Expression Signatures Of Exosomal Long Non-coding RNAs In Urine Serve As Novel Non-invasive Biomarkers For Diagnosis And Recurrence Prediction Of Bladder Cancer

Objective:Exosomes released by tumor cells contain oncogenic long non-coding RNAs(IncRNAs),which can reflect the pathophysiology of tumors.Studies showed that bladder cancer(BC)cells can secrete exosomes into urine and facilitate tumor progression.Therefore,exosome-shuttled bioactive lncRNAs have aroused great interest of researchers to study whether it is possible to use exosomal lncRNAs as novel diagnostic and prognostic indicators for cancer.However,only few urinary exosome(UE)-derived lncRNAs are characterized as potential biomarkers in BC patients.In this research,we intend to establish a UE-derived lncRNA model for detection of BC by extracting and analyzing the expression signatures of UE-derived ncRNAs,and to explore the predictive value of differentially expressed exosomal lncRNAs in the recurrence of BC.Methods:1.Exosomes were extracted from urine of BC patients and healthy controls and confirmed using transmission electron microscopy(TEM),Western blotting analysis,nanoparticle tracking analysis(NTA)and flow cytometry.2.Then,eight lncRNAs,including MALAT1,PCAT-1,SPRY4-IT1,UCA1,MEG3,H19,UBC1 and TUG1,were selected as candidate moleculars according to their functional roles in tumorigenesis,development,and metastasis of BC.3.During the training phase,exosomal lncRNAs was extracted from the urine samples of 104 patients with BC and 104 healthy controls.Then,qRT-PCR was used to analyze the expressions of the above 8 candidate IncRNAs,and IncRNAs that were abnormally expressed in the urinary exosomes of BC were screened out.Finally,the relative expressions(2-??Ct)of the selected exosomal lncRNAs were put into multivariate logistic regression model to establish the diagnostic panel for BC.The diagnostic accuracy of the panel was evaluated by receiver-operating characteristic(ROC)curves.4.During the validation phase,the differentially expressed UE-derived lncRNAs were further validated in another 80 patients with BC and 80 healthy controls,and the diagnostic efficiency of the panel was measure by ROC analysis.In addition,we also compared the diagnostic performance between the panel and urine cytology for BC detection.5.To investigate the prognostic value of the differentially expressed IncRNAs for BC recurrence,a total of 80 patients with BC were followed-up,which were divided into the nonmuscle-invasive bladder cancer(NMIBC)group(50 cases)and the muscle-invasive bladder cancer(MIBC)group(30 cases).The median follow-up time was 57 months.Kaplan-Meier survival analysis was used to analyze the relationship between the differentially expressed lncRNAs and the recurrence-free survival(RFS)of BC patients in the two groups.Then,multiple variables including the selected exosomal IncRNAs,age of patients,gender of patients,tumor stage,tumor grade and lymph node metastasis were incorporated into the Cox proportional hazards regression model to find independent risk factors that could be used to predict recurrence of BC.Results:1.In the training set,the expressions of eight candidate lncRNAs in 104 BC patients and 104 healthy controls were assessed by qRT-PCR.MALAT1,PCAT-1 and SPRY4-IT1 were significantly up-regulated in BC patients compared with the healthy controls(all at P<0.001).The diagnostic accuracy of MALAT1,PCAT-1 and SPRY4-IT1,measured by AUC,was 0.844,0.832 and 0.760,respectively.A three-IncRNA panel based on this result was established for BC diagnosis:Logit(P)=0.6577-0.0695×MALAT1-0.0686×PCAT-1-0.0015×SPRY4-IT1,and the diagnostic performance of the established three-IncRNA panel was 0.854(95%CI=0.799-0.899,sensitivity = 70.2%and specificity =85.6%).2.These three lncRNAs were further verified in an independent validation set,containing 80 BC patients and 80 healthy controls.Similarly,the dysregulated expression trend was consistent between the training set and the validation set,revealing no significant difference.The corresponding AUCs of the three lncRNAs(MALAT1,PCAT-1 and SPRY4-IT1)were 0.785,0.810 and 0.799,respectively.The AUC of the three-lcRNA panel for BC detection was 0.813(95%CI=0.744-0.870,sensitivity =62.5%and specificity = 85.0%),which was significantly higher than that of the urine cytology(AUC=0.619,95%CI =0.539-0.694,sensitivity = 25%and specificity=98.7%,P<0.001).3.In the NMIBC group,Kaplan-Meier survival analysis showed that patients with up-regulated MALAT1 and PCAT-1 had a significantly lower RFS compared with their corresponding counterparts(P =0.002 and P<0.001,respectively).Multivariate analysis revealed that exosomal IncRNA PCAT-1 and tumor stage were independent indicators for predicting recurrence of NMIBC(P =0.018 and P = 0.036,respectively).However,none of the three dysregulated lncRNAs were correlated with the recurrence of MIBC patients(all at P>0.05).Conclusions:1.We successfully established a three-IncRNA panel for BC diagnosis through extracting and analyzing UE-derived lncRNAs.2.We identified that exosomal PCAT-1 could act as an independent risk factor for recurrence prediction of NMIBC.3.Our study provides a new strategy to explore effective and reliable urinary biomarkers for detection and recurrence prediction of BC.