![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Division of Epidemiology, School of Public Health, and Institute of Human Genetics, University of Minnesota, Minneapolis, MN 55454
ABSTRACT
INTRODUCTION
p53
BRCA1
BRCA2
GENETIC FACTORS AND NON-MENDELIAN FAMILIAL BREAST CANCER
ATAXIA TELANGIECTASIA
HRAS1
ESTROGEN RECEPTOR
ESTROGEN METABOLISM
FOOTNOTES
LITERATURE CITED
ABSTRACT 
Aggregation of breast cancer in families is an established risk factor associated with increased incidence of the disease, which is a leading cause of morbidity and mortality among women in this country. Three genes have now been identified that confer increased susceptibility in families with a clear hereditary (i.e., Mendelian) pattern of expression: BRCA1, BRCA2 and p53. However, a significant number of women have an identified family history of breast cancer without clear Mendelian patterns of disease. Such patterns are consistent with the effect of more common genes with lower associated risk. Some evidence is available to implicate three additional genes that fit this category: AT, ESR and HRAS1. An area of active interest is genetically mediated variation in the metabolism of estrogens, a process controlled by several genes, each with more modest effects. The interaction of genes and environmental factors in breast cancer pathogenesis is of considerable public health importance.
KEY WORDS: breast cancer · genes · risk factors · nutrition · reviewINTRODUCTION
It has been firmly established that at least a portion of the occurrence of breast cancer can be ascribed to inherited susceptibility. There are currently three genes (p53, BRCA1 and BRCA2) known that, when inherited in an altered form, confer very high lifetime risks (high penetrance in genetic terms) and high relative risks (in epidemiologic terms). Families segregating a mutation in one of these genes demonstrate an autosomal dominant pattern of susceptibility. These genes seem to be an infrequent cause of breast cancer in the general population and are associated primarily with early-onset breast cancer (Ford and Easton 1995
).
p53
) but not all (Santibanez-Koref et al. 1991) cases of the Li-Fraumeni syndrome. However, this syndrome is rare in the population and has been estimated to account for less than 1% of cases of breast cancer, even at young ages (Sidransky et al. 1992).
BRCA1
) and was recently cloned (Miki et al. 1994). Genetic linkage studies have shown that almost all multiple-case breast-ovarian cancer families are counted segregating mutations in BRCA1 (Narod et al. 1995). At age 40 y, women who inherit a mutation in BRCA1 are at roughly 200-fold greater risk of breast cancer than the general population. This risk decreases somewhat with age, but remains elevated roughly 15-fold to age 70 y. Ford et al. (1995) used two population-based samples of families and estimated the frequency of BRCA1 mutations to be 0.0006 [95% confidence interval (CI) 0.0002 to 0.001] and the proportion of breast cancer cases due to BRCA1 to be 5.3% below age 40 y, 2.2% between ages 40 and 49 y, and 1.1% between ages 50 and 70 y. They concluded that "the majority of breast cancer families with less than four cases and no ovarian cancer are not due to rare highly penetrant genes such as BRCA1 but are more likely to be due to either chance or to more common genes of lower penetrance." Langston et al. (1996) studied 80 women with onset of breast cancer before the age of 35 y and found mutations in eight and alterations of unknown functional importance in another four. Intriguingly, two of the six women with mutations did not report a family history of breast or ovarian cancer in first- or second-degree relatives, but the number of relatives at risk and their ages were not presented. Fitzgerald et al. (1996) obtained similar estimates of alterations in a sample of 30 women with breast cancer onset before age 30 y. The complete phenotype associated with mutations in this gene is not known, but a recent report of 33 families with evidence for linkage to BRCA1 revealed statistically significant excesses of colon cancer and prostate cancer but no deficits or excesses for any other sites (Ford et al. 1994).
BRCA2
). Like BRCA1, BRCA2 is thought to account for approximately 40% of the families with apparent hereditary predisposition to breast cancer and to confer a high risk of early-onset breast cancer in females. Although the gene was localized in families that also had cases of male breast cancer, affected males are also observed in families with BRCA1 mutations. Analysis of different tumor types from five families showing evidence of linkage to BRCA2 revealed loss of heterozygosity of the wild-type chromosomal region in tumors of the prostate, ovary, cervix, colon and ureter (Gudmundsson et al. 1995). This information suggests a broader phenotype may be more appropriate for BRCA2 than BRCA1.
), or roughly 300,000 women. Thus, factors that influence penetrance of the gene need to be identified. In particular, those factors that prevent elderly mutation carriers from developing cancer despite their inherited high predisposition might prove to be equally important and relevant to women without such strong susceptibility. The role of environmental factors, especially modifiable ones such as diet, is an important area of research that needs to be addressed.
GENETIC FACTORS AND NON-MENDELIAN FAMILIAL BREAST CANCER
, Langston et al. 1996). In particular, according to data from the Surveillance, Epidemiology and End Results program, fully 80% of all breast cancers occur in women over the age of 55 y (National Cancer Institute 1990). Given that late-onset breast cancer consistently demonstrates familial aggregation and that the clustering does not seem to be explained by chance or measured nongenetic risk factors (Chen et al. 1994), it is reasonable to hypothesize that genes other than BRCA1, BRCA2 and p53 might be contributing factors. In fact, there are at least three genes (AT, ESR, HRAS1) for which there is some evidence for an association with breast cancer. Variations in these genes are characterized as having low penetrance and low relative risks and might be involved in the non-Mendelian familial clustering of breast cancer that occurs at more typical ages at onset. In addition, there is evidence that variations or alterations in these genes are more common in the general population than mutations in BRCA1, BRCA2 or p53. Even with the associated lower relative risks, however, the total number of cases in the population attributable to inherited alterations in these genes may be considerable.
ATAXIA TELANGIECTASIA
). Four studies have shown that female relatives of AT patients (obligate gene carriers) are at increased risk of breast cancer. A review of these studies (Easton 1994) produced a combined estimate of relative risk of breast cancer associated with AT heterozygosity of 3.9 (95% CI 2.1 to 7.2). It is important to verify these observations. Because the AT gene has been cloned (Savitsky et al. 1995), it is now possible to base exposure classification on genotype data and extend the study samples to women outside of AT families. It is relevant to note that a relative risk of this magnitude is unlikely to yield multiple cases of early-onset breast cancer in a family in a pattern consistent with simple Mendelian segregation. Wooster et al. (1993) studied 16 multiplex breast cancer families and found no evidence for genetic linkage between breast cancer and chromosome 11q markers (the physical location of the AT gene). Although AT patients are known to have radiosensitivity, they are also more susceptible to cell damage from hydroxy radicals. Thus, it would seem appropriate to investigate the risk of breast cancer among women who carry an altered AT gene and who have low dietary exposure to antioxidants; susceptible individuals may not develop breast cancer unless they receive relevant exposure.
HRAS1
). A meta-analysis of 23 studies has shown that rare HRAS1 alleles are associated with a 1.9-fold (95% CI 1.6 to 2.3) increased risk of cancer (Krontiris et al. 1993). Statistically significant elevated risk estimates were evident for cancers of the breast (1.7), bladder (2.3), colon/rectum (2.2) and lung (1.6) and for leukemia (2.3). Just as for AT, these magnitudes of relative risk are probably insufficient for hereditary manifestation of susceptibility. Indeed, Hall et al. (1990) found no evidence of linkage of HRAS1 to breast cancer susceptibility in 12 high risk families with a Mendelian pattern of disease. Because of the relative high frequency of the rare alleles, however, Krontiris et al. (1993) estimated that as many as one in 11 cancers may be attributed to this factor. Because the biologic mechanism underlying the apparent increased breast cancer risk among women carrying rare HRAS1 alleles is unknown, examination of possible interactions of rare HRAS1 alleles with aspects of diet is perhaps premature at this time.
ESTROGEN RECEPTOR
). At least three published studies implicate the ESR in familial breast cancer. Zuppan et al. (1991) reported genetic linkage of the ESR locus with late-onset breast cancer, but the results were not statistically significant. Andersen et al. (1994) compared genomic and tumor DNA of 274 breast cancer patients with DNA from 204 controls for restriction fragment length polymorphisms and found suggestive evidence that the frequency of the shorter alleles was 1.4 times higher among breast cancer patients (95% CI 1.03 to 1.85). These alleles tended to be associated with later age at onset, but the results were not statistically significant. Roodi et al. (1995) examined the entire coding region of the ESR gene in 118 receptor-positive and 70 receptor-negative breast cancers. Mutations were observed in only two cases (1%), but several neutral polymorphisms (i.e., base pair substitutions that did not alter the encoded amino acid) were identified. Intriguingly, the polymorphism at codon 325 was strongly associated (P < 0.0005) with a positive family history of breast cancer. The fact that certain vitamins act as ligands to the ESR suggests that simultaneous consideration of dietary intake and genetic constitution could be informative in further elucidation of the role of the ESR gene.
ESTROGEN METABOLISM
). It has been repeatedly shown that estrogens can induce and promote mammary tumors in rodents (Bernstein 1993). Epidemiologic studies have consistently demonstrated that early menarche, late menopause, nulliparity, late age at first full-term pregnancy, and obesity in postmenopausal women are associated with significant excess risks of breast cancer (Henderson and Bernstein 1996). These risk factors are thought to exert their effect via longer periods of exposure to estrogens secreted by the ovaries.
-hydroxyestrone (16-OHE1) or 2-hydroxyestrone (2-OHE1) (Martucci and Fishman 1993). The latter exhibits virtually no peripheral estrogenic effect and in fact may act as an anti-estrogen in estrogen-sensitive tissues (Schneider et al. 1984). The former, however, retains uterotropic activity (Yu and Fishman 1985) and, unlike the parent estrogen E2, possesses initiator and promoter activity in normal mammary epithelial cells (Telang et al. 1992). It has been shown that 16-OHE1 can covalently bind to the estrogen receptor and produce prolonged biochemical or physiological response in animals (Bradlow et al. 1992). Laboratory experiments have shown that 16-OHE1 is elevated in explant cultures of mammary terminal duct lobular units of breast cancer patients (Osborne et al. 1993). Epidemiologic studies in humans have indicated that breast cancer patients have substantially lower 2- to 16-OHE1 ratios than controls (Schneider et al. 1982). Taken together, these data suggest that genetic variation in enzyme activity may be related to breast cancer risk.
-hydroxylases, other enzymes (e.g., 17-hydroxysteroid dehydrogenase, aromatase) are also directly involved in estrogen synthesis and clearance. Many proteins or peptides are involved in estrogen regulation (e.g., follicle-stimulating hormone and luteinizing hormone) and transport (e.g., sex hormone-binding globulin), and these compounds are in turn regulated by other factors. It is certain that the complex process of estrogen regulation, synthesis, transport and clearance is determined, to a large extent, by genetic factors. Therefore, it is conceivable that endogenous estrogens and their metabolism patterns may explain interindividual variation in risk of breast cancer. This hypothesis, however, has not been adequately investigated.
FOOTNOTES
LITERATURE CITED
-hydroxylation in human breast tissue: a potential biomarker of breast cancer risk.
J. Natl. Cancer Instit.
1993;
85:1917-1920 [Medline]-hydroxyestrone.
Biochemistry
1985;
29:8017-8021This article has been cited by other articles:
|
|
 |
|
  S. S. Coughlin and M. Piper Genetic Polymorphisms and Risk of Breast Cancer Cancer Epidemiol. Biomarkers Prev., November 1, 1999; 8(11): 1023 - 1032. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
