Abstract
OBJECTIVE: To examine the association of the APOE ε2ε4 genotype with incident Alzheimer disease (AD), mild cognitive impairment (MCI), cognitive decline, and AD pathology in older adults. METHODS: We used data from 2,151 older adults of European ancestry who were free of dementia at baseline and underwent structured annual clinical evaluation in a longitudinal study for incident AD and MCI, and cognitive decline. Postmortem examination in decedents documented pathologic AD and quantified β-amyloid and neurofibrillary tangles. Participants were stratified into 4 groups based on APOE genotyping: ε2ε4, ε4 (ε4ε4, ε4ε3), ε2 (ε2ε2, ε2ε3), with ε3ε3 carriers serving as the reference group. We used Cox proportional hazards models to examine the association of APOE genotype with incident AD and MCI. Linear mixed-effect models were used to examine the association with cognitive decline. Logistic and linear regression models were used to examine AD pathology. All the models controlled for age, sex, and education. RESULTS: Of the 2,151 participants included in this study, ε2ε4 accounted for 2.1%, ε3/4 and 4/4 21.8%, ε2/3 and 2/2 14.0%, and ε3ε3 62.1%. We did not observe a difference in the risk of AD for ε2ε4 compared to ε3ε3. In cases without cognitive impairment at baseline, ε2ε4 carriers had an increased risk of incident MCI (hazard ratio 2.13, 95% confidence interval 1.34-3.39, p = 0.002) and a faster rate of cognitive decline (estimate -0.047, SE 0.018, p = 0.008) compared to ε3ε3 carriers. In decedents (n = 1,100), ε2ε4 showed a 3-fold increased odds of pathologic AD and a higher β-amyloid load than ε3ε3. CONCLUSION: APOE ε2ε4 genotype in older adults is associated with risk of MCI, cognitive decline, and a greater burden of AD pathology, especially β-amyloid.