Gliomas are primary malignant brain tumors. Monocytes have been proved to actively participate in tumor growth. Weighted gene co-expression network analysis was used to identify meaningful monocyte-related genes for clustering. Neural network and SVM were applied for validating clustering results. Somatic mutation and copy number variation were used for defining the features of identified clusters. Differentially expressed genes (DEGs) between the stratified groups after performing elastic regression and principal component analyses were used for the construction of risk scores. Monocytes were associated with glioma patients' survival and exhibited high predictive value. The prognostic value of risk score in glioma was validated by the abundant expression of immune checkpoint and metabolic profile. Additionally, high risk score was positively associated with the expression of immunogenic and antigen presenting factors, which indicated high immune infiltration. A prognostic model based on risk score demonstrated high accuracy rate of receiver operating characteristic curves. Compared with previous studies, our research dissected functional roles of monocytes from large-scale analysis. Findings of our analyses strongly support an immune modulatory and prognostic role of monocytes in glioma progression. Notably, monocyte could be an effective predictor for therapy responses of glioma patients.
Keywords: glioma microenvironment; immune infiltration; immunotherapy; machine learning; monocyte; prognostic model.
Copyright © 2021 Zhang, Dai, Wu, Wang, Cao, Zhang, Wang, Zhang and Cheng.