{
  "abstract": "Introduction Insomnia may be a risk factor for breast cancer, but existing evidence is inconsistent, particularly in non-European populations. This study explores the association and genetic underpinnings of clinical insomnia and breast cancer across multiple ancestral populations within the ‘All of Us’ Research Program.Methods We used data from ‘All of Us’, a diverse longitudinal US cohort study with genomic data linked to electronic health records. Insomnia was defined based on clinical diagnoses and/or medication use. We examined associations between insomnia and breast cancer incidence among women of European (n=98,544), African American (n=26,456) and Hispanic ancestry (n=22,058) using multivariable Cox regression. Further, genome-wide association studies (GWAS) for insomnia were conducted in each ancestral population, through single variant testing of the whole genome sequencing (short read) data.Results Fully adjusted Cox regression models showed an inverse relationship between clinical insomnia and breast cancer in European ancestry (HR=0.82; 95%CI=0.74, 0.92) and African ancestry groups (HR=0.71; 95%CI=0.54,0.94), and a weaker association in Hispanics (HR=0.85; 95%CI=0.61;1.20) ( figure 1). GWAS confirmed known variants and identified novel ones (figure 2). Variants in MEIS1, a gene linked to restless leg syndrome, were associated with clinical insomnia in European ancestry in ‘All of Us’ (P value<1x10-9), but not in other ancestry groups (P value>5x10-8). A novel variant in WSB1 was identified in European (P value= 5.05x10-41), African (P value=2.92x10-81) and Hispanic ancestry groups (P value=2.17x10-11), suggesting evidence for conservation across ancestries.Discussion Consistent but unexpected protective associations between clinical insomnia and breast cancer were observed across ancestries. A genetic variant in WSB1 was strongly associated with clinical insomnia across ancestries. WSB1 is a regulator of metastatic disease in hormone receptor-negative breast cancer. Ongoing work aims to better understand the role of this novel variant in insomnia and breast cancer, and to clarify causal mechanisms.Abstract O23 Figure 1Forest plot of multivariable estimates for association between insomnia and breast cancer risk among different ancestral groups in All of UsAbstract O23 Figure 2Manhattan plot for clinical insomnia genome-wide association study among different ancestral groups in All of Us",
  "authors": [
    {
      "affiliations": [
        "University of Bristol, Bristol, UK",
        "University of Oxford, Oxford, UK"
      ],
      "name": "Rebecca Richmond"
    },
    {
      "affiliations": [
        "University of Bristol, Bristol, UK"
      ],
      "name": "Bryony Hayes"
    },
    {
      "affiliations": [
        "University of Exeter, Exeter, UK"
      ],
      "name": "Alok Singh"
    },
    {
      "affiliations": [
        "University of Exeter, Exeter, UK"
      ],
      "name": "Marina Vabistsevits"
    },
    {
      "affiliations": [
        "University of Oxford, Oxford, UK"
      ],
      "name": "David Ray"
    },
    {
      "affiliations": [
        "University of Exeter, Exeter, UK"
      ],
      "name": "Michael Weedon"
    }
  ],
  "title": "O23 Assessing the role of insomnia in breast cancer risk across multiple ancestries within the ‘all of us’ research program",
  "uid": "598f3ac3-a9e7-5864-863e-b14eab527573"
}