The role of the consumption of fat, animal protein and vitamins on breast-cancer risk was investigated in a hospitalbased case-control study of 924 patients (409 cases and 515 controls) in Montpellier (France). A dietary history questionnaire, administered by interview, comprising 55 key food items as well as beverage consumption, and including food frequencies and portion sizes, was used to measure the intake of total fat and its constituents, animal protein. retinol, p-urotene, vitamin E and alcohol consumption. The questionnaire also elicited information on relevant medical history and personal characteristics. All food items which showed significantly elevated odds ratio (high-fat cheese, desserts and chocolate and processed pork meat) in a multivariate analysis contained a high proportion of animal fat. This is reflected in the nutrient analysis, which showed a significant linear trend as well as an elevated odds ratio for the highest tertile of consumption of total fat [OR, = I .6 (I. l-2.2)], animal fat [OR, = 1.6 (1.1-2.211, saturated fat [OR, = 1.9 (1.3-2.6)] and mono-unsaturated fat [OR, I .7 (I .2-2.511. For post-menopausal women, there is a particularly strong association with saturated fat [OR, = 3.3 (I .C7.8)] in a multivari-
ate analysis including all other significant nutrients. There is no evidence of an increase of risk with the intake of animal protein and no evidence of risk reduction with increased consumption of vegetables, p-carotene or vitamin E. Along with some recent studies, our results give support to the hypothesis that dietary fat is a risk factor in breast carcinogenesis.