WASHINGTON--Addressing an American Gastroenterological Association (AGA)
symposium on NSAIDs during Digestive Disease Week, Dr. Robert Sandler,
of the University of North Carolina, Center for Gastrointestinal Biology
and Disease, posed two questions:
- Do these compounds prevent colorectal cancer?
- Should we use them?
Quoting his favorite social philosopher, former NY Yankee catcher Yogi
Berra, Dr. Sandler advised, "You can observe a lot by watching."
To answer the first question, Dr. Sandler reviewed a series of case-control
studies, cohort studies, and clinical trials in which researchers assembled
subjects using or not using NSAIDs regularly and then "watched"
Results of the case-control studies are remarkably consistent, with
all showing that NSAIDs do have a protective effect against colorectal
Among the cohort studies, all but one show a protective effect. Results
of the randomized controlled trials, although few, are similar. Dr. Sandler
cautioned that no randomized trial has lasted longer than five years, and
colon cancer may take up to 10 years to develop.
Should We Use Them?
While the answer to the first question, appears to be "Yes,"
the second question is not answered simply. Dr. Sandler cautioned against
extrapolating from results of studies involving high-risk patients to average-risk
Also, there is some evidence of decreased efficacy as length of use
in-creases. Results are showing a growing number of persons whose polyps
regress but who nevertheless develop colorectal cancer. "We may be
preventing polyps, but not cancer," he said.
Finally, since the drugs are not completely without side effects, the
risks involved in taking NSAIDs must be considered. "As gastroenterologists,
we often suffer from tunnel vision," he warned. "We tend to look
only at colorectal cancer, ignoring the fact that when we give a systemic
agent, it has an effect on a number of different organs."
He called attention to the Guide to Clinical Preventive Services, published
by the US Preventive Services Task Force in 1996, which declined to recommend
aspirin therapy to prevent myocardial infarction (MI) in asymptomatic patients
because the "balance of risks and benefits of these therapies . .
. is not resolved."
Death from MI is six times more prevalent than death from colorectal
cancer, Dr. Sandler said. "If you can't recommend using aspirin to
prevent MI, I don't think you can recommend it routinely to prevent colorectal
For individuals at average risk for development of colorectal cancer,
Dr. Sandler argued that 70% to 90% of cases can be prevented by regular
colonoscopic surveillance. For those with a family history of colon cancer
or polyps, he recommends endoscopic surveillance. For patients with familial
adenomatous polyposis (FAP), surgery is the most effective approach.
He did suggest, however, that aspirin therapy might be appropriate in
two settings. For patients with frequent polyps, the use of aspirin or
other NSAIDs could reduce the necessary interval for colon-oscopy and might
also reduce the size of the polyps. He also recommended the use of NSAIDs
in the context of randomized controlled trials designed to learn more about
dose, duration, and toxicity.
The ideal chemopreventive should be safe, effective, and nontoxic over
the long haul. While aspirin and the nonsteroidals are promising, they
are risky, he said. Until more definitive chemopreventive studies are available,
Dr. Sandler recommends steering average-risk patients to a sensible low-fat
diet, regular exercise, and avoidance of obesity. And hold the aspirin.