Clinical Application Of Digital PCR Technology For Diagnosis Of Urothelial Carcinoma

Background and purpose:digital PCR,with its high sensitivity and precision,can detect target gene mutations with low nucleic acid content in liquid biopsies in clinical research on tumors.It has clinical value for early diagnosis of cancer patients,real-time monitoring of patient treatment efficacy,and prediction of tumor recurrence and metastasis,thereby achieving personalized and precise treatment for cancer patients.In this study,we first established a stable digital PCR reaction system by optimizing it using postoperative tissue specimens with high tumor purity from liver cancer patients.We also validated the use of digital PCR to detect hot mutation sites to improve the prognostic efficacy evaluation of liver cancer patients.To further apply digital PCR to clinical practice,alleviate the trauma caused by invasive detection methods for urinary tract cancer patients,and assist in improving the diagnostic efficacy of urinary tract cancer through urinary cytology examination,we collected urine samples from urinary tract cancer patients in a noninvasive manner and conducted digital PCR detection of their hot mutation sites,exploring the diagnostic efficacy of using digital PCR technology combined with urinary cytology examination for urinary tract cancer.Method:Firstly,we extracted DNA from 66 postoperative tissue samples of liver cancer with high tumor purity.TERT promoter mutations detected by Sanger sequencing and digital PCR.To establish the interpretation criteria of digital PCR for mutation samples,we established the limit of blank(LoB)and limit of detection(LoD),and continuously adjusted the type and concentration of special chemical reagents to optimize the reaction conditions of digital PCR.The consistency between Sanger sequencing and digital PCR was compared using Cohen’s Kappa index.Kaplan-Meier curves and Cox proportional hazard models were plotted based on prognostic data,A multivariate Cox regression was used to construct Nomogram models for OS and DFS,and column charts for predicting OS and DFS were generated using the "rms" package in R software.Time-dependent receiver operating characteristic curves(ROC),calibration curves,and decision curve analysis(DCA)were used to verify the accuracy and clinical effectiveness of the prediction model.The study included 214 patients with NCC cytology-sent urine,14 patients with hematuria from Henan Provincial People’s Hospital,and 48 normal controls.Urine samples were collected and urinary cfDNA was extracted.Digital PCR was used to detect TERT C228T/C250T mutations in urine cfDNA.When setting the criteria for judgment,LoB and LoD values were established.The statistically significant results of the univariate analysis were included in a multiple logistic regression model to identify independent factors affecting the diagnosis of urothelial carcinoma and clinical characteristics influencing TERT mutations in urothelial carcinoma patients,and a predictive model was established.The diagnostic efficacy of the predictive model was verified using ROC curves and calibration curves.In addition,urine samples were collected from 27 patients with UC and urinary cfDNA was extracted.Droplet digital Polymerase Chain Reaction(ddPCR)was used to detect FGFR3 S249C mutations in urine cfDNA,ROC analysis was employed to evaluate the diagnostic efficacy of FGFR3 S249C mutations in urinary cfDNA for patients with unclear UC diagnosis based on urine cytology.Results:In hepatocellular carcinoma,39.39%(26/66)of TERT C228T mutations were detected by Sanger sequencing.The threshold for determining mutation status by digital PCR was 0.55 cp/μL,and 45.45%(30/66)of the samples were found to carry TERT C228T mutations.The two methods had good agreement(concordance=93.939,Kappa=0.876).Digital PCR had a sensitivity of 100%and a specificity of 90%for detecting TERT C228T mutations.Compared with wild-type TERT C228T,the survival rate of HBV-related HCC patients with mutant-type TERT C228T was decreased,and the risk of recurrence was increased.Predictive models for overall survival(OS)and disease-free survival(DFS)were constructed for HBV-related HCC patients,with C indices of 0.7651 and 0.6899,respectively.Using digital PCR to detect urinary cfDNA in urothelial carcinoma patients,the detection limits for TERT C228T and C250T were 0.546 cp/μL and 0.545 cp/μL,respectively.The mutation rate of TERT C228T was 30.8%(66/214)in NCC patients,0%in the blood and urine control groups at Henan People’s Hospital,while the mutation rate of TERT C250T was 9.3%(20/214)in NCC patients,7.14%(1/14)in the blood and urine control group at Henan People’s Hospital,and 0%in the normal control group.The mutation rates of TERT C228T/C250T in the three groups were 36.4%(78/214),7.14%(1/14),and 0%,respectively.The mutation status of TERT C228T or C250T was associated with patient age(P=0.049),cytology classification(P=0.001),borderline cytology(P<0.001),follow-up results(P<0.001),and infiltration(P=0.019).The mutation copy number concentration of TERT C228T/C250T increased with cytology classification,and was higher in the muscle-invasive UC and follow-up result UC groups.Multivariate logistic regression showed that age(P=0.013),TERT C228T/C250T(P<0.001),and suspicious cancer cells(P<0.001)were independent factors affecting UC.The detection rates of UC with unclear cytology and TERT C228T/C250T mutation and without mutation were 30.5%(50/164)and 22.6%(37/164),respectively.The detection rate of UC with unclear cytology can be improved by the mutation of TERT C228T/C250T in urinary shedding cells.The detection rate of FGFR3 S249C mutation in cfDNA from urine samples of UC patients is 25.9%(7/27).According to the Paris system(TPS)classification of urinary cytology,the detection rate of FGFR3 S249C mutation in high-grade urothelial carcinoma(HGUC)is 18.8%(3/16),50%(4/8)in suspicious HGUC(SHGUC),and 0(0/3)in HGUC-negative cases(NHGUC).The mutation rate of FGFR3 S249C is 25.0%(1/4)in non-invasive UC and 31.6%(6/19)in invasive UC.In samples diagnosed with SHGUC by urine cytology,the detection rate of FGFR3 S249C mutation determined by ddPCR is 50%(4/8),and the area under the ROC curve is 0.781 with a 95%CI(0.407,1.000).This indicates that it has a good auxiliary role in the diagnosis of UC.Conclusion:The use of digital PCR to detect TERT promoter mutations is expected to become a potential molecular biomarker for evaluating postoperative recurrence in HCC patients.Combining digital PCR with urinary cytology examination can improve the diagnostic efficiency of urothelial carcinoma.These research findings indicate that the predictive models constructed using digital PCR for detecting liver cancer and urothelial carcinoma have certain reference value for clinical physicians to develop precise treatment strategies.