MeReC Rapid Review:
9 May 2012
A meta-analysis of 51 RCTs found that patients taking daily aspirin for the primary or secondary prevention of CV events had a lower risk of death from cancer than those who did not take aspirin, especially after 5 years, although the absolute reduction in risk was small. Patient-level data from six of the CV primary prevention trials also suggest those taking low-dose daily aspirin beyond three years have a lower risk of developing cancer. In the first three years, the reduced absolute risk of major vascular events is offset by a similar increase in the absolute risk of major bleeding. There are limitations to these findings and a cautious approach in needed when applying them to practice.
Level of evidence:
Level 1 (good quality patient-oriented evidence) according to the SORT criteria.
It is still premature to consider routine administration of daily aspirin to reduce the risk of developing cancer or of dying from it, especially when balancing the benefits against risks of taking aspirin. It is not yet clear what groups of patients might benefit most and be at the lowest risk from the harms of aspirin. Health professionals should be ready to advise those considering taking aspirin to prevent cancer on their risk of vascular events and of extracranial bleeds over time. In particular, they should note that aspirin was not found to have any effect on risk of death from cancer until at least 5 years of follow-up.
What is the background to this?
Several earlier studies have suggested that aspirin might reduce the risk of cancer, especially that of the gastrointestinal tract. A previous meta-analysis of eight randomised controlled trials (RCTs) found that patients treated with daily aspirin for primary or secondary prevention of cardiovascular (CV) disease were less likely to die from many common cancers during those trials. This meta-analysis was discussed in MeReC Rapid Review 2221. However, important questions remained, in particular a key clinical issue: the balance of risks and benefits of daily low-dose aspirin in primary prevention, and the evolution of these risks over time.
This new meta-analysis considered data from 51 RCTs in 77,549 patients who were given daily aspirin (any dose) for either primary or secondary prevention of CV disease. Mean duration of follow up in these trials ranged from 0.5 years to 8.2 years. The authors first assessed the effect of aspirin on cancer death. Then, looking only at six primary prevention trials of daily low-dose aspirin, they studied the effect of low-dose aspirin on cancer incidence in the context of its overall risks and benefits with duration of use.
What does this study claim?
Allocation to aspirin was associated with fewer deaths due to cancer versus control (odds ratio [OR] 0.85, 95% confidence interval [CI] 0.76 to 0.96, p=0.008, 34 RCTs, n=69,224). However, when stratified by time from randomisation to cancer death, there was no statistically significant effect from aspirin during the first 5 years of follow-up (p>0.18). After 5 or more years’ follow-up the OR for cancer death was 0.63 (95% CI 0.49 to 0.82, p=0.0005) in the patients randomised to aspirin. The maximum duration of follow-up was not stated, nor is it clear if follow-up continued beyond the duration of the trial (i.e. when patients may have discontinued aspirin).
Individual patient data from six primary prevention trials of low-dose aspirin suggest that taking 75 mg to 100 mg aspirin/day was associated with a reduced rate of new cancer diagnoses compared to controls after at least 3 years’ follow-up.
Looking at overall risks and benefits, during 0–2.9 years’ follow-up, low-dose aspirin was associated with 204 fewer major vascular events but 133 additional major extracranial bleeds per 100,000 patients per year compared with control (OR 0.82, 95% CI 0.72 to 0.92 and OR 1.95, 95% CI 1.47 to 2.59, respectively). However, there was no statistically significant difference in the rate of new cancer diagnoses From 3 years’ follow-up onwards, low-dose aspirin use did not significantly affect the risk of major vascular events or major extracranial bleeds. However, during 3–4.9 years’ follow-up it was associated with 219 fewer new cancers diagnosed per 100,000 patients per year compared with control (OR 0.81, 95% CI 0.67 to 0.98). After 5 or more years’ follow-up, low-dose aspirin was associated with 480 fewer new cancers diagnosed per 100,000 patients per year compared with control (OR 0.70, 95% CI 0.56 to 0.88). See Table in Study details below. The greatest benefit on cancer diagnosis was in people scheduled to take aspirin for 5 years or more.
This meta-analysis adds to the growing body of evidence available showing that aspirin has a beneficial effect on cancer. Although it reported that taking daily aspirin is associated with a lower incidence of cancer, as well as a reduced risk of cancer death, it has several limitations and a cautious approach is needed when applying the results to practice.
Firstly, the studies included were in people at risk of CV disease and so the results cannot be extrapolated to the rest of the population. Also, they were only designed to look at vascular events and not specifically designed to evaluate cancer risk. Importantly, and as discussed in the accompanying editorial, two large RCTs which provide important evidence were excluded. These studies (the Women’s Health Study n=39,876 and Physicians’ Health Study n=22,071) are the largest RCTs of aspirin designed to assess aspirin’s effects in the primary prevention of cancer: neither found an association between aspirin use and reduced overall cancer or mortality, or a reduced risk of colorectal cancer after 10 to 12 years. However, these RCTs investigated the effects of low-dose aspirin taken on alternate days. The authors of the meta-analysis discussed here suggest that this difference in dosing is important, but this is not conclusively established.
Another important limitation of the meta-analysis is that information was not obtained about cancer screening or surveillance in the trials included, so it is possible that earlier diagnosis of cancer or precancerous precursors due to aspirin-associated bleeding may complicate the profile of cancer incidence reported over time. In addition, the data on cancer incidence, and the overall risks and benefits of taking aspirin over time, come only from the six primary prevention trials of low-dose aspirin and not all 51 RCTs.
The authors of this meta-analysis have attempted to build on findings from the previous meta-analysis (see also MeReC Rapid Review No. 2221) and provide answers to some outstanding questions. In particular, this meta-analysis provides some useful information on the short-term risks and benefits of daily aspirin, and suggests that, for the first 3 years, there is no benefit on cancer risk from taking daily low-dose aspirin. However, the reduced risk of major vascular events in the first 3 years is offset by an increased risk of major bleeding. Furthermore, aspirin was not found to have any statistically significant effect on the risk of death from cancer, compared to controls, until at least 5 years of follow-up. This is consistent with the previous meta-analysis, which also found no benefit on cancer death during the first 5 years. Even so, these risk profiles stratified by time should be interpreted with caution. The evolution of these risks may reflect that, for example, people who have not had a major bleed in the first 3 years of aspirin treatment, and therefore remain in a study after this time, may be less likely to have a major aspirin-associated bleed in subsequent years, because of their individual risk profile. Also, the maximum duration of follow-up was not stated, nor is it clear if follow-up continued beyond the duration of the trial (i.e. when patients may have discontinued aspirin).
Any potential benefits of aspirin on cancer need to be balanced against its risks, especially of fatal and non-fatal bleeds, including haemorrhagic stroke. Health professionals should be ready to advise those considering taking aspirin to prevent cancer on their risk of vascular events and of extracranial bleeds over time (see Table in Study details below). It is not yet clear what groups of patients might benefit most and be at the lowest risk from the harms of aspirin, especially as the absolute benefits and risks depend on the baseline risks of the patient. The Patient Decision Aid (PDA) for aspirin for primary prevention also may be helpful as it considers the benefits and harms of aspirin for people who are at different baseline risks of CV disease.
This meta-analysis was published alongside two other systematic reviews on aspirin and cancer. One study showed that taking aspirin reduces the risk of cancer with distant metastasis. The other study reported that, from observational studies, aspirin reduces the long-term risk of several cancers and the risk of distant metastasis. It also found that estimates from observational studies of the effect of aspirin on long-term incidence and mortality due to cancer, and the effects on metastasis are consistent with those from RCTs.
In conclusion, although this study adds to the evidence on aspirin, it is still premature to consider routine administration of daily aspirin to reduce the risk of developing cancer or of dying from it, especially when balancing the benefits against risks of taking aspirin, (particularly in the first few years). Indeed, this study is hypothesis generating and not hypothesis testing. A well-designed, prospective RCT, with a full intention-to-treat analysis and long-term follow-up over several years would ideally be required to establish whether or not aspirin has a place in preventing cancer.
Rothwell PM, Price JF, Fowkes FJR, et al. Short-term effects of daily aspirin on cancer incidence, mortality, and non-vascular death: analysis of the time course of risks and benefits in 51 randomised controlled trials. Lancet 21 March 2012 doi:10.1016/S0140-6736(11)61720-0
Meta-analysis of data from eligible RCTs
Patients, intervention and comparison
51 RCTs of 77,549 patients randomly assigned to daily aspirin (any dose) or no aspirin for primary or secondary prevention of CV events. Trials of patients given aspirin for the treatment or secondary prevention of cancer or colonic polyps were excluded.
Selected outcomes and results
Death due to cancer
Allocation to aspirin was associated with fewer deaths due to cancer versus control (OR 0.85, 95% CI 0.76 to 0.96, p=0.008, 34 RCTs, n=69,224). However, when stratified by time from randomisation to cancer death, there was no statistically significant effect from aspirin during the first 5 years of follow-up (p > 0.18). After 5 or more years’ follow-up the OR for cancer death was 0.63 (95% CI 0.49 to 0.82, p=0.0005) in the patients randomised to aspirin.
In the 12 primary prevention trials (n=42,356), taking aspirin did not have a statistically significant effect on all-cause mortality (p = 0.06) This is because, even though the risk of non-vascular death was reduced (OR 0.88, 95% CI 0.78 to 0.98, p=0.02), the risk of vascular death (which included fatal haemorrhages) was not (OR 0.99, 95% CI 0.87 to 1.12).
Table Risks and benefits of 75 mg –100 mg aspirin daily in primary CV prevention (6 RCTs, n=35,535, duration of scheduled treatment in trial 1 –10.5 years )
OR (95% CI)
Difference in aspirin group vs control, per 100,000 patients per year
|New cancers diagnosed||Major vascular events||Major extra-cranial bleeds||New cancers diagnosed||Major vascular events||Major extra-cranial bleeds|
|0–2.9 years||1.01 (0.88 to 1.15)||0.82 (0.72 to 0.92)||1.95 (1.47 to 2.59)||NS difference||204 fewer||133 extra|
|3–4.9 years||0.81 (0.67 to 0.98)||1.00 (0.84 to 1.20)||1.37 (0.87 to 2.14)||219 fewer||NS difference||NS difference|
|5 years or more||0.70 (0.56 to 0.88)||0.93 (0.74 to 1.16)||0.63 (0.34 to 1.16)||480 fewer||NS difference||NS difference|
NS difference = no significant difference
The authors state that this study was completely independent of any pharmaceutical company or other commercial interest.
Please comment on this rapid review using our feedback form.
Make sure you are signed up to NPC Email updates — the free email alerting system that keeps you up to date with the NPC news and outputs relevant to you.