Regular aspirin intake only reduces incidence of colorectal tumors that overexpress COX-2

May 24 2007

Aspirin therapy's ability to reduce the risk of colorectal cancer, an association seen in a large number of studies, appears to depend on the drug's inhibition of the COX-2 enzyme, the action that also underlies aspirin's usefulness for treating pain and inflammation.

In the May 24 New England Journal of Medicine, investigators from Massachusetts General Hospital (MGH), Dana-Farber Cancer Institute and Brigham and Women's Hospital report that regular aspirin intake only reduced the incidence of colorectal tumors that overexpress COX-2.

"We knew that aspirin can block COX-2 function and that COX-2 is present in the vast majority of colorectal tumors but not in normal colon tissue," explains Andrew Chan, MD, MPH, of the MGH Gastrointestinal Unit, the paper's lead author. "Therefore we hypothesized that, if blocking the COX-2 pathway was the mechanism underlying aspirin-associated risk reduction, it should preferentially reduce the incidence of those tumors that rely on COX-2."

To investigate that theory, the research team compiled data from two ongoing prospective research studies , the Nurses Health Study (NHS) and the Health Professionals Follow-Up Study (HPFS). Both studies gather comprehensive health information on their participants every two years, which is analyzed for associations between factors such as diet and the incidence of several diseases. Both the NHS, which enrolls more than 120,000 female registered nurses, and the HPFS, following more than 50,000 men employed in the health professions, have previously found associations between aspirin intake and reduced colorectal cancer risk.

For the current study, the researchers focused on almost 83,000 NHS participants and about 47,000 HPFS participants for whom necessary information was available. They received permission to acquire medical records and pathology reports from those who had reported being diagnosed with colorectal cancer, then retrieved more than 600 pathology specimens from participants whose diagnoses could be confirmed. The samples were analyzed for expression of COX-2.

As seen in previous reports, among the more than 120,000 participants, those who took at least two standard aspirin tablets a week had about three-quarters the risk of colorectal cancer that aspirin non-users did. However, analysis of tumor samples showed that reduction in risk only applied to tumors that expressed COX-2. The incidence of COX-2-negative tumors was virtually the same among those who did and did not take aspirin.

"These results will allow us to test another hypothesis: that in patients who have had colorectal cancer or polyps in the past, expression of COX-2 in the earlier lesion might indicate those for whom aspirin could reduce the risk of recurrence," says Charles Fuchs, MD, MPH, of Dana-Farber, the study's senior author. "Answering that will be our next target."

Both researchers note that current evidence does not support generally recommending aspirin therapy to reduce colorectal cancer risk. "For most people, the best way to prevent colorectal cancer is through screening, which we know saves lives by allowing us to treat polyps before they become cancers," says Chan.

Fuchs adds, "An individual who has had colorectal cancer in the past is at higher risk for subsequent tumors, and that might be someone who should discuss the advisability of taking aspirin with his or her primary care physician. We hope that our future research will further clarify who would benefit most from regular aspirin therapy and that understanding the mechanism of this effect will lead to new preventive and treatment strategies."

http://www.mgh.harvard.edu