Identification of Qualitative Characteristics of Immunosuppression in Sepsis Based on Immune-Related Genes and Immune Infiltration Features
Ni Zeng,Zaijin Jian,Junmei Xu,Tian Peng,Guiping Hong,Feng Xiao
DOI: https://doi.org/10.1016/j.heliyon.2024.e29007
IF: 3.776
2024-01-01
Heliyon
Abstract:ObjectiveSepsis is linked to high morbidity and mortality rates. Consequently, early diagnosis is crucial for proper treatment, reducing hospitalization, and mortality rates. Additionally, over one-fifth of sepsis patients still face a risk of death. Hence, early diagnosis, and effective treatment play pivotal roles in enhancing the prognosis of patients with sepsis.MethodThe study analyzed whole blood data obtained from patients with sepsis and control samples sourced from three datasets (GSE57065, GSE69528, and GSE28750). Commonly dysregulated immune-related genes (IRGs) among these three datasets were identified. The differential characteristics of these common IRGs in the sepsis and control samples were assessed using the REO-based algorithm. Based on these differential characteristics, samples from eight Gene Expression Omnibus (GEO) databases (GSE57065, GSE69528, GSE28750, GSE65682, GSE69063, GSE95233, GSE131761, and GSE154918), along with three ArrayExpress databases (E-MTAB-4421, E-MTAB-4451, and E-MTAB-7581), were categorized and scored. The effectiveness of these differential characteristics in distinguishing sepsis samples from control samples was evaluated using the AUC value derived from the receiver operating characteristic curve (ROC) curve. Furthermore, the expression of IRGs was validated in peripheral blood samples obtained from patients with sepsis through qRT-PCR.ResultsAmong the three training datasets, a total of 84 common dysregulated immune-related genes (IRGs) were identified. Utilizing a within-sample relative expression ordering (REOs)-based algorithm to analyze these common IRGs, differential characteristics were observed in three reverse stable pairs (ELANE-RORA, IL18RAP-CD247, and IL1R1-CD28). In the eight GEO datasets, the expression of ELANE, IL18RAP, and IL1R1 demonstrated significant upregulation, while RORA, CD247, and CD28 expression exhibited notable downregulation during sepsis. These three pairs of immune-related marker genes displayed accuracies of 95.89% and 97.99% in distinguishing sepsis samples among the eight GEO datasets and the three independent ArrayExpress datasets, respectively. The area under the receiver operating characteristic curve ranged from 0.81 to 1.0. Additionally, among these three immune-related marker gene pairs, mRNA expression levels of ELANE and IL1R1 were upregulated, whereas the levels of CD247 and CD28 mRNA were downregulated in blood samples from patients with sepsis compared to normal controls.ConclusionThese three immune-related marker gene pairs exhibit high predictive performance for blood samples from patients with sepsis. They hold potential as valuable auxiliary clinical blood screening tools for sepsis.