Current issues of ACP Journal Club are published in Annals of Internal Medicine


Aspirin prevented new colorectal adenomas in patients with previous colorectal cancer


ACP J Club. 2003 Nov-Dec;139:73. doi:10.7326/ACPJC-2003-139-3-073

Related Content in this Issue
• Companion Abstract and Commentary: Chemoprophylaxis with aspirin (81 mg daily) reduced the incidence of colorectal adenomas in persons at risk

Clinical Impact Ratings

GIM/FP/GP: 6 stars

Gastroenterology: 6 stars

Source Citation

Sandler RS, Halabi S, Baron JA, et al. A randomized trial of aspirin to prevent colorectal adenomas in patients with previous colorectal cancer. N Engl J Med. 2003;348:883-90. [PubMed ID: 12621132]



In patients with previous colorectal cancer, is aspirin effective for preventing the occurrence of new colorectal adenomas?


Randomized (allocation concealed*), blinded (investigators, patients, clinicians, data collectors, and outcome assessors),* placebo-controlled trial with median follow-up of 31 months (The Colorectal Adenoma Prevention Study).


39 cancer centers in the United States.


635 patients who were 30 to 80 years of age (mean age 62 y, 52% men) and previously had curative resection of early-stage colon or rectal cancer and colonoscopy to the cecum with removal of all polyps within 4 months before study entry. Exclusion criteria included recent use of aspirin or nonsteriodal anti-inflammatory drugs, poor general health, expected survival < 5 years, pregnancy or nursing, familial polyposis, invasive cancer, and cardiovascular disease. Follow-up was 81%.


After a 3-month run-in period, patients were allocated to enteric-coated aspirin, 325 mg/d (n = 317), or identical placebo (n = 318).

Main outcome measures

Primary outcomes were the detection of adenomas after randomization, time to detection of a first adenoma, and proportion with advanced adenomas (≥ 1 cm in diameter or villous components).

Main results

The aspirin and placebo groups had a similar mean number of colonoscopic examinations (1.60 vs 1.68, P = 0.13). The aspirin group was associated with a lower mean number of adenomas detected during the study than the placebo group (0.30 vs 0.49, P = 0.003). Furthermore, fewer patients in the aspirin group had ≥ 1 adenoma detected during the study than did those in the placebo group (Table). The groups did not differ for proportions of patients with advanced adenomas.


In patients with previous colorectal cancer, aspirin was effective for preventing the occurrence of new colorectal adenomas.

*See Glossary.

Source of funding: National Institutes of Health. Aspirin and placebo tablets provided by Bayer.

For correspondence: Dr. R.S. Sandler, University of North Carolina, Chapel Hill, NC, USA. E-mail

Table. Aspirin vs placebo for colorectal adenomas in previous colorectal cancer†

Outcome Aspirin Placebo Adjusted hazard ratio (95% CI)‡ NNT (CI)
≥ 1 colorectal adenoma detected 17% 27% 0.64 (0.43 to 0.94) 9 (6 to 29)

†Abbreviations defined in Glossary; NNT and CI calculated from data in article.
‡Hazard ratio adjusted for time to a first colonoscopic examination, number of colonoscopic examinations, cancer stage, age, and sex.


A series of animal experiments as well as epidemiologic studies dating back to the early 1980s have supported the protective effects of nonsteroidal anti-inflammatory drugs against the development of colorectal adenomas. Over time, these studies have been consistent and positive, and now, with the studies by Baron and colleagues and Sandler and colleagues, we can add 2 well-designed, randomized, placebo-controlled trials to this body of evidence.

In the study by Baron and colleagues, 2 doses of aspirin, 81 mg and 325 mg daily, were compared with placebo. A modest 19% reduction in the relative risk for any adenoma was found in patients who received aspirin, 81 mg/d, but no significant protection was found in those who received aspirin, 325 mg/d, when compared with placebo. In the study by Sandler and colleagues, only 1 dose of aspirin (325 mg/d) was evaluated. A 36% reduction in the relative risk for ≥ 1 polyp was found. The greater benefit observed in the study by Sandler and colleagues may be attributed to differences in the populations of the 2 studies. Both recruited patients at “high risk” for polyps, but Sandler and colleagues selected patients with a history of colorectal cancer, whereas Baron and colleagues selected those with a history of polyps. The side effects from regular use of aspirin were similar to placebo in the study by Sandler and colleagues. However, in the study by Baron and colleagues the incidence of stroke was of concern: There were none among those who received placebo, 2 among those who received aspirin, 81 mg/d, and 5 among those who received aspirin, 325 mg/d (P = 0.06).

The results from these studies raise several important questions: First, how does aspirin protect against adenoma formation? In a separate study, it has been proposed that aspirin might increase the catabolism of carcinogenic polyamines (1). The most commonly proposed mechanism of action is inhibition of cyclooxygenase enzymes and subsequent inhibition of prostaglandin synthesis. Prostaglandin E2 is the most abundant prostaglandin in colorectal tumors and can block apoptosis (2). It has also been reported that the doses of aspirin from 81 mg to 650 mg daily are equally effective at inhibiting the production of prostaglandin E2 in the rectal mucosa (3). Similarly, no dose effect is evident for aspirin used in cardiovascular prophylaxis (4). Therefore, if a physician were to recommend aspirin to prevent colorectal polyps, the smaller dose would seem justifiable and may minimize rates of such other adverse events as gastrointestinal bleeding.

In the study by Sandler and colleagues, aspirin protected against the development of new adenomas overall but had no effect on larger (> 1 cm) polyps or polyps with villous features. This differs from the Baron study, where the only significant effect was a reduced incidence of advanced adenomas among patients randomized to receive aspirin, 81 mg/d. How can this be? Whether this observation is because of differences in the colonic milieu of persons who have had colon cancer compared with those who have had previous adenomas, other undisclosed factors, or chance is yet undetermined. Other curious findings were noted in the subgroup analyses of the study by Baron and colleagues. Patients who were younger than the median age of 57 years and women seemed to be more likely to benefit. Explanations for these findings are elusive at present and will require further study before our understanding of the association between aspirin and colorectal adenomas is complete.

Aspirin should not replace endoscopic surveillance or be recommended for everyone. However, for persons at high risk for adenomas, low-dose aspirin can be added with assurance that it is reasonably effective and safe.

James E. Shaw, MD, MPH
Virginia Commonwealth University Health System
Richmond, Virginia, USA


1. Martinez ME, O’Brien TG, Fultz KE, et al. Pronounced reduction in adenoma recurrence associated with aspirin use and a polymorphism in the ornithine decarboxylase gene. Proc Natl Acad Sci USA. 2003;100:7859-64. [PubMed ID: 12810952]

2. Sheng H, Shao J, Morrow JD, Beauchamp RD, DuBois RN. Modulation of apoptosis and Bcl-2 expression by prostaglandin E2 in human colon cancer cells. Cancer Res. 1998;58:362-6. [PubMed ID: 9443418]

3. Sample D, Wargovich M, Fischer SM, et al. A dose-finding study of aspirin for chemoprevention utilizing rectal mucosal prostaglandin E(2) levels as a biomarker. Cancer Epidemiol Biomarkers Prev. 2002;11:275-9. [PubMed ID: 11895877]

4. U.S. Preventive Services Task Force. Aspirin for the primary prevention of cardiovascular events: recommendations and rationale. Ann Intern Med. 2002;136:157-60. [PubMed ID: 11790071]