Int J Clin Oncol (2014) 19:782–792 DOI 10.1007/s10147-014-0742-y
REVIEW ARTICLE
Next‑generation prostate‑specific antigen test: precursor form of prostate‑specific antigen Kazuto Ito · Yuji Fujizuka · Kiyohide Ishikura · Bernard Cook
Received: 1 August 2014 / Accepted: 7 August 2014 / Published online: 20 August 2014 © Japan Society of Clinical Oncology 2014
Abstract An urgent need exists to develop a more sophisticated screening system in order to improve diagnostic accuracy of clinically significant cancer and also to reduce the drawbacks of prostate-specific antigen (PSA) screening including overdetection and overtreatment. The most promising next-generation PSA test, which can improve the management of prostate cancer, may be proenzyme PSA (proPSA) or precursor PSA (pPSA). proPSA has pro-leader peptide sequences of seven or less amino acids and previous studies demonstrated that [−2]proPSA, which contains only a 2-amino-acid propeptide leader, could be more useful not only to distinguish between men with and without cancer, but also between tumors with aggressive features with performance exceeding other classical PSA-related indices including ratio of free PSA to total PSA (%f-PSA) and PSA density. Recently, it was demonstrated that baseline [−2]proPSA-related indices were independent factors to predict pathological reclassification at one year or several years after entering active surveillance. Furthermore, a retrospective study suggested that [−2]proPSA might be a useful predictive marker for future developing clinically manifested prostate cancer as well as aggressive tumors. ProPSA-related indices may have the potential for developing a more ideal risk classification for men at risk for
K. Ito (*) · Y. Fujizuka Department of Urology, Gunma University Graduate School of Medicine, 3‑39‑22 Showamachi, Maebashi, Gunma 3718511, Japan e-mail: kzito@gunma‑u.ac.jp K. Ishikura Beckman Coulter Japan, Tokyo, Japan B. Cook Beckman Coulter, Brea, USA
13
prostate cancer, with a screening system maintaining the sensitivity of detecting clinically significant prostate cancer while saving cost, individualized treatment strategies, and follow-up procedures of active surveillance or active treatments. At a minimum, proPSA will be one of the most important new markers on the prostate cancer management in the near future. Keywords Prostate cancer · Prostate-specific antigen · proPSA
Introduction Prostate specific antigen (PSA) is a serum biomarker that has many important roles including screening for prostate cancer, objective assessment of tumor aggressiveness of prostate cancer, deciding appropriate treatment strategies, and predicting treatment outcomes. In Japan, epidemiological research predicted that the incidence of prostate cancer will increase to 118,000 cases between 2025 and 2029, which would make it the most frequent male cancer followed by stomach cancer, lung cancer, and colorectal cancer [1]. Therefore, there is an urgent need to take useful countermeasures against prostate cancer. Recently, large prospective randomized controlled trials, the European Randomized Study of Screening for Prostate Cancer (ERSPC) [2] and the Göteborg study [3], revealed that PSA-based screening could reduce mortality due to prostate cancer. In the Göteborg study, the mortality reduction in the screening group during a 14-year median follow-up was very high at 44 % according to the intentionto-screen analysis. The merits of introducing PSA-based screening could be mortality reduction and prevention of progression to
Int J Clin Oncol (2014) 19:782–792
metastatic disease, which may decrease quality of life. However, some participants in the screening program suffer drawbacks in terms of overdetection and overtreatment. Therefore, it would be equally necessary to develop a more sophisticated screening system in order to improve the diagnostic accuracy of clinically significant cancer as well as reducing the drawbacks of PSA screening. The most promising next-generation PSA test, which can improve the management of prostate cancer, is proenzyme PSA (proPSA) or precursor PSA (pPSA), which includes pro-leader peptide sequences of seven, five, four, or two amino acids and is more often associated with peripheral zone cancer than transition zone hyperplasia in prostate tissue [4]. proPSA is usually associated with cancer patients rather than with non-cancer patients when measured in the serum [5]. Here, we focus on the discovery phase, technical aspect of measurements, stability of molecule in the serum, clinical utility, and future perspectives on proPSA.
proPSA: fractions of free PSA isoforms PSA molecules enriched from human prostate cancer serum were examined using hydrophobic interaction chromatography-high performance liquid chromatography (HICHPLC). Research demonstrated that some of the uncomplexed PSA, free PSA, is an inactive proPSA, and proPSA makes up a significant fraction of the free PSA (approximately 25 %) in pooled serum samples of a prostate cancer patient [4]. Other studies revealed proPSA has no or very low enzymatic activity, and this zymogen form of PSA is cleaved and activated by human kallilrein 2 (hk2) [6, 7]. A subsequent analysis demonstrated that a specific form of proPSA, [−2]proPSA, with a serine-arginine leader peptide is a more elevated portion of free PSA forms in the serum of prostate cancer patients compared to biopsy-negative men, ranging from 25−95 and 6−19 %, respectively [8]. There are other forms of free PSA in the prostate gland and in serum. Benign PSA (BPSA) is usually associated with the transition zone of the prostate as is intact, nonnative PSA (iPSA), which is not internally clipped but is an enzymatically inactive form that does not form a complex with serum protease inhibitors. BPSA and iPSA constitute a mean of 28 and 39 % of the free PSA fraction, respectively, in a sample set of 157 men with biopsy-proven cancer, with PSA values from 4−10 ng/ml. Another distinct form of free PSA is proPSA. In the same serum set, proPSA represented 33 % of the free PSA [9]. Additional investigation with HIC-HPLC revealed that isoforms of proPSA can be separated into three groups as shown in Fig. 1 [10]. Peak 3 is native proPSA, [−7]proPSA, with the full 7-amino acid pro-leader peptide consisting of APLILSR, coeluted with the slightly truncated [−5]proPSA
783
Fig. 1 HIC-HPLC elution profile of PSA and proPSA forms. Each PSA form was purified previously by HPLC (Ref. [8]), and an overlay chromatogram of the peaks is presented. (Modified from the original Fig. 1 in Ref. [10]; Reproduced with permission from the American Association for Clinical Chemistry)
containing a 5-amino acid pro-leader peptide with LILSR. Because [−7]proPSA and [−5]proPSA are recognized simultaneously with the same antibody, these are designated [−5, −7]proPSA. Peak 2 is [−4]proPSA, containing a 4-amino acid pro-leader peptide ILSR. Peak 1 eluted after PSA is the most truncated form of proPSA, [−2]proPSA, which contains only a 2-amino acid leader sequence, SR. Each proPSA is enzymatically inactive, and these isoforms are unable to form a complex with α1-antichymotrypsin.
The molecular structure of proPSA in the prostate gland and in serum Figure 2 depicts the processing of the molecular structure of PSA in the prostate and serum. PSA is initially synthesized as a premature form of PSA including pre-protein with 261 amino acid (pre-proPSA) [4, 6–12]. One form of proPSA is co-translationally generated by the removal of the 17-amino acid leader sequence of pre-proPSA, that includes 244 amino acids in the native form. This native molecule of proPSA, [−7]proPSA with the 7-amino acid pro-leader peptide, is secreted in the acinar region of the prostate. The protease, hk2, removes all of the amino acid pro-leader peptide located in the N-terminal of [−7]proPSA and generates mature PSA. The partial removal of amino acids in the N-termonal of [−7]proPSA generates truncated forms of proPSA, which are [−5]proPSA, [−4]proPSA, and [−2]proPSA. In those proPSA isoforms, [−2]proPSA only has a two-amino acid leader peptide remaining. Consequently, hk2 is unable to cleave [−2]proPSA to mature PSA, and this molecule is stable in serum. [−2]proPSA tends to accumulate in the prostate tissue, especially in the cancer cells, which may have some impairment in the activation system from proPSA to mature PSA. Furthermore, the prostate cancer cells lack the
13
784
Int J Clin Oncol (2014) 19:782–792
Fig. 2 Intraprostatic and intravascular molecular forms of preproPSA, proPSA, active mature PSA and those derivatives
basal cell layer and invade small blood vessels. Therefore, proPSA may leak more from the malignant acinar gland compared to the normal prostatic gland.
Immunohistochemical distribution of proPSA in benign, high‑grade prostatic intraepithelial neoplasia (HG‑PIN) and malignant prostatic tissues In previous immunohistochemical analysis, all types of prostatic gland excluding atrophic benign gland stained positive with the monoclonal antibody to [−2]proPSA and [−5, −7]proPSA [13]. Intensity of immunostaining by monoclonal antibody of [−5, −7]proPSA was similar in malignant gland tissue, HG-PIN, and benign prostatic gland. On the other hand, intensity of immunostaining by monoclonal antibody to [−2]proPSA was evident in malignant epithelium compared to benign. There were no differences in intensity of immunostaining between low-grade and high-grade cancer. However, it is hypothesized that high-grade cancer may
13
increase more in the tumor volume and tend to invade microvascular systems more often compared to low-grade tumors. Therefore, it is also speculated that [−2]proPSA in highgrade cancer cells may tend to leak more to the circulation system compared to low-grade cancer cells.
Overview of biological and clinical features of [−2], [−4] and [−5, −7]proPSA The modification mechanism from [−7]proPSA to [−5], [−4], and [−2]proPSA isoforms as well as biological roles of proPSA has not been identified. However, hk2 can cleave the leader peptide of [−5, −7]proPSA and activate them rapidly, whereas [−4]proPSA is cleaved and activated slowly. Compared to these isoforms, [−2]proPSA only has a two-amino acid leader peptide remaining. Consequently, hk2 is unable to cleave [−2]proPSA to mature PSA, and consequently this molecule is the most stable in serum [8].
Int J Clin Oncol (2014) 19:782–792
Several clinical studies have also shown that [−5, −7]proPSA has no additional value over free PSA or total PSA in identifying prostate cancer [14, 15]. Other studies suggest the [−4]proPSA performs less well than [−2]proPSA in distinguishing between prostate cancer serum samples and benign serum samples [10]. Therefore, we focus here on the clinical value of [−2]proPSA.
Measurement of [−2]proPSA There is an automated immunoassay kit available to measure [−2]proPSA in serum, the Access Hybritech p2PSA (Beckman Coulter, Brea, USA). This is a two-site immunoenzymatic sandwich assay using Access analyzers. A sample is added to a reaction vessel with monoclonal antiPSA alkaline phosphatase conjugate, and paramagnetic particles coated with a monoclonal anti-[−2]proPSA antibody. The [−2]proPSA in the sample binds to the immobilized monoclonal anti-[−2]proPSA during the solid phase while, at the same time, the monoclonal anti-PSA alkaline phosphatase conjugate reacts with different antigenic sites on the [−2]proPSA molecule. After incubation in a reaction vessel, materials bound to the solid phase are held in a magnetic field while unbound materials are washed away. The chemiluminescent substrate is then added to the vessel and light generated by the reaction is measured. The light production is directly proportional to the concentration of [−2]proPSA in the sample. The measurement of [−2]proPSA by itself does not bring diagnostic efficiency. Access Hybritech p2PSA is used in combination with the Access Hybritech PSA and free PSA assays to calculate the Prostate Health Index, or phi. This multivariate index is √ calculated as ([−2]proPSA/freePSA) × PSA [16].
Stability of [−2]proPSA in whole blood and serum In general, free PSA is less stable than total PSA in serum samples. To maintain the quality of serum markers, it is necessary to know how to handle blood samples in a clinical setting. There have been studies that investigate the stability of proPSA in whole blood and serum after drawing from the body [17, 18]. Recently, Igawa et al. systematically investigated the stability of total PSA, free PSA, and [−2]proPSA in the serum and whole blood under various storage conditions, which included room temperature (21 °C) and refrigeration (4 °C) and set the time of measurement at one, 3, 8, and 24 h after drawing blood samples [18]. They investigated the stability of these markers in the serum and whole blood at room temperature or at 4 °C between 0 and 24 h after drawing blood from patients. The measured whole-blood [−2]proPSA increased in a
785
time-dependent manner, particularly at room temperature. Therefore, whole-blood samples at room temperature should be centrifuged within one hour after drawing and should be measured within 24 h. The stability of [−2]proPSA within 24 h was acceptable both at 4 °C and at room temperature. If it is difficult to measure within 24 h after drawing blood and to separate into serum samples, it is recommended that serum samples are stored in a deep freezer at −70 °C. Clinical utility of [−2]proPSA before prostate biopsy Several studies have investigated the diagnostic significance of [−2]proPSA-related indices including %p2PSA using the formula ({[−2]proPSA pg/ml/free PSA ng/ml × 1,000} × 100), p2PSA/%free PSA using the formula ({[−2]proPSA pg/ml/free PSA ng × PSA ng × 1,000} × 100), or Prostate Health Index (phi) using the formula mentioned above [16, 19–39]. Almost all headto-head comparisons revealed that diagnostic accuracy of [−2]proPSA-related indices were more evident than classical PSA-related serum indices including the ratio of free PSA to total PSA (%f-PSA) [16, 19, 20, 22–28, 30–32, 34–39] and also prostate volume-adjusted indices including PSA density and PSA density adjusted by transition zone prostate volume [35]. Recently, several studies demonstrated that %p2PSA and/or phi were more accurate laboratory-based indices for predicting positive prostate biopsy outcomes compared with classical serum indices, such as total PSA and %free PSA in men undergoing an initial prostate biopsy [16, 19, 20, 22–26, 28, 30–32, 34–39] as well as in those scheduled for repeat biopsy [27, 32] (Table 1). A single study showed outstanding diagnostic accuracy of prostate dimensionadjusted PSA-related indices including total or transition zone prostate volume-adjusted [−2]proPSA-related indices, such as %p2PSA density (%p2PSA/total prostate volume), phi density, %p2PSA transition zone volume density (TZD) using the formula (%p2PSA/transition zone volume), and phi TZD (Table 2) [35]. When sensitivity was fixed at 90 %, unnecessary biopsies could have been avoided in approximately 50 % of men when transrectal ultrasonography was used to calculate phi density and phi TZD. Several studies have demonstrated a positive correlation between tumor aggressiveness and [−2]proPSArelated indices [16, 24, 25, 30, 31, 35]. The Gleason score increased with increasing [−2]proPSA-related indices like [−2]proPSA, %p2PSA and/or phi in men with prostate cancer in the PSA range of 0.29–18.24 [24]. In a cohort including men with a family history of prostate cancer, the Gleason score increased with increasing [−2]proPSA-related
13
13
USA, Austria
Catalona et al. [19]
Clinical features
6 biopsy cores (n = 794), 10 biopsy cores (n = 297) Skoll et al. USA At least 10 [20] biopsy cores Stephan Caucasian 8–12 biopsy et al. [21] (Germany) cores Le et al. [22] USA Normal DRE Jansen et al. The Nether- At least 6 [23] land biopsy cores Austria USA At least 10 Skoll et al. [24] core template biopsy Catalona USA Normal DRE et al. [16] Guazzoni Italy 18–22 biopsy et al. [25] cores Lughezzani Italy At least 18 et al. [26] biopsy cores Lazzeri et al. Italy Males with 1 [27] or 2 previous negative biopsy Ferro et al. Italy Normal [28] DRE/18 biopsy cores At least 12 Lazzeri et al. Caucasian biopsy cores [29] (five European countries) Stephan Germany, At least 10 et al. [30] France biopsy cores Family Lazzeri et al. Five history of [31] European prostate cancer countries
Race/ Country
References
51.0 % 44.9 %
2.5–10.0 2.0–10.0 2.0–9.7 20.–10.0
2.0–10.0 2.0–10.0 0.5–19.9 0.3–46.4
2.0–20.0
2.0–10.0
74 405 351 429
892 268 729 222
151
646
1362 1.6–8.0 158
1.1–57.5
40.9 %
2.0–10.0
475
31.8 %
31.9 %
38.4 %
39.9 %
48.2 %
49.5 % 45.5 %
41.0 % 55.8 %
55.6 %
56.2 %
2.0–10.0
89
% positive biopsy
41.8 %
Range of PSA (ng/ml)
1091 2.0–10.0
n
Table 1 Diagnostic accuracy of laboratory-based [−2]proPSA-related indices
%f-PSA
%[−2]proPSA
90.0 %
90.0 %
90.0 %
90.0 %
90.1 %
NA
90.0 %
NA
90.0 % 80.0 %
88.5 % 90.0 %
90.0 %
90.0 %
90.0 %
37.9 %
33.6 %
22.8 %
38.0 %
40.4 %
NA
38.8 %
NA
33.9 % 44.9 %
48.6 % 31.8 %
41.7 %
41.0 %
21.0 %
0.730
0.720
0.670
0.730
0.720
NA
0.760
NA
0.695 0.700
0.760 0.716
0.780
0.730
0.650
90.0 %
90.0 %
90.0 %
90.0 %
90.1 %
90.0 %
90.0 %
95.0 %
90.0 % NA
88.5 % 90.0 %
NA
NA
NA
23.0 %
35.4 %
19.4 %
36.0 %
25.2 %
27.0 %
42.9 %
16.0 %
31.1 % NA
64.9 % 34.7 %
NA
NA
NA
0.730
0.740
0.670
0.770
0.670
0.700
0.760
0.703
0.709 NA
0.770 0.750
NA
NA
NA
90.0 %
90.0 %
90.0 %
90.0 %
91.6 %
90.0 %
90.0 %
95.0 %
90.0 % 80.0 %
88.5 % 90.0 %
90.0 %
90.0 %
90.0 %
19.5 %
15.0 %
21.5 %
34.0 %
13.9 %
17.8 %
20.0 %
8.4 %
11.9 % 34.6 %
40.5 % 22.2 %
45.5 %
18.0 %
13.0 %
0.600
0.610
0.640
NA
0.600
0.620
0.580
0.648
0.576 0.660
0.680 0.675
0.770
0.530
0.602
Sensitivity Specificity AUC-ROC Sensitivity Specificity AUC-ROC Sensitivity Specificity AUC-ROC
phi
Classical laboratory-based PSArelated indices
Laboratory-based [−2]proPSA-related indices
786 Int J Clin Oncol (2014) 19:782–792
Germany
Italy
Italy
Stephan et al. [32]
Scattoni et al. [33]
Perdona et al. [34]
Five European countries
China
Italy
Fossati et al. Five [39] European countries
Lughezzani et al. [38]
Ferro et al. [36] Ng et al. [37]
Ito et al. [35] Japan
Race/ Country
References
Table 1 continued
Age
REVIEW ARTICLE
Next‑generation prostate‑specific antigen test: precursor form of prostate‑specific antigen Kazuto Ito · Yuji Fujizuka · Kiyohide Ishikura · Bernard Cook
Received: 1 August 2014 / Accepted: 7 August 2014 / Published online: 20 August 2014 © Japan Society of Clinical Oncology 2014
Abstract An urgent need exists to develop a more sophisticated screening system in order to improve diagnostic accuracy of clinically significant cancer and also to reduce the drawbacks of prostate-specific antigen (PSA) screening including overdetection and overtreatment. The most promising next-generation PSA test, which can improve the management of prostate cancer, may be proenzyme PSA (proPSA) or precursor PSA (pPSA). proPSA has pro-leader peptide sequences of seven or less amino acids and previous studies demonstrated that [−2]proPSA, which contains only a 2-amino-acid propeptide leader, could be more useful not only to distinguish between men with and without cancer, but also between tumors with aggressive features with performance exceeding other classical PSA-related indices including ratio of free PSA to total PSA (%f-PSA) and PSA density. Recently, it was demonstrated that baseline [−2]proPSA-related indices were independent factors to predict pathological reclassification at one year or several years after entering active surveillance. Furthermore, a retrospective study suggested that [−2]proPSA might be a useful predictive marker for future developing clinically manifested prostate cancer as well as aggressive tumors. ProPSA-related indices may have the potential for developing a more ideal risk classification for men at risk for
K. Ito (*) · Y. Fujizuka Department of Urology, Gunma University Graduate School of Medicine, 3‑39‑22 Showamachi, Maebashi, Gunma 3718511, Japan e-mail: kzito@gunma‑u.ac.jp K. Ishikura Beckman Coulter Japan, Tokyo, Japan B. Cook Beckman Coulter, Brea, USA
13
prostate cancer, with a screening system maintaining the sensitivity of detecting clinically significant prostate cancer while saving cost, individualized treatment strategies, and follow-up procedures of active surveillance or active treatments. At a minimum, proPSA will be one of the most important new markers on the prostate cancer management in the near future. Keywords Prostate cancer · Prostate-specific antigen · proPSA
Introduction Prostate specific antigen (PSA) is a serum biomarker that has many important roles including screening for prostate cancer, objective assessment of tumor aggressiveness of prostate cancer, deciding appropriate treatment strategies, and predicting treatment outcomes. In Japan, epidemiological research predicted that the incidence of prostate cancer will increase to 118,000 cases between 2025 and 2029, which would make it the most frequent male cancer followed by stomach cancer, lung cancer, and colorectal cancer [1]. Therefore, there is an urgent need to take useful countermeasures against prostate cancer. Recently, large prospective randomized controlled trials, the European Randomized Study of Screening for Prostate Cancer (ERSPC) [2] and the Göteborg study [3], revealed that PSA-based screening could reduce mortality due to prostate cancer. In the Göteborg study, the mortality reduction in the screening group during a 14-year median follow-up was very high at 44 % according to the intentionto-screen analysis. The merits of introducing PSA-based screening could be mortality reduction and prevention of progression to
Int J Clin Oncol (2014) 19:782–792
metastatic disease, which may decrease quality of life. However, some participants in the screening program suffer drawbacks in terms of overdetection and overtreatment. Therefore, it would be equally necessary to develop a more sophisticated screening system in order to improve the diagnostic accuracy of clinically significant cancer as well as reducing the drawbacks of PSA screening. The most promising next-generation PSA test, which can improve the management of prostate cancer, is proenzyme PSA (proPSA) or precursor PSA (pPSA), which includes pro-leader peptide sequences of seven, five, four, or two amino acids and is more often associated with peripheral zone cancer than transition zone hyperplasia in prostate tissue [4]. proPSA is usually associated with cancer patients rather than with non-cancer patients when measured in the serum [5]. Here, we focus on the discovery phase, technical aspect of measurements, stability of molecule in the serum, clinical utility, and future perspectives on proPSA.
proPSA: fractions of free PSA isoforms PSA molecules enriched from human prostate cancer serum were examined using hydrophobic interaction chromatography-high performance liquid chromatography (HICHPLC). Research demonstrated that some of the uncomplexed PSA, free PSA, is an inactive proPSA, and proPSA makes up a significant fraction of the free PSA (approximately 25 %) in pooled serum samples of a prostate cancer patient [4]. Other studies revealed proPSA has no or very low enzymatic activity, and this zymogen form of PSA is cleaved and activated by human kallilrein 2 (hk2) [6, 7]. A subsequent analysis demonstrated that a specific form of proPSA, [−2]proPSA, with a serine-arginine leader peptide is a more elevated portion of free PSA forms in the serum of prostate cancer patients compared to biopsy-negative men, ranging from 25−95 and 6−19 %, respectively [8]. There are other forms of free PSA in the prostate gland and in serum. Benign PSA (BPSA) is usually associated with the transition zone of the prostate as is intact, nonnative PSA (iPSA), which is not internally clipped but is an enzymatically inactive form that does not form a complex with serum protease inhibitors. BPSA and iPSA constitute a mean of 28 and 39 % of the free PSA fraction, respectively, in a sample set of 157 men with biopsy-proven cancer, with PSA values from 4−10 ng/ml. Another distinct form of free PSA is proPSA. In the same serum set, proPSA represented 33 % of the free PSA [9]. Additional investigation with HIC-HPLC revealed that isoforms of proPSA can be separated into three groups as shown in Fig. 1 [10]. Peak 3 is native proPSA, [−7]proPSA, with the full 7-amino acid pro-leader peptide consisting of APLILSR, coeluted with the slightly truncated [−5]proPSA
783
Fig. 1 HIC-HPLC elution profile of PSA and proPSA forms. Each PSA form was purified previously by HPLC (Ref. [8]), and an overlay chromatogram of the peaks is presented. (Modified from the original Fig. 1 in Ref. [10]; Reproduced with permission from the American Association for Clinical Chemistry)
containing a 5-amino acid pro-leader peptide with LILSR. Because [−7]proPSA and [−5]proPSA are recognized simultaneously with the same antibody, these are designated [−5, −7]proPSA. Peak 2 is [−4]proPSA, containing a 4-amino acid pro-leader peptide ILSR. Peak 1 eluted after PSA is the most truncated form of proPSA, [−2]proPSA, which contains only a 2-amino acid leader sequence, SR. Each proPSA is enzymatically inactive, and these isoforms are unable to form a complex with α1-antichymotrypsin.
The molecular structure of proPSA in the prostate gland and in serum Figure 2 depicts the processing of the molecular structure of PSA in the prostate and serum. PSA is initially synthesized as a premature form of PSA including pre-protein with 261 amino acid (pre-proPSA) [4, 6–12]. One form of proPSA is co-translationally generated by the removal of the 17-amino acid leader sequence of pre-proPSA, that includes 244 amino acids in the native form. This native molecule of proPSA, [−7]proPSA with the 7-amino acid pro-leader peptide, is secreted in the acinar region of the prostate. The protease, hk2, removes all of the amino acid pro-leader peptide located in the N-terminal of [−7]proPSA and generates mature PSA. The partial removal of amino acids in the N-termonal of [−7]proPSA generates truncated forms of proPSA, which are [−5]proPSA, [−4]proPSA, and [−2]proPSA. In those proPSA isoforms, [−2]proPSA only has a two-amino acid leader peptide remaining. Consequently, hk2 is unable to cleave [−2]proPSA to mature PSA, and this molecule is stable in serum. [−2]proPSA tends to accumulate in the prostate tissue, especially in the cancer cells, which may have some impairment in the activation system from proPSA to mature PSA. Furthermore, the prostate cancer cells lack the
13
784
Int J Clin Oncol (2014) 19:782–792
Fig. 2 Intraprostatic and intravascular molecular forms of preproPSA, proPSA, active mature PSA and those derivatives
basal cell layer and invade small blood vessels. Therefore, proPSA may leak more from the malignant acinar gland compared to the normal prostatic gland.
Immunohistochemical distribution of proPSA in benign, high‑grade prostatic intraepithelial neoplasia (HG‑PIN) and malignant prostatic tissues In previous immunohistochemical analysis, all types of prostatic gland excluding atrophic benign gland stained positive with the monoclonal antibody to [−2]proPSA and [−5, −7]proPSA [13]. Intensity of immunostaining by monoclonal antibody of [−5, −7]proPSA was similar in malignant gland tissue, HG-PIN, and benign prostatic gland. On the other hand, intensity of immunostaining by monoclonal antibody to [−2]proPSA was evident in malignant epithelium compared to benign. There were no differences in intensity of immunostaining between low-grade and high-grade cancer. However, it is hypothesized that high-grade cancer may
13
increase more in the tumor volume and tend to invade microvascular systems more often compared to low-grade tumors. Therefore, it is also speculated that [−2]proPSA in highgrade cancer cells may tend to leak more to the circulation system compared to low-grade cancer cells.
Overview of biological and clinical features of [−2], [−4] and [−5, −7]proPSA The modification mechanism from [−7]proPSA to [−5], [−4], and [−2]proPSA isoforms as well as biological roles of proPSA has not been identified. However, hk2 can cleave the leader peptide of [−5, −7]proPSA and activate them rapidly, whereas [−4]proPSA is cleaved and activated slowly. Compared to these isoforms, [−2]proPSA only has a two-amino acid leader peptide remaining. Consequently, hk2 is unable to cleave [−2]proPSA to mature PSA, and consequently this molecule is the most stable in serum [8].
Int J Clin Oncol (2014) 19:782–792
Several clinical studies have also shown that [−5, −7]proPSA has no additional value over free PSA or total PSA in identifying prostate cancer [14, 15]. Other studies suggest the [−4]proPSA performs less well than [−2]proPSA in distinguishing between prostate cancer serum samples and benign serum samples [10]. Therefore, we focus here on the clinical value of [−2]proPSA.
Measurement of [−2]proPSA There is an automated immunoassay kit available to measure [−2]proPSA in serum, the Access Hybritech p2PSA (Beckman Coulter, Brea, USA). This is a two-site immunoenzymatic sandwich assay using Access analyzers. A sample is added to a reaction vessel with monoclonal antiPSA alkaline phosphatase conjugate, and paramagnetic particles coated with a monoclonal anti-[−2]proPSA antibody. The [−2]proPSA in the sample binds to the immobilized monoclonal anti-[−2]proPSA during the solid phase while, at the same time, the monoclonal anti-PSA alkaline phosphatase conjugate reacts with different antigenic sites on the [−2]proPSA molecule. After incubation in a reaction vessel, materials bound to the solid phase are held in a magnetic field while unbound materials are washed away. The chemiluminescent substrate is then added to the vessel and light generated by the reaction is measured. The light production is directly proportional to the concentration of [−2]proPSA in the sample. The measurement of [−2]proPSA by itself does not bring diagnostic efficiency. Access Hybritech p2PSA is used in combination with the Access Hybritech PSA and free PSA assays to calculate the Prostate Health Index, or phi. This multivariate index is √ calculated as ([−2]proPSA/freePSA) × PSA [16].
Stability of [−2]proPSA in whole blood and serum In general, free PSA is less stable than total PSA in serum samples. To maintain the quality of serum markers, it is necessary to know how to handle blood samples in a clinical setting. There have been studies that investigate the stability of proPSA in whole blood and serum after drawing from the body [17, 18]. Recently, Igawa et al. systematically investigated the stability of total PSA, free PSA, and [−2]proPSA in the serum and whole blood under various storage conditions, which included room temperature (21 °C) and refrigeration (4 °C) and set the time of measurement at one, 3, 8, and 24 h after drawing blood samples [18]. They investigated the stability of these markers in the serum and whole blood at room temperature or at 4 °C between 0 and 24 h after drawing blood from patients. The measured whole-blood [−2]proPSA increased in a
785
time-dependent manner, particularly at room temperature. Therefore, whole-blood samples at room temperature should be centrifuged within one hour after drawing and should be measured within 24 h. The stability of [−2]proPSA within 24 h was acceptable both at 4 °C and at room temperature. If it is difficult to measure within 24 h after drawing blood and to separate into serum samples, it is recommended that serum samples are stored in a deep freezer at −70 °C. Clinical utility of [−2]proPSA before prostate biopsy Several studies have investigated the diagnostic significance of [−2]proPSA-related indices including %p2PSA using the formula ({[−2]proPSA pg/ml/free PSA ng/ml × 1,000} × 100), p2PSA/%free PSA using the formula ({[−2]proPSA pg/ml/free PSA ng × PSA ng × 1,000} × 100), or Prostate Health Index (phi) using the formula mentioned above [16, 19–39]. Almost all headto-head comparisons revealed that diagnostic accuracy of [−2]proPSA-related indices were more evident than classical PSA-related serum indices including the ratio of free PSA to total PSA (%f-PSA) [16, 19, 20, 22–28, 30–32, 34–39] and also prostate volume-adjusted indices including PSA density and PSA density adjusted by transition zone prostate volume [35]. Recently, several studies demonstrated that %p2PSA and/or phi were more accurate laboratory-based indices for predicting positive prostate biopsy outcomes compared with classical serum indices, such as total PSA and %free PSA in men undergoing an initial prostate biopsy [16, 19, 20, 22–26, 28, 30–32, 34–39] as well as in those scheduled for repeat biopsy [27, 32] (Table 1). A single study showed outstanding diagnostic accuracy of prostate dimensionadjusted PSA-related indices including total or transition zone prostate volume-adjusted [−2]proPSA-related indices, such as %p2PSA density (%p2PSA/total prostate volume), phi density, %p2PSA transition zone volume density (TZD) using the formula (%p2PSA/transition zone volume), and phi TZD (Table 2) [35]. When sensitivity was fixed at 90 %, unnecessary biopsies could have been avoided in approximately 50 % of men when transrectal ultrasonography was used to calculate phi density and phi TZD. Several studies have demonstrated a positive correlation between tumor aggressiveness and [−2]proPSArelated indices [16, 24, 25, 30, 31, 35]. The Gleason score increased with increasing [−2]proPSA-related indices like [−2]proPSA, %p2PSA and/or phi in men with prostate cancer in the PSA range of 0.29–18.24 [24]. In a cohort including men with a family history of prostate cancer, the Gleason score increased with increasing [−2]proPSA-related
13
13
USA, Austria
Catalona et al. [19]
Clinical features
6 biopsy cores (n = 794), 10 biopsy cores (n = 297) Skoll et al. USA At least 10 [20] biopsy cores Stephan Caucasian 8–12 biopsy et al. [21] (Germany) cores Le et al. [22] USA Normal DRE Jansen et al. The Nether- At least 6 [23] land biopsy cores Austria USA At least 10 Skoll et al. [24] core template biopsy Catalona USA Normal DRE et al. [16] Guazzoni Italy 18–22 biopsy et al. [25] cores Lughezzani Italy At least 18 et al. [26] biopsy cores Lazzeri et al. Italy Males with 1 [27] or 2 previous negative biopsy Ferro et al. Italy Normal [28] DRE/18 biopsy cores At least 12 Lazzeri et al. Caucasian biopsy cores [29] (five European countries) Stephan Germany, At least 10 et al. [30] France biopsy cores Family Lazzeri et al. Five history of [31] European prostate cancer countries
Race/ Country
References
51.0 % 44.9 %
2.5–10.0 2.0–10.0 2.0–9.7 20.–10.0
2.0–10.0 2.0–10.0 0.5–19.9 0.3–46.4
2.0–20.0
2.0–10.0
74 405 351 429
892 268 729 222
151
646
1362 1.6–8.0 158
1.1–57.5
40.9 %
2.0–10.0
475
31.8 %
31.9 %
38.4 %
39.9 %
48.2 %
49.5 % 45.5 %
41.0 % 55.8 %
55.6 %
56.2 %
2.0–10.0
89
% positive biopsy
41.8 %
Range of PSA (ng/ml)
1091 2.0–10.0
n
Table 1 Diagnostic accuracy of laboratory-based [−2]proPSA-related indices
%f-PSA
%[−2]proPSA
90.0 %
90.0 %
90.0 %
90.0 %
90.1 %
NA
90.0 %
NA
90.0 % 80.0 %
88.5 % 90.0 %
90.0 %
90.0 %
90.0 %
37.9 %
33.6 %
22.8 %
38.0 %
40.4 %
NA
38.8 %
NA
33.9 % 44.9 %
48.6 % 31.8 %
41.7 %
41.0 %
21.0 %
0.730
0.720
0.670
0.730
0.720
NA
0.760
NA
0.695 0.700
0.760 0.716
0.780
0.730
0.650
90.0 %
90.0 %
90.0 %
90.0 %
90.1 %
90.0 %
90.0 %
95.0 %
90.0 % NA
88.5 % 90.0 %
NA
NA
NA
23.0 %
35.4 %
19.4 %
36.0 %
25.2 %
27.0 %
42.9 %
16.0 %
31.1 % NA
64.9 % 34.7 %
NA
NA
NA
0.730
0.740
0.670
0.770
0.670
0.700
0.760
0.703
0.709 NA
0.770 0.750
NA
NA
NA
90.0 %
90.0 %
90.0 %
90.0 %
91.6 %
90.0 %
90.0 %
95.0 %
90.0 % 80.0 %
88.5 % 90.0 %
90.0 %
90.0 %
90.0 %
19.5 %
15.0 %
21.5 %
34.0 %
13.9 %
17.8 %
20.0 %
8.4 %
11.9 % 34.6 %
40.5 % 22.2 %
45.5 %
18.0 %
13.0 %
0.600
0.610
0.640
NA
0.600
0.620
0.580
0.648
0.576 0.660
0.680 0.675
0.770
0.530
0.602
Sensitivity Specificity AUC-ROC Sensitivity Specificity AUC-ROC Sensitivity Specificity AUC-ROC
phi
Classical laboratory-based PSArelated indices
Laboratory-based [−2]proPSA-related indices
786 Int J Clin Oncol (2014) 19:782–792
Germany
Italy
Italy
Stephan et al. [32]
Scattoni et al. [33]
Perdona et al. [34]
Five European countries
China
Italy
Fossati et al. Five [39] European countries
Lughezzani et al. [38]
Ferro et al. [36] Ng et al. [37]
Ito et al. [35] Japan
Race/ Country
References
Table 1 continued
Age
