Construction and Comprehensive Prognostic Analysis of a lncRNA–miRNA–mRNA Regulatory Network and Tumor Immune Cell Infiltration in Colorectal Cancer

• Main Authors: Xiong Guo, Xiaolong Liang, Yujun Wang, Anqi Cheng, Chuan Qin, Han Zhang, Ziwei Wang
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2021-07-01
• Series: Frontiers in Genetics
• Subjects: colorectal cancer; ceRNA network; infiltrating immune cells; prognostic model; bioinformatics analysis
• Online Access: https://www.frontiersin.org/articles/10.3389/fgene.2021.652601/full

Colorectal cancer (CRC) is a malignant tumor with high morbidity and mortality worldwide. Recent studies have shown that long noncoding RNAs (lncRNAs) play an important role in almost all human tumors, including CRC. Competitive endogenous RNA (ceRNA) regulatory networks have become hot topics in cancer research. Tumor-infiltrating immune cells (TICs) have also been reported to be closely related to the survival and prognosis of CRC patients. In this study, we used the lncRNA–miRNA–mRNA regulatory network combined with tumor immune cell infiltration to predict the survival and prognosis of 598 CRC patients. First, we downloaded the lncRNA, mRNA, and miRNA transcriptome data of CRC patients from The Cancer Genome Atlas (TCGA) database and identified differentially expressed genes through “limma” package of R software. The ceRNA regulatory network was established by using the “GDCRNATools” R package. Then, univariate Cox analysis and least absolute shrinkage and selection operator analysis were performed to identify the optimal prognostic network nodes, including SRPX, UST, H19, SNHG7, hsa-miR-29b-3p, and TTYH3. Next, we analyzed the differences in 22 types of TICs between 58 normal subjects and 206 CRC patients and included memory CD4 T cells, dendritic cells and neutrophils in the construction of a prognostic model. Finally, we identified the relationship between the ceRNA prognostic model and the infiltrating immune cell prognostic model. In conclusion, we constructed two prognostic models that provide insights on the prognosis and treatment strategy of CRC.