Carotene, Vitamins a &E, and Risk of Melanoma

Willett, Walter

Abstract

We propose to conduct a case-control study to test the hypotheses that increased plasma levels of beta-carotene, alpha-tocopherol, and retinol and increased red cell selenium reduce the risk of malignant melanoma. In addition we will attempt to confirm the recent report that exposure to fluorescent lighting increases the risk of melanoma. In vitro and animal studies as well as epidemiologic observations suggest that selenium and vitamin A or its precursor beta-carotene may reduce the risk of some types of cancer. There are scanty data that relate intake of these substances to malignant melanoma, in part because even large prospective studies accumulate cases slowly. However, melanoma is of particular interest since carotene modifies the effect of ultra-violet radiation, a possible risk factor for melanoma, and topical retinoids can cause regression of melanoma metastases. We plan to enrol all new patients attending the Massachusetts General Hospital Pigmented Lesion Clinic. Over a two year period we expect that about 200 of these patients will be found to have malignant melanoma. Patients not found to have melanoma will serve as controls. Blood specimens will be drawn at the first clinic visit to determine plasma levels of carotene, retinol, alpha-tocopherol, and lipids (used to adjust the levels of fat soluble vitamins). Selenium levels will be determined in red cells by neutron activation. To provide an independent assessment of nutritional status, a dietary questionnaire, which has been validated for assessing intake of these vitamins, will be used to measure these and other nutrients. The relationships between biochemical measures of nutritional status and the occurrence of melanoma will be examined in both crude, stratified, and multiple logistic regression analyses. The study design provides an 90% propability of detecting a relative risk of 2.0 for extreme tertiles of exposures.