Identification Of Prognostic Key Genes Associated With Liver Cancer Microenvironment And Construction Of Prognostic Model

Background and aimsDespite significant advances in multimodality treatments,hepatocellular carcinoma(HCC)remains one of the most common malignant tumors.Identification of novel prognostic biomarkers and molecular targets is urgently needed.This study aims to identify potential prognostic key genes associated with tumor microenvironments and construct a prognostic model of HCC by combining with clinical features.MethodsBased on the gene expression profile of HCC patients in The Cancer Genome Atlas(TCGA),the ESTIMATE algorithm was used to calculate the infiltration level of immune cells and stromal cells(i.e.,immune score and stromal score)in each patient's tumor sample.Differentially expressed genes(DEGs)between the high immune/stromal score group and the low immune/stromal score group were then screened,and the biological processes and signaling pathways involved in these DEGs were analyzed using bioinformatics.Univariate cox analysis was then used to screen for prognostic DEGs,and these prognostic genes were verified by four other HCC cohorts,and Least absolute shrinkage and selection operator(LASSO)regression was used to further screen for key prognostic genes.Finally,these key genes were used to construct prognostic risk scores and prognostic models.The discrimination and consistency of the prognostic model was evaluated by C index and calibration curve.ResultsHCC patients with a high immune/stromal score had better survival benefits than patients with a low score.A total of 899 DEGs were identified and found to be involved in immune responses and extracellular matrices,147 of which were associated with overall survival.Subsequently,53 of 147 survival-related DEGs were validated in additional cohorts.Finally,ten genes(STSL2,TMC5,DOK5,RASGRP2,NLRC3,KLRB1,CD5 L,CFHR3,ADH1 C,and UGT2B15)were identified as key prognostic genes and used to build a prognostic model,which shows a good performance in predicting overall survival.ConclusionsThis study identified key prognostic genes related to HCC microenvironment and provided some valuable directions for finding molecular targets for HCC therapy and biomarkers for prognosis prediction.On the other hand,the prognostic model constructed may help clinicians to implement risk stratification and personalized management of cancer patients and guide subsequent treatment.