Previous work in breast cancer patients has indicated an inverse relationship between the risk of relapse and the alpha-linolenic acid (18:3n-3) level in adipose breast tissue. To determine whether low alpha-linolenic levels in patients with aggressive breast cancer resulted from lower 18:3n-3 dietary intake and/or increased metabolism of stored 18:3n-3, we analyzed the fatty acid composition of mammary adipose tissue during tumor growth in a rat model of mammary carcinogenesis. Rats were fed a diet containing 10% fat as rapeseed oil (in which 9% of total fatty acids is 18:3n-3). one-half of the rats received an injection of nitrosomethylurea (NMU) to initiate mammary tumors. In control and NMU-treated groups, three to five animals were sacrificed every three weeks during the five-month experimental time. tumor growth was followed by weekly palpation of the animals and by the measure of total tumor mass and number in sacrificed rats. Mammary tumor and adipose tissues were sampled in sacrificed rats. We found that although mammary adipose tissue fatty acid profile changed throughout the experiment, there was no difference in fatty acid profile between control and NMU-treated rats of the same age. In the NMU-treated group, 18:3n-3 level remained identical throughout the experimental period, irrespective of tumor burden. These data show that, in this model, mammary tumor growth does not modify stored fatty acid levels, including 18:3n-3. this suggests that decreased 18:3n-3 level in patients with poor prognosis is not a consequence of tumor burden but more likely depends on decreased dietary intake.