Mendelian randomization study for the roles of IL-18 and IL-1 receptor antagonist in the development of inflammatory bowel disease

IL-1 and IL-18 play important roles in intestine barrier integrity maintenance and inflammatory response. However, their net effects on the risk of IBD are still inconclusive. Here, we used Mendelian randomization (MR) approaches to investigate the causal associations of IL-18 and IL-1Ra (receptor antagonist) on the risks of IBD and subtypes. For IL-18, both three-sample and two-sample MR approaches were used for the causal inferences. In three-sample MR, three single nucleotide polymorphisms (SNPs) and the effect values were extracted from two quantitative trait loci (pQTL) datasets with non-overlapping populations. In two-sample MR, we extracted genetic instruments information from the same larger pQTL dataset. For IL-1Ra, we applied the two-sample MR method with summary-statistics from the larger pQTL dataset. Summary-level results of three large IBD/CD/UC genome-wide association studies in European ancestry were employed. Inverse-variance weighted method, various sensitivity analyses and meta-analysis were performed to give causal estimates, detect heterogeneity and correct for outliers. We observed consistent positive causal effects of IL-18 on all three major outcomes using three-sample MR, with meta-analyses odds ratios (ORs) equal to 1.240 (IBD), 1.199 (CD) and 1.274 (UC) respectively. The two-sample MR demonstrated similar results. Moreover, genetically predicted IL-1Ra is inversely associated with the risk of IBD/UC/CD with ORs equal to 0.915 (IBD), 0.902 (CD) and 0.899 (UC) respectively in meta-analyses. This study suggested genetically predicted IL-18 and IL-1Ra level are causally associated with an increased and decreased risk of IBD and subtypes.