Purpose: To review new evidence on the benefits and harms of risk assessment, genetic counseling, and genetic testing for BRCA-related cancer in women for the U.S. Preventive Services Task Force.
Data Sources: MEDLINE and PsycINFO (January 2002 to December 31, 2012), Cochrane Central Register of Controlled Trials and Cochrane Database of Systematic Reviews (4th Quarter 2012), Scopus, and reference lists were searched for English-language studies of benefits and harms of risk assessment, genetic counseling, genetic testing, and interventions to reduce BRCA-related cancer and mortality.
Data Synthesis: Thirteen general risk models, such as the Gail model, are modest predictors of individual risk for breast cancer (c-statistic, 0.55 to 0.65). Five familial risk models for nongenetics specialists to guide referrals to genetic counseling accurately predict individual risk for BRCA mutations (c-statistic, >0.80). No studies reported harms of risk assessment. Sixteen studies indicated that genetic counseling decreases cancer worry, anxiety, and depression; increases the accuracy of risk perception; and decreases intention for mutation testing.
Thirty-two new studies and 38 earlier studies provided data for meta-analysis estimates of the prevalence and penetrance of BRCA mutations. Prevalence varies by population: 0.2 to 0.3 percent in general populations, 3 percent in women with breast cancer, 6 percent in women with breast cancer onset before age 40 years, 10 percent in women with ovarian cancer, and 20 percent in high-risk families. Among Ashkenazi Jewish women, prevalence is 2 percent in unselected populations and 10 percent in high-risk families. The penetrance of BRCA mutations differs by test result. Breast cancer penetrance to age 70 years if the test is positive is 46 to 71 percent for BRCA1 or BRCA2; ovarian cancer penetrance is 41 to 46 percent for BRCA1 and 17 to 23 percent for BRCA2. No estimates were available for women with variants of uncertain significance. The standardized incidence rate for breast cancer is 3.81 (95% CI, 3.06 to 4.75) for uninformative negative test results and 1.13 (95% CI, 0.81 to 1.58) for true negative results. Estimates for ovarian cancer were highly heterogeneous. Breast cancer worry and anxiety increased after testing in women with positive results and decreased in others, although results differed across studies. Risk perception improved after receiving test results.
No trials of the effectiveness of intensive screening for breast or ovarian cancer in women who are mutation carriers have been published. False-positive rates, unnecessary imaging, and unneeded surgery were higher in women undergoing intensive screening. Most women experienced no anxiety after screening with magnetic resonance imaging, mammography, or clinical breast examination, although women recalled for additional testing had transient anxiety. There are no trials of risk-reducing medications specifically in women who are mutation carriers. Tamoxifen and raloxifene reduced invasive breast cancer by 30 to 68 percent in placebo-controlled trials enrolling women with various levels of risk; tamoxifen had a greater effect than raloxifene in a head-to-head trial. Results suggested that reduction was greater in women with more relatives with breast cancer, but confidence intervals overlapped and results were not specific for women who are mutation carriers. Tamoxifen and raloxifene increased thromboembolic events and tamoxifen increased endometrial cancer and cataracts. In high-risk women and women who are mutation carriers, risk-reducing mastectomy reduced breast cancer by 85 to 100 percent and breast cancer mortality by 81 to 100 percent; risk-reducing salpingo-oophorectomy reduced breast cancer by 37 to 100 percent, ovarian cancer by 69 to 100 percent, and all-cause mortality by 55 to 100 percent. Some women experienced physical complications of surgery, postsurgical symptoms, or changes in body image; some had improved anxiety.
Limitations: Including only English-language articles and studies applicable to the United States; varying number, quality, and applicability of studies.
Conclusions: Risk assessment using familial risk models to guide referrals is accurate. Genetic counseling reduces distress, improves risk perception, and reduces intention for testing. Genetic testing provides risk estimates for specific populations depending on test results. A true negative test indicates no increased risk for breast cancer. The effectiveness of intensive screening is not known, but it increases false-positive results and procedures. Tamoxifen and raloxifene reduce risk for breast cancer, but have adverse effects. Risk-reducing mastectomy and salpingo-oophorectomy are effective in reducing breast and ovarian cancer. Several evidence gaps remain and additional studies are necessary to better inform practice.