Background Although studies have shown that patients with rheumatoid arthritis (RA) are at a higher risk of developing myasthenia gravis (MG), the causal relationship and shared genetic basis between these two diseases have not been fully investigated. The purpose of this study is to uncover the potential bidirectional causality between RA and MG, and to explore their shared genetic factors and possible pathogenic mechanisms. Methods First, we utilized genome-wide association (GWAS) data from the IEU Open GWAS project, employing the online analysis platform MRBASE and applying four Mendelian randomization (MR) methods (Inverse Variance Weighted regression, Weighted Median, MR-Egger, and Weighted Mode) to explore the bidirectional causal relationship between RA and MG. Subsequently, we extracted transcriptomic data for RA and MG from the GEO database and used differential expression analysis, weighted gene coexpression network analysis (WGCNA), machine learning, and gene set enrichment analysis (GSEA) to identify key hub genes and their associated pathways. Furthermore, we employed the CIBERSORT method to analyze the immune cell infiltration in both diseases. Ultimately, based on these identified hub genes, we constructed a diagnostic model—nomogram—to aid in the diagnosis and prediction of the diseases. Result RA is significantly associated with an increased risk of MG (Odds Ratio [OR]: 1.353, 95% Confidence Interval [CI]: 1.081 to 1.693, P = 0.008). However, there is insufficient evidence to support the hypothesis that MG increases the risk of RA. Through differential expression analysis and WGCNA methods, we collectively identified 18 key shared genes. Further, using two machine learning approaches, we ultimately identified 4 core hub genes (CDC42EP2, FKBP5, CD79A, and TDP1), which have great value in the diagnosis of RA and MG and are closely related to immune cell infiltration. Conclusion Our study has unveiled the bidirectional causality between RA and MG, and identified shared molecular characteristics, highlighting the potential for developing targeted therapeutic strategies.<Table Float="No" ID="Taba"> <tgroup cols="2"> <colspec align="left" colname="c1" colnum="1" /> <colspec align="left" colname="c2" colnum="2" /> <tbody> <row> <entry nameend="c2" namest="c1"> Key Points • Our study shows a significant link between RA and increased MG risk, suggesting a bidirectional causal relationship. • We identified 18 key shared genes between RA and MG through differential expression and WGCNA, and pinpointed 4 core hub genes (CDC42EP2, FKBP5, CD79A, TDP1) using several machine learning algorithms. These genes are valuable for diagnosis and associated with immune cell infiltration. • We developed a diagnostic nomogram based on the hub genes, which could aid in diagnosing and predicting RA and MG, guiding clinical practice and personalized medicine. </entry> </row> </tbody> </tgroup> </Table>

Genetic and molecular underpinnings of the link between rheumatoid arthritis and myasthenia gravis: Insights from GWAS and transcriptomic analyses

期刊论文
发表日期：

2025年7月19日

查看全文

Jian Huang,
Lu Wang,
Xiaodong Hu,
Tianrui Wang,
Yingze Zhang

摘要

Background

Although studies have shown that patients with rheumatoid arthritis (RA) are at a higher risk of developing myasthenia gravis (MG), the causal relationship and shared genetic basis between these two diseases have not been fully investigated. The purpose of this study is to uncover the potential bidirectional causality between RA and MG, and to explore their shared genetic factors and possible pathogenic mechanisms.

Methods

First, we utilized genome-wide association (GWAS) data from the IEU Open GWAS project, employing the online analysis platform MRBASE and applying four Mendelian randomization (MR) methods (Inverse Variance Weighted regression, Weighted Median, MR-Egger, and Weighted Mode) to explore the bidirectional causal relationship between RA and MG. Subsequently, we extracted transcriptomic data for RA and MG from the GEO database and used differential expression analysis, weighted gene coexpression network analysis (WGCNA), machine learning, and gene set enrichment analysis (GSEA) to identify key hub genes and their associated pathways. Furthermore, we employed the CIBERSORT method to analyze the immune cell infiltration in both diseases. Ultimately, based on these identified hub genes, we constructed a diagnostic model—nomogram—to aid in the diagnosis and prediction of the diseases.

Result

RA is significantly associated with an increased risk of MG (Odds Ratio [OR]: 1.353, 95% Confidence Interval [CI]: 1.081 to 1.693, P = 0.008). However, there is insufficient evidence to support the hypothesis that MG increases the risk of RA. Through differential expression analysis and WGCNA methods, we collectively identified 18 key shared genes. Further, using two machine learning approaches, we ultimately identified 4 core hub genes (CDC42EP2, FKBP5, CD79A, and TDP1), which have great value in the diagnosis of RA and MG and are closely related to immune cell infiltration.

Conclusion

Our study has unveiled the bidirectional causality between RA and MG, and identified shared molecular characteristics, highlighting the potential for developing targeted therapeutic strategies.

Key Points

• Our study shows a significant link between RA and increased MG risk, suggesting a bidirectional causal relationship.

• We identified 18 key shared genes between RA and MG through differential expression and WGCNA, and pinpointed 4 core hub genes (CDC42EP2, FKBP5, CD79A, TDP1) using several machine learning algorithms. These genes are valuable for diagnosis and associated with immune cell infiltration.

• We developed a diagnostic nomogram based on the hub genes, which could aid in diagnosing and predicting RA and MG, guiding clinical practice and personalized medicine.