Tuesday December 7 2010
The doses examined here were low, at only 75mg a day
“A daily dose of the ‘wonder drug’ aspirin is the simplest way to avoid dying from the most common cancers,” says the Daily Express.
This news story is based on a study of data from eight clinical trials in more than 25,000 people, which recorded daily aspirin use and deaths from cancer. In total, there were 674 deaths from cancer during the studies’ follow-up periods. People taking aspirin were less likely to die from cancer than those who weren’t taking aspirin.
Overall, this study was well-conducted and its findings will probably be taken into account with other evidence in the next review of the clinical guidance for cancer prevention. On its own, however, the study does not present strong enough evidence for aspirin to be universally recommended. This is because long-term aspirin is associated with a risk of internal bleeding, particularly in the elderly. Different people may also have different levels of benefit.
People who want to start taking aspirin should speak to their GP first. Importantly, the doses examined here were low, at only 75mg a day, which is a quarter of what over-the-counter pills for pain relief contain.
Where did the story come from?
The study was carried out by researchers from the University of Oxford, the University of Edinburgh, the University of Dundee, Kumamoto University, Japan, and the London School of Hygiene and Tropical Medicine. No specific sources of funding are mentioned. It was published in the peer-reviewed medical journal The Lancet.
The story was reported by a large number of newspapers. The accuracy of the coverage varied, with some newspapers reporting that aspirin reduced the risk of developing cancer, when the study actually only looked at the risk of dying from cancer. Most newspapers sensibly caution that there are risks associated with taking aspirin long term and that people should check with a doctor before starting to take it regularly.
What kind of research was this?
The researchers say that research in animals suggests that aspirin inhibits the occurrence or development of tumours, but the evidence in humans has so far been lacking. Observational studies in people have had conflicting results, with some showing that aspirin reduces the risk of some cancers, but only weaker associations being found in more rigorous studies. There is a similar picture for randomised trials of aspirin for cancer prevention, with some trials finding that aspirin cuts the risk of colorectal cancer, but others finding little evidence of benefit for other cancers.
There have been several clinical trials of aspirin for reducing the risk of vascular events (such as angina). These trials often follow people for several years and reliably include details on the cause of death of any participants. Here, the researchers combined data from these studies to examine which people died from cancer while taking part in these trials.
Combining the results of several studies in this way is a valid approach for searching for associations. It should be noted, however, that none of the original trials were designed to study cancer risk, and some of the included studies did not follow the patients for as many years as others. As such, these results ideally need confirmation in further research specifically designed to answer the question of whether aspirin can prevent cancer.
What did the research involve?
The researchers identified clinical trials in which any dose of aspirin had been compared with no aspirin, including trials in which aspirin was being compared with another antiplatelet drug or an antithrombotic drug, for example warfarin. Studies were identified from a number of research databases and were selected if they had followed people for an average (median) of at least four years. The researchers then contacted the authors of the original studies to obtain data on individual patients. The analysis was limited to looking at fatal cancers only, as the data for these were more reliable and easier to obtain.
The studies identified by the researchers were:
- The Thrombosis Prevention Trial (TPT, 5,085 people)
- The British Doctors’ Aspirin Trial (BDAT, 5,139 people)
- The UK Transient Ischaemic Attack Trial (UK-TIA, 2,435 people)
- The Early Treatment Diabetic Retinopathy Study (ETDRS, 3,711 people)
- The Swedish Angina Pectoris Aspirin Trial (SAPAT, 2,035 people)
- Japanese Primary Prevention of Atherosclerosis with Aspirin for Diabetes (JPAD, 2,539 people)
- Prevention of Progression of Arterial Disease and Diabetes (POPADAD, 1,276 people)
- Aspirin for Asymptomatic Atherosclerosis (AAA, 3,310 people)
In all the studies data were available on cause of death during the study, but in the three UK-based studies (TPT, BDAT, UK-TIA) the researchers were able to obtain additional long-term data by checking national death certification and cancer registration databases. Data from individual patients were not available for one trial (SAPAT), so these could not be included in the detailed analyses.
What were the basic results?
In total, there were 674 deaths from cancer in 25,530 trial participants. Pooling data from all eight studies showed that people who were allocated to receive aspirin were significantly less likely to die from cancer than people who were not (odds ratio (OR) = 0.79, 95% confidence interval (CI) 0.68 to 0.92, p=0.003). This represents the 21% reduction in risk or odds of death quoted by the newspapers.
Looking at different types of cancer in the seven studies with individual patient data, the researchers found no association between taking aspirin and risk of death from cancer in patients who had been followed up for five years or fewer. Limiting the analysis to patients with more than five years of follow-up showed a reduction in the odds of death from all cancers combined (hazard ratio (HR) = 0.66, 95% CI 0.50 to 0.87, p=0.003). This association was also observed in the combined data for deaths from all gastrointestinal cancers (HR=0.46, 95% CI 0.27 to 0.77, p=0.003) and all solid tumours (HR=0.64, 95% CI 0.49 to 0.85, p=0.002) in patients with more than five years’ follow-up.
Limiting the analysis to the three UK studies in which additional long-term data were available and patients had received aspirin for at least five years (10,502 patients) confirmed that daily aspirin was associated with a lower risk of all cancers (HR=0.78, 95% CI 0.70 to 0.87, p<0.0001). These patients had been followed for up to 20 years.
How did the researchers interpret the results?
The researchers report that “aspirin reduced the risk of death due to cancer by about 20% during the trials”. They point out that this reduction is mostly due to reduction in deaths after five years of treatment. They point out that the original trials involved different types of people, which suggests that the results should be applicable to the general population. They note that aspirin did not seem to be any more beneficial at doses greater than 75mg a day.
Overall, this study was well-conducted and its findings will probably be taken into account with other evidence during the next review of the clinical guidance for cancer prevention. On its own, however, the study does not present strong enough evidence for aspirin to be universally recommended. When looking at the results, the following should be taken into consideration:
- The eight studies on which this research is based were originally designed to look at aspirin for the prevention of vascular events (such as stroke). The results are, therefore, not as robust as they would have been had the researchers specifically set out to look at the effect of aspirin on cancer. Clinical trials with this aim may give a better insight and enable researchers to look at cases of cancer and not just deaths from cancer.
- Although the pooled number of people included from the studies is large, the number of deaths from some types of cancer was comparatively small (although, as expected, this increased with age). To understand the extent to which aspirin is protective against specific cancer types, further clinical trials or prospective studies designed to look at those specific diseases are needed.
- The authors point out that the original trials didn't include sufficient women for them to assess whether there was any association with breast cancer or any other gynaecological cancers, and this is an area for further research.
With any medicine, there is a balance between the risks and benefits of that treatment, and the benefits need to outweigh the potential harms. The issue here is that taking aspirin increases the risk of internal bleeding, particularly in the elderly. Aspirin benefits people who are at risk of cardiovascular disease, but the benefits for healthy people are still unclear.
People who want to start taking aspirin should speak to their GP first. Importantly, the doses in these studies were low, at only 75mg a day, which is a quarter of what over-the-counter pills for pain relief contain.