However, leaving aside issues of classification, the principal aim of the present study was to attempt to define certain factors that may be driving, or determining, such phenotypic variations. Comparisons across subtypes of demographic and disease-specific information (age of onset, age of death, disease duration and brain weight, and presence of family history) failed to show significant differences between the pathological subgroups. The fact that one particular phenotype

was not associated with increasing age at onset, or duration of disease, compared with (any of) the others, lends support to the argument that the phenotypes are not a continuum of one another but instead exist as separate entities. Nevertheless, gender ratios did appear BAY 57-1293 nmr to differ between the group 1 and group 2 phenotype, in that women were over-represented (65%) in group 1 (with less extensive CAA) and were under-represented (43%) in group 2 (where CAA was on the whole more severe). One possible reason for this could be that group 1 cases were older (at death) than those in group 2, and as such would reflect relative longevities of male and women – it being well known that older subjects with AD are more likely to be female. However, as mentioned above there were no significant differences in the age structure of

the Groups. Another reason might relate to Fenbendazole the suggestion [32] that oestrogen has a neuroprotective effect and therefore might JQ1 cost afford some protection against more widespread CAA. However, another study [33] suggested that oestrogen fails to protect endothelial cells in the same way it protects neurones, glial cells, and smooth muscle cells, and this might therefore facilitate the progression of CAA. The present study has heuristic value in that it proposes that four separate patterns of Aβ deposition with regard to SP and CAA exist. Such a classification has not been done previously. For many years, the

diagnostic focus of AD has been given to the presence of NFT (Braak and Braak Staging) or neuritic plaques (SP) (CERAD), or both of these pathological entities [12]. Building more subtle CAA classifications into pathological diagnostic criteria may have value in assigning diagnostic accuracy, particularly in cases where SP density may be low, and may not meet pathological ‘thresholds’ under current criteria. However, beyond this, identification of AD patients with severe CAA may have value in predicting those cases at risk of cerebral haemorrhage [16], or defining patients suitable for immunotherapy. In present trials, it has been shown that while plaque Aβ load can be drastically reduced following immunotherapy, this seems to be at the expense of increased CAA [34].