Rationalizing cyclooxygenase (COX) inhibition for maximal efficacy and minimal adverse events.

New information indicates that cyclooxygenase-2 (COX-2) is constitutively expressed in several tissues, including brain, lung, pancreas, kidney, and ovary, and plays an important role in renal and gastrointestinal function. Selective COX-2 inhibition has been associated in animal studies with impairment of ulcer healing and renal function and inhibition of prostacyclin, an effect that inhibits vasodilation without inhibiting platelet aggregation. The clinical consequences, if any, of these effects remain to be determined in long-term studies in humans. The premise that selective COX-2 inhibitors will cause less gastrointestinal toxicity than nonsteroidal antiinflammatory drugs that inhibit both COX isoforms needs to take into account the low toxicity of nabumetone. The gastrointestinal safety profile of this nonacidic, dual COX inhibitor that does not undergo enterohepatic circulation has been evaluated in extensive clinical trials. The data submitted to the US Food and Drug Administration in the New Drug Application for nabumetone (Relafen), the comparative trials subsequently completed, the published databases of the comparative gastrointestinal toxicity of various nonsteroidal anti-inflammatory drugs (NSAIDs), and the meta-analysis published in this issue of The American Journal of Medicine (Schoenfeld, page 48S) indicate that nabumetone has the lowest incidence of gastrointestinal toxicity among the extensively studied NSAIDs. Overall, the incidence is approximately 10-fold less than with comparator drugs. This rate is an appropriate current reference against which the gastrointestinal toxicity of COX-2 inhibitors can be compared.