[Prophylactic effect of a selective COX-2 inhibitor celecoxib on carcinogen-induced gastric premalignant lesions in rats].

BACKGROUND & OBJECTIVE: Although we have previously showed that a selective cyclooxygenase-2(COX-2) inhibitor celecoxib prevents gastric cancer development in a rat model of gastric carcinogenesis, the role of celecoxib on gastric premalignant lesions remains unknown. This study was to explore whether celecoxib was effective for the prevention of premalignancy, and further to clarify its mechanism.

METHODS

Ninety-four male Wistar rats were divided into 5 groups. Group A (n=12) was fed with water only; group B (n=16) with daily 10 mg/kg celecoxib; group C (n=22) with 100 microg/ml N-methyl-No-nitro-N-nitrosoguanidine (MNNG); group D (n=22) with 100 microg/ml MNNG and daily 10 mg/kg celecoxib; group E (n=22) with 100 microg/ml MNNG and daily 3 mg/kg indomethacin. The rats in groups B to E were given 10% sodium chloride in the initial 6 weeks, and the rats in groups C to E were given 100 microg/ml MNNG in drinking water to induce premalignant lesions in the stomach. Six rats in group A, 8 in group B, 10 in group C, 10 in group D, and 10 in group E were killed at week 16, and others were killed at week 24. The occurrence rates of gastric premalignant lesions in the groups were compared. The mRNA and protein levels of COX-1 and COX-2 in gastric mucosa were determined by real-time polymerase chain reaction (PCR) and immunohistochemistry; prostaglandin E2 (PGE2) level was measured by an ELISA-based assay.

RESULTS

Ninety-three rats were studied. In week 16 and week 24, the occurrence rates of glandular atrophy in groups C, D, and E had no significant difference (P>0.05). In week 16, gastric mucosal dysplasia was not detected in groups C, D, and E; at week 24, the occurrence rates of dysplasia were 75% (9/12) on group C, 25% (3/12) in group D, and 46% (5/11) in group E. The occurrence rate of gastric mucosal dysplasia was significantly lower in group D than in group C (25% vs. 75%, P=0.039); there was no significant difference between group E and group C (46% vs. 75%, P=0.214). At week 16 and week 24, there was no significant difference in COX-1 expression between treatment groups and control group. The mRNA and protein levels of COX-2 in group C (3.29+/-1.50 and 3.41+/-0.94) were significantly higher than those in other groups (P<0.001). There was no significant difference in PGE2 level between groups C, D, and E (P>0.05).

CONCLUSIONS

Celecoxib effectively inhibits the development of gastric mucosal dysplasia in rats induced by MNNG, but has no effect on the PGE2 level in the gastric mucosa, indicating that the anti-neoplastic activities of celecoxib may be independent of COX-2.