Genome‐wide association study of plasma apolipoprotein E levels and risk of dementia in older adults

Abstract

AbstractBackgroundThe APOE 2/3/4 polymorphism is the greatest risk factor for Alzheimer’s disease (AD). This polymorphism is also associated with variation in plasma ApoE levels and explains about 20‐25% of its genetic variance; while APOE*4 lowers, APOE*2 increases ApoE levels. Lower plasma ApoE levels have also been reported to be a risk factor of future AD and all dementia, independent of the APOE 2/3/4 polymorphism (Ann Neurol 2015;77:301). To our knowledge, no genome‐wide association study (GWAS) has been reported on plasma ApoE levels. This study was aimed to identify novel genetic factors affecting plasma ApoE levels as well as to confirm if baseline plasma ApoE levels predict future AD dementia in an older and longitudinally followed cohort of the Ginkgo Evaluation of Memory (GEM) study.MethodBaseline plasma ApoE levels were measured using an immunoturbidimetric assay in 3,031 older subjects (aged 72 to 96 years; 94.1% White). Genome‐wide genotype was performed on the available 2,737 DNA samples (96% White). Due to the small number in other ethnicities, we only included the 2,852 White subjects (2,412 cognitively normal, 440 with incident AD dementia) for plasma ApoE analysis. GWAS analysis on plasma ApoE levels was performed on 2,580 White subjects where both genotype and ApoE level data were available.ResultAs expected, the risk for AD increased from E2/2 through to E4/4 genotypes (p for trend =1.7E‐10). There was no significant difference in plasma ApoE levels between cognitively normal and AD dementia subjects (p=0.55). GWAS analysis revealed the expected associations of APOE*2 (β= 1.12; p= 3.57E‐79) and APOE*4 (β= ‐0.35; p= 2.21E‐11) with plasma ApoE levels. In addition, we observed 5 novel genome‐wide significant associations on chromosomes 1 (β= 0.921; p= 5.36E‐10), 4 (β= 0.796; p= 4.31E‐08;), 7 (β= 0.722; p= 9.64E‐09), 11 (β= 0.268; p= 2.58E‐08), and 20 (β= 1.379; p=8.25E‐09).ConclusionWe have identified multiple novel associations affecting plasma ApoE levels in addition to the established APOE association. However, we could not replicate the reported association of plasma ApoE levels with future AD, probably because of the older nature of our cohort.