A genome-wide meta-analysis of migraine in more than 59,000 cases and 313,000 controls reveals 29 new loci, increasing the total number of risk loci to 42. P. Gormley1,2, V. Anttila2,3, M. Muona4, A. Palotie1,2,4 on behalf of the International Headache Genetics Consortium (IHGC) 1) Psychiatric and Neurodevelopmental Genetics Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA; 2) Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA, USA; 3) Analytical and Translational Genetics Unit, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA; 4) Institute for Molecular Medicine Finland (FIMM), University of Helsinki, Helsinki, Finland.
Migraine is a complex disorder of neurogenic and/or neurovascular origin affecting around 14% of the population worldwide. It is characterized by chronic, recurrent headaches, often with related symptoms including nausea, photophobia, and phonophobia. One third of cases experience a sensory aura just prior to the headache, usually a visual phenomenon known as scintillating scotoma but other forms also exist. Family and twin studies estimate high heritability for migraine, pointing to a strong genetic component of the disease. Despite this, relatively little is known about the molecular mechanisms that contribute to migraine pathophysiology. Understanding has been limited because, to date, only 13 genome-wide significant risk loci have been identified for common migraine (Nature Genetics, 2013, 45:912). In this study we aimed to increase our understanding by carrying out a large-scale meta-analysis of migraine, consisting of 59,043 cases and 313,781 controls collected from 6 tertiary headache clinics and 26 population-based cohorts from multiple locations throughout the world. We performed a fixed-effects meta-analysis of genome-wide association data in 32 cohorts from the International Headache Genetics Consortium (IHGC) after imputing missing genotypes in each dataset to a common 1000 Genomes reference panel (March 2012 release, Phase I, v3). Our primary analysis investigated all general forms of common migraine, including self-reported cases and those diagnosed according to International Headache Society (IHS) diagnostic criteria. For the common migraine phenotypes we now implicate 42 risk loci in total, replicating 10 of the 13 previously reported associations and detect 29 new loci at levels of genome-wide significance. In follow-up analyses we investigated migraine sub-types and found 7 out of the 42 loci could be associated specifically to migraine without aura. No loci were robustly associated to migraine with aura, reinforcing the suggestion that rare variation plays more of a role in this form of the disease. In summary, we report 29 new risk loci and increase the total number of genome-wide significant loci to 42 for the common forms of migraine. Preliminary results point to a role for many molecular processes, including cell-signaling and ion-channel dysfunction, but also implicate several genes with previous vascular-disease associations, thus supporting a role for both neuropathic and neurovascular mechanisms in migraine pathophysiology.
You may contact the first author (during and after the meeting) at