The chemopreventive efficacies of raloxifene and nimesulide, an anti-estrogen but with anti-androgen action and a cyclooxygenase-2 (COX-2) selective inhibitor, respectively, were evaluated in probasin/SV40 T antigen (Tag) transgenic (TG) rats. The treatment groups were placebo, nimesulide (400 p.p.m. in basal diet p.o.), raloxifene (slow-release pellets implanted s.c., 5 mg/kg/day), raloxifene (5 mg/kg/day) plus nimesulide (400 p.p.m.), and raloxifene (10 mg/kg/day) plus nimesulide (400 p.p.m.). Animals were killed at 17 weeks of age, and prostate tissues were harvested and weighed by lobes. Tissues were evaluated by histology, immunohistochemistry, and western blot analyses and blood was collected to measure the testosterone levels. All the animals in the placebo group had tumors in each lobe compared with only 43% each in the dorsolateral (DLP) and anterior prostate (AP) of the animals treated with raloxifene (10 mg/kg/day) plus nimesulide. The total prostate weights and adenocarcinoma portions were significantly reduced in the three raloxifene-treated groups, whereas atrophic glands were increased. There were no significant differences between the nimesulide alone and placebo groups or between the raloxifene (5 mg/kg/day) alone and raloxifene (5 mg/kg/day) plus nimesulide group, suggesting a lack of cancer preventive effects of the COX-2 inhibitor in this animal model. PCNA positive rates in ventral prostate (VP) and DLP, and androgen receptor (AR) levels in VP were significantly reduced in the three raloxifene-treated groups. Furthermore, circulating testosterone was decreased after raloxifene (10 mg/kg/day) plus nimesulide treatment. These results demonstrate that raloxifene, but not nimesulide, inhibits prostate carcinogenesis in SV40 Tag TG rats associated with a decline in circulating testosterone levels and a loss of AR expression, as well as an inhibition of cell proliferation.