Premium
APOE ‐stratified, genome‐wide, gene‐based analysis of episodic memory trajectories in a multi‐ethnic sample of 24,769 elderly
Author(s) -
Gao Yizhe,
Vardarajan Badri N.,
ReyesDumeyer Dolly,
Sariya Sanjeev,
Mayeux Richard,
Tosto Giuseppe,
Barral Sandra
Publication year - 2020
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.045997
Subject(s) - episodic memory , genome wide association study , genetic association , longitudinal study , dementia , psychology , cognition , genetics , biology , medicine , single nucleotide polymorphism , gene , genotype , disease , psychiatry , pathology
Background Understanding genetic modulation of episodic memory will yield insight into the biological mechanisms underlying normal cognitive aging and dementia. Genetic studies based on longitudinal measures of episodic memory are scarce, predominantly focused on candidate genes, and have interrogated common genetic variation. We aimed to identify common and rare genetic variation that may be responsible for uncharacterized genetic risk underlying individual differences in longitudinal performance on memory. Method A sample of 24,769 elderly individuals was ascertained from eight independent cohorts (ADGC_NACC, NIA‐LOAD, ROSMAP, WHICAP, LLFS, CHAP, ADNI and UK Biobank). Data on study participants included socio‐demographic variables, longitudinal scores on episodic memory (2 to 15 follow‐up visits), clinical diagnosis of cognitive impairment and genome‐wide imputed genotyped data using the haplotype reference consortium for accurate imputation of low‐frequency genetic variants. The Latent Class Mixed Model was used to derive episodic memory trajectories within each of the study cohorts. The slope of residualized episodic memory scores was used as outcome in an APOE stratified (non‐ APOE stratified, APOE _ε4 and APOE _nonε4 sub‐samples) genome‐wide gene‐based analyses. Common (minimum allele frequency, MAF, between 5%‐10%), rare (1% < MAF ≤ 5%) and ultra‐rare variants (≤1% MAF) were analyzed. Association results from each study cohort within each sub‐sample were meta‐analyzed. Result Consistent with previous studies, the majority of the study participants maintain their memory performance over time (ranging from 51% to 98%). The strongest association signal for the episodic memory trajectories (p = 4.8 × 10 −8 ) was achieved at chromosomal region 6p22 among in the APOE _nonε4 sub‐sample. Meta‐analysis results in the non‐ APOE stratified and APOE _ε4 sub‐samples also yielded several chromosomal regions associated with memory performance over time, although significance level was diminished (p∼ 10 −6 ). Conclusion To our knowledge, this is the largest gene‐based GWAS meta‐analysis of high quality imputed common, rare and ultra‐rare variants influence on episodic memory performance over time. Identifying genes associated with progression of episodic memory performance over time in older adulthood provides the possibility of identifying at‐risk population subgroups that can benefit from possible interventions to enhance cognitive function.