Screening Of Tumor-Associated Antigens Based On Oncomine Database And Evaluation Of Diagnostic Value Of Autoantibodies In Lung Cancer

Lung cancer is a malignant tumor with its morbidity and mortality ranking first in China and the world.The average 5-year survival rate of patients with lung cancer is less than 15%,the postoperative survival rate in early diagnosed patients can reach 80%,so early detection and early diagnosis are very important for the treatment and prognosis of patients with lung cancer.Autoantibodies against tumor-associated antigens(TAAs)can occur in the early stages of tumorigenesis in a patient's serum,which are expected to be taken as the serum markers for the early diagnosis of lung cancer.ObjectiveThe present study aims to screen and identify lung cancer-associated antigens by the Oncomine database and the enzyme-linked immunosorbent assay(ELISA)method.This study evaluated the diagnostic value of autoantibody detection and autoantibody markers combined with traditional tumor markers in patients with lung cancer at early stage.Methods1.Screening potential tumor-associated antigens using Oncomine databaseAccording to the screening conditions,the Oncomine gene data were in-depth excavated,and genes significantly dysregulated in non-small cell lung cancer tissues were selected,compared with normal control tissues.Genes significantly dysregulated were selected according to the P-value and expression differential multiple selection criteria.Finally,16 candidate lung cancer-associated antigens were selected.2.Detection the levels of 16 autoantibodies in serum by indirect ELISA in the testing setIndirect ELISA was used to detect the 16 lung cancer-associated autoantibodies in 184 patients with lung cancer and 184 healthy individuals.The levels of autoantibodies in lung cancer and normal human serum were compared,the OD value with a specificity more than 90% and the highest Youden index was used as the cut-off value.Chi-square test was used to compare the differences of the positive rates of autoantibodies between the lung cancer group and the normal control group.The ROC curve was used to evaluate the diagnostic value.3.Validation set with a large scale of serum samples was used to further verify the preliminarily determined antigens446 patients with lung cancer,446 healthy individuals and 119 patients with benign pulmonary disease were selected and tested for lung cancer-associated antigens identified in the previous step.One-way analysis of variance and SNK method were used to compare the differences of the autoantibody levels among three groups.Chi-square test was used to compare the differences of the positive rates of autoantibodies between the lung cancer group and the normal control group.4.Evaluation of diagnostic value of autoantibody detection for lung cancerDiagnostic tests were used to evaluate the diagnostic authenticity and reliabilityof autoantibody detection for lung cancer,and the relationship between autoantibodies and clinical features of lung cancer patients was analyzed.The parallel combination of autoantibody markers and traditional tumor markers was used to evaluate the diagnostic value of combined detection for early lung cancer.Results1.Sixteen genes highly expressed in lung cancer tissues were selected from the Oncomine database and the corresponding proteins were used as candidate lung cancer-associated antigens,which were HMGB3,MMP12,GREM1,ZWINT,NUSAP1,RRM2,SULF1,CEP55,CDC20,SPP1,CCNB1,GOLM1,PRDX4,COL11A1,PCLAF,MDK.2.The results of ELISA showed that except for CCNB1,COL11A1,GOLM1,PCLAF,PRDX4 and MDK,the levels of autoantibodies against the other 10 TAAs(HMGB3,MMP12,GREM1,ZWINT,NUSAP1,RRM2,SULF1,CEP55,CDC20,SPP1)in the serum of patients with lung cancer were significiantly higher than normal control group(P<0.05).ROC curve analysis showed that the AUC results of the 5 autoantibodies against HMGB3,MMP12,GREM1,ZWINT and NUSAP1 were all greater than 0.7.3.According to the positive rate of the autoantibodies in lung cancer group and the AUC,the above five lung cancer related antigens were selected for further verification.ELISA results showed that the levels of the five autoantibodies were highly expressed in the lung cancer group,compared with the normal control group(P<0.05).Except for NUSAP1,the levels of the other four autoantibodies in the lung cancer group were higher than those in the lung benign disease group(P<0.05).The positive rates of the five autoantibodies in the lung cancer group were significantly higher than that in the normal control group(P<0.05).Among them,HMGB3 autoantibodies have the highest sensitivity of 34.08%.4.Clinical characteristics analysis showed that the positive rate of autoantibody against HMGB3 in patients with lung cancer at early stage was significantly higher(I/II)than that in patients at advanced stage(III/IV)(P<0.05).The positive rate ofautoantibody against MMP12 in the elderly group(?60 years)was significantly higher than that in the non-elderly group(<60 years)(P<0.05).Other factors,such as pathological type,family history and distant metastasis,had no effect on the positive rate in patients with lung cancer(P>0.05).5.The positive rates of CYFRA21-1,CEA and CA125 in patients with lung cancer at early stage were 35.3%,28.6% and 20.0%,respectively.The detection rate of lung cancer increased to 70.6%,61.9% and 60.0% when autoantibody against HMGB3 was combined with CYFRA21-1,CEA and CA125 in parallel,respectively.Conclusions1.This study firstly found that the autoantibodies agianst HMGB3,MMP12,GREM1,ZWINT and NUSAP1 could be used as markers for the diagnosis of lung cancer.Among them,autoantibody against HMGB3 has potential value for the diagnosis of patients with lung cancer at early stage.2.Combined detection of HMGB3 autoantibodies and traditional tumor markers can significantly increase the detection rate of early-stage lung cancer.