Identification of Immune-Related Biomarkers Linked to Systemic Lupus Erythematosus and Dilated Cardiomyopathy through Integrated Bioinformatics Analysis and Multiple Machine Learning Algorithms

Gaijie Li¹Liwen Lin¹Shushu Wang¹Kachun Lu¹KaMan Szeto¹Guiting Zhou^2*Xianwen Tang^3*Chuanjin Luo^4,5*

• ¹First Clinical Medical College, Guangzhou University of Chinese Medicine, Guangzhou, Guangdong Province, China
• ²Guangzhou University of Chinese Medicine, Guangzhou, Guangdong Province, China
• ³Shenzhen Hospital of Beijing University of Traditional Chinese Medicine, Shenzhen, Guangdong, China
• ⁴Department of Cardiology, The First Affiliated Hospital of Chinese Medicine, Guangzhou University of Chinese Medicine, Guangzhou, Guangdong Province, China
• ⁵Guangdong Clinical Research Academy of Chinese Medicine, Guangzhou, China

Background: Epidemiological evidence indicates that up to 50% of systemic lupus erythematosus (SLE) patients exhibit cardiac involvement, suggesting a potential strong association between SLE and dilated cardiomyopathy (DCM). This study aims to identify SLE-related genes that may contribute to DCM development and to discover potential biomarkers for early DCM diagnosis in SLE patients. Methods: We obtained expression profile datasets for dilated cardiomyopathy DCM and SLE from the Gene Expression Omnibus (GEO) database. Through differential expression analysis and weighted gene co-expression network analysis (WGCNA), we screened for candidate biomarkers shared between DCM and SLE and constructed a diagnostic nomogram. The diagnostic performance and effectiveness of the nomogram were evaluated using external datasets and qPCR. Additionally, we performed single-gene set enrichment analysis (GSEA) on key genes to elucidate their potential roles in SLE-related DCM. Finally, we applied the CIBERSORT algorithm to assess immune cell infiltration in both DCM and SLE patients. Results: Through DEG and WGCNA in the DCM and SLE datasets, we identified a total of 141 key module genes and 24 commonly expressed differentially expressed genes. Enrichment analysis revealed that these 24 genes were primarily involved in inflammation, cell apoptosis, and immune regulation. Through machine learning algorithms and dataset validation, we further identified the HERC6 and IFI44L genes as important diagnostic markers for SLE-related DCM. Experimental validation supports the key role of HERC6, IFI44L, and RSAD2 in SLE-related cardiac dysfunction. Additionally, we developed a nomogram for DCM based on these two genes, and the results showed that both genes exhibited AUC values greater than 0.84. Simultaneously, single-GSEA and immune infiltration analysis indicated immune dysfunction in both DCM and SLE, with both HERC6 and IFI44L significantly associated with immune cell infiltration. Furthermore, connectivity map (cMAP) analysis identified α -linolenic acid as a potential therapeutic agent for treating DCM. Conclusion: Our study identifies HERC6 and IFI44L as diagnostic markers for DCM in SLE and suggests α-linolenic acid as a potential therapeutic agent.