Epidemiological observations and laboratory research have suggested that nonsteroidal anti-inflammatory drugs (NSAIDs) reduce the risk of colon cancer and that the inhibition of colon carcinogenesis by NSAIDs is mediated through the modulation of prostaglandin production by rate-limiting enzymes known as cyclooxygenases (COXs). Because traditional NSAIDs inhibit both COX-1 and COX-2, these drugs induce side effects, such as gastrointestinal ulceration and renal toxicity, through the inhibition of the constitutive COX-1. Overexpression of COX-2 has been observed in colon tumors; therefore, specific inhibitors of COX-2 could serve as chemopreventive agents. Our previous study has shown that celecoxib, an inhibitor of COX-2, while sparing COX-1, inhibited azoxymethane (AOM)-induced colon tumorigenesis when administered during both initiation and postinitiation stages, ie., celecoxib administered continuously before, during, and after carcinogen treatment. This study examined the dose-response effect of celecoxib when administered during the initiation and postinitiation stages. In addition, the chemopreventive effects of high-dose celecoxib administered during the promotion/progression stage of colon carcinogenesis, ie., continuous celecoxib administration beginning 14 weeks after the carcinogen treatment, was determined in male F344 rats. We also measured the steady-state levels of celecoxib in the plasma of animals given this inhibitor. Groups of 5-week-old male F344 rats were fed either a control diet or experimental diets containing 500, 1000, or 1500 ppm celecoxib. At 7 and 8 weeks of age, rats scheduled for carcinogen treatment were injected s.c. with AOM at a dose rate of 15 mg/kg body weight/week. Groups of animals destined for the promotion/ progression study and initially receiving the control diet were switched to a diet containing 1500 ppm celecoxib beginning 14 weeks after the second AOM treatment. All rats remained on their respective dietary regimens until the termination of the study, ie., 52 weeks, and were then sacrificed. Colon tumors were evaluated histopathologically. Administration of 500, 1000, or 1500 ppm celecoxib during the initiation and postinitiation stages significantly inhibited the incidence (P < 0.01 to P < 0.0001) as well as the multiplicity (P < 0.01 to P < 0.0001) of adenocarcinomas of the colon in a dose-dependent manner. Importantly, administration of 1500 ppm celecoxib during the promotion/progression stage also significantly suppressed the incidence and multiplicity of adenocarcinomas of the colon (P < 0.01). Also, administration of celecoxib to the rats during the initiation and postinitiation periods and throughout the promotion/progression stage strongly suppressed colon tumor volume (P < 0.0002 to P < 0.001). The steady-state plasma concentration of celecoxib increases somewhat with the dose. Thus, in this model system, the chemopreventive efficacy of celecoxib is dose-dependent when this COX-2 inhibitor is administered during the initiation and postinitiation periods. This study provides the first evidence that celecoxib is also very effective when it is given during the promotion/progression stage of colon carcinogenesis, indicating that the chemopreventive efficacy is achieved during the later stages of colon tumor development. This suggests that celecoxib may potentially be an effective chemopreventive agent for the secondary prevention of colon cancer in patients with familial adenomatous polyposis and sporadic polyps.