Prophylactic salpingo-oophorectomy (SO), which is recommended in BRCA1/2 mutation carriers, still needs to be reappraised.
Prophylactic salpingo-oophorectomy (SO), which is recommended in BRCA1/2 mutation carriers, still needs to be reappraised.
In all, 89 BRCA1/BRCA2 mutation carriers underwent SO between 1994–2004. Past medical and familial history, SO, results and survival after SO were analyzed.
The series consisted of 56 BRCA1 and 33 BRCA2 mutation carriers. All but 1 had a family history of breast (BC) and/or ovarian cancer; 42 BRCA1 and 31 BRCA2 had a personal history of BC. The median age at SO was 44 (BRCA1) and 49.5 (BRCA2) years for women without previous BC (not significant) and 48 (BRCA1) and 53 BRCA2) years (P = .03) for women with previous BC. Occult ovarian (n = 2) and/or fallopian (n = 3) carcinomas were found in 4 patients (4.5%): 1 experienced recurrence (4 years), 2 are disease-free (26 and 38 months of follow-up), and 1 died from BC (12 months). Among the other 69 patients with previous BC (median follow-up, 42 months), 14 developed ipsilateral or contralateral BC and 8 developed metastatic disease. Among the 16 patients without previous BC (median follow-up, 27 months), 3 developed BC. Of the 89 patients, 85 are still alive: 3 died from BC and 1 died from pancreatic cancer. No peritoneal malignancy was observed.
This study shows that prophylactic SO remains an important option for BRCA1/2 mutation carriers as asymptomatic ovarian/fallopian cancers were found in 4.5% of patients. However, a longer follow-up and larger series are required to more precisely evaluate the benefits of this procedure in terms of BC incidence, peritoneal malignancy, or recurrence. Cancer 2007. © 2007 American Cancer Society.
Germline mutations of the BRCA1 and BRCA2 genes are associated with a high risk of both breast and ovarian cancers.1–6 Identification of these genes about 10 years ago constituted a breakthrough in the management of women with a family history of breast and/or ovarian cancer,7, 8 as genetic testing can now reassure women who have not inherited a BRCA1/2 mutation running in their family and guides the management of women who have inherited this mutation.
In a combined analysis of 22 population studies, the cumulative risk of breast cancer (BC) at 70 years of age was estimated to be 65% (95% confidence interval [CI]: 51–75) and 45% (95% CI: 33–54) in BRCA1 and BRCA2 mutation carriers, respectively.9 The 10-year risk of contralateral BC is more than 5-fold higher than that observed after sporadic cancer (37% compared with 7%).10, 11 However, the risk of ipsilateral disease, at least during the first 10 years after treatment of a first BC, does not appear to be increased.11, 12 According to this same study, the cumulative risk of ovarian cancer at 70 years of age was 39% (95% CI: 18–54) and 11% (95% CI: 2.4–19) for BRCA1 and BRCA2 mutation carriers, respectively, whereas the risk in the age-matched general population is about 1%.9 However, the risks of both breast and ovarian cancers have been reported to be slightly modified by the site of the gene mutation and by either genetic or nongenetic modifying factors.9, 13, 14 Ovarian cancer risks are low before the age of 40 and 50 years in BRCA1 and BRCA2 mutation carriers, respectively. In addition to the risk of ovarian cancer, there is a low risk of fallopian tube (FTC) and primary peritoneum carcinomas (PPC). These 2 cancers not only have similar clinical symptoms, but also present an almost identical pathological appearance to that of ovarian carcinoma. The lifetime risks of FTC and PPC, estimated from the frequency of BRCA1 and BRCA2 mutations in a population study of FTC and PPC, were 0.6% and 1.3%, respectively.15 Ovarian cancers associated with BRCA1 or BRCA2 mutations are more likely than sporadic tumors to be serous cystadenocarcinomas with an aggressive pattern and are expected to have a poor prognosis.16
Due to the absence of reliable methods for early ovarian cancer detection and the high mortality of advanced ovarian cancer,17 bilateral prophylactic salpingo-oophorectomy (SO) has been recommended for BRCA1 or BRCA2 mutation carriers at the age of 40 or between the ages of 35 and 39 if childbearing is complete18, 19 (and http://www.nice.org.uk). In various series, SO has been demonstrated to achieve a 96% reduction of the ovarian cancer risk and about a 50% reduction of the risk of either a first BC in women with no previous BC or contralateral BC in women with a previous BC.10, 20–22
Prophylactic SO has consequently been offered to high-risk women undergoing genetic testing at Institut Curie. The primary objective of this retrospective study was to evaluate the frequency of ovarian, fallopian, and peritoneal carcinomas diagnosed at the time of SO and for a mean follow-up of 40 months in a series of 89 women carrying a BRCA1 or a BRCA2 mutation. Breast cancer occurrence or recurrence after prophylactic SO was also evaluated. A secondary objective was to analyze the demographic data of a population of BRCA1/2 mutation carriers who underwent prophylactic SO during the last decade.
This retrospective study analyzed demographic and clinical information regarding the 89 women with a BRCA1 or BRCA2 germline mutation who underwent a prophylactic SO in Institut Curie of Paris between December 1994 and October 2004.
Genetic testing for BRCA1 and/or BRCA2 mutations is regularly proposed to women with BC (index cases, ie, the first family member in whom complete screening of BRCA1/2 genes had been performed), followed at Institut Curie, and presenting a family history of breast and/or ovarian cancer or women with BC before the age of 36, or to relatives of women in whom a BRCA1 or a BRCA2 mutation had been identified. A family history was defined as either 2 first-degree relatives with cancer: 1) at least 1 with invasive BC before the age of 41 years, or 2) 1 with ovarian cancer at any age, or 3 first- or second-degree relatives from the same lineage with invasive BC or ovarian cancer at any age, where the index case constituted 1 of the affected cases. Genetic testing was performed in the Genetics Department of Institut Curie. The procedure used for molecular analysis, as well as the mode of information collection, such as family cancers, age at cancer diagnosis of relatives, and age at death or current age, have been reported previously.23, 24 Prophylactic SO was offered to these patients according to the recommendations of the French INSERM-FNCLCC collective expertise updated in 2004.19 At Institut Curie, ultrasound surveillance of the pelvis and abdominal cavity before SO has been regularly performed once a year since 2002. Serum CA125 evaluation was not a surveillance examination in high-risk women. However, it was performed in most cases before surgery, and was performed systematically since 2002. After SO, all these women were subsequently followed at our institute. Their clinical, pathologic, and outcome data were collected retrospectively by reviewing the patients' medical records.
As required, all subjects gave their informed consent for genetic testing and for prophylactic SO. This study was approved by our Institutional Review Board.
Bilateral prophylactic SO was performed by laparoscopy-assisted surgery or laparotomy when indicated. The surgical protocol included complete surgical resection of the ovaries and fallopian tubes as far as their insertion into the cornua of the uterus. Systematic peritoneal washings and biopsies of the pelvic peritoneum, gutter peritoneum, and omentum have been recommended as part of the procedure since 2002. The specimens were removed in an endoscopic sac to avoid intraperitoneal spillage for pathologic analysis. Careful macroscopic examination of the ovaries and fallopian tubes was performed by the pathologist before fixation in order to detect any abnormality that needed to be specifically analyzed. After fixation, systematic pathologic microsectioning and histopathologic examination of serial sections of all ovarian and fallopian tube tissue were performed.
A χ2 or Fisher exact test for qualitative variables and Student test (comparisons of means) for quantitative variables were used to compare patient characteristics. Survival was determined from the date of prophylactic SO to the date of event (death, recurrence, metastatic disease) or last follow-up. Kaplan-Meier estimates were calculated to assess overall survival and recurrence rates.25, 26 Statistical analyses were performed with R2.01 software from CRAN PROJECT (www.r-project.org).
In this series of 89 women who underwent prophylactic SO, 88 had a family history of BC and/or ovarian cancer, and 1 woman presented early-onset BC (34 years) with no family history of BC or ovarian cancer. Fifty-six and 33 women were BRCA1 and BRCA2 mutation carriers, respectively (Table 1). Thirty-seven (66%) of the BRCA1 carriers and 12 (36%) of the BRCA2 carriers had at least 1 first- or second-degree relative affected with ovarian cancer (P = .004).
|Breast and ovarian cancer||33||12||45|
|Personal breast cancer|
|No previous BC||14||2||16|
|BC after SO*||6||1||7|
|Progression of previous BC||8||3||11|
Sixteen of the 89 women had no personal history of BC: 14 were BRCA1 carriers and 2 were BRCA2 carriers (Table 1). The mean age at SO was 44 years and 49.5 years in BRCA1 and BRCA2 mutation carriers, respectively (not significant).
Seventy-three patients had a history of BC, 42 were BRCA1 mutation carriers, and 31 were BRCA2 mutation carriers (Table 1). The median age at BC diagnosis was 45 years (range, 22–63). At the time of SO, all patients were disease-free from their BC. The mean age at SO was 48 years and 53 years in BRCA1 and BRCA2 mutation carriers, respectively (P = .03).
The mean age at BRCA1 or BRCA2 mutation diagnosis was 51 years (range, 33–69). The mean latency period between announcement of a BRCA1/2 mutation and SO in the 89 patients was 7.6 ± 3.5 months. Between December 1994 and December 2000 (73 months), 36 of 89 prophylactic SO were performed (ie, an average of 6 per year), and between January 2001 and October 2004 (46 months), 53 of 89 SO were performed (ie, an average of 14 per year).
Ultrasound examination results during the 6 months before SO were available in 46 women and did not reveal any abnormality. Preoperative serum CA125 assay was performed in 39 cases and was within the normal range in all cases.
Prophylactic SO was performed by laparoscopy-assisted surgery in 75 women (84%) and open surgery in 14 women (16%). No death occurred and only 1 patient (1.1%) presented intraoperative bowel perforation related to the laparoscopic procedure, due to a trocar injury, which required temporary (6 months) colostomy followed by stomy closure with no late complications.
Histopathology examination of ovaries and fallopian tubes was strictly normal in 65 cases (73%). Benign lesions were detected in 20 women (median age, 45.5 years; range, 39–63): 14 serous cysts, 5 ovarian endometriotic cysts, and 1 dermoid cyst, all less than 10 mm in diameter and which were only discovered at the time of histopathologic examination. Peritoneal cytology and biopsies, available in 33 of these 85 women, were normal.
Although surgical exploration was described as normal, carcinomas were diagnosed in 4 patients (4.5%). Occult carcinomas were ovarian (n = 2) and/or involved the fallopian tube (n = 3). Data concerning these 4 patients and tumor characteristics are summarized in Table 2. Two of the 3 women with occult fallopian tube carcinoma were BRCA1 mutation carriers and 1 carried a BRCA2 mutation. Two patients presented FIGO stage Ia fallopian tube cancer, 1 patient presented both FIGO stage IIIc fallopian tube carcinoma and ovarian cystadenocarcinoma, and the fourth patient presented FIGO stage IIc ovarian cystadenocarcinoma. None of these 4 patients had an abnormal ultrasound examination or elevated serum CA125 level in the 3 months before surgery. All 4 patients had a personal history of BC. Peritoneal cytology and biopsies available in these 4 patients were normal.
|Patient||Family history||Mutation status||Breast cancer||Age at SO||Histopathology (FIGO stage)||Follow-up|
|1||Breast cancer only||BRCA1 c.4389delC/p.Tyr1463TyrfsX3||Unilateral, dg 44 y||46 y||Fallopian tube adenocarcinoma (Ia)||Breast cancer recurrence, death 47 y|
|2||Breast-ovarian cancer||BRCA1 C442_547del106/p.Gln148_Gly 183>AspfsX51||Unilateral, dg 51 y||55 y||Bilateral ovarian cystadenocarcinoma (IIc)||Alive free of disease, 57 y|
|3||Breast cancer only||BRCA1 c.68delAG/p.Glu23ValfsX17 185delAGter39||Bilateral, dg 46, 49 y||53 y||Fallopian tube adenocarcinoma (Ia)||Alive free of disease, 56 y|
|4||Breast-ovarian cancer||BRCA2 c.8140C>T/p.Gln2714X*||Unilateral, dg 49 y||54 y||Bilateral ovarian cystadenocarcinoma and fallopian tube adenocarcinoma (IIIc)||Alive liver ovary metastasis, 59 y|
Eighty-five of these 89 women are still alive: 3 died from BC and 1 died from pancreatic cancer. At 5 years, Kaplan-Meier estimates were: 95.7% (95% CI: 0.03) for survival, 94.1% (95% CI: 0.03) for local BC recurrence-free interval, 93.9% (95% CI: 0.03) for contralateral BC-free interval, and 88% (95% CI: 0.05) for metastasis-free interval.
In the 4 patients with occult ovarian and/or fallopian carcinomas diagnosed at the time of SO, 1 experienced recurrence, 2 are disease-free, and 1 of the 2 patients with fallopian tube cancer without ovarian cancer died from progression of a previous BC 12 months after SO (Table 2). The patient with both fallopian tube and ovarian carcinoma experienced recurrence with peritoneal and liver metastases 4 years after SO. The serum CA125 level remained normal during follow-up and the patient is still alive 14 months after disease recurrence. The other 2 patients are free of breast and ovary disease after 26 and 38 months of follow-up.
Among the 73 women with a history of BC, the 69 patients who were free of fallopian tube or ovarian carcinoma at SO remained free of peritoneal malignancy throughout follow-up, with a median of 42 months (range, 1–121). However, 14 of these 69 patients developed local BC recurrence (10) or contralateral BC (4), and 7 presented metastatic disease from their BC. Three patients died from their BC 22, 33, and 95 months after SO.
One patient, a BRCA1 mutation carrier (c.2063_ 2066del4/p.Thr688_Ser689>IlefsX12), died at 67 years of age from pancreatic cancer diagnosed 51 months after SO that was performed at 62 years of age.
The 16 women with no history of BC, with a median follow-up of 27 months (range, 6–89 months), were free of peritoneal malignancy. However, 3 patients developed BC after SO. All these patients are still alive with no peritoneal disease.
In this series of 89 women with a BRCA1 or BRCA2 germline mutation who underwent prophylactic SO during a 10-year period, 4 occult ovarian and/or fallopian tube cancers were diagnosed at the time of SO (4.5%). Three of these 4 carcinomas involved the fallopian tubes. Two of theses carcinomas were FIGO stage Ia, and 2 were FIGO stage IIc and IIIc. Three carcinomas occurred among the 56 BRCA1 carriers (5.3%) of our series and 1 among the 33 BRCA2 carriers (3%). The 5.3% frequency of carcinomas at the time of surgery in the BRCA1 carriers of our series is in agreement with recent studies by Olivier et al.27 and Finch et al.,28 as 8.6% and 6.4% of BRCA1 carriers undergoing SO were diagnosed with ovarian and/or fallopian tube carcinomas. Powell et al.29 showed that a rigorous operative and pathologic SO protocol increases the occult ovarian carcinoma detection rate. Pathologists in our institute have adopted a strict protocol of histopathologic examination of serial sections of ovarian and fallopian tube tissues since the beginning of prophylactic SO procedures. These 4 patients with occult cancer had a family history of BC, whereas 2 of the 3 BRCA1 carriers had no family history of ovarian cancer. The mother of the BRCA2 carrier with ovarian carcinoma and FTC had a history of both BC and ovarian cancer. The BRCA2 mutation, c.8140C>T/p.Gln2714X, was located outside the Ovarian Cancer Cluster Region (OCCR) of the BRCA2 gene. The OCCR is a region of mutations associated with a 2-fold higher risk of ovarian cancer compared with the risk of mutations located outside the OCCR.30 These observations illustrate that BRCA1 and BRCA2 mutations alone are strong indicators of ovarian cancer risk.
No primary peritoneal disease was diagnosed in our study of 85 women with no occult carcinoma at the time of surgery and with a mean follow-up of 40 months. This is not unexpected in the light of previous studies reporting a low risk of PPC.27, 31, 32 In the recent study by Finch et al.33 in women who had undergone SO, the annual risk of PPC was estimated to be 0.217% in BRCA1 carriers and 0.167% in BRCA2 carriers. On the basis of these estimates, less than 1 case would have been expected in our series. However, various studies, including the study by Finch et al.,33 have suggested that the risk of PPC may be overestimated due to undiagnosed occult ovarian or fallopian tube carcinomas at the time of surgery, which would be considered to be PPC when they become clinically apparent.29, 34 It should be stressed that in the series by Olivier et al.27 based on 38 patients who had undergone bilateral oophorectomy without associated salpingectomy (mean follow-up, 45 months), 3 of the 26 BRCA1 carriers developed PPC, whereas no case of PPC was observed in the 90 women in whom follow-up was limited to 12 months but who underwent SO. Prospective studies examining the frequency of PPC in women who have undergone prophylactic SO with thorough pathologic examination to identify any occult carcinomas are urgently needed.
Current ovarian screening strategies remain inadequate to identify early ovarian and/or fallopian tube cancer, as illustrated by the normal pre-SO ultrasound examination or CA125 level of the 4 patients with occult carcinomas. Prophylactic SO has therefore been demonstrated to effectively reduce the risk of ovarian and fallopian tube cancer and has been largely recommended in various countries in women with BRCA1/2 mutations. Prophylactic SO is associated with a low risk of operative complications. No surgical or anesthetic complications were observed in our series, except for 1 patient in whom an intraoperative bowel perforation was related to the laparoscopic procedure due to a trocar injury, requiring temporary colostomy with no late complications and complete recovery. An increasing rate of prophylactic SO has been recorded during the 1994–2004 decade. During the 2001–2004 period, an average of 14 SO per year were performed in contrast with the average of 6 SO per year during the 1994–2000 period. This increase in the number of SO is related to many additive factors: 1) the increasing use of BRCA1/2 genetic testing in recent years, resulting in an increased number of identified mutations; 2) the higher number of tests in relatives; 3) the publication of recommendations for prophylactic SO in BRCA1/2 mutation carriers; and perhaps 4) the better acceptance of SO by high-risk women. These factors may be reflected by the increasing number of women with no family history of ovarian cancer in the 2001–2004 period compared with the 1994–2000 period (54.7% vs 30.6%; P = .025). In addition, among the 16 women with no previous personal history of BC, a strong family history of ovarian cancer with at least 1 first-degree relative affected before the age of 50 was present in the 5 women who underwent SO before 2001 compared with 4 out of 11 after 2001. A recent report35 has also shown that women with a family history of ovarian cancer are more likely to undergo a risk-reduction surgical option than other patients. The good acceptance of SO is underlined by the short latency period, 7.6 ± 3.5 months in our series, between announcement of a BRCA1/2 mutation and SO.
The growing impact of a positive BRCA1/2 test result on the choice of SO observed in our series is in agreement with the results reported by Schwartz et al.,36 who examined the choice of preventive options in a series of 289 high-risk women. We observed a very short latency period between announcement of the positive genetic test results and prophylactic SO. The mean age at BRCA1/2 mutation diagnosis was 51 years and the mean age at SO was 48 years in BRCA1 mutation carriers and 53 years in BRCA2 mutation carriers (P = .03). This different mean age at SO was expected, as the risk of ovarian cancer is low before the age of 50 in BRCA2 carriers.9, 37 This risk is estimated to be less than 1% in many studies, including studies in which index cases were selected according to a positive family history of cancer.6, 9
Eighty-two percent of women who underwent SO in our series had a personal history of BC. This high percentage is related to various factors: 1) most women were index cases; 2) due to the high risk of BC, many relatives carrying a BRCA1 mutation and, to a lesser extent, relatives carrying a BRCA2 mutation, had already developed BC at the time of SO; or 3) women with a history of BC more readily accept SO. As shown in the studies by Scheuer et al.38 and Meijers-Heijboer et al.,39 prophylactic SO has been demonstrated to offer about a 50% reduction of the risk of either a first BC in women with no previous BC or contralateral BC in women with a previous BC.10, 11, 20–22 In the present series, 7 BCs occurred after SO: 3 BCs among the 16 women with no personal history of BC, and 4 contralateral BCs among patients with a history of BC. Due to the limited sample size and short follow-up of our series, mainly composed of patients with previous BC, no conclusions can be drawn concerning the effect of SO on BC in these patients. However, it must be stressed that in the study by Eisen et al.40 the greatest impact of SO on BC was observed for SO performed before 40 years of age and for early-onset BC in BRCA1 carriers. A less marked risk reduction was observed for BRCA2 carriers.
Eight patients in the present study also presented metastatic disease from previous BC and 3 patients died from their BC 22, 33, and 95 months after SO. The impact of SO on reduction of the cancer-specific mortality is still unclear, and a recent study reported that prophylactic SO was associated with a decrease in overall mortality, but not in breast or ovarian cancer-specific mortality.41 One BRCA1 carrier died from pancreatic carcinoma, observed more frequently in BRCA2 carriers but also known to be associated with BRCA1 mutation.42
Quality of life, although not studied in our series, is not expected to be markedly altered according to the study by Madalinska et al.43 The main expected impact of SO is hormone deprivation. Low-dose hormone replacement therapy (HRT) does not appear to decrease the preventive effect of SO on the BC risk.44 In women younger than 50 years of age with no history of BC, HRT may be prescribed45 in the presence of severe menopausal symptoms. The age at which SO should be recommended is still a subject of controversy, particularly in BRCA2 carriers. We consider that SO can be postponed until the age of 50.45
In conclusion, this study shows that prophylactic SO remains an important option for women at risk of hereditary breast or gynecologic cancer, particularly in patients with a personal history of BC, as asymptomatic ovarian and/or fallopian tube cancers were found in 4.5% of patients. This study indicates the need to improve the information of gynecologists and women in order to increase the SO rate in BRCA1/2 mutation carriers with no personal history of cancer. Moreover, a larger series with a longer follow-up is necessary to more precisely evaluate the benefits of this procedure, particularly in terms of considering its effects on BC in BRCA1/2 mutation carriers.
We thank Dr. Krishna Clough, who initiated this work. We thank the patients for giving access to their follow-up.