Regular aspirin use for at least 10 years appears to reduce the risk of colorectal cancers that overexpress cyclooxygenase-2, Dr. Andrew T. Chan and his associates reported.
The researchers mailed questionnaires every 2 years to 121,701 women in the Nurses' Health Study and 51,529 men in the Health Professionals Follow-Up Study to determine aspirin use and incidence of colorectal cancer. The women (age range at entry, 30–55 years) received the survey starting in 1976, and the men (age range at entry, 40–75 years) received it starting in 1986.
More detailed questions on aspirin use, including frequency and amount, were added in 1980 for the women and in 1992 for the men. For women, regular aspirin use was defined as taking two or more 325-mg aspirin tablets per week; for men, it was defined as using aspirin at least twice a week (N. Engl. J. Med. 2007;356:2131–42).
Medical and pathology reports were obtained for participants who reported colorectal cancers, said Dr. Chan of Massachusetts General Hospital and Harvard Medical School, Boston, and his associates.
During follow-up, 636 specimens sufficient for immunohistochemical analysis were obtained from patients with confirmed colorectal cancer. Of the 636 tumors, 423 (67%) were COX-2 positive, or had moderate or strong expression of the enzyme.
The researchers analyzed the association between the expression of COX-2 in the tumors and the patients' use of aspirin, and calculated multivariate relative risk after adjustment for factors including age, gender, smoking, BMI, exercise, family history of colorectal cancer, history of polyps, and meat and alcohol consumption.
The multivariate relative risk of colorectal cancer for aspirin users vs. nonregular users was a significant 0.64 for COX-2-positive cancers and a nonsignificant 0.96 for COX-2-negative tumors. Thus, aspirin use was of benefit only in tumors in which COX-2 was overexpressed. However, this benefit was not seen until aspirin had been used for more than 10 years.
A greater amount of aspirin use also was associated with lower incidence of COX-2-positive disease, with more than five tablets per week associated with significantly fewer such cancers; the association was not significant for COX-2-negative disease. This “is consistent with the results of studies in which higher doses of aspirin were required to inhibit COX-2 than to inhibit COX-1,” the investigators noted.
The results of this observational study “suggest that the anticancer benefit of aspirin is mediated, at least in part, by inhibition of COX-2,” Dr. Chan and his associates concluded.
In an accompanying editorial, Dr. Sanford D. Markowitz of Case Western Reserve University, Cleveland, pointed out that aspirin use has its own risk of adverse effects (N. Engl. J. Med. 2007;356:2195–8).
Researchers “need to ask whether there are alternative strategies for targeting the COX pathway that have better efficacy or lower rates of adverse effects,” Dr. Markowitz said. Inhibitors of the COX-2-generated prostaglandin PGE2 receptors or synthases “might provide better specificity for the prevention of colon cancer and, hence, reduced adverse effects,” he suggested.