Table 2 Characteristics of RCTs and summary of effects of non-aspirin NSAIDs on the incidence of recurrent colorectal adenomas

Arber 2006

(The Prevention of Colorectal Sporadic Adenomatous Polyps (Pre SAP) study) [37]

Multi-national

≈3 years (1 and 3 years after the baseline examination)

Age – median, 61 years; range, 30–92 years; % male: 66; subjects with history of adenomas; and documented clean colon post-polypectomy.

Celecoxib 400 mg/day (n = 933); placebo (n = 628)

Primary outcome: adenoma recurrence at year 1, 3, or both

Secondary outcomes: advanced adenomas recurrence at year 1, 3, or both; cardiovascular outcomes and adverse events (advanced adenoma defined as adenoma ≥1.0 cm (villous or tubule-villous histology); high-grade dysplasia; Intra-mucosal carcinoma or invasive cancer)

11% excluded from analysis as no follow-up colonoscopy at year 1 or year 3

78% of participants reported taking the majority of their study medications, with similar compliance between arms

Relative Risk (95% CI)

Celecoxib 400 mg versus placebo

Any adenoma: 0.64 (0.56–0.75)

Advanced adenoma: 0.49 (0.33–0.73)

Bertagnolli 2006 (The Adenoma Prevention with Celecoxib (APC) trial) [38]

Multi-national

3 years (1 and 3 years after the baseline examination)

Age – median, 59 years; range, 31–88 years; % male: 68; subjects with history of adenomas; and documented clean colon post-polypectomy.

Celecoxib 400 mg/day (n = 685);

Celecoxib 800 mg/day (n = 671);

Placebo (n = 679)

Primary outcome: adenoma recurrence at year 1, 3, or both

Secondary outcomes: advanced adenomas recurrence at year 1, 3, or both; number of adenomas; size of largest adenoma; adenoma burden; cardiovascular outcomes and adverse events (advanced adenoma defined as adenoma ≥1.0 cm (villous or tubule-villous histology); high-grade dysplasia; Intra-mucosal carcinoma or invasive cancer)

10% excluded from analysis as no follow-up colonoscopy at year 1 or year 3

≈ 68% of participants reported taking the majority of their study medications at least 80% of the time, with similar compliance between arms.

Relative Risk (95% CI)

Celecoxib 400 versus placebo

Any adenoma: 0.67 (0.59–0.77)

Advanced adenoma: 0.43 (0.31–0.61)

Celecoxib 800 versus placebo

Any adenoma: 0.55 (0.48–0.64)

Advanced adenoma: 0.34 (0.24–0.50)

Summary for celecoxib, 400 mg/day versus placebo

Relative Risk (95% CI)

Any adenoma: 0.66 (0.59–0.72)

Advanced adenoma: 0.45 (0.33–0.57)

Baron 2006

(The Adenomatous Polyp PRevention On Vioxx (APPROVe) Trial [28]

Multi-national

3 years (1 and 3 years after the baseline examination)

Age – mean, 59.4 years; range, 40–86 years; % male: 62.3%; subjects with history of adenomas; and documented clean colon post-polypectomy.

Rofecoxib 25 mg/day (n = 1277); placebo(n = 1293)

Primary outcome: ≥1 adenoma at year 1 or 3 on colonoscopy

Secondary outcomes: number of adenomas; advanced adenomas; death; cardiovascular and gastrointestinal events (advanced adenoma defined as adenoma ≥1.0 cm (villous or tubule-villous histology); high-grade dysplasia; Intra-mucosal carcinoma or invasive cancer)

7% excluded from analysis as no follow-up colonoscopy

More than 87% of subjects reported taking at least 90% of their tablets in the third year of the trial

Relative Risk (95% CI)

Rofecoxib versus placebo

Any adenoma: 0.76 (0.69–0.83)

Advanced adenoma: 0.56 (0.42–0.75)