ArticlesEffect of daily aspirin on long-term risk of death due to cancer: analysis of individual patient data from randomised trials
Introduction
In the developed world, the lifetime risk of cancer is about 40%, and rates are increasing in the developing world.1 In Europe, about 3·2 million new cancers present each year, with about 1·7 million deaths,2 and there are more than 1·5 million new cases each year in the USA.3 By contrast with treatment of cancer, there has been little progress in use of drugs in prevention of the disease. However, several lines of evidence suggest that long-term use of aspirin might reduce the risk of some cancers, particularly gastrointestinal tumours. Aspirin reduces incidence or growth rate, or both, of several cancers in animal models,4, 5, 6 mediated at least in part by inhibition of the cyclo-oxygenase (COX) enzymes and reduced production of prostaglandins and other inflammatory mediators, but these findings might not be applicable to humans. Observational studies in humans also suggest that aspirin reduces risk of certain cancers,5, 6, 7, 8 but results have been conflicting, with more rigorous studies yielding weaker associations.8 Moreover, observational studies have proved to be unreliable in determining risks and benefits of medications in the past,9, 10 and there is trial evidence that one antiplatelet drug might have adverse effects on cancer outcomes.11
Nevertheless, long-term follow-up of randomised trials has shown that aspirin does reduce the risk of colorectal cancer after a delay of several years,12, 13 probably by reducing precancerous adenomas,14 possibly by inhibition of COX-2.15 However, proof of an effect on other cancers is lacking. 10-year follow-up of the Women's Health Study, a randomised trial of 100 mg of aspirin on alternate days versus control, showed no reduction in incidence of cancer.16 However, aspirin also failed to prevent colorectal adenomas in this study, which is consistent with observational studies suggesting that daily aspirin is required for prevention of cancer.5, 6, 7, 8 Observational studies also suggest that use of aspirin for at least 5 years is required before reductions in risk of cancer are observed,5, 6, 7, 8 and the effect of aspirin on risk of colorectal cancer on follow-up of randomised trials was greatest in patients with duration of trial treatment of 5 years or longer.12, 13
We therefore determined the effect of aspirin on risk of fatal cancer by analysis of individual patient data for deaths due to cancer during randomised trials of daily aspirin versus control (done originally for primary or secondary prevention of vascular events) in which the median duration of scheduled trial treatment was at least 4 years. We studied fatal cancers only, in the first instance, because cause of non-vascular deaths was reliably determined in most aspirin trials, and we also aimed to determine the effect of any reduction in cancer deaths on overall all-cause mortality. In three trials done in the UK,17, 18, 19 we also determined any delayed effects of aspirin on the 20-year risk of death due to cancer by long-term post-trial follow-up.
Section snippets
Search strategy and selection criteria
We searched for randomised trials of aspirin versus control that had a mean or median scheduled trial treatment period of at least 4 years and a range extending beyond 5 years. Eligible trials had investigated the effects of randomised allocation to: aspirin (any dose) versus no aspirin in the absence of another agent; or aspirin (any dose) versus no aspirin in the presence of another antiplatelet agent or antithrombotic agent, if the other agent was used in the same way in the aspirin and no
Trials
Of eight eligible randomised trials of aspirin versus control (webappendix p 1) with a mean duration of scheduled treatment before the end of the trial of 4 years or more, two had been done in primary prevention of vascular disease,17, 19 one in secondary prevention after recent vascular events,18 and five in groups with increased vascular risk without previous vascular events (type 1 or 2 diabetes;20 type 2 diabetes;21 stable angina;22 diabetes with asymptomatic peripheral arterial disease;23
Discussion
We showed previously that treatment with aspirin for longer than 5 years reduced the long-term risk of colorectal cancer.12, 13 In analyses of nearly 2000 cancer deaths, we now show that aspirin also reduces deaths due to several other common cancers (panel). First, we showed by meta-analysis that aspirin reduced the risk of death due to cancer by about 20% during the trials. Second, by analysis of individual patient data we showed that this benefit was due mainly to a delayed reduction of
References (44)
- et al.
Estimates of the cancer incidence and mortality in Europe in 2006
Ann Oncol
(2007) - et al.
Aspirin, salicylates, and cancer
Lancet
(2009) - et al.
Aspirin and non-steroidal anti-inflammatory drugs for cancer prevention: an international consensus statement
Lancet Oncol
(2009) - et al.
Those confounded vitamins: what can we learn from the differences between observational versus randomised trial evidence?
Lancet
(2004) - et al.
Effect of aspirin on long-term risk of colorectal cancer: consistent evidence from randomised and observational studies
Lancet
(2007) - et al.
Long-term effect of aspirin on colorectal cancer incidence and mortality: 20-year follow-up of five randomised trials
Lancet
(2010) - et al.
Double-blind trial of aspirin in primary prevention of myocardial infarction in patients with stable chronic angina pectoris
Lancet
(1992) - et al.
Reliability of cancer registration data in Scotland, 1997
Eur J Cancer
(2002) - et al.
Death certificates provide an adequate source of cause of death information when evaluating lung cancer mortality: an example from the Mayo Lung Project
Lung Cancer
(2009) - et al.
Colorectal cancer deaths as determined by expert committee and from death certificate. The Minnesota Study
J Clin Epidemiol
(1999)