The role of cortical connectivity in Alzheimer's disease pathogenesis: a review and model system

Here we review current evidence in support of the cortical disconnection/cortical connectivity model of Alzheimer disease (AD) pathogenesis, a model which predicts that one of the first events in AD is damage to the entorhinal cortex and/or subiculum resulting in the disconnection of the hippocampal formation and neocortex, and the subsequent progression of the disease in a stepwise fashion along cortico-cortical connections. Much of the evidence for this model has been obtained from studies involving the limbic system where investigators have demonstrated a precise correspondence between established patterns of connectivity and the degenerative changes associated with AD. In addition, some studies of the distribution of neuritic plaques (NP) and neuro-fibrillary tangles (NFT) in the neocortex and subcortical structures have yielded corroborative data. The validity of the cortical disconnection/connectivity model in the neocortex remains to be established or refuted. We propose that testing of this model can be accomplished with systematic studies of the laminar and regional distribution of NP and NFT in a series of sequentially interconnected cytoarchitectural regions that also form part of two functional hierarchies--the paralimbic and occipitotemporal visual systems. To adequately control for variation between brains affected by AD, it is imperative that such studies be conducted in a large but varied population of AD cases exhibiting differences in several variables, including clinical and/or neuropathological severity of the disease, temporal duration of the disease, and clinical/neuropsychological profile. We believe that further understanding of the relationship between characteristic AD pathology and intrinsic anatomico-functional circuits will contribute not only to our comprehension of AD pathogenesis but also to our general knowledge of the human brain.