Chronic inflammation is a tissue-specific process implicated in several diseases of an aging population, including cancer, cardiovascular disease, and arthritis. Cyclooxygenase-2 (COX-2) is a mediator of acute and chronic inflammation, and drugs designed to specifically target this enzyme have achieved widespread clinical use. Unfortunately, randomised trials of selective COX-2 inhibitors for cancer prevention have shown that beneficial effects in one type of tissue can be accompanied by toxic effects in another. These trials documented a significant reduction in adenoma formation in patients at high risk for colorectal cancer, with reductions in advanced disease occurrence from 28–66% over 3 years. As a result, these studies provided important evidence for the involvement of COX-2 in early colorectal tumorigenesis. In the same patients, however, these placebo-controlled clinical trials revealed a little-understood relation between COX-2 and maintenance of cardiovascular integrity. During the 3 years of treatment, patients who received selective COX-2 inhibitors were 1·3–3·4-times more likely to have serious cardiovascular events than those treated with placebo. This article will discuss the biological rationale for using selective COX-2 inhibitors in cancer chemoprevention, and outline new avenues of research into toxic effects and tissue specificity that are necessary to allow their successful use in patients at risk for colorectal cancer.