Complex phenotypic analysis to identify genes which contribute to or modify the development of Alzheimer's disease.
Late-onset Alzheimer's disease (LOAD) is a heritable disorder. It is invariably characterised by a decline in cognitive abilities, however, marked variation in behavioural symptoms and age at onset are observed between sufferers. This clinical heterogeneity may be genetically modified, hence, may provide a productive avenue of exploration for those seeking to unravel the genetic aetiology of LOAD. This thesis employed a sequential three stage approach to search for loci implicated in the development of genetically influenced features of the disease. Behavioural symptoms in 1,120 unrelated individuals with LOAD were assessed using the Neuropsychiatric Inventory. The 12 symptom domain scores were subjected to principal components analysis. Three interpretable components were identified, comprising: "frontal lobe dysfunction", "psychosis" and "mood". These components remained stable when taking account of disease severity. The familiality of clinical variation was assessed. Affected siblings from 388 families were characterised in terms of aggression, psychosis and mood disturbances. Age at onset data were available for affected siblings from 458 families. Familial clustering was found for age at onset, psychosis, aggression and mild depression, with the strongest evidence noted for age at onset and psychosis. Major depression and a combined phenotype of depression with anxiety showed limited evidence of familial aggregation. Covariate linkage analysis was employed to search for loci which may influence clinical variation in LOAD. This included a sample of 513 affected relative pairs. Increases in LOD were observed with age at onset (chromosome 1, 2,12,19 and 21), aggression (chromosome 9), psychosis (chromosome 7 and 15) and minor depression (chromosome 21). Understanding factors associated with behavioural symptoms and age at disease onset may lead to the achievable goal of disease modification. These findings support the hypothesis that clinical variation in AD is genetically modified, setting the stage for future linkage and association studies.
Online Research @ Cardiff (http://orca.cf.ac.uk/54267/1/U584070.pdf)