Progress in Polygenic Composite Scores in Alzheimer's and Other Complex Diseases

Abstract

Advances in high-throughput genotyping and next-generation sequencing (NGS) coupled with larger sample sizes brings the realization of precision medicine closer than ever. Polygenic approaches incorporating the aggregate influence of multiple genetic variants can contribute to a better understanding of the genetic architecture of many complex diseases and facilitate patient stratification. This review addresses polygenic concepts, methodological developments, hypotheses, and key issues in study design. Polygenic risk scores (PRSs) have been applied to many complex diseases and here we focus on Alzheimer's disease (AD) as a primary exemplar. This review was designed to serve as a starting point for investigators wishing to use PRSs in their research and those interested in enhancing clinical study designs through enrichment strategies.

Keywords: Alzheimer’s disease; heritability; linkage disequilibrium; polygenic hazard score; polygenic risk score; statistical power.

Figure 1.

Key Figure. Polygenic risk score calculation. Step1) SNP selection (with or without filtering), Step 2) weight calculation: Candidate SNPs can be assigned a weight of 1 (PRS is a simple sum of SNP alleles) or weighed using existing GWAS-derived effect sizes. Alternatively, one can re-calculate the SNP weights (re-weighting), that is, estimate new weights by including the SNPs in a regression model (e.g., Cox). Penalization techniques (either frequentist e.g., Lasso or Bayesian e.g., LDpred) can also be used for re-weighting. These methods can achieve SNP selection and weight estimation simultaneously, by setting some of the SNP weights to zero. Penalization methods can be either applied on the filtered or on the original SNP list.

Figure 2.

Factors affecting PRS accuracy. Disease related factors (e.g., heritability, functional annotation, LD structure, and number of informative SNPs) as well as study design aspects (e.g., sample size, p-value threshold for SNP selection, and sampling variability), affect the power and performance of PRS. Depending on the hypothesis tested and the disease characteristics, improved PRS performance is possible via the appropriate sample size, SNP selection threshold and LD control.