Objective: Upper tract urothelial carcinoma(UTUC)accounts for 5-10% of all urothelial neoplasms.Due to the lack of early and specific symptoms and effective diagnostic tools,it is mostly diagnosed in late stages and has a poor prognosis.Therefore,a sensitive and non-invasive diagnostic method for UTUC is urgently needed.In this study,we aimed to detect urine DNA mutations(TERT、FGFR3)and NRN1 methylation,to find an optimal model for their combination with clinical risk factors in the diagnosis of UTUC,and to investigate the diagnostic value of urine DNA mutations(TERT 、 FGFR3)and NRN1 methylation in upper urinary tract uroepithelial carcinoma.Methods: A total of 402 patients with hematuria from 6 hospitals in Hunan,China,including 95 UTUC patients and 307 non-UTUC patients,were enrolled in this study.Collect into a group of patients with preoperative morning urine samples extracted DNA and analysis of FGFR3 and TERT mutation and NRN1 methylation,gather relevant clinical data into the group of patients at the same time,the molecular marker detection results and clinical risk factors in statistical analysis,logistic regression forecasting model analysis to predict the relationship between the variables and UTUC and optimal model is set up,Beijing queue(n=76)for external validation.Results: The AUC of NRN1 methylation combined with gene mutations(TERT and FGFR3)was 0.958(95%CI: 0.933-0.975),the sensitivity was 91.58%,and the specificity was 94.79%.The combined detection group showed good detection efficiency in UTUC and other malignant tumors(AUC=0.920)and UTUC and benign diseases(AUC=0.975).In addition,logistic regression analysis of important characteristics including age,gene mutation status(TERT,FGFR3)and NRN1 methylation level yielded an optimal model with sensitivity of93.68%,specificity of 94.44%,AUC=0.970 and NPV=98.6%.In external validation,AUC was 0.971,sensitivity 95.83%,specificity 92.31%.Conclusion: In this study,the epigenetic molecular marker NRN1 was used for the detection of UTUC for the first time,and its superior performance was found.Genetic mutations(TERT,FGFR3)and NRN1 methylation in urine DNA can be used as biomarkers for the diagnosis of UTUC,and logistic regression model can be established by combining them with clinical risk factors to evaluate the risk of UTUC in hematuria patients,providing help for auxiliary diagnosis and treatment of UTUC. |