APOE loss‐of‐function variants: Compatible with longevity and associated with resistance to Alzheimer’s Disease pathology

Background The ε4 variant of APOE is the strongest genetic risk factor for late‐onset Alzheimer’s disease (AD). It remains unclear if the ε4 protein increases AD risk through a gain of abnormal function, or whether it functions less well than the ε3 protein. Framed bluntly: Would knocking down ε4 in an ε4 carrier increase or decrease AD risk? Method We searched whole‐exomes and whole‐genomes from ∼47,000 older controls and AD cases in the Alzheimer’s Disease Sequencing Project (ADSP) looking for early loss‐of‐function variants on APOE. Five of 26,605 older controls carried a loss‐of‐function variant while only one of 20,856 AD cases did (Figure.1A/B). We reasoned that if ε4 does not function as well as ε3, then loss‐of‐function of either ε3 or ε4 should have an ε4‐like impact on risk. However, if ε4 is inherently detrimental, loss‐of‐function of ε4, specifically, should have a protective effect. Result Loss of ε3 was seen in two cognitively healthy ε3 homozygotes, last assessed at ages 82 and 88. This suggests that the ε3/null effective genotype is not a strong driver of AD risk. Subject 6‐nominally ε3/ε4 but effectively ε4/null ‐developed symptoms at 75 and died at 87 with autopsy‐confirmed AD, with an age‐at‐onset similar to typical ε3/ε4 (μ = 73.5y), but later typical than ε4/ε4 (μ = 69.7y). This suggests that the loss of ε3 does not result in an ε4‐like phenotype, as seen with Subjects 3 and 5. The most illustrative phenotypes are seen in Subjects 1 and 2 who are nominally ε3/ε4 but effectively ε3/null. Subject 2 was cognitively healthy at 79, with normal levels of amyloid and tau in cerebrospinal fluid at 76, while 2/3rds of ε3/ε4 are amyloid‐positive by 75. Subject 1 was assessed as cognitively normal at 90, died the same year, and had no amyloid pathology, and no amyloid deposits in the cerebral blood vessels, making him a prominent outlier compared to 1700 ε3/ε4 individuals in NACC (Figure.1C). Conclusion Taken together, these results provide the strongest human genetics evidence yet available to support the hypothesis that ε4 drives AD risk through a gain of abnormal function and that reducing ε4 should be expected to reduce AD risk.