Prostate Cancer and Prostatic Disease (2014) 17, 70–74 & 2014 Macmillan Publishers Limited All rights reserved 1365-7852/14 www.nature.com/pcan
ORIGINAL ARTICLE
[-2]proPSA is an early marker for prostate cancer aggressiveness I Heidegger1, H Klocker1, E Steiner1, V Skradski1, M Ladurner1, R Pichler1, G Scha¨fer2, W Horninger1 and J Bektic1 BACKGROUND: The aim of the study was to evaluate the correlation between preoperative [-2]proPSA, the Gleason Score (GS) and the risk of non-organ-confined (NOC) disease in patients undergoing radical prostatectomy (RP). METHODS: Beckman Coulter Access immunoassay was used to study serum specimens of 381 patients enrolled in a prostate cancer (PCa) early detection program. Inclusion criteria were three or more available serum specimens over 4 years before diagnosis. The values obtained were correlated with the GSs and pathological stages of specimens obtained at RP. RESULTS: [-2]proPSA levels were significantly higher in the cancer group (n ¼ 208) than in the benign group (n ¼ 173). Already 4 years before diagnosis [-2]proPSA differed significantly between PCa and benign prostate in all measured time points, however, highest prediction value was 2 and 1 years before diagnosis (Po0.001). When stratified [-2]proPSA levels according to GS of RP specimens, [-2]proPSA was highest in patients with XGS8 and lowest in those with pGS6. The difference in [-2]proPSA values between GSX8 and GSp7 was highly significant (Po0.01) already 3 years before diagnosis. Investigating the correlation between extraprostatic extension and the preoperative [-2]proPSA levels we found preoperative [-2]proPSA values significantly higher in men with NOC PCa compared with organ-confined (OC) cancers. The highest predictive value of [-2]proPSA to differ between OC and extraprostatic extension was found 3 and 2 years before RP. CONCLUSIONS: Patients with high [-2]proPSA levels in the years before cancer diagnosis are at a higher risk of having aggressive PCas. Thus, the [-2]proPSA should be included in the treatment decision-making for managing screen-detected PCa. Prostate Cancer and Prostatic Disease (2014) 17, 70–74; doi:10.1038/pcan.2013.50; published online 29 October 2013 Keywords: [-2]proPSA; Gleason score; screening; cancer aggressiveness
INTRODUCTION Prostate cancer (PCa) is the leading cancer entity in men and the second leading cause of cancer death among men in European countries.1 Thus, early detection of PCa is one of the most important issues in health care. Currently, PSA is the only broadly used routine molecular marker for diagnosis and prognosis of PCa.2 However, PSA is known to be an organ-specific but not cancerspecific marker, thus PSA values could be also elevated in the benign prostatic hypertrophy, in urogenital infections or after mechanical manipulations such as digital rectal examination.2,3 Consequently, the low-predictive value and a low specificity of total PSA (tPSA) measurement are one of its major drawbacks in the clinical setting thereby limiting early diagnosis of PCa. Emilozzi et al. for example described that only 20–40% of men with PSA values between 4 and 10 ng ml 1 will be found to be diagnosed with PCa.4 Otherwise, although men with PSA values below 4 ng ml 1 were traditionally not biopsied, it has been showed that 20–30% of these will also have PCa.5,6 Therefore, more specific markers that can accurately detect as well as differentiate patients with significant from insignificant PCa are of urgent need. Observational evidence began to point to a substantial burden of associated overdiagnosis and overtreatment triggered by PSA testing. Although the PSA screening test cannot differ between lethal and non-lethal PCa, it allows men with potentially lethal disease to be diagnosed and treated in the early stages of their disease. However, men with nonlethal disease do not benefit from their diagnosis and subsequent treatment. Consequently, those men in whom PCa was never destined to cause clinical symptoms are overdiagnosed. Apart of the psychological distress when
cancer is diagnosed, the short-term and long-term adverse effects of treatment have to be considered.7 In general, PSA belongs to the family of kallikreins encoded by numerous genes on chromosome 19q13.4 produced in the prostatic ductal and acinar epithelium. Initially, PSA is produced with a 17 amino-acid leader sequence to produce the inactive precursor enzyme proPSA. After secreted into the prostatic ducts, proPSA is activated into PSA by hk2 and hk4.8,9 In serum, 70–90% of tPSA is present as a complexed form PSA with a number of different endogenous protease inhibitors, preventing damage of protease activity of PSA. However, 10–30% of tPSA is present in an uncomplexed form. This so-called free PSA (fPSA) is composed of three different types of enzymatically inactive PSA: benign PSA, intact inactive PSA and proPSA.8,10 Preclinical studies found proPSA significantly elevated in serum extracts of patients with PCa. Immunohistochemical staining of 47 PCa cases, for example, showed a positive staining for proPSA in all cases.11 The promising results from preclinical studies led to several clinical studies investigating proPSA as serum marker for PCa detection. The amount of proPSA was shown to be increased in PCa and to be more associated with aggressive forms of the malignant disease.12,13 This includes native proPSA containing a 7 amino-acid pro-leader peptide ([-7]proPSA) and proPSA forms with truncated pro peptides containing 4 ([-4]proPSA) and 2 ([-2]proPSA) amino acids. Among all proPSA isoforms, the [-2]proPSA form has received the most attention because it was the primary form found in tumor extracts and it was preferentially more concentrated in cancer tissue than in benign glands.11 Otherwise, benign PSA seems to be associated
1 Department of Urology, Medical University Innsbruck, Innsbruck, Austria and 2Department of Pathology, Medical University Innsbruck, Innsbruck, Austria. Correspondence: J Bektic, Department of Urology, Medical University Innsbruck, Anichstr. 35, Innsbruck 6020, Austria. E-mail: [email protected] Received 29 April 2013; revised 30 August 2013; accepted 5 September 2013; published online 29 October 2013
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
71 with the benign disease, but its exact role in prostatic diseases still has to be identified.14 Catalona et al. showed that %proPSA measurement significantly improved the specificity for PCa detection in a PSA range between 2 and 4 ng ml 1 thereby decreasing the number of unnecessary biopsies.13 Other research found in a patient cohort of 119 men that by using %proPSA, 75% of cancers might be detected with 59% of unnecessary biopsies being spared using %proPSA.15 Recent studies integrated [-2]proPSA measurements in multivariate models and found an improvement of predictive values between PCa and benign prostate upon integrating [-2]proPSA into their analyses. Sokoll et al., for example, were able to show that [-2]proPSA was superior to fPSA% in the overall diagnostic utility (area under the curve (AUC): 0.73 vs 0.61).16 Multivariate analyses of Jansen et al. showed that the addition of [-2]proPSA to a logistic regression model consisting of tPSA and fPSA% significantly increased PCa prediction value and specificity.17 In addition, publications showed that integration of [-2]proPSA in risk calculation systems improved PCa prediction compared with standard prediction markers PSA, fPSA% and age.8 Loeb et al., for example, recently showed in a prospective multicenter study including 892 patients that the phi-index ([-2]proPSA/fPSAxPSA½) directly reflects the PCa risk.18 To date, only a few studies address proPSA as a biomarker for predicting cancer aggressiveness. Therefore, the present study has been conducted to investigate the impact of [-2]proPSA levels on PCa aggressiveness years before diagnosis. Briefly, we showed for the first time that [-2]proPSA levels are able to distinguish highly aggressive from non-aggressive PCa already 3 years before diagnosis. Thus, [-2]proPSA could serve as an early detection marker for aggressive PCa.
MATERIALS AND METHODS Patient characteristics Our patient cohort included 381 serum samples of (benign 173 and cancer 208) men enrolled in a PCa screening project who underwent at least one ultrasound-guided prostate biopsy at the Department of Urology, Medical University of Innsbruck, between 1993 and 2006 because of elevated PSA levels. All patients were Caucasian race. All patients diagnosed with PCa underwent open radical retropubic prostatectomy (RP) at our institution followed by a histopathological examination of prostate specimens at the Department of Pathology, Medical University Innsbruck. In 1993, a screening project using PSA as the only screening test was launched in the Federal State of Tyrol, Austria. Age-referenced PSA levels in combination with percent-fPSA of less than 22% were used as biopsy criteria.19 Since October 1995, in addition to PSA, percent-fPSA o18% has been used for PCa detection.20 Moreover, patients with PSA levels higher than 10 ng ml 1 or suspicious findings on digital rectal examinations were advised to undergo biopsy. Owing to changes in the biopsy protocol, the number of biopsies obtained increased from 6 cores (January 1993 to October 1995) to 10 (November 1995 to March 2000) and then to 15 cores (March 2000 to July 2006).21 The selection for this study was based on serial serum specimens obtained three or more available serum specimens over 4 years before diagnosis collected between January 1993 and July 2006. Patient characteristics are shown in Table 1. Organ-confined disease (OC) was defined as ppT2c tumors, however, non-organ-confined disease (NOC) was defined as XpT3a tumors (extraprostatic extension). Concerning Gleason Score (GS) 7 patients, there was no differentiation between 3 þ 4 and 4 þ 3 scores because the number of 4 þ 3 tumors was too low.
Sample collection and preparation Samples were obtained from patients serum stored immediately after centrifugation at 70 1C until thawed for the [-2]proPSA measurement. All samples were thawed only once. Storage time ranged from 2 to 15 years with a median storage time of 5 years. PSA and fPSA levels were determined by Access Hybritech PSA and fPSA assays (Beckman Coulter, Fullerton, CA, USA).22 The immunoassay for the & 2014 Macmillan Publishers Limited
Table 1.
Demographic data
Patient number (%) Race Age (years) mean Median
Total
Benign
Cancer
n ¼ 381 (100%) Caucasian 62.3 61.5
n ¼ 173 (45.4%) Caucasian 62 61
n ¼ 208 (54.6%) Caucasian 62.5 62
8.5
17.2
8.0
15.9
11.9
15.8
14.1
17.1
2.6
3.7
3.8
4.0
3.7
3.9
2.8
3.5
2[pro]PSA (pg ml 1) mean 1 Year before diagnosis 2 Years before diagnosis 3 Years before diagnosis 4 Years before diagnosis PSA (ng ml 1) mean 1 Year before diagnosis 2 Years before diagnosis 3 Years before diagnosis 4 Years before diagnosis
detection of [-2]proPSA was a commercial dual monoclonal sandwich assay (Accesss Hybridech pPSA assay, Beckman Coulter) in a microtiter plate format using a biotinylated capture anti-PSA monoclonal antibody and Europium-labeled proPSA-specific monoclonal antibody for detection with a Victor 1420 multilabel counter (Perkin Elmer, Gaithersburg, MD, USA). The assay was performed according to the manufacture instructions. This assay had less than 0.2% cross-reactivity with mature PSA and had a limit of quantitation in serum of 0.015 ng ml 1. The development of this monoclonal antibody to [-2]proPSA has been described previously.23,24
Statistical analyses The statistical distribution of the different data sets was determined using the Kolmogorov–Smirnov test. All data sets followed a Gaussian distribution. Therefore, Student’s t-test was applied to calculate the statistical significance of differences between treatment groups. P-values below 0.05 were considered significant (*Po0.05; **Po0.01; ***Po0.001). Bars and error bars in the histograms represent mean values±s.d. AUC for pre-diagnostic -2[pro]PSA and PSA have been calculated. Multivariate logistic regression analysis was performed to determine associations of pre-diagnostic -2[pro]PSA and PSA and the outcome (SPSS software version 20; SPSS, Chicago, IL, USA).
RESULTS Of our cohort of 381 patients who underwent ultrasound-guided prostate biopsy because of elevated PSA levels, 54.6% (n ¼ 208) were diagnosed with PCa, however, 45.4% (n ¼ 173) of patients showed a benign histology. All patients diagnosed with PCa underwent RP. The median age of PCa patients was 62 years, the median age of benign patients was 60 years (Table 1). All patients (n ¼ 208) with needle biopsy-verified PCa, 52.9% (n ¼ 110) had pGS 6 in RP specimens, 35.6% (n ¼ 74) had GS 7 and 11.5% (n ¼ 24) had GS X8 tumors (Table 2). Mean [-2]proPSA levels (pg ml 1) and also mean PSA levels (ng ml 1) are shown in Table 1. Our data show that [-2]proPSA levels were significantly higher in the cancer group than in the benign (non-cancerous) group (Figure 1). Already 4 years before histological verified diagnosis [-2]proPSA was able to differ significantly between PCa and benign prostate in all measured time points (1–4 years before diagnosis; Po0.01). In addition, our data show the highest prediction value of [-2]proPSA in differing between Prostate Cancer and Prostatic Disease (2014), 70 – 74
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
72 Table 2. Overview about Gleason Scores and histology of prostatectomy specimens (n ¼ 208) Gleason Score prostatectomy specimen n ¼ 208 (100%) 6 110 7 74 X8 24 Histology prostatectomy specimen n ¼ 208 (100%) Organ confined, total n ¼ 166 (77.4%) pT2a 38 pT2b 54 pT2c 74 Non-organ-confined, total n ¼ 42 (22.6%) pT3a XpT3b
(52.9) (35.6) (11.5) (18.2) (24.6) (34.6)
40 (19) 2 (0.6)
Figure 2. [-2]proPSA differs between Gleason Scores (GS) of prostate cancer. [-2]proPSA levels (pg ml 1) were measured 4, 3, 2 and 1 years before ultrasound-guided prostate biopsy. Prostate cancer GS has been determined in radical prostatectomy specimens. Bars and error bars in the histograms represent mean values±s.d. n (GS 6) ¼ 110, n (GS 7) ¼ 74, n (GS 8) ¼ 24. ***Po0.001.
Figure 1. [-2]proPSA differs significantly between cancerous and benign (non-cancerous) prostate. [-2]proPSA levels (pg ml 1) were measured 4, 3, 2 and 1 years before ultrasound-guided prostate biopsy. Bars and error bars in the histograms represent mean values±s.d. n ¼ 173 for benign, n ¼ 208 for cancer. **Po0.01, ***Po0.001.
cancerous and benign prostate 2 and 1 years before diagnosis (Po0.001; Figure 1). When stratified [-2]proPSA levels according to GS obtained in the RP specimens, [-2]proPSA was highest in patients with XGS 8 (mean ¼ 37.00 pg ml 1) and lowest in those with pGS 6 (mean ¼ 11.93 pg ml 1; Figure 2). The difference in [-2]proPSA values between GSX8 and GSp7 was highly significant (Po0.001) already 3 years before diagnosis. However, there was no significant difference between [-2]proPSA levels in GS 7 and GSp6 PCa. Furthermore, we investigated if there is a correlation between extraprostatic extension and the preoperative [-2]proPSA levels measured in a time frame of 4 years before RP. 77.4% of RP specimens were OC, whereas 22.6% of RP specimens showed an extraprostatic extension. The distribution of histopathological status after RP is shown in Table 2. Our data show that preoperative [-2]proPSA values were significantly higher in men with NOC PCa (XpT3a tumors) compared with OC cancers (ppT2c tumors) 4, 3, 2 and 1 years before diagnosis (Po0.01; Figure 3). The highest predictive value of [-2]proPSA to distinguish between OC and extraprostatic extension was found 3 and 2 years before RP. PSA is currently the standard marker used for PCa detection as well as for prediction of PCa aggressiveness. Therefore, we also investigated the impact of PSA in PCa prediction within our patient cohort. In striking contrast to [-2]proPSA, in our patient cohort, PSA was not able to predict PCa from benign cases (Supplementary Figure 1). Also concerning GS (Supplementary Prostate Cancer and Prostatic Disease (2014), 70 – 74
Figure 3. [-2]proPSA differs between organ-confined (OC) and nonorgan confined (NOC) prostate cancers. [-2]proPSA levels (pg ml 1) were measured 4, 3, 2 and 1 years before diagnosis. NOC was defined as extraprostatic tumor extension (XpT3a), OC disease was defined as ppT2c tumors. Bars and error bars in the histograms represent mean values±s.d. n (OC) ¼ 166, n (NOC) ¼ 42. **Po0.01, ***Po0.001.
Figure 2) as well as organ confinement (Supplementary Figure 3), PSA was not able to differ between the groups in our patient cohort. AUC curves reveal a positive predicting value of [-2]proPSA for predicting GSX8 and GSp7 PCa also after controlling for PSA (Table 3). Also concerning extraprostatic extension, AUC show a positive predictive value for [-2]proPSA after controlling for PSA (Table 4). However, multivariate analyses did not show considerable impact on prediction of high-risk and NOC PCa, which might be explained because of the low number of high-risk and NOC cancers included in the study (Tables 5 and 6). In summary, these data provide evidence that [-2]proPSA is able to differ between benign and malignant prostates and predicts PCa aggressiveness several years before diagnosis. However, in our patient cohort PSA did not differ cancerous from benign prostate as well as PCa aggressiveness several years before diagnosis. DISCUSSION For PCa screening, evidence of mortality reduction was shown by PSA-based screening in the European Randomized Study of Screening for Prostate Cancer and the overlapping Goeteborg & 2014 Macmillan Publishers Limited
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
73 AUC for 2[pro]PSA and PSA Gleason Score 8 vs Gleason Score p7
Table 3.
Years before diagnosis 1 2 3 4
Year before diagnosis Years before diagnosis Years before diagnosis Years before diagnosis
AUC 2[pro]PSA
AUC PSA
AUC 2[pro]PSA þ PSA
0.852 0.740 0.777 0.795
0.534 0.604 0.546 0.607
0.823 0.733 0.738 0.765
Abbreviation: AUC, area under the curve.
Table 4. AUC for 2[pro]PSA and PSA organ-confined vs non-organconfined prostate cancer Years before diagnosis 1 2 3 4
year before diagnosis years before diagnosis years before diagnosis years before diagnosis
AUC 2[pro]PSA
AUC PSA
AUC 2[pro]PSA þ PSA
0.623 0.795 0.661 0.638
0.584 0.662 0.561 0.657
0.622 0.780 0.635 0.632
Abbreviation: AUC, area under the curve.
Table 5. Multivariate analysis of 2[pro]PSA and PSA predicting Gleason Score 8 vs Gleason Score p7 Years before diagnosis
OR
P-value
95% CI
1 Year before diagnosis 2[pro]PSA PSA
1.064 0.512
0.006 0.92
1.018–1.112 0.234–1.115
2 Years before diagnosis 2[pro]PSA PSA
1.023 0.985
0.032 0.843
1.002–1.045 0.848–1.144
3 Years before diagnosis 2[pro]PSA PSA
1.055 0.875
0.011 0.362
1.012–1.099 0.656–1.166
4 Years before diagnosis 2[pro]PSA PSA
1.013 1.041
0.266 0.779
0.990–1.035 0.786–1.378
Abbreviations: CI, confidence interval; OR, odds ratio.
trial.25 However, because of increasing use of PSA testing over the last two to three decades, the proportion of low-risk tumors has increased considerably. Treatment of these so-called overdiagnosed cases will inevitably lead to overtreatment and its potential side effects, thereby negatively affecting the patients’ quality of life. Therefore, more specific markers that can accurately detect as well as differentiate patients with insignificant from significant PCa, who need therapy are urgently needed. The identification of molecular forms of PSA, such as ‘free’ PSA, ‘benign’ PSA or ‘intact’ PSA, has been described to enhance the specificity of PSA.8 The precursor form of tPSA, so-called ‘proPSA’, has been identified as promising new biomarker for distinguishing between men with non-cancerous prostatic diseases and with PCa.8,26 In general, proPSA is a component of fPSA in the serum of men and it is expressed in the peripheral zone of the prostate where PCa is located preferentially. Thus, several studies investigated isoforms of [-2]proPSA as potential marker for differing benign prostatic diseases from PCa.8 & 2014 Macmillan Publishers Limited
Table 6. Multivariate analysis of 2[pro]PSA and PSA predicting organ-confined vs non-organ-confined prostate cancer Years before diagnosis
OR
P-value
95% CI
1 Year before diagnosis 2[pro]PSA PSA
0.984 0.978
0.085 0.773
0.967–1.002 0.843–1.136
2 Years before diagnosis 2[pro]PSA PSA
0.974 0.975
0.019 0.586
0.952–0.996 0.891–1.067
3 Years before diagnosis 2[pro]PSA PSA
0.969 0.913
0.070 0.300
0.936–1.003 0.768–1.085
4 Years before diagnosis 2[pro]PSA PSA
1.006 0.715
0.638 0.012
0.983–1.029 0.550–0.930
Abbreviations: CI, confidence interval; OR, odds ratio.
In the present study, we analyzed the impact of [-2]proPSA in a study cohort of 381 patients over a time course of 4 years. In summary, our data reveal for the first time that [-2]proPSA is a marker for PCa aggressiveness already several years before diagnosis. Our data show that [-2]proPSA levels were significantly higher in the cancer group than in the benign group already 4 years before diagnosis. These findings are in line with data from Mikolajczyk et al. describing that [-2]proPSA is a potent marker for differing between PCa and benign prostate.23 Other studies investigating the relationship between PSA levels and [-2]proPSA predictive value found that [-2]proPSA has the best predictive value in a PSA range of 2 to 4 ng ml 1.27 We found that [-2]proPSA is a reliable marker for PCa aggressiveness: [-2]proPSA levels were highest expressed in XGS 8 carcinomas and lowest in pGS 6 carcinomas (Pp0.01). Moreover, our data clearly indicate that the preoperative [-2]proPSA values correlate significantly with extraprostatic extension of PCa. These findings are in line with other studies where the amount of [-2]proPSA was shown to be increased in aggressive forms of PCa.13,16 Another retrospective study, including data from our department showed that [-2]proPSA is able to differentiate aggressive cancers (XGS 7 or greater and/or extracapsular tumor extension) from GS 6 carcinomas.13 Recently, a prospective multicenter study confirmed that the addition of [-2]proPSA could improve PCa detection at initial biopsy in a tPSA range of 2–10 ng ml 1.28 However, none of these studies investigated the impact of [-2]proPSA years before diagnosis of PCa. Our data clearly show that [-2]proPSA is able to differentiate highly significant (Po0.01) between GSX8 and GSp7 tumors already 3 years before diagnosis. The highest predictive value of [-2]proPSA to distinguish between OC and extraprostatic extension was found 2 years before diagnosis of PCa. However, according to our knowledge, our study is the first study reporting a positive predictive value of [-2]proPSA for clinical significant PCa already years before diagnosis (1–4 years). Thus, [-2]proPSA measurement could be used as marker to detect aggressive forms of PCa several years earlier leading to a timely and consequently more effective therapy of the disease. Several studies reveal that 30–80% of cancers diagnosed by screening would never have compromised patients’ health if they had remained undetected. Therefore, our findings raise [-2]proPSA to an prediction marker that is able to differentiate non-aggressive and aggressive PCa where treatment is of urgent need. Consequently, patients harboring tumors with possible aggressive features, which would most likely benefit from treatment, could Prostate Cancer and Prostatic Disease (2014), 70 – 74
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
74 be detected. This would reduce overtreatment of non-aggressive forms of PCa, which could be accompanied by side effects including urinary incontinence, erectile dysfunction or hormonal impacts depending on the choice of treatment. Currently, there are no biomarkers available for early detection of aggressive PCa, however, urinary biomarkers including PCA3, alpha-methylacyl-CoA racemase, the TMPRSS2-ERG fusion gene and microseminoprotein-beta are discussed.29 For the future, [-2]proPSA measurement could be a predictive biomarker for aggressive PCa used for clinical routine because it is easy to use and measurable in blood serum of patients. Strength of this study is the long-term follow-up character and the large patient collective. However, the present study also has several limitations like the lack of a multicenter character, thus randomized controlled multicenter studies are urgently needed to confirm the present findings. Another possible limitation represents the two different biopsy techniques used between 1996– 2000 and 2000–2012 as already described earlier by our group.30 CONCLUSION In summary, we found that [-2]proPSA is an early potent marker for differing between PCa and benign prostate several years before diagnosis. In the present study, we show for the first time the impact of [-2]proPSA to distinguish aggressive from nonaggressive forms of PCa already years before diagnosis. This finding identifies [-2]proPSA as possible predictive biomarker useful in clinical routine to carry out significant PCa that need therapy. [-2]proPSA would help to avoid overtreatment in nonaggressive and thus nonsignificant PCa entities. Further studies with a larger patient collective and with prospective multicenter character are needed to proof these findings.
CONFLICT OF INTEREST The authors declare no conflict of interest.
REFERENCES 1 Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011; 61: 69–90. 2 Heidenreich A, Bellmunt J, Bolla M, Joniau S, Mason M, Matveev V et al. [EAU guidelines on prostate cancer. Part I: screening, diagnosis, and treatment of clinically localised disease]. Actas Urol Esp 2011; 35: 501–514. 3 Nelson WG, De Marzo AM, Isaacs WB. Prostate cancer. N Engl J Med 2003; 349: 366–381. 4 Emiliozzi P, Longhi S, Scarpone P, Pansadoro A, DePaula F, Pansadoro V. The value of a single biopsy with 12 transperineal cores for detecting prostate cancer in patients with elevated prostate specific antigen. J Urol 2001; 166: 845–850. 5 Catalona WJ, Partin AW, Finlay JA, Chan DW, Rittenhouse HG, Wolfert RL et al. Use of percentage of free prostate-specific antigen to identify men at high risk of prostate cancer when PSA levels are 2.51 to 4 ng/ml and digital rectal examination is not suspicious for prostate cancer: an alternative model. Urology 1999; 54: 220–224. 6 Roehl KA, Antenor JA, Catalona WJ. Serial biopsy results in prostate cancer screening study. J Urol 2002; 167: 2435–2439. 7 Strope SA, Andriole GL. Prostate cancer screening: current status and future perspectives. Nat Rev Urol 2010; 7: 487–493. 8 Hori S, Blanchet JS, McLoughlin J. From prostate-specific antigen (PSA) to precursor PSA (proPSA) isoforms: a review of the emerging role of proPSAs in the detection and management of early prostate cancer. BJU Int 2012; 112: 717–728. 9 Ward AM, Catto JW, Hamdy FC. Prostate specific antigen: biology, biochemistry and available commercial assays. Ann Clin Biochem 2001; 38(Pt 6): 633–651.
10 Djavan B. Validity and legacy of prostate-specific antigen (PSA) and PSA-based parameters and isoforms in the new millennium. Eur Urol 2010; 57: 928–929. 11 Chan TY, Mikolajczyk SD, Lecksell K, Shue MJ, Rittenhouse HG, Partin AW et al. Immunohistochemical staining of prostate cancer with monoclonal antibodies to the precursor of prostate-specific antigen. Urology 2003; 62: 177–181. 12 Catalona WJ, Bartsch G, Rittenhouse HG, Evans CL, Linton HJ, Amirkhan A et al. Serum pro prostate specific antigen improves cancer detection compared to free and complexed prostate specific antigen in men with prostate specific antigen 2 to 4 ng/ml. J Urol 2003; 170(6 Pt 1): 2181–2185. 13 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): 2239–2244. 14 Mikolajczyk SD, Millar LS, Marker KM, Wang TJ, Rittenhouse HG, Marks LS et al. Seminal plasma contains "BPSA," a molecular form of prostate-specific antigen that is associated with benign prostatic hyperplasia. Prostate 2000; 45: 271–276. 15 Sokoll LJ, Chan DW, Mikolajczyk SD, Rittenhouse HG, Evans CL, Linton HJ et al. Proenzyme psa for the early detection of prostate cancer in the 2.5-4.0 ng/ml total psa range: preliminary analysis. Urology 2003; 61: 274–276. 16 Sokoll LJ, Sanda MG, Feng Z, Kagan J, Mizrahi IA, Broyles DL et al. A prospective, multicenter, National Cancer Institute Early Detection Research Network study of [-2]proPSA: improving prostate cancer detection and correlating with cancer aggressiveness. Cancer Epidemiol Biomarkers Prev 2010; 19: 1193–1200. 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: 921–927. 18 Loeb S, Sokoll LJ, Broyles DL, Bangma CH, van Schaik RH, Klee GG et al. Prospective, Multi-Center Evaluation of the Beckman Coulter Prostate Health Index Using WHO Calibration. J Urol 2012; 189: 867–870. 19 Oesterling JE, Jacobsen SJ, Chute CG, Guess HA, Girman CJ, Panser LA et al. Serum prostate-specific antigen in a community-based population of healthy men. Establishment of age-specific reference ranges. JAMA 1993; 270: 860–864. 20 Reissigl A, Horninger W, Fink K, Klocker H, Bartsch G. Prostate carcinoma screening in the county of Tyrol, Austria: experience and results. Cancer, 1997; 80: 1818–1829. 21 Bartsch G, Horninger W, Klocker H, Pelzer A, Bektic J, Oberaigner W et al. Tyrol Prostate Cancer Demonstration Project: early detection, treatment, outcome, incidence and mortality. BJU Int 2008; 101: 809–816. 22 Laffin RJ, Chan DW, Tanasijevic MJ, Fischer GA, Markus W, Miller J et al. Hybritech total and free prostate-specific antigen assays developed for the Beckman Coulter access automated chemiluminescent immunoassay system: a multicenter evaluation of analytical performance. Clin Chem 2001; 47: 129–132. 23 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: 6958–6963. 24 Finlay JA, Day JR, Evans CL, Carlson R, Kuus-Reichel K, Millar LS et al. Development of a dual monoclonal antibody immunoassay for total human kallikrein 2. Clin Chem 2001; 47: 1218–1224. 25 Zhu X, Albertsen PC, Andriole GL, Roobol MJ, Schroder FH, Vickers AJ. Risk-based prostate cancer screening. Eur Urol 2012; 61: 652–661. 26 Stephan C, Jung K, Lein M, Diamandis EP. PSA and other tissue kallikreins for prostate cancer detection. Eur J Cancer 2007; 43: 1918–1926. 27 Stephan C, Kahrs AM, Cammann H, Lein M, Schrader M, Deger S et al. A [-2]proPSA-based artificial neural network significantly improves differentiation between prostate cancer and benign prostatic diseases. Prostate 2009; 69: 198–207. 28 Lazzeri M, Haese A, de la TA, Palou RJ, McNicholas T, Lughezzani G et al. Serum isoform [-2]proPSA derivatives significantly improve prediction of prostate cancer at initial biopsy in a total PSA range of 2-10 ng/ml: a multicentric European study. Eur Urol 2013; 63: 986–994. 29 Prensner JR, Rubin MA, Wei JT, Chinnaiyan AM. Beyond PSA: the next generation of prostate cancer biomarkers. Sci Transl Med 2012; 4: 127rv3. 30 Pelzer AE, Colleselli D, Bektic J, Steiner E, Ramoner R, Mitterberger M et al. Pathological features of Gleason score 6 prostate cancers in the low and intermediate range of prostate-specific antigen level: is there a difference? BJU Int 2008; 101: 822–825.
Supplementary Information accompanies the paper on the Prostate Cancer and Prostatic Diseases website (http://www.nature.com/pcan)
Prostate Cancer and Prostatic Disease (2014), 70 – 74
& 2014 Macmillan Publishers Limited
ORIGINAL ARTICLE
[-2]proPSA is an early marker for prostate cancer aggressiveness I Heidegger1, H Klocker1, E Steiner1, V Skradski1, M Ladurner1, R Pichler1, G Scha¨fer2, W Horninger1 and J Bektic1 BACKGROUND: The aim of the study was to evaluate the correlation between preoperative [-2]proPSA, the Gleason Score (GS) and the risk of non-organ-confined (NOC) disease in patients undergoing radical prostatectomy (RP). METHODS: Beckman Coulter Access immunoassay was used to study serum specimens of 381 patients enrolled in a prostate cancer (PCa) early detection program. Inclusion criteria were three or more available serum specimens over 4 years before diagnosis. The values obtained were correlated with the GSs and pathological stages of specimens obtained at RP. RESULTS: [-2]proPSA levels were significantly higher in the cancer group (n ¼ 208) than in the benign group (n ¼ 173). Already 4 years before diagnosis [-2]proPSA differed significantly between PCa and benign prostate in all measured time points, however, highest prediction value was 2 and 1 years before diagnosis (Po0.001). When stratified [-2]proPSA levels according to GS of RP specimens, [-2]proPSA was highest in patients with XGS8 and lowest in those with pGS6. The difference in [-2]proPSA values between GSX8 and GSp7 was highly significant (Po0.01) already 3 years before diagnosis. Investigating the correlation between extraprostatic extension and the preoperative [-2]proPSA levels we found preoperative [-2]proPSA values significantly higher in men with NOC PCa compared with organ-confined (OC) cancers. The highest predictive value of [-2]proPSA to differ between OC and extraprostatic extension was found 3 and 2 years before RP. CONCLUSIONS: Patients with high [-2]proPSA levels in the years before cancer diagnosis are at a higher risk of having aggressive PCas. Thus, the [-2]proPSA should be included in the treatment decision-making for managing screen-detected PCa. Prostate Cancer and Prostatic Disease (2014) 17, 70–74; doi:10.1038/pcan.2013.50; published online 29 October 2013 Keywords: [-2]proPSA; Gleason score; screening; cancer aggressiveness
INTRODUCTION Prostate cancer (PCa) is the leading cancer entity in men and the second leading cause of cancer death among men in European countries.1 Thus, early detection of PCa is one of the most important issues in health care. Currently, PSA is the only broadly used routine molecular marker for diagnosis and prognosis of PCa.2 However, PSA is known to be an organ-specific but not cancerspecific marker, thus PSA values could be also elevated in the benign prostatic hypertrophy, in urogenital infections or after mechanical manipulations such as digital rectal examination.2,3 Consequently, the low-predictive value and a low specificity of total PSA (tPSA) measurement are one of its major drawbacks in the clinical setting thereby limiting early diagnosis of PCa. Emilozzi et al. for example described that only 20–40% of men with PSA values between 4 and 10 ng ml 1 will be found to be diagnosed with PCa.4 Otherwise, although men with PSA values below 4 ng ml 1 were traditionally not biopsied, it has been showed that 20–30% of these will also have PCa.5,6 Therefore, more specific markers that can accurately detect as well as differentiate patients with significant from insignificant PCa are of urgent need. Observational evidence began to point to a substantial burden of associated overdiagnosis and overtreatment triggered by PSA testing. Although the PSA screening test cannot differ between lethal and non-lethal PCa, it allows men with potentially lethal disease to be diagnosed and treated in the early stages of their disease. However, men with nonlethal disease do not benefit from their diagnosis and subsequent treatment. Consequently, those men in whom PCa was never destined to cause clinical symptoms are overdiagnosed. Apart of the psychological distress when
cancer is diagnosed, the short-term and long-term adverse effects of treatment have to be considered.7 In general, PSA belongs to the family of kallikreins encoded by numerous genes on chromosome 19q13.4 produced in the prostatic ductal and acinar epithelium. Initially, PSA is produced with a 17 amino-acid leader sequence to produce the inactive precursor enzyme proPSA. After secreted into the prostatic ducts, proPSA is activated into PSA by hk2 and hk4.8,9 In serum, 70–90% of tPSA is present as a complexed form PSA with a number of different endogenous protease inhibitors, preventing damage of protease activity of PSA. However, 10–30% of tPSA is present in an uncomplexed form. This so-called free PSA (fPSA) is composed of three different types of enzymatically inactive PSA: benign PSA, intact inactive PSA and proPSA.8,10 Preclinical studies found proPSA significantly elevated in serum extracts of patients with PCa. Immunohistochemical staining of 47 PCa cases, for example, showed a positive staining for proPSA in all cases.11 The promising results from preclinical studies led to several clinical studies investigating proPSA as serum marker for PCa detection. The amount of proPSA was shown to be increased in PCa and to be more associated with aggressive forms of the malignant disease.12,13 This includes native proPSA containing a 7 amino-acid pro-leader peptide ([-7]proPSA) and proPSA forms with truncated pro peptides containing 4 ([-4]proPSA) and 2 ([-2]proPSA) amino acids. Among all proPSA isoforms, the [-2]proPSA form has received the most attention because it was the primary form found in tumor extracts and it was preferentially more concentrated in cancer tissue than in benign glands.11 Otherwise, benign PSA seems to be associated
1 Department of Urology, Medical University Innsbruck, Innsbruck, Austria and 2Department of Pathology, Medical University Innsbruck, Innsbruck, Austria. Correspondence: J Bektic, Department of Urology, Medical University Innsbruck, Anichstr. 35, Innsbruck 6020, Austria. E-mail: [email protected] Received 29 April 2013; revised 30 August 2013; accepted 5 September 2013; published online 29 October 2013
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
71 with the benign disease, but its exact role in prostatic diseases still has to be identified.14 Catalona et al. showed that %proPSA measurement significantly improved the specificity for PCa detection in a PSA range between 2 and 4 ng ml 1 thereby decreasing the number of unnecessary biopsies.13 Other research found in a patient cohort of 119 men that by using %proPSA, 75% of cancers might be detected with 59% of unnecessary biopsies being spared using %proPSA.15 Recent studies integrated [-2]proPSA measurements in multivariate models and found an improvement of predictive values between PCa and benign prostate upon integrating [-2]proPSA into their analyses. Sokoll et al., for example, were able to show that [-2]proPSA was superior to fPSA% in the overall diagnostic utility (area under the curve (AUC): 0.73 vs 0.61).16 Multivariate analyses of Jansen et al. showed that the addition of [-2]proPSA to a logistic regression model consisting of tPSA and fPSA% significantly increased PCa prediction value and specificity.17 In addition, publications showed that integration of [-2]proPSA in risk calculation systems improved PCa prediction compared with standard prediction markers PSA, fPSA% and age.8 Loeb et al., for example, recently showed in a prospective multicenter study including 892 patients that the phi-index ([-2]proPSA/fPSAxPSA½) directly reflects the PCa risk.18 To date, only a few studies address proPSA as a biomarker for predicting cancer aggressiveness. Therefore, the present study has been conducted to investigate the impact of [-2]proPSA levels on PCa aggressiveness years before diagnosis. Briefly, we showed for the first time that [-2]proPSA levels are able to distinguish highly aggressive from non-aggressive PCa already 3 years before diagnosis. Thus, [-2]proPSA could serve as an early detection marker for aggressive PCa.
MATERIALS AND METHODS Patient characteristics Our patient cohort included 381 serum samples of (benign 173 and cancer 208) men enrolled in a PCa screening project who underwent at least one ultrasound-guided prostate biopsy at the Department of Urology, Medical University of Innsbruck, between 1993 and 2006 because of elevated PSA levels. All patients were Caucasian race. All patients diagnosed with PCa underwent open radical retropubic prostatectomy (RP) at our institution followed by a histopathological examination of prostate specimens at the Department of Pathology, Medical University Innsbruck. In 1993, a screening project using PSA as the only screening test was launched in the Federal State of Tyrol, Austria. Age-referenced PSA levels in combination with percent-fPSA of less than 22% were used as biopsy criteria.19 Since October 1995, in addition to PSA, percent-fPSA o18% has been used for PCa detection.20 Moreover, patients with PSA levels higher than 10 ng ml 1 or suspicious findings on digital rectal examinations were advised to undergo biopsy. Owing to changes in the biopsy protocol, the number of biopsies obtained increased from 6 cores (January 1993 to October 1995) to 10 (November 1995 to March 2000) and then to 15 cores (March 2000 to July 2006).21 The selection for this study was based on serial serum specimens obtained three or more available serum specimens over 4 years before diagnosis collected between January 1993 and July 2006. Patient characteristics are shown in Table 1. Organ-confined disease (OC) was defined as ppT2c tumors, however, non-organ-confined disease (NOC) was defined as XpT3a tumors (extraprostatic extension). Concerning Gleason Score (GS) 7 patients, there was no differentiation between 3 þ 4 and 4 þ 3 scores because the number of 4 þ 3 tumors was too low.
Sample collection and preparation Samples were obtained from patients serum stored immediately after centrifugation at 70 1C until thawed for the [-2]proPSA measurement. All samples were thawed only once. Storage time ranged from 2 to 15 years with a median storage time of 5 years. PSA and fPSA levels were determined by Access Hybritech PSA and fPSA assays (Beckman Coulter, Fullerton, CA, USA).22 The immunoassay for the & 2014 Macmillan Publishers Limited
Table 1.
Demographic data
Patient number (%) Race Age (years) mean Median
Total
Benign
Cancer
n ¼ 381 (100%) Caucasian 62.3 61.5
n ¼ 173 (45.4%) Caucasian 62 61
n ¼ 208 (54.6%) Caucasian 62.5 62
8.5
17.2
8.0
15.9
11.9
15.8
14.1
17.1
2.6
3.7
3.8
4.0
3.7
3.9
2.8
3.5
2[pro]PSA (pg ml 1) mean 1 Year before diagnosis 2 Years before diagnosis 3 Years before diagnosis 4 Years before diagnosis PSA (ng ml 1) mean 1 Year before diagnosis 2 Years before diagnosis 3 Years before diagnosis 4 Years before diagnosis
detection of [-2]proPSA was a commercial dual monoclonal sandwich assay (Accesss Hybridech pPSA assay, Beckman Coulter) in a microtiter plate format using a biotinylated capture anti-PSA monoclonal antibody and Europium-labeled proPSA-specific monoclonal antibody for detection with a Victor 1420 multilabel counter (Perkin Elmer, Gaithersburg, MD, USA). The assay was performed according to the manufacture instructions. This assay had less than 0.2% cross-reactivity with mature PSA and had a limit of quantitation in serum of 0.015 ng ml 1. The development of this monoclonal antibody to [-2]proPSA has been described previously.23,24
Statistical analyses The statistical distribution of the different data sets was determined using the Kolmogorov–Smirnov test. All data sets followed a Gaussian distribution. Therefore, Student’s t-test was applied to calculate the statistical significance of differences between treatment groups. P-values below 0.05 were considered significant (*Po0.05; **Po0.01; ***Po0.001). Bars and error bars in the histograms represent mean values±s.d. AUC for pre-diagnostic -2[pro]PSA and PSA have been calculated. Multivariate logistic regression analysis was performed to determine associations of pre-diagnostic -2[pro]PSA and PSA and the outcome (SPSS software version 20; SPSS, Chicago, IL, USA).
RESULTS Of our cohort of 381 patients who underwent ultrasound-guided prostate biopsy because of elevated PSA levels, 54.6% (n ¼ 208) were diagnosed with PCa, however, 45.4% (n ¼ 173) of patients showed a benign histology. All patients diagnosed with PCa underwent RP. The median age of PCa patients was 62 years, the median age of benign patients was 60 years (Table 1). All patients (n ¼ 208) with needle biopsy-verified PCa, 52.9% (n ¼ 110) had pGS 6 in RP specimens, 35.6% (n ¼ 74) had GS 7 and 11.5% (n ¼ 24) had GS X8 tumors (Table 2). Mean [-2]proPSA levels (pg ml 1) and also mean PSA levels (ng ml 1) are shown in Table 1. Our data show that [-2]proPSA levels were significantly higher in the cancer group than in the benign (non-cancerous) group (Figure 1). Already 4 years before histological verified diagnosis [-2]proPSA was able to differ significantly between PCa and benign prostate in all measured time points (1–4 years before diagnosis; Po0.01). In addition, our data show the highest prediction value of [-2]proPSA in differing between Prostate Cancer and Prostatic Disease (2014), 70 – 74
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
72 Table 2. Overview about Gleason Scores and histology of prostatectomy specimens (n ¼ 208) Gleason Score prostatectomy specimen n ¼ 208 (100%) 6 110 7 74 X8 24 Histology prostatectomy specimen n ¼ 208 (100%) Organ confined, total n ¼ 166 (77.4%) pT2a 38 pT2b 54 pT2c 74 Non-organ-confined, total n ¼ 42 (22.6%) pT3a XpT3b
(52.9) (35.6) (11.5) (18.2) (24.6) (34.6)
40 (19) 2 (0.6)
Figure 2. [-2]proPSA differs between Gleason Scores (GS) of prostate cancer. [-2]proPSA levels (pg ml 1) were measured 4, 3, 2 and 1 years before ultrasound-guided prostate biopsy. Prostate cancer GS has been determined in radical prostatectomy specimens. Bars and error bars in the histograms represent mean values±s.d. n (GS 6) ¼ 110, n (GS 7) ¼ 74, n (GS 8) ¼ 24. ***Po0.001.
Figure 1. [-2]proPSA differs significantly between cancerous and benign (non-cancerous) prostate. [-2]proPSA levels (pg ml 1) were measured 4, 3, 2 and 1 years before ultrasound-guided prostate biopsy. Bars and error bars in the histograms represent mean values±s.d. n ¼ 173 for benign, n ¼ 208 for cancer. **Po0.01, ***Po0.001.
cancerous and benign prostate 2 and 1 years before diagnosis (Po0.001; Figure 1). When stratified [-2]proPSA levels according to GS obtained in the RP specimens, [-2]proPSA was highest in patients with XGS 8 (mean ¼ 37.00 pg ml 1) and lowest in those with pGS 6 (mean ¼ 11.93 pg ml 1; Figure 2). The difference in [-2]proPSA values between GSX8 and GSp7 was highly significant (Po0.001) already 3 years before diagnosis. However, there was no significant difference between [-2]proPSA levels in GS 7 and GSp6 PCa. Furthermore, we investigated if there is a correlation between extraprostatic extension and the preoperative [-2]proPSA levels measured in a time frame of 4 years before RP. 77.4% of RP specimens were OC, whereas 22.6% of RP specimens showed an extraprostatic extension. The distribution of histopathological status after RP is shown in Table 2. Our data show that preoperative [-2]proPSA values were significantly higher in men with NOC PCa (XpT3a tumors) compared with OC cancers (ppT2c tumors) 4, 3, 2 and 1 years before diagnosis (Po0.01; Figure 3). The highest predictive value of [-2]proPSA to distinguish between OC and extraprostatic extension was found 3 and 2 years before RP. PSA is currently the standard marker used for PCa detection as well as for prediction of PCa aggressiveness. Therefore, we also investigated the impact of PSA in PCa prediction within our patient cohort. In striking contrast to [-2]proPSA, in our patient cohort, PSA was not able to predict PCa from benign cases (Supplementary Figure 1). Also concerning GS (Supplementary Prostate Cancer and Prostatic Disease (2014), 70 – 74
Figure 3. [-2]proPSA differs between organ-confined (OC) and nonorgan confined (NOC) prostate cancers. [-2]proPSA levels (pg ml 1) were measured 4, 3, 2 and 1 years before diagnosis. NOC was defined as extraprostatic tumor extension (XpT3a), OC disease was defined as ppT2c tumors. Bars and error bars in the histograms represent mean values±s.d. n (OC) ¼ 166, n (NOC) ¼ 42. **Po0.01, ***Po0.001.
Figure 2) as well as organ confinement (Supplementary Figure 3), PSA was not able to differ between the groups in our patient cohort. AUC curves reveal a positive predicting value of [-2]proPSA for predicting GSX8 and GSp7 PCa also after controlling for PSA (Table 3). Also concerning extraprostatic extension, AUC show a positive predictive value for [-2]proPSA after controlling for PSA (Table 4). However, multivariate analyses did not show considerable impact on prediction of high-risk and NOC PCa, which might be explained because of the low number of high-risk and NOC cancers included in the study (Tables 5 and 6). In summary, these data provide evidence that [-2]proPSA is able to differ between benign and malignant prostates and predicts PCa aggressiveness several years before diagnosis. However, in our patient cohort PSA did not differ cancerous from benign prostate as well as PCa aggressiveness several years before diagnosis. DISCUSSION For PCa screening, evidence of mortality reduction was shown by PSA-based screening in the European Randomized Study of Screening for Prostate Cancer and the overlapping Goeteborg & 2014 Macmillan Publishers Limited
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
73 AUC for 2[pro]PSA and PSA Gleason Score 8 vs Gleason Score p7
Table 3.
Years before diagnosis 1 2 3 4
Year before diagnosis Years before diagnosis Years before diagnosis Years before diagnosis
AUC 2[pro]PSA
AUC PSA
AUC 2[pro]PSA þ PSA
0.852 0.740 0.777 0.795
0.534 0.604 0.546 0.607
0.823 0.733 0.738 0.765
Abbreviation: AUC, area under the curve.
Table 4. AUC for 2[pro]PSA and PSA organ-confined vs non-organconfined prostate cancer Years before diagnosis 1 2 3 4
year before diagnosis years before diagnosis years before diagnosis years before diagnosis
AUC 2[pro]PSA
AUC PSA
AUC 2[pro]PSA þ PSA
0.623 0.795 0.661 0.638
0.584 0.662 0.561 0.657
0.622 0.780 0.635 0.632
Abbreviation: AUC, area under the curve.
Table 5. Multivariate analysis of 2[pro]PSA and PSA predicting Gleason Score 8 vs Gleason Score p7 Years before diagnosis
OR
P-value
95% CI
1 Year before diagnosis 2[pro]PSA PSA
1.064 0.512
0.006 0.92
1.018–1.112 0.234–1.115
2 Years before diagnosis 2[pro]PSA PSA
1.023 0.985
0.032 0.843
1.002–1.045 0.848–1.144
3 Years before diagnosis 2[pro]PSA PSA
1.055 0.875
0.011 0.362
1.012–1.099 0.656–1.166
4 Years before diagnosis 2[pro]PSA PSA
1.013 1.041
0.266 0.779
0.990–1.035 0.786–1.378
Abbreviations: CI, confidence interval; OR, odds ratio.
trial.25 However, because of increasing use of PSA testing over the last two to three decades, the proportion of low-risk tumors has increased considerably. Treatment of these so-called overdiagnosed cases will inevitably lead to overtreatment and its potential side effects, thereby negatively affecting the patients’ quality of life. Therefore, more specific markers that can accurately detect as well as differentiate patients with insignificant from significant PCa, who need therapy are urgently needed. The identification of molecular forms of PSA, such as ‘free’ PSA, ‘benign’ PSA or ‘intact’ PSA, has been described to enhance the specificity of PSA.8 The precursor form of tPSA, so-called ‘proPSA’, has been identified as promising new biomarker for distinguishing between men with non-cancerous prostatic diseases and with PCa.8,26 In general, proPSA is a component of fPSA in the serum of men and it is expressed in the peripheral zone of the prostate where PCa is located preferentially. Thus, several studies investigated isoforms of [-2]proPSA as potential marker for differing benign prostatic diseases from PCa.8 & 2014 Macmillan Publishers Limited
Table 6. Multivariate analysis of 2[pro]PSA and PSA predicting organ-confined vs non-organ-confined prostate cancer Years before diagnosis
OR
P-value
95% CI
1 Year before diagnosis 2[pro]PSA PSA
0.984 0.978
0.085 0.773
0.967–1.002 0.843–1.136
2 Years before diagnosis 2[pro]PSA PSA
0.974 0.975
0.019 0.586
0.952–0.996 0.891–1.067
3 Years before diagnosis 2[pro]PSA PSA
0.969 0.913
0.070 0.300
0.936–1.003 0.768–1.085
4 Years before diagnosis 2[pro]PSA PSA
1.006 0.715
0.638 0.012
0.983–1.029 0.550–0.930
Abbreviations: CI, confidence interval; OR, odds ratio.
In the present study, we analyzed the impact of [-2]proPSA in a study cohort of 381 patients over a time course of 4 years. In summary, our data reveal for the first time that [-2]proPSA is a marker for PCa aggressiveness already several years before diagnosis. Our data show that [-2]proPSA levels were significantly higher in the cancer group than in the benign group already 4 years before diagnosis. These findings are in line with data from Mikolajczyk et al. describing that [-2]proPSA is a potent marker for differing between PCa and benign prostate.23 Other studies investigating the relationship between PSA levels and [-2]proPSA predictive value found that [-2]proPSA has the best predictive value in a PSA range of 2 to 4 ng ml 1.27 We found that [-2]proPSA is a reliable marker for PCa aggressiveness: [-2]proPSA levels were highest expressed in XGS 8 carcinomas and lowest in pGS 6 carcinomas (Pp0.01). Moreover, our data clearly indicate that the preoperative [-2]proPSA values correlate significantly with extraprostatic extension of PCa. These findings are in line with other studies where the amount of [-2]proPSA was shown to be increased in aggressive forms of PCa.13,16 Another retrospective study, including data from our department showed that [-2]proPSA is able to differentiate aggressive cancers (XGS 7 or greater and/or extracapsular tumor extension) from GS 6 carcinomas.13 Recently, a prospective multicenter study confirmed that the addition of [-2]proPSA could improve PCa detection at initial biopsy in a tPSA range of 2–10 ng ml 1.28 However, none of these studies investigated the impact of [-2]proPSA years before diagnosis of PCa. Our data clearly show that [-2]proPSA is able to differentiate highly significant (Po0.01) between GSX8 and GSp7 tumors already 3 years before diagnosis. The highest predictive value of [-2]proPSA to distinguish between OC and extraprostatic extension was found 2 years before diagnosis of PCa. However, according to our knowledge, our study is the first study reporting a positive predictive value of [-2]proPSA for clinical significant PCa already years before diagnosis (1–4 years). Thus, [-2]proPSA measurement could be used as marker to detect aggressive forms of PCa several years earlier leading to a timely and consequently more effective therapy of the disease. Several studies reveal that 30–80% of cancers diagnosed by screening would never have compromised patients’ health if they had remained undetected. Therefore, our findings raise [-2]proPSA to an prediction marker that is able to differentiate non-aggressive and aggressive PCa where treatment is of urgent need. Consequently, patients harboring tumors with possible aggressive features, which would most likely benefit from treatment, could Prostate Cancer and Prostatic Disease (2014), 70 – 74
[-2]proPSA as early marker for cancer aggressiveness I Heidegger et al
74 be detected. This would reduce overtreatment of non-aggressive forms of PCa, which could be accompanied by side effects including urinary incontinence, erectile dysfunction or hormonal impacts depending on the choice of treatment. Currently, there are no biomarkers available for early detection of aggressive PCa, however, urinary biomarkers including PCA3, alpha-methylacyl-CoA racemase, the TMPRSS2-ERG fusion gene and microseminoprotein-beta are discussed.29 For the future, [-2]proPSA measurement could be a predictive biomarker for aggressive PCa used for clinical routine because it is easy to use and measurable in blood serum of patients. Strength of this study is the long-term follow-up character and the large patient collective. However, the present study also has several limitations like the lack of a multicenter character, thus randomized controlled multicenter studies are urgently needed to confirm the present findings. Another possible limitation represents the two different biopsy techniques used between 1996– 2000 and 2000–2012 as already described earlier by our group.30 CONCLUSION In summary, we found that [-2]proPSA is an early potent marker for differing between PCa and benign prostate several years before diagnosis. In the present study, we show for the first time the impact of [-2]proPSA to distinguish aggressive from nonaggressive forms of PCa already years before diagnosis. This finding identifies [-2]proPSA as possible predictive biomarker useful in clinical routine to carry out significant PCa that need therapy. [-2]proPSA would help to avoid overtreatment in nonaggressive and thus nonsignificant PCa entities. Further studies with a larger patient collective and with prospective multicenter character are needed to proof these findings.
CONFLICT OF INTEREST The authors declare no conflict of interest.
REFERENCES 1 Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin 2011; 61: 69–90. 2 Heidenreich A, Bellmunt J, Bolla M, Joniau S, Mason M, Matveev V et al. [EAU guidelines on prostate cancer. Part I: screening, diagnosis, and treatment of clinically localised disease]. Actas Urol Esp 2011; 35: 501–514. 3 Nelson WG, De Marzo AM, Isaacs WB. Prostate cancer. N Engl J Med 2003; 349: 366–381. 4 Emiliozzi P, Longhi S, Scarpone P, Pansadoro A, DePaula F, Pansadoro V. The value of a single biopsy with 12 transperineal cores for detecting prostate cancer in patients with elevated prostate specific antigen. J Urol 2001; 166: 845–850. 5 Catalona WJ, Partin AW, Finlay JA, Chan DW, Rittenhouse HG, Wolfert RL et al. Use of percentage of free prostate-specific antigen to identify men at high risk of prostate cancer when PSA levels are 2.51 to 4 ng/ml and digital rectal examination is not suspicious for prostate cancer: an alternative model. Urology 1999; 54: 220–224. 6 Roehl KA, Antenor JA, Catalona WJ. Serial biopsy results in prostate cancer screening study. J Urol 2002; 167: 2435–2439. 7 Strope SA, Andriole GL. Prostate cancer screening: current status and future perspectives. Nat Rev Urol 2010; 7: 487–493. 8 Hori S, Blanchet JS, McLoughlin J. From prostate-specific antigen (PSA) to precursor PSA (proPSA) isoforms: a review of the emerging role of proPSAs in the detection and management of early prostate cancer. BJU Int 2012; 112: 717–728. 9 Ward AM, Catto JW, Hamdy FC. Prostate specific antigen: biology, biochemistry and available commercial assays. Ann Clin Biochem 2001; 38(Pt 6): 633–651.
10 Djavan B. Validity and legacy of prostate-specific antigen (PSA) and PSA-based parameters and isoforms in the new millennium. Eur Urol 2010; 57: 928–929. 11 Chan TY, Mikolajczyk SD, Lecksell K, Shue MJ, Rittenhouse HG, Partin AW et al. Immunohistochemical staining of prostate cancer with monoclonal antibodies to the precursor of prostate-specific antigen. Urology 2003; 62: 177–181. 12 Catalona WJ, Bartsch G, Rittenhouse HG, Evans CL, Linton HJ, Amirkhan A et al. Serum pro prostate specific antigen improves cancer detection compared to free and complexed prostate specific antigen in men with prostate specific antigen 2 to 4 ng/ml. J Urol 2003; 170(6 Pt 1): 2181–2185. 13 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): 2239–2244. 14 Mikolajczyk SD, Millar LS, Marker KM, Wang TJ, Rittenhouse HG, Marks LS et al. Seminal plasma contains "BPSA," a molecular form of prostate-specific antigen that is associated with benign prostatic hyperplasia. Prostate 2000; 45: 271–276. 15 Sokoll LJ, Chan DW, Mikolajczyk SD, Rittenhouse HG, Evans CL, Linton HJ et al. Proenzyme psa for the early detection of prostate cancer in the 2.5-4.0 ng/ml total psa range: preliminary analysis. Urology 2003; 61: 274–276. 16 Sokoll LJ, Sanda MG, Feng Z, Kagan J, Mizrahi IA, Broyles DL et al. A prospective, multicenter, National Cancer Institute Early Detection Research Network study of [-2]proPSA: improving prostate cancer detection and correlating with cancer aggressiveness. Cancer Epidemiol Biomarkers Prev 2010; 19: 1193–1200. 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: 921–927. 18 Loeb S, Sokoll LJ, Broyles DL, Bangma CH, van Schaik RH, Klee GG et al. Prospective, Multi-Center Evaluation of the Beckman Coulter Prostate Health Index Using WHO Calibration. J Urol 2012; 189: 867–870. 19 Oesterling JE, Jacobsen SJ, Chute CG, Guess HA, Girman CJ, Panser LA et al. Serum prostate-specific antigen in a community-based population of healthy men. Establishment of age-specific reference ranges. JAMA 1993; 270: 860–864. 20 Reissigl A, Horninger W, Fink K, Klocker H, Bartsch G. Prostate carcinoma screening in the county of Tyrol, Austria: experience and results. Cancer, 1997; 80: 1818–1829. 21 Bartsch G, Horninger W, Klocker H, Pelzer A, Bektic J, Oberaigner W et al. Tyrol Prostate Cancer Demonstration Project: early detection, treatment, outcome, incidence and mortality. BJU Int 2008; 101: 809–816. 22 Laffin RJ, Chan DW, Tanasijevic MJ, Fischer GA, Markus W, Miller J et al. Hybritech total and free prostate-specific antigen assays developed for the Beckman Coulter access automated chemiluminescent immunoassay system: a multicenter evaluation of analytical performance. Clin Chem 2001; 47: 129–132. 23 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: 6958–6963. 24 Finlay JA, Day JR, Evans CL, Carlson R, Kuus-Reichel K, Millar LS et al. Development of a dual monoclonal antibody immunoassay for total human kallikrein 2. Clin Chem 2001; 47: 1218–1224. 25 Zhu X, Albertsen PC, Andriole GL, Roobol MJ, Schroder FH, Vickers AJ. Risk-based prostate cancer screening. Eur Urol 2012; 61: 652–661. 26 Stephan C, Jung K, Lein M, Diamandis EP. PSA and other tissue kallikreins for prostate cancer detection. Eur J Cancer 2007; 43: 1918–1926. 27 Stephan C, Kahrs AM, Cammann H, Lein M, Schrader M, Deger S et al. A [-2]proPSA-based artificial neural network significantly improves differentiation between prostate cancer and benign prostatic diseases. Prostate 2009; 69: 198–207. 28 Lazzeri M, Haese A, de la TA, Palou RJ, McNicholas T, Lughezzani G et al. Serum isoform [-2]proPSA derivatives significantly improve prediction of prostate cancer at initial biopsy in a total PSA range of 2-10 ng/ml: a multicentric European study. Eur Urol 2013; 63: 986–994. 29 Prensner JR, Rubin MA, Wei JT, Chinnaiyan AM. Beyond PSA: the next generation of prostate cancer biomarkers. Sci Transl Med 2012; 4: 127rv3. 30 Pelzer AE, Colleselli D, Bektic J, Steiner E, Ramoner R, Mitterberger M et al. Pathological features of Gleason score 6 prostate cancers in the low and intermediate range of prostate-specific antigen level: is there a difference? BJU Int 2008; 101: 822–825.
Supplementary Information accompanies the paper on the Prostate Cancer and Prostatic Diseases website (http://www.nature.com/pcan)
Prostate Cancer and Prostatic Disease (2014), 70 – 74
& 2014 Macmillan Publishers Limited
