Methyl-group dietary intake and risk of breast cancer among African-American women: a case–control study by methylation status of the estrogen receptor alpha genes

Objectives: Recent molecular studies show that the absence of estrogen receptor (ER) α gene expression in breast cancer is associated with methylation of the CpG island located in the 5′ region and the first exon of the ER α gene. Because CpG island methylation is an early event in carcinogenesis and because a methyl-deficient diet may lead to abnormal DNA methylation including CpG island methylation, we hypothesized that a methyl-deficient diet is more likely to be associated with breast cancer with methylated ER α gene CpG islands. This study aimed to test this hypothesis in African-American women using a case–control design. Methods: Cases were 304 African-American women pathologically diagnosed with breast cancer during 1995–1998 who lived in three Tennessee counties. Controls were 305 African-American women without breast cancer, who were selected through random-digit dialing and frequency matched to cases by 5-year age range and county of residence. Information on dietary intake and other risk factors was collected through telephone interviews. Dietary methyl-components were defined based on folate and methionine intakes and alcohol consumption. Tumor tissue samples were collected for measuring methylation status of the ER α gene. Results: Our results showed that the odds ratio (OR) estimates for lower dietary folate intake were 2.0 (95% confidence interval, CI: 0.8–4.8) for cases with a methylated ER α gene, 0.6 (95% CI: 0.3–1.5) for cases with an un-methylated ER α gene, and 1.6 (95% CI: 0.7–3.8) for cases with unknown methylation status (presumably including cases with both methylated and un-methylated genes). However, low methionine intake appeared more likely to be associated with tumors with unknown methylation status and high level of alcohol consumption seemed more likely to be related to tumors with un-methylated genes. Conclusions: These results did not show a pattern consistent with the study hypothesis that methyl-deficient diets are more likely related to breast cancer with a methylated ER gene.