Can Aspirin Reduce Cancer Risk and Mortality?
The prospect is too enticing to dismiss: a single pill—a cheap one, too—that, when taken regularly, can reduce the risk not only of heart attack or stroke, but also of developing or dying from some types of cancer.
The pill is aspirin, and a host of studies have suggested that it can prevent and possibly reduce the risk of dying from some cancers. The data in support of aspirin’s anticancer effects recently received another boost, with one of the largest analyses to date—a meta-analysis of eight clinical trials involving regular aspirin use, published online December 6, 2010, in The Lancet—showing a substantial reduction in mortality for a number of different cancers.
As is often the case, though, the story is not that simple. None of the trials included in the meta-analysis, for example, was designed specifically to assess whether aspirin reduces cancer incidence or mortality. And although many studies, including a few clinical trials, support the notion that aspirin or other nonsteroidal anti-inflammatory drugs (NSAIDs) can reduce the risk of developing or dying from some cancers, other studies, including clinical trials, have reached a different conclusion.
It would certainly be welcome news to say this century-old pill may be the next great clinical cancer advance. But at this point, the sentiments among researchers in the field are mixed, and no medical organization has issued guidelines or recommendations on aspirin’s use as an anticancer therapy. For the moment, at least, the emergence of aspirin into a cancer prevention role seems to be on hold.
Lots of Data, but Still Unknowns
The meta-analysis, led by Peter Rothwell, M.D., Ph.D., from the University of Oxford, found that after 5 years of follow-up, trial participants who took aspirin daily—regardless of dose—for a mean of 4 years had a 44 percent reduced risk of dying from cancer compared with participants who took a placebo. The largest decrease in risk was for gastrointestinal cancers.
One of the strong points of the meta-analysis, noted Dr. Rothwell, is that it included 20 years of follow-up data for individual participants in three of the eight trials, rather than relying solely on combined data from the original studies. In this subset of patients, there was a 20 percent lower risk of dying from cancer after 20 years, with the largest benefit seen again for gastrointestinal cancers. Overall in the meta-analysis, trial participants who took aspirin had a decreased risk of dying from relatively rarer but highly deadly cancers such as pancreatic and esophageal, as well as from common cancers such as lung and prostate. The longer the duration of aspirin use in the trials, the greater the benefit.
Recently, a similar meta-analysis from the same research team, this one focused specifically on the effect of aspirin use on colorectal cancer, showed large and statistically significant reductions in both incidence and death from the disease. The finding, the study team pointed out, is consistent with results from numerous studies that have shown regular NSAID use reduces or prevents the growth of precancerous growths in the colon.
The mounting evidence of aspirin’s strong anticancer effect is hard to ignore, argued Dr. Rothwell. “At a minimum, in combination with the appropriate screening, the data indicate that aspirin would be highly effective in preventing colon cancer,” he said.
Dr. Rothwell is not alone in his opinion. “I do believe there is a role for using [NSAIDs] at a subtherapeutic level for various types of tumors,” said Randall Harris, M.D., Ph.D., from Ohio State University. Dr. Harris led a prospective observational study of participants in the Women’s Health Initiative (WHI), which found that women who took aspirin or ibuprofen at least twice a week for 5 years or more had a reduced risk of breast cancer. “The data are strong for colon cancer and reasonably strong for breast cancer,” he said.
That sentiment, though, is not universal.
“It would be premature to recommend that people start taking aspirin specifically to prevent cancer,” said Eric Jacobs, Ph.D., strategic director of pharmacoepidemiology for the American Cancer Society, in an e-mail. Potential side effects have to be taken into consideration, he continued, noting that even low dosages of aspirin “can substantially increase the risk of serious gastrointestinal bleeding.”
Reasons for Caution
Much of the hesitation, explained Asad Umar, D.V.M., Ph.D., of NCI’s Division of Cancer Prevention, stems from the lack of data from appropriately designed randomized clinical trials focused specifically on a cancer outcome. “Only when you do a randomized clinical trial do you get the complete picture of what’s going on,” he said, including potential benefits and harms.
Dr. Umar cited the experience with a different NSAID, the COX-2 inhibitor celecoxib (Celebrex). Celecoxib is approved by the FDA as an adjunct to surgery to prevent polyp formation in people with a disorder called familial adenomatous polyposis (FAP) that puts them at an increased risk for colorectal cancer. Early, smaller clinical trials to test whether daily use of celecoxib could prevent the return of colon polyps in people without FAP didn’t reveal any meaningful side effects from the drug, and did show significant reductions in polyp formation and growth. It was only in much larger trials, including the NCI-funded Adenoma Prevention with Celecoxib trial—which assessed effects of using the drug daily for 3 years—that evidence emerged for potential adverse cardiovascular effects of regular, long-term treatment with celecoxib.
The trials included in The Lancet meta-analysis, Dr. Rothwell noted, had more fatal bleeding events among participants who took a placebo than among those who took aspirin (60 versus 40). The analysis also showed a more than 8 percent reduction in all-cause mortality among aspirin users by the end of the trials included in the analysis, “nearly all of which was due to the reduction in cancer deaths,” he said.
Despite the promising data from these studies, John Baron, M.D. of the Dartmouth Medical School said he is not yet ready to endorse aspirin’s widespread use for cancer prevention. “We need to consider all of the benefits and harms that can accrue from aspirin use in various population groups,” he said. “We don’t quite have that yet.”
Filling in the Blanks
In a consensus statement published in 2009, an international group of researchers reviewed the published data on aspirin’s effects on cancer incidence and mortality and laid out what they determined to be the most important clinical questions that still need to be answered. These questions included how long people should take aspirin for the optimum effect, the age when they should start taking it, and the best dose, among others.
Dr. Baron, a statement co-author, admitted that it will be difficult to answer these questions. “The most important issue that needs to be dealt with immediately is the dose issue,” he said.
Opinions on dose vary. A daily low-dose aspirin (such as 81 mg) is likely to be the proper choice, Dr. Rothwell said. But Dr. Harris suggested that a higher dose, up to 325 mg at least twice a week, would also be effective.
In The Lancet meta-analysis, the risk of dying from cancer decreased regardless of the aspirin dose. But observational data from a recent Harvard study showed that, at least in people already treated for colorectal cancer, the extent of the risk reduction seemed to increase with dose. “That certainly needs to be considered,” Dr. Baron said, “because side effects from aspirin also increase with dose, although not strikingly.”
Typically, clinical trials could help to provide more definitive information on dose. But large cancer prevention trials of aspirin are very difficult to conduct, Dr. Umar acknowledged. A trial using cancer mortality as an endpoint would require a very long study with a large number of participants to demonstrate any effect. And many people are already taking aspirin for cardiac prevention and pain relief, Dr. Harris said, so gathering enough participants who don’t take aspirin for the sake of comparison would also pose a difficulty.
As far as clinical trials are concerned, Dr. Umar said, “We need to develop carefully designed studies, identify high-risk groups and surrogate endpoints, and get a genomic signature of response in addition to clinical endpoints.”
Several trials focused on high-risk groups are planned or ongoing. In the United Kingdom, for example, participants are being enrolled in the phase III AspECT trial, which will test whether aspirin in combination with a drug that prevents and treats acid reflux (and may prevent or limit gastrointestinal bleeding) can reduce the risk of esophageal cancer in people with Barrett esophagus, which can be a precursor to the cancer.
Dr. Rothwell predicts that the findings from these trials and other studies will “firm up the evidence” in support of aspirin for cancer prevention.
In the meantime, Dr. Jacobs advised, the average person needs to proceed with an abundance of caution. “Decisions about aspirin use should be made by balancing the risks against the benefits,” he said, “so any decision about daily aspirin use should be made only in consultation with your health care professional.”