Background: Colon cancer is the second leading type of cancer in developed countries. Studies have shown that aspirin in a dosage of at least 500 mg per day is effective in preventing colon cancer, although higher dosages may be contraindicated in persons at risk of bleeding. Rothwell and colleagues performed a meta-analysis to determine whether lower-dose aspirin regimens are effective in the primary prevention of colon cancer.

The Study: The authors evaluated four trials performed in the 1980s and 1990s, with end points focusing on cardiovascular outcomes. Studies with at least 1,000 participants and a median scheduled treatment period of at least 2.5 years were included. All of the studies were randomized and double-blinded except one that used physician volunteers as participants. The authors reviewed death registries, histology reports, and cancer registries to determine cause of death. Persons whose primary cause of death was colon cancer were counted, and control versus intervention groups were compared. Kaplan-Meier analysis of survival curves, log-rank test to assess significance, and hazard ratios to assess incidence and risk of death were performed. Results were stratified by the dosage of aspirin (75 to 300 mg versus 500 to 1,200 mg), the duration of treatment (at least 2.5 years versus 5 years), and by the location of colorectal cancer if discovered (colon proximal to the splenic flexure, descending colon, sigmoid colon, rectosigmoid junction, and rectum).

Results: A total of 14,033 participants were assigned to control or intervention groups across studies. There was no notable heterogeneity among studies; data were pooled with a median treatment duration of 6 years and median time to follow-up of 18.3 years. There was evidence for increased risk reduction with a longer treatment period, but no evidence of earlier diagnosis in the aspirin versus control groups when colorectal cancer presented during the trial. Pooled results showed that aspirin given at any dosage with a mean treatment duration of 5.8 years reduced long-term risk of colon cancer, with no benefit for rectal cancer. When the results could be divided based on location of the colon cancer, there was a significant benefit for prevention of colon cancer incidence and mortality occurring proximally to the sigmoid flexure (hazard ratio [HR] = 0.45, mortality HR = 0.34), but no effect was found in the distal colon. For persons using aspirin for at least five years, the risk of proximal colon cancer was decreased by 70 percent (HR = 0.35, mortality HR = 0.24) and there was also a decrease in the risk of rectal cancer (HR = 0.58, mortality HR = 0.47). The benefit of aspirin on colon cancer prevention was not increased with dosages greater than 75 mg per day. In patients treated with 75 to 300 mg of aspirin per day for five years, there was a 1.76 percent absolute reduction in the risk of fatal colorectal cancer.

Conclusion: The authors conclude that using aspirin in a dosage of at least 75 mg per day for five years or more may help prevent proximal colon cancer and rectal cancer. Although previous studies showed benefit only with higher dosages of aspirin, the authors suggest that patients may successfully decrease risk of colorectal cancer using lower dosages.

editor's note: Evidence regarding which agents are effective in preventing colon cancer is limited. For example, there is insufficient evidence to support fiber supplementation and antioxidant use for primary prevention of colorectal cancer, and moderate evidence to support hormone therapy.¹ This study shows that aspirin in lower dosages may prevent colon cancer and related mortality. However, an accompanying editorial by Benamouzig and Uzzan notes some weaknesses of the study. The authors assert that there is still controversy about whether lowering the dosage of aspirin from 300 mg or 500 mg per day to 75 mg per day decreases the risk of bleeding.² Also, the question remains if these results are applicable to women; two of the four studies enrolled only men.² Conversely, the maximal duration of the trials was seven years, but this study looks at the 20-year mortality rate, potentially underestimating the beneficial effects of aspirin. Surely some patients previously randomized to control groups chose to take daily aspirin once their trial was completed. Additionally, because an analysis of aspirin-related mortality was not performed, it seems the important question of safety remains.—j.p. and sumi sexton, Associate Editor, American Family Physician