cancer had a recorded first- or second-degree relative with ovarian cancer.
We previously reported on this cohort in 2014 (3). At that time, 32 incident peritoneal cancers were diagnosed among 3513 oophorectomized women after a mean of 3.5 years of follow-up. In the earlier study, we estimated the annual risk of peritoneal cancer to be 0.20% for BRCA1 and 0.10% for BRCA2 mutation carriers. The current study now includes 55 incident cases of peritoneal cancer. With a larger study population and longer average follow-up and more recent follow-up, we now have revised these annual risk estimates downward (to 0.14% and 0.06%, respectively). To a large extent, this is because the annual risk of peritoneal cancer declined in recent years in carriers of either gene mutation. In both studies, the mean age at oophorectomy was about the same (47 years and 49 years).
Ninety-nine participants in the overall cohort had a carcinoma of the ovary or tubes diagnosed after the baseline questionnaire and were therefore excluded from the study (88 with BRCA1 and 11 with BRCA2). The median age of diagnosis was 50.6 years for BRCA1 carriers and 55.8 years for BRCA2 carriers. The majority of these were occult carcinomas. This number (99) exceeds the number of peritoneal cancers (n = 55) and emphasizes the need to recommend and conduct oophorectomy at the appropriate age. (Ideally, the routine practice of early oophorectomy for women known to carry a mutation will be sufficiently high that, in the future, the number of peritoneal cancers will outnumber the number of ovarian/fallopian cancers.)
Our study has several strengths. This was a large series of women, and all peritoneal cancers were diagnosed in the follow-up period. We excluded diagnoses of peritoneal cancer reported before or at oophorectomy because it is difficult to distinguish these from ovarian or fallopian cancers. We also excluded women who had occult disease (ovarian, fallopian, or peritoneal cancer) diagnosed at the time of oophorectomy for the same reason. We considered any report of ovarian or peritoneal cancer in the follow-up period post-oophorectomy to be peritoneal.
The study also has several limitations. We did not have details on the pathology findings in the tubes at the time of oophorectomy. We did not collect pathology data for benign or pre-invasive tubal lesions including serous tubal intra-epithelial carcinomas (STICs). However, none of the patients reported chemotherapy use at the year of oophorectomy. The practice of routine salpingo-oophorectomy became widespread after 2005. Of the women who developed peritoneal cancer, only 1 was documented to have her tubes intact at the time of peritoneal cancer. Although this is the largest prospective study of peritoneal cancer conducted to date, the number of events in specific subgroups was small, and the hazard ratio estimates are imprecise. The number of peritoneal cancers among BRCA2 carriers was small (n = 8).
Our data support the hypothesis that peritoneal cancers arise from cells disseminated from the fallopian tubes. In favor of the tubal hypothesis, we observed a sharp decrease in the incidence of peritoneal cancer for women who underwent their oophorectomy after 2005. There has been increased effort since 2005 to ensure that all preventive surgeries include the removal of both fallopian tubes and ovaries (ie, bilateral salpingo-oophorectomy) and to examine the tubal fimbria closely for the presence of STICs and small invasive cancers (SEE-FIM protocol) (10). We do not have data regarding how often this protocol was adhered to in this cohort. Further, we saw a lower annual risk in women who had an oophorectomy before age 45 than in those who had it after age 45. There were no peritoneal cancers diagnosed among 357 women with a BRCA1 mutation who had an oophorectomy before age 35. There were no cases of peritoneal cancer diagnosed among 722 BRCA2 carriers who had an oophorectomy before age 45. If the cancer developed de novo in the peritoneum, then we would not expect the age of oophorectomy or year of oophorectomy to be predictive. These data support the existing recommendations for early oophorectomy in women with a BRCA mutation.
The decline in the incidence of peritoneal cancer in recent years supports the premise that the dissemination of premalignant cells to the peritoneum might occur during surgery—if the dissemination occurred before surgery, we would not expect to see a sharp decline in incidence with more recent oophorectomies. Peritoneal cancers were diagnosed up to 20 years post-oophorectomy. This suggests that for some patients, there is a period of dormancy of precancerous cells in situ before progression to frank cancer. Amazu et al. (11) reported a case of peritoneal cancer in a BRCA1 carrier 15 years post-oophorectomy. Retrospective pathology review revealed occult fallopian tube cancer. This case supports the theory of a dormant period of peritoneal cancer before clinical detection.
A small percentage of fallopian tubes removed from carriers at preventive surgery are found to harbor STICs; these have the appearance of malignant lesions but have not breached the basement membrane. STICs are rare in the non-carrier population (12) but are a putative precursor to high-grade serous ovarian cancer and peritoneal cancer among mutation carriers. The prevalence of STICs in BRCA mutation carriers at the time of BSO has been estimated to be between 2.0% and 3.8% (13-18). In a series of 45 women with primary peritoneal cancer who had sampling of the fallopian tubes at the time of diagnosis, a STIC was identified in about 40% (7). In a recent report of oophorectomy patients, Steenbeek et al. (16) followed BRCA1 and BRCA2 carriers with and without a STIC found at surgery. There were 14 peritoneal cancers diagnosed among 115 women with a STIC (annual risk 2.8%), and there were 11 peritoneal cancers diagnosed among 2741 women without a STIC (annual risk 0.09%). The 10-year risk of peritoneal cancer was 27.5% for women with a STIC and 0.9% for women without a STIC. Powell et al. followed 17 women with high-grade intraepithelial neoplasms (14 in BRCA1 carriers, 3 in BRCA2) found incidentally at surgery (19). After a median follow-up of 80 months, there was 1 recurrence, and all 17 patients were alive at the end of the study. A systematic review by Linz et al. (20) estimated the 10-year risks of peritoneal cancer in women with STICS to be 21.6%. These data suggest that the majority of women who had peritoneal cancer had a prior STIC. Given the high risk of peritoneal cancer in women with STICs proposed by Steenbeek et al. and Linz et al., it is important that these estimates be confirmed in future studies. There is a need for large collaborative studies to follow women prospectively to estimate the annual incidence of invasive peritoneal cancer in women with and without STICs detected at oophorectomy. It would also be of interest to determine whether women with STICs benefit from adjuvant chemotherapy.
Our data are consistent with the model of Crum and colleagues (21), who suggest that some peritoneal cancers arise from STICs and some arise de novo or as a result of nonmalignant precursor cells that pass from the fallopian tube to the peritoneum early in life with malignant transformation occurring later (precursor escape). It is notable that we did not see a difference in the risk of peritoneal cancer associated with oral contraceptive use, parity, breastfeeding, or tubal ligation. These have all been shown to reduce the risk of ovarian cancer in women with BRCA1 mutations in previous studies of this cohort (22,23). This may be because of sample size limitations or a difference in the natural history of peritoneal and ovarian cancer. In summary, the results support the recommendation that BRCA1 mutation carriers have preventive salpingo-oophorectomy before age 40 and BRCA2 mutation carriers before age 45. The residual risk of peritoneal cancer is small and, fortunately, has declined over the past 2 decades. This is likely due to improved surgical techniques. The benefit of salpingo-oophorectomy is strong as it diminishes ovarian cancer mortality, breast cancer mortality, and all-cause mortality (3,4), and it should be encouraged for carriers of either mutation.
Acknowledgements
We are grateful for the contributions of the women who participated in this study, without whom this research would not be possible. We acknowledge the Hereditary Ovarian Cancer Clinical Study Group, study staff, students, and volunteers who assisted with data collection and data entry: Ellen MacDougall, Shana Kim, Clotilde Ngwa, Aiman Syeda, Anasua Kundu, Nurun Nahar, Abigail Sims, Alexandra Parco, Christine Zhu, Cindy Zhang, Elizabeth Hall, Lisa Asbroek, Izzar Linares, Shaelyn Laurie, Kamrun Urmi, Amina Mahmood, Mayra Gholizadeh, Nazia Awan, Neelam Dehal, Pooja Chaudhary, Pooja Patel, Yasmin Tehrani, Seetha Venkatewsaran, Seema Mehta, Jasdeep Brar, Marsela Supriadi, Jenani Anantharajah, Grace Li, Hannah Horvath, Laavanya Somasundaram, Anne Matip, Forough Armaghan, Mohamed Bekkouche, Yasaman Ghazi, Qadriy Naimi, Liao Jia, Li Quan, Martina Delle Marchette, Serena Negri, Cristina Dell’Oro, and Alessandra Inzoli.
Contributor Information
Steven A Narod, Women’s College Research Institute, Women’s College Hospital, Toronto, ON, Canada; Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada.
Jacek Gronwald, International Hereditary Cancer Center, Department of Genetics and Pathology, Pomeranian Medical University, Szczecin, Poland.
Beth Karlan, Department of Obstetrics and Gynecology, David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA, USA.
Pal Moller, Inherited Cancer Research Group, The Norwegian Radium Hospital, Department for Medical Genetics; and Department of Tumor Biology, Institute of Cancer Research, The Norwegian Radium Hospital; Oslo University Hospital, Oslo, Norway.
Tomasz Huzarski, International Hereditary Cancer Center, Department of Genetics and Pathology, Pomeranian Medical University, Szczecin, Poland.
Nadine Tung, Beth Israel Deaconess Medical Center, Boston, MA, USA.
Amber Aeilts, Division of Human Genetics, The Ohio State University Medical Center, Comprehensive Cancer Center, Columbus, OH, USA.
Andrea Eisen, Toronto-Sunnybrook Regional Cancer Center, Toronto, ON, Canada.
Susan Randall Armel, Princess Margaret Cancer Centre, University of Toronto, ON, Canada.
Christian F Singer, Department of Obstetrics and Gynecology and Comprehensive Cancer Center, Medical University of Vienna, Vienna, Austria.
William D Foulkes, Program in Cancer Genetics, Department of Oncology and Human Genetics, McGill University, Montréal, QC, Canada.
Susan L Neuhausen, Department of Population Sciences, Beckman Research Institute of City of Hope, Duarte, CA, USA.
Olufunmilayo Olopade, Department of Medicine and Human Genetics, University of Chicago, Chicago, IL, USA.
Tuya Pal, University Medical Center/Vanderbilt-Ingram Cancer Center, Vanderbilt University, Nashville, TN, USA.
Robert Fruscio, Department of Medicine and Surgery, University of Milan Bicocca, IRCCS San Gerardo, Monza, Italy.
Kelly Metcalfe, Women’s College Research Institute, Women’s College Hospital, Toronto, ON, Canada; Bloomberg School of Nursing, University of Toronto, Toronto, ON, Canada.
Rebecca Raj, Women’s College Research Institute, Women’s College Hospital, Toronto, ON, Canada.
Michelle Jacobson, Women’s College Research Institute, Women’s College Hospital, Toronto, ON, Canada.
Ping Sun, Women’s College Research Institute, Women’s College Hospital, Toronto, ON, Canada.
Jan Lubinski, International Hereditary Cancer Center, Department of Genetics and Pathology, Pomeranian Medical University, Szczecin, Poland.
Joanne Kotsopoulos, Women’s College Research Institute, Women’s College Hospital, Toronto, ON, Canada; Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada.
Data availability
Data are available from the corresponding author to researchers who meet the Ethics Committee criteria for access to confidential data. Available data are dependent on patient consent and legal constrictions from each contributing site.
Author contributions
Steven A. Narod, MD (Conceptualization; Formal analysis; Supervision; Writing—original draft; Writing—review & editing), Ping Sun, PhD (Formal analysis; Writing—review & editing), Michelle Jacobson, MD (Data curation; Writing—review & editing), Rebecca Raj, BSc (Formal analysis; Writing—review & editing), Kelly Metcalfe, PhD (Data curation; Writing—review & editing), Robert Fruscio, MD (Data curation; Writing—review & editing), Tuya Pal, MD (Data curation; Writing—review & editing), Olufunmilayo Olopade, MD (Data curation), Susan L. Neuhausen, MD (Data curation; Writing—review & editing), Jan Lubinski, MD (Data curation; Writing—review & editing), William Foulkes, MD (Data curation; Writing—review & editing), Susan R. Armel, MD (Data curation; Writing—review & editing), Andrea Eisen, MD (Data curation; Writing—review & editing), Amber Aeilts, MSc (Data curation; Writing—review & editing), Nadine Tung, MD (Data curation; Writing—review & editing), Tomasz Huzarski, MD (Data curation; Writing—review & editing), Pal Moller, MD (Data curation; Writing—review & editing), Beth Karlan, MD (Data curation; Writing—review & editing), Jacek Gronwald, MD (Data curation; Writing—review & editing), Christian F. Singer, MD (Data curation; Writing—review & editing), Joanne Kotsopoulos, PhD (Conceptualization; Methodology; Visualization; Writing—review & editing).
Funding
Steven A. Narod is the recipient of a Canada Research Chair (Tier I). Joanne Kotsopoulos is the recipient of a Canada Research Chair (Tier II). This work was supported by the Canadian Institutes of Health Research (FDN 154275), Canadian Cancer Society Research Institute grant (703058), and the Peter Gilgan Centre for Women’s Cancers at Women’s College Hospital, in partnership with the Canadian Cancer Society.
Conflicts of interest
The authors declare that they have no conflict of interest.
References
- 1. Kuchenbaecker KB, Hopper JL, Barnes DR, et al. ; BRCA1 and BRCA2 Cohort Consortium. Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation carriers. JAMA. 2017;317(23):2402-2416. [DOI] [PubMed] [Google Scholar]
- 2. 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] [PubMed] [Google Scholar]
- 3. Finch A, Beiner M, Lubinski J, et al. ; Hereditary Ovarian Cancer Clinical Study Group. Salpingo-oophorectomy and the risk of ovarian, fallopian tube, and peritoneal cancers in women with a BRCA1 or BRCA2 mutation. JAMA. 2006;296(2):185-192. [DOI] [PubMed] [Google Scholar]
- 4. Finch APM, 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] [PMC free article] [PubMed] [Google Scholar]
- 5. Tobacman JK, Greene MH, Tucker MA, Costa J, Kase R, Fraumeni JF Jr. Intra-abdominal carcinomatosis after prophylactic oophorectomy in ovarian-cancer-prone families. Lancet. 1982;2(8302):795-797. [DOI] [PubMed] [Google Scholar]
- 6. Piver MS, Jishi MF, Tsukada Y, Nava G.. Primary peritoneal carcinoma after prophylactic oophorectomy in women with a family history of ovarian cancer: a report of the Gilda Radner Familial Ovarian Cancer Registry. Cancer. 1993;71(9):2751-2755. [DOI] [PubMed] [Google Scholar]
- 7. Carlson JW, Miron A, Jarboe EA, et al. Serous tubal intraepithelial carcinoma: its potential role in primary peritoneal serous carcinoma and serous cancer prevention. J Clin Oncol. 2008;26(25):4160-4165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. 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. Epub 2018 Nov 30. PMID: 30503267. [DOI] [PubMed] [Google Scholar]
- 9. Dubeau L. The cell of origin of ovarian epithelial tumors and the ovarian surface epithelium dogma: does the emperor have no clothes? Gynecol Oncol. 1999;72(3):437-442. [DOI] [PubMed] [Google Scholar]
- 10. Medeiros F, Muto MG, Lee Y, et al. The tubal fimbria is a preferred site for early adenocarcinoma in women with familial ovarian cancer syndrome. Am J Surg Pathol. 2006;30(2):230-236. [DOI] [PubMed] [Google Scholar]
- 11. Amuzu S, Fu L, Demko N, et al. Long-term tumour dormancy in a BRCA1 heterozygote. J Med Genet. 2023;60(1):33-35. [DOI] [PubMed] [Google Scholar]
- 12. Samimi G, Trabert B, Geczik AM, Duggan MA, Sherman ME.. Population frequency of serous tubal intraepithelial carcinoma (STIC) in clinical practice using SEE-Fim protocol. JNCI Cancer Spectr. 2018;142(4):pky061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Abay M, Ozgen L, Yalcin Y, Ozerkan K.. Clinical significance of risk-reducing salpingo-oophorectomy in patients with BRCA1/2 mutation. J Gynecol Obstet Hum Reprod. 2023;52(8):102642. doi: 10.1016/j.jogoh.2023.102642. [DOI] [PubMed] [Google Scholar]
- 14. 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]
- 15. Saccardi C, Zovato S, Spagnol G, et al. Efficacy of risk-reducing salpingo-oophorectomy in BRCA1-2 variants and clinical outcomes of follow-up in patients with isolated serous tubal intraepithelial carcinoma (STIC). Gynecol Oncol. 2021;163(2):364-370. [DOI] [PubMed] [Google Scholar]
- 16. 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] [PMC free article] [PubMed] [Google Scholar]
- 17. Kang O-J, Lee S-W, Kim J-H, et al. Pathological findings and long-term prognosis in Korean BRCA1/2 mutation carriers undergoing risk-reducing salpingo-oophorectomy. Int J Gynecol Cancer. 2023;33(11):1743-1749. [DOI] [PubMed] [Google Scholar]
- 18. Visvanathan K, Shaw P, May BJ, et al. Fallopian tube lesions in women at high risk for ovarian cancer: a multicenter study. Cancer Prev Res (Phila). 2018;11(11):697-706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Powell CB, Swisher EM, Cass I, et al. Long term follow up of BRCA1 and BRCA2 mutation carriers with unsuspected neoplasia identified at risk reducing salpingo-oophorectomy. Gynecol Oncol. 2013;129(2):364-371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Linz VC, Löwe A, van der Ven J, Hasenburg A, Battista MJ.. Incidence of pelvic high-grade serous carcinoma after isolated STIC diagnosis: a systematic review of the literature. Front Oncol. 2022;12:951292. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Yoon JY, Chapel DB, Goebel E, et al. Genomic catastrophe, the peritoneal cavity and ovarian cancer prevention. J Pathol. 2022;257(3):255-261. [DOI] [PubMed] [Google Scholar]
- 22. Narod SA, Risch H, Moslehi R, et al. Oral contraceptives and the risk of hereditary ovarian cancer. Hereditary Ovarian Cancer Clinical Study Group. N Engl J Med. 1998;339(7):424-428. [DOI] [PubMed] [Google Scholar]
- 23. McLaughlin JR, Risch HA, Lubinski J.. Reproductive risk factors for ovarian cancer in carriers of BRCA1 or BRCA2 mutations: a case-control study. Lancet Oncol. 2007;8(1):26-34. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Data are available from the corresponding author to researchers who meet the Ethics Committee criteria for access to confidential data. Available data are dependent on patient consent and legal constrictions from each contributing site.