Construction And Verification Of A LncRNAs Prognostic Risk Model In Hepatocellular Carcinoma

Objective: Despite the existence of various therapies for HCC,hepatocellular carcinoma still has a high mortality rate and recurrence rate,which seriously endangers people's health.Many studies have shown that long noncoding RNAs(lncRNA)are closely associated with the occurrence and development of various tumors and have the potential to be prognostic markers.Some studies have reported that lncRNA-related prognostic models have been used to predict overall survival(OS)and recurrence-free survival(RFS)in patients with hepatocellular carcinoma(HCC).Moreover,cirrhosis is an important prognostic risk factors in patients with liver cancer.However,no one has constructed a prognostic lncRNA model only in patients with cirrhotic HCC.Therefore,it is necessary to screen new potential lncRNAs prognostic markers to determine the prognosis of hepatocellular carcinoma patients with cirrhosis,and to implement individualized treatment.Methods: The probe expression profile dataset(GSE14520–GPL3921)from the Gene Expression Omnibus(GEO),which included 203 cirrhotic HCC samples,was reannotated and the lncRNAs and mRNA expression dataset was obtained.The patients were randomly assigned to either the training set(n = 103)and testing set(n= 100)at a 1:1 ratio by R software package.Univariate cox regression and the least absolute shrinkage and selection operator(LASSO)model were applied to screen lncRNAs linked to the OS of cirrhotic HCC in the training set.The lncRNAs having significant correlation with OS were then selected and the multivariate Cox regression model was implemented to construct the prognostic score model.Whether or not this model was related to RFS in the training set was simultaneously determined.The testing set was used to validate the lncRNA risk score model.A risk score based on the lncRNA signature was used for stratified analysis of different clinical features to test their prognostic performance.The prognostic lncRNAs related protein genes were identified by the co-expression matrix of lncRNA-mRNA,and the function of these lncRNAs was predicted through the enrichment of these co-expression genes.Results: 1.A total of 203 cases of hepatocellular carcinoma with cirrhosis were put in the study,including 103 cases in the training cohort and 100 cases in the validation cohort.There was no significant difference in the clinical case characteristics between the two groups.Four lncRNAs were selected to construct the prognostic risk score model,including AC093797.1,POLR2J4,AL121748.1 and AL162231.4.Risk scores of all patients in the training cohort were calculated,with a median risk score of 0.968 as the cut-off value.Patients with a risk score greater than 0.968 were in the high-risk group,otherwise the low-risk group.The risk score model was able to predict the OS and RFS of cirrhotic HCC in the training cohort and was verified in the validation cohort.2.The results of subgroup analysis in both the training cohort and the validation cohort showed that the prognostic risk score model had a good prognostic risk stratification effect for OS and RFS in patients with TNM(Tumor,Node,Metastasis system)stages I-II,Barcelona Clinic Liver Cancer(BCLC)stages 0-A,and solitary tumors.3.Functional enrichment showed that four lncRNAs may be involved in the metabolism of amino acid,lipid,glucose metabolic pathways of hepatocellular carcinoma.Conclusion: 1.We identified four potential lncRNAs biomarkers associated with the prognosis of cirrhotic HCC and constructed a risk scoring model.2.This prognostic risk model may have the ability to stratify the OS and RFS of HCC patients with cirrhosis in BCLC stage 0-A,isolated tumor,and TNM stage I-II.3.These four new lncRNAs may also serve as potential therapeutic targets.