Ursodeoxycholic acid counteracts celecoxib in reduction of duodenal polyps in patients with familial adenomatous polyposis: a multicentre, randomized controlled trial

BACKGROUND Due to prophylactic colectomy, mortality in patients with familial adenomatous polyposis (FAP) has changed, with duodenal cancer currently being the main cause of death. Although celecoxib reduces duodenal polyp density in patients with FAP, its long-term use may increase the risk of cardiovascular events and alternatives need to be explored. Preclinical studies suggest that the combination of celecoxib with ursodeoxycholic acid (UDCA) is a potentially effective strategy. We performed a randomized, double-blind, placebo-controlled trial to investigate the effect of celecoxib and UDCA co-treatment on duodenal adenomatosis in patients with FAP. METHODS Patients with FAP received celecoxib (400 mg twice daily) and UDCA (1000-2000 mg daily, ~20-30 mg/kg/day, n=19) or celecoxib and placebo (n=18) orally for 6 months. Primary outcome was drug efficacy, assessed by comparing duodenal polyp density at pre- and post-intervention by blinded review of endoscopic recordings. As secondary outcomes, cell proliferation, apoptosis, and COX-2 levels in normal duodenal mucosa were assessed by immunohistochemistry or real-time quantitative polymerase chain reaction. RESULTS In intention-to-treat analysis, deceased polyp density was observed after celecoxib/placebo treatment (p=0.029), whereas increased polyp density was observed after celecoxib/UDCA treatment (p=0.014). The difference in change in duodenal polyp density was statistically significant between the groups (p=0.011). No changes in secondary outcomes were observed. Thirty patients (81%) reported one or more adverse events, 16 patients (84%, Common Toxicity Criteria for Adverse Events version 3.0 (CTCAE) grade 1-3) treated with celecoxib/UDCA and 14 patients (78%, CTCAE grade 1-2) treated with celecoxib/placebo. Nine patients (24%) discontinued intervention prematurely, 5 patients (26%) treated with celecoxib/UDCA and 4 patients (22%) treated with celecoxib/placebo. CONCLUSIONS Celecoxib reduces duodenal polyp density in patients with FAP, and unexpectedly, high dose UDCA co-treatment counteracts this effect. The benefit of long term use of celecoxib for duodenal cancer prevention needs to be weighed against the (risk of) adverse events. TRIAL REGISTRATION http://ClinicalTrials.gov, identifier NCT00808743.