In studies of breast and endometrial cancer, the interrelationships of fat and calorie intake, body size and shape, physical activity, and endogenous hormones are being investigated. In two case-control studies of breast cancer, in young U.S. women and in Asian-American women, height was a strong predictor of risk. Adiposity was associated with decreased risk in the young U.S. women and the youngest Asian-American women. However, risk steadily increased with both recent adiposity and recent weight gain in the older Asian-American women. Contrary to expectation, blood lipids were predictive of reduced risk of endometrial cancer, especially in older women. Two studies of the natural history of cervical neoplasia, in Oregon and Costa Rica, are utilizing new, accurate measures of human papillomavirus infection in evaluating whether vitamins C and E, carotenoids, and folate are protective. The reason why risk of many cancers is consistently associated with increased vegetable and fruit intake is a major focus. A case-control study of diet and lung cancer and a cohort study of upper aerodigestive tract cancers with stored serum samples suggest that alpha-carotene, beta-carotene, and possibly other carotenoids and phytochemicals may play important roles. In two case-control studies of esophageal cancer, carried out in the U.S. among blacks and whites and in China, increased vegetable and fruit intake was protective; this improved diet may explain recent decreases in the incidence of this cancer in China. Stored sera from a prospective study provided no support for the hypotheses that low vitamin D levels are associated with increased risk of colon and prostate cancers. Heterocyclic aromatic amines (HCAs), mutagens formed by cooking meat at high temperatures, have been linked to colon and other cancers. To facilitate assessment in epidemiologic studies, a database for five HCAs in foods was developed; and potential biomarkers and metabolic activation were investigated in a metabolic study. Food frequency instruments have been improved by using cognitive psychology approaches and updated national dietary data in their development. Several methodologic studies are assessing measurement errors and biases in dietary, anthropometric, and nutrient data.
