Xinghong Liu ^1* Yi Peng ² Ling Guo ¹ Weilan Xiong ³ Weijiang Liao ¹ Fan Jiangang ¹

• ¹ University of Electronic Science and Technology of China, Chengdu, China
• ² Department of Otolaryngology Head and Neck Surgery, Second People's Hospital of Chengdu,China, Chengdu, China
• ³ Chengdu University of Traditional Chinese Medicine, Chengdu, Sichuan Province, China

Extensive efforts have been made to explore members of the IL-10 family as potential therapeutic strategies for various diseases; however, their biological role in chronic rhinosinusitis with nasal polyps (CRSwNP) remains underexplored. Gene expression datasets GSE136825, GSE179265, and GSE196169 were retrieved from the Gene Expression Omnibus (GEO) for analysis. Candidate genes were identified by intersecting differentially expressed genes (DEGs) between the CRSwNP and control groups (DEGs_all) with those between the high-and low-score groups within the CRSwNP cohort (DEGs_NP). Biomarker selection was performed using the Least Absolute Shrinkage and Selection Operator (LASSO), Support Vector Machine Recursive Feature Elimination (SVM-RFE), and the Boruta algorithm. Further refinement of biomarkers was carried out using receiver operating characteristic (ROC) analysis, with genes demonstrating an area under the curve (AUC) greater than 0.7 being considered significant. Genes exhibiting consistent expression trends and significant differences across both GSE136825 and GSE179265 were selected as potential biomarkers. Cell-type annotation was performed on GSE196169, and the expression profiles of the biomarkers across various cell types were analyzed. A competing endogenous RNA (ceRNA) network and a biomarker- drug interaction network were also established. Additionally, the mRNALocater database was utilized to determine the cellular localization of the identified biomarkers. The intersection of 1817 DEGs_all and 24 DEGs_NP yielded 15 candidate genes. Further filtering through LASSO, SVM-RFE, and Boruta led to the identification of seven candidate biomarkers: PRB3, KRT16, MUC6, SPAG4, FGFBP1, NR4A1, and GSTA2. Six of these genes demonstrated strong diagnostic performance in GSE179265, while four biomarkers, showing both significant differences and consistent expression trends, were validated in both GSE179265 and GSE136825. Single-cell sequencing analysis of GSE196169 revealed seven distinct cell types, including endothelial cells, with the biomarkers predominantly expressed in epithelial cells. The ceRNA network comprised nine nodes and eleven edges, with only FGFBP1 exhibiting a complete lncRNA-miRNA-mRNA interaction. This study identifies several novel biomarkers and their associated drugs for CRSwNP therapy, as well as potential therapeutic targets, such as spiperone and arnenous acid, identified through molecular docking. Ultimately, this work underscores the identification of four IL-10 familyrelated biomarkers, providing a theoretical foundation for future clinical research in CRSwNP.