Genome-wide association study of panic disorder reveals genetic overlap with neuroticism and depression

Andreas J. Forstner, Swapnil Awasthi, Christiane Wolf, Eduard Maron, Angelika Erhardt, Darina Czamara, Elias Eriksson, Catharina Lavebratt, Christer Allgulander, Nina Friedrich, Jessica Becker, Julian Hecker, Stefanie Rambau, Rupert Conrad, Franziska Geiser, Francis J. McMahon, Susanne Moebus, Timo Hess, Benedikt C. Buerfent, Per HoffmannStefan Herms, Stefanie Heilmann-Heimbach, Ingrid Kockum, Tomas Olsson, Lars Alfredsson, Heike Weber, Georg W. Alpers, Volker Arolt, Lydia Fehm, Thomas Fydrich, Alexander L. Gerlach, Alfons Hamm, Tilo Kircher, Christiane A. Pané-Farré, Paul Pauli, Winfried Rief, Andreas Ströhle, Jens Plag, Thomas Lang, Hans Ulrich Wittchen, Manuel Mattheisen, Sandra Meier, Andres Metspalu, Katharina Domschke, Andreas Reif, Iiris Hovatta, Nils Lindefors, Evelyn Andersson, Eco J.C. de Geus, Brenda W.J.H. Penninx, Jürgen Deckert, J. Schumacher, NESDA

Research output: Contribution to journalArticleAcademicpeer-review


Panic disorder (PD) has a lifetime prevalence of 2–4% and heritability estimates of 40%. The contributory genetic variants remain largely unknown, with few and inconsistent loci having been reported. The present report describes the largest genome-wide association study (GWAS) of PD to date comprising genome-wide genotype data of 2248 clinically well-characterized PD patients and 7992 ethnically matched controls. The samples originated from four European countries (Denmark, Estonia, Germany, and Sweden). Standard GWAS quality control procedures were conducted on each individual dataset, and imputation was performed using the 1000 Genomes Project reference panel. A meta-analysis was then performed using the Ricopili pipeline. No genome-wide significant locus was identified. Leave-one-out analyses generated highly significant polygenic risk scores (PRS) (explained variance of up to 2.6%). Linkage disequilibrium (LD) score regression analysis of the GWAS data showed that the estimated heritability for PD was 28.0–34.2%. After correction for multiple testing, a significant genetic correlation was found between PD and major depressive disorder, depressive symptoms, and neuroticism. A total of 255 single-nucleotide polymorphisms (SNPs) with p < 1 × 10−4 were followed up in an independent sample of 2408 PD patients and 228,470 controls from Denmark, Iceland and the Netherlands. In the combined analysis, SNP rs144783209 showed the strongest association with PD (pcomb = 3.10 × 10−7). Sign tests revealed a significant enrichment of SNPs with a discovery p-value of <0.0001 in the combined follow up cohort (p = 0.048). The present integrative analysis represents a major step towards the elucidation of the genetic susceptibility to PD.

Original languageEnglish
JournalMolecular Psychiatry
Publication statusE-pub ahead of print - 1 Jan 2019

Cite this