The Causal Effect Of Interleukin-17 On The Risk Of Psoriatic Arthritis: A Mendelian Randomization Study

Objective. To determine causal associations between genetically predicted TNF-alpha, IL-12p70 and IL-17 levels and risk of PsA.Methods. The publicly available summary-level findings from genome-wide association studies (GWAS) was used to identify loci influencing normal physiological concentrations of TNF-alpha, IL-12p70 and IL-17 (n=8293) among healthy individuals as exposure and a GWAS for PsA from the UK Biobank (PsA = 900, control = 462 033) as the outcome. A two-sample Mendelian randomization (MR) analysis was performed using the inverse-variance weighted (IVW), weighted median and MR-Egger regression methods. Sensitivity analysis and MR-Egger regression analysis were performed to evaluate the heterogeneity and pleiotropic effects of each variant.Results. Single-nucleotide polymorphisms (SNPs) at genome-wide significance from GWASs on TNF-alpha, IL-12p70 and IL-17 were identified as the instrumental variables. The IVW method indicated a causal association between increased IL-17 level and risk of PsA (beta=-0.00186 per allele, s.e.=0.00043, P=0.002). Results were consistent in the weighted median method (beta=-0.00145 per allele, s.e.=0.00059, P=0.014) although the MR-Egger method suggested a non-significant association (beta =-0.00133 per allele, s.e.=0.00087; P=0.087). Single SNP MR results revealed that the C allele of rs117556572 was robustly associated with risk of PsA (beta=0.00210, s.e.=0.00069, P=0.002). However, no evidence for a causal effect was observed between TNF-alpha, IL-12p70, decreased IL-17 levels and risk of PsA.Conclusion. Our findings provide preliminary evidence that genetic variants predisposing to higher physiological IL-17 level are associated with decreased risk of PsA.