Construction and validation of a prognosis predicting model for thyroid carcinoma patients based on disulfidptosis-related lncRNA
Objective To explore the effect of disulfidptosis-related long non-coding RNA(DRLncs)expression on prognosis of thyroid carcinoma(THCA)patients,and to construct a prognosis prediction model based on DRLncs.Methods Sample data of THCA were obtained from The Cancer Genome Atlas(TCGA)database.Thirty-three disulfidptosis-related genes(DRGs)were retrieved from published articles.The obtained 505 THCA samples(total sample cohort)were randomly divided into a training group(253 cases)and a validation group(252 cases).The association of DRLncs with the overall survival of patients in the training group was analyzed with univariate Cox regression analysis.Least absolute shrinkage and selection operator(LASSO)regression were used to determine the DRLncs with the least deviation.The influcing factors of THCA prognosis were analyzed with multivariate Cox regression,with which a prognostic model for THCA DRLncs was constructed.According to the median risk score 3 cohorts of patients were divided into high-risk and low-risk groups.The efficacy of the prognostic model was evaluated with ROC curve and the indepent risk factors affecting patient prognosis were identified by univarate and multivariate Cox regression analysis.In addition,gene ontology(GO)function and Kyoto encyclopedia of genes and genomes(KEGG)signaling pathway enrichment analysis,immune checkpoint and related function analysis,tumor mutation burden analysis,and drug sensitivity analysis were performed on the high-and low-risk groups of the total sample cohort.Results A total of 1 488 long non-coding RNAs were identified through co-expression analysis,associated with 22 known DRGs.Univariate Cox regression revealed 41 DRLncs related to the prognosis of THCA.LASSO regression and multivariate Cox regression analysis showed that 8 DRLncs were associated with the prognosis of THCA patients,which were used to construct the prognostic model.Survival analysis revealed that in all 3 cohorts,the survival probability of the high-risk group was significantly lower than that of the low-risk group(all P＜0.05).Age and risk score were independent risk factors for predicting patient prognosis of THCA.The ROC curve indicated that the AUCs of this model for predicting the 3-and 5-year survival periods of patients were 0.783 and 0.864 respectively.A total of 255 differentially expressed genes between high-and low-risk groups in total sample cohort were identified,which were exhibited significant enrichment in nine functional categories,including copper ion stress response,presynaptic activity,neuropeptide hormone activity,as well as ten additional functions such as neuroactive ligand-receptor interactions.In the high-risk group,13 immune-related functions including promoting inflammation,type Ⅰ interferon,and quasi-inflammation were inhibited.In terms of drug sensitivity,patients in the high-risk group were more sensitive to axitinib,BI-2536,and ribociclib than patients in the low-risk group.Conclusion Eight DRLncs related to the prognosis of THCA have been identified by bioinformatics analyses,and the constructed risk prediction model can effectively predict the prognosis of patients with THCA.
万方数据
维普
