Abstract B126: Does adult soy intake influence total estrogen levels and estrogen metabolism patterns?

B126 Background Soy foods are a dietary staple in Asian countries and may play a role in this region’s historically low rates of breast cancer incidence. Isoflavones found in soy have structural similarities to estrogens and could exert protective effects by interacting with estrogen receptors, binding proteins, or metabolic enzymes. In a sample of Asian-American women we test whether soy is associated with urinary concentrations of estrogens and estrogen metabolites (EM). Methods This analysis was conducted among controls from a population-based case-control study of breast cancer among women of Chinese, Japanese and Filipino ancestry, aged 20-55 years, and living in San Francisco-Oakland (CA), Los Angeles (CA) and Oahu (HI). Of 966 controls, we included the 571 (59%) women who donated a 12-hour urine and had not been pregnant or lactating or used exogenous hormones in the last 6 months. At urine collection, 168 women were postmenopausal, 233 were pre-menopausal in mid-luteal phase, and 170 were in other categories. Fifteen EM were measured simultaneously with a highly sensitive and reproducible liquid chromatography-tandem mass spectrometry method. Participants had been queried about their usual adult frequency of intake of tofu and other selected soy foods. Robust regression was used to model the associations of soy intake (in tertiles) with EM measures. Models were adjusted for age, ethnicity, body mass index (BMI), and birthplace (Asia/West). Results No statistically significant associations were seen between soy intake and urinary concentrations of estrone, estradiol, estriol, or total EM in postmenopausal women or premenopausal women. In postmenopausal but not premenopausal women, we found significant trends across tertiles of soy intake for EM grouped by metabolic pathways. The 2- and 4-hydroxylated EM and the methylated catechols, each expressed as a percent of total EM, increased with soy intake, while relative levels of EM in the 16-hydroxylation pathway decreased. Models adjusted for age and ethnicity were similar. Factor analysis was used to confirm EM groupings. Among postmenopausal women, factor analysis identified four independent EM patterns, termed ‘catechols’, ‘methylated catechols’, ‘16-hydroxylation pathway EM’, and ‘parent estrogens’ based on the predominant EM. Factors for premenopausal women were similar. Discussion Our findings do not support the hypothesis that soy intake influences endogenous production of estrogens but they do suggest that soy intake can modify estrogen metabolism in postmenopausal women. Increased ratios of 2-hydroxyestrone to 16α-hydroxyestrone have been associated with lower breast cancer risk in some epidemiologic studies. Methylation of catechol estrogens results in deactivation of these metabolites and prevents conversion into genotoxic quinones. Thus, the trends by soy intake represent variations in EM metabolism that could reduce breast cancer risk. Citation Information: Cancer Prev Res 2008;1(7 Suppl):B126.