Objective
Despite substantial survival benefits of risk-reducing mastectomy (RRM) and risk-reducing bilateral salpingo-oophorectomy (RRBSO) among BRCA mutation carriers, only a minority elect to undergo these procedures. This study investigates factors that might influence decision making regarding prophylactic surgeries among women with BRCA mutations.
Study Design
Unaffected BRCA mutation carriers who were counseled at our center and either underwent prophylactic surgery or participated in a high-risk surveillance program at our institution from 1998 through 2010 were included in the study. Medical records were reviewed and data collected included age, family history, parity, mutation type, history of breast biopsy or cosmetic surgery, and uptake of prophylactic surgeries.
Results
Among 136 unaffected women with BRCA mutations, uptake of RRM was 42% and uptake of RRBSO was 52%. Family history of first- and second-degree relatives being deceased from breast cancer was predictive of uptake of RRM and of RRBSO (odds ratio [OR], 11.0; P = .005; and OR, 15.8; P = .023, respectively), and history of a mother lost to pelvic cancer was predictive of uptake of RRBSO (OR, 7.9; P = .001). Parity also predicted both RRM and RRBSO uptake (OR, 4.2; P = .001; and OR, 5.4; P = .003, respectively). Age at the time of genetic testing and history of breast biopsy or cosmetic surgery were not predictive of RRM uptake.
Conclusion
Perceptions of cancer risk are heavily influenced by particular features of an individual’s family history and may be motivators in preventive surgery more than actual cancer risk estimations themselves. Awareness of subtle factors beyond the statistical risk for cancers is relevant when counseling at-risk women.
The risks of breast and ovarian cancer in women with mutations in the BRCA1 or BRCA2 gene are elevated substantially, with an estimated lifetime risk of breast cancer of 56-84%, and a lifetime risk of ovarian cancer of up to 54% for BRCA1 and 27% for BRCA2. Currently, the management options for women with BRCA mutations are somewhat limited and include: (1) increased surveillance to maximize early detection of cancers, with or without the use of antihormonal agents for cancer prevention, or (2) prophylactic surgeries to reduce cancer risk.
The most effective cancer prevention strategy for individuals at high risk is prophylactic surgery. Risk-reducing mastectomy (RRM) reduces breast cancer risk among BRCA mutation carriers by at least 95%. Risk-reducing bilateral salpingo-oophorectomy (RRBSO) reduces breast cancer risk by 50% when completed before menopause and reduces ovarian/fallopian tube/peritoneal cancer risk by at least 80%. The combination of both RRM and RRBSO at age 40 years has been shown to improve survival more than either intervention alone.
Despite the substantial survival benefits, only a portion of unaffected women with BRCA mutations elect to undergo prophylactic surgeries. In an international study by Metcalfe et al, the United States had the highest uptake of all countries with 36.3% of 703 BRCA mutation carriers undergoing RRM and 71.1% undergoing RRBSO. Poland had the lowest rates with 2.7% for RRM and 34.9% for RRBSO. The variations in uptake rate of prophylactic surgeries worldwide may be influenced by a number of factors including differences in health care access, provider recommendations, and social mores. Individually, perception of risk, body image, and sexuality clearly influence decisions, although studies have shown that the majority of women who have undergone prophylactic mastectomy have high levels of satisfaction with the procedure and no significant changes in levels of self-esteem. Los Angeles, CA, is a demographic with very high access and cultural permissiveness toward plastic surgery, making it an interesting study population for investigation of uptake of preventive mastectomy and breast reconstruction.
The purpose of our study was to: (1) quantify the uptake of prophylactic surgery among BRCA mutation carriers at our center; and (2) investigate factors that might influence decision making regarding prophylactic surgery including age, parity, and family history of cancer. With our large cohort of BRCA mutation carriers and long-standing interest in barriers to risk reduction, we sought to evaluate history of cancer among extended family members and total cancer burden as potential influential factors in the decision-making process.
Materials and Methods
Study subjects were women without a personal history of cancer who had been identified as carriers of deleterious mutations in the BRCA1 and/or BRCA2 genes and were prospectively followed up in the Gilda Radner breast and ovarian cancer screening program. The Gilda Radner registry follows up women who are carriers of a BRCA mutation, both with and without a personal history of cancer. Women who had a personal history of cancer were excluded from this analysis. Year of genetic testing ranged from 1998 through 2010, and study subjects were divided into 2 cohorts: tested before and after 2005. This midpoint in the testing decade was chosen for comparison because in 2005 national guidelines were published regarding BRCA testing and there were significant changes in insurance coverage for testing, greatly increasing awareness and availability of genetic testing.
Study subjects were followed up after testing for a minimum of 1 year and up to 11 years. Medical records, operative reports, and family pedigrees were reviewed after approval from the Institutional Review Board at Cedars-Sinai Medical Center. Data collected included each woman’s age, ethnicity, parity, mutation type, history of breast biopsy, previous cosmetic surgeries, and the uptake of prophylactic surgeries. Pedigree analysis was conducted to assess cancer penetrance and degree of relationship of affected individuals to the proband. First-degree relatives (mother, father, sibling, offspring) and/or second-degree relatives (aunts or uncles, and/or grandparents) with cancer were included. Cancer sites included in our analysis were those associated with hereditary breast and ovarian cancer syndrome: breast, ovarian, fallopian tube, peritoneal, prostate, pancreatic, and melanoma. Family members with a nonspecific diagnosis of cancer were excluded from further analysis.
All data are presented as means and SD or counts and percentages where appropriate. Differences between groupings were compared using t tests or χ 2 tests. Unadjusted odds ratios (ORs) and their 95% confidence intervals were computed using logistic regression modeling. Using forward-stepwise selection methods, an adjusted logistic regression model of a set of significant factors in predicting likelihood of women to have received RRM was constructed. All tests were 2-sided with the level of statistical significance set at P < .05. All statistical analysis was performed using SAS v9.3 (SAS Institute, Cary, NC).
Results
There were 136 unaffected women with BRCA mutations and adequate follow-up time who were included in the study population. A total of 71% were self-identified as being of Ashkenazi Jewish background and the remainder were non-Jewish Caucasian (19%), Hispanic (2%), African American (1%), or other or mixed race (8%). None were first-degree relatives of each other. In all, 65% reported having had some college education or higher ( Table 1 ).
| Demographic | Total (N = 136) | Mastectomy (n = 57) | No mastectomy (n = 79) | P value | OR (95% CI) | |||
|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | |||
| Age at time of genetic testing, y (mean ± SD) | 44.3 ± 11.3 | 39.9 ± 9.9 | 40.2 ± 12.7 | .889 | 1.0 (1.0–1.0) | |||
| Age at time of surgery, y (mean ± SD) | 43.7 ± 8.1 | |||||||
| Age ≥50 y | 44 | 32 | 18 | 32 | 26 | 33 | .870 | 0.9 (0.5–2.0) |
| Caucasian | 115 | 85 | 43 | 75 | 72 | 91 | .841 | 1.2 (0.2–6.8) |
| Other races | 21 | 15 | 14 | 25 | 7 | 9 | ref | |
| Ashkenazi Jewish a | 96 | 71 | 41 | 72 | 55 | 70 | .649 | 1.2 (0.5–2.7) |
| Non-Jewish | 40 | 29 | 16 | 28 | 24 | 30 | ref | |
| Some college or higher (n = 97) | 63 | 65 | 30 | 88 | 44 | 70 | .042 | 3.2 (1.0–10.5) |
| Genetic testing ≥2005 | 61 | 45 | 30 | 53 | 31 | 39 | .121 | 1.7 (0.9–3.4) |
| BRCA1 | 82 | 60 | 37 | 65 | 45 | 57 | .687 | 1.4 (0.7–2.8) |
| BRCA2 | 51 | 38 | 19 | 33 | 32 | 41 | ref | |
| BRCA1 and BRCA2 | 3 | 2 | 1 | 2 | 2 | 3 | — | |
| Parity ≥1 | 85 | 63 | 45 | 79 | 40 | 51 | .001 | 3.7 (1.7–7.9) |
| RRBSO | 71 | 52 | 36 | 63 | 35 | 44 | .030 | 2.2 (1.1–4.3) |
| Mother with breast cancer | 57 | 42 | 26 | 46 | 31 | 39 | .457 | 1.3 (0.7–2.6) |
| Mother deceased from breast cancer | 21 | 15 | 14 | 25 | 7 | 9 | .012 | 3.3 (1.3–8.9) |
| Sister with breast cancer | 19 | 14 | 7 | 12 | 12 | 15 | .629 | 0.8 (0.3–2.1) |
| First-degree family death from breast cancer | 22 | 16 | 14 | 25 | 8 | 10 | .028 | 2.9 (1.1–7.5) |
| Both first- and second-degree family death from breast cancer | 12 | 9 | 10 | 18 | 2 | 3 | .008 | 8.2 (1.7–39.0) |
| Mother with pelvic cancer | 55 | 40 | 18 | 32 | 37 | 47 | .074 | 0.5 (0.3–1.1) |
| Mother deceased from pelvic cancer | 35 | 26 | 14 | 25 | 21 | 17 | .790 | 0.9 (0.4–2.0) |
| Sister with pelvic cancer | 4 | 3 | 3 | 5 | 1 | 1 | .309 | 4.3 (0.4–42.7) |
| Only maternal second-degree affected (any BRCA-related cancer) | 4 | 3 | 2 | 4 | 2 | 3 | .740 | 1.4 (0.2–10.3) |
| Only paternal second-degree affected (any BRCA-related cancer) | 5 | 4 | 2 | 4 | 3 | 4 | .930 | 0.9 (0.2–5.7) |
| Personal history of breast biopsy | 26 | 22 | 10 | 24 | 16 | 21 | .729 | 1.2 (0.5–2.9) |
| Previous cosmetic surgery | 21 | 18 | 7 | 17 | 14 | 19 | .787 | 0.9 (0.3–2.4) |
Risk-reducing mastectomy
The uptake of RRM in our population was 42% (57/136). Of these 57 women, 37 (65%) had a BRCA1 mutation, 19 (33%) had a BRCA2 mutation, and 1 (2%) had both BRCA1 and BRCA2 mutations, the same proportion seen in women who did not undergo RRM ( P = .687). There were no significant differences in the mean age at time of testing among those who had RRM and those who did not (39.9 ± 9.9 vs 40.2 ± 12.7 years; P = .889), however, women without RRM were less parous (parity ≥1 79% vs 51%; P < .001; OR, 3.7). The mean age at time of RRM was 43.7 ± 8.1 years. Uptake of RRM did not increase at age >50 years (32% vs 33%; P = .870) ( Table 1 ). The majority of subjects who opted for RRM (72%) had the procedure within 2 years of genetic testing.
Significantly more women in the RRM group had lost their mother to breast cancer than those who did not have RRM (25% vs 9%; P = .012; OR, 3.3). In addition, more women in the RRM group reported any first-degree relative (mother or sister) being deceased from breast cancer than those who did not have RRM (25% vs 10%; P = .028; OR, 2.9). Those with both first- and second-degree relatives deceased from breast cancer were also more likely to have RRM than those who did not (18% vs 3%; P = .008; OR, 8.2) ( Tables 1 and 2 ). There were no significant differences in uptake of RRM based upon personal history of breast biopsy or previous cosmetic surgery ( Table 1 ). In multivariable modeling the most significant factors to predict the likelihood of RRM uptake were having both first- and second-degree relatives deceased from breast cancer (OR, 11.0; P = .005), already having had at least 1 childbirth (OR, 4.2; P = .001), and more recent genetic testing (OR, 2.8; P = .014) ( Table 2 ).
| Variable | OR (95% CI) | P value |
|---|---|---|
| Both first- and second-degree relative death from breast cancer | 11.0 (2.1–57.9) | .005 |
| Parity ≥1 | 4.2 (1.8–10.0) | .001 |
| Genetic testing >2005 | 2.8 (1.2–6.4) | .014 |
Risk-reducing bilateral salpingo-oophorectomy
The uptake of RRBSO in our population was 52% (71/136). Of these women, 46 (65%) had BRCA1 mutations, 24 (34%) had BRCA2 mutations, and 1 (1%) had both, with a similar distribution in the non-RRBSO group ( P = .492). Women who had RRBSO were older at the time of genetic testing than those who did not (44.7 ± 9.4 vs 34.9 ± 11.7 years; P < .001) and more parous (parity ≥1 82% vs 42%; P < .001) ( Table 3 ). The mean age at time of RRBSO was 45.2 ± 8.4 years. The proportion of women aged ≥50 years who had RRBSO was 51% as compared to only 12% uptake among those aged <50 years ( P < .001; OR, 7.3). The majority who opted for RRBSO had the procedure within 2 years of genetic testing.
| Demographic | Total (N = 136) | RRBSO (n = 71) | No RRBSO (n = 65) | P value | OR (95% CI) | |||
|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | |||
| Age at time of genetic testing, y (mean ± SD) | 44.3 ± 11.3 | 44.7 ± 9.4 | 34.9 ± 11.7 | < .001 | 1.2 (1.1–1.2) | |||
| Age at time of surgery, y (mean ± SD) | 45.2 ± 8.4 | |||||||
| Age ≥50 y | 44 | 32 | 36 | 51 | 8 | 12 | < .001 | 7.3 (3.1–17.6) |
| Caucasian | 115 | 85 | 63 | 89 | 52 | 80 | .164 | 2.0 (0.8–5.1) |
| Other races | 21 | 15 | 8 | 11 | 13 | 20 | ref | |
| Ashkenazi Jewish a | 96 | 71 | 51 | 72 | 45 | 69 | .368 | 1.4 (0.7–3.2) |
| Non-Jewish | 40 | 29 | 20 | 28 | 20 | 31 | ref | |
| Some college or higher (n = 97) | 63 | 65 | 39 | 76 | 35 | 76 | .965 | 1.0 (0.4–2.5) |
| Genetic testing ≥2005 | 61 | 45 | 24 | 34 | 37 | 57 | .007 | 0.4 (0.2–0.8) |
| BRCA1 | 82 | 60 | 46 | 65 | 36 | 55 | .353 | 1.4 (0.7–2.8) |
| BRCA2 | 51 | 38 | 24 | 34 | 27 | 42 | ref | |
| BRCA1 and BRCA2 | 3 | 2 | 1 | 1 | 2 | 3 | — | |
| Parity ≥1 | 85 | 63 | 58 | 82 | 27 | 42 | < .001 | 5.8 (2.6–12.9) |
| Mastectomy | 57 | 42 | 36 | 51 | 21 | 32 | .030 | 2.2 (1.1–4.3) |
| Mother with breast cancer | 57 | 42 | 27 | 38 | 30 | 46 | .338 | 0.7 (0.4–1.4) |
| Mother deceased from breast cancer | 21 | 15 | 16 | 23 | 6 | 9 | .061 | 2.6 (0.9–7.3) |
| Sister with breast cancer | 19 | 14 | 12 | 17 | 7 | 11 | .306 | 1.7 (0.6–4.6) |
| First-degree family death from breast cancer | 22 | 16 | 16 | 23 | 6 | 9 | .041 | 2.9 (1.0–7.8) |
| Both first- and second-degree family death from breast cancer | 12 | 9 | 11 | 15 | 1 | 2 | .020 | 11.7 (1.5–93.6) |
| Mother with pelvic cancer | 55 | 40 | 37 | 52 | 18 | 28 | .004 | 2.8 (1.4–5.8) |
| Mother deceased from pelvic cancer | 35 | 26 | 29 | 41 | 6 | 9 | < .001 | 6.8 (2.6–17.8) |
| Sister with pelvic cancer | 4 | 3 | 3 | 4 | 1 | 2 | .374 | 2.8 (0.3–27.9) |
| Only maternal second-degree affected (any BRCA-related cancer) | 4 | 3 | 1 | 1 | 3 | 5 | .296 | 0.3 (0.1–2.9) |
| Only paternal second-degree affected (any BRCA-related cancer) | 5 | 4 | 1 | 1 | 4 | 6 | .178 | 0.2 (0.1–2.0) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
