Prognostic Value of Immune-Related genes in the Tumor Microenvironment of Glioma - Analyzed by Integrated Bioinformatics [post]

Yali Zhong, Xiaobin Luo, Fubing Yang, Xinling Song
2021 unpublished
Object: Immune related genes play an important role in the process of tumor genesis and development. Therefore, we aim to find the Immune genes which are related to the prognosis of glioma patients, and to explore the infiltration of Immune cells in glioma microenvironment. Methods We downloaded the data of the glioma samples from the CGGA database, and performed batch correction to screen the primary glioma samples for subsequent analysis. Then the ESTIMATE algorithm was used to deal with the
more » ... tromal scores and Immune scores of the primary glioma samples, and the difference was analyzed. Then the common Immune related genes (IRGs) were obtained by intersecting with the Immune genes in the ImmPort database. Moreover, we used common IRGs to construct protein-protein interaction (PPI) networks, from which we screened the top 30 genes with high connectivity, and Lasso regression was used to screen the IRGs. Lastly, we obtained the combined genes, which were overlapped both in the top 30 high-connection genes and Lasso regression genes. The final genes were used to construct COX risk prediction models. The accuracy of the model were verified by the TCGA glioma data, and the model genes were analyzed for Immune-related pathways, as well as the Hallmark and KEGG enrichment. Additionally, we used CIBERSOFT algorithm to estimate the Immune cell content of the samples, and analyzed the differences, correlations and survival of the Immune cells in high and low risk groups. Results Firstly, a total of 117 IRGs were obtained from the gene sets, which were overlapped in the data of Stromal score, Immune score and ImmPort database. Secondly, the top 30 genes were selected after the PPI network, and another 26 genes were screened out after the Lasso regression algorithm. And then, six coexist IRGs were obtained from the intersecting sets. Furthermore, the COX risk prediction model was constructed and tested, showing that the overall survival rate of the high-risk group was about 50% of that of the low-risk group. We observed that the high-risk group were enriched in Immune response and Immune process. Most importantly, in KEGG pathways, the high-risk groups were mainly enriched in p53 signaling pathway, JAK-STAT signaling pathway, pathways in cancer and cell cycle. By estimating the Immune cell contents, we also found that the Immune cell Plasma cells, T cells CD8, T cells CD4 naïve, T cells regulatory (Tregs), Macrophages M0 and Neutrophils were higher in high-risk groups, when compared to the low-risk group, with significant difference. Finally, the correlation analysis showed that the degree of Immune infiltration in high-risk groups was related to T cells regulatory (Tregs), Macrophages M0 and Neutrophils. Conclusion A COX risk prediction model of 6 genes was successfully constructed, which was enriched in Immune-related pathways. Meanwhile, survival analysis and TCGA data validation revealed significant differences in the model genes in the overall survival of the glioma patients, and the degree of Immune infiltration in the model was associated with T cells regulatory (Tregs), Macrophages M0 and Neutrophils.
doi:10.21203/ fatcat:ssvmhihrnrcj3aygcefywhyqmu