Judy E. Garber, MD, MPH
Director
Dana-Farber Cancer Risk and Prevention Clinic
Dana-Farber Cancer Institute
Associate Professor of Medicine
Harvard Medical School
Boston, Massachusetts
The last decade has witnessed the evolution of cancer genetics from a concept to an established aspect of patient care. Familial breast cancer has been recognized for more than a century, but today it is possible to use the tools of cancer genetics to both recognize women without a dramatic family history who still have a remarkable breast cancer risk, and to distinguish women within a strongly affected family who did not inherit the burden of an unusual breast and ovarian cancer risk. Among the continuing challenges in breast cancer genetics is that additional unidentified genes may confer lesser but important risks; this would account for the majority of families with a strong breast cancer history that remain genetically undefined today.
BRCA1 and BRCA2, the genes that underlie the Hereditary Breast-Ovarian Cancer Syndrome (HBOS), account for less than 5% of all breast cancers. However, they account for a greater proportion of breast cancer in younger women. The Centers for Disease
Control and Prevention (CDC) strongly encourages the collection of accurate information on family history of cancer and other diseases. This is particularly important for breast cancer, where specific features can provide clues to the presence of BRCA1/2 mutations in the family: multiple cases of breast cancer or ovarian cancer in one side of the family, or both breast and ovarian cancer in the family, and particularly younger ages at breast cancer diagnosis. The paternal lineage can be the source of BRCA1/2 mutations, so a diagnosis of ovarian cancer in a paternal grandmother may have significant implications for her descendents.
When a family history suggests that a BRCA1/2 mutation may be present, the best person to test is the one with the greatest chance of carrying a mutation: the woman diagnosed with breast cancer at the youngest age or with ovarian cancer, or a man with breast cancer. If no such person is available or interested in being tested, then other close relatives may certainly be tested, but a negative result is more difficult to interpret in the absence of an identified mutation in affected family members. Testing in the US most often involves determining the sequence of nucleotides in both genes from a single tube of blood, looking for insertions, deletions, or other errors that can lead to a prematurely shortened protein product. The technology is extremely accurate, particularly when performed robotically to minimize typical human errors. However, certain kinds of mutations are not detected by this technology: specifically, deletions of larger segments of DNA. It has been estimated that such large deletion mutations may account for up to 15% of BRCA1/2 mutations, a concern when a negative test result is obtained in a family that has many features consistent with HBOS. It is anticipated that a supplemental test designed to detect the largest deletions may soon be available. Issues such as these highlight the value of having patients who are considering testing meet with a genetic counselor or other trained health professional to review the risks, benefits, limitations, and concerns associated with genetic testing for hereditary breast cancer susceptibility.
Once a mutation is identified in one family member, relatives may be tested to see if they share that specific mutation. The testing is then definitive, so a negative result puts a woman’s breast cancer risk back to a more typical range. A positive result then identifies individuals as having remarkable cancer risks and risk management options can be discussed.
A variation on this testing strategy is used for women of Eastern European Jewish ancestry. These women have a higher chance of carrying one of three specific founder mutations, and so are generally tested for a mutation at one of these three sites first. Then, depending on the strength of the personal and family cancer history, those testing negative for the three-site test may or may not be tested for mutations elsewhere in the two genes. Individuals undergoing testing because one of these “founder” mutations has been identified in a relative will generally be tested for all three instead of only the mutation previously identified in a family member because of the higher frequency of these mutations in this population (1 in 40) compared to the combined prevalence of all mutations in most other populations (1 in 800).
Why be tested at all? Fortunately, many women and families with BRCA1/2 mutations have been willing to participate in research, some of the most important of which has been conducted in New York. The table below summarizes the cancer risks associated with heritable mutations in these genes:
| Gene |
Breast Cancer Risk to age 80 |
Ovarian Cancer Risk to age 80 |
Other Cancers |
| BRCA1 |
50%--85% (increased risk before age 50) |
20%--40% |
Male breast cancer  d; small increased risk of prostate/colon |
| BRCA2 |
50%--85% |
10%--20% |
Male breast cancer 6%--10% lifetime; prostate, pancreas and melanoma d |
|
General population risks to age 80: breast 10%; ovary < 2%.
Several papers in 2004 compared imaging strategies for early detection of breast cancer in BRCA1/2 mutation carriers. Breast MRI was nearly twice as sensitive in these series, with remarkably low sensitivity for mammography and clinical breast examination. The question is whether this was because the study cohorts were younger, and mammography is known to be less sensitive in younger women (perhaps because of denser breast parenchyma) or whether mutation-associated breast cancers are intrinsically different and less visible on mammogram. Breast MRI is also associated with a significant false-positive rate, especially in inexperienced hands. Finally, these studies had relatively short follow-up, and so could not look at the potential impact of these modalities on breast cancer outcome or mortality. Intense research is in progress to improve the options for early detection of ovarian cancer, since transvaginal ultrasound and serum CA-125, the current standards, are known to offer poor test characteristics in many groups.
Important research has demonstrated that the cancer risks associated with BRCA1/2 mutations can be modified by lifestyle factors such as exercise. The most effective ways to reduce the high breast and ovarian cancer risks available today are surgical: prophylactic removal of the ovaries reduces ovarian and fallopian tube cancer risk by > 90%; prophylactic mastectomies with or without reconstruction similarly reduce breast cancer risk. In addition, prophylactic oophorectomy performed when a woman is premenopausal will reduce subsequent breast cancer risk by about 50% over her lifetime, a powerful effect. Certainly, surgical approaches are effective, but the decisions to undergo these surgeries are not easy, are intensely personal, and all require consideration of other issues as well. For mastectomies, reconstruction options, loss of breast sensation, body image issues, and sense of sexual identity predominate. For premenopausal women who are considering oophorectomies, menopausal issues (including loss of libido, sexual discomfort, and vasomotor symptoms) must be faced, and surgical decisions about laparoscopic versus alternative approaches and whether or not to remove the uterus must be evaluated.
The reduction in breast cancer risk by removal of the premenopausal ovaries suggests that the mechanism is reduction in hormones, or at least elimination of cycling hormones. Most studies addressing this issue have included predominantly BRCA1 mutation carriers, yet 80% of the breast cancers occurring in BRCA1 carriers are negative for estrogen and progesterone receptors. Similar hormone receptor negative sporadic breast cancers are not prevented by tamoxifen, for example. These data raise questions about BRCA1-associated tumors and their biology. BRCA1-associated breast tumors have also been shown to comprise a subset of basal-like tumors by microarray and immunohistochemical features, which may have important implications for tumor biology and treatment. Basal-like tumors account for 20%--30% of all breast cancers (but 80% of BRCA1-associated breast cancers), are thought to arise from a specific population of stem cells, and are almost always negative for estrogen and progesterone receptors, and HER2/neu overexpression. Research protocols evaluating chemotherapeutic regimens for BRCA1-associated tumors based on an understanding of the consequences of defects in the DNA-repair functions of BRCA1 are in progress. BRCA2-associated tumors do not appear to have unique histopathologic phenotypes as of this writing. Studies evaluating tumor-specific behavior and prognosis must now treat the two genetic groups independently.
Knowledge of one's BRCA1/2 status can have powerful implications for a woman's own cancer risks and management options. Many of these options are influenced by the ages at which women become aware of their genetic status. In addition, the fact that genetic information may have implications for siblings, children, and other family members can be burdensome for families, but can also be empowering, permitting individuals to avoid breast and ovarian cancer diagnoses and, we hope, cancer deaths.
