Lack of reliable evidence for a distinctive ε4-related cognitive phenotype that is independent from clinical diagnostic status: findings from the Australian Imaging, Biomarkers and Lifestyle Study

Individuals who carry the apolipoprotein E ε4 polymorphism have an increased risk of late-onset Alzheimer's disease. However, because possession of the ε4 allele confers an increased risk for the diagnosis of dementia, it has proven problematic in older individuals to dissociate the influence of ε4 on cognitive capacity per se as distinct from its influence on clinical diagnostic status. We report a statistical approach that attempts to partial out the influence of diagnostic group membership (Alzheimer's disease, mild cognitive impairment, healthy control) from the influence of apolipoprotein ε4 genetic status on cognitive functioning. The cognitive phenotype hypothesis predicts that ε4-positive individuals will show cognitive deficits (relative to matched ε4-negative individuals) independent of the development of Alzheimer's disease. By contrast, the prodromal/preclinical Alzheimer's disease hypothesis proposes that the effect of apolipoprotein E status on cognitive performance is a function of the increased risk of dementia in individuals with the ε4 allele. We evaluated these hypotheses in the Australian Imaging, Biomarkers and Lifestyle cohort (n = 1112). We first determined whether previously reported findings concerning ε4 status and age-related neuropsychological performance could be explained by the inadvertent recruitment of people with mild cognitive impairment into the healthy control group. We then tested each diagnostic group in isolation to identify any neuropsychological patterns that may be attributed to the ε4 allele. Finally, as interactions between the ε4 allele and age have previously been reported in cognitive functioning within healthy elderly populations, we attempted to determine whether the inclusion of mild cognitively impaired individuals in the sample may drive this relationship. An extensive battery of standardized, well-validated neuropsychological tasks was administered to a final sample of 764 healthy control subjects, 131 individuals with mild cognitive impairment and 168 individuals with Alzheimer's disease. The effect of the ε4 allele on cognitive performance was assessed using a statistical mediation analysis and supplemented with Bayesian methods to address a number of the limitations associated with Fisherian/Neyman-Pearsonian significance testing. Our findings support the prodromal/preclinical Alzheimer's disease hypothesis and do not support the concept of a distinctive ε4-related cognitive phenotype.