Prevention of bladder cancer in the FHIT knock-out mouse model with rofecoxib, a COX-2 inhibitor

INTRODUCTION and OBJECTIVE: Epidemiological studies have indicated that non-steroidal, anti-inflammatory drugs may have a role in the prevention of human cancers. Molecular based targeting of the COX-2 isoform has lead to the development of COX-2 selective inhibitors such as rofecoxib. Previous studies have shown that administration of COX-2 inhibitors, while decreasing overall morbidity do not decrease the incidence of pre-neoplastic lesions in the murine model. Our laboratory has demonstrated that mice lacking the Fhit tumor suppressor gene are more susceptible than controls to the carcinogenic effect of N-Butyl-N-(-4-hydroxybutyl)-nitrosamine (BBN). Our hypothesis is that rofecoxib will inhibit or slow the development of BBN-induced urinary bladder cancers in mice lacking the Fhit tumor suppressor gene. METHODS: 208 mice consisting of 50 Fhit +/+, 63 Fhit +/ -, and 95 Fhit -/-, were divided into five treatment groups: 1) 0.1% BBN and rofecoxib; 2) 0.01 % BBN and rofecoxib (only Fhit -/- mice); 3) rofecoxib alone; 4) 0.1% BBN alone; and 5) 0.01% BBN alone (only Fhit -/- mice). Rofecoxib was administered at the concentration of 150 ppm in mouse chow. BBN was dissolved in water. At the end of the 15th week-treatment the mice were sacrificed and their bladders collected for histological analysis. The mice fed 150 ppm rofecoxib had blood samples taken twice during the 13 weeks to verify the blood serum concentration of rofecoxib. RESULTS: The mean concentration of rofecoxib in the blood serum was 0.37-µg/mL (± 0.15). The mice that received only rofecoxib and no BBN presented normal bladders, regardless of the genotype. The association between Fhit genotype, rofecoxib treatment, presence of preneoplastic lesions and bladder tumors, was evaluated via the two sided Fisher's exact test ( of 0.05). Comparing the means of each group the pair wise Ps were as follows : Group 1 vs. Group 4 p=1 for Fhit +/+; p=0.687 for Fhit -/-; p=0.107 for Fhit +/-. Comparing Group 2 vs. Group 5 p= 0.114. The difference between the groups was not statistically significant. There was a significant increase in neoplastic lesions in the Fhit+/- and Fhit -/- vs. Fhit +/+ mice after BBN treatment. CONCLUSIONS: Our results confirmed that Fhit null mice, both Fhit +/- and Fhit -/-, are exquisitely sensitive to the carcinogenic effect of BBN and that the Fhit-KO mouse is an extremely valid model to study in vivo bladder cancer tumorigenesis. Our preliminary results suggest also that rofecoxib does not provide a therapeutic defense in our chemically induced mouse bladder cancer model