Genome-Wide Association Study of Incident Dementia in a Community-Based Sample of Older Subjects

Background: Alzheimer's disease (AD) is a complex disease influenced by the environment and genetics; however, much of the genetic component remains unaccounted for. Objective: The purpose of this work was to use genome-wide association analyses to detect genetic associations with incident AD in a sample of older adults aged 75 and above. Methods: We performed a genome-wide association study (GWAS) on genome-wide genotyped and imputed data (14,072,053 variants) on the Gingko Evaluation of Memory (GEM) study sample consisting of 424 incident dementia (mean age = 84.46 +/- 3.91) and 2,206 non-demented (mean age = 84.55 +/- 3.23) subjects. Results: The established association of APOE*4 carriers with AD was confirmed in this community-based sample of older subjects (odds ratio (OR) = 2.22; p = 9.36E-14) and was stronger in females (OR = 2.72; p = 1.74E-10) than in males (OR = 1.88; p = 2.43E-05). We observed a novel genome-wide significant (GWS) locus on chromosome 12 near ncRNA LOC105369711/rs148377161 (OR = 3.31; p = 1.66E-08). In addition, sex-stratified analyses identified two novel associations in males: one near ncRNA LOC729987/rs140076909 on chromosome 1 (OR = 4.51; p = 3.72E-08) and the other approaching GWS near ncRNA LOC105375138/rs117803234 on chromosome 7 (OR = 3.76; p = 6.93E-08). Conclusion: The use of community-based samples of older individuals and incident dementia as a phenotype may be a helpful approach for the identification of novel genes for AD, which may not be detected in standard case-control studies. Replication of these signals and further studies of these regions and genes will help to provide a clearer picture for their role in AD.