BACKGROUND Evidence suggests that steroid hormones may affect the natural history of colon cancer. METHODS Baseline levels of estrogen receptors, polyamines, and ornithine decarboxylase in colonic mucosa, and blood estradiol were measured in 10 normal Sprague-Dawley outbread female rats. Therefore, 151 rats were fed a 15% fat diet and divided into three groups. Rats in the control group (n = 20) received weekly s.c injections of the 1,2-Dimethylhydrazine-HCl (DMH) vehicle. To induce colon cancer, 131 rats received weekly subcutaneous injections of DMH (20 mg/kg). Of these 131 rats, 65 also ingested 0.5 microgram/g tamoxifen, daily. Half of the rats in each group were sacrificed at 14 weeks, the remainder at 28 weeks. All measurements were repeated at these times and tumor incidence was calculated. RESULTS The number of rats with tumors was 41% higher (P = .07) in rats treated with DMH vs those treated with tamoxifen and DMH (72.7% vs 51.5%). Tumor cells in both groups had higher levels of polyamines and ornithine decarboxylase activities (P = .03 to P < .001) and lower levels of estrogen receptors (P = .005) to P < .001) compared to adjacent normal colonic mucosa. Estrogen receptors were not detected in the colons of the rats in the control group. No correlations were found between estradiol and estrogen receptors in normal (r = .01, P = .95) or tumor (r = .03, P = .86) cells, or between polyamines or ornithine decarboxylase and estrogen receptors in normal (r = .01 to .14, P = .63 to .95) or tumor (r = .07 to .26, P = .16 to .86) cells. CONCLUSIONS Tamoxifen reduced the incidence of DMH-induced colon cancer in rats and may thus have chemopreventive effects. Although it was not statistically significant, further studies are justified to continue this line of research.