OTHER GENES INVOLVED IN BREAST CANCER SUSCEPTIBILITY
A number of studies have now tested for BRCA1 and BRCA2 mutations in population-based series of breast cancer cases, and thus the contribution of these two genes to familial aggregation can be assessed. Peto et al. (1999) identified 30 BRCA1/2 mutations in 617 breast cancer patients diagnosed before age 46. Interestingly, only five of their mothers or sisters had had breast cancer, compared with 64 in the relatives of the 587 noncarriers. After allowing for the number of breast cancers that would be expected at population rates, and assuming a mutation sensitivity of 64 %, this would equate to approximately 16 % of the observed familial risk being due to BRCA1 and BRCA2. In a previous linkage study, over 80 % of families with six or more cases of breast cancer were found to be linked to either BRCA1 or BRCA2, but the proportion fell to 40 % in families with four or five cases, leaving room for other genes (Ford et al. 1998).
In the last few years, four genes have emerged as strong candidate breast cancer susceptibility genes – CHEK2, BRIP1, ATM, and PALB2 (Foulkes 2008) – and the candidacy of other so-called “moderate-penetrance” breast cancer susceptibility genes (such as RAD50 and NBS1) have been advanced. It is important to recognize that identified mutations in all these genes are either too low in frequency or of too low penetrance to account for more than a small part of the remaining familial cases. In particular, BRIP1 appears to have a very limited role (Seal et al. 2006; Wong et al. 2011), but has recently been associated with ovarian carcinoma. In families with a strong family history of breast cancer, these moderate-risk alleles could be clinically useful (Byrnes et al. 2008). The same authors have suggested that the breast cancer penetrance for at least some PALB2 alleles is as high as some BRCA2 alleles (Erkko et al. 2008; Southey et al. 2010).
Approximately 50 % of women with Li–Fraumeni syndrome develop breast cancer, particularly at young ages. The RR for breast cancer in women under 45 years of age in Li–Fraumeni syndrome families has been estimated at 17.9 (Garber et al. 1991). However, the proportion of cases of breast cancer overall is likely to be extremely small because fewer than 1 % of breast cancer cases can be demonstrated to have germline TP53 mutations (Borresen et al. 1992), and even among early-onset breast cancer, TP53 mutations have a minor role (Sidransky et al. 1992).
Other genes associated with a moderate to high penetrance for breast cancer include PTEN (Cowden Syndrome), STK11 (Peutz–Jeghers syndrome), and CDKN2A (hereditary malignant melanoma). The contribution of mutations in all of these three genes to breast cancer incidence is negligible, but the risks for carriers can be as high as 50 %. Klinefelter syndrome (XXY) is also associated with an increased risk for breast cancer.
Epidemiological data have demonstrated an increased RR of early-onset breast cancer in the female relatives of cases of ataxia telangiectasia, and the proportion of breast cancer due to heterozygosity for ATM mutations has been estimated to be 7 % overall, a smaller proportion at later ages at diagnosis than at earlier ages. Initial studies had suggested that a very small proportion of breast cancer cases, even in radiosensitive subjects, was attributable to mutations in this gene. One missense mutation, ATM*7271T>G, has been associated with a high risk of breast cancer in a few families (Chenevix-Trench et al. 2002), but it is not frequently implicated in hereditary breast cancer (Szabo et al. 2004). The most comprehensive genetic epidemiological study suggested that the RR for breast cancer in
ATM carriers is 2.2 (95 % CI 1.2–4.3), but is nearly 5 for those diagnosed under 40 years of age (Thompson et al. 2005). This study was supported by a definitive molecular study which identified deleterious mutations in 12 of 443 probands from UK familial breast cancer pedigrees and in 2 in 521 controls, leading to an estimated RR of 2.37 (95 % CI, 1.51-3.78) (Renwick et al. 2006), a number remarkably similar to the estimate from epidemiological studies. This study also emphasized the need for very large studies when looking at rare alleles with moderate risks. The previous negative studies listed above were underpowered to detect relative risks as low as 2.4. RR between 2.0 and 5.0 and have been termed moderate-risk alleles, whereas those below this are generally grouped together as low-risk alleles.
A rare novel type of mutation resulting in somatic mosaicism has been uncovered by next-generation sequencing. Truncating somatic mutations in PPM1D have been identified in small percentage of breast and ovarian cancers, but interestingly, they appear not to be heritable. They are only present in lymphocytes and not in the Tumours occurring in those carrying these mutations. Also, they are not present in the offspring of carriers thus far tested and identifying PPM1D mutation carriers will be a challenge. The relative risk for breast and ovarian cancer, however, is substantial (likely>20) and would warrant preventive intervention (Ruark et al. 2013).
Several score of low-risk breast cancer susceptibility alleles have been discovered through powerful genome-wide association studies (GWAS) (Easton et al. 2007), and there are probably hundreds more such alleles, but currently their usefulness in the clinic is very limited. In the future, it is hoped that risk calculation programs, such as BOADICEA (Antoniou et al. 2004), and perhaps BRCAPRO, will be able to incorporate these genotypes into the risk model. If so, low-risk alleles could be of considerable interest in planning screening protocols (Pharoah et al. 2008). Until that time, lower-risk alleles are more relevant to the clinician as modifiers of high-penetrance alleles. For example, SNPs at ESR1, TOX3, and one on 2q (not in a gene) have been found to modify breast cancer risk in BRCA1 carriers. By contrast, so far 11 SNPs modify risk in BRCA2-related breast cancer. This is probably related to the fact that more SNPs are known to modify ER-positive breast cancer than are known to modify ER-negative breast cancer. Several alleles also modify risk for ovarian cancer in BRCA1 or BRCA2 mutation carriers (Barnes and Antoniou 2012). Clearly, many more such loci will be identified, and perhaps this will be of clinical use in the next decade or so, but currently the effect sizes are not large enough to warrant altering clinical recommendations for mutation carriers.