The Value Of Urine PCA3 Detection Based On ⁶⁸Ga-PSMA PET/CT In The Diagnosis Of Prostate Cancer

Background:Prostate cancer（PCa）seriously endangers the health of middle-aged and elderly men,and its early diagnosis is still a difficult problem to be explored in clinical practice.Compared with traditional imaging techniques such as CT and MRI,⁶⁸Ga-PSMA PET/CT showed higher accuracy in the diagnosis of PCa.PCA3 is a prostate cancer-specific antigen,which is overexpressed in prostate cancer tissues,and has shown a good clinical application prospect in the diagnosis of PCa.Objective:To evaluate the accuracy of ⁶⁸Ga-PSMA PET/CT and urine PCA3detection in the diagnosis of PCa,and to combine the two detection results to determine the high-risk and low-risk groups of PCa,and to reduce clinically unnecessary prostate biopsy.Combined with other clinical data such as serum PSA,PV and PSAD,a more efficient PCa diagnostic model was established.Materials and Methods:The clinical data（serum PSA,PV and PSAD）of suspected PCa patients with serum PSA﹥4ng/ml were collected from February2020 to January 2022 in the Department of Urology,Nanjing First Hospital affiliated to Nanjing Medical University.⁶⁸Ga-PSMA PET/CT was performed to determine the SUVmax value of the lesions.PCA3 scores were obtained from urine PCA3 detection through the method of quantitative real-time PCR.The diagnostic efficacy of ⁶⁸Ga-PSMA PET/CT and urine PCA3 detection for PCa was analyzed based on the pathological results of prostate biopsy.The double-positive and double-negative results were analyzed to identify the high-risk and low-risk groups of PCa.The risk factors of PCa were identified through univariate and multivariate analysis.A model for predicting PCa was established,and then draw a nomogram to visualize the data model.The performance of the model for predicting PCa was assessed by the area under the ROC curve（the AUC value）.The net benefit of the model was evaluated by the DCA curve.The reliability of the model was evaluated by internal validation of the calibration diagram.Results:A total of 98 patients were included.The sensitivity of SUVmax value in the diagnosis of PCa was 80.5%,the specificity was 87.7%,the PPV was 82.5%,the NPV was 86.2%,and the diagnostic accuracy was 84.7%;and urine PCA3 detection corresponded to 73.2%%,57.9%,55.6%,75%and 64.3%respectively.The PPV for double-positive results was 92.6%,and the NPV for double-negative results was90.3%.For patients with serum PSA of 4-10ng/ml,the double-negative results had the highest sensitivity and NPV:90.0%and 88.9%,respectively;the double-positive results showed the highest specificity of 93.3%;SUVmax had the most excellent PPV of 71.4%.For patients with serum PSA of 10-20 ng/ml,double-negative results had the highest sensitivity and NPV:92.3%and 92.9%,respectively;the double-positive results showed the highest specificity and PPV:95.8%and 88.9%,respectively.For patients with serum PSA of 20-50ng/ml,the sensitivity,specificity,PPV,and NPV of SUVmax in diagnosing PCa were all 100%.For patients with serum PSA﹥50ng/ml,the detection rate of PCa was 100%.Univariate analysis showed that SUVmax（P=0.000）,PCA3 score（P=0.000）,PSAD（P=0.010）,PV（P=0.028）and serum PSA（P=0.017）had significantly differences between PCa patients and non-PCa patients.The difference was statistically significant,and the corresponding AUC values under the ROC curve were:0.893,0.759,0.686,0.621and 0.617,respectively.The multivariate logistic regression analysis model included SUVmax（P=0.000）and PCA3 score（P=0.006）.The AUC value of the model was0.929,which was higher than 0.893 for SUVmax and 0.759 for PCA3 score.A visual nomogram model was established,and the optimal threshold probability was calculated to be 0.589.The sensitivity of the model to diagnose PCa was 80.5%,the specificity was 98.2%,the PPV was 97.1%,the NPV was 87.5%,and the diagnostic accuracy was 90.8%.The DCA curve yielded that the nomogram model provided a net benefit of 0.23 for the patient at the optimal threshold probability.The internal validation of the calibration plot demonstrated the good reliability of the nomogram.Conclusions:SUVmax value of lesions detected by ⁶⁸Ga-PSMA PET/CT examination and PCA3 scores have good diagnostic performance for PCa,and double-positive and double-negative results can identify high-risk and low-risk patients and reduce clinically unnecessary biopsies.SUVmax,PCA3 score,PSAD,PV,and serum PSA all had the ability to predict PCa individually,with decreasing performance in order.The nomogram model including SUVmax and PCA3 scores has better diagnostic performance for PCa and can improve the efficiency of prostate biopsy.The model has good reliability after internal verification.