Genome-wide association study of image-based emphysema scoring in the Swedish CArdioPulmonary bioImage Study (SCAPIS) suggests two new risk loci in smokers

Abstract

Abstract Background Despite the high prevalence and clinical significance of emphysema, few genetic risk loci have been consistently replicated. We conducted a genome-wide association study (GWAS) of CT-based emphysema, with a particular focus on non-smoking-related genetic determinants. Methods We analyzed 25,639 individuals of European ancestry from the SCAPIS national cohort, aged 50–65 years, of which 51% were never-smokers. Emphysema was assessed through semi-quantitative visual scoring of CT scans. GWAS was performed in the whole sample and stratified on smoking status. We also examined the association of previously reported emphysema- and lung function-related variants with emphysema in our dataset. Results Emphysema criteria were fulfilled for 1,479 participants (5.6%), with higher prevalence among current ( N = 576, 18.2%) and former smokers ( N = 612, 6.5%) compared to never-smokers ( N = 263, 2.0%). We identified three independent genetic loci for emphysema in smokers and no signals in never-smokers. The strongest signal was observed in the well-established nicotinic acetylcholine receptor cluster (CHRNA5-A3-B4) locus on chromosome 15. Additionally, we discovered novel associations near the dysferlin (DYSF) gene on chromosome 2 and in an intergenic region on chromosome 3. By assessing previously lung phenotype-associated variants we also found evidence supporting association with emphysema in smokers for variants in the EFEMP1/MIR217HG/PNPT1 locus on chromosome 2, previously linked to reduced FEV 1 /FVC ratio. Conclusion This study, based on the largest unselected population sample to date, provides novel insights into the genetic architecture of emphysema. However, no signals were detected in never-smokers despite the large sample-size, likely due to the low prevalence of emphysema in that group. The proposed genetic risk loci require external replication.