Assessing the potential causal influence of myasthenia gravis on neurodegenerative diseases via multivariable Mendelian randomization.

Myasthenia gravis (MG), an autoimmune condition known for impairing neuromuscular signaling, has increasingly been implicated in broader neurological dysfunctions. Recent studies point toward a possible connection between autoimmune and neurodegenerative processes. However, whether MG contributes causally to the onset of major neurodegenerative disorders such as Alzheimer disease (AD), Parkinson disease (PD), and amyotrophic lateral sclerosis (ALS) remains unclear. This study utilizes Mendelian randomization (MR) to explore the potential causal influence of MG on these disorders from a genetic standpoint. A univariable Mendelian randomization (UVMR) framework was employed using summary-level data from genome-wide association studies (GWAS) to evaluate the effect of MG on the risk of AD, PD, and ALS. To confirm the robustness of the association between MG and AD, 2 independent AD GWAS datasets were incorporated for external replication, followed by a meta-analysis to combine the evidence. Additionally, multivariable Mendelian randomization (MVMR) was conducted to adjust for smoking behavior as a potential confounding factor. The UVMR analysis revealed a statistically significant causal relationship between MG and increased susceptibility to AD (odds ratio (OR): 1.037; 95% confidence interval (CI): 1.007-1.068; P = .016). No significant causal effects were observed for PD (OR: 1.019; 95% CI: 0.964-1.077; P = .509) or ALS (OR: 1.055; 95% CI: 0.977-1.140; P = .171). The association between MG and AD was consistently validated in 2 independent datasets (ieu-a-297: OR = 1.084; 95% CI: 1.017-1.156; P = .013; ieu-b-2: OR = 1.054; 95% CI: 1.006-1.104; P = .027). Meta-analysis reinforced the evidence supporting MG as a risk factor for AD (OR: 1.047; 95% CI: 1.023-1.072; P < .001). Furthermore, MVMR adjusting for smoking confirmed that MG independently contributes to AD risk (OR: 1.037; 95% CI: 1.006-1.069; P = .020). This study provides robust genetic evidence suggesting that MG is a causal and independent risk factor for AD. These findings highlight a novel link between autoimmunity and neurodegeneration, offering new directions for mechanistic and therapeutic research.