Controversies surrounding diet and breast cancer.

What we know about prevention of breast cancer is related to lifetime oestrogen exposure and exposures to specific oestrogens at vulnerable periods of life. This can be influenced by diet. The strongest indicator of a diet-related effect to date is the fairly consistent increase in breast cancer among women who are tall or obese (Hunter & Willett, 1993). The other dietary factors summarized in Table 1 are less strongly associated with breast-cancer risk in epidemiological studies. The relationship between fat and breast-cancer risk has been extensively studied but remains somewhat uncertain. Fat, as a contributor to energy intakes and energy imbalance, is probably a factor in the higher breast-cancer rates in Western countries. Beyond its role as an energy source, the evidence for an independent effect of dietary fat on breast-cancer risk is weak. More focused analyses of the role of individual fatty acids, and on lipid-related pesticide exposures, may reveal strong effects which are currently masked by the use of inadequate exposure measures, as well as by measurement error. Currently, there is substantial evidence of a weak relationship with alcohol consumption, even at frequencies of drinking of less than once daily. The evidence of a protective role for antioxidants is weaker for breast cancer than for other cancers. This might by expected in a cancer which is not strongly associated with cigarette smoking. Specific foods are being studied for other potentially-active ingredients which may be involved in hormone metabolism, but conclusive results for soyabean or cruciferous vegetables are not yet available. Studying these relationships will continue to be a challenge for researchers because of the difficulties in measuring dietary exposures, which is complicated by the uncertainty of the relevant time frame for exposure assessment. While substantial attention has been focused on studying diet in relation to incidence, the potential for diet to reduce recurrence of breast cancer is thoroughly under-studied. There is little reason to believe that the factors which influence the incidence of breast cancer, perhaps during childhood and puberty, are the same as those which affect recurrence in adulthood. In this area, the very limited evidence available suggests that study of biologically-active fatty acids is promising.