We examined the effect of voluntary physical exercise (running wheels) on pancreatic carcinogenicity of N-nitrosobis-(2-oxopropyl) amine (BOP) in groups of female Syrian hamsters fed a high-fat (HF) diet in which corn oil was 24.6% of the diet or a low-fat (LF) diet in which corn oil was 4.5% of the diet. Each group was divided into an exercising (EX) group (LF-EX and HF-EX) and a sedentary (S) group (LF-S and HF-S). All hamsters were treated with BOP (20 mg/kg body wt) weekly for two weeks beginning four weeks after the experimental diets, which were fed from weaning. A modified glucose tolerance test was performed before the BOP injections and then again at 20 and 40 weeks, and the levels of glucose, insulin-like growth factor I, and insulin were determined in the plasma samples. At the end of the experiment, serum levels of lipid metabolites were also examined in six hamsters from each group. The experiment was terminated 44 weeks after the BOP treatment. Pancreatic ductal/ductular adenocarcinoma incidence was significantly higher in hamsters fed the HF diet (HF-S and HF-EX) than in those fed the LF diet (LF-S and LF-EX). In all groups, glucose and insulin-like growth factor I levels remained within the normal range throughout the experiment, whereas insulin and lipid metabolite levels were significantly elevated in all hamsters fed the HF diet (HF-S and HF-EX). Exercise significantly reduced the insulin level in the group fed the HF diet but did not influence the cancer burden, possibly by the generation of reactive lipid metabolites. Overall, the results showed that voluntary physical exercise does not influence the promotional action of the HF diet on pancreatic carcinogenesis in hamsters. This action could be attributed to the ability of the HF diet to increase the secretion of insulin, which has a growth-promoting and mitogenic effect on pancreatic cells, and to the effect of an HF diet or physical exercise in producing excess free radicals.