Friday October 28 2011
The study was in people with a genetic risk of bowel cancer
Aspirin cuts the risk of bowel cancer in people with inherited susceptibility to the disease, The Guardian has today reported. The newspaper said that a study of long-term aspirin use found it cut the risk of bowel cancer by more than 60% in these individuals.
The news is based on research that examined how effectively aspirin prevented bowel cancer in over 800 patients with Lynch syndrome, a rare genetic condition that raises the risk of several types of cancer.
The researchers gave half the participants aspirin and the other half a dummy placebo for two years, looking at how many people from each group had developed bowel cancer in the years that followed. When the researchers analysed data on those participants that had completed the full two years of treatment, they found that the group taking aspirin had 63% lower rate of the disease in the 5-10 years that participants were followed.
This well conducted trial was the first to look at aspirin preventing cancer in this way. It suggested yet another use for the humble aspirin pill, which already has proven benefits in fighting heart disease and preventing blood clots. However, it should be remembered that the results apply only to people with this specific genetic condition, which is behind around 2-7% of bowel cancers. Also, regular aspirin is not suitable for everyone as it can cause side effects such as ulcers and stomach bleeds.
Where did the story come from?
The study was carried out by researchers from numerous universities and hospitals throughout the UK and around the world. The research was funded by a number of governmental bodies, medical institutions and research foundations. The research was partially funded by Bayer Pharma, a pharmaceutical company that manufactures aspirin.
The study was published in the peer-reviewed medical journal The Lancet.
The media has generally reported the story accurately and appropriately. However, several news outlets reported that the research involved people with a family history of the disease. This could be misleading as the participants had a specific genetic condition that predisposed them to the disease and were not recruited solely based on people in their family having a history of bowel cancer. While those with the condition would have inherited the faulty genes involved from their parents, it does not necessarily mean their parents had the condition.
What kind of research was this?
This was a randomised control trial that looked at aspirin’s ability to prevent bowel cancer in a group of patients with a condition called Lynch syndrome. The syndrome is a type of inherited cancer in which people carry a faulty DNA mismatch repair (MMR) gene, which would ordinarily aid the repair of DNA mutations that occur during DNA replication, which can lead to cancer. The condition is also called hereditary nonpolyposis colorectal cancer (HNPCC).
These faulty genes predispose those with the condition to polyps and colorectal cancer but also cancers in other locations, including:
- the stomach
- small intestine
- gallbladder ducts
- upper urinary tract
People with this genetic condition are at significantly higher risk of developing bowel cancer than the general population. It is estimated that approximately 2-7% of colorectal cancer cases are caused by the syndrome. Patients with a genetic predisposition that puts them at higher risk of developing a disease are often enrolled in trials such as this to ensure a sufficient number of cases will appear during the relatively short timeline.
The particular study design used - a randomised control trial - ensures that patients are randomly assigned to take either the genuine treatment or an inactive placebo and then follows them over time. Since they have an equal chance of receiving either treatment this should lead to an equal distribution of people with factors that might otherwise bias the results.
Since randomised control trials remove the influence of these ‘confounding’ factors and assess people before they start their treatment, they are capable of establishing a cause-and-effect relationship. This means that, if conducted and analysed correctly, the study design itself ensures we can be fairly confident that any difference in bowel cancer rates between the two groups is due to treatment with aspirin.
What did the research involve?
The researchers recruited 937 people with Lynch Syndrome to participate in the research. They randomised the participants to receive either 300mg of aspirin twice a day or a placebo. Of these, 76 participants (8%) opted out of the study due to aspirin sensitivity or history of ulcers.
The researchers predicted that those who received aspirin would go on to have lower rates of bowel cancer than those who received placebo. Two years after treatment began they compared rates of bowel cancer in the two groups, as well as rates of other cancers common to Lynch syndrome. They followed-up participants for a number of years to see if they developed a cancer, which was up to 10 years after randomisation for some early recruits.
The data was analysed based on the 861 participants enrolled in the aspirin portion of the study. The researchers adjusted for variables that might account for the rates of cancer, such as sex and the cancer history of each participant since randomisation. They then calculated the Incident Rate Ratio (IRR) to estimate the effect of aspirin. IRR is a measure that expresses how the rates of new cancer cases between two groups relate to each other.
The data was first analysed on an intention-to-treat (ITT) basis, where all participants who were randomised were included in the analysis regardless of whether or not they stopped or started aspirin during the study or otherwise complied with the treatment regimen. This has the advantage of being more like treatment under real-world conditions, where some patients may not take their medication as prescribed or stop taking it all together. It also prevents bias from influencing the results, as people who do not comply with treatment may be different from those who remain in the study in ways that impact their risk of developing bowel cancer.
They also performed a second analysis that included only those participants who completed the full treatment regimen of two years or more.
What were the basic results?
Data analysis revealed that out of the 861 people enrolled, 20 developed bowel cancer within two years after randomisation, 10 people from each treatment group. After 10 or more years, however, a further 28 people had developed the disease, eight from the aspirin group and 20 from the placebo group. This equates to an overall 18 cases in the aspirin group and 30 cases in the placebo group.
Intention-to-treat analysis revealed a protective effect of treatment, with the aspirin group having a 44% lower rate of bowel cancer than the placebo group (IRR 0.56, 95% CI 0.32 to 0.99, p=0.05).
The restricted analysis, based on only those who had complied with the treatment regimen, revealed that:
- patients who completed two-or-more years of aspirin treatment had a 63% lower rate of bowel cancer compared to patients who took the placebo pill for two-or-more years (IRR 0.37, 95% CI 0.18 to 0.78, p=0.008)
- patients who completed less than two years of aspirin treatment showed no significant difference in bowel cancer rates compared to those who took the placebo for two or more years (IRR 0.90, 95% CI 0.42 to 1.91, p=0.77)
How did the researchers interpret the results?
The researchers concluded that their study provided clear evidence of aspirin’s effectiveness at preventing bowel cancer among Lynch syndrome patients. They conclude that ‘the case for prescription of aspirin to this high-risk group is clear.’ They said the results are consistent with more than 20 years of observational study findings that show a reduced risk of developing bowel cancer among those who regularly consume aspirin.
This was a well-designed long-term trial. It examined the effect of regular aspirin consumption on bowel cancer rates in a specific group of patients with a raised risk of developing bowel and other cancers. The results indicate that regular treatment with aspirin is an effective method of preventing bowel cancer in this group of high-risk patients.
The study had several strengths, particularly related to design of the trial. For example, even at the end of the study neither the participants nor investigators were made aware of which individuals had received aspirin and which had received placebo. This helps ensure an unbiased analysis of long-term follow-up data, increasing the confidence we can have in the results.
There are, however, several things to consider when interpreting the results:
- This study examined treating patients with a genetic condition that gives them significantly higher risk of developing bowel cancer than the general population. It is unclear whether, based on this study, aspirin would have a protective effect for everybody, and what the size that effect would be.
- Aspirin treatment is not suitable for everyone as it has several known side effects. These include an increased risk of developing ulcers and hemorrhagic stroke. Before treatment began, 8% of those initially recruited had to be excluded due to a medical history suggesting they could be prone to side effects.
- No information was provided regarding side effects in those who went on to receive treatment. Data on such side effects is necessary to ensure that the benefits of treatment outweigh the risks.
- This trial had few observed cases of bowel cancer, even after 10 years of follow-up. Splitting the participants into subgroups based on compliance further reduces the number of cases included in the analysis, which decreases the certainty surrounding these results.
- Intention-to-treat analysis has the advantage of being more like treatment under real-world conditions, where some patients may not take their medication as prescribed. It also prevents bias from influencing the results, as people who do not comply with treatment may be different from those who remain in the study in ways that impact their risk of developing bowel cancer. As the largest effect was seen in patients who had complied with treatment for two years or more it makes generalising this result to a real-world setting difficult, as many people in real life stop their therapy.
- The publication did not indicate whether or not participants continued taking aspirin after the close of the treatment period or whether any participants from the placebo group started taking aspirin. Such patient choices may have influenced the results.
- The researchers say that, importantly, the mechanism by which aspirin protects against cancer development long after patients stop taking the drug is not known. Such knowledge of the natural history of the disease and mechanism of action could be important for designing future trials.
The researchers have planned a follow-up trial with thousands of participants in order to address these points. The trial will also compare multiple doses of aspirin, in order to identify the lowest effective dose. Ensuring that the lowest possible dose is used should help to balance the risk of developing ulcers with the benefit of preventing cancer.