Development and validation of prognostic models for colon adenocarcinoma based on combined immune-and metabolism-related genes

Abstract

BackgroundThe heterogeneity of tumor tissue is one of the reasons for the poor effect of tumor treatment, which is mainly affected by the tumor immune microenvironment and metabolic reprogramming. But more research is needed to find out how the tumor microenvironment (TME) and metabolic features of colon adenocarcinoma (COAD) are related.MethodsWe obtained the transcriptomic and clinical data information of COAD patients from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Consensus clustering analysis was used to identify different molecular subtypes, identify differentially expressed genes (DEGs) associated with immune-and metabolism-related genes (IMRGs) prognosis. Univariate and multivariable Cox regression analysis and Lasso regression analysis were applied to construct the prognostic models based on the IMRG risk score. The correlations between risk scores and TME, immune cell infiltration, and immune checkpoint genes were investigated. Lastly, potential appropriate drugs related to the risk score were screened by drug sensitivity analysis.ResultsBy consensus clustering analysis, we identified two distinct molecular subtypes. It was also found that the multilayered IMRG subtypes were associated with the patient’s clinicopathological characteristics, prognosis, and TME cell infiltration characteristics. Meanwhile, a prognostic model based on the risk score of IMRGs was constructed and its predictive power was verified internally and externally. Clinicopathological analysis and nomogram give it better clinical guidance. The IMRG risk score plays a key role in immune microenvironment infiltration. Patients in the high-risk groups of microsatellite instability (MSI) and tumor mutational burden (TMB) were found to, although with poor prognosis, actively respond to immunotherapy. Furthermore, IMRG risk scores were significantly associated with immune checkpoint gene expression. The potential drug sensitivity study helps come up with and choose a chemotherapy treatment plan.ConclusionOur comprehensive analysis of IMRG signatures revealed a broad range of regulatory mechanisms affecting the tumor immune microenvironment (TIME), immune landscape, clinicopathological features, and prognosis. And to explore the potential drugs for immunotherapy. It will help to better understand the molecular mechanisms of COAD and provide new directions for disease treatment.

Analysis of DHX32 as diagnostic gene and independent prognostic factor of lung squamous cell carcinoma

“…2.7 Drug sensitivity analyses GDSC (https://www.Cancerrxgene.org) was used to access the chemotherapy drugs of LUSC. The IC50 of the drug in each sample was calculated using the R-package 'pRRophetic' (version 0.5) [11] . Target genes of drugs with signi cant sensitivity differences were predicted through the DrugBank database (https://go.drugbank.com/).…”

Development and verification of a combined immune- and cancer-associated fibroblast related prognostic signature for colon adenocarcinoma

“…The relative predictive ability of the immune-and CAFassociated model described in this study was compared with the predictive ability of four previously-described prognostic models (12)(13)(14)(15). To ensure the comparability of these signatures, the same method for calculating and converting the RS was applied to the entire TCGA-COAD cohort.…”