Breast and ovarian cancer risk perception after prophylactic salpingo-oophorectomy due to an inherited mutation in the BRCA1 or BRCA2 gene
Steven Narod, Women’s College Research Institute, 790 Bay Street, 7th Floor, Toronto, ON M5G 1N8, Canada.
Tel.: 416 351 3765;
fax: 416 351 3767;
It is often recommended that women who carry a mutation in the BRCA1 or BRCA2 gene have their ovaries and fallopian tubes removed to reduce their risk of gynecologic cancer. The aim of this study was to evaluate women’s perception of their risk of breast and ovarian cancer before and after prophylactic salpingo-oophorectomy. We surveyed 127 women who carry a BRCA1 or BRCA2 mutation and who underwent prophylactic salpingo-oophorectomy at the University Health Network, Toronto. Subjects were asked to estimate their risks of breast and ovarian cancer before and after surgery. Their perceived risks of cancers were then compared with published risks, based on their mutation status. BRCA1 carriers estimated their risk of breast cancer risk to be, on average, 69% before surgery and 41% after surgery. They estimated their risk of ovarian cancer to be 55% before surgery and 11% after surgery. BRCA2 carriers estimated their risk of breast cancer to be 69% prior to surgery and 45% after surgery and their perceived risk of ovarian cancer to be 43% before surgery and 8% after surgery. Compared with published risk figures, the perceived risk of ovarian cancer before prophylactic salpingo-oophorectomy was overestimated by 47% of BRCA1 mutation carriers and by 61% of BRCA2 mutation carriers. Most women who have undergone genetic counseling and subsequently choose prophylactic salpingo-oophorectomy accurately perceive their risk of breast cancer. However, in this study, many women overestimated their risk of ovarian cancer, particularly women who carry a BRCA2 mutation.
Barbara Bowles Biesecker, email: email@example.com
Prophylactic salpingo-oophorectomy is recommended to women at high risk of ovarian and fallopian tube cancer (1). The procedure has been shown to decrease the risk of gynecologic cancers in women of all ages and of breast cancer when carried out in premenopausal years (2–6).
The goals of cancer genetic counseling include the introduction of strategies to prevent or reduce the risk of cancer and an achievement of an accurate understanding of cancer risk (7–9). Many studies have focused on changes in risk perception among women with an increased susceptibility to cancer in response to genetic counseling and genetic testing (9–13). Studies have also examined how personal and family histories of cancer relate to perception of risk and the uptake of screening and preventive measures (13–15). There are few studies that examine the perception of risk following an intervention to reduce cancer risk (14, 16).
We asked women who carry a BRCA1 or BRCA2 mutation, and who chose to undergo prophylactic salpingo-oophorectomy, to estimate their risk of developing breast and ovarian cancer before and after surgery.
Materials and methods
Women with a BRCA1 or BRCA2 mutation who elected to undergo prophylactic salpingo-oophorectomy at The University Health Network from 1 January 2000 to 31 May 2006 were invited to participate. Subjects were contacted from December 2002 until May 2007. To introduce the study, a letter of invitation and an information sheet were mailed from the gynecologic oncologist who performed the surgery. The subject was then contacted by telephone, and the study was explained. Each study subject who elected to participate was mailed a questionnaire assessing satisfaction with their decision to undergo prophylactic salpingo-oophorectomy and their perception of risk, approximately 1 year following surgery. In this questionnaire, women were asked to estimate what they believed their lifetime risk of breast cancer and ovarian cancer to be before, and after, prophylactic salpingo-oophorectomy by marking along a scale from 0% to 100%. A second follow-up questionnaire was sent approximately 3 years after surgery to a subset of these women (with sufficient follow-up time). Women who did not return their questionnaires were contacted one time by telephone as a reminder. All subjects had genetic counseling as part of the genetic testing process prior to participation in this study, including a discussion of risks for cancer and the reduction in risk associated with prophylactic salpingo-oophorectomy. The study was approved by the Research Ethics Board at the University Health Network.
Published estimates of risk for breast cancer and ovarian cancer in women who carry BRCA1 or BRCA2 mutations were used as standards and compared with the perceived risks in this study. These estimates of risk for breast cancer are similar for BRCA1 and BRCA2 and range from 45% to 87% (17, 18). The risk of ovarian cancer ranges from 39% to 54% for BRCA1 and 11% to 23% for BRCA2(17–20). Studies have found that breast cancer risk can be reduced by up to 50% if salpingo-oophorectomy is performed prior to menopause and that ovarian cancer risk can be reduced by up to 95% (2–6, 21).
All data, including perception of cancer risks, were coded and entered in an Access database. The statistical package SAS (version 9.1) was used for analysis. The paired t-test procedure was used to compare perceived risk at 1 and 3 years post-surgery. The unpaired t-test procedure was used to compare the perceived change in breast cancer risk for women under the age of 50 years and for women above the age of 50 years. The Pearson correlation coefficient was calculated to evaluate the relationship between age at the time of surgery and perceived risk of breast and ovarian cancer before and after surgery. Statistical significance was defined at the level of p ≤ 0.05. All statistical tests were two tailed.
A total of 188 women with a BRCA1 or BRCA2 mutation underwent prophylactic salpingo-oophorectomy at a Canadian hospital between 1 January 2000 and 31 May 2006. One hundred and eighty women were contacted to participate in a study of the impact of the prophylactic surgery (five women were excluded from the study because they were diagnosed with occult ovarian or fallopian tube cancer at the time of the surgery; one woman died of breast cancer prior to mailing of the follow-up questionnaire; and two women were not contacted due to illness). Of the 180 women, 127 (71%) women agreed to participate and returned the questionnaire regarding risk perception, between 6 months and 4 years after surgery. Eight (4%) women declined to participate, and 45 (25%) women agreed to participate in the study but did not return the questionnaire. The questionnaires were completed on average 17 months (range 6–48.5 months) after surgery. The characteristics of the study subjects are described in Table 1.
Table 1. Characteristics of study subjects
|Number of subjects (n)||75||52|
|Age at surgery in years, mean (range)||45.3 (35–63)||48.8 (37–67)|
|Months since surgery||16.4 (6–39)||16.9 (10.5–48.5)|
|Previous breast cancer||48% (n = 36)||41% (n = 21)|
As expected, women in this study perceived a significant reduction in risk of breast and ovarian cancer as a result of surgery (Table 2). The perceived reduction in risk from before to after surgery was significant, with a p-value of <0.0001 for risk of both breast and ovarian cancer.
Table 2. Perceived risks for breast and ovarian cancer as reported after prophylactic salpingo-oophorectomy
|Breast risk before surgerya,b||69 (34) (10–100)||69 (29) (20–100)|
|Breast risk after surgerya,b||41 (34) (0–99)||45 (29) (0–85)|
|Ovary Risk before Surgerya||55 (74) (4–100)||43 (51) (5–92)|
|Ovary risk after Surgerya||11 (74) (0–90)||8 (51) (0–95)|
The range of perceived risks was broad before and after surgery for both breast and ovarian cancer (Table 2). The great majority of BRCA1 carriers (30 of 34 or 88%) estimated their breast cancer risk to be within 5% of the actual range for BRCA1 (45–87%). This number was similar for BRCA2; 25 of 29 (86%) women estimated their risk within the actual range (45–87%).
Twenty-eight of 74 (38%) BRCA1 carriers estimated their ovarian cancer risk prior to surgery within 5% of the range of published risk figures. Forty-seven percent overestimated their risk, and 16% underestimated their risk. Nineteen of 51 (37%) BRCA2 carriers estimated their risk within 5% of published risk figures; 61% overestimated their risk, and 2% underestimated this risk. The range of perceived risk for ovarian cancer after surgery was 0% to 95%. Twenty-eight of 74 (38%) BRCA1 carriers estimated their risk of ovarian cancer after surgery to be less than 5%. Thirteen of the 74 (18%) estimated their risk to be 0. Twenty-seven of 51 (53%) BRCA2 carriers estimated their risk to be less than 5%, and 11 of the 51 (22%) estimated their risk to be 0.
There was no significant correlation between current age and perceived risk of breast cancer, before or after surgery, or between age and perceived risk of ovarian cancer after surgery. However, there was a significant negative correlation between age and perceived risk of ovarian cancer before surgery (−0.20, p = 0.02). Women under the age of 50 years at the time of surgery perceived a significantly greater risk reduction from prophylactic surgery (mean 30% reduction in risk) compared with women over the age of 50 years (mean 13% reduction in risk) (p = 0.02) (women with a personal history of breast cancer or prophylactic mastectomy were excluded from this analysis).
A number of women were surveyed twice to determine if their perception of risk was stable over time. Forty-three women in this study completed two follow-up questionnaires; the first at a mean of 12 months post-surgery (range 6–18 months) and the second at a mean of 37 months post-surgery (range 35–42 months). Their perception of their breast and ovarian cancer risks did not change significantly over this time period (Table 3). Perceived risks reported by BRCA1 and BRCA2 carriers were combined for this analysis. When BRCA1 and BRCA2 were analyzed separately, there were no differences with the exception of BRCA2 carrier’s perception of their ovarian cancer risk before surgery. Their perception of risk increased from a mean of 31% at 1 year to 43% at 3 years post-surgery (n = 18, p = 0.02).
Table 3. Perceived risk of breast and ovarian cancer at 1 and 3 years following surgery
|Follow-up time: (months) mean (range)||12 (6–18)||37 (35–42)|
|Breast risk before surgerya (n = 17) (%)||64||65|
|Breast risk after surgerya (n = 17) (%)||39||28|
|Ovarian risk before surgery (n = 43) (%)||44||47|
|Ovarian risk after surgery (n = 43) (%)||5||7|
One of the purposes of genetic counseling and genetic testing for BRCA1 and BRCA2 is to educate women about cancer risk and to facilitate the selection of cancer risk reduction strategies. Studies reporting on perception of risk for breast and ovarian cancer, before and after genetic counseling, show inconsistent findings. Braithwaite et al. performed a review of five controlled trials and 16 prospective studies that evaluated the impact of genetic counseling (22). They found that knowledge of cancer genetics improved through counseling, but not all prospective studies showed that risk estimation was more accurate after counseling.
A study by Cull et al. evaluated women’s perception of ovarian cancer risk prior to attending a familial ovarian cancer clinic (8). Perceived risk was compared with actual risk based on family history. They found that more women underestimated than overestimated their ovarian cancer risk. Huiart et al. studied the effects of genetic consultation on perception of familial risk of breast and ovarian cancer in low-, moderate- and high-risk women (12). They found that genetic counseling did not affect the perceived risk of the women at high familial risk (who perceived their risk to be high prior to, and after, genetic consultation). McInerney-Leo et al. looked at whether perception of breast and ovarian cancer risk changes following genetic counseling and receipt of genetic test results (9). Perception of cancer risk was assessed prior to testing and 6–9 months after testing. They found that those who tested positive did not have a change in perception of breast cancer risk but did have a significant increase in perception of ovarian cancer risk. These studies examined perception of cancer risk in individuals with a family history of cancer. In contrast, this study evaluates perception of cancer risk only in women found to carry a mutation in the BRCA1 or BRCA2 gene. In addition, we evaluated the impact of prophylactic surgery on risk estimation.
Our study shows that the women understood the effectiveness of the surgery for both breast and ovarian cancer risk. There was no relationship between age at surgery and perceived risk of breast cancer. Perceived risk of ovarian cancer before surgery was inversely correlated with age. Patients are not routinely counseled that lifetime risk is based on current age, and this was evident in their perception of risk. A greater reduction in breast cancer risk was reported by women who were under the age of 50 years at the time of surgery compared with women who were greater than the age of 50 years, consistent with the understanding that premenopausal prophylactic oophorectomy is associated with a decrease in breast cancer risk, whereas postmenopausal oophorectomy is not.
Although perceived risk of ovarian cancer decreased significantly after surgery, a large range was reported. Of note, approximately 20% of both BRCA1 and BRCA2 carriers believed that their ovarian cancer risk was eliminated following prophylactic surgery; although the risk of peritoneal cancer is small, this risk should be included in the counseling discussion.
Based on estimates taken at 1 and 3 years post-surgery in this cohort, we found that perceived breast and ovarian cancer risk did not change over time. This contrasts with previous studies that have shown that perceived risk often reverts to pre-counseling levels (10, 23). Watson et al. examined perception of breast cancer risk before and after genetic counseling (10). They showed that although perception of breast cancer risk was more accurate directly after counseling, the number of subjects reporting the correct risk declined at 1 year following counseling. Lloyd et al. reported similar results. They found that despite clinic attendance, 66% of women continued to overestimate or underestimate their lifetime risk of developing cancer. However, in both of these studies, the women were not known BRCA1 or BRCA2 carriers.
A limitation of this study is the retrospective design. Study subjects were not asked to report their risk perception before undergoing surgery. As a result, the perception of risk of women who elected to have screening instead of surgery was not evaluated in this study. They may have had lower risk perception reflected by the willingness to forgo primary prevention in favor of screening. In addition, the pre-surgery perception of risk is reported months after surgery, a possible source of recall bias. The response rate may also introduce bias. Twenty-five percent of women agreed to participate but did not return the questionnaire. Lastly, although the data support that perception of risk was stable over time, the range of time since surgery is broad (6–48.5 months). Perception of risk may have been different if subjects were surveyed at exactly 1, 2 and 3 years after surgery.
Women who learn that they carry a BRCA1 or BRCA2 mutation and subsequently go on to have prophylactic salpingo-oophorectomy have an accurate perception of their breast cancer risk, but many have an inflated perception of their ovarian cancer risk before and after surgery. This risk perception appears to be stable over time. Women who continue to perceive themselves at high risk for ovarian or ovarian-like cancers after prophylactic salpingo-oophorectomy may benefit from additional counseling.
This study is supported by the Toronto Ovarian Cancer Research Network with funds raised by the Toronto Fashion Show, the Canadian Institutes of Health Research, The Kristi Piia Callum Memorial Fellowship in Ovarian Cancer Research, and the University of Toronto Open Fellowship.