Thursday August 12 2010
Aspirin use carries a small risk of stomach bleeding
Men may be running out to buy aspirin following a story in the Daily Mail, which reported: “Aspirin 'cuts prostate cancer risk by 30%'.” It claimed that scientists have found that a standard aspirin tablet taken daily has a powerful protective effect against the disease.
This study compared the history of aspirin use in 1,001 men with prostate cancer and 942 men without the disease. It found that those who had taken aspirin in the past and those who took daily low-dose aspirin were less likely to have prostate cancer. Limitations to this type of study and the mixed findings of previous studies mean that it is difficult to say conclusively whether aspirin reduces the risk of prostate cancer. The best way to determine this would be a long-term randomised controlled trial that compared aspirin to a placebo.
Potential benefits of aspirin in reducing the risk of cancer would need to be weighed up against potential harms, as aspirin increases the risk of gastrointestinal (stomach) bleeding.
Where did the story come from?
The study was carried out by researchers from the Fred Hutchinson Cancer Research Center and other research centres in the USA. It was funded by the National Cancer Institute, the US Department of Defense, the Fred Hutchinson Cancer Research Center and the National Human Genome Research Institute. The study was published in the peer-reviewed American Journal of Epidemiology.
The Mirror, Daily Express and Daily Mail reported this research accurately. The Mail importantly noted that other studies have had mixed findings, and that aspirin can increase the risk of stomach bleeding. None of the reports noted the researchers’ call for further research.
What kind of research was this?
This was a case-control study which looked at the effects of aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs) on the risk of prostate cancer.
This type of study design is often used to look at relatively rare outcomes, such as cancer, as it is quicker to do than a prospective cohort study and requires fewer people. However, because this type of study asks people about their exposures in the past (how much aspirin they took in this case), there is always a risk that people may not be able to recall this accurately, which could affect the results. People who have a disease may also recall their exposures differently to those who do not, especially if they think the exposure in question may be related to their disease. In addition, the groups being compared (those with and without prostate cancer) may differ in ways other than aspirin use, which may account for the results. To test whether aspirin can reduce the risk of prostate cancer, and whether this outweighs any harms of aspirin, it would be best to carry out a randomised controlled trial in which aspirin was compared to a placebo.
What did the research involve?
Men who had been diagnosed with prostate cancer (cases) were enrolled between 2002 and 2005 in King County Washington. Another group of men who had never had prostate cancer (controls) were also recruited. The researchers asked these men about their past use of aspirin and other NSAIDs, and looked at whether the proportion of men taking aspirin in the cases and controls differed. If more men without cancer took aspirin, this would suggest that aspirin could have a protective effect.
The prostate cancer group comprised 1,001 white and African-American men aged 35-74, while the control group consisted of 942 men. Men with cancer were identified through a cancer registry database, which also provided information about the stage of their cancer. Participants completed questionnaires about their lifetime use of aspirin, other NSAIDs and paracetamol. Current use was defined as use in the year before their diagnosis with cancer or a similar reference date for the controls.
The questionnaire also asked the participants about their lifestyle, family history of cancer, medical history, use of certain medications and prostate cancer screening history (whether they had had a prostate-specific antigen (PSA) test and digital rectal examination).
Analyses took into account age, race, smoking status, education, income, marital status, prostate cancer screening history (none, digital rectal examination only and PSA testing) and various indications and contraindications for using NSAIDs (reasons for or against taking the drugs).
What were the basic results?
Men with prostate cancer were more likely to be African-American, to have a first-degree relative with prostate cancer and to have had prostate cancer testing before their diagnosis.
Just under half of the cases (48.4%) and just over half of the controls (51.6%) had ever taken aspirin.
The researchers found that men who had ever used aspirin were 18% less likely to have prostate cancer than those who had never used aspirin (odds ratio [OR] 0.82, 95% confidence interval [CI] 0.68 to 0.99). Men who had taken aspirin for over five years were 24% less likely to have prostate cancer than non-users (OR 0.76, 95% CI 0.61 to 0.96). Men who took a daily low dose (81mg) of aspirin were 29% less likely to have prostate cancer than those who never used aspirin (OR 0.71, 95% CI 0.56 to 0.90).
Men currently using aspirin were 21% less likely to have prostate cancer than non-users (OR 0.79, 95% CI 0.65 to 0.96). Former users of aspirin did not differ in their risk of prostate cancer compared with non-users.
Taking other NSAIDs or paracetamol was not related to the risk of developing prostate cancer.
How did the researchers interpret the results?
The researchers concluded, “these results contribute further evidence that aspirin may have [preventative] activity against prostate cancer and highlight the need for additional research”.
This study suggests that there is a link between aspirin use and risk of prostate cancer. There are some points to note:
- As men were required to recall their past aspirin use over their entire lifetime, there may have been some inaccuracies in these estimates, which may have affected the results.
- The authors note that other studies have had mixed results, and although most found a reduction in risk, a few found increased risk of prostate cancer with aspirin. The best way to reconcile these differences would be to carry out a systematic review to assess all such studies and, if their methods are similar enough, to pool their results.
- Some men in the control group may have had prostate cancer that had not yet been detected, and this could affect results.
- There may have been differences other than aspirin use between the cases and controls which affected their risk of being diagnosed with prostate cancer. For example, men who had prostate cancer were more likely to have been tested for prostate cancer before their diagnosis. The differences seen between groups in aspirin use could reflect in part that men who receive screening are more likely to have a cancer found than those who do not. Although the researchers took this and other differences into account in their analyses, they and other factors could still have had an effect.
- Further to this, though the researchers adjusted for race in their analyses, it would have been beneficial to match controls to cases on the basis of ethnicity, in addition to age. Both increased age and ethnicity are recognised risk factors for prostate cancer, with men of black African and Caribbean backgrounds having increased risk. This was demonstrated in this study as a higher proportion of men with prostate cancer were of African-American ethnicity.
- It is difficult to be certain whether aspirin use preceded the development of prostate cancer among cases, as the cancer may have been present before the men’s diagnosis.
- About a third of controls who were asked to participate did not do so. If these men had differing aspirin use to those who did participate, this could affect results.
Ideally, to determine whether low-dose aspirin reduces the risk of developing prostate cancer, a randomised controlled trial would be needed. Aspirin increases the risk of gastrointestinal bleeding, particularly in the elderly, and any potential benefits would need to be weighed up against these risks.