| ObjectiveOvarian cancer(OC)is the most lethal gynecologic malignancy.Due to the lack of typical symptoms and the absence of specific and effective screening tools,the survival rate of OC patients is low,with a 5-year survival rate of less than 30%.Among the various pathological types of OC,Epithelial Ovarian Cancer(EOC)is the most common,accounting for approximately 80%of ovarian malignancies.Studies have shown that the detection and analysis of circulating tumor cells(CTCs)are important for the early diagnosis,efficacy monitoring and prognosis assessment of EOC.In this study,we proposed to construct a purification method for EOC CTCs in the blood of ovarian cancer patients and quantify the characteristic mRNAs of EOC-derived CTCs.By efficiently capturing epithelial phenotype and mixed epithelial-mesenchymal phenotype CTCs in peripheral blood samples from EOC patients and molecularly analyzing the characteristic mRNA genes embedded in CTCs,a highly sensitive and specific diagnostic model was established,with the aim of achieving timely diagnosis and efficacy assessment of EOC.MethodsIn this study,we achieved affinity capture of epithelial phenotype and mixed epithelial-mesenchymal phenotype OC CTCs cells by multi-antibody system(antiEpCAM,anti-N-cadherin,anti-MSLN)modified magnetic nanoparticles(MNPs).Firstly,the magnetic beads were synthesized by the solvent heat method,and the non-specific adhesion of leukocytes and other components in peripheral blood was inhibited by wrapping poly carboxybetaine methacrylate(pCBMA)on the surface of the beads;secondly,streptavidin(SA)was modified by activating the carboxyl group on the CBMA polymer;finally,antibodies were modified on the surface of the beads by the recognition of biotin and SA.Based on the heterogeneity of CTCs surface marker expression,this study explored the optimal amount of antibody modification on immunomagnetic beads using three ovarian cancer tumor cell lines,OVCAR-3(epithelial phenotype CTCs model),SK-OV-3 and HO-8910(both are mixed epithelial-mesenchymal phenotype CTCs model).The OVCAR-3 model cells before and after capture were characterized by field emission environment scanning electron microscopy(SEM).In this study,the target genes with high expression in EOC tissues were screened by the Tumor Genome Atlas(TCGA),Genome-Tissue Expression Database(GTEx),GSE4122 dataset from the Gene Expression Omnibus(GEO)and Human Protein Data Bank(HPA)immune cell dataset,and validated by extracting RNA from OVCAR-3 cells,SK-OV-3 cells and healthy female peripheral blood mononuclear cells(PBMCs).RNA from OVCAR-3 cells,SK-OV-3 cells and healthy female peripheral blood mononuclear cells(PBMCs)was extracted,reverse transcribed into cDNA products,and validated by droplet digital PCR(ddPCR).In this study,17 peripheral blood samples were collected from patients with EOC before treatment,and 30 peripheral blood samples were collected from a subset of these patients during the course of treatment,and 30 peripheral blood samples were also collected from patients with benign ovarian tumors.EOC CTCs in clinical samples were captured by multi-antibody modified MNPs,and the expression levels of the target genes were subsequently detected by quantifying the characteristic mRNAs of EOC-derived CTCs,and the characteristic genes with excellent diagnostic performance were screened out by ROC(Receiver Operating Characteristic)curve,and then a multi-factor logistic regression analysis was performed to construct an EOC CTC RNA detection model.The diagnostic performance of the EOC CTC RNA assay model was evaluated,and the ROC curve was plotted to analyze the sensitivity,specificity,and accuracy of the model,and compared with the commonly used serum tumor biomarker CA125.In addition,the model was used to analyze peripheral blood samples from EOC patients before surgical treatment and before chemotherapy cycle to evaluate the performance of the model in terms of treatment response and prognosis monitoring.Results1.In this study,the multi-antibody modified MNPs were prepared with high specific capture performance and anti-non-specific adhesion ability.The capture performance of the multi-antibody system was investigated using OC cell lines OVCAR-3,SK-OV-3 and HO-8910.The results showed that the optimal cell capture efficiency was achieved when the beads were modified with single antibodies of EpCAM,N-cadherin and MSLN at the dosages of 200 ng,100 ng and 100 ng,respectively;the capture efficiency of the multiantibody modified MNPs modified with the optimal dosages of the three antibodies for OVCAR-3,SK-OV-3 and HO-8910 cells reached 82.69%± 2,82.64%± 2.79%and 87.44%±2.85%,respectively.What’s more,the multi-antibody modified MNPs showed good anti-adhesion performance against PBMCs in peripheral blood with a non-specific adhesion rate of 0.79%±0.07%.2.Nine OC characteristic high expression gene mRNAs,i.e.,BCAM,KLK10,MSLN,LYPD1,MUC16,SCGB2A1,WFDC2,KLK7 and FOLR1 mRNA,screened from the public database,which were highly expressed in ovarian cancer tissues and cell lines,and low in normal ovarian tissues and healthy female PBMCs.The differences were statistically significant(P<0.05).Among the nine target mRNAs,BCAM,MSLN and FOLR1 had better diagnostic performance,and their areas under the ROC curves(AUROC)reached 0.85,0.89 and 0.87,respectively(P<0.0001).The mathematical formula of the EOC CTC RNA diagnostic model was obtained by multi-factor logistic regression analysis of the expression of the three mRNAs,BCAM,MSLN and FOLR1:EOC CTC Score=-3.904+0.482 × BCAM+0.314 × MSLN+1.7 × FOLR1.The gene name represents the expression of the gene.3.The sensitivity,specificity,accuracy and AUROC of the EOC CTC RNA diagnostic model reached 86.7%,87.5%,87.2%and 0.9588(P<0.0001,95%CI=0.911.00)when identifying EOC and benign ovarian tumors.The specificity,accuracy and overall diagnostic performance of this model were better than those of the commonly used serum tumor biomarker CA125(specificity:56.7%;accuracy:70.2%;AUROC:0.7539),and the difference was statistically significant(P<0.0001).4.The EOC CTC Score was significantly lower in patients with EOC during treatment(n=17)than in patients with untreated EOC(n=30,P<0.0001).The EOC CTC Scores of the four EOC patients before surgery and before the chemotherapy cycle showed a progressive decrease with treatment measures,and the trend was consistent with the CA125,HE4 and ROMA index.The performance of the EOC CTC RNA diagnostic model in patients with EOC-36 and EOC-26 suggested that the model has good potential for real-time monitoring of patient progression and treatment evaluation.ConclusionIn summary,this study formed a method to quantify the characteristic mRNAs of EOC-derived CTCs by combining multi-antibody system,magnetic separation technique and ddPCR detection technique,and constructed a diagnostic model of EOC CTC RNA by analyzing the molecular characteristics of mRNAs derived from EOC CTCs.The model can discriminate EOC with high specificity and sensitivity,and is expected to be a reliable aid for timey diagnosis of EOC.In addition,the model showed good stability and sensitivity in dynamically monitoring the therapeutic effect of EOC patients,which is expected to play a potential role in the efficacy monitoring and recurrence prediction of EOC patients. |
PDF Full Text Request