NUTRITIONAL DETERMINANTS OF BREAST CANCER RISK

点击翻译按钮获取中文摘要

Although a series of international population studies and animal experiments suggest that dietary fat may increase the risk of breast cancer, epidemiologic studies conducted to date have yielded inconsistent results. However, few epidemiologic studies have measured specific fatty acid intakes or have controlled for other nutritional factors, particularly antioxidants, that may modify the effect of fat intake. A case-control study is proposed to address a series of hypotheses relating biochemical and other indicators of dietary intake to the risk of breast cancer among post-menopausal women. Specifically, it is hypothesized that the following factors increase the risk of breast cancer: 1) diets high in specific fatty acids such as linoleic and trans fatty acids 2) diets low in selenium 3) diets high in total fat 4) the combination of high intake of polyunsaturated fatty acids and low intake of antioxidants (selenium, vitamins A and E and specific carotenoid fractions) or low level of plasma glutathione peroxidase activity. Cases and controls will be post-menopausal women who are patients of the Breast Evaluation Center of the Dana Farber Cancer Institute and its surgical staff. Over three years 1080 subjects - 360 women with newly diagnosed local or regional breast cancer and two control subjects for each case (one with no breast disease requiring biopsy and another with benign breast disease) - will be enrolled. Specific fatty acids will be measured by capillary column gas-liquid chromatography of fat from subcutaneous fat aspiration specimens; selenium by neutron activation analysis of toenails; vitamins A, E and specific carotenoid fractions by high-pressure liquid chromatography of plasma; and total dietary fat, fiber, protein, vitamin C and other nutrients by a validated semi-quantitative food frequency questionnaire. The study will also provide an opportunity to evaluate the micronuclei test, a possible measure of DNA damage. The effects and interactions of specific dietary factors will be studied using crude, stratified, and multivariate analyses. The study design conservatively provides a 90 percent probability of detecting relative risks of 2.0 for extreme quartiles of exposure.