Many popular dietary supplements are enriched in polyphenols such as the soy isoflavones, tea catechins, and resveratrol (from grape skins), each of which has been shown to have chemopreventive activity in cellular models of cancer. The proanthocyanidins, which are oligomers of the catechins, are enriched in grape seeds and form the basis of the dietary supplement grape seed extract (GSE). Evidence suggests that the proanthocyanidins may be metabolized to the monomeric catechins. This study was carried out to determine whether GSE added to rodent diets protected against carcinogen-induced mammary tumorigenesis in rats and whether this was affected by the composition of the whole diet. Female rats were begun on 5%, 1.25%, or 0% (control) GSE-supplemented diets at age 35 d. At age 50 d they were administered 7,12-dimethylbenz[a]anthracene (DMBA) in sesame oil at 80 mg/kg body weight. They were weighed and monitored weekly for tumor development until 120 d after DMBA administration. Administration of GSE in AIN-76A diet did not show any protective activity of GSE against DMBA-induced breast cancer. However, administration of GSE in a laboratory dry food diet (Teklad 4% rodent diet) resulted in a 50% reduction in tumor multiplicity. In similar experiments, genistein administered in AIN-76A diet also failed to show chemopreventive activity against the carcinogen N-methyl-N-nitrosourea; however, when administered at the same dose in the Teklad 4% rodent diet, genistein exhibited significant chemopreventive activity (44-61%). These results demonstrate that GSE is chemopreventive in an animal model of breast cancer; moreover, the diet dependency of the chemopreventive activity for both GSE and genistein suggests that whether or not a compound is chemopreventive may depend on the diet in which the agent is administered.