{
  "abstract": "Introduction Transcriptional risk scores (TRS), an integrated score of genomic susceptibility variants and gene expression data (eQTL), have been shown in selected diseases to offer clinically useful prognostic information, but have never been applied in multiple sclerosis (MS).Methods We integrated MS susceptibility genetic variants with eQTL data to compute CNS and immune specific TRS. We also calculated a “classical” polygenic risk score (PRS) for MS susceptibility. We compared those scores between groups of deeply phenotyped treatment naive MS patients longtiduinally followed since 1985.Results MS susceptibility variants are associated with expression of 240 genes, of which 109 immune/blood and 88 CNS were specific to an organ system. In a well validated MS cohort (n=1,077), immune TRS was found to distinguish patients with a short (<2 years) versus long (>/=2 years) interval between 1st and 2nd relapse (p=0.03). Last-recorded Age-Related Multiple Sclerosis Severity Score was also associated with immune TRS, (p=0.03) while no differences were found using traditional PRS.Conclusions This data confirms that MS susceptibility variants alter gene expression in both the CNS and immune system. TRS may be more helpful than PRS in identifying MS phenotypes. This provides opportunities to further explore prognostic markers and druggable pathways.mattupcott37@gmail.com",
  "authors": [
    {
      "affiliations": [
        "Division of Psychological Medicine and Clinical Neuroscience, School of Medicine, Cardiff University"
      ],
      "name": "Upcott Mae"
    },
    {
      "affiliations": [
        "Department of Neurology, University Hospital of Wales, Cardiff, UK",
        "Division of Psychological Medicine and Clinical Neuroscience, School of Medicine, Cardiff University"
      ],
      "name": "Tallantyre Emma C"
    },
    {
      "affiliations": [
        "Department of Neurology, University Hospital of Wales, Cardiff, UK",
        "Division of Psychological Medicine and Clinical Neuroscience, School of Medicine, Cardiff University"
      ],
      "name": "Robertson Neil P"
    },
    {
      "affiliations": [
        "School of Medicine, University of Nottingham, UK",
        "Department of Neurology, Queen’s Medical Centre, Nottingham University Hospital NHS Trust"
      ],
      "name": "Kreft Karim L"
    }
  ],
  "title": "19 Transcriptional risk scores identify subgroups of multiple sclerosis patients more accurately than polygenic risk scores",
  "uid": "1162d059-d2be-573a-a5eb-6055badd442d"
}