Relationship Between Vitamin E and Polyunsaturated Fatty Acids in Breast Cancer Nutritional and Metabolic Aspects zyx MARIETTE GERBER, MD, PHD,* SYLVIA RICHARDSON, PHD,? PIERRETTE CRASTES DE PAULET, PHD,$ HENRI PUJOL, MD,* AND ANDRE CRASTES DE PAULET MD, PHDS zyxw Intake of vitamin E, total lipids, total cholesterol, and fatty acids were analyzed with the blood levels of vitamin E, total cholesterol, triglycerides, and the serum distribution of fatty acids in a hospital-based population of zyxwvutsrqp 120 patients and 109 controls. In regard to nutritional intake, the only significant differences involve saturated and monounsaturated fatty acid consumption, which is more elevated in postmenopausal patients than in postmenopausal controls. Vitamin E and total cholesterol blood levels are significantly higher in patients than in controls, where the difference is that vitamin E is independent from cholesterol level in premenopausal women only. Fatty acid serum distribution is comparable in both samples, with the exception of arachidonic acid, which is significantly lower in premenopausal patients than in pre- menopausal controls. Two multivariate regression analyses of the plasma vitamin E levels of patients and controls were done with menopausal status and nutrients as independent variables for the first analysis, and with menopausal status and all blood analytes for the second one. The regression coefficients for total cholesterol and triglycerides are statistically significant for both samples, whereas a positive association between vitamin E plasma level and sunflower oil consumption and between vitamin E plasma level and serum linoleic acid distribution is significant for patients only. Furthermore, the multiple regression shows that, when adjusted for analyte variables, plasma vitamin E levels are higher in premenopausal than in postmenopausal patients. In addition, plasma lipid peroxidation, evaluated by malondialdehyde measure- ment, is shown to be significantly lower in patients than in controls. Malondialdehyde level is associated with a significant lower odds ratio (OR) after multivariate tertile analysis (OR for the highest tertile: 0.51; 95% CI: 0.29-0.89). Together, these findings are consistent with a picture of lower lipid peroxidation in patients than in controls. zyxwvuts Cancer 64:2347-2353, 1989. T HAS BEEN REPORTED in several animal studies'3293 that I peroxide radicals derived from the peroxidation of polyunsaturated fatty acids (PUFA), might be involved in oncogenesis or tumorigenesis either &rectly4 or through prostaglandin synthe~is.~ Similarly, a protective role has From the Centre Paul Lamarque, Montpellier, France, *Centre Paul Lamarque, Montpellier, France, TINSERM U 170, Villejuif, France, and the $Facult& de MMecine, Institut de Biologie, Montpellier, France, Supported by the Ligue Nationale Contre le Cancer. The authors thank Drs. Cadillac, Frerebeau, Labauge, and Vlahovitch, from Neurology and Neuro-Surgery wards (HBpital Gui-de-Chauliac), and their nursing staff for collaboration in control sample recruitment; Anne Gravelat for her work zyxwvutsr as interviewer and all the medical, secretarial, and nursing staffof Professor Pujol, Centre Paul Lamarque; and Y. Pioch, at Centre Paul Lamarque, and S. Cenee at the U. 170, for their technical work. Address for reprints: Mariette Gerber, MD, Centre Paul Lamarque, 34094 Montpellier Cedex, France. Accepted for publication June 14, 1989. been reported for vitamin E6 because of its antioxidant capacity. Few studies of humans have addressed the question of the specific role of fatty acids (FA) in relation to breast cancer (BC). Population correlation show a cor- relation of BC mortality with saturated fat or animal fat intake, whereas the association is weaker or nonexistent with polyunsaturated fat or vegetal fat. In addition, it has been shown* that in countries with a high lipid intake, but where this intake is mainly olive oil, rich in mono- unsaturated FA, the BC incidence is lower than in coun- tries where the equivalent amount of fat comes from an- other source. A case-control study' suggested a negative association for BC risk with lipid intake in a country where olive oil is the major source of fat. These data agree with the results of a follow-up study" that suggest a protective role for monounsaturated fatty acid. In that study, satu- 2347