Neuroinflammation is a central feature of Alzheimer disease (AD). It involves an innate immune reaction of sufficient intensity that self attack on neurons occurs. This phenomenon is best described as autotoxicity to distinguish it from classical autoimmunity which involves cloning of peripheral lymphocytes. Many compounds have been identified in AD brain which are known to promote and sustain inflammatory responses. They include beta-amyloid protein; the pentraxins C-reactive protein and amyloid P; complement proteins; the inflammatory cytokines interleukin-1, interleukin-6 and tumor necrosis factor-alpha; the protease inhibitors alpha-2-macroglobulin and alpha-1-antichymotrypsin; and the prostaglandin generating cyclooxygenases COX-1 and COX-2. Orally effective agents which can counteract the influence of these inflammatory stimulators should be effective in treating AD. Epidemiological evidence, coupled with results from pilot clinical trials, suggest there is great promise for traditional COX-1 inhibiting NSAIDs. Inhibitors of mediators closer to the core processes might offer even greater therapeutic promise. Some theoretical opportunities are suggested, based on intervention in the action of the above mentioned mediators.