The Prostate 39:123–129 (1999) Elevation of Serum Levels of Urokinase-Type Plasminogen Activator and Its Receptor Is Associated With Disease Progression and Prognosis in Patients With Prostate Cancer Hideaki Miyake,* Isao Hara, Kazuki Yamanaka, Kazuo Gohji, Soichi Arakawa, and Sadao Kamidono Department of Urology, Kobe University School of Medicine, Kobe, Japan BACKGROUND. Several investigators have revealed that urokinase-type plasminogen activator (uPA) and its receptor (uPAR) are overexpressed in serum as well as in tumor tissues in patients with various types of cancer. In this study, we examined whether the serum levels of uPA and uPAR could be used as predictors of the progression and prognosis of prostate cancer. METHODS. Serum levels of uPA and uPAR in 54 healthy controls, 62 patients with benign prostatic hypertrophy (BPH), and 72 patients with prostate cancer were measured by a sandwich enzyme immunoassay. RESULTS. The mean serum levels of uPA and uPAR in patients with prostate cancer were significantly higher than those in healthy controls and patients with BPH. Furthermore, the serum uPA and uPAR levels in prostate cancer patients with metastasis were significantly elevated compared with those in patients without metastasis. Among patients who underwent radical prostatectomy, the serum levels of uPA and uPAR in patients with pathologically organ-confined disease were significantly lower than in those with advanced disease. The overall survival rate of prostate cancer patients with elevated serum levels of either uPA or uPAR, or of both, was significantly lower than that of patients with normal serum levels of uPA and uPAR. CONCLUSIONS. The results of this study indicate that the elevation of serum levels of either uPA or uPAR, or of both, could be used as new predictors of progression and prognosis in patients with prostate cancer. Prostate 39:123–129, 1999. © 1999 Wiley-Liss, Inc. KEY WORDS: prostate cancer; serum; uPA; uPA receptor; progression; prognosis INTRODUCTION Prostate cancer is the most frequent malignancy in men of the United States, and is the second-leading cause of cancer-related death . Recent development in the fields of diagnosis and follow-up of prostate cancer using prostate-specific antigen (PSA) have contributed to early detection and accurate prediction of prognosis. However, despite these advances, more than 50% of patients still show evidence of advanced disease at time of diagnosis, and serum levels of PSA are not correlated with clinical course in some patients with hormone-refractory prostate cancer ; hence, there is a pressing need to develop a novel diagnostic © 1999 Wiley-Liss, Inc. and monitoring system in order to further improve the prognosis of prostate cancer patients. Cancer cells possess a high degree of proteolytic activity, resulting in the acquisition of the ability to degrade and subsequently invade surrounding normal tissues . Several proteolytic enzyme systems are reported to be involved in the degradation of the extracellular matrix and basement membrane, and among them, the urokinase-type plasminogen activa*Correspondence to: Hideaki Miyake, now at the Department of Cancer Endocrinology, British Columbia Cancer Agency, 600 West 10th Ave., Vancouver, British Columbia V5Z 4E6, Canada. Received 5 August 1998; Accepted 30 October 1998 124 Miyake et al. tor (uPA)-mediated plasminogen activation pathway is one of the most important cascades for the invasion of cancer cells . uPA is produced and secreted by various types of cancer cells, and binds with high affinity to its specific cell surface receptor (uPAR). uPA exerts its effect by converting the proenzyme plasminogen into the widely acting serine protease plasmin, which cleaves the extracellular components, including fibronectin, laminin, and collagen. The binding of uPA to uPAR increases its enzymatic activity and further allows a focal and directional proteolysis of the extracellular matrix . Thus, not only uPA but also uPAR may play a crucial role in the invasion and metastasis of cancer cells. Using clinical tissue samples and tumor cell lines, several investigators revealed that the level of uPA in prostate cancer correlates with an aggressive and invasive phenotype [6–8]. In addition, a number of experimental findings clearly demonstrated the important role of the uPA/uPAR system in the malignant progression of human and rodent prostate cancer models [9–12]. Crowly et al.  showed that displacement of uPAR-bound uPA with an active-site mutant uPA blocks spontaneous metastasis of a human prostate cancer cell line in nude mice. Achbarou et al.  showed that introduction of the uPA gene into prostate cancer cells enhanced skeletal metastasis of the cells. These findings suggest that uPA and uPAR may serve as useful markers and prognostic factors for prostate cancer progression. Recently, elevated levels of uPA and uPAR in various body fluids of cancer patients, such as sera, ascites, and pleural effusions, were reported [13–15]. However, there has been no report regarding serum levels of uPA and uPAR in human prostate cancer patients. Therefore, in the present study, we measured the serum levels of uPA and uPAR in patients with prostate cancer, and analyzed the relationship between these levels and several clinicopathological factors. MATERIALS AND METHODS Patient Characteristics Between January 1993–December 1997, serum samples were obtained before any clinical treatment from 72 patients with prostate cancer (age 47–89 years, median 71) and 62 patients with BPH (age 53–84 years, median 69). Diagnosis was confirmed histopathologically by transrectal needle biopsy under the guidance of transrectal ultrasonography. Sera were also collected from 70 healthy males (age 51–82 years, median 67) who had no suspicious findings of prostate cancer in tests including digital rectal examination and mea- surement of serum PSA levels. Informed consent was obtained from all patients and healthy controls for measuring serum uPA and uPAR. Patient characteristics are summarized in Table I. Thirty-five patients received hormonal therapy owing to bone metastases. Thirty-one patients underwent retropubic radical prostatectomy (RRP), and among them, 15 patients had pathologically confirmed organconfined disease (pT1N0M0 or pT2N0M0), and 16 patients had nonorgan-confined disease (pT3 艋 and/or N+). The remaining 6 prostate cancer patients without evidence of distant metastases did not receive radical surgery and underwent hormonal therapy due to severe heart or liver disease. The stage was determined according to the UICC TNM classification , and tumor differentiation was determined by Gleason score . Measurement of Serum Levels of uPA and uPAR The concentrations of uPA and uPAR were determined using quantitative sandwich EIA kits for human uPA and uPAR (American Diagnostica, Greenwich, CT), respectively. For each analysis, 100 l of 1:5 diluted serum samples were added to microtiter plates coated with a purified polyclonal antibody against human uPA or uPAR. Bound uPA or uPAR was detected by an additional biotinylated anti-uPA or anti-uPAR antibody, respectively. After addition of streptavidinconjugated horseradish peroxidase, peroxidasemediated conversion of 3,3⬘,5,5⬘-tetramethylbenzidine was measured as an optical density at 450 nm with a microculture plate reader (Becton Dickinson Labware, Lincoln Park, NJ). All analyses and calibrations were performed in duplicate. Each microtiter plate included recombinant human uPA or uPAR standards for calibration. The blank value was subtracted from the duplicate readings for each standard and sample. A standard curve was created using Stat View 4.02 (Abacus Concepts, Inc., Berkeley, CA) by plotting the logarithm of the mean absorbance of each sample vs. the sample concentration. The rank correlation coefficients were determined between the serum uPA and PSA levels, and between the serum uPAR and PSA levels according to the Spearman rank test with Statgraphics (Statistical Graphics, Rockville, MD). Statistics Differences in the serum levels of uPA and uPAR between controls and prostate cancer patients were calculated by the Mann-Whitney U-test. Differences in elevation of serum uPA and uPAR levels were determined by chi-square test. The overall survival of pros- Serum uPA and uPAR in Prostate Cancer 125 TABLE I. Serum uPA and uPAR Levels in Healthy Controls and Patients With BPH and Prostate Cancer* uPA (ng/ml) Number of subjects (%) Healthy controls BPH Prostate cancer Overall Clinical stage T1 T2 T3 T4 Metastasisa Negative Positive Lymph node metastasis Negative Positive Bone metastasis Negative Positive Gleason score 2–4 5–7 8–10 Mean ± SD (range) uPAR (ng/ml) P value Mean ± SD (range) P value 70 (100) 62 (100) 362 ± 90 (198–511) 389 ± 163 (201–883) 72 (100) 453 ± 188 (245–971) 5.5 ± 2.8 (1.5–11.4) 6 (8) 35 (49) 27 (38) 4 (5) 377 ± 172 (245–702) 411 ± 156 (261–883) 523 ± 199 (257–971) 451 ± 85 (401–636) <0.05 3.9 ± 2.7 (1.6–6.2) 4.8 ± 2.7 (1.5–9.7) 6.9 ± 3.2 (1.8–11.4) 6.5 ± 2.4 (2.1–8.8) 25 (35) 47 (65) 391 ± 177 (257–935) 486 ± 194 (245–971) <0.05 4.2 ± 2.6 (1.5–10.7) 6.2 ± 2.9 (1.6–11.4) <0.01 54 (75) 18 (25) 449 ± 193 (245–949) 465 ± 180 (266–971) NS 5.4 ± 2.8 (1.5–10.9) 5.8 ± 2.4 (1.6–11.4) NS 37 (51) 35 (49) 400 ± 179 (245–935) 509 ± 191 (257–971) <0.05 4.6 ± 2.2 (1.5–10.7) 6.4 ± 3.0 (1.7–11.4) <0.01 20 (28) 23 (32) 29 (40) 431 ± 175 (245–889) 459 ± 185 (269–959) 463 ± 192 (277–971) NS 5.1 ± 2.3 (1.5–10.1) 5.6 ± 2.7 (1.6–11.4) 5.7 ± 2.9 (1.6–11.0) NS NS 3.9 ± 1.2 (1.4–5.5) 4.2 ± 1.5 (1.7–7.2) <0.05 NS <0.005 <0.01 *NS, not significant. a Either lymph node metastasis or bone metastasis, or both. tate cancer patients was examined by the KaplanMeier method, and differences were evaluated by logrank test. The prognostic significance of some factors was assessed by the Cox proportional hazards regression model. A difference with P < 0.05 was considered significant. in prostate cancer patients with metastasis (bone and/ or lymph node metastasis) were significantly higher than those in patients without metastasis (P < 0.05 and P < 0.01, respectively). However, serum levels of uPA and uPAR in prostate cancer patients were not related to Gleason score. RESULTS Elevation of Serum uPA and uPAR Serum uPA and uPAR Levels In healthy controls, the mean serum levels of uPA and uPAR were 362.4 ± 89.7 ng/ml (range, 198–511 ng/ml) and 3.9 ± 1.2 ng/ml (range, 1.4–5.5 ng/ml), respectively, and the variations in levels of both uPA and uPAR were not related to age (data not shown). There were no significant differences in serum uPA or uPAR levels between healthy controls and patients with BPH. In contrast, serum levels of uPA and uPAR in patients with prostate cancer were significantly higher than those in patients with BPH (P < 0.05 and P < 0.005, respectively) (Table I). Serum uPA and uPAR levels in patients with T3 or T4 prostate cancer were significantly higher than those in patients with T1 or T2 prostate cancer (P < 0.05 and P < 0.01, respectively). Serum levels of uPA and uPAR The cutoff values for normal serum levels of uPA and uPAR were determined by the mean level of healthy controls plus 2 standard deviations; therefore, the normal values of uPA and uPAR were less than 541.8 ng/ml and 6.3 ng/ml, respectively. According to these criteria, the specificities of the elevation of serum levels of uPA and uPAR in healthy controls were 100% and 100%, respectively, and those in patients with BPH were 90% and 95%, respectively (data not shown). The elevation of serum levels of uPA and uPAR in patients with prostate cancer and BPH is shown in Table II. The elevation of serum levels of uPA and uPAR in prostate cancer patients was significantly higher than in patients with BPH (P < 0.005 and P < 0.001, respectively). In addition, the differences of elevation of serum uPA and uPAR levels between 126 Miyake et al. TABLE II. Elevation of Serum Levels of uPA and uPAR in Patients With BPH and Prostate Cancer* Elevation (%) Number of subjects (%) BPH Prostate cancer Overall Clinical stage T1 T2 T3 T4 Metastasisa Negative Positive Gleason score 2–4 5–7 8–10 uPA (%) 62 (100) uPAR (%) P value 6 (10) uPA or uPAR (%) P value P value 3 (5) <0.005 8 (13) <0.001 <0.001 72 (100) 25 (35) 29 (40) 39 (54) 6 (8) 35 (49) 27 (38) 4 (5) 1 (17) 7 (20) 16 (59) 1 (25) <0.005 0 (0) 6 (17) 21 (78) 2 (50) <0.001 1 (17) 12 (34) 24 (89) 2 (50) <0.001 25 (35) 47 (65) 4 (16) 21 (45) <0.05 4 (16) 25 (53) <0.005 7 (28) 32 (68) <0.005 20 (28) 23 (32) 29 (40) 5 (25) 8 (35) 12 (41) NS 5 (25) 10 (43) 14 (48) NS 8 (40) 13 (57) 18 (62) NS *NS, not significant. a Either lymph node metastasis or bone metastasis, or both. TABLE III. Serum uPA and uPA Levels in Patients With BPH and Prostate Cancer Who Underwent Radical Prostatectomy* uPA (ng/ml) BPH Organ-confined prostate cancera Nonorgan-confined prostate cancerb Number of subjects Mean ± SD 62 389 ± 163 15 401 ± 147 16 517 ± 166 P value NS uPAR (ng/ml) Mean ± SD 4.2 ± 1.5 P value NS 4.4 ± 2.5 <0.05 6.4 ± 2.7 Elevation (%) uPA (%) 6 (10) P value NS 3 (20) <0.05 4 (25) uPAR (%) 3 (5) P value NS 2 (13) NS 8 (50) uPA or uPAR (%) 8 (13) P value NS 4 (27) <0.05 10 (63) <0.05 *NS, not significant. a pT1N0M0 or pT2N0M0. b pT3 艋 or N+. prostate cancer patients with and without metastasis were also significant (P < 0.05 and P < 0.005, respectively). However, there were no significant differences in elevation of serum uPA and uPA levels between patients with Gleason grade 2–4 prostate cancer and patients with Gleason grade 5–7 or Gleason grade 8–10 prostate cancer. Relationship Between Pathological Findings and Serum Levels of uPA and uPAR There were no significant differences in serum uPA or uPAR levels between patients with BPH and organconfined prostate cancer (pT1N0M0 or pT2N0M0); however, the differences in serum uPA and uPAR lev- els between patients with organ-confined disease and patients with nonorgan-confined disease (pT3 艋 or N+) were significant (P < 0.05 and P < 0.05, respectively) (Table III). There were no significant differences in elevation of serum level of uPA among BPH, organ-confined disease, and nonorgan-confined disease; however, the difference in the elevation of uPAR between organconfined disease and nonorgan-confined disease was significant (P < 0.05) (Table III). Relationship of Serum uPA and uPAR to PSA The relationships between serum uPA and PSA levels and between serum uPAR and PSA levels in pros- Serum uPA and uPAR in Prostate Cancer Fig. I. Relationships between serum uPA, uPAR, and PSA. A: Serum uPA level and PSA. B: Serum uPAR level and PSA. tate cancer patients are shown in Figure 1A,B, respectively. There was no correlation between serum uPA level and PSA (r = 0.08) or between serum uPAR and PSA (r = 0.01). Survival of Prostate Cancer Patients According to Elevation of Serum uPA and uPAR The overall survival rate of prostate cancer patients with an elevated level of serum uPA was not significantly different from that of patients with a normal level (Fig. 2A), and the difference in overall survival rate between prostate cancer patients with an elevated serum level of uPAR and patients with a normal level was also not significant (Fig. 2B). However, the overall 127 Fig. 2. Overall survival rates of prostate cancer patients according to serum uPA and uPAR status. A: uPA. B: uPAR. C: Either uPA or uPAR, or both. The overall survival rates of patients with elevated serum levels of either uPA or uPAR, or of both, were significantly lower than the overall survival rate of patients with normal levels of both uPA and uPAR. survival rate of prostate cancer patients with elevated serum levels of either uPA or uPAR, or of both, was significantly lower than the survival rate of patients with normal levels of both uPA and uPAR (P < 0.05) (Fig. 2C). To compare the predictive value of serum uPA and uPAR status in prognosis, univariate and multivariate analyses with the Cox regression hazards model were performed (Table IV); however, the elevation of serum levels of either uPA or uPAR, or of both, was not an independent prognostic predictor. The analyses further revealed that bone and/or lymph node metastasis was strongly associated with overall survival (univariate analysis, P < 0.01; multivariate analysis, P < 0.05), 128 Miyake et al. TABLE IV. Association of Various Factors With Overall Survival Determined by Cox Univariate and Multivariate Proportional Hazards Model* Univariate Variables Serum uPA and uPAR status (normal vs. elevated)a Age (years) (<61 vs. ⱖ61) Clinical stage (T1 vs. T2 vs. T3 vs. T4) Gleason score (<8 vs. ⱖ8) PSA level (<10 ng/ml vs. ⱖ10 ng/ml) Metastasis (negative vs. positive)b Multivariate Hazard ratio 95% CI P value Hazard ratio 95% CI P value 1.89 1.19 1.98 1.66 1.39 2.02 0.97–2.88 0.58–2.69 1.11–3.20 0.85–2.81 0.61–2.96 1.12–3.23 0.044 0.74 0.041 0.12 0.16 0.0079 0.64 1.34 2.14 0.61 1.89 3.56 0.21–1.82 0.55–3.11 0.88–2.21 0.21–1.70 0.70–1.91 1.19–8.67 0.091 0.30 0.16 0.081 0.15 0.019 *CI, confidence interval. a Elevation of serum levels of either uPA or uPAR, or both. b Either lymph node metastasis or bone metastasis, or both. regardless of uPA and uPAR status, age, PSA level, clinical stage, and Gleason score. DISCUSSION A number of clinical studies have demonstrated that uPA and/or uPAR are overexpressed in several types of human malignant tumors, including prostate cancer, and that the expression level of the uPA system closely correlates with tumor progression [6,8,18,19]. Several experimental findings also demonstrated that the uPA-uPAR system plays a crucial role in invasion and metastasis of prostate cancer cells both in vitro and in vivo [9–12]. Furthermore, recent studies revealed that serum levels of uPA and uPAR appear to be indicative of malignancy in gastric and ovarian cancer [13–15]. Therefore, in the present study, serum levels of uPA and uPAR were examined in healthy controls, and in BPH and prostate cancer patients, and an investigation of whether they could be used as predictors of disease progression and prognosis in prostate cancer patients was conducted. In this series, we showed for the first time that both serum uPA and uPAR levels in patients with prostate cancer were significantly higher than levels in patients with BPH or in healthy controls. In addition, the elevations of serum uPA and uPAR were much higher in prostate cancer patients with metastases than in those without metastases. Soff et al.  and Rabbani et al.  demonstrated that prostate cancer metastases are suppressed by the inhibition of the uPA system, using human and rat prostate cancer cells, respectively. Considering these findings and our present results, we conclude that the uPA and uPAR system contributes to prostate cancer development, and that serum uPA and uPA levels can be used to predict prostate cancer progression. In order to accurately distinguish prostate cancer patients with organ-confined disease from those with nonorgan-confined disease preoperatively, several methods have been introduced into clinical practice, such as computerized tomography, magnetic resonance imaging, PSA density, and histological grading of biopsy specimens . However, even if these methods are combined, it remains difficult to correctly identify patients with nonorgan-confined disease. In the present study, the elevation of serum levels of either or both uPA and uPAR in patients with nonorgan-confined disease was significantly higher than that in patients with organ-confined disease. These findings indicate that levels of serum uPA and uPAR are useful markers for the detection of patients with nonorgan-confined disease. In this series, no significant correlation was observed between serum uPA and PSA levels, or between serum uPAR and PSA levels. Therefore, serum uPA and uPAR levels can be regarded as clinical factors independent of serum PSA. It is of particular interest whether the elevations of serum uPA and/or uPAR concentrations are associated with the poor prognosis of prostate cancer patients. In this study, we showed that the overall survival rate of prostate cancer patients with either serum uPA or uPAR elevation, or both serum uPA and uPAR elevation, was significantly lower than that of patients with normal levels of uPA and uPAR. These findings suggest that measurement of serum uPA and uPAR may provide additional information in determining prognosis and patient treatment strategy, i.e., aggressive therapy should be performed in patients with elevated serum levels of uPA and/or uPAR, whereas patients with normal levels of both uPA and uPAR may not need to undergo intensive additional therapies if the initial treatment is successfully performed. Serum uPA and uPAR in Prostate Cancer In conclusion, our results suggest that uPA and/or uPAR expression is associated with the progression of prostate cancer, and that serum uPA and uPAR levels indicate the degree of prostate cancer progression and patient prognosis. Hence, measurement of serum uPA and uPAR could serve as useful adjuncts to conventional diagnostic tools. REFERENCES 1. Parker SL, Tong T, Bolden S, Wingo PA. Cancer statistics, 1996. CA Cancer J Clin 1996;45:5–27. 2. Oesterling JE. Prostate specific antigen: a critical assessment of the most useful tumor marker for adenocarcinoma of the prostate. J Urol 1991;145:907–923. 3. Fidler IJ. Critical factors in the biology of human cancer metastasis: twenty-eight G.H.A. Clowes Memorial Award Lecture. Cancer Res 1990;50:6130–6138. 4. Stetler-Stevenson WG, Aznavoorian S, Liotta LA. Tumor cell interactions with the extracellular matrix during invasion and metastasis. Annu Rev Cell Biol 1993;9:541–573. 5. Testa JE, Quigley JP. The role of urokinase-type plasminogen activator in aggressive tumor cell behavior. Cancer Metastasis Rev 1990;9:353–367. 6. Gaylis FD, Keer HN, Wilson MJ, Kwaan HC, Sinha AA, Kozlowski JM. Plasminogen activators in human prostate cancer cell lines and tumors. Correlation with aggressive phenotype. J Urol 1988;142:193–198. 7. Keer HN, Gaylis FD, Kozlowski JM, Kwaan HC, Bauer KD, Akouchi A, Sinha AA, Wilson MJ. Heterogeneity in cellular levels of expression of plasminogen activators in human prostate cancer cell lines. Invasive cells contain and secrete more uPA. Prostate 1991;18:201–204. 8. Festuccia C, Vincentini C, di Pasquale AB, Aceto G, Zazzeroni F, Miano L, Bologna M. Plasminogen activator activities in shortterm tissue cultures of benign prostatic hyperplasia and prostatic carcinoma. Oncol Res 1995;7:131–138. 9. Crowly CW, Cohen RL, Liu G, Shuman MA, Levinson AD. Prevention of metastasis by inhibition of the urokinase receptor. Proc Natl Acad Sci USA 1993;90:5021–5025. 10. Achbarou A, Kaiser S, Tremblay G, Ste-Marie L-G, Brodt P, Goltzman D, Rabbani SA. Urokinase overproduction results in 129 increased skeletal metastasis by prostate cancer cells in vivo. Cancer Res 1994;54:2372–2377. 11. Soff GA, Sanderowitz J, Gately S, Verrusio E, Weiss I, Brem S, Kwaan HC. Expression of plasminogen activator inhibitor type 1 by human prostate carcinoma cells inhibits primary tumor growth, tumor-associated angiogenesis, and metastasis to lung and liver in an athymic mouse model. J Clin Invest 1995;96: 2593–2600. 12. Rabbani SA, Harakidas P, Davidson DJ, Henkin J, Mazar AP. Prevention of prostate-cancer metastasis in vivo by a novel synthetic inhibitor of urokinase-type plasminogen activator (uPA). Int J Cancer 1995;63:840–845. 13. Mizukami IF, Faulkner NE, Gyetko MR, Sitrin RG, Todd RF III. Enzyme-linked immunoabsorbent assay detection of a soluble form of urokinase plasminogen activator receptor in vivo. Blood 1995;86:203–211. 14. Rahr HB, Sørensen JV, Larsen JF, Svenden Jensen F, Bredahl C. Plasminogen activators and plasminogen activator inhibitor before and after surgery in patients with and without gastric malignancy. Haemostasis 1995;25:248–256. 15. Sier CFM, Stephens R, Bizik J, Mariani A, Bassan M, Pedersen N, Frigerio L, Ferrari A, Danø K, Brunner N, Blasi F. The level of urokinase-type plasminogen activator receptor is increased in serum of ovarian cancer patients. Cancer Res 1998;58:1843–1849. 16. Spiessl B, Beahrs OH, Hermanek P, Hutter RVP, Scheibe O, Sobin LH, Wagner G. UICC, TNM atlas, 3rd ed. Berlin: Springer-Verlag; 1992. 17. Gleason DF. Histologic grading and clinical staging of carcinoma of the prostate. In: Tannenbaum M, editor. Urologic pathology: the prostate. Philadelphia: Lea and Febiger; 1977. p 171–197. 18. Duffy MJ, Reilly D, O’Sullivan C, O’Higgins N, Fennelly JJ, Andreasen P. Urokinase plasminogen activator, a new and independent prognostic marker in breast cancer. Cancer Res 1990;50:6827–6829. 19. Landau BJ, Kwaan HC, Verrusio EN, Brem SS. Elevated levels of urokinase-type plasminogen activator and plasminogen activator inhibitor type-1 in malignant human brain tumors. Cancer Res 1994;54:1105–1108. 20. Partin AW, Yoo J, Carter HB, Pearson JD, Chan DW, Epstein JI, Walsh PC. The use of prostatic specific antigen, clinical stage and Gleason score to predict pathological staging in men with localized prostate cancer. J Urol 1993;150:110–114.