First Genome-Wide Analysis in Pediatric Multiple Sclerosis (MS) Confirms a Role for Adult MS Risk Variants and Reveals New Candidates (S29.001)

Abstract

Objective: We conducted the first GWAS in pediatric MS and tested previously established adult MS-associated candidate genes for association. Background: Up to 10[percnt] of MS patients have symptoms before age 18 (i.e. pediatric-onset MS). Similar to adults, both genetic and environmental risk factors have been implicated. In addition to HLA-DRB1*15:01, results from GWAS in adults with MS have identified ~110 non-MHC MS variants. Methods: All subjects had whole genome profiles (Illumina or Affymetrix chips). Genotype imputation and very rigorous quality control criteria were applied. SHAPEIT2 for phasing was followed by IMPUTE2 (1000 Genomes Phase 3 integrated variant set, October, 2014). Each individual was characterized for genetic ancestry using MDS (PLINK) and STRUCTURE analysis; cases and controls were closely matched on genetic ancestry. Results: We assembled samples from 732 MS cases with onset <18 years (70[percnt] female, mean age of onset 13.9 (+/- 3.5) years, 46[percnt] non-white) and 27,187 controls for genetic studies. Results from whites-only analysis (394 cases, 10,875 controls) showed HLA-DRB1*1501 was strongly associated with MS (OR=2.62, 95[percnt] CI: 2.23-3.07, p=2.1x10-32). Further, 36 of 110 non-MHC SNPs (33[percnt]) were significant (p<0.05); rs2744148 (chr 16), rs3007421 (chr 1) and rs1800693 (chr 12) demonstrated strongest evidence of association with pediatric MS, with larger effects compared to adults: OR=2.94, 95[percnt] CI: 2.41-3.57, p=5.2 x10-27; OR=2.46, 95[percnt] CI: 2.02-2.98, p=3.0 x 10-19; and OR=1.74, 95[percnt] CI: 1.51-2.01, p=4.7 x 10-14, respectively. A weighted genetic risk score (wGRS, 110 variants) was strongly associated with pediatric MS (2.0 x 10-16). Results from GWAS showed at least three genomic regions harbored SNPs with p<10-7. Fine mapping and trans-ethnic meta-analyses are underway. Conclusions: Variants within genes involved in adult MS are also risk factors in children and suggest that similar biological processes are present in both groups. Genetic effects may be stronger in pediatric MS overall.