Aspirin or NSAIDs for the Primary Prevention of Colorectal Cancer

David A. Johnson, MD, FACG, FACP


April 06, 2007

The Use of Aspirin for Primary Prevention of Colorectal Cancer: A Systematic Review Prepared for the U.S. Preventive Services Task Force

Dube C, Rostom A, Lewin G, et al
Ann Intern Med. 2007;146:365-375


Nonsteroidal Anti-inflammatory Drugs and Cyclooxygenase-2 Inhibitors for Primary Prevention of Colorectal Cancer: A Systematic Review Prepared for the U.S. Preventive Services Task Force

Rostom A, Dube C, Lewin G, et al
Ann Intern Med. 2007;146:376-389


Routine Aspirin or Nonsteroidal Anti-inflammatory Drugs for the Primary Prevention of Colorectal Cancer: U.S. Preventive Services Task Force Recommendation Statement

Ann Intern Med. 2007;146:361-364


[Editor's Note: Dr. Johnson's summary and commentary collaboratively address these 3 reports on the use of selective and nonselective cyclooxygenase-2 inhibitors in the prevention of colorectal cancer.]

Colorectal cancer remains a very prevalent and deadly disease despite growing emphasis on early screening. Indeed, colorectal cancer is the second and third leading cause of cancer-related deaths in men and women in the United States, respectively. In 2006, it was estimated that there were approximately 149,000 new cases of colorectal cancer and approximately 52,000 colorectal cancer-related deaths (see[1,2] It has been proposed that a chemoprevention strategy may serve to complement a colorectal cancer screening strategy. Multiple studies have suggested that there is an upregulation of cyclooxygenase (COX)-2 in colorectal cancer. In line with this, the possible benefit of agents that may impede this upregulation has been proposed -- in particular, as supported by observational studies suggesting a lower incidence of colorectal cancer in patients taking either selective COX-2 inhibitors or nonselective COX inhibitors (eg, aspirin). Additionally, more recent chemoprevention trials have focused on the selective use of COX-2 inhibitors in the prevention of the development of colorectal adenomas.

The use of both selective or nonselective COX-2 inhibitors, however, is not without potential risk. These agents have been implicated in both gastrointestinal and -- more recently -- highly publicized, cardiovascular complications.[3] Accordingly, the US Preventive Services Task Force commissioned these 2 systematic reviews to profile the risk-benefit ratios and to provide the basis for a recommendation about the potential use of these agents as a chemopreventive strategy for colorectal cancer. These reviews included all identifiable studies (randomized, cohort, and case-control) in which the effect of aspirin or other nonselective or selective COX-2 inhibitors was evaluated for a possible impact on the incidence of colorectal cancer. All studies included in the reviews were assessed for quality and heterogeneity in the definitions of regular use, dose, and duration, before statistically pooling any data to evaluate the effect of these agents.

In the first systematic review, the study authors concluded that the regular use of aspirin modestly reduced the relative risk for colorectal adenomas. In randomized controlled trials, the relative risk reduction for aspirin users was 0.82 (95% confidence interval [CI], 0.7-0.95). In cohort studies the risk reduction was 0.72 (95% CI, 0.61-0.85), and in the case-control studies, the relative risk reduction was 0.87 (95% CI, 0.77-0.98). For cancer reduction, the 2 randomized controlled trials of low-dose aspirin did not demonstrate a protective effect. In the cohort studies, they found a relative risk reduction of 22% associated with the use of aspirin. For the case-control studies, aspirin use of 1-3 years' duration showed a nonsignificant trend in favor of aspirin use, although there was a significant protective effect (relative risk, 0.68 [95% CI, 0.54-0.87]) associated with longer duration of use, particularly for periods longer than 10 years. The effect of aspirin dose was assessed in one study as well; there were significant relative risk reductions of 40% observed with aspirin doses of 300 and 325 mg/day, but not for lower doses. However, there were gastrointestinal complications associated with aspirin use. Controlled trials demonstrated a 1.6- to 2.5-fold increase in the relative risk of gastrointestinal bleeding for aspirin users vs nonaspirin users. Additionally, upper gastrointestinal symptoms of nausea and dyspepsia increased by a factor of 2 among individuals in the aspirin group.

In the second systematic review, the focus was on nonsteroidal anti-inflammatory drugs (NSAIDs) other than aspirin as well as the more specific COX-2 inhibitors. Overall, there was fairly good evidence to suggest a positive effect of nonaspirin NSAIDs on the incidence of colorectal cancer. There was good and supportive evidence to suggest a significant effect of these agents on colorectal cancer and adenoma incidence. In the single cohort study that evaluated the effect of nonaspirin NSAIDs on colorectal cancer mortality, there was no supporting evidence to demonstrate a beneficial effect. Colorectal cancer incidence was reduced with nonaspirin NSAID use in cohort studies (relative risk, 0.61 [95% CI, 0.48-0.77]) and case-control studies (relative risk, 0.70 [95% CI, 0.63-0.78]). The colorectal adenoma incidence was also significantly reduced with nonaspirin NSAID use in cohort studies (relative risk, 0.64 [95% CI, 0.48-0.74]) and with the more specific COX-2 inhibitors in randomized controlled studies (relative risk, 0.72 [95% CI, 0.68-0.77]). The trade-off for the potential benefit associated with the use of these agents was the potential risk for the identified complications. The ulcer complication rate associated with nonaspirin NSAIDs was reported as 1.5%/year. Although it was recognized that the use of COX-2 inhibitors somewhat reduced this risk in the short-term studies, it was clear that with multiyear use, there were higher ulcer complication rates when compared with placebo. Perhaps more reflective of recent news, the study authors also found that the COX-2 inhibitors increased the relative risk of cardiovascular events (relative risk, 1.86 [95% CI, 1.33-2.59]).

These very comprehensive systematic reviews provided the basis for the recommendations by the US Preventive Task Force. Recognizably, there was considerable heterogeneity in the dose, duration, and frequency of use of the various drugs in these 2 reviews, limiting the pooling of the study data. Despite this observation, these systematic reviews were believed to provide sufficient data to issue the overall official recommendations.

On the basis of these data, the Task Force concluded that "there is poor-quality evidence that aspirin and NSAID use leads to a reduction in colorectal cancer mortality." Furthermore, they found good-quality evidence that aspirin use increased the incidence of gastrointestinal bleeding (in a dose-related manner), and additionally, that aspirin use increased the incidence of hemorrhagic stroke. They found good-quality evidence that there were significant off-setting potential harms associated with the use of NSAIDs, including perforation, ulcer, bleeding, renal failure, and hypertension. The potential harms were particularly evident in elderly patients.

This compendium of reports should provide guidance to physicians considering NSAID therapy for the prevention of colorectal cancer. Although these agents have potential and well-recognized benefits, on the basis of the currently available evidence, their use cannot be justified for the primary prevention of colorectal cancer.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.