Polygenic risk score in postmortem diagnosed sporadic early-onset Alzheimer's disease

Abstract

Sporadic early-onset Alzheimer's disease (sEOAD) exhibits the symptoms of late-onset Alzheimer's disease but lacks the familial aspect of the early-onset familial form. The genetics of Alzheimer's disease (AD) identifies APOEε4 to be the greatest risk factor; however, it is a complex disease involving both environmental risk factors and multiple genetic loci. Polygenic risk scores (PRSs) accumulate the total risk of a phenotype in an individual based on variants present in their genome. We determined whether sEOAD cases had a higher PRS compared to controls. A cohort of sEOAD cases was genotyped on the NeuroX array, and PRSs were generated using PRSice. The target data set consisted of 408 sEOAD cases and 436 controls. The base data set was collated by the International Genomics of Alzheimer's Project consortium, with association data from 17,008 late-onset Alzheimer's disease cases and 37,154 controls, which can be used for identifying sEOAD cases due to having shared phenotype. PRSs were generated using all common single nucleotide polymorphisms between the base and target data set, PRS were also generated using only single nucleotide polymorphisms within a 500 kb region surrounding the APOE gene. Sex and number of APOE ε2 or ε4 alleles were used as variables for logistic regression and combined with PRS. The results show that PRS is higher on average in sEOAD cases than controls, although there is still overlap among the whole cohort. Predictive ability of identifying cases and controls using PRSice was calculated with 72.9% accuracy, greater than the APOE locus alone (65.2%). Predictive ability was further improved with logistic regression, identifying cases and controls with 75.5% accuracy.

Keywords: Genotyping; NeuroChip; NeuroX; Polygenic risk score (PRS); Sporadic early-onset Alzheimer's disease (sEOAD).

Fig. 1

Proportion of diagnosed sEOAD cases and controls at each decile determined by a range of score. The figure breaks down the range of PRS into deciles. The range of scores which make up each decile are depicted, as well as the number of cases and controls, and the percentage of individuals which fall into each decile. Controls are right skewed, whereas cases demonstrate left skewness. These figures were produced from PRS of 408 sEOAD cases and 436 controls. The embedded table lists the decile ranges with the number of cases and controls in each decile along with the proportion of the cohort which make up each decile. Abbreviations: PRS, polygenic risk score; sEOAD, sporadic early-onset Alzheimer’s disease.

Fig. 2

Distribution of polygenic risk score among sEOAD cases and controls with average scores at each decile. The range of PRS obtained for cases and controls is distributed into deciles. The range of coverage of each decile is shown in the bar plot together with the proportion of cases and controls which make up each decile. The average scores for all cases and controls are indicated by the thick bar, while the short horizontal bars show PRS for each individual. Average scores at each decile are indicated as hollow circles for cases and filled circles for controls. Abbreviations: PRS, polygenic risk score; sEOAD, sporadic early-onset Alzheimer’s disease.

Fig. 3

Results of logistic regression with an area under the receiver operating characteristic curve (AUC) for alternative risk-scoring models in sEOAD. For this analysis, the APOE locus was defined as a 500 kb region surrounding the APOE gene, and the scores produced by PRSice for this model are based on the SNPs within that region; PRS represents the score produced for all SNPs present on both the NeuroX array and in the base data set. The relevant variables included sex together with the number of APOE ε2 allele and/or APOE ε4 allele. As shown in the table, a nonsignificant Hosmer-Lemeshow p-value suggests that the model is suitable for using as a predictive tool. Nagelkerke’s R2 can also be used to identify the best model for risk prediction; the higher the value of R2 the greater the predictive accuracy of each model. This approach identified sex, ε2, ε4 + PRS as the best model for calculating risk in our sEOAD cohort as the largest AUC value is produced from the combination of variables. Abbreviations: PRS, polygenic risk score; sEOAD, sporadic early-onset Alzheimer’s disease; SNPs, single nucleotide polymorphisms.