Breast cancer, which presents the highest global incidence of all female cancers, is caused by the interaction of genetic and environmental factors. Among the latter, diet has attracted considerable attention, as it is a modifiable risk factor and thus offers an opportunity to design preventive strategies. Nevertheless, only alcohol consumption has been unequivocally related to increased breast cancer risk. Despite the failure of observational studies in human populations to clearly define the nature of the relationship between specific nutrient exposures and breast cancer risk, in vivo and in vitro studies strongly suggest its existence. Moreover, studies at the molecular level have identified the putative action mechanism by which the nutritional constituents of specific foodstuffs may exert protective or enhancing effects with respect to breast cancer risk. The inadequate experimental design of some observational studies, or the occurrence of measurement errors and/or recall bias during data collection, or insufficient follow-up and subject characterization, may underlie these controversies. By improving the methods used to study the relationship between diet and breast cancer risk, and by applying new technologies linked to novel approaches such as "nutrigenomics," it might be possible to derive effective recommendations for breast cancer prevention and thus improve anti-cancer treatment.