It is estimated that between 5% and 10% of women with breast cancer have an inherited mutation in a cancer susceptibility gene.[1] Mutations in the BRCA1 and BRCA2 genes account for most of these cases. However, BRCA1/2 mutations are most likely to be identified in families with multiple cases of ovarian and earlyonset breast cancer.[2] The genetic basis of cancer in families in which multiple women have developed breast cancer, even at an early age, appears to be much more heterogeneous given that many do not harbor mutations in BRCA1 or BRCA2.[2] For example, one study found that among 237 families containing at least four cases of female breast cancer, only 65% were attributable to BRCA1 or BRCA2 mutations.[3] While other hereditary cancer syndromes also exist and should be considered based on suggestive family history constellations, together these account for only a small proportion of hereditary breast cancer. These syndromes, including Li-Fraumeni, Cowden, and Peutz-Jeghers, have distinct clinical features and are reviewed elsewhere.[4,5] Mutations in BRCA1, BRCA2, and the genes associated with the aforementioned syndromes all confer substantially elevated risks of breast and other cancers. But other genes have been identified that appear to confer more modest breast cancer risks, including ATM and CHK2, and their contribution to the genetic epidemiology of familial and hereditary breast cancer has yet to be fully elucidated.[2] Molecular and Population Genetics BRCA1 and BRCA2 function as tumor-suppressor genes and fall under the umbrella of "caretaker genes," given their role in DNA damage signaling and repair, transcriptional regulation, and cell-cycle regulation.[1] The BRCA1 gene is located on chromosome 17q21 and spans 5.6 kilobases of genomic DNA, comprising 1,863 amino acids.[1] BRCA2, on chromosome 13q12 is much larger, encompassing 10.2 kilobases and coding for 3,418 amino acids.[1] Given the large size of these genes, it is perhaps not surprising that over 1,500 distinct mutations and variants have been documented in each gene, including five rarely occurring genomic rearrangements in BRCA1 that are not detectable by sequencing.[6-8]. Some recurrent mutations have been identified in specific geographic or ethnic groups. Most notably, in Ashkenazi Jewish individuals (descended from central or eastern Europe), two BRCA1 mutations (187delAG, 5385insC) and one BRCA2 mutation (6174delT) have been identified with a population frequency of 1 in 40 (compared to the incidence in the general population of up to 1 in 500).[5,9] It is relatively rare for high-risk Jewish families to harbor a mutation other than one of these three.[10,11] In the United States, virtually all clinical testing is performed by the commercial laboratory that holds the gene patents, Myriad Genetic Laboratories. As of February 2005, full gene sequencing of BRCA1 and BRCA2 costs approximately $3,000, and includes analysis for the five BRCA1 rearrangements. Testing for the three recurrent mutations observed in Ashkenazi Jews is $415, and single mutation analysis is $350. Many insurance companies will pay all or part of the costs of BRCA1/2 testing, usually on the basis of medical necessity to the person requesting testing. Risk Assessment A three-generation pedigree, as shown in Figure 1, is a convenient form of recording family history to determine whether an individual is at high risk of cancer. Age at diagnosis, current age, age and cause of death, surgical procedures (including those performed for benign conditions such as removal of the ovaries and uterus), precancerous conditions, and relevant environmental exposures should also be documented, as well as the ethnic/ racial background of the patient's grandparents (eg, "Irish"). Features of the family history that are suggestive of a BRCA1/2 mutation include the presence of (a) two or more women on the same side of the family diagnosed with breast cancer prior to age 50, (b) breast and ovarian cancer in the same woman, (c) breast and ovarian cancer in two or more relatives on the same side of the family, and (d) male breast cancer, especially with a family history of breast and/or ovarian cancer. In addition, Jewish women diagnosed with breast cancer (especially premenopausally) or ovarian cancer, even in the absence of a family history of these cancers, should be referred for genetic counseling and consideration of genetic testing. Mutations in the genes are passed down in an autosomal dominant fashion, such that each child of a parent with a BRCA1 or BRCA2 mutation has a 50% chance of also testing positive. Thus, it is important to assess family history on both the paternal and maternal side and to be aware that transmission of the mutated gene may occur through males. Specific referral guidelines, which may be useful in the identification of high-risk patients, have been developed.[12,13] Quantitative Assessments
Although a qualitative impression of risk is useful for making determinations about who may be referred for genetic counseling, quantitative assessments of carrier probability may be useful for patients who are considering BRCA1/2 testing, especially in light of the potential expense. The most easily referenced estimates are empiric, based on the reported family history of over 30,000 women tested clinically through Myriad Genetic Laboratories.[10,14] Data are organized into two easy-to-read tables, which may be downloaded onto palm or handheld personal computers. Probabilities of testing positive for a BRCA1 or BRCA2 mutation are provided based on whether or not the patient is of Ashkenazi Jewish ancestry, personal history of breast or ovarian cancer, and family history of these cancers by age at diagnosis of breast cancer (ie, less than age 50 or not). Important limitations of these data are the lack of independent verification of cancer diagnoses, and the fact that family history beyond that observed in first- and second-degree relatives may not have been routinely recorded. Other Risk Assessment Models
Other models for predicting BRCA1/2 carrier probability have been developed.[15] Thus far, only the BRCAPRO model has been subject to rigorous validation.[16] This model, which is available at no cost after completing a licensing agreement, can be downloaded from a comprehensive breast cancer risk assessment package known as CancerGene.[17] It relies on estimates of BRCA1/2 mutation prevalence and penetrance to apply Bayesian (a priori) probabilities of carrier status. Because a full-pedigree structure (of both affected and unaffected individuals, including all current ages, ages at diagnosis, and death) is required, providing rapid risk assessments at the time of the initial patient inquiry may not be feasible. Each model has its own set of strengths and limitations, and ease of use within the clinic. However, one critical caveat is that it is important for clinicians to always consider the limitations in pedigree-based risk assessment such as small family size, few females, limited family history information, and death at young ages. In addition, the decision to undergo genetic testing may not always be based merely on the likelihood of testing positive, but on the patient's worry about testing positive and the degree of reassurance, if any, that may be obtained as a result of pursuing testing. As discussed below, it is valuable to elicit these concerns during the course of genetic counseling. Genetic Counseling and Testing Although healthy individuals are often likely to seek out genetic counseling for the purposes of being tested themselves, it is ideal to initiate BRCA1/2 testing in the relative who is most likely to test positive on the basis of a personal history of ovarian cancer or early-onset breast cancer. In Figure 1, the proband (indicated by an arrow) is not the family member who is most likely to test positive, given that she was diagnosed with breast cancer after menopause. However, because there are no other living affected relatives available to test, it is still preferable to test her before testing healthy relatives concerned about their risk. Possible Outcomes of Testing
There are three possible outcomes of BRCA1/2 testing.[18] Again referring to Figure 1, it is possible that the proband will test positive for a deleterious mutation in the BRCA1 or BRCA2 gene. In that instance, there may be medical implications for her (as discussed later) and relatives such as her daughters and son (not pictured). Given that her maternal histo- ry is noncontributory, she would be counseled that her paternal relatives have a 50% chance of carrying the mutated gene (or 25% if they are children of her paternal uncles). Testing for these individuals would involve analysis only for the deleterious mutation. If they test negative for the familial mutation, they could be counseled that a true negative result means that their risk of developing cancer is thought to be equivalent to that of the general population, although personal risk factors and the cancer history of the other side of their family would also need to be assessed. A third outcome of testing arises when no risk-conferring mutation in BRCA1 or BRCA2 is identified in the proband. This result is uninformative because it does not rule out the possibility that the proband still may have an inherited susceptibility to breast cancer, either due to a mutation that could not be detected by conventional methods or to a mutation in another gene. The probability of the former is estimated to be up to 15%.[8] In Figure 1, given the patient's later age at diagnosis, it is also possible that she may have developed sporadic breast cancer in a family that does in fact harbor a BRCA1 or BRCA2 mutation. Another type of uninformative result arises when a genetic variant or DNA change is identified that cannot be classified as deleterious or benign. In such instances, healthy relatives should not be tested for the variant for clinical purposes, as it does not serve to clarify their cancer risk. Counseling about cancer risk in "uninformative families" is complex and should be tailored to the specific history noted in the family.[18] Counseling and Consent
The genetic counseling and informed consent process should include a thorough discussion encompassing risk assessment, psychosocial assessment, and a review of test result inter- pretation, management guidelines, and family implications.[19] In addition, patients should consider the potential benefits, limitations, and risks associated with genetic testing. Fortunately, most studies have shown that at least in the context of genetic counseling and testing obtained in research settings, there do not appear to be significant adverse psychosocial effects among women who learn their results.[ 20] However, this does not diminish the potentially life-altering impact of genetic testing for an individual woman and her family, who may be faced with difficult decisions about screening and prevention as well as the various emotional responses to living with an increased cancer risk and the possible implications to future generations. Cancer Risks Breast and Ovarian Cancer
Cancer risks associated with mutations in BRCA1 and BRCA2 are quite variable depending on the population studied (Table 1).[3,21-31] Estimates from high-risk, registry-based kindreds and clinic-based populations have tended to yield the highest cancer risks of breast and ovarian cancer. For example, studies in such groups have found that the lifetime breast cancer risks in BRCA1 and BRCA2 carriers range from 73% to 87%.[21,22] The lifetime risk of ovarian cancer in BRCA1 carriers is approximately 40%, and in BRCA2 carriers, 25%.[3,21,22] Although data from unselected individuals (ie, those not studied on the basis of a strong family history of these malignancies) have revealed much lower estimates than those quoted above, a recent pooled analysis of data from 22 studies involving 8,139 index cases from unselected families demonstrated that the average risk of breast cancer was 65% in BRCA1 carriers and 45% in BRCA2 carriers.[23] The average cumulative risk of ovarian cancer was 39% and 11% in BRCA1 and BRCA2 carriers, respectively.[23] A novel approach to deriving cancer risk in carriers was undertaken by the New York Breast Cancer Study Group.[24] Instead of using "likelihood" approaches to estimating mutation status of relatives, BRCA1/2 testing was actually performed in relatives of unselected Jewish breast cancer patients with positive BRCA1/2 test results. The 71% cumulative risk of breast cancer to age 70 that they identified in carriers is comparable to that found in the highest-risk kindreds, as is the risk of ovarian cancer in BRCA1 carriers (46%).[24] The BRCA2 risk was lower, at 12% by age 70, although it increased to 23% by age 80.[24] There do appear to be differences in age-specific risks for breast and ovarian cancer between BRCA1 and BRCA2 carriers. Overall, the risk of breast and ovarian cancer among women under age 50 is lower in BRCA2 carriers than in BRCA1 carriers, but at all ages in both groups, the risk is significantly increased over that of the general population.[3,23] That said, ovarian cancer rarely occurs in women under age 30. Risk for this cancer begins to rise appreciably after age 35, throughout the 40s, and after age 50.[23] Second Malignancies After Breast Cancer
It is well established that BRCA1/2 carriers who have had unilateral breast cancer are at elevated risk of developing contralateral primary breast cancers. Studies have shown that the lifetime risk of such occurrences is between 40% and 65%, with the risk at 10 years postdiagnosis being as high as 30%.[21,22,25-27] By comparison, sporadic breast cancer patients face a 20-year cumulative risk of contralateral breast cancer of up to 20%.[28] With respect to the risk of metachronous ipsilateral breast cancer in carriers, the largest studies performed to date demonstrate that the 10-year actuarial risk is roughly between 11% and 14%, which is similar to the risk for young breast cancer patients without an identified inherited susceptibility.[ 27] Continued investigation in large cohorts over an extended period of time will be important to further clarify these risks and determine the incidence of late events. For women with early-stage breast cancer who have a good long-term prognosis, there is concern not only about the risk of a second breast cancer but also the risk of ovarian cancer. As expected, these patients have a very high 10-year actuarial risk of ovarian cancer after their diagnosis, which is approximately 13% in BRCA1 carriers and 7% in BRCA2 carriers.[29] In the cohort of stage I breast cancer patients studied by Metcalfe et al, 25% of their mortality over 10 years was attributed to ovarian cancer.[29] Therefore, management strategies in breast cancer patients need to emphasize the importance of aggressively addressing their risk of ovarian cancer. Other Cancers
The most substantial risks of cancer conferred by mutations in BRCA1 and BRCA2 are for breast and ovarian cancer. However, female mutation carriers should be informed that there are medical implications to at-risk male relatives with respect to prostate cancer and breast cancer.[30] Although the implications for screening are not clear, the risk of prostate cancer appears to be most significant in BRCA1 and BRCA2 carriers under age 65; however, the absolute risk is difficult to pinpoint.[30] In addition, other cancers have been reported to occur in excess, with overall low absolute rates (ie, less than 10%). For example, both BRCA1 and BRCA2 carriers have an increased risk of pancreatic cancer, whereas risks of melanoma, stomach cancer, and possibly other cancers appear to be elevated in BRCA2 carriers only.[21,25,30] Of note, by order of magnitude, the risk of fallopian tube cancer in BRCA1 and BRCA2 carriers is also quite significant (relative risk of 120, translating to a cumulative risk of 3%),[21] although pathologically, these cancers are very similar to ovarian cancers. Most studies have not shown an excess risk of colon cancer in carriers. Ongoing research will clarify the tumor spectrum associated with BRCA1/2 mutations. Summary
In the decade or so since the cloning of BRCA1 and BRCA2, an abundance of data have been published about cancer risk in BRCA1/2 carriers. While there are certainly some disparities in the estimated cancer risks, in many instances the confidence intervals within studies may be wide, and between studies, may overlap significantly. Patients should be informed about how cancer risks are derived and that precision in risk estimates is not possible. It is clear, however, that relative to the general population, BRCA1/2 carriers face significantly elevated risks of breast and ovarian cancer that need to be addressed through appropriate and aggressive management strategies. Risk Modifiers Genetic Factors
One potential explanation for the wide variability in cancer risks among mutation carriers is the effect of individual risk modifiers on penetrance. It is possible that specific mutations in BRCA1 or BRCA2 may have a different effect on risk based on their effect on the protein product or the location of the mutation within the gene.[32] For example, women with a mutation in the ovarian cancer cluster region (nucleotides 3059-4075 and 6503-6629) of the BRCA2 gene appear to have a higher risk of ovarian cancer and a diminished risk of breast cancer compared to women with mutations in other parts of the gene.[33,34] Variations in other genes are also important considerations. For instance, polymorphisms in genes that are related to the metabolism of sex hormones or DNA repair, such as the androgen receptor gene, AIB1, RAD51, and HRAS1, may affect risk in carriers, although the data are too preliminary to be utilized for tailored risk assessments in carriers.[32] Nongenetic Factors
Reproductive, environmental, and lifestyle factors may also affect risk in carriers. Rapid proliferation of breast epithelial cells occurs during puberty and pregnancy, which may result in the loss of somatic BRCA1 or BRCA2 alleles.[35] Thus, it is of interest to determine what impact reproductive factors such as parity, pregnancy, and breast-feeding have on cancer risk among carriers, especially given the tendency for high-risk women to develop premenopausal breast cancer. A neither strong nor consistent association of these risk factors with risk has emerged. Data are variable as to the effect of age at first full-term pregnancy, with some studies indicating no association and others suggesting a protective effect of late age at first full-term pregnancy.[36,37] Interestingly, data have shown that tubal ligation reduces the risk of ovarian cancer in BRCA1 but not BRCA2 carriers, and that this effect is magnified in carriers who also used oral contraceptives.[32] The specific mechanism by which tubal ligation may reduce risk has not been elucidated. Additionally, it is important to note that many of the studies examining the significance of cancer risk modifiers in mutation carriers are limited by small sample size and thus must be viewed as preliminary in nature.

• Hormones-Studies suggest that exogenous hormones may also affect risk. Although data in the general population and in women with a family history of ovarian cancer have demonstrated that birth control pills reduce the risk of ovarian cancer, two large studies in mutation carriers reached opposite conclusions about this issue.[32] With respect to breast cancer risk, one large case-control study found an elevated risk of breast cancer in BRCA1 carriers who used the pill prior to 1975, before age 30 years, or for more than 5 years.[38] No elevation in risk of breast cancer was observed for BRCA2 carriers.[38] Thus, mutation carriers who are considering initiating use of oral contraceptives must consider the limitations in data about the effect on breast and ovarian cancer risk, and must also weigh the other potential benefits and risks associated with such use. Another hormone of interest is tamoxifen(Drug information on tamoxifen), an antiestrogenic drug used in the adjuvant treatment of estrogenreceptor- positive breast cancer and as a chemopreventive agent in highrisk women. A large retrospective case-control study demonstrated that, similar to the effect observed in sporadic breast cancer patients, tamoxifen reduced the risk of contralateral breast cancer in BRCA1 and BRCA2 carriers overall by 50%, with a greater effect seen in the former (62% vs 37%).[39] This difference is especially interesting given that BRCA1 carriers tend to develop estrogen-receptor- negative breast cancers.[40] Indeed, analysis of 19 BRCA1/2 mutation-positive participants in the National Surgical Adjuvant Breast and Bowel Project prevention trial (NSABP-P1) who developed breast cancer after randomization to tamoxifen or placebo showed that tamoxifen reduced the incidence in BRCA2, but not BRCA1, carriers.[41] However, the small sample size and lack of statistical significance of these results reduces the generalizability of these findings to clinical practice. More data are needed to adequately assess the effectiveness of tamoxifen in mutation carriers as a preventive measure to reduce the incidence of first or second breast cancers. One observation that is consistent with tamoxifen's efficacy in reducing the risk of breast cancer is the finding that oophorectomy in carriers less than age 50 (ie, premenopausal) has an effect of similar magnitude. For example, a retrospective study of 99 BRCA1/2 carriers followed for a median of 8.8 years revealed that oophorectomy in premenopausal women reduced their risk of breast cancer by 53%.[42] Among women who have had breast cancer, the effect of oophorectomy on the risk of second breast cancers appears to be similar.[26] In addition, tamoxifen and oophorectomy appear to have synergistic effects, given that BRCA1/2-positive breast cancer patients who completed both interventions had lower risks of second primary breast cancers than women who underwent one or the other treatment.[26,39] However, as discussed later, the primary indication for oophorectomy is to reduce the risk of ovarian cancer, but the compelling evidence for its role in breast cancer risk management is reason to reinforce to carriers the benefit of obtaining this surgery prior to menopause, so that maximum benefits can be realized.

• Diet and Exercise-Many BRCA1/2 mutation carriers inquire about the effect of diet and exercise on breast cancer risk. Although there is an association between alcohol(Drug information on alcohol) intake and breast cancer risk in the general population, there are no data about this or other dietary factors in carriers.[43] However, it has been postulated that given the genes' role in DNA repair, studies of drugs or nutritional supplements that reduce oxidative damage may be of interest.[32] One study showed that higher levels of physical activity during adolescence and having a normal weight at age 21 were associated with a delayed onset of breast cancer.[ 24] This area of modifiable risk factors, as with the others discussed above, will require further investigation before such information can be adequately considered in risk estimates or management guidelines. Nonetheless, given the other known health benefits of physical exercise, for example, it is reasonable to encourage women to pursue this activity.