A family history of breast cancer in a first-degree relative is reported in 13% of women with the disease.[21] However, only 1% of women have multiple affected relatives, a history
suggestive of a highly penetrant germ-line mutation.
About 25% of familial cancers (or around 3% of all breast cancers) can be attributed to two highly penetrant autosomaldominant genes: BRCA1 and BRCA2 ( Table 23-3 ). The
probability of breast cancer associated with a mutation in these genes increases if there are multiple affected first-degree relatives, if individuals are affected before menopause and/or
have multiple cancers, if there is a case of male breast cancer, or if family members also develop ovarian cancer. The general lifetime breast cancer risk for female carriers is 60% to 85%,
and the median age at diagnosis is about 20 years earlier compared to women without these mutations. The penetrance (i.e., the number of carriers who actually develop breast cancer) can
vary with the specific type of mutation present. Mutated BRCA1 also markedly increases the risk of developing ovarian carcinoma, which is as high as 20% to 40%. BRCA2 confers a
smaller risk for ovarian carcinoma (10% to 20%) but is associated more frequently with male breast cancer. BRCA1 and BRCA2 carriers are also susceptible to other cancers, such as colon,
prostate, and pancreas, but to a lesser extent.
Although BRCA1 and BRCA2 do not show sequence homology, they function in similar pathways and interact with the same multiprotein complexes. Both act as tumor suppressors, as it
is a loss of function that confers the risk of malignancy. A wide variety of functions have been suggested for these proteins, including transcriptional regulation, cell-cycle control,
ubiquitin-mediated protein degradation pathways, and chromatin remodeling. A key function for both appears to be their role in protecting the genome from damage by halting the cell
cycle and promoting DNA damage repair in a complex process that is not yet fully understood. BRCA1 is phosphorylated in response to damage and may transduce DNA damage signals
from checkpoint kinases to effector proteins. BRCA1 is also bound with BRCA2 and RAD51 in a nuclear dot complex—presumably the site of DNA repair.[33] BRCA2 can bind directly
to DNA and functions in homologous recombination for the error-free repair of double-strand DNA breaks.[34] Why loss of these functions specifically affects the breast is unclear. Perhaps
the intermittent proliferation of breast epithelium (as opposed to the constitutive proliferation of other epithelia such as colon or skin) makes this organ more susceptible to the
accumulation of genetic damage, or possibly, other cell types have additional mechanisms for DNA repair that the breast lacks. BRCA1, but not BRCA2, interacts with the ER and is
involved in X chromosome inactivation—two features that may be related to its gender-specific risk.[35] Interestingly, male breast cancers are markedly increased only in families carrying
BRCA2 mutations.
Both genes have a total length of over 80 kb, and hundreds of different mutations distributed throughout the coding region have been reported for each one. The frequency of mutations is
only 0.1% to 0.2% in the general population. Some mutations diminish the function of the genes and increase cancer risk, whereas others might be unimportant sequence variants. Genetic
testing is difficult and often inconclusive unless several family members are affected or unless the individual belongs to an ethnic group with a known high incidence of specific mutations.
[36] For example, people of Ashkenazi Jewish descent have a 2% to 3% risk of three specific mutations. Identification of carriers of clinically significant mutations is important, as
prophylactic mastectomy and/or oophorectomy can reduce the risk of cancer mortality. [31] [37] [38]
In hereditary carcinomas, one mutant BRCA allele is inherited, and the second allele is inactivated by somatic mutation. Although BRCA1 and BRCA2 mutations are rarely found in
sporadic tumors, about 50% of such tumors have decreased or absent expression of BRCA1. In most cases, this is accomplished by a combination of loss of heterozygosity (LOH) and
methylation of the promoter to inactivate both alleles. [39] Hypermethylation of the promoter is detected in 13% of unselected carcinomas but is more common in medullary carcinomas
(67% of tumors) and mucinous carcinomas (55% of tumors)—histologic subtypes that are more commonly found in BRCA1 carriers. A similar mechanism has not yet been described for
BRCA2.
BRCA1-associated breast cancers are more commonly poorly differentiated, have a syncytial growth pattern with pushing margins, have a lymphocytic response, and do not express
hormone receptors or overexpress HER2/neu (an epidermal growth factor receptor that is commonly overexpressed in breast cancer, to be discussed later), as compared to sporadic breast
carcinomas. BRCA2-associated breast carcinomas do not have a distinct morphologic appearance. Initial results using gene expression RNA profiling have revealed that BRCA1, BRCA2,
and subtypes of sporadic cancers can be recognized by their gene expression patterns[40] [41] ( Box 23-1 ). Sporadic carcinomas with an mRNA profile similar to BRCA1 carcinomas have
been termed "basal-like" carcinomas owing to the expression of genes that are characteristic of myoepithelial or possible breast progenitor cells. These results demonstrate that a subset of
sporadic carcinomas have biologic similarities to hereditary carcinomas.
Genetic susceptibility due to other known genes is much less common, and together this group accounts for fewer than 10% of hereditary breast carcinomas. [42] Only five have been
studied sufficiently to be worth noting. Mutations in the cell-cycle checkpoint kinase gene (CHEK2), which is an important component of the recognition and repair of DNA damage and
which activates BRCA1, may account for 5% of familial cases. [43] The risk for a mutation carrier may be as low as 20%. Women with the Li-Fraumeni syndrome (due to a germ-line
mutation in the p53 gene) have an 18-fold higher risk of developing breast cancer before the age of 45. Mutations in p53 also occur in 19% to 57% of sporadic breast carcinomas. Cowden
syndrome ("multiple hamartoma syndrome" due to a mutation of the PTEN gene on chromosome 10q) confers a 25% to 50% lifetime risk of breast cancer in affected women. Mutations in
the PTEN gene are rare in sporadic carcinomas, but LOH is found in 11% to 41%. Further studies will be necessary to determine whether the function of the other allele is altered (e.g., by
methylation). Women with Peutz-Jeghers syndrome (caused by truncating mutations in the LKBI gene) are at increased risk for breast cancer. There is, as yet, no evidence that this gene
plays a role in sporadic carcinoma. The role of the ATM gene in breast cancer susceptibility in ataxia telangiectasia carriers has been intensively studied owing to the high frequency of
carriers in the population (approximately 7%) and the increased sensitivity to radiation exposure leading to concerns about screening mammography. Studies have had mixed results, some
showing an increased risk and others not showing an association. The risk might be dependent on the type of germ-line mutation (e.g., truncating versus missense). Mutations in the ATM
gene in sporadic carcinomas are rare.
All of these genes considered together still leave at least two-thirds of familial risk unexplained. The search for a putative "BRCA3" gene of high penetrance has, as yet, been unsuccessful,
and such a gene might not exist.[44] A polygenic model
in which many weakly penetrant genes (perhaps dozens or hundreds) act in combination to create a spectrum of risk could explain the majority of familial breast cancers, as well as risk in
the general population.[45] [46] [47] This model suggests that most breast cancers arise in a minority of women carrying combinations of these susceptibility genes. The identification of
these genes might allow better stratification of women into low-risk and high-risk groups, which would help to focus efforts toward prevention and early detection in these women. Yet to
be determined are the number of genes that could be involved, the nature of interactions among these genes (e.g., additive or multiplicative), the interaction with environmental factors, and
the possible role of protective alleles. Candidates for such genes have been identified by their ability to modify the expression of known genes such as BRCA1.
Genome-wide approaches (e.g., microarray technology; Box 23-1 ) could play an important role in identifying this potentially very large group of susceptibility genes. One current
approach classifies hereditary cancers by mRNA profiling in the hopes that cancers arising due to the same germline mutation (or mutations) will have similar patterns, as has been
demonstrated with BRCA1 and BRCA2. [48] If true, this would simplify linkage analysis by identifying groups of families likely to carry similar mutations.
Many studies have confirmed that some of the genes involved in hereditary breast cancer (e.g., BRCA1 and p53) are also important in many sporadic cancers. It is hoped that the continued
investigation of the wide variety of naturally occurring mutations and combinations of mutations will provide important clues to breast cancer pathogenesis.