原发性肝癌患者的临床结局与治疗反应预测模型：基于失巢凋亡和免疫基因

打开文本图片集

Abstract： Objective Toestablisha prognostic modelfor primary liver cancer （PLC）using bioinformatics methods.Methods Basedon the data from 404 patients in the Cancer Genome Atlas （TCGA）database，we constructed a prognostic model integratingthedierentiallyexpressedgenes，anoikisandimmune-relatedgenes （DAs）usinguivariateCoxregreioand theLASSO-Cox approach.Thepredictiveabilityof the model was evaluatedusing Kaplan-Meiermethodand receiveroperating characteristiccurves，andanomogramwasdevelopedtofacilitateitsclinicalaplications.Genesetenrichment analysis （GSEA）wasperformedtoexploretheassociatedpathwaysandrelationshipbetweentheDAIsandthetumor immune microenvironment，and the half-maximal inhibitory concentration （IC50） of liver cancer drugs was calculated using the "pRRophetic"Rpackage.WealsodetectedtheexpressionofSEMA7Ainpairedtumorandadjacenttissesfromlivercancer patients.Results Weconstructedandvalidatedaprognosticmodelbasedon7AIs（NR4A3，EMA7A，IL11AR，C5，GF andSPP1），andobainedconsistentresultsinboththeTCGAtrainingcohortandGEOvalidationcohort （GSE14520），wrethe patients in the low-risk group were characterized by more favorable clinical outcomes and immune status.Byintegrating this prognostic signature withclinicalinformation，acompositenomogram was generated.Somatic mutationanalysisshowed that TTN，TP53，andC1mutaionsaccountedforthelargestproportionoftotalmutationsandtepatientsitelow-isklow-TMBgrouphadhighersurvivalrate.Drug sensitivityanalysis revealed diferences insensitivitytochemotherapeutic agents between high-and low-riskgroupsandbetweenTP53 mutationsand non-mutations.Inclinicaltissue specimens， SEMA7Aexpressionwassignificantlyhigher inlivercancertissesthanintheadjacenttisses.Concusions Weestablisheda newprognosticmodelbasedonDAIsforpredictingclinicaloutcomesandtherapeuticresponseofpatients withprimaryliver cancer.

Keywords： liver cancer;anoikis;immune-related genes; prognosismodel; bioinformatics

原发性肝癌（PLC）是全球常见的消化系统恶性肿瘤，其发病率和死亡率居高不下。（剩余16352字）