| Background:Lung cancer is the malignant disease with the highest cancer mortality rate in the world.CT(Computed Tomography)examination and detection of serum carcinoembryonic antigen levels are commonly used for screening,early detection and dynamic monitoring of lung cancer progression.In this study,we firstly evaluated the clinical application of these two methods,and then screened and analyzed patients’ driver genes based on the whole exome sequencing results of enrolled patients,and investigated the relationship between the rare mutated B3GNT3 gene and immune infiltration of lung adenocarcinoma,aiming to provide a new theoretical basis for the precise treatment of lung adenocarcinoma.Methods:In this study,we collected clinical data from 87 patients with pathologically confirmed non-small cell lung cancer who attended Zhujiang Hospital of Southern Medical University from January 2016 to December 2018 consecutively.We analyzed the relationship between serum carcinoembryonic antigen(CEA)levels and long-term prognosis.87 patients were excluded from cases of central lung cancer,combined with obstructive pneumonia and pleural effusion,etc.The chest CT image features of 66 patients eligible for CT image analysis were analyzed by applying Mimics(Materialise’s interactive medical image control system)software which was used to evaluate the CT image characteristics of patients with non-early stage non-small cell lung cancer,with a particular focus on the inner and peripheral areas of the lesion.Whole exon sequencing(abbreviated as WES)was performed on tumor and paracancerous tissue specimens from patients who had undergone surgery.For non-surgical advanced NSCLC patients included in the study,and for all patients,peripheral venous blood was drawn as needed for subsequent treatment or follow-up,and circulating tumor DNA(ctDNA)was isolated and sequenced by WES.In this study,the WES results were used to map the mutations in non-small cell lung cancer and to analyze the driver genes in the tumors of NSCLC patients.We also performed WES sequencing on tumor samples obtained from five patients at different times and performed clonal evolutionary analysis based on the WES results to further characterize the heterogeneity within NSCLC tumors.Finally,we performed driver gene analysis in the enrolled lung adenocarcinoma patients and analyzed the interrelationship between B3GNT3 and immune infiltration of lung adenocarcinoma by bioinformatics analysis and cytological experiments for the detected B3GNT3 gene,which has an unclear mechanism of influence on lung adenocarcinoma pathogenesis.Result:A total of 87 patients with NSCLC were included in this study,and all patients had serum test results of CEA;according to the inclusion and exclusion criteria of CT image analysis,after excluding 21 patients with lung cancer,the CT image data of 66 patients were included for study.All patients underwent WES sequencing of tumor specimens,and we removed samples with high sequencing duplication rate and tumor samples with sequencing depth less than 100X,and obtained a total of 174 WES datasets.We detected a total of 6785 mutated genes,including low frequency mutated genes such as B3GNT3、CLDN1、CLDN2.1.Among the 87 NSCLC patients,61 had CEA>5.0 μg/L and 26 had CEA≤5.0μg/L.After Kaplan-Meier curve survival analysis,the results confirmed that the P value>0.05,and the difference in overall survival time between the two groups was not statistically significant.2.We obtained 99 groups of standard lung CT Imaging data of "Digital Imaging and Communications in Medicine(Dicom)" for peripheral lung cancer patients.Mimics software was used to conduct 3D reconstruction of 99 groups of CT image data collected to obtain 3D images of nodular or mass NSCLC lesions.CT images were imported into ImageJ software from the layer with the largest tumor diameter and the least interference factors.The tumor interface was automatically identified by computer,and the areas of interest were automatically separated for correction.In our study,there was no statistically significant difference in individual CT image features between early(stage Ⅰ)and non-early(stage Ⅱ-Ⅳ)NSCLC patients.In the combined analysis of chest CT features(burr sign,lobectomy sign,pleural draft sign,vacuole sign,air bronchus,etc.),there was a significant increase in the incidence of two or more CT image features in stage Ⅱ-Ⅳ patients(P=0.01372).3.We performed tumor purity and ploidy analysis on the WES results of tumor tissue samples and ctDNA,and found considerable concordance between them.After analyzing the filtered mutation results,we found that the tumor tissue or ctDNA samples had the highest number of missense mutations compared to the paracancerous tissue and were mostly single nucleotide polymorphisms(SNPs),with the highest percentage of base C mutations to base A or T.In total,we detected 6785 genes with mutations.Among them,TP53,EGFR,and KRAS genes are the recognized driver genes of lung adenocarcinoma,and we also detected rare mutated genes such as B3GNT3,whose role needs to be further investigated.4.We performed WES sequencing after obtaining tumor samples from 5 patients at different time points.Based on the analysis of the WES results of these 5 patients,we found that for secondary cloning,the results of detection using ctDNA were highly consistent with the results of clonal analysis of tissue samples.5.In the lung adenocarcinoma patients enrolled in this study,the incidence of B3GNT3 gene mutation was 1.33%,which was generally consistent with the incidence in previous studies(1.4%).The results of Kaplan-Meier curve survival analysis showed that the overall survival rate of patients with high expression of B3GNT3 gene was lower than that of patients with low expression.Serum cells were enriched in tumor samples from the low B3GNT3 group,whereas M0 macrophages were enriched in tumor samples from the high B3GNT3 group.B3GNT3 may be associated with CTLA-4 and CD276(B7-H3).We performed a transwell assay and found that knockdown of B3GNT3 significantly inhibited the invasive ability of two lung adenocarcinoma cell lines,A549 and PC9.Conclusions:Conventional examinations and tests for non-small cell lung cancer,including CT image characterization and CEA testing,have significant limitations in predicting the effectiveness of NSCLC treatment or assessing patient prognosis.In this study,we mapped gene mutations in 87 NSCLC patients.The WES technique combines efficiency,accuracy,and economy at the same time.The WES technique can monitor the changes of tumor subclones in NSCLC patients,and the use of ctDNA to detect the changes of tumor subclones in patients is highly consistent with the clonal analysis of tissue samples.The WES technique derived from ctDNA can be used as a prognostic assessment of patients and WES technology derived from ctDNA can be used as a tool for patient prognosis assessment and tumor surveillance.WES technology can detect genes with relatively low mutation rates,including B3GNT3.Our study demonstrated that B3GNT3 is highly associated with immune infiltration of lung adenocarcinoma and is an important biomarker for prognosis of LUAD. |