Abstract
Background & Aims: Cirrhosis is a leading cause of liver-related mortality and a multifactorial disease. To date, the complex genetic architecture of non-viral cirrhosis has not been fully explored. Cross-trait genetic correlations can elucidate the common genetic etiology of genetically correlated phenotypes. This study aims to identify polygenic and pleiotropic traits associated with cirrhosis using the linkage disequilibrium score regression analysis.
Methods: We conducted genome-wide association analysis of 9,622,842 imputed SNPs on 3,368 non-viral cirrhosis cases and 258,258 controls, and cross-trait analysis between non-viral cirrhosis and various polygenic and pleiotropic traits using the UK Biobank cohort study. We further performed sensitivity analyses by removing genomic regions of alcohol intake, smoking behaviors, and obesity. We observed multiple traits showing robust genetic correlations (rg) with non-viral cirrhosis.
Results: We found strong genetic correlations between the genetic architectures of non-viral cirrhosis and clinical/physiologic factors, including BMI (rg=0.82), alanine aminotransferase (0.71), diabetes (0.70), number of cigarettes currently smoked daily (0.67), amount of alcohol drunk on a typical drinking day (0.60), insomnia (0.59), gout (0.57), depression (0.50), apoliprotein-A (-0.33), HDL cholesterol (-0.49). Exclusion of genomic regions associated with alcohol intake, smoking behaviors, and obesity demonstrated consistent directions and persistent associations in genetic patterns. The inheritability of cirrhosis on the observed scale showed 0.56%.
Conclusions: This study provides a comprehensive assessment of the shared genetic architecture of non-viral cirrhosis predisposition and numerous polygenic and pleiotropic traits, most notably BMI, alanine aminotransferase, and diabetes. These findings provide new information on underlying comorbid conditions that can increase the non-viral cirrhosis risk.