The primary drawback of serum PSA as a biomarker of PCa screening is the low specificity, especially in the PSA “gray area” (4.0–10.0 ng/mL). Compared with Western populations, its specificity for PCa seemed to be overall lower in Chinese populations. About 25% of patients with PSA in gray area were diagnosed as PCa, but approximately 40% in US clinical practice . The data from a collaborative report of the Asian population agreed with the viewpoint, which revealed that the detection rate of PCa is general below 25% in Asian men with serum PSA 4–10 ng/mL, even below 10% in Indonesia men . In this study, about 18.5% cases were positive biopsy outcomes. In addition, our results suggested that the diagnostic performance of PSAD seems to be better than PSA, The AUCs of PSA and PSAD for predicting PCa were 0.540 (95%CI: 0.451–0.628) and 0.680 (95%CI: 0.592–0.759), respectively. A recent study showed similar results that the AUC for predicting PCa was 0.55 for PSA and 0.69 for PSAD in 2,162 men with serum PSA 4–10 ng/mL, of whom 56% were African American . The significant difference in AUC was found in this report, but it was not shown in the present study.
The diagnostic value of urine PCA3 gene-based analysis for PCa has been extensively investigated in western populations. In this study, the diagnosis performances of PCA3 were evaluated in the urinary sediments from 113 Chinese patients with serum PSA 4–10 ng/ml. The PCA3 showed an AUC of 0.728 in the urinary sediments, which is slightly lower than that reported in two previous studies, where the AUC values of PCA3 were 0.734 and 0.750 in Chinese males with PSA 4–10 ng/ml [8, 13]. It should be noted that these three studies all suggested that the diagnostic accuracy of PCA3 was superior to that of PSA. The AUC values of PCA3 remain inconsistent among some studies of Western populations with PSA 4–10 ng/ml. Rigau M et al.  found an AUC of only 0.61 for PCA3 and Perdonà S et al.  described a high AUC of 0.873. This may be associated with the valuation procedures of the level of PCA3 mRNA, in addition to other variables in these studies, such as ethnic, small sample sizes. The PCA3 was generally measured with the Progensa PCA3 assay In Western studies, but with quantitative real-time RT-PCR assay in Chinese studies. However, Filella et al. thought that there was no significant difference between two methods used to assess the diagnostic effectiveness of PCA3 .
In the present study, the diagnostic value of PCA3 in the urinary exosomes was also assessed in 103 patients with serum PSA 4–10 ng/ml. Exosomes are the internal vesicles secreted from various cell types, including cancer cells. They are 30–200 nm in diameter, and contain the proteins, mRNAs and microRNAs of the derived cells. Nilsson et al. firstly reported that PCA3 could be detected in exosomes in urine of PCa patients in 2009 . Since then, few studies have investigated the diagnostic value of PCA3 in exosomes for PCa. In the study of Dijkstra et al., 24 (80%) of 30 patients with PSA > 3 ng/ml meet the predetermined cutoff of the sufficient amount of mRNA to assess the diagnostic performance of PCA3 . The rate of assessable exosomes samples was similar to the results of Hendriks et al. and the present study . However, the AUC of PCA3 in urinary exosomes was 0.64 (95% CI 0.41–0.88), that is lower than our results 0.740 (95% CI: 0.644–0.821)). The difference of AUC values may result from the heterogenicities in two studies, such as ethnic, small sample sizes. Consequently, the diagnostic value of PCA3 in urinary exosomes should be further evaluated. It should be noted that the exciting results have been recently reported in a series of studies of ExoDx Prostate test, which was used to predict the outcomes of initial biopsy and high-grade PCa through analyzing the TMPRSS2: ERG fusion gene and PCA3 in the exosomes of first-catch urine samples without prostate manipulation [19–21]. ExoDx Prostate test has been commercially available as a urinary test for the detection of PCa in Western countries. However, the report of Chinese populations using ExoDx Prostate test has not been found.
There are several limitations in this study should be acknowledged. First, the manual quantitative real-time RT-PCR is used to test PCA3 and PSA mRNA in urine samples, instead of the commercial PROGENSA® PCA3 test. Second, it is a single-center trial with limited participants. More convincing results should come from future large sample size and multicenter studies.