Genetic susceptibility to breast and endometrial cancer

Abstract

Hormones are central in the carcinogenic process in the breast and in the uterine epithelium. Individual genetically determined variation in the response to hormonal influence may alter susceptibility to breast and endometrial cancers. Many small studies of this hypothesis have generated inconclusive results. Since the effect of any genetic variant is expected to be modest, large studies are needed to draw reliable conclusions. Also, there may be interaction between genetic and lifestyle factors that needs to be considered. To adress these issues, we conducted a large population-based case-control study that incorporated genetic and lifestyle exposures. We investigated common variants in the genes for the estrogen metabolizing enzymes catechol-Omethyl transferase (COMT) and cytochrome P450 1B1 (CYP1B1) and in the estrogen (ESR1), androgen (AR) and vitamin D receptors (VDR), selected from previous literature, in relation to breast and endometrial cancer risk. We genotyped 1569 breast cancer patients, 707 endometrial cancer patients and 1729 partly shared population controls. All participants had previously provided extensive information via a questionnaire about lifestyle factors such as reproductive history, body size, and use of menopausal hormone preparations.

We found that a COMT allele that confers high enzyme activity (c.324/474A>G G) was associated with increased risk for lobular breast cancer, but not with breast cancer risk overall. The CYP1B1 variants, c.355G>T, c.4326C>G, and c.4390A>G, were not associated with breast cancer risk. The common ESR1 c.454-351A>G allele was associated with relatively higher cancer risk compared with the rare allele, although in breast cancer only when considered in a haplotype with c.975C>G. The endometrial cancer risk in women homozygous for the rare c.454-351A>G allele was just half of the risk in those homozygous for the common allele. The AR CAGn and VDR An were not associated with breast cancer risk. In exploratory subgroup analyses stratified according to lifestyle factors, we found that in women with diabetes mellitus the high activity COMT allele was associated with an increased risk for breast cancer. In women who had used menopausal hormones for at least four years the CYP1B1 c.4326C>G G allele conferred increased breast cancer risk. The association between the ESR1 c.454-35 1A>G and c.975C>G AC haplotype and breast cancer was stronger among women with a BMI above 30. Women who carried two short alleles of VDR An had a halved risk for breast cancer, irrespective of parity.

The most persuasive association in this work was the similar relation between variation in ESR1 and breast and endometrial cancer. A stronger association with endometrial cancer was expected due to its apparently lower degree of etiological complexity relative to breast cancer. The estrogen receptor is crucial in estrogen stimulation and thus the prior probability of this gene being linked to breast and endometrial cancer risk was high. The remaining results indeed add to the body of evidence regarding breast and endometrial cancer susceptibility but the statistically significant associations may well be due to chance.