Effect of aspirin on long-term risk of colorectal cancer: consistent evidence from randomised and observational studies.

BACKGROUND: Randomised trials have shown that aspirin reduces the short-term risk of recurrent colorectal adenomas in patients with a history of adenomas or cancer, but large trials have shown no effect in primary prevention of colorectal cancer during 10 years' follow-up. However, the delay from the early development of adenoma to presentation with cancer is at least 10 years. We aimed to assess the longer-term effect of aspirin on the incidence of cancers.
METHODS: We studied the effect of aspirin in two large randomised trials with reliable post-trial follow-up for more than 20 years: the British Doctors Aspirin Trial (N=5139, two-thirds allocated 500 mg aspirin for 5 years, a third to open control) and UK-TIA Aspirin Trial (N=2449, two-thirds allocated 300 mg or 1200 mg aspirin for 1-7 years, a third placebo control). We also did a systematic review of all relevant observational studies to establish whether associations were consistent with the results of the randomised trials and, if so, what could be concluded about the likely effects of dose and regularity of aspirin use, other non-steroidal anti-inflammatory drugs (NSAID), and the effect of patient characteristics.
RESULTS: In the randomised trials, allocation to aspirin reduced the incidence of colorectal cancer (pooled HR 0.74, 95% CI 0.56-0.97, p=0.02 overall; 0.63, 0.47-0.85, p=0.002 if allocated aspirin for 5 years or more). However, this effect was only seen after a latency of 10 years (years 0-9: 0.92, 0.56-1.49, p=0.73; years 10-19: 0.60, 0.42-0.87, p=0.007), was dependent on duration of scheduled trial treatment and compliance, and was greatest 10-14 years after randomisation in patients who had had scheduled trial treatment of 5 years or more (0.37, 0.20-0.70, p=0.002; 0.26, 0.12-0.56, p=0.0002, if compliant). No significant effect on incidence of non-colorectal cancers was recorded (1.01, 0.88-1.16, p=0.87). In 19 case-control studies (20 815 cases) and 11 cohort studies (1 136 110 individuals), regular use of aspirin or NSAID was consistently associated with a reduced risk of colorectal cancer, especially after use for 10 years or more, with no difference between aspirin and other NSAIDs, or in relation to age, sex, race, or family history, site or aggressiveness of cancer, or any reduction in apparent effect with use for 20 years or more. However, a consistent association was only seen with use of 300 mg or more of aspirin a day, with diminished and inconsistent results for lower or less frequent doses.
INTERPRETATION: Use of 300 mg or more of aspirin a day for about 5 years is effective in primary prevention of colorectal cancer in randomised controlled trials, with a latency of about 10 years, which is consistent with findings from observational studies. Long-term follow-up is required from other randomised trials to establish the effects of lower or less frequent doses of aspirin.