Non-steroidal anti-inflammatory drugs (NSAIDs) have been considered for treatment and prevention of Alzheimer's disease (AD) for more than two decades. Biochemical markers in the brains of individuals with AD suggest that inflammation might be a driving cause of the disease that can be suppressed by drug treatment. In addition, a subgroup of widely used NSAIDs inhibits generation of the pathogenic amyloid-beta(1-42) peptide (Abeta42) independently of the inflammatory cyclooxygenase (COX) pathway. Here, we summarize evidence showing that the efficacy of NSAIDs in AD might be attributable to either anti-inflammatory or anti-amyloidogenic activities, and we acknowledge the possibility that current NSAIDs could be neuroprotective through synergistic mechanisms. Ongoing drug development efforts are concentrating on improvement of the COX-independent Abeta42-lowering activity to prevent amyloid pathology and secondary inflammatory reactions and to avoid the clinical side-effects associated with inhibition of COX.