PURPOSE: Angiogenesis plays a key role in the growth and metastasis of malignant tumor. Angiogenesis is reportedly enhanced by prostaglandins (PGs). Cyclooxygenase (COX)-2 is an inducible enzyme that catalyzes the formation of PGs from arachidonic acid. The COX enzyme system is composed of two isoenzymes, COX-1 and COX-2. Recent sources of experimental and epidemiological evidence suggest a significant role for the COX enzymes, particularly COX-2, in the pathogenesis of breast cancer. COX-2 overexpression in a murine mammary gland is sufficient to cause tumor formation. We performed our study to determine the effect of COX-2 inhibitor in a in vivo mouse mammary tumor (MMT) cell line. METHODS: In order to test our study, 24 C57BL/6 type mice (Jackson Laboratory, Bar Harbor, USA) were randomized to receive 35 days of either placebo supplemented diet (n=11) or a 1,500ppm celecoxib (CELEBREX, Pfizer Inc. St. Louis, USA) supplemented diet (n=13) beginning at day 0. At 14 days after the beginning day, 30 microliter of a 1% India ink solution that contained 500,000 of MMT cells or dye alone (control) was intradermally inoculated at each flank (day 14). The animals were sacrificed 21 days later (day 35) and skin specimens were harvested/processed for quantification of the microvessel density (MVD) that was associated with each inoculated site. The aortas that were isolated according to each treatment group at the time of animal sacrifice were used to create identical aortic ring angiogenesis assays (media 199 supplemented with 20% FBS). Explants were evaluated for 14 days in culture to determine both the rate of angiogenesis initiation (% of explants exhibiting the angiogenic phenotype) and the neovessel growth rate (using a subjective angiogenic index score for the wells exhibiting initiation). Analysis of variance (ANOVA) was used to evaluate the differences between groups for each assay. RESULTS: According to the immunohistochemical staining, celecoxib administration resulted in a parallel decrease in the MVD at both the control and MMT inoculated sites (22% and 21%, p = 0.025 and p = 0.010 respectively). On the aortic ring assay, the dietary treatment group was not significantly inhibited compared with the placebo group (75% and 63.3%, respectively, p = NS). However, dietary celecoxib administration significantly inhibited the angiogenic index of the neovessel growth rate (5.0 +/- 2.38 and 8.9 +/- 3.44, respectively, p < 0.001). CONCLUSION: These results suggest that a selective COX-2 inhibitor had an antiangiogenic effect on the in vivo tumor cells. We will perform more investigations of a selective COX-2 inhibitors, and these may will be crucial drugs to use as new chemotherapy agents for treating in cancer.