VEGETABLES, FRUITS, AND CAROTENOIDS Previously, we demonstrated that diets high in vegetables, fruits, and carotenoids were strongly associated with reduced risk of lung and upper aerodigestive tract cancers. To explore the individual contributions of the major carotenoids, we conducted nested case-control studies in the Honolulu Heart Program cohort. Low serum levels of several carotenoids were associated with increased risk of lung, oral-pharyngeal, esophageal, and laryngeal cancer, with low alpha-carotene the most predictive. Recently, in the Nurses and Health Professionals cohorts, using repeated dietary measures, we found that high vegetable and fruit intake was associated with only a modestly lower risk of lung cancer in women and no risk reduction in men. Our analyses suggested that the strength of this widely accepted relationship might have been exaggerated because confounding by smoking was not always adequately controlled. Since recent publications have also questioned whether vegetable intake can reduce the risk of colorectal cancer, we are exploring opportunities to continue to evaluate the importance of vegetables and fruits in cancer etiology, with emphasis on prospective studies, range and variety in intake, and accurate assessment of exposure. Initially, we will explore vegetable and fruit intake and blood carotenoid levels with respect to colorectal polyp progression in the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial cohort. Lycopene, the major source of which is tomato products, has been linked to prostate cancer in several studies. In a case-control study conducted among U.S. Blacks and Whites, we found that high blood lycopene was protective in both races, and that serum levels were lower in the Blacks, suggesting that low lycopene might contribute to the substantially higher rates of prostate cancer among U.S. Blacks. The protective influence of lycopene-rich diets has been most consistently seen in US studies so we are now exploring its role in a cohort of Finnish men with different dietary patterns. ONE-CARBON METABOLISM Disruption of one-carbon metabolism can interfere with DNA synthesis, repair, and methylation and thus promote carcinogenesis. Efficient one-carbon metabolism requires not only folate but also vitamins B6 and B12 and optimal activity of multiple enzymes, such as methylene tetrahydrofolate reductase and methionine synthase. In a community-based case-control study of invasive cervical cancer in five U.S. areas, we found that low serum and red blood cell folate were only modestly predictive of increased risk. However, elevated serum homocysteine was strongly and significantly predictive (RR's~2-3), suggesting that pervasive problems in one-carbon metabolism might be involved. We are now assessing the contribution of B-vitamin deficiencies and polymorphisms in key one-carbon metabolism genes. In addition, we will be evaluating the importance of one-carbon metabolism in the etiology of brain and colorectal cancer. Dietary and vitamin supplement information are being analyzed in a multicenter case-control study of brain cancer, and will be complemented by assays for genetic polymorphisms. In the PLCO cohort, 2000 men and women have been diagnosed with colorectal adenomatous polyps. The dietary information, serum and DNA collected in this study will allow us to explore the role of one-carbon metabolism in the progression of colorectal cancer. Since folic acid fortification has already been shown to decrease homocysteine levels in populations with a variety of nutritional and genetic impairments, our results may suggest targeted supplementation and/or fortification schemes that can reduce cancer risk. BREAST CANCER International variation in breast cancer incidence and migrant studies indicate that modifiable factors play a major role in breast cancer etiology although the specific lifestyles and environmental exposures remain elusive. We designed a large, population-based case-control study of breast cancer in Asian-American women to take advantage of their diversity in diet and breast cancer risk. Childhood, adolescent, and adult exposures were assessed by interviewing both study participants and their mothers. To complement the extensive interview information, body size and shape were measured, and blood and urine samples were collected. Endogenous hormones, phytoestrogens, growth factors, micronutrients, and fatty acids have been or will be measured in these specimens. We found a 6-fold gradient in breast cancer risk by migration history within our study population, comparable to the international differences in breast cancer incidence rates. Adult exposures substantially influenced risk. Furthermore, there was no evidence of an especially susceptible period during menarche or early reproductive life. Increased adiposity and weight gain in the decade preceding diagnosis were critical determinants of risk. Thus, excess weight may function as a late stage promoter, and weight maintenance or reduction as an adult may have a significant and rapid impact on breast cancer risk. We are currently exploring which endogenous hormones, hormone metabolites, growth factors, and dietary patterns are most correlated with the 6-fold gradient in risk, and predictive of increased breast cancer incidence. Circulating estrogens were only weakly associated with increased risk, while circulating androgens were inversely associated in both pre- and postmenopausal women. Decreased intake of soy was associated with a doubling of breast cancer risk. We now want to examine the broader dietary patterns associated with soy intake; in particular, its relationship vegetable, fruit, and grain consumption; the relative contribution of childhood and adult soy intake; and what specific isoflavones and lignans seem important. Since height was a strong and consistent predictor of risk in these Asian-American women, we are seeking biologic explanations, with emphasis initially on insulin-like growth factors.
