More than 30 independent single-nucleotide polymorphisms (SNPs) have been associated with Alzheimer's disease (AD) risk by genome-wide association studies (GWAS) in European. We aimed to confirm these SNPs in Chinese Han and investigate the utility of these genetic markers. We randomly divided 459 sporadic AD (SAD) patients and 751 cognitively normal controls into two sets (discovery and testing). Thirty-three SAD risk-associated SNPs were firstly tested in the discovery set. Significant SNPs were used to calculate genetic risk score (GRS) in the testing set. Predictive performance of GRS was evaluated using the area under the receiver operating characteristic curve (AUC). In the discovery set, 6 SNPs were confirmed (P = 7.87 x 10(-11)~0.048), including rs9349407 in CD2AP, rs11218343 in SORL1, rs17125944 in FERMT2, rs6859 in PVRL2, rs157580 and rs2075650 in TOMM40. The first three SNPs were associated with SAD risk independent of APOE genotypes. GRS based on these three SNPs were significantly associated with SAD risk in the independent testing set (P = 0.002). The AUC for discriminating cases from controls was 0.58 for GRS, 0.60 for APOE, and 0.64 for GRS and APOE. Our data demonstrated that GRS based on AD risk-associated SNPs may supplement APOE for better assessing individual risk for AD in Chinese.
Keywords: Alzheimer’s disease; Gerotarget; association; genetic risk score; risk prediction; single nucleotide polymorphism.