BOXIT—A Randomised Phase III Placebo-controlled Trial Evaluating the Addition of Celecoxib to Standard Treatment of Transitional Cell Carcinoma of the Bladder (CRUK/07/004)

Abstract Background Non–muscle-invasive bladder cancer (NMIBC) has a significant risk of recurrence despite adjuvant intravesical therapy. Objective To determine whether celecoxib, a cyclo-oxygenase 2 inhibitor, reduces the risk of recurrence in NMIBC patients receiving standard treatment. Design, setting, and participants BOXIT (CRUK/07/004, ISRCTN84681538) is a double-blinded, phase III, randomised controlled trial. Patients aged ≥18 yr with intermediate- or high-risk NMIBC were accrued across 51 UK centres between 1 November 2007 and 23 July 2012. Intervention Patients were randomised (1:1) to celecoxib 200 mg twice daily or placebo for 2 yr. Patients with intermediate-risk NMIBC were recommended to receive six weekly mitomycin C instillations; high-risk NMIBC cases received six weekly bacillus Calmette-Guerin and maintenance therapy. Outcome measurements and statistical analysis The primary endpoint was time to disease recurrence. Analysis was by intention to treat. Results and limitations A total of 472 patients were randomised (236:236). With median follow-up of 44 mo (interquartile range: 36–57), 3-yr recurrence-free rate (95% confidence interval) was as follows: celecoxib 68% (61–74%) versus placebo 64% (57–70%; hazard ratio [HR] 0.82 [0.60–1.12], p  = 0.2). There was no difference in high-risk (HR 0.77 [0.52–1.15], p  = 0.2) or intermediate-risk (HR 0.90 [0.55–1.48], p  = 0.7) NMIBC. Subgroup analysis suggested that time to recurrence was longer in pT1 NMIBC patients treated with celecoxib compared with those receiving placebo (HR 0.53 [0.30–0.94], interaction test p  = 0.04). The 3-yr progression rates in high-risk patients were low: 10% (6.5–17%) and 9.7% (6.0–15%) in celecoxib and placebo arms, respectively. Incidence of serious cardiovascular events was higher in celecoxib (5.2%) than in placebo (1.7%) group (difference +3.4% [–0.3% to 7.2%], p  = 0.07). Conclusions BOXIT did not show that celecoxib reduces the risk of recurrence in intermediate- or high-risk NMIBC, although celecoxib was associated with delayed time to recurrence in pT1 NMIBC patients. The increased risk of cardiovascular events does not support the use of celecoxib. Patient summary Celecoxib was not shown to reduce the risk of recurrence in intermediate- or high-risk non–muscle-invasive bladder cancer (NMIBC), although celecoxib was associated with delayed time to recurrence in pT1 NMIBC patients. The increased risk of cardiovascular events does not support the use of celecoxib.