Aster Harder

SUMMARY 239 & In Chapter 6 the aim was to identify genetic susceptibility loci for cluster headache.To this end the DNA of 840 cluster headache patients and 1,457 controls was genotyped, quality controlled and imputated and compared to each other with a logistic regression model. An association was found with cluster headache for four independent loci rs11579212 (near RP11-815M8.1), rs6541998 (near MERTK), rs10184573 (near AC093590.1), and rs2499799 (near UFL1/FHL5), these loci collectively explain 7.2% of the variance of the phenotype. Three of the four loci replicated in an independent cohort of 144 cluster headache cases and 1,800 controls from Norway. Contrary to what is normally found in a GWAS, very large effect sizes were found in our study (all with an odds ratio of >1.2), indicating that the risk for cluster headache seems driven by a limited number of loci. In addition, the effect of the four loci on the expression of genes was investigated and this yielded 16 genes that seem involved in the pathophysiology of the disease. The expression of these genes was investigated in a previously performed RNA-sequencing dataset, by which a differential expression of POLR1B and TMEM87B in cluster headache patients compared to controls was found. In a separate correlation analysis, there was an indication of a genetical overlap of cluster headache with migraine. At a similar time, a UK-Swedish cohort also performed a GWAS in cluster headache patients based in 852 UK cases and 5,614 controls as well as 591 Swedish cases and 1,134 controls. This parallel study found four loci that were in linkage disequilibrium with our loci. This provides support that the four loci are genuine risk loci for cluster headache. Data of our study were also combined with the UK-Sweden data in a meta-analysis in a post-scriptum of the chapter. This meta-analysis resulted in three additional loci becoming genome-wide significant. The research for Chapter 7 is a further expansion on the meta-analysis of cluster headache performed in Chapter 6. In Chapter 7 the genetic data of the two earlier studies (Netherlands/ Norway and UK/Sweden) together with another previously published small Italian GWAS study and five novel European cohorts was jointly analysed. In total, the meta-analysis encompassed 4,043 cluster headache patients and 21,729 controls from ten cohorts, all of European ancestry. Seven genome-wide significant loci were identified, of which three are novel rs2402176 (WNT2), rs57866767 (PLCE1) and rs11172113 (LRP1) and four previously identified rs17011182, (DUSP10), rs13399108 (MERTK), rs6714578 (FTCDNL1) and rs9486725 (FHL5). Downstream bioinformatics analyses showed enrichment for artery and brain tissue. Furthermore, correlation analyses showed a genetic overlap of cluster headache with migraine, cigarette smoking, risk-taking behaviour, ADHD, depression, and musculoskeletal pain. An in-depth analysis of the overlap with migraine showed that cluster headache and migraine share three genetic risk loci. This suggests that the two disorders are genetically partially the same and partially distinct. Furthermore, the potential causality of smoking on cluster headache was investigated with a two-sample Mendelian randomization analysis based on the summary statistics. This analysis indicated a causal effect of smoking intensity on cluster headache. This effect of smoking on cluster headache has potential clinical implications. In a secondary trans-ancestry meta-analysis we added 734 cases and 9,846 controls of East Asian ancestry to our meta-analysis. One additional genome-wide significant cluster headache locus, in CAPN2, was identified when adding the East Asian cohort in an

RkJQdWJsaXNoZXIy MTk4NDMw