Gene Expression Signatures Of Peripheral Blood As The Biomarkers Of Stomach Cancer And Cervical Cancer

PurposeTo screen potential biomarkers for stomach cancer and cervical cancer from gene expression profiles of peripheral blood of the patients.MethodsPeripheral blood samples were collected into PAXgene tubes from 21 stomach cancer(SC)patients,11 cervical cancer(CC)patients,21 cervical intraepithelial neoplasia(CIN)patients and 36 healthy controls(HC).The extracted mRNA from samples was used for high-throughput sequencing.Gene expression profiles of samples were analysed using bioinformatics methods.Differentially expressed genes(DEGs)were identified between SC and HC,and between CC,CIN and HC.Biological process analysis of Gene Ontology(GO)and Kyoto encyclopedia of genes and genomes(KEGG)pathway enrichment analysis for DEGs were carried out using DAVID online analysis platform.Candidate DEGs were screened for SC and CC detection.Peripheral blood samples were collected into PAXgene tubes from 25 SC patients,83 CC patients,32 CIN patients and 86 HC.Primers were designed for candidate DEGs.The expression levels of candidate DEGs in new samples were detected using quantitative real-time polymerase chain reaction(qRT-PCR)technology,and genes with significant differences in expression were screened.Receiver operating characteristic(ROC)curve and Logistic regression analysis were performed to evaluate their cancer detection potential,and detection models of SC and CC were initially established.ResultsWe identified 7263 DEGs between SC patients and HC.Their biological processes and pathways included immune response,inflammatory response,apoptotic process,T cell receptor signaling pathway and interferon-gamma-mediated signaling pathway.We identified 1008 DEGs between CC patients and CIN patients.Their biological processes and pathways included immune response,apoptotic process,signal transduction,cell killing and inflammatory response.We identified 381 DEGs between CIN patients and HC.Their biological processes and pathways included immune response,T cell activation,defense response,inflammatory response and interferon-gamma-mediated signaling pathway.118 DEGs were identified between CC patients and HC,and had no definite biological process and pathway.30 DEGs between SC patients and HC were screened and validated in 73 new samples using qRT-PCR.3 genes(C1QB,MMP9 and SMIM1)had significant differences.Logistic regression analysis revealed that the combination of these 3 genes were able to detect SC patients with a sensitivity of 92%,a specificity of 72.92% and an area under curve(AUC)of 0.8717.59 DEGs among CC patients,CIN patients and HC were screened and validated in 161 new samples using qRT-PCR.9 genes had significant differences between at least 2 groups of samples.Logistic regression analysis revealed that the combination of 4 genes(CIR1,GPR84,GZMB and NDUFA1)were able to detect cervical lesions(including CC and CIN)with a sensitivity of 82.61%,a specificity of 87.83% and an AUC of 0.8981.In addition,the combination of 3 genes(AGAP1,GZMB and NUAK1)were able to distinguish CC patients and CIN patients with a sensitivity of 53.13%,a specificity of 96.39% and an AUC of 0.7786.ConclusionThe gene expression profiles of peripheral blood in SC patients and CC patients were significantly different from that in HC.C1 QB,MMP9 and SMIM1 showed potential value in detecting SC.CIR1,GPR84,GZMB and NDUFA1 showed potential value in detecting cervical lesions.AGAP1,GZMB and NUAK1 showed potential value in distinguishing CC patients and CIN patients.Screening DEGs from gene expression profiles of peripheral blood was a promising non-invasive detection method for cancer.