Prostate-specific antigen (PSA) is an established tumor marker for the diagnosis of patients with prostate cancer. The aim of the study was to evaluate the performance of [-2]proenzyme PSA ([-2]proPSA) and prostate health index (PHI) tumor markers in the differential diagnosis between benign prostatic diseases and prostate cancer.
Total PSA (tPSA), free PSA (fPSA) and [-2]proPSA were measured using antibody-based sandwich enzyme-linked immunosorbent assay with a chemiluminescent detection system in 110 patients, with a tPSA of 1.6–8.0 µg/L. The PHI and %[-2]proPSA were calculated from the PSA values mentioned above. The results were compared with histopathological examination results following a transrectal ultrasound-guided biopsy of the prostate.
For the prediction of a malignant histopathological result, the specificity at the 90% sensitivity level was 24.3% for [-2]proPSA, 32.4% for %[-2]proPSA, 28.4% for PHI, 18.9% for tPSA and 28.4% for the free-to-total PSA ratio. The area under the curve for [-2]proPSA, %[-2]proPSA, PHI, tPSA and the free-to-total PSA ratio was 0.663, 0.749, 0.742, 0.616 and 0.625, respectively.
Our study found a moderate improvement over tPSA and %fPSA in detecting prostate cancer using the [-2]proPSA assay in patients with a tPSA range of 1.6–8.0 µg/L.
Prostate cancer (PCa) is the most frequent cancer among males and the second most common cause of death in men in the USA and the EU , . The incidence rate of PCa has increased dramatically in the last 25 years due to the adoption of the tumor marker prostate specific antigen (PSA) into clinical practice .
PSA is the most widely used serum tumor biomarker for the early detection of PCa. PSA is an organ- but not a cancer-specific tumor marker. Besides PCa, PSA levels are also elevated in acute prostatitis and benign prostatic hyperplasia (BPH) . This limitation in specificity leads to substantial rates of false positive results from PSA tests, resulting in unnecessary prostate biopsies, especially in the PSA range of 4–10 µg/L.
Several PCa biomarkers were evaluated to overcome these limitations of PSA . One of these markers is PSA circulating in the unbound form – free PSA (fPSA), which is used in calculating the percentage of fPSA against the total PSA (tPSA) ratio (%fPSA), which helps distinguish benign conditions from prostate cancer .
Molecular isoforms of PSA are intensively investigated as possible effective tools for the early diagnosis of PCa . Free PSA is a mixture of circulating inactive forms unattached to plasma proteins. fPSA includes the subforms BPH-associated PSA (BPSA), inactive PSA (iPSA) and proenzyme PSA (proPSA) , . BPSA and iPSA are associated with benign tissue, while proPSA is considered a promising PCa serum marker , . Four different proPSA isoforms exist in serum, named [-2]proPSA, [-4]proPSA, [-5]proPSA and [-7]proPSA depending on whether the length of the pro-leader peptide sequences is seven, five, four or two amino acids. [-2]proPSA was found to be more accurate than tPSA and fPSA in differentiating PCa from benign disease in a prospective prostate cancer screening study in patients with PSA ranging from 2 to 10 µg/L. [-2]proPSA showed a higher specificity compared to tPSA and emerged as a promising marker for PCa detection , , .
The [-2]proPSA-based Beckman Coulter Prostate Health Index (PHI), defined as ([-2]proPSA/fPSA)×sqrtPSA[-2], and %[-2]proPSA, defined as [([-2]proPSA/fPSA)×100], were developed as important derivatives of [-2]proPSA . Several investigations showed that [-2]proPSA, %[-2]proPSA and PHI are higher in PCa than in benign prostate conditions and that their use significantly improved cancer detection compared to tPSA and %fPSA , , also having the potential of decreasing the number of unnecessary (negative) prostate biopsies , . Furthermore, PHI seems to be associated with PCa aggressiveness and might guide the selection of patients suitable for active surveillance PCa protocol and determine the point at which radical PCa therapy should be offered to these patients , .
The aim of our study was to evaluate the [-2]proPSA and PHI tumor markers in the differential diagnosis between benign prostatic diseases and PCa and compare it with PSA, an established clinical marker.
Materials and methods
The study was approved by the National Medical Ethics Committee of the Republic of Slovenia and complied with the principles of the World Medical Association Declaration of Helsinki regarding ethical conduct of research involving human subjects. An informed consent was obtained from each participant. The study was conducted at the urological departments of two Slovenian University medical centres in Ljubljana and Maribor. Patients included in the study were routinely scheduled for transrectal ultrasound-guided biopsy (TRUS) of the prostate by the referring urologist due to an elevated PSA concentration (ranging between 1.6 and 8.0 µg/L). The exclusion criteria were a history of previous malignant disease, an indwelling urinary catheter, a previous positive biopsy for PCa, medical therapy involving dutasteride, finasteride or androgen replacement, a history of instrumental procedures on the lower urinary tract that increase PSA levels up to 3 months before sample collection, an inconclusive histopathological result, a history of transurethral resection of the prostate or open prostatectomy and acute prostatitis.
A total of 110 patients were prospectively enrolled in our study from January 2013 to September 2014.
A blood sample for analysis was obtained just before TRUS and was allowed to clot before the serum was separated by centrifugation. Serum samples were centrifuged and frozen at −20°C within at least 3 h of blood drawing . At both sites, archival serum samples were analyzed. tPSA, fPSA and [-2]proPSA (Beckman Coulter [-2]proPSA) were measured on the Beckman Coulter Access 2 immunoassay analyzer (Beckman Coulter, Brea, CA, USA) using three dual monoclonal sandwich assays using Hybritech antibodies and a chemiluminescent detection system [the US Food and Drug Association (FDA)-approved]. All assays were performed from one sample cup using the World Health Organization (WHO) standard calibrated access tPSA and fPSA immunoassays or the automated [-2]proPSA assay. The analytical performance of the measurements assessed against control materials from Beckman Coulter showed values within the allowed recommended limits. tPSA, fPSA and [-2]proPSA results were obtained in single determinations.
%[-2]proPSA was calculated as ([-2]proPSA ng/L/10)/fPSA µg/L, and the PHI was then calculated according to the following formula: ([-2]proPSA/fPSA)×√PSA.
Before the biopsy, digital rectal examination of the prostate was performed and the volume of the prostate was measured by a TRUS examination. TRUS was performed using the protocol established in both institutions, targeting the lateral parts of the peripheral zone of both prostatic lobes. Ten biopsy cores were taken in each patient.
Biopsy cores were analyzed in the histopathology laboratory of the Institute for Pathology at the Medical Faculty of Ljubljana, Slovenia and the Department of Pathology of the University Medical Center Maribor, Slovenia. Reported results were divided into a no-evidence-of-malignancy (NOMA) category [BPH, prostatitis and high-grade prostatic intraepithelial neoplasia (HGPIN)] and a malignant (PCa and suspicious of PCa) category.
Data was analyzed using SPSS v.21.0 (IBM Corp., Armonk, NY, USA) and MedCalc v 16.4.3 (Software bvba, Ostend, Belgium). The measured concentrations of tPSA, fPSA, [-2]proPSA, %fPSA, %[-2]proPSA and PHI were compared using the Mann-Whitney U test for the two groups of patients.
tPSA, fPSA, [-2]proPSA, %fPSA, %[-2]proPSA and PHI were compared by calculating specificity at the 90% and 95% sensitivity levels and cut-off values were determined.
The performance of the tumor markers was also compared based on the receiver operating characteristic (ROC) curve by calculating the area under the curve (AUC) for each tumor marker, and pairwise comparison of the ROC curves was performed with the method of DeLong et al.
Out of 110 patients, 74 had no evidence of malignancy in histopathology and in 36, PCa was found. There were 48 (64.8%) patients with BPH, 17 (23.0%) with prostatitis and 9 (12.2%) with HGPIN in the NOMA group. In the malignant group, two patients (5.6%) had a suspicious-of-PCa histopathological result, 16 patients (44.4%) had Gleason 6 PCa, 11 patients (30.6%) had Gleason 7 PCa, three patients (8.3%) had Gleason 8 PCa and in four patients (11.1%), Gleason 9 PCa was found. The age and prostate volume were significantly different between the NOMA and malignant groups of patients (Table 1).
|Patient characteristics||NOMA group median (interquartile range)||Malignant group median (interquartile range)||Mann-Whitney U (sig.)|
|Number of patients||74||36|
|Age, years||64.0 (60.8–68.0)||67.0 (63.3–71.8)||976.00 (p=0.023)|
|Prostate volume, cm3||39.2 (29.6–54.3)||30.3 (24.3–47.6)||947.50 (p=0.014)|
NOMA, no evidence of malignancy. Bold indicates p<0.05.
The values of the tumor markers in the study were statistically significantly different between the groups of patients with NOMA and malignant histopathological results, except for the concentration of fPSA, for which no significant difference was confirmed (Table 2).
|Tumor marker||NOMA group median (interquartile range)||Malignant group median (interquartile range)||Mann-Whitney U (sig.)|
|Number of patients||74||36|
|tPSA, µg/L||4.34 (3.47–5.59)||5.03 (3.85–7.60)||1023.00 (p=0.049)|
|fPSA, µg/L||0.69 (0.53–1.01)||0.69 (0.53–1.07)||1283.50 (p=0.757)|
|[-2]proPSA, ng/L||10.33 (8.00–14.01)||13.64 (9.56–20.09)||897.50 (p=0.006)|
|%fPSA, %||0.163 (0.125–0.226)||0.143 (0.102–0.185)||999.00 (p=0.034)|
|%[-2]proPSA, %||62.63 (43.60–91.41)||118.68 (71.97–166.18)||668.00 (p=0.000)|
|PHI||32.67 (24.19–40.30)||51.42 (35.50–67.25)||686.00 (p=0.000)|
tPSA, total PSA; fPSA, free PSA; [-2]proPSA, [-2]proenzyme PSA; %fPSA, ratio of free PSA to total PSA; %[-2]proPSA, ratio of [-2]proPSA to fPSA; PHI, Prostate Health Index; NOMA, no evidence of malignancy. Bold indicates p<0.05.
%[-2]proPSA was found to have the highest specificity of 32.4% and 14.8% of the tumor markers in the study, at 90% and 95% sensitivity, respectively, followed by PHI and %fPSA with equal specificity of 28.4% and 12.2% at 90% and 95% sensitivity, respectively (Table 3).
|Tumor marker||Sensitivity||Specificity at 90% sensitivity||Cut-off value at 90% sensitivity||Sensitivity||Specificity at 95% sensitivity||Cut-off value at 95% sensitivity|
|tPSA||90.0||18.9||3.3 µg/L||95.0||6.8||2.7 µg/L|
|fPSA||90.0||5.41||1.56 µg/L||95.0||4.1||1.74 µg/L|
|[-2]proPSA||90.0||24.3||8.0 ng/L||95.0||14.9||6.6 ng/L|
tPSA, total PSA; fPSA, free PSA; [-2]proPSA, [-2]proenzyme PSA; %fPSA, ratio of free PSA to total PSA; %[-2]proPSA, ratio of [-2]proPSA to fPSA; PHI, Prostate Health Index.
%[-2]proPSA was found to have the largest AUC of all the tumor markers in the study, closely followed by PHI. The serum concentrations of tPSA and %fPSA had nearly the same performance, while the AUC for the concentration of fPSA was low (0.518) (Table 4 and Figure 1).
|Area under the curve|
|Tumor marker||AUC||95% confidence interval||Significance level p Area=0.5|
|Lower bound||Upper bound|
AUC, area under the curve; tPSA, total PSA; fPSA, free PSA; [-2]proPSA, [-2]proenzyme PSA; %fPSA, ratio of free PSA to total PSA; %[-2]proPSA, ratio of [-2]proPSA to fPSA; PHI, Prostate Health Index.
Pairwise comparison of ROC curves showed statistically significant differences between the AUC for %[-2]proPSA against tPSA, fPSA and %fPSA, and PHI against fPSA and %fPSA (Table 5).
|AUC difference (sig.)||tPSA||fPSA||[-2]proPSA||%fPSA||%[-2]proPSA|
|[-2]proPSA||0.0471 (0.4719)||0.145 (0.1505)|
|%fPSA||0.009 (0.9018)||0.107 (0.0404)||0.0381 (0.6722)|
|%[-2]proPSA||0.133 (0.0225)||0.231 (0.0030)||0.0861 (0.0639)||0.124 (0.0259)|
|PHI||0.127 (0.0784)||0.224 (0.0019)||0.0794 (0.1149)||0.117 (0.0456)||0.0068 (0.7510)|
AUC, area under the curve; tPSA, total PSA; fPSA, free PSA; [-2]proPSA, [-2]proenzyme PSA; %fPSA, ratio of free PSA to total PSA; %[-2]proPSA, ratio of [-2]proPSA to fPSA; PHI, Prostate Health Index. Bold indicates p<0.05.
The present study has focused on the performance of [-2]proPSA as a novel tumor marker for PCa. Ideally, a disease marker should have high sensitivity complemented by high specificity in order to identify as many patients with a disease as possible, while simultaneously preventing healthy subjects being subjected to unnecessary invasive and costly procedures. tPSA – an established PCa marker – was found to have a specificity of only 18.9% at the 90% sensitivity level, which is a minimal acceptable sensitivity level for a tumor marker. The main reason for the low specificity at the required sensitivity levels stems from the fact that tPSA is organ-specific but not PCa-specific . A considerable number of men in the general population have tPSA elevation due to BPH. Bohnen et al. have shown that 72% of men in the general population with no PCa and tPSA between 2.1 and 2.5 µg/L had a prostate volume above 30 cm3 and 66% of men with tPSA between 7.1 and 10.0 µg/L had a prostate volume above 50 cm3 .
In our study, [-2]proPSA was also detected in men with NOMA prostate disease; however, it was significantly higher in PCa patients (median concentrations 10.33 ng/L vs. 13.64 ng/L, p=0.006). When [-2]proPSA was normalized through fPSA and tPSA in %[-2]proPSA and PHI, the difference was even more significant (p=0.000), resulting in improvement of sensitivity and specificity against tPSA and fPSA. A meta-analysis of 17 studies with 6912 participants found high sensitivities for [-2]proPSA, %[-2]proPSA and PHI (90%, 89% and 90%, respectively), while specificities were low (13%, 32% and 31%, respectively) . In contrast, our results show higher specificity for [-2]proPSA (24.3%) and comparable specificity for %[-2]proPSA and PHI (32.4% and 28.4%, respectively) at the 90% sensitivity level. Reported values of sensitivity/specificity among different studies are largely dependent on the selection of the cut-off concentration of tumor markers, which is often dictated by the aim of the study. Lowering the cut-off concentration increases sensitivity but decreases specificity. In the above meta-analysis, the reported cut-off concentrations of [-2]proPSA were in the range of 7–9 ng/L , which is comparable with our study, where the cut-off concentration of [-2]proPSA at 90% sensitivity was 8.0 ng/L.
Calculating the ROC curve and AUC allows for comparison of the performance of different tumor markers in the study and across studies without the bias introduced by setting the cut-off concentration. The AUC of tPSA (0.616) in our study was not significantly different from the AUC of %fPSA (0.625). The AUC of %[-2]proPSA (0.749) and PHI (0.742) were significantly higher than the AUC of fPSA and %fPSA, and the AUC of %[-2]proPSA was also significantly higher than the AUC of tPSA; however, they were still significantly lower than the reported AUC by the meta-analysis of Pecoraro et al. (0.95, 0.89 and 0.95 for [-2]proPSA, %[-2]proPSA and PHI, respectively) . Other studies reported a similar AUC to ours for [-2]proPSA (0.62) , %[-2]proPSA (0.768) and PHI (0.781) .
One of the reasons for the improved performance of [-2]proPSA and its derivatives is its differential distribution in anatomical prostate zones. While tPSA does not seem to be differentially distributed in prostate zones, [-2]proPSA is elevated in cancerous and noncancerous parts of the prostate peripheral zone (PZ), but almost no [-2]proPSA (<0.2%) is present in the prostate transition zone (TZ) . In contrast, BPSA is found predominantly in the TZ of patients with nodular BPH in patients with or without PCa . When the main source of increased PSA is TZ, as in BPH, [-2]proPSA, %[-2]proPSA and PHI remain low, whereas in patients with PCa, which predominantly grows in the PZ, [-2]proPSA, %[-2]proPSA and PHI are significantly higher. When [-2]proPSA is produced exclusively in cancerous parts of the PZ, it would represent an ideal tumor marker with nearly 100% sensitivity and specificity limited only by the sensitivity of the marker laboratory assay to detect even minimal concentrations produced by cancerous cells.
Our study found a moderate improvement over tPSA and %fPSA in detecting PCa using the [-2]proPSA assay and PHI in patients with a tPSA range of 1.6–8.0 µg/L. Its adoption into clinical practice depends on cost-effectiveness and on the rate of development of new genetically based PCa markers, for example microRNAs, which promise nearly ideal performance (AUC=0.989) when used in panels . Not all studies, however, have found microRNAs to have such perfect performance, and some studies failed to find clinical value of these markers in PCa , .
The authors wish to thank Beckman Coulter for the donation of reagents for measuring tPSA, fPSA and [-2]proPSA.
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: None declared.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.
1. Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9–29. Search in Google Scholar
2. Arnold M, Karim-Kos HE, Coebergh JW, Byrnes G, Antilla A, Ferlay J, et al. Recent trends in incidence of five common cancers in 26 European countries since 1988: analysis of the European cancer observatory. Eur J Cancer 2015;51:1164–87. Search in Google Scholar
3. Stamey TA, Yang N, Hay AR, McNeal JE, Freiha FS, Redwine E. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med 1987;317:909–16. Search in Google Scholar
4. Bernstein LH, Rudolph RA, Pinto MM, Viner N, Zuckerman H. Medically significant concentrations of prostate-specific antigen in serum assessed. Clin Chem 1990;36:515–8. Search in Google Scholar
5. Velonas VM, Woo HH, dos Remedios CG, Assinder SJ. Current status of biomarkers for prostate cancer. Int J Mol Sci 2013;14:11034–60. Search in Google Scholar
6. Christensson A, Bjork T, Nilsson O, Dahlen U, Matikainen MT, Cockett AT, et al. Serum prostate specific antigen complexed to alpha 1-antichymotrypsin as an indicator of prostate cancer. J Urol 1993;150:100–5. Search in Google Scholar
7. Stephan C, Jung K, Lein M, Sinha P, Schnorr D, Loening SA. Molecular forms of prostate-specific antigen and human kallikrein 2 as promising tools for early diagnosis of prostate cancer. Cancer Epidemiol Biomarkers Prev 2000;9:1133–47. Search in Google Scholar
8. Mikolajczyk SD, Marks LS, Partin AW, Rittenhouse HG. Free prostate-specific antigen in serum is becoming more complex. Urology 2002;59:797–802. Search in Google Scholar
9. Mikolajczyk SD, Marker KM, Millar LS, Kumar A, Saedi MS, Payne JK, et al. A truncated precursor form of prostate-specific antigen is a more specific serum marker of prostate cancer. Cancer Res 2001;61:6958–63. Search in Google Scholar
10. Mikolajczyk SD, Millar LS, Wang TJ, Rittenhouse HG, Marks LS, Song W, et al. A precursor form of prostate-specific antigen is more highly elevated in prostate cancer compared with benign transition zone prostate tissue. Cancer Res 2000;60:756–9. Search in Google Scholar
11. Catalona WJ, Bartsch G, Rittenhouse HG, Evans CL, Linton HJ, Horninger W, et al. Serum pro-prostate specific antigen preferentially detects aggressive prostate cancers in men with 2 to 4 ng/ml prostate specific antigen. J Urol 2004;171(6 Pt 1):2239–44. Search in Google Scholar
12. Le BV, Griffin CR, Loeb S, Carvalhal GF, Kan D, Baumann NA, et al. [-2]Proenzyme prostate specific antigen is more accurate than total and free prostate specific antigen in differentiating prostate cancer from benign disease in a prospective prostate cancer screening study. J Urol 2010;183:1355–9. Search in Google Scholar
13. Jansen FH, Roobol M, Jenster G, Schroder FH, Bangma CH. Screening for prostate cancer in 2008 II: the importance of molecular subforms of prostate-specific antigen and tissue kallikreins. Eur Urol 2009;55:563–74. Search in Google Scholar
14. Mikolajczyk SD, Rittenhouse HG. Pro PSA: a more cancer specific form of prostate specific antigen for the early detection of prostate cancer. Keio J Med 2003;52:86–91. Search in Google Scholar
15. Loeb S, Catalona WJ. The Prostate Health Index: a new test for the detection of prostate cancer. Ther Adv Urol 2014;6:74–7. Search in Google Scholar
16. Guazzoni G, Nava L, Lazzeri M, Scattoni V, Lughezzani G, Maccagnano C, et al. Prostate-specific antigen (PSA) isoform p2PSA significantly improves the prediction of prostate cancer at initial extended prostate biopsies in patients with total PSA between 2.0 and 10 ng/ml: results of a prospective study in a clinical setting. Eur Urol 2011;60:214–22. Search in Google Scholar
17. Jansen FH, van Schaik RH, Kurstjens J, Horninger W, Klocker H, Bektic J, et al. Prostate-specific antigen (PSA) isoform p2PSA in combination with total PSA and free PSA improves diagnostic accuracy in prostate cancer detection. Eur Urol 2010;57:921–7. Search in Google Scholar
18. Filella X, Gimenez N. Evaluation of [-2] proPSA and Prostate Health Index (phi) for the detection of prostate cancer: a systematic review and meta-analysis. Clin Chem Lab Med 2013;51:729–39. Search in Google Scholar
19. Abrate A, Lughezzani G, Gadda GM, Lista G, Kinzikeeva E, Fossati N, et al. Clinical use of [-2]proPSA (p2PSA) and its derivatives (%p2PSA and Prostate Health Index) for the detection of prostate cancer: a review of the literature. Korean J Urol 2014;55:436–45. Search in Google Scholar
20. Makarov DV, Isharwal S, Sokoll LJ, Landis P, Marlow C, Epstein JI, et al. Pro-prostate-specific antigen measurements in serum and tissue are associated with treatment necessity among men enrolled in expectant management for prostate cancer. Clin Cancer Res 2009;15:7316–21. Search in Google Scholar
21. Semjonow A, Kopke T, Eltze E, Pepping-Schefers B, Burgel H, Darte C. Pre-analytical in-vitro stability of [-2]proPSA in blood and serum. Clin Biochem 2010;43:926–8. Search in Google Scholar
22. Bohnen AM, Groeneveld FP, Bosch JL. Serum prostate-specific antigen as a predictor of prostate volume in the community: the Krimpen study. Eur Urol 2007;51:1645–52. Search in Google Scholar
23. Pecoraro V, Roli L, Plebani M, Trenti T. Clinical utility of the (-2)proPSA and evaluation of the evidence: a systematic review. Clin Chem Lab Med 2016;54:1123–32. Search in Google Scholar
24. Ng CF, Chiu PK, Lam NY, Lam HC, Lee KW, Hou SS. The Prostate Health Index in predicting initial prostate biopsy outcomes in Asian men with prostate-specific antigen levels of 4-10 ng/mL. Int Urol Nephrol 2014;46:711–7. Search in Google Scholar
25. Mikolajczyk SD, Millar LS, Wang TJ, Rittenhouse HG, Wolfert RL, Marks LS, et al. “BPSA,” a specific molecular form of free prostate-specific antigen, is found predominantly in the transition zone of patients with nodular benign prostatic hyperplasia. Urology 2000;55:41–5. Search in Google Scholar
26. Bertoli G, Cava C, Castiglioni I. MicroRNAs as Biomarkers for Diagnosis, Prognosis and Theranostics in Prostate Cancer. Int J Mol Sci 2016;17:421. Search in Google Scholar
28. Witwer KW. Circulating microRNA biomarker studies: pitfalls and potential solutions. Clin Chem 2015;61:56–63. Search in Google Scholar
©2016 Walter de Gruyter GmbH, Berlin/Boston
This article is distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.