Many studies over the past two decades have suggested, but not proved, that taking aspirin helps protect against colorectal cancer. The evidence, however, hasn’t been strong enough to persuade public-health experts to recommend the drug for that purpose.
An experiment answering the question definitively probably will not be done because it would cost too much and take too long. For millions of people taking aspirin to lower the risk of heart attack, the question is largely irrelevant. Whether this study will change physicians’ advice to others is uncertain.
“I believe there is already strong evidence for aspirin use and reduced risk for colorectal cancer,” said Eric Jacobs, an epidemiologist at the American Cancer Society. “The question is really about the overall balance of risks and benefits” — something not addressed in the new study.
Aspirin causes one additional bleeding ulcer for every 1,000 people using it for a year. It also increases the risk of bleeding in the brain, especially in the elderly. In 2007, the U.S. Preventive Services Task Force, a federally appointed panel of experts, recommended against its use “to prevent colorectal cancer in individuals at average risk.”
Asad Umar, an expert on gastrointestinal cancer prevention at the National Cancer Institute, said that for people with Lynch syndrome, daily aspirin use is worthwhile. However, he said, his personal belief is that “the data is not really there” to recommend its use for colon cancer prevention in most people.
About 50 to 70 percent of people with Lynch syndrome develop colorectal cancer. The disease occurs earlier and progresses more rapidly, making them a group in whom preventive measures can be more easily studied than the general population.
In the new study, published in The Lancet, researchers randomly assigned 861 people in 16 countries who had Lynch syndrome to take two aspirin tablets a day or identical-looking placebos. After slightly more than two years, there was no difference between the two groups in cancer incidence.
But after about five years, there was a difference. Eighteen people taking aspirin developed cancer compared with 30 taking a placebo. Not every participant stayed with the program. When only those who took aspirin for two years or more were considered, the reduction in cancer rate was about 60 percent.
People with Lynch syndrome are at higher risk for some other cancers — uterine, pancreatic, stomach — and aspirin cut their risk for those, too.
About 15 percent of cancers in the general population result from mutations in the genes affected in Lynch syndrome — a sort of DNA spell-checker known as “mismatch-repair.” Other studies, however, have shown that aspirin lowers the risk of colorectal cancer resulting from other mutations, too.
“My strong suspicion is that this is not something specific to mismatch-repair,” said John Burn, a physician at Newcastle University in England, who led the study. “I think aspirin is a broader anti-cancer agent.”
A study in The Lancet early this year of 25,000 people in eight different clinical trials found that aspirin reduced cancer deaths by 21 percent. The trials were set up to look at aspirin’s effect on preventing heart attacks, not cancer.
The effect on cancer of the esophagus, pancreas, brain and lungs was seen in about five years, but the reduction in colon and several other cancers appeared later. Numerous studies have also shown that the effect remains for at least a few years after a person stops taking aspirin.
Aspirin is part of a family of chemical compounds called salicylates, which are found in many wild plants. They are part of the plant immune system that helps identify and eliminate infected cells.
One hypothesis is that, in human beings, aspirin helps cells die when they suffer irreparable damage, a process called “apoptosis.” Burn said he suspects that action is occurring at the level of intestinal stem cells, which divide to create new colon cells as old ones are sloughed off.
“That would be the sort of effect that would result in a long-term benefit,” he said.
The Lancet paper and news release accompanying it say the study is the first randomized controlled trial to assess aspirin’s effect on cancer prevention as “the primary endpoint.” However, that does not appear to be the case.
The Women’s Health Study, conducted from 1992 to 2004 by Harvard University researchers, assigned roughly 40,000 American women to take aspirin or a placebo. The number of cancers and heart attacks was counted.
“That’s a good point,” Burn said. “I think we probably overstated our case there.”