Association between clinical characteristics and risk-reduction interventions in women who underwent BRCA1 and BRCA2 testing
A single-institution study
Women who are at increased risk for breast and ovarian cancers, especially BRCA1 and BRCA2 mutation carriers, face a myriad of risk-reduction options, including increased surveillance, chemoprevention, prophylactic oophorectomy, and prophylactic mastectomy. However, little is known about which clinical, demographic, or cancer-related factors are associated with risk-reduction interventions.
The authors conducted a retrospective review of records for 554 women who had undergone testing at The University of Texas M. D. Anderson Cancer Center between 2000 and 2006 for deleterious BRCA1 and BRCA2 gene mutations. Data were collected on the risk-reduction interventions these women adopted after they underwent genetic testing. These data were tested for associations with demographic and clinical characteristics.
Among the 554 women who underwent genetic testing for BRCA mutation, 78 were found to have a deleterious mutation in the BRCA1 gene, and 54 had a mutation in the BRCA 2 gene. Of the 554 women, 85 underwent prophylactic mastectomy, 30 prophylactic oophorectomy, and 52 both surgeries; 387 women opted for surveillance. Women who had BRCA mutations, a history of breast cancer or ductal carcinoma in situ (DCIS), or previous breast biopsies were more likely to have prophylactic surgery. Women with a family history of ovarian cancer were more likely to undergo prophylactic oophorectomy. Women with a personal history of ovarian cancer or advanced breast cancer were more likely to undergo surveillance only. Women with breast cancer who had had a total mastectomy as part of their prior breast cancer treatment underwent prophylactic mastectomy more frequently than women who either had breast-conserving surgery or no history of breast cancer. In multivariate analysis, only positive BRCA mutation carrier status was associated with having had prophylactic surgery. In addition, breast cancer history was significantly associated with prophylactic mastectomy.
Women who were BRCA carriers, women who had a history of breast cancer, DCIS, or breast biopsy, or had a family history of ovarian cancer were more likely to have undergone surgery for cancer risk reduction. Women with ovarian cancer or advanced breast cancer were more likely to have undergone surveillance. Cancer 2006. © 2006 American Cancer Society.
About 5% to 10% of all breast cancers and 25% to 40% of breast cancers that occur in women younger than 35 years of age are attributable to a hereditary cause; 60% to 75% of these cancers are caused by an inheritable mutation in the BRCA1 or BRCA2 gene.1, 2 Women who test positive for deleterious BRCA1 and BRCA2 gene mutations are at an increased risk for developing breast and ovarian cancer. Specifically, BRCA1 and BRCA2 mutation carriers have a lifetime breast cancer risk of 60% to 80%2, 1 and up to a 40% lifetime ovarian cancer risk.
Women who have an increased risk for breast and ovarian cancers are advised to consider risk-reduction strategies that vary considerably in terms of invasiveness and efficacy. Currently available risk-reduction options include surveillance (breast self-exam, clinical breast exam, mammography, and breast magnetic resonance imaging [MRI]), chemoprevention, prophylactic oophorectomy, and prophylactic mastectomy.3 Surveillance, the least invasive option, targets early detection rather than prevention of breast and ovarian cancers. Although this strategy generally has no direct physical or toxicological side effects, it can have negative consequences, such as increased worry about cancer and unnecessary biopsies.4, 5 Furthermore, the effect of surveillance on survival in high-risk women is unknown. Chemoprevention by means of tamoxifen use reduces a women's risk of breast cancer by almost 50%, but it is associated with increased risk for endometrial cancer, thromboembolic events, cataracts, and menopausal symptoms.6 Surgical options are highly invasive but also highly effective. Prophylactic oophorectomy reduces risk for not only ovarian cancer but also for breast cancer (by 60%).7 The side effects of prophylactic oophorectomy included decreases in endocrine function and sexual function compared with women who did not choose prophylactic oophorectomy.8 Prophylactic mastectomy reduces the risk of developing breast cancer by more than 90%,9 but it is the least acceptable risk reduction option for many women.10
Little is known about factors related to the selection of risk-reduction interventions among women at increased risk for hereditary breast or ovarian cancer. This report describes our retrospective analysis of women who underwent genetic counseling and testing for BRCA1 and BRCA2 mutations at 1 comprehensive cancer center. We examined the women's demographic and clinical characteristics and features of any prior cancers to determine their association with risk-reduction interventions.
MATERIALS AND METHODS
This study was approved by the University of Texas M. D. Anderson Cancer Center institutional review board. The study population included women who were seen at M. D. Anderson Cancer Center between 2001 and 2005 for genetic counseling and evaluation of their risk for hereditary breast or ovarian cancer. The current analysis was confined to those women who underwent genetic testing for mutations in the BRCA1 and BRCA2 genes and who had medical records available at M. D. Anderson Cancer Center. All BRCA testing was performed by Myriad Genetics Laboratories, Inc (Salt Lake City, UT). BRCA test results were categorized as either positive or negative. Test results that indicated a variance of uncertain significance were categorized as negative, because women with these findings are usually advised similarly to women who have negative test results.
Demographic and clinical information regarding the women who were tested for BRCA1 and BRCA2 were collected from the medical records at M. D. Anderson Cancer Center and the Cancer Genetics Department database. For patients who had a history of previous breast and/or ovarian cancer, we also collected data related to disease characteristics and surgical treatments.
Clinical decision groups were categorized into surveillance, prophylactic mastectomy, prophylactic oophorectomy, chemoprevention with tamoxifen, and a combination of surgeries (mastectomy and oophorectomy). Patient characteristics were tabulated within each clinical decision group, and the chi-square test or Fisher exact test was used to test for associations. The analysis was repeated in the subgroup of patients who had a history of breast cancer, with the addition of variables describing their disease characteristics and treatments. Multivariate logistic regression modeling was used to further analyze associations between clinical characteristics and risk-reduction strategies undertaken.
Between January 2001 and December 2005, 1055 patients underwent genetic counseling for hereditary breast and ovarian cancer syndrome at M. D. Anderson Cancer Center. Of these women, 627 underwent genetic testing for BRCA1 or BRCA2 mutations. Medical records for 554 of these 627 women were available for review.
Table 1 shows the demographic and clinical characteristics of the 554 women whose records were reviewed. BRCA1 mutations were detected in 78 women and BRCA2 mutations in 54 women; 410 women were negative for both mutations. The majority of patients categorized themselves as Caucasian or Jewish (84.1%); the remainder was 9.4% Hispanic, 3.8% Black, and 2.7% Asian. Of the patients tested, 64% had a personal history of breast cancer, whereas 14% had a personal history of ovarian cancer; 22 (4%) patients had a personal history of both types of cancer. A family history of breast cancer was more common (81.5%) than a personal history of the disease; 14% of all women in the study had a family history of ovarian cancer. Family history was subdivided into 2 categories, strong and weak. A strong family history of breast cancer was defined as having a first-degree relative who developed breast cancer before the age of 50 years or having more than 1 family member (at least 1 of whom was a first-degree relative) affected by breast cancer. A weak family history of breast cancer was having any family history of breast cancer that did not fit criteria for a strong family history.
Table 1. Demographic and Clinical Characteristics of Women Tested for BRCA1 and BRCA2 Mutations
|Both negative||410 (74)|
| BRCA1||78 (14)|
| BRCA2||54 (12)|
|History of breast cancer|
| None||202 (36)|
| 1 Primary||300 (54)|
| More than 1 primary||52 (10)|
|History of ovarian cancer|
| Yes||77 (14)|
| No||477 (86)|
| Yes||116 (21)|
| No||438 (89)|
|Family history of breast cancer|
| None||103 (18.5)|
| Weak||186 (33.5)|
| Strong||263 (48)|
|Family history of ovarian cancer|
| None||477 (86)|
| Yes||77 (14)|
|OCP or HRT use|
| None||50 (11.3)|
| OCP only||269 (61.3)|
| HRT only||38 (8.6)|
| Both||83 (18.8)|
|History of breast biopsy|
| No||152 (27.4)|
| Yes||402 (72.6)|
|Breast Cancer Specific Characteristics|| |
|Age at cancer diagnosis|
| <50 y||260 (74)|
| ≥50 y||91 (26)|
|Estrogen receptor status|
| Negative||81 (27)|
| Positive||222 (73)|
| Negative||178 (76)|
| Positive||55 (24)|
| 0 or I|| |
| II||110 (34)|
| III or IV||170 (52)|
|Complete mastectomy for breast cancer|
| No||92 (26)|
| Yes||260 (74)|
Among the patients who had a personal history of breast cancer, 44 tested positive for BRCA1 mutations, 30 tested positive for BRCA2 mutations, and 271 tested negative for both mutations. The median age of breast cancer diagnosis was 43 years (range, 21–75 years). The majority of tumors were estrogen receptor positive (73%) and Her2/Neu negative (76%). At diagnosis, 87% of patients had stage I or II disease. Most (74%) patients opted for a total mastectomy rather than breast-conserving surgery and radiation therapy, and 60% underwent antiestrogen adjuvant therapy.
The risk-reduction interventions adopted by the women were as follows: 52 (9.4%) underwent both prophylactic mastectomy and prophylactic oophorectomy, 85 (15.3%) had prophylactic mastectomy, 30 (5.4%) had prophylactic oophorectomy, and 387 (69.9%) had surveillance. Seven patients chose chemoprevention, 1 patient received both prophylactic surgeries and chemoprevention, and 2 patients received prophylactic oophorectomy and chemoprevention.
Table 2 shows the use of risk-reduction interventions relative to cancer history and BRCA mutation status. Among breast cancer patients, 62.6% of BRCA1 or BRCA2 carriers underwent prophylactic surgery, whereas 32.0% of those with BRCA-negative cancer underwent prophylactic surgery. The median age of mastectomy was 43 years for BRCA1 mutation carriers and 44 years for BRCA2 mutation carriers, the difference was not statistically different. Among ovarian cancer patients, the majority underwent surveillance. The 2 ovarian cancer patients who were BRCA mutation carriers and who underwent prophylactic mastectomy had the surgery before their diagnosis of ovarian cancer. Among the 22 women who had both breast and ovarian cancer, 60.0% of the BRCA mutation carriers underwent prophylactic mastectomy (1 underwent both prophylactic surgeries); 22.0% of the BRCA mutation-negative women underwent prophylactic surgery. In women with no history of breast or ovarian cancer, 56.8% of BRCA mutation carriers underwent surveillance, compared with 85.7% of BRCA mutation noncarriers. Chemoprevention with the use of tamoxifen was not included in the analysis because there were too few patients to perform the analysis.
Table 2. BRCA Status, Cancer History, and Use of Risk-reducing Interventions
| BRCA positive||1 (1.5)||26 (40)||5 (7.6)||10 (15)||23 (35)|
| BRCA negative||0 (0)||19 (7)||57 (22)||8 (3)||178 (68)|
| BRCA positive||3 (11.5)||0 (0)||2 (7.7)||1 (3.8)||20 (77)|
| BRCA negative||0 (0)||0 (0)||0 (0)||1 (3)||30 (97)|
|Breast and ovarian cancer|
| BRCA positive||0 (0)||1 (10)||5 (50)||0 (0)||4 (40)|
| BRCA negative||0 (0)||0 (0)||1 (11)||1 (11)||7 (78)|
| BRCA positive||2 (5.4)||5 (13.5)||4 (10.8)||5 (13.5)||21 (56.8)|
| BRCA negative||4 (3.6)||1 (0.9)||7 (6.3)||4 (3.6)||96 (85.7)|
Patient characteristics were evaluated for possible associations with any of the risk-reduction interventions. Data for the group as a whole are reported in Table 3. BRCA mutation carriers were more likely than noncarriers to undergo either both prophylactic surgeries or prophylactic oophorectomy alone. BRCA mutation carriers also were less likely to undergo surveillance than their BRCA mutation-negative counterparts.
Table 3. Clinical Characteristics Associated With Uptake of Risk Reduction Interventions
|BRCA status|| || || || ||<.0001|
| Both negative||20 (4.9)*||65 (15.9)||14 (3.4)*||311 (75.8)*|| |
| BRCA1||17 (21.8)*||11 (14.1)||9 (11.5)*||41 (52.6)*|| |
| BRCA2||15 (27.8)*||5 (9.3)||7 (13.0)*||27 (50)*|| |
|Breast cancer|| || || || ||<.0001|
| None||6 (3)*||15 (7.4)*||11 (5.4)||170 (84.1)|| |
| 1 Primary||41 (13.7)*||66 (22)*||14 (4.7)||179 (59.7)|| |
| More than 1||5 (9.6)*||3 (5.8)*||5 (9.6)||39 (75)|| |
|Ovarian cancer|| || || || ||<.0001|
| None||51 (10.7)*||76 (15.9)||27 (5.7)||323 (67.7)*|| |
| Yes||1 (1.3)*||8 (10.4)||3 (3.9)||65 (84.4)*|| |
|DCIS|| || || || ||.01|
| None||39 (8.9)||56 (12.8)*||22 (5.0)||321 (73.2)*|| |
| Yes||13 (11.2)||28 (24.1)*||8 (6.9)||67 (57.8)*|| |
|FH ovarian cancer|| || || || ||.0001|
| None||25 (6.9)*||65 (18.1)*||13 (3.6)*||257 (71.4)|| |
| Yes||27 (13.9)*||19 (9.8)*||17 (8.8)*||131 (67.5)|| |
|OCP or HRT|| || || || ||.03|
| None||6 (12)||10 (20)||2 (4.0)*||32 (64)|| |
| OCP only||20 (7.4)||46 (17.1)||15 (5.6)*||188 (69.8)|| |
| HRT only||1 (2.6)||7 (18.4)||1 (2.6)*||29 (76.3)|| |
| Both||8 (9.6)||7 (8.4)||11 (13.3)*||57 (68.7)|| |
|Breast biopsy|| || || || ||<.0001|
| None||6 (3.9)*||5 (3.3)*||7 (4.6)||134 (88.1)|| |
| Yes||46 (11.4)*||79 (19.7)*||23 (5.7)||254 (63.1)|| |
|Age at breast cancer diagnosis|| || || || ||.12|
| <50 y||37 (14.2)||56 (21.5)||13 (5.0)||154 (59.3)|| |
| ≥50 y||8 (8.8)||13 (14.3)||6 (6.6)||64 (70.4)|| |
|Stage of breast cancer|| || || || ||.046|
| 0 or I||21 (19.1)*||19 (17.3)||8 (7.3)||62 (56.4)|| |
| II||17 (10.0)*||30 (17.6)||8 (4.7)||115 (67.6)|| |
| III or IV||4 (8.7)*||15 (32.6)||2 (4.3)||25 (54.4)|| |
Table 3 also shows that patients with a history of breast cancer were more likely to undergo surgical intervention than patients without a history of breast cancer. Patients with lower stage disease (stage 0, I, or II) underwent prophylactic surgeries at a higher rate than patients with higher stage cancer. Patients who underwent total mastectomy as part of their cancer treatment also underwent both prophylactic surgeries (either both surgeries or prophylactic mastectomy of the contralateral breast) than patients who had previously undergone lumpectomy, radiation, or a combination of the 2 therapies.
Patients with a history of ovarian cancer were more likely to undergo surveillance than patients without any history of ovarian cancer. Women who had a family history of ovarian cancer were more likely than other patients to undergo any surgical option. Patients who had used both oral contraceptive pills (OCP) and hormone replacement therapy (HRT) were more likely to undergo prophylactic oophorectomy than patients who had used either OCP or HRT alone or neither of these treatments. Ethnicity, marital status, parity, and family history of breast cancer were not independently associated with specific risk-reduction interventions.
Multivariate logistic regression modeling was performed to further evaluate variables associated with use of prophylactic mastectomy, prophylactic oophorectomy, or surveillance. BRCA mutation-positive status was positively associated with use of prophylactic surgeries (odds ratio, 5.96; 95% confidence interval, 2.74–12.97) and negatively associated with the use of surveillance (odds ratio, 0.22; 95% confidence interval, 0.13–037). A history of breast cancer was significantly associated with prophylactic mastectomy of the contralateral breast.
Women who carry a mutation in the BRCA1 or BRCA2 gene have up to an 80% chance of developing breast cancer over their lifetime.1, 11 These women develop breast cancer at a significantly younger age than women with sporadic breast cancer. BRCA1 mutation carriers have an 18% risk, and BRCA2 mutation carriers have a 15% risk of developing breast cancer before they are 40 years of age.11 Our study retrospectively reviewed all the women who were tested for these mutations and counseled about their risk for familiar or gene mutation-associated cancers at M. D. Anderson to determine whether any of the women's clinical or demographic characteristics could be associated with risk-reduction strategies.
Women who undergo genetic testing for hereditary breast and ovarian cancer syndrome are counseled on strategies they can use to reduce their risk of subsequent cancers. Prophylactic mastectomies reduce risk of developing breast cancer by 90%, but these procedures are considered too aggressive by many women.9, 12 The complication rate for prophylactic mastectomy was reported to be as high as 64% in one study,13 and the potential for both short-term and long-term psychosocial consequences14 must be considered. In previous studies of BRCA mutation carriers, 0% to 54% chose to undergo prophylactic mastectomies.15–17 In our patient population, 36% underwent prophylactic mastectomy, with or without prophylactic oophorectomy, if they tested positive for mutations in either BRCA1 or BRCA2, whereas only 20% did so if they tested negative for these mutations. In our analysis, BRCA mutation-positive carrier status was associated with use of prophylactic mastectomy. Women with a history of breast cancer regardless of their BRCA status were more likely to undergo a contralateral prophylactic mastectomy if they had a complete mastectomy as part of their treatment. These results are consistent with those in a previous study, which showed that BRCA mutation carriers with breast cancer chose bilateral mastectomy over breast-conserving surgery.18 Women without a history of invasive breast cancer but who had ductal carcinoma in situ (DCIS) or had undergone a breast biopsy also were more likely to have prophylactic mastectomies than women who did not have DCIS or biopsy; these women may have chosen the aggressive approach because of their more intimate fear of breast cancer. Women who undergo a mastectomy as part of their treatment may be more open to surgical intervention and removal of the breast.
Prophylactic oophorectomy not only reduces risk of developing ovarian cancer, but also reduces the chance of developing breast cancer, especially in premenopausal women who are BRCA mutation carriers.19–21 In all women in the study and in subgroup analysis, a family history of ovarian cancer was associated with prophylactic oophorectomy. We found that among women without breast cancer, BRCA2 mutation carriers were more likely to undergo both prophylactic surgeries or prophylactic oophorectomy than were BRCA1 mutation carriers. This result was somewhat surprising given that women with a BRCA1 mutation have a lifetime risk of 39% for ovarian cancer, whereas BRCA2 mutation carriers have only an 11% risk for ovarian cancer;11 it is unclear why BRCA2 mutation carriers tended to choose prophylactic oophorectomy more than BRCA1 mutation carriers in our study.
The National Surgical Adjuvant Breast and Bowel Project (NSABP) study P-1 showed that chemoprevention by tamoxifen reduces the risk of breast cancer by 50% in high-risk women.6 In BRCA mutation carriers who have already had breast cancer, initial treatment with tamoxifen reduces the risk of contralateral breast cancer.22 Another small study found that tamoxifen reduced the incidence of breast cancer in BRCA2 mutation carriers by 62% (not significant), but tamoxifen had no effect on BRCA1 mutation carriers.23 In our study population, only 10 (2%) women chose to take tamoxifen as chemoprevention, a rate much lower than that in other reports.24, 25 We surmise that this low rate of tamoxifen use may be related to its side effects or because of the finding that the majority of women in our study population had a history of breast or ovarian cancer, and tamoxifen is not offered to these patients as a chemopreventive agent. In addition, many of these women received tamoxifen or an aromatase inhibitor as part of treatment for their breast or ovarian cancer, and so they might have been unwilling or unable to consider using systemic agents for risk reduction. It is also possible that because this is a single-institution study, a selection bias might have played a role.
Although detection of breast cancer due to ongoing surveillance is not preventive, it may be of benefit to detection of breast cancer at an earlier stage, and, thus, survival may be increased. However, it has not been shown yet that screening increases survival in high-risk women. The Cancer Genetics Studies Consortium for Surveillance of Women at High Risk for the Development of Breast Cancer recommends monthly breast self-exams, biannual clinical breast exams, and annual mammograms beginning between the ages of 25 and 35 years for women who carry BRCA mutations.26 Mammograms reduce breast cancer mortality in the general population by detecting cancers early, but they are less sensitive in younger women who have denser breast tissue and more specifically in those who are BRCA mutation carriers. Magnetic resonance imaging (MRI) of the breast is more sensitive than mammography in detecting breast cancer, but MRI is less specific and results in higher rates of unnecessary biopsy and cancer worry.27 In BRCA mutation carriers, who tend to get breast cancer at a younger age, MRI detects a greater number of breast cancers than mammography and clinical breast exams.28, 29 Our institution advocates MRI of the breast in all women who have BRCA mutations but not in women who test negative for these mutations; thus, intensive surveillance, including MRI, would occur more frequently in BRCA mutation carriers. Women in our study population used surveillance more often than an aggressive risk-reduction intervention when they had a history of ovarian cancer or stage III or IV breast cancer. Presumably, these women had a higher risk of death due to their current cancer than any cancer yet to arise and, thus, considered the more aggressive preventive options, such as prophylactic surgery, unreasonable.
Our study was a retrospective chart review, and it is limited in several ways, for instance, long-term follow-up and outcome of different risk reduction options is not analyzed. Furthermore, although associations between demographic or clinical characteristics and chosen risk-reduction strategies may shed light on the reasoning behind a woman's decision, researching other factors that may come into play is warranted. Through a prospective questionnaire study, we are currently evaluating the outcome of these patients, their reasons for choosing certain risk-reduction strategies, and their satisfaction with their decisions.