Constitutive cyclooxygenase (COX-1) and inducible cyclooxygenase (COX-2): rationale for selective inhibition and progress to date.

While a great deal has been discovered concerning the potential physiological and pathological role of prostanoids, much is left to be determined. The widespread distribution of both COX-1 and COX-2 coupled with the capacity of most vascular beds, smooth muscle, as well as leukocytes to respond to prostanoids make drawing generalities difficult. The problems with the majority of currently used NSAIDs are clear and ulcerogenic liability is of obvious concern. Interestingly enough, the mechanism of that damage is still the subject of controversy as illustrated by the recent review and hypothesis of Somasundaram et al. In this treatise, the suggestion is made that the initial gastric damage is the result of uncoupling of oxidative phosphorylation which is independent but simultaneous with COX inhibition. At least two currently marketed NSAIDs have improved G.I. liability (nabumetone and etodolac) with efficacy equivalent to other more ulcerogenic NSAIDs. These drugs appear to have achieved that by a mechanism distinct from selective inhibition of COX-2. Whether or not selective COX-2 inhibitors will demonstrate an improved profile over these compounds remains to be shown. Unfortunately, clinical experience with nimsulide and CGP 28238 suggest that NSAID-like toxicity may still be an issue. The promise of selective COX-2 inhibitors remains largely untested. It is with great interest and expectation that the clinical evaluation of the more selective compounds of different structural types is awaited.