Prostaglandin G/H synthetase or cyclooxygenase (COX) was discovered in the 1970s, when it was noted that aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) inhibited its proinflammatory activity. Subsequent studies in humans revealed that COX-2, one of the two COX isoforms, participated in colon cancer progression (familial adenomatous polyposis [FAP], Gardner’s syndrome, and sporadic colon cancer) and that its inhibition by the NSAIDs could have a preventive effect. These studies generated a great deal of scientific and commercial enthusiasm culminating in the approval of celecoxib in 1999 by the US Food and Drug Administration for the prevention of colorectal polyps in patients with FAP. COX-2 is the key in the metabolism of arachidonic acid and it catabolizes the biosynthesis of five function-specific prostaglandins (PG) including PGE2, PGD2, PGF2a, PGI2, and thromboxane. 9 As opposed to COX-1, which is physiologically constitutively expressed in a variety of tissues, COX-2 expression is absent under normal conditions but its expression is rapid in response to proinflammatory cytokines and certain hormones. COX-2 expression is aberrantly upregulated in a wide variety of human preneoplastic and neoplastic (breast, gastrointestinal, bladder, lung, and prostate) conditions. COX-2 overexpression can lead to the production of PGE2 and to a lesser extent thromboxane, both implicated in angiogenesis, cell proliferation, cell survival, migration, invasion, and modulation of host immune cells. Importantly, COX-2–derived PGE2 can cross-talk with other molecular pathways important for cancer progression (Fig 1). An example would be the activation of the epidermal growth factor receptor pathway responsible for increased cell motility, invasion, and metastases (Fig 1). Since 1992, there have been approximately 10,962 COX-2– related publications and many studies of human cancer have associated its overexpression with poor patient outcome as well as with large tumor size and lymph node and distant metastases. In this issue of the Journal, de Maat et al report on COX-2 in gastric cancer by adopting a novel approach to judge the impact of “underexpression or lack of expression” on patient outcome. To our