Exploring the relationship between pyroptosis, infiltrating immune cells and Kawasaki disease with resistance to intravenous immunoglobulin (IVIG) via bioinformatic analysis

Kawasaki disease (KD) is a kind of vasculitis predominantly afflicting children younger than five. Although intravenous immunoglobulin (IVIG) has been regarded as the first-line therapy, there are some children unresponsive to it, resulting in higher risk of coronary artery aneurysms (CAA), the most severe complication of KD. Pyroptosis is an inflammatory apoptosis, which resembles the traits of IVIG-resistance. Therefore, our research aims to find relationships between KD with IVIG-resistance and pyroptosis, and provide the underlying mechanisms of IVIG-resistance.The transcriptome data of three datasets were downloaded from Gene Expression Omnibus (GEO) database. CIBERSORTx and WGCNA were combined to identify the coexpression gene network correlated with the up-regulated immune cells in KD, using differentially expressed genes (DEGs) overlapped in GSE68004 and GSE73461. The key genes in hub module were intersected with pyroptosis-related genes (PRGs). Then KD patients were divided into subgroups according to the expression of remaining genes, along with the construction of risk score (RS) based on the least absolute shrinkage and selection operator (LASSO) regression analysis. Besides, we explored the clinical value of RS between IVIG-responsive and -resistant KD patients in GSE16797. In addition, the biological pathways between subgroups were evaluated using Gene Set Variation Analysis (GSVA).A total of 4246 DEGs and three immune cells, including Monocytes, M0 macrophage, and neutrophils, were analyzed with P < 0.05 between KD and healthy controls (HCs). The lightcyan module was the hub module based on WGCNA, and only NLRC4, CASP1, CASP4, GSDMD, IL1B and PYCARD in the hub module were overlapped with PRGs. Then KD patients in GSE68004 were stratified into two clusters on the basis of the expression levels of six genes. RS was built with five out of six genes (exclude PYCARD) according to the LASSO analysis, which could differentiate C1 from C2, IVIG-responsive from -resistant KD patients. Besides, the high-risk group (C1) tended to be with increased levels of inflammation, immune responses and infiltration of neutrophils according to the analysis of GSVA and CIBERSORTx.We built a pyroptosis-related RS to evaluate the degree of pyroptosis and infiltrating immune cells in subgroups of KD, and associated it with the responsiveness to IVIG, which might help us to further understand the pathological process during IVIG-nonresponse.