Alzheimer's disease (AD) is the most common cause of dementia in developed countries. AD is characterized pathologically by the presence of senile plaques and neurofibrillary tangles (NFTs), the major constituents of which are the amyloid beta protein (Abeta) and tau protein, respectively. Several epidemiological studies have reported moderately increased risks of AD in diabetic patients compared with general population. In diabetes mellitus, the formation and accumulation of advanced glycation end products (AGEs) progress. Recent understandings of this process have confirmed that AGEs - their receptor (RAGE) interactions may play a role in the pathogenesis of diabetic vascular complications and neurodegenerative disorders including AD. Indeed, it has been demonstrated that AGEs can be identified immunohistochemically to be present in both senile plaques and NFTs from patients with AD. Glycation of Abeta markedly enhances its aggregation in vitro, and the glycation of tau, in addition to hyperphosphorylation, appears to enhance the formation of paired helical filaments. Further, RAGE has been found a specific cell surface receptor for Abeta peptite, thus eliciting neuronal cell perturbation. The active participation of RAGE in the pathogenesis of AD has also been confirmed in RAGE-overexpressed transgenic mice. Moreover, we have recently found that glyceraldehyde-derived AGEs, one of the representative ligands for RAGE, exerted cytopathic effects on cultured neuronal cells and that neurotoxic effect of diabetic serum was completely blocked by neutralizing antibodies against glyceraldehydes-derived AGEs. These observations led us to hypothesize that serum or cerebrospinal fluid (CSF) levels of glyceraldehyde-derived AGEs could become a promising biomarker for early detection of AD. We also would like to propose the possible ways of testing our hypothesis. Are the concentrations of glyceraldehyde-derived AGEs in serum or CSF elevated early in the course of dementia? Are these levels correlated with disease severity and progression, especially in patients with diabetes? These clinical studies clarify whether use of serum or CSF levels of glyceraldehyde-derived AGEs as a biomarker for AD might enable more effective diagnosis and treatment of patients with this devastating disorder.