Low-Penetrance Genotypes and Gene-Environment Interactions
Family history of breast cancer in a first-degree relative is a consistent risk factor; risk increases with earlier age at diagnosis in first-degree relatives and with the number of relatives affected. In the Nurses’ Health Study, for instance, the relative risk associated with a maternal diagnosis before age 40 years was 2.1, and the relative risk associated with maternal diagnosis after age 70 years was 1.5. The relative risk was 2.5 for women with both an affected mother and at least one affected sister. Most women with breast cancer do not have a family history of the disease in a first-degree relative; however, only 5% to 10% of breast cancers are estimated to be attributable to the inheritance of rare, highly penetrant, germ-line mutations in genes, although this proportion is higher at younger ages of diagnosis. Mutations in BRCA1 and BRCA2 are responsible for most of these inherited breast cancers; mutations in the gene for the tumor protein p53 (causing Li-Fraumeni syndrome) and in PTEN (causing Cowden disease) account for a small proportion of inherited breast cancers. Mutations in each of these genes occur in fewer than 1% of the population.
Polymorphisms are usually defined as a sequence variant in a gene that occurs in more than 1% of alleles. Polymorphisms in genes that code for enzymes, receptors, or other proteins that act in metabolic pathways of potential relevance to breast cancer may influence the function of these proteins and thus create between-person differences in metabolic activity that may alter risk of breast cancer. Candidates include genes for carcinogen-metabolizing enzymes, steroid hormone-metabolizing enzymes, and receptors such as the estrogen and progesterone receptors. If these polymorphisms cause only modest increases in risk, or confer risk only in conjunction with exposure to carcinogens, they would not cause noticeable familial aggregation. Because these polymorphisms may be very common (the homozygous deletion in the glutathione-S-transferase mu gene occurs in approximately 50% of whites), their population-attributable risks may be large even if the relative risks are modest.
Most exogenous carcinogens are metabolized by enzymes, which may change them to more DNA-damaging forms ordetoxify them, often by attaching glutathione or other groups that promote excretion of the modified compounds. Many of the genes that encode these enzymes display phenotypic variation (individuals differ in the activity of the enzymes), and for many of these enzymes, the alterations in DNA sequence that are responsible for the phenotypic variation are known. A growing number of studies have examined whether these different genotypes are associated with breast cancer risk or interact with environmental exposures (e.g., cigarette smoking).
Cigarette smoking is a route of exposure to aryl aromatic amine carcinogens that can be activated by the cytochrome P450 1A2 (CYP1A2) and NAT1 and NAT2 (N-acetyltransferase 1 and 2) genes. The NAT2 gene has four major alleles in whites. Individuals homozygous for any combination of the three slow acetylator alleles are phenotypically slow at the acetylation step; those who have at least one copy of the rapid acetylator allele are rapid acetylators (approximately 45% of whites). In more than 10 studies, the prevalence of slow acetylators was similar among cases of breast cancer and among controls; thus, no independent main effect of this genotype is seen. In a case control study of 185 postmenopausal cases and 213 controls, among women who were slow acetylators, those in the highest quartile of number of cigarettes smoked were at increased breast cancer risk (odds ratio, 4.4; 95% confidence interval, 1.3 to 14.8), suggesting that smoking was a risk factor for breast cancer among slow acetylators. Subsequent studies that included 466 cases and 498 cases did not confirm this finding; indeed, in the latter study, an increased risk was observed among rapid acetylators who were recent smokers. The latter study also examined a polymorphism in NAT1 that may code for a rapid acetylation form of this enzyme; no main effect was observed, although among postmenopausal women an increase in risk was observed among recent smokers with the putative rapid allele. NAT2 is also important in the metabolic activation of heterocyclic amines formed during the high-temperature cooking of animal protein, and heterocyclic amines have been found to be mammary carcinogens in some rodent models. In the two published studies, however, no interaction was seen between meat intake and NAT2 genotype.
Polycyclic aromatic hydrocarbons are animal mammary carcinogens also found in cigarette smoke. The cytochrome P450 1A1 (CYP1A1) gene product is involved in metabolism of polycyclic aromatic hydrocarbons and is polymorphic, although the exact functional significance of these polymorphisms is unclear. No association was observed with a variant in exon 7 among 96 cases of breast cancer. In data based on 216 cases, a significant increase in risk among postmenopausal women with the exon 7 variant was observed, and the increase in risk was greater among lighter than among heavier smokers. In a study of 466 cases, no main effect of the exon 7 polymorphism was observed, and little evidence was found of an interaction with pack-years of smoking. This study, however, showed a suggestion of an interaction between an MspI polymorphism and smoking: Both long duration and early onset of smoking were associated with increased risk among carriers of this polymorphism. No main effect of either of these polymorphisms or two additional polymorphisms was observed among 164 white and 59 African-American cases. Thus, no polymorphisms in CYP1A1 have been independently associated with breast cancer risk; larger studies are needed to assess potential interactions with cigarette smoking.