基于乙醇相关基因构建肝细胞癌预后风险模型OA

Objective To construct a prognostic risk model for hepatocellular carcinoma(HCC)based on the expressions of ethanol-re-lated genes.Methods The data of 335 HCC patients and 61 healthy controls in The Cancer Genome Atlas Liver Hepatocellular Carcinoma(TCGA-LIHC)dataset were standardized and subjected to differential analysis,identifying 4 362 differentially expressed genes(DEGs).These genes were compared with ethanol-related genes from the Comparative Toxicogenomics Database(CTD),yielding 157 overlapping genes associated with ethanol and HCC.Univariate Cox and LASSO regression analyses were used to identify genes related to the over-all survival(OS)of HCC patients,and a prognostic risk model was constructed.A cohort of 221 HCC patients was acquired from the Gene Expression Omnibus(GEO)database to serve as an external validation set.Univariate and multivariate Cox regression analyses were per-formed to identify independent prognostic factors for OS.Kaplan-Meier and time-dependent receiver operating characteristic(timeROC)curves were used to assess the model's prediction accuracy.The abundance differences of immune cells were analyzed using CIBERSORT,and correlations between the selected genes and immune cells were examined.Results Using Cox and LASSO regression analyses,17 ethanol-related genes associated with the OS of HCC patients were identified,including pregnancy zone protein(PZP),abnormal spin-dle microtubule assembly(ASPM),Apelin(APLN),olfactomedin like 2B(OLFML2B),cadherin EGF LAG seven-pass G-type receptor 3(CELSR3),microtubule-associated protein tau(MAPT),growth hormone receptor(GHR),Harvey rat sarcoma viral oncogene homolog(HRAS),G protein subunit alpha Z(GNAZ),fatty acid binding protein 5(FABP5),ATP citrate lyase(ACLY),DnaJ heat shock protein family(Hsp40)member C6(DNAJC6),stratifin(SFN),ring finger protein 157(RNF157),aldo-keto reductase family 1 member B10(AKR1B10),solute carrier family 2 member 2(SLC2A2),and ubiquitin C-terminal hydrolase L1(UCHL1).A prognostic risk model for HCC was con-structed based on the expressions of the 17 genes,and risk scores were calculated.Patients were divided into high-and low-risk groups based on the median risk score,with the low-risk group showing better OS in both TCGA and GEO databases(both P＜0.01).Risk score was an independent prognostic factor for the OS of HCC patients(P＜0.01).timeROC curves showed risk score had good predictive accura-cy for predicting 1-,3-,and 5-year survival of HCC patients in both TCGA and GEO databases[all area under the curve(AUC)＞0.65].Immune infiltration analysis indicated that the high-risk group was characterized by higher abundances of resting natural killer(NK)cells,macrophages M0,and T regulatory cells,whereas the low-risk group was predominantly composed of eosinophils,resting mast cells,and resting memory CD4+T cells(all P＜0.05).Correlations were found between the 17 ethanol-related genes and immune cells,especially rest-ing NK cells and M2 macrophages(both P＜0.05).Risk scores correlated positively with M0 macrophages(r=0.55,P＜0.05)and regulatory T cells(r=0.48,P＜0.05),and negatively with resting memory CD4+T cells(r=-0.40,P＜0.05).Conclusions This study developed a prog-nostic model based on the expressions of 17 ethanol-related genes for HCC patients.The model's risk score is an independent prognostic factor for HCC,and demonstrates strong predictive value for the OS of HCC.