Coxibs and Alzheimer's disease: should they stay or should they go?

The neuropathology of Alzheimer's disease (AD) is characterized by deposits of amyloid beta (Abeta) peptides and neurofibrillary tangles, but also, among other aspects, by signs of a chronic inflammatory process. Epidemiological studies have shown that long-term use of nonsteroidal antiinflammatory drugs (NSAIDs) reduces the risk of developing AD and delays its onset. The classic target of NSAIDs is the prevention of cyclooxygenase (COX) activation. The main mechanism of action of COXs is the synthesis of prostaglandins, some of which have potent inflammatory activity. The discovery of two isoforms of this enzyme, COX-1 and COX-2, and that the latter is inducible by inflammatory cytokines supported the hypothesis that its inhibition would result in a potent antiinflammatory effect and led to the rapid development of selective COX-2 inhibitors, collectively called coxibs. Based on this rationale, some coxibs have been used in clinical trials for AD patients, but all the results obtained so far have been negative. Here, we review our knowledge in terms of COX-2 in the central nervous system, COX-2 and Abeta formation, and finally COX-2 and AD pathogenesis to understand the reasons why these drugs have failed and whether there is any scientific support to keep them as therapeutic tools for this chronic disease.