Objectives: Declines in brain glucose metabolism have been described early in Alzheimer's disease (AD), and there is evidence that a genetic predisposition to AD contributes to accelerate this process. The epsilon 4 (e4) allele of the apolipoprotein E (ApoE) gene has been implicated as a major risk factor in this process. The aim of this FDG-PET study was to assess the ApoE e4 dose related effect on regional cerebral glucose metabolism (METglc) in clinical AD patients, with statistical voxel based methods.
Methods: Eighty six consecutive mild to moderate AD patients included in the Network for Efficiency and Standardisation of Dementia Diagnosis database underwent FDG-PET scans at rest. PCR was used to determine the ApoE genotype. Patients were grouped as e4 non-carriers (n = 46), e3/e4 (n = 27) and e4/e4 (n = 13) carriers. A voxel-based mapping program was used to compare each AD subgroup with a database of 35 sex and age matched controls (p<0.001, corrected for cluster extent) and also to compare between the subgroups (p<0.001, uncorrected).
Results: No difference was found as to age at examination, age at onset, sex, disease duration, educational level, or severity of dementia between AD subgroups. Compared with controls, all AD subgroups had equivalent METglc reductions in the precuneus, posterior cingulate, parietotemporal, and frontal regions. Direct comparisons between AD subgroups indicated that patients with at least one e4 allele had METglc reductions within additional associative and limbic areas compared with e4 non-carriers.
Conclusions: The present FDG-PET study showed different metabolic phenotypes related to the ApoE genotype in clinical AD patients, as revealed with voxel based statistical methods. The results suggest a generalised disorder in e4 carriers impairing metabolism globally, in addition to the more localised changes typical of AD patients.