Identification of a pyroptosis-related lncRNA risk model for predicting prognosis and immune response in colon adenocarcinoma

Abstract

Background: Colon adenocarcinoma (COAD) is one of the most common malignant tumors and is diagnosed at an advanced stage with a poor prognosis worldwide. Pyroptosis is involved in the initiation and progression of tumors. This research focused on constructing a pyroptosis-related ceRNA network to generate a reliable risk model for risk prediction and immune infiltration analysis of COAD. Methods: Transcriptome data, miRNA-sequencing data, and clinical information were downloaded from the TCGA database. First, differentially expressed mRNAs (DEmRNAs), miRNAs (DEmiRNAs), and lncRNAs (DElncRNAs) were identified to construct a pyroptosis-related ceRNA network. Second, a pyroptosis-related lncRNA risk model was developed applying univariate Cox regression analysis and least absolute shrinkage and selection operator method (LASSO) regression analysis. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) enrichment analyses were utilized to functionally annotate RNAs contained in the ceRNA network. In addition, Kaplan-Meier analysis, receiver operating characteristic (ROC) curves, univariate and multivariate Cox regression, and nomogram were applied to validate this risk model. Finally, the relationship of this risk model with immune cells and immune checkpoint blockade (ICB)-related genes was analyzed. Results: A total of 5373 DEmRNAs, 1159 DElncRNAs, and 355 DEmiRNAs were identified. A pyroptosis-related ceRNA regulatory network containing 132 lncRNAs, 7 miRNAs, and 5 mRNAs was constructed, and a ceRNA-based pyroptosis-related risk model including 11 lncRNAs was built. The tumor tissues were classified into high- and low-risk groups according to the median risk score. Kaplan-Meier analysis showed that the high-risk group had a shorter survival time; ROC analysis, independent prognostic analysis, and nomogram further indicated the risk model was a significant independent prognostic factor what had an excellent ability to predict patients’ risk. Moreover, immune infiltration analysis indicated that the risk model was related to immune infiltration cells (i.e., B cell naïve, T cell follicular helper, macrophage M1) and ICB-related genes (i.e., PD-1, CTLA4, HAVCR2). Conclusions: This pyroptosis-related lncRNA risk model possessed good prognostic value, and the ability to predict the outcome of ICB immunotherapy in COAD.