GENOME-WIDE ANALYSIS OF 102,084 MIGRAINE CASES IDENTIFIES 123 RISK LOCI AND SUBTYPE-SPECIFIC RISK ALLELES 181 8 Even though we observed links between our new risk loci and known target genes of effective migraine drugs, the accurate gene prioritization at risk loci remains challenging. First, robust finemapping would require accurate LD information,36 which is typically lacking in meta-analyses and further distorted from reference panels by variation in effective sample size across variants. Second, computational approaches to gene prioritization require further methodological work104 and extension to additional sources of functional data in order to provide more robust and comprehensive gene prioritization results. Another limitation of our study is that a large proportion of migraine diagnoses are self-reported. Therefore, we cannot rule out misdiagnosis, such as, e.g., tension headache being reported as migraine, which could overemphasize genetic factors related to general pain mechanisms and not migraine per se. Regardless, the high genetic correlation that we observed supports a strong phenotypic concordance between the study collections that also included deeply phenotyped clinical cohorts from headache specialist centers, which were instrumental for the migraine subtype analyses. While the subtype data provided convincing evidence of both loci with genetic differences and other loci with genetic overlap between subtypes, larger samples are still needed to achieve a more accurate picture of the similarities and differences in genetic architecture behind the subtypes. To conclude, we report the largest GWAS meta-analysis of migraine to date, detecting 123 risk loci. We demonstrated that both vascular and central nervous systems are involved in migraine pathophysiology, supporting the notion that migraine is a neurovascular disease. Our subtype analysis of migraine with aura and migraine without aura shows that these migraine subtypes have both shared risk alleles and risk alleles that appear specific to one subtype. In addition, new loci include two targets of recently developed and effective migraine treatments. Therefore, we expect that these and future GWAS data will reveal more of the heterogeneous biology of migraine and potentially point to new therapies against migraine that currently is a leading burden for population health throughout the world.
RkJQdWJsaXNoZXIy MTk4NDMw