Assessing the causal relationship between inflammatory cytokines and myasthenia gravis: A bidirectional and two-sample Mendelian randomization study

Abstract

Abstract Background: Myasthenia gravis (MG) is categorized as an autoimmune disorder. A myriad of preceding investigations have elucidated the interrelation between inflammation and MG. Nevertheless, the precise array of inflammatory cytokines and their underlying mechanisms of action remain enigmatic. Objective: The primary intent of this research was to discern and elucidate the potential causal associations between 41 inflammatory cytokines and MG, employing a bidirectional and two-sample Mendelian randomization (MR) paradigm. Methods: Summary statistics pertaining to MG were procured from a genome-wide association study (GWAS) comprising 232 cases and 217,056 controls of European descent. Concurrently, genetic variants were sourced from an avant-garde GWAS centered on inflammatory cytokines, with an inclusive sample of 8,293 European subjects. An array of techniques including Inverse Variance Weighted (IVW), MR-Egger, weighted median, simple mode, and Weighted Mode were harnessed to assess the putative causal interlinkages between the 41 inflammatory cytokines and MG. Primarily, IVW outcomes formed the basis for deliberations on causality. Additionally, the MR-Egger approach was utilized to ascertain the potential horizontal pleiotropy of instrumental variables, and Cochran's Q statistics were employed to gauge the heterogeneity intrinsic to instrumental variables (IV). Results: Findings derived from the IVW model posited that IL-10 (OR=0.46, 95% CI=0.300.72, P=0.0006), IL-17 (OR=0.43, 95% CI=0.210.85, P=0.015), MIP1a (OR=2.24, 95% CI=1.014.97, P=0.048), MIP1b (OR=0.84, 95% CI=0.720.99, P=0.033), and SDF1a (OR=2.32, 95% CI=1.124.81, P=0.024) are potentially implicated in MG susceptibility. Concurrently, cytokines such as bNGF (OR=1.03, 95% CI=1.001.07, P=0.038, PFDR=0.778) and MIF (OR=0.97, 95% CI=0.95~1.00, P=0.033, PFDR=0.778) are ostensibly resultant sequelae of MG pathology. Upon the application of the False Discovery Rate (FDR) correction, a profound association was discerned solely for IL-10 in relation to MG (PFDR=0.025). Contrastingly, multiple inflammatory cytokines, including IL-17, MIP1a, MIP1b, and SDF1a, failed to meet the threshold post-FDR correction (with PFDR＞0.05), as did bNGF and MIF. Conclusions: The Mendelian randomization study, anchored in genetic variation methodologies, substantiates that IL-10 potentially holds significant relevance in the etiological landscape of MG. Contrarily, no discernible evidence was found to implicate other inflammatory cytokines in the post-onset progression of MG.