Advertisement

Cathepsin E, Maspin, Plk1, and Survivin Are Promising Prognostic Protein Markers for Progression in Non-Muscle Invasive Bladder Cancer

      Bladder cancer is a common cancer with particularly high recurrence after transurethral resection. In this study, we investigated the prognostic value of the protein expression of cathepsin E, maspin, polo-like kinase 1 (Plk1), and survivin in patients with stage Ta and T1 urothelial carcinomas. Transcripts from the four genes encoding these proteins were previously included in gene expression signatures for outcome prediction for Ta/T1 bladder cancer. We used three different tissue microarrays with 693 non-muscle invasive urothelial carcinomas from Danish, Swedish, and Spanish patient cohorts with long-term follow-up. Protein expression was measured by immunohistochemistry, and antibody specificity was validated by Western blotting. In the Danish patient cohort, we found the expression of cathepsin E, maspin, Plk1, and survivin to be significantly associated with progression to stage T2 to T4 bladder cancer (for each marker: log-rank test; P < 0.001). Multivariate Cox regression analysis identified cathepsin E (P < 0.001), Plk1 (P = 0.021), maspin (P = 0.001), and survivin (P = 0.001) as independent prognostic markers. Furthermore, maspin, survivin, and cathepsin E expression significantly subgrouped patients already stratified by European Organization for Research and Treatment of Cancer risk scores. Finally, we successfully validated the results in tumors from 410 patients from both Sweden and Spain. We conclude that all four protein markers may have prognostic value in non-muscle invasive bladder cancer for guiding optimal treatment of patients. Additional prospective studies are needed for further validation of the clinical relevance of this marker panel.
      Bladder cancer is the fifth most common cancer in the Western world. In the US alone, an estimated 70,000 new patients were diagnosed in 2010.
      • Jemal A.
      • Siegel R.
      • Xu J.
      • Ward E.
      Cancer statistics, 2010.
      About 70% of urothelial bladder tumors present as non-muscle invasive (NMIBC) and 30% as muscle-invasive cancers (MIBC) in the Western hemisphere. The risk of recurrence after primary, transurethral resection of the tumor is very high: in 60% of cases, at least one recurrence is diagnosed within 5 years.
      • Kiemeney L.A.
      • Witjes J.A.
      • Verbeek A.L.
      • Heijbroek R.P.
      • Debruyne F.M.
      The clinical epidemiology of superficial bladder cancer Dutch South-East Cooperative Urological Group.
      The non-muscle invasive tumors of the bladder form a heterogeneous group, spanning from noninvasive papillary tumors with low-grade dysplasia to superficially invasive tumors with high-grade dysplasia and concomitant carcinoma in situ. In the latter group, the risk of disease progression to a muscle-invasive stage reaches 60% at long-term follow-up.
      • Cookson M.S.
      • Herr H.W.
      • Zhang Z.F.
      • Soloway S.
      • Sogani P.C.
      • Fair W.R.
      The treated natural history of high risk superficial bladder cancer: 15-year outcome.
      • Cheng L.
      • Cheville J.C.
      • Neumann R.M.
      • Leibovich B.C.
      • Egan K.S.
      • Spotts B.E.
      • Bostwick D.G.
      Survival of patients with carcinoma in situ of the urinary bladder.
      The histopathological parameters used in the clinic cannot precisely predict the individual disease course, and no molecular markers are currently in clinical use for predicting tumor recurrence or later progression.
      • Ehdaie B.
      • Theodorescu D.
      Predicting tumor outcomes in urothelial bladder carcinoma: turning pathways into clinical biomarkers of prognosis.
      Bladder cancer patients must therefore be monitored thoroughly for disease recurrence and progression by urine and cystoscopy examinations. This involves a heavy cost to society, and bladder cancer carries the highest cost per patient from diagnosis to death; the estimated total in the US exceeds 3.4 billion dollars annually.
      • Hong Y.M.
      • Loughlin K.R.
      Economic impact of tumor markers in bladder cancer surveillance.
      The European Organization for Research and Treatment of Cancer (EORTC) risk tables for estimation of recurrence and progression risk are routinely used in the clinic.
      • Sylvester R.J.
      • van der Meijden A.P.
      • Oosterlinck W.
      • Witjes J.A.
      • Bouffioux C.
      • Denis L.
      • Newling D.W.
      • Kurth K.
      Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials.
      They are based on clinical and histopathological risk factors. However, molecular markers may add further prognostic value to the risk tables, and their use may enhance our ability to individualize treatment, an ambition that has gained momentum in bladder cancer research during the past decade.
      • Dyrskjot L.
      Classification of bladder cancer by microarray expression profiling: towards a general clinical use of microarrays in cancer diagnostics.
      One possibility is to link gene expression to known outcomes for groups of bladder cancer patients,
      • Dyrskjot L.
      • Zieger K.
      • Real F.X.
      • Malats N.
      • Carrato A.
      • Hurst C.
      • Kotwal S.
      • Knowles M.
      • Malmstrom P.U.
      • de la Torre M.
      • Wester K.
      • Allory Y.
      • Vordos D.
      • Caillault A.
      • Radvanyi F.
      • Hein A.M.
      • Jensen J.L.
      • Jensen K.M.
      • Marcussen N.
      • Orntoft T.F.
      Gene expression signatures predict outcome in non-muscle-invasive bladder carcinoma: a multicenter validation study.
      • Blaveri E.
      • Simko J.P.
      • Korkola J.E.
      • Brewer J.L.
      • Baehner F.
      • Mehta K.
      • Devries S.
      • Koppie T.
      • Pejavar S.
      • Carroll P.
      • Waldman F.M.
      Bladder cancer outcome and subtype classification by gene expression.
      Our group has previously identified a gene expression signature that successfully predicts subsequent disease progression to MIBC.
      • Dyrskjot L.
      • Zieger K.
      • Kruhoffer M.
      • Thykjaer T.
      • Jensen J.L.
      • Primdahl H.
      • Aziz N.
      • Marcussen N.
      • Moller K.
      • Orntoft T.F.
      A molecular signature in superficial bladder carcinoma predicts clinical outcome.
      This gene expression signature proved to be highly significantly associated with progression-free survival (P < 0.001) and cancer-specific survival (P < 0.001) in a multicenter validation study of tumors from 294 patients with NMIBC. Furthermore, the expression signature was shown to be an independent, statistically significant variable associated with disease progression (hazard ratio: 2.3; P = 0.007).
      • Dyrskjot L.
      • Zieger K.
      • Real F.X.
      • Malats N.
      • Carrato A.
      • Hurst C.
      • Kotwal S.
      • Knowles M.
      • Malmstrom P.U.
      • de la Torre M.
      • Wester K.
      • Allory Y.
      • Vordos D.
      • Caillault A.
      • Radvanyi F.
      • Hein A.M.
      • Jensen J.L.
      • Jensen K.M.
      • Marcussen N.
      • Orntoft T.F.
      Gene expression signatures predict outcome in non-muscle-invasive bladder carcinoma: a multicenter validation study.
      However, the expression level of the transcripts does not necessarily associate with the expression level of the encoded proteins.
      • Orntoft T.F.
      • Thykjaer T.
      • Waldman F.M.
      • Wolf H.
      • Celis J.E.
      Genome-wide study of gene copy numbers, transcripts, and protein levels in pairs of non-invasive and invasive human transitional cell carcinomas.
      Our aim in the present study was therefore to investigate the prognostic value of four of the most differentially expressed transcripts in the expression signature (BIRC5, SERPINB5, PLK1, and CTSE) at the protein level in a cohort of 283 Danish patients with NMIBC (182 Ta and 101 T1). After this, the markers were validated using TMAs from Sweden and Spain holding 141 and 269 primary tumors from NMIBC patients with long-term follow-up.

      Materials and Methods

      Patient Follow-Up

      Patients were followed by control cystoscopies in a routine schedule. Follow-up was censored at the time of the most recent cystoscopy. The follow-up included progression-free survival, disease-specific survival, and overall survival. Tumor progression was defined as progression to stage T2 to T4 verified by pathological examination, and measured from the time of primary surgery to the time of the proven event. Patients were censored due to death or discontinued follow-up after 5 years without disease recurrences.

      Danish Patient Cohort and Tumor Material

      All patients gave their written informed consent, and the study was approved by the Central Denmark Region Committees on Biomedical Research Ethics (1994/2920). The 283 patients selected for this retrospective study debuted with non-muscle invasive urothelial bladder tumors between 1979 and 2007. Patients who underwent cystectomy before pathological evidence of progression were excluded from the study. We retrieved formalin-fixed, paraffin-embedded (FFPE) tumor blocks from the primary tumors of the 283 selected patients at the Institute of Pathology, Aarhus University Hospital, NBG, Denmark. The original hematoxylin-eosin (HE) stained sections were reviewed by one experienced uropathologist (B.P.U.) who re-evaluated the stage and grade of each tumor blinded to the original diagnosis. Grading was performed according to World Health Organization 2004 classification. Furthermore, the uropathologist identified the area of each paraffin-embedded bladder tumor from which a 0.6-mm tissue core was taken. The cores were placed in a recipient paraffin block using a manual tissue-microarrayer (Beecher Instruments, Sun Prairie, WI). This was done according to the method developed by Kononen et al.
      • Kononen J.
      • Bubendorf L.
      • Kallioniemi A.
      • Barlund M.
      • Schraml P.
      • Leighton S.
      • Torhorst J.
      • Mihatsch M.J.
      • Sauter G.
      • Kallioniemi O.P.
      Tissue microarrays for high-throughput molecular profiling of tumor specimens.

      Swedish Patient Cohort and Tumor Material

      Primary tumors from 141 patients (67 Ta and 74 T1, 47 low grade and 94 high grade) were positioned on TMAs with 1-mm core biopsies. The patients were treated between 1984 and 2005 in the Uppsala University Hospital, Uppsala, Sweden.

      Spanish Patient Cohort and Tumor Material

      Primary tumors from 269 patients (21 Ta and 248 T1, 103 low grade and 166 high grade) were positioned on TMAs with either 0.6-mm or 1-mm core biopsies. The patients were treated between 1994 and 2008 in the Hospital Central de Asturias and Fundació Puigvert. For further clinical and histopathological characteristics of the three patient cohorts, see Table 1.
      Table 1Clinical and Histopathological Characteristics for the Danish, Swedish, and Spanish Patient Cohorts
      Danish cohort (n = 283)Swedish cohort (n = 141)Spanish cohort (n = 269)
      Median follow-up time in months for all patients (range)103 (2–263)72 (1–193)99 (3–205)
      Median follow-up time in months for progressing patients (range)22 (1–210)21 (1–96)18 (3–126)
      Median follow-up time in months for non-progressing patients (range)117 (53–250)80 (60–193)104 (60–205)
      Median age (range)68 (32–86)70 (31–96)68 (25–89)
      Male–female ratio3.643.906.47
      Stage
       Ta182 (64%)67 (48%)21 (8%)
       T1101 (36%)74 (52%)248 (92%)
      Grading WHO 2004
       Low grade + PUNLMP183 (65%)47 (33%)103 (38%)
       High grade100 (35%)94 (67%)166 (62%)
      Tumor size
       <3 cm177 (63%)NA187 (70%)
       ≥3 cm72 (25%)78 (29%)
       Unspecified34 (12%)4 (1%)
      Growth pattern
       Papillary248 (88%)NA230 (86%)
       Solid20 (7%)22 (8%)
       Mixed14 (5%)17 (6%)
      Multiplicity
       One169 (60%)NANA
       More87 (31%)
       Unspecified27 (9%)
      Concomitant CIS
       Yes95 (34%)NA70 (26%)
       No185 (65%)199 (74%)
       Unspecified3 (1%)
      Adjuvant therapy (BCG or MMC)
       Yes70 (25%)NA193 (72%)
       No213 (75%)76 (28%)
      Number of progression events to stage pT2-4 bladder cancer
       All113 (40%)38 (27%)65 (24%)
       Ta54 (30%)10 (15%)0 (0%)
       T159 (58%)28 (38%)65 (26%)
      CIS, carcinoma in situ; PUNLMP, papillary urothelial neoplasm of low malignant potential; NA, not applicable; WHO, World Health Organization.

      Immunohistochemistry and TMA Scoring

      After blocking the endogenous peroxidase with hydrogen peroxide for 10 minutes, the cores were demasked with heat-induced epitope retrieval. Nonspecific binding was blocked with 1% bovine serum albumin (Albumin Fraction V; AppliChem, Darmstadt, Germany). The TMA slide was incubated with the primary antibody for 60 minutes at room temperature. The DAKO EnVision+ visualization system (DakoCytomation, Glostrup, Denmark) was used to visualize the antigen through a chromogen reaction. Immunohistochemistry was performed essentially as described by Heebøll et al,
      • Heeboll S.
      • Borre M.
      • Ottosen P.D.
      • Andersen C.L.
      • Mansilla F.
      • Dyrskjot L.
      • Orntoft T.F.
      • Torring N.
      SMARCC1 expression is upregulated in prostate cancer and positively correlated with tumour recurrence and dedifferentiation.
      using rabbit polyclonal antibodies to cathepsin E (ab36996; 1:1200), maspin (ab22354; 1:400), and survivin (ab469; 1:1600) and mouse monoclonal antibody to polo-like kinase 1 (Plk1) (ab17056; 1:1000; Abcam, Cambridge, UK). All immunohistochemical analyses were performed in the same laboratory (Department of Molecular Medicine, Aarhus University Hospital, Denmark). All TMA slides were scanned using a Hamamatsu Nanozoomer scanner (Hamamatsu Corporation, Hamamatsu City, Japan). The high-resolution images were scored manually using VIS visualization software (Visiopharm A/S; Hoersholm, Denmark). The immunohistochemical stainings were scored blinded to clinical outcome by two observers (N.F. and L.D.). In case of interobserver disagreement, a consensus scoring was obtained. For each tumor, the intensity of the stained carcinoma cells was scored as none, low, or high, together with the relative percentage of carcinoma cells carrying the particular intensity. Receiver-operating characteristic curves were then used for dichotomization using best-performance cut-point.
      • Soreide K.
      Receiver-operating characteristic curve analysis in diagnostic, prognostic and predictive biomarker research.
      All cut-points were defined in the Danish patient cohort and then applied to the Spanish and Swedish patient cohort. Scoring was evaluated based on the receiver-operating characteristic, curve dichotomization as follows: cathepsin E; positive cytoplasmic expression, 80% cut-point; maspin: positive nuclear expression, 20% cut-point; Plk1: positive cellular expression (both cytoplasmic and nuclear), 10% cut-point; survivin: high nuclear expression, 10% cut-point.

      Western Blot Analysis

      The specificity of the antibody against survivin was validated previously by Als et al.
      • Als A.B.
      • Dyrskjot L.
      • von der Masse H.
      • Koed K.
      • Mansilla F.
      • Toldbod H.E.
      • Jensen J.L.
      • Ulhoi B.P.
      • Sengelov L.
      • Jensen K.M.
      • Orntoft T.F.
      Emmprin and survivin predict response and survival following cisplatin-containing chemotherapy in patients with advanced bladder cancer.
      The specificity of the antibodies against maspin, cathepsin E, and Plk1 was validated on 4% to 12% SDS PAGE and Western blot analysis of extracts from either corresponding fresh-frozen bladder cancer samples from the patients in the study, or from human cell line HeLa S3 and bladder cancer cell line J82 (see Supplemental Figure S1 at http://ajp.amjpathol.org). Primary antibodies used: rabbit polyclonal against cathepsin E (ab36996), 1:1000; rabbit polyclonal against maspin (ab22354) 1:300, and rabbit polyclonal against survivin (ab469) 1:5000; mouse monoclonal against Plk1: (ab17056), 1:1000. Secondary antibodies used: swine anti-rabbit horseradish peroxidase conjugated (P0217), 1:3000; goat anti-mouse horseradish peroxidase conjugated (P0447), 1:3000 (both DakoCytomation). Loading control used: polyclonal antibody against β-actin horseradish peroxidase conjugated (ab49900), 1:30,000 (Abcam Inc., Cambridge, UK). Protein extraction was done in radioimmunoprecipitation assay buffer, and protein concentration measurement was subsequently performed following a modified Bradford protocol (Bio-Rad, Hercules, CA). Every lane was loaded with 25 μg of protein. Western blot analysis was performed essentially as described in Mansilla et al.
      • Mansilla F.
      • Birkenkamp-Demtroder K.
      • Kruhoffer M.
      • Sorensen F.B.
      • Andersen C.L.
      • Laiho P.
      • Aaltonen L.A.
      • Verspaget H.W.
      • Orntoft T.F.
      Differential expression of DHHC9 in microsatellite stable and instable human colorectal cancer subgroups.

      Statistical Analysis

      We used Stata 10.0 statistical analysis software (Stata Corporation, College Station, TX) for calculation of log-rank tests for equality of survival function, Kaplan-Meier survival curves, and univariate and multivariate Cox regression analysis. Variables with a P value <0.05 in univariate analysis were included in multivariate analysis to identify variables with independent significance. The predictive accuracy of prognostic models was quantified with Harrell's concordance index. The assumptions of proportional hazards were verified.

      Results

      A total of 283 patients with primary NMIBC (182 Ta, 101 T1) fulfilled the inclusion criteria for the study in the Danish patient cohort; the follow-up in the cohort ranged from 2 to 263 months, with a median follow-up time of 103 months. One hundred thirteen patients showed disease progression to MIBC during a median follow-up time of 22 months (1 to 210 months). The 170 patients without progression to MIBC had a median follow-up time of 117 months (53 to 250 months). The clinical and histopathological characteristics are listed in Table 1.

      Expression of Cathepsin E, Maspin, Survivin, and Plk1 in the Danish Cohort

      We found cathepsin E to be localized predominantly in the cytoplasm of the carcinoma cells. The staining was distinct and confined to the carcinoma cells. Maspin was found expressed in both the cytoplasm and nuclei of the carcinoma cells. In many cases, we observed a mutually exclusive staining pattern for the nuclear and cytoplasmic expression of maspin. Survivin was found expressed both in the cytoplasm and the nuclei of the carcinoma cells, and the staining was confined mostly to the carcinoma cells. Plk1 expression was localized both in the nucleus and in the cytoplasm of the carcinoma cells, and the staining was confined mostly to the carcinoma cells (see Supplemental Figure S1 at http://ajp.amjpathol.org). The distribution of clinical and histopathological characteristics according to protein expression is listed in Supplemental Table S1 at http://ajp.amjpathol.org. The immunohistochemistry scoring resulted in the following interobserver statistics for the Danish cohort (free marginal Kappa): cathepsin E 0.86, maspin 0.73, survivin 0.66, and Plk1 0.67.

      Prognostic Value of the Markers in the Danish Cohort

      We found the expression of cathepsin E, maspin, Plk1, and survivin to be significantly correlated with progression (all markers: P < 0.001) (Table 2 and Figure 1). When stratifying for age, stage, grade, growth pattern, and adjuvant treatment in multivariate Cox regression analysis, we identified cathepsin E (P < 0.001), maspin (P < 0.001), Plk1 (P = 0.021), and survivin (P < 0.001) to be independent prognostic markers for progression to MIBC (Table 2). Furthermore, the expression of cathepsin E, maspin, Plk1, and survivin was significantly associated with disease-specific survival (see Supplemental Table S2 at http://ajp.amjpathol.org) and overall survival (see Supplemental Table S3 at http://ajp.amjpathol.org). Kaplan-Meier survival curves for the expression of the four proteins in relation to disease-specific survival and overall survival are shown in Supplemental Figures S2 and S3 at http://ajp.amjpathol.org.
      Table 2Univariate and Multivariate Cox Regression Analysis for the Danish Cohort; Progression-Free Survival
      Danish Ta/T1 cohortDanish Ta cohortDanish T1 cohort
      VariableHR (95% CI)P valuePAHR (95% CI)P valuePAHR (95% CI)P valuePA
      Univariate Cox regression analysis and prediction accuracy for each variable
      Age (per 5-year increment)1.18 (1.07–1.31)0.00160.3%1.13 (0.98–1.31)0.08758.9%1.11 (0.95–1.30)0.17755.8%
      Sex1.03 (0.65–1.61)0.98150.1%1.10 (0.58–2.11)0.76450.4%1.09 (0.58–2.06)0.78350.8%
      Stage (T1 vs Ta)3.01 (2.07–4.38)<0.00165.6%
      Grading WHO 2004
       High/vs low/PUNLMP3.42 (2.34–4.98)<0.00167.1%7.59 (2.66–21.67)<0.00154.2%1.93 (0.60–6.18)0.26752.8%
      Growth pattern solid/mixed vs papillary2.35 (1.44–3.83)0.00156.0%0.67 (0.16–2.74)0.57350.2%2.06 (1.17–3.61)0.01256.8%
      Adjuvant treatment0.52 (0.29–0.90)0.01955.7%0.91 (0.46–1.81)0.79252.0%0.24 (0.09–0.67)0.00658.5%
      Concomitant CIS1.41 (0.96–2.06)0.07753.0%1.71 (0.98–2.98)0.05955.6%0.73 (0.44–1.23)0.24056.4%
      Size (≥ vs <3 cm)1.06 (0.67–1.67)0.81051.7%0.60 (0.26–1.36)0.21953.6%1.36 (0.77–2.42)0.28853.6%
      Multiplicity (one vs more)1.12 (0.74–1.70)0.58451.3%1.04 (0.54–2.01)0.89749.3%1.11 (0.64–1.91)0.71851.5%
      Cathepsin E0.32 (0.18–0.59)<0.00159.1%0.54 (0.26–1.12)0.09756.9%0.21 (0.07–0.67)0.00858.2%
      Maspin0.39 (0.26–0.58)<0.00162.2%0.31 (0.17–0.54)<0.00165.2%0.74 (0.42–1.32)0.30854.0%
      Plk12.21 (1.44–3.38)<0.00157.4%3.19 (1.71–5.94)<0.00160.4%1.17 (0.66–2.10)0.59052.2%
      Survivin2.74 (1.82–4.13)<0.00162.2%3.14 (1.74–5.65)<0.00163.0%1.55 (0.88–2.73)0.12956.2%
      Signature: high vs low6.79 (3.78–12.2)<0.00172.6%6.76 (3.20–14.29)<0.00172.0%3.49 (1.33–9.18)0.01163.6%
      Signature: intermediate vs low3.06 (1.75–5.36)<0.00164.3%2.41 (1.18–4.89)0.01561.1%2.92 (1.10–7.76)0.03261.9%
      Signature: high vs intermediate2.21 (1.41–3.45)0.00160.0%3.18 (1.61–6.29)0.00163.7%1.27 (0.71–2.29)0.42154.4%
      PA model without markers73.5%67.3%68.7%
      Multivariate analysis and prediction accuracy for the PA model including prognostic markers
      Cathepsin E0.31 (0.17–0.58)<0.00176.3%0.52 (0.25–1.08)0.07869.4%0.24 (0.07–0.76)0.01573.9%
      Maspin0.44 (0.29–0.67)<0.00175.9%0.31 (0.18–0.55)<0.00176.2%0.68 (0.38–1.22)0.19969.5%
      Plk11.68 (1.08–2.61)0.02175.0%3.04 (1.62–5.69)0.00169.7%1.06 (0.59–1.90)0.85372.6%
      Survivin2.25 (1.46–3.49)<0.00176.5%3.35 (1.85–6.07)<0.00172.5%1.56 (0.89–2.76)0.12270.7%
      Signature: high vs low6.48 (3.28–12.8)<0.00183.1%6.37 (2.98–13.61)<0.00184.1%2.66 (1.00–7.14)0.05074.2%
      Signature: intermediate vs low2.60 (1.46–4.64)0.00177.2%2.24 (1.19–4.91)0.01575.3%2.09 (0.76–5.73)0.15480.4%
      Signature: high vs intermediate2.08 (1.29–3.34)0.00373.1%2.96 (1.48–5.92)0.00270.3%1.44 (0.80–2.60)0.22669.3%
      PA model including all four markers78.6%77.7%74.5%
      Only risk factors with P values <0.05 were included in the multivariate analysis. The molecular markers were included separately and not in combination. PA model included all variables. Bold indicates P values <0.05.
      PA, prediction accuracy (Harrell's concordance index); PUNLMP, papillary urothelial neoplasm of low malignant potential; WHO, World Health Organization.
      Figure thumbnail gr1
      Figure 1Correlation between outcome and marker expression in the Danish training cohort. Kaplan-Meier curves showing progression-free survival as a function of protein expression of cathepsin E (A), maspin (B), Plk1 (C), and survivin (D) in the Danish patient cohort. Progression-free survival as a function of protein expression of cathepsin E (E) in the Danish T1 patient cohort and of maspin (F), Plk1 (G), and survivin (H) in the Danish Ta patient cohort.
      When subdividing the patient cohort by stage, maspin (P < 0.001), Plk1 (P < 0.001), and survivin (P < 0.001) were identified in univariate Cox regression analysis and log-rank test as prognostic markers for progression in patients with Ta bladder tumors (Table 2 and Figure 1). Cathepsin E was identified in univariate Cox regression analysis (P = 0.008) and log-rank test (P = 0.003) as a prognostic marker for T1 bladder cancer (Table 2 and Figure 1). Using Harrell's concordance index, we calculated the predictive accuracy of the clinical and histopathological parameters of the patient cohort. The concordance index is the probability that, when randomly selecting two patients, the patient with the worse outcome is actually predicted to have a worse outcome.
      • Harrell Jr., F.E.
      • Califf R.M.
      • Pryor D.B.
      • Lee K.L.
      • Rosati R.A.
      Evaluating the yield of medical tests.
      This measure ranges from 0.5 (chance) to 1.0 (perfect ability to rank patients). In addition, we calculated the prediction accuracy when including the protein markers, both separately and in combination. For the total Danish cohort, the addition of the four markers to the existing clinical and histopathological parameters raised the predictive accuracy by 6.5%, for Ta-tumors only 13.4%, and for T1-tumors only 7.8% (Table 2). An improved predictive accuracy was also seen when using disease-specific survival or overall survival as endpoints in the analysis (see Supplemental Tables S2 and S3 at http://ajp.amjpathol.org).

      EORTC Risk Stratification

      We then applied the EORTC risk table for progression to the Danish Ta/T1 cohort and found that the cohort was split into a low-risk group and a high-risk group (Figure 2A). Using these low- and high-risk EORTC-based groups, we found the expression of survivin (log-rank test; P = 0.004) and maspin (log-rank test; P < 0.001) to significantly subgroup the EORTC low-risk patients (Figure 2, B and C). We could subgroup the EORTC low-risk patients even further by combining the survivin and maspin expression in the tumors from the EORTC low-risk patients (Figure 2D). Consequently, when using survivin and maspin in combination, we identified a patient group with very low risk of progression [4.5% (3 of 66) in the first 48 months after diagnosis]. The remaining EORTC low-risk patients had a moderate risk of progression [17% (9 of 53) in the first 48 months after diagnosis] (log-rank test; P < 0.001; Figure 2D). The expression of cathepsin E (log-rank test; P = 0.001) subdivided the EORTC high-risk patients (Figure 2E). This use of the cathepsin E expression in tumors from EORTC high-risk patients identified a subgroup of patients with a lowered risk of progression MIBC [16% (3 of 19) in the first 48 months after diagnosis]. The cathepsin E low-expression subgroup of EORTC high-risk patients had a very high risk of progression to MIBC [47% (39 of 83) in the first 48 months after diagnosis].
      Figure thumbnail gr2
      Figure 2EORTC risk stratification and marker expression in the Danish training cohort. A: Kaplan-Meier curve showing progression-free survival in the Danish Ta/T1 cohort as a function of the EORTC progression score. Kaplan-Meier curve showing progression-free survival in the EORTC low-risk group (progression score 0 to 6) as a function of the dichotomized expressions of survivin (B), maspin (C), and maspin and survivin in combination (D). E: Kaplan-Meier curve showing progression-free survival in the EORTC high-risk group (progression score 7 to 23) as a function of the dichotomized expression of cathepsin E.

      Validation in Swedish and Spanish Patient Cohorts

      The prognostic value of cathepsin E, maspin, survivin, and Plk1 expression was validated in two independent NMIBC cohorts from Sweden and Spain with long-term follow-up. Clinical and histopathological characteristics are listed in Table 1. The cut-points applied for each protein marker were identical to those established in the Danish cohort. The following interobserver statistics were obtained (free marginal Kappa): the Swedish cohort—cathepsin E 0.87, maspin 0.71, survivin 0.74, and Plk1 0.78; the Spanish cohort—cathepsin E 0.81, maspin 0.74, survivin 0.81, and Plk1 0.69.
      Using univariate Cox regression analysis, we identified cathepsin E (P = 0.002), maspin (P < 0.001), Plk1 (P = 0.003), and survivin (P = 0.002) to show significant association with disease progression to MIBC in the Spanish–Swedish validation cohort (Table 3). Predictive accuracy was raised 8.0% when including the four prognostic markers for predicting progression to MIBC (Table 3). Kaplan-Meier survival curves for the Spanish–Swedish validation cohort are shown in Figure 3. Multivariate Cox regression analysis showed cathepsin E (P = 0.019) and survivin (P = 0.014) to significantly associate with disease progression when adjusting for other known clinical and histopathological risk factors (Table 3). It was not possible to subgroup the Spanish and Swedish patient cohort based on the EORTC criteria due to missing clinical parameters (Table 1).
      Table 3Univariate and Multivariate Cox Regression Analysis for the Swedish and Spanish Cohorts
      Univariate Cox regression analysis for the Spanish-Swedish validation cohortMultivariate Cox regression analysis for the Spanish-Swedish validation cohort
      VariableHR (95% CI)P valuePAHR (95% CI)P valuePA
      Age (per 5-year increment)1.13 (1.02–1.25)0.02157.4%
      Sex1.09 (0.65–1.83)0.74750.9%
      Stage (T1 vs Ta)2.79 (1.45–5.35)0.00256.9%
      Grading WHO 2004 High versus low/PUNLMP3.03 (1.84–4.98)<0.00161.3%
      Cathepsin E0.33 (0.17–0.67)0.00257.5%0.43 (0.21–0.87)0.01967.5%
      Maspin0.43 (0.27–0.67)<0.00161.0%0.64 (0.39–1.03)0.06768.9%
      Plk11.91 (1.24–2.94)0.00358.4%1.47 (0.94–2.28)0.08668.0%
      Survivin2.00 (1.30–3.07)0.00258.3%1.73 (1.12–2.69)0.01468.5%
      Signature: high versus low4.38 (2.35–8.15)<0.00167.2%2.89 (1.46–5.71)0.00273.1%
      Signature: intermediate versus low2.07 (1.06–4.02)0.03359.6%1.49 (0.75–2.98)0.25864.7%
      Signature: high versus intermediate2.04 (1.26–3.32)0.00458.3%1.82 (1.11–2.99)0.01869.1%
      PA model without markers66.4%
      PA model incl. all four markers71.7%
      Only risk factors with P values <0.05 were included in the multivariate analysis. The molecular markers were included separately and not in combination. PA model included all variables. Bold indicates P values <0.05.
      PA, prediction accuracy (Harrell's concordance index); PUNLMP, papillary urothelial neoplasm of low malignant potential; WHO, World Health Organization.
      Figure thumbnail gr3
      Figure 3Correlation between outcome and marker expression in the Spanish–Swedish– validation cohort. Kaplan-Meier curves showing progression-free survival in the Spanish–Swedish validation cohort as a function of dichotomized protein expression of cathepsin E (A), maspin (B), Plk1 (C), and survivin (D). The protein marker cut points defined by the Danish cohort were applied to the Spanish–Swedish validation cohort.
      We then generated a low-, intermediate-, and high-risk signature for the Danish cohort based on the combined expression of the four protein markers examined. This signature significantly split the patient cohort into three groups with different prognoses (log-rank test; P < 0.001) (Figure 4A). All three risk groups differed significantly from each other in both univariate analysis and multivariate analysis (Table 2). We successfully validated this risk signature in the Spanish–Swedish validation cohort (Figure 4B, log-rank test; P < 0.001). The three risk signatures differed significantly from each other in univariate analysis and in multivariate analysis for high-risk signature versus low-risk signature (P = 0.002), and for high-risk signature versus intermediate-risk signature (P = 0.018) (Table 3).
      Figure thumbnail gr4
      Figure 4Molecular risk signatures based on combined marker expression. A: Kaplan-Meier curves showing progression-free survival in the Danish Ta/T1 cohort as a function of high -, intermediate -, or low-risk signatures based on protein expression of cathepsin E, maspin, PLK-1, and survivin. B: Kaplan-Meier curves showing progression-free survival in the Spanish–Swedish Ta/T1 cohort as a function of high-, intermediate-, or low-risk signatures based on protein expression of cathepsin E, maspin, PLK-1, and survivin. Only patients with information from three or four markers were included in the analysis. Low-risk signature: 0 or 1 marker for poor outcome. Intermediate-risk signature: 2/4 markers for poor outcome. High-risk signature: 2/3, 3, or 4 markers for poor outcome.

      Discussion

      For a molecular marker to be useful in the clinic, it has to add additional prognostic accuracy to already existing prognostic clinical and histopathological parameters, thereby giving the clinicians an improved possibility to individualize the treatment. Adding prognostic molecular markers to a risk table of clinical and pathological parameters, eg, the EORTC risk table, which is currently available for patient risk management in the clinic, may have great potential in future clinical practice. In addition, optimization of already existing risk tables may also be necessary, as it was recently published that the EORTC risk table may overestimate the actual risk of progression to MIBC.
      • Fernandez-Gomez J.
      • Madero R.
      • Solsona E.
      • Unda M.
      • Martinez-Pineiro L.
      • Ojea A.
      • Portillo J.
      • Montesinos M.
      • Gonzalez M.
      • Pertusa C.
      • Rodriguez-Molina J.
      • Camacho J.E.
      • Rabadan M.
      • Astobieta A.
      • Isorna S.
      • Muntanola P.
      • Gimeno A.
      • Blas M.
      • Martinez-Pineiro J.A.
      The EORTC tables overestimate the risk of recurrence and progression in patients with non-muscle-invasive bladder cancer treated with bacillus Calmette-Guerin: external validation of the EORTC risk tables.
      In this study, we identified cathepsin E, maspin, Plk1, and survivin as candidate prognostic protein markers in patients with NMIBC. Our group has previously found the associated gene transcripts (CTSE, SERPINB5, PLK1, and BIRC5) to be among the most differentially expressed transcripts in gene expression signatures for prediction of subsequent disease progression.
      • Dyrskjot L.
      • Zieger K.
      • Real F.X.
      • Malats N.
      • Carrato A.
      • Hurst C.
      • Kotwal S.
      • Knowles M.
      • Malmstrom P.U.
      • de la Torre M.
      • Wester K.
      • Allory Y.
      • Vordos D.
      • Caillault A.
      • Radvanyi F.
      • Hein A.M.
      • Jensen J.L.
      • Jensen K.M.
      • Marcussen N.
      • Orntoft T.F.
      Gene expression signatures predict outcome in non-muscle-invasive bladder carcinoma: a multicenter validation study.
      • Dyrskjot L.
      • Zieger K.
      • Kruhoffer M.
      • Thykjaer T.
      • Jensen J.L.
      • Primdahl H.
      • Aziz N.
      • Marcussen N.
      • Moller K.
      • Orntoft T.F.
      A molecular signature in superficial bladder carcinoma predicts clinical outcome.
      Interestingly, we showed how the EORTC low-risk patients could be significantly subgrouped based on the survivin and maspin expression. This ability to identify a subgroup of patients with very low risk of progression to MIBC is a step toward individualized treatment of these cancer patients. Equally interesting, the EORTC high-risk patients were statistically significantly subgrouped based on the cathepsin E expression in the Danish patient cohort. We were not able to subgroup the Swedish and Spanish patient cohort based on the EORTC criteria due to missing clinical parameters. Using Harrell's concordance index, we showed how the four protein markers added prognostic value to the existing clinical and histopathological parameters, thereby giving the clinician more accurate tools for guiding the management of bladder cancer patients in the future.
      We then validated the four protein markers in an independent cohort of NMIBC patients with long-term follow-up. Finally, we derived a combined risk signature on the basis of the protein expressions in the Danish cohort capable of dividing patients in low-, intermediate-, and high-risk groups. This signature was validated in the Spanish–Swedish patient cohort. The successful validation of the four protein markers in the Spanish–Swedish NMIBC cohort is promising, seen in the light of possible different ischemia time, tissue fixation, and tissue handling. However, a multicenter study with numerous prospectively collected specimens is needed before possible clinical usage.
      • Pepe M.S.
      • Feng Z.
      • Janes H.
      • Bossuyt P.M.
      • Potter J.D.
      Pivotal evaluation of the accuracy of a biomarker used for classification or prediction: standards for study design.
      The role of cathepsin E in cancer has become increasingly clear. This intracellular aspartic proteinase has been shown to prevent tumor growth and metastasis,
      • Kawakubo T.
      • Okamoto K.
      • Iwata J.
      • Shin M.
      • Okamoto Y.
      • Yasukochi A.
      • Nakayama K.I.
      • Kadowaki T.
      • Tsukuba T.
      • Yamamoto K.
      Cathepsin E prevents tumor growth and metastasis by catalyzing the proteolytic release of soluble TRAIL from tumor cell surface.
      to inhibit angiogenesis, and to augment the immune response.
      • Shin M.
      • Kadowaki T.
      • Iwata J.
      • Kawakubo T.
      • Yamaguchi N.
      • Takii R.
      • Tsukuba T.
      • Yamamoto K.
      Association of cathepsin E with tumor growth arrest through angiogenesis inhibition and enhanced immune responses.
      Our study found the cytoplasmatic expression of cathepsin E to associate statistically significantly with a low risk of disease progression to T2 to T4 bladder cancer. This is in accordance with previous findings by our group, stating that the CTSE transcript is overexpressed in Ta/T1 bladder cancer patients that have a less aggressive course.
      • Dyrskjot L.
      • Zieger K.
      • Kruhoffer M.
      • Thykjaer T.
      • Jensen J.L.
      • Primdahl H.
      • Aziz N.
      • Marcussen N.
      • Moller K.
      • Orntoft T.F.
      A molecular signature in superficial bladder carcinoma predicts clinical outcome.
      Previously, Thykjaer et al
      • Thykjaer T.
      • Workman C.
      • Kruhoffer M.
      • Demtroder K.
      • Wolf H.
      • Andersen L.D.
      • Frederiksen C.M.
      • Knudsen S.
      • Orntoft T.F.
      Identification of gene expression patterns in superficial and invasive human bladder cancer.
      showed an association between the mRNA expression of the CTSE transcript and the expression of cathepsin E in bladder cancer. This association was also shown by Blaveri et al
      • Blaveri E.
      • Simko J.P.
      • Korkola J.E.
      • Brewer J.L.
      • Baehner F.
      • Mehta K.
      • Devries S.
      • Koppie T.
      • Pejavar S.
      • Carroll P.
      • Waldman F.M.
      Bladder cancer outcome and subtype classification by gene expression.
      in an article that found cathepsin E to be overexpressed in non-muscle invasive bladder cancer compared with muscle-invasive bladder cancer. However, Wild et al
      • Wild P.J.
      • Herr A.
      • Wissmann C.
      • Stoehr R.
      • Rosenthal A.
      • Zaak D.
      • Simon R.
      • Knuechel R.
      • Pilarsky C.
      • Hartmann A.
      Gene expression profiling of progressive papillary noninvasive carcinomas of the urinary bladder.
      have found high cathepsin E expression to associate with shorter disease-specific survival.
      In 1994, the expression of maspin was reported to be down-regulated in invasive and metastatic breast cancer cells.
      • Zou Z.
      • Anisowicz A.
      • Hendrix M.J.
      • Thor A.
      • Neveu M.
      • Sheng S.
      • Rafidi K.
      • Seftor E.
      • Sager R.
      Maspin, a serpin with tumor-suppressing activity in human mammary epithelial cells.
      This discovery made maspin a candidate for extensive investigation in several cancer tissues, including bladder cancer. It has been shown to be a member of the mammary serpin protease inhibitor family and has been shown to be an inhibitor of angiogenesis in vitro and in vivo. Its presence in both the cytoplasm and nucleus of cancer cells was first described in 1997.
      • Pemberton P.A.
      • Tipton A.R.
      • Pavloff N.
      • Smith J.
      • Erickson J.R.
      • Mouchabeck Z.M.
      • Kiefer M.C.
      Maspin is an intracellular serpin that partitions into secretory vesicles and is present at the cell surface.
      This simultaneous cytoplasmic and nuclear presence has afterward been shown in several types of cancers.
      • Maass N.
      • Teffner M.
      • Rosel F.
      • Pawaresch R.
      • Jonat W.
      • Nagasaki K.
      • Rudolph P.
      Decline in the expression of the serine proteinase inhibitor maspin is associated with tumour progression in ductal carcinomas of the breast.
      • Machtens S.
      • Serth J.
      • Bokemeyer C.
      • Bathke W.
      • Minssen A.
      • Kollmannsberger C.
      • Hartmann J.
      • Knuchel R.
      • Kondo M.
      • Jonas U.
      • Kuczyk M.
      Expression of the p53 and Maspin protein in primary prostate cancer: correlation with clinical features.
      Mohsin et al
      • Mohsin S.K.
      • Zhang M.
      • Clark G.M.
      • Craig A.D.
      Maspin expression in invasive breast cancer: association with other prognostic factors.
      first stressed the importance of the subcellular distribution of maspin. They found that the nuclear maspin expression correlated with a good prognosis and that the cytoplasmatic expression of maspin correlated with a poor prognosis. Our immunohistochemical analysis of the subcellular expression of maspin supports these findings. Three earlier studies have investigated the clinical relevance of the expression of maspin in non-muscle invasive urothelial bladder tumors: Blandamura et al,
      • Blandamura S.
      • D'Alessandro E.
      • Giacomelli L.
      • Guzzardo V.
      • Battanello W.
      • Repele M.
      • Ninfo V.
      Expression of maspin in papillary Ta/T1 bladder neoplasms.
      Friedrich et al,
      • Friedrich M.G.
      • Toma M.I.
      • Petri S.
      • Cheng J.C.
      • Hammerer P.
      • Erbersdobler A.
      • Huland H.
      Expression of Maspin in non-muscle invasive bladder carcinoma: correlation with tumor angiogenesis and prognosis.
      and Sugimoto et al.
      • Sugimoto S.
      • Maass N.
      • Takimoto Y.
      • Sato K.
      • Minei S.
      • Zhang M.
      • Hoshikawa Y.
      • Junemann K.P.
      • Jonat W.
      • Nagasaki K.
      Expression and regulation of tumor suppressor gene maspin in human bladder cancer.
      Blandamura et al
      • Blandamura S.
      • D'Alessandro E.
      • Giacomelli L.
      • Guzzardo V.
      • Battanello W.
      • Repele M.
      • Ninfo V.
      Expression of maspin in papillary Ta/T1 bladder neoplasms.
      and Friedrich et al
      • Friedrich M.G.
      • Toma M.I.
      • Petri S.
      • Cheng J.C.
      • Hammerer P.
      • Erbersdobler A.
      • Huland H.
      Expression of Maspin in non-muscle invasive bladder carcinoma: correlation with tumor angiogenesis and prognosis.
      did not find maspin to be an independent predictor of progression to muscle-invasive bladder cancer. Interestingly, in these three studies, the subcellular distribution of maspin in bladder carcinoma cells was not described in relation to prognosis.
      Survivin is a multifunctional protein. It has been shown to inhibit apoptosis, to regulate cell division, and to enhance angiogenesis.
      • Duffy M.J.
      • O'Donovan N.
      • Brennan D.J.
      • Gallagher W.M.
      • Ryan B.M.
      Survivin: a promising tumor biomarker.
      Studies have shown an association between the expression of survivin in non-muscle invasive bladder cancer and progression to muscle-invasive disease.
      • Yin W.
      • Chen N.
      • Zhang Y.
      • Zeng H.
      • Chen X.
      • He Y.
      • Wang X.
      • Zhou Q.
      Survivin nuclear labeling index: a superior biomarker in superficial urothelial carcinoma of human urinary bladder.
      • Ku J.H.
      • Kwak C.
      • Lee H.S.
      • Park H.K.
      • Lee E.
      • Lee S.E.
      Expression of survivin, a novel inhibitor of apoptosis, in superficial transitional cell carcinoma of the bladder.
      • Karam J.A.
      • Lotan Y.
      • Ashfaq R.
      • Sagalowsky A.I.
      • Shariat S.F.
      Survivin expression in patients with non-muscle-invasive urothelial cell carcinoma of the bladder.
      It has been discussed whether total, nuclear, or cytoplasmic survivin expression should be evaluated in relation to clinical outcome. Our study identified a high nuclear survivin expression to be statistically significant and independently correlated to disease progression, disease-specific survival, and overall survival.
      Plk1 is involved in the regulation of cell division, DNA damage repair pathways, apoptosis, and the progression of the cell cycle.
      • van Vugt M.A.
      • Medema R.H.
      Getting in and out of mitosis with Polo-like kinase-1.
      The loss of Plk1 expression has been shown to induce pro-apoptotic pathways and inhibit growth.
      • Ando K.
      • Ozaki T.
      • Yamamoto H.
      • Furuya K.
      • Hosoda M.
      • Hayashi S.
      • Fukuzawa M.
      • Nakagawara A.
      Polo-like kinase 1 (Plk1) inhibits p53 function by physical interaction and phosphorylation.
      Furthermore, NIH3T3 cell lines have been shown to form tumors in nude mice due to Plk1 overexpression.
      • Smith M.R.
      • Wilson M.L.
      • Hamanaka R.
      • Chase D.
      • Kung H.
      • Longo D.L.
      • Ferris D.K.
      Malignant transformation of mammalian cells initiated by constitutive expression of the polo-like kinase.
      Plk1 has been reported as a target for treatment,
      • Rodel F.
      • Keppner S.
      • Capalbo G.
      • Bashary R.
      • Kaufmann M.
      • Rodel C.
      • Strebhardt K.
      • Spankuch B.
      Polo-like kinase 1 as predictive marker and therapeutic target for radiotherapy in rectal cancer.
      • Gleixner K.V.
      • Ferenc V.
      • Peter B.
      • Gruze A.
      • Meyer R.A.
      • Hadzijusufovic E.
      • Cerny-Reiterer S.
      • Mayerhofer M.
      • Pickl W.F.
      • Sillaber C.
      • Valent P.
      Polo-like kinase 1 (Plk1) as a novel drug target in chronic myeloid leukemia: overriding imatinib resistance with the Plk1 inhibitor BI 2536.
      Interestingly, knock-down of survivin mRNA and Plk1 mRNA using small interfering RNA has been shown to significantly reduce the tumor size in mice.
      • Seth S.
      • Matsui Y.
      • Fosnaugh K.
      • Liu Y.
      • Vaish N.
      • Adami R.
      • Harvie P.
      • Johns R.
      • Severson G.
      • Brown T.
      • Takagi A.
      • Bell S.
      • Chen Y.
      • Chen F.
      • Zhu T.
      • Fam R.
      • Maciagiewicz I.
      • Kwang E.
      • McCutcheon M.
      • Farber K.
      • Charmley P.
      • Houston Jr., M.E.
      • So A.
      • Templin M.V.
      • Polisky B.
      RNAi-based therapeutics targeting survivin and PLK1 for treatment of bladder cancer.
      One previous study has examined immunohistochemical Plk1 overexpression in bladder cancer in relation to outcome.
      • Yamamoto Y.
      • Matsuyama H.
      • Kawauchi S.
      • Matsumoto H.
      • Nagao K.
      • Ohmi C.
      • Sakano S.
      • Furuya T.
      • Oga A.
      • Naito K.
      • Sasaki K.
      Overexpression of polo-like kinase 1 (PLK1) and chromosomal instability in bladder cancer.
      Here, they found a trend of Plk1 overexpression in patients with shorter progression-free survival.
      The treatment of NMIBC has been gradually intensified during the time span in which the patient cohorts have been followed (1979 to 2007), and this may introduce a bias. However, here we found no significant difference in progression-rates when comparing the latest 50% of cases of bladder cancer to the former 50% (results not shown). The distribution of stage in the validation cohort does not completely resemble the distribution in the Danish cohort. The validation cohort should ideally include more patients with Ta tumors. However, despite this bias, the markers examined in this study still have prognostic potential in both Ta and T1 bladder cancer and in NMIBC in general.
      The findings in this study are in concert with the results from our previous gene expression signatures for disease progression. The validation of the results in additional patient cohorts from Sweden and Spain strengthens the results considerably because cut-point determination and scoring was defined in one Danish training cohort, and then validated in an independent test cohort. We conclude that all four protein markers may be clinically relevant for better guiding the treatment of patients with non-muscle invasive urothelial carcinoma. Addition of molecular markers to the existing EORTC risk table may be a step toward further individualized treatment. Multicenter prospective studies are needed for further validation of the clinical relevance of the four protein markers.

      Acknowledgments

      We are grateful to Susanne Bruun, Pamela Celis, Connie Sørensen, Gitte Stougaard, and Mette Christensen for their excellent technical assistance.

      Supplementary data

      • Supplemental Figure S1

        Validation of antibody specificity by Western blotting. A: A band at 42 kDa was observed that is equal to the predicted molecular mass of cathepsin E. B: A band at 42 kDa was observed that is equal to the predicted molecular mass of maspin. C: A band at 68 kDa was observed that is equal to the predicted molecular mass of Plk1. HeLaS3 and HU609 cell lines were used as Plk1 positive controls as no patient samples were available. Beta actin was used as loading control. L, patient sample with low/no immunohistochemical expression of the examined protein. h, patient sample with an immunohistochemical high expression of the examined protein; MW, molecular weight. Immunohistochemical staining of selected 0.6 mm tissue microarray core biopsies stained with antibodies against cathepsin E, maspin, Plk1, and survivin, magnification x 20. Ta PUNLMP (D), T1 high grade (E), T1 high grade (F), Ta low grade (G), T1 high grade (H), Ta low grade (I), T1 high grade (J), Ta low grade (K).

      • Supplemental Figure S2

        Kaplan-Meier curves showing disease-specific survival as a function of protein expression of cathepsin E (A), maspin (B), Plk1 (C), and survivin (D) in the Danish patient cohort. Kaplan-Meier curves subdivided by stage, for cathepsin E (E) in the Danish T1 patient cohort, maspin (F), Plk1 (G), and survivin (H) in the Danish Ta patient cohort. A minimal follow-up of 5 years before censoring was only defined for progression-free survival.

      • Supplemental Figure S3

        Kaplan-Meier curves showing overall survival as a function of protein expression of cathepsin E (A), maspin (B), Plk1 (C), and survivin (D) in the Danish patient cohort. Kaplan-Meier curves subdivided by stage, for cathepsin E (E) in the Danish T1 patient cohort, maspin (F), Plk1 (G), and survivin (H) in the Danish Ta patient cohort (H). A minimal follow-up of 5 years before censoring was only defined for progression-free survival.

      References

        • Jemal A.
        • Siegel R.
        • Xu J.
        • Ward E.
        Cancer statistics, 2010.
        CA Cancer J Clin. 2010; 60: 277-300
        • Kiemeney L.A.
        • Witjes J.A.
        • Verbeek A.L.
        • Heijbroek R.P.
        • Debruyne F.M.
        The clinical epidemiology of superficial bladder cancer.
        Br J Cancer. 1993; 67: 806-812
        • Cookson M.S.
        • Herr H.W.
        • Zhang Z.F.
        • Soloway S.
        • Sogani P.C.
        • Fair W.R.
        The treated natural history of high risk superficial bladder cancer: 15-year outcome.
        J Urol. 1997; 158: 62-67
        • Cheng L.
        • Cheville J.C.
        • Neumann R.M.
        • Leibovich B.C.
        • Egan K.S.
        • Spotts B.E.
        • Bostwick D.G.
        Survival of patients with carcinoma in situ of the urinary bladder.
        Cancer. 1999; 85: 2469-2474
        • Ehdaie B.
        • Theodorescu D.
        Predicting tumor outcomes in urothelial bladder carcinoma: turning pathways into clinical biomarkers of prognosis.
        Expert Rev Anticancer Ther. 2008; 8: 1103-1110
        • Hong Y.M.
        • Loughlin K.R.
        Economic impact of tumor markers in bladder cancer surveillance.
        Urology. 2008; 71: 131-135
        • Sylvester R.J.
        • van der Meijden A.P.
        • Oosterlinck W.
        • Witjes J.A.
        • Bouffioux C.
        • Denis L.
        • Newling D.W.
        • Kurth K.
        Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials.
        Eur Urol. 2006; 49: 466-475
      1. discussion 475–477
        • Dyrskjot L.
        Classification of bladder cancer by microarray expression profiling: towards a general clinical use of microarrays in cancer diagnostics.
        Expert Rev Mol Diagn. 2003; 3: 635-647
        • Dyrskjot L.
        • Zieger K.
        • Real F.X.
        • Malats N.
        • Carrato A.
        • Hurst C.
        • Kotwal S.
        • Knowles M.
        • Malmstrom P.U.
        • de la Torre M.
        • Wester K.
        • Allory Y.
        • Vordos D.
        • Caillault A.
        • Radvanyi F.
        • Hein A.M.
        • Jensen J.L.
        • Jensen K.M.
        • Marcussen N.
        • Orntoft T.F.
        Gene expression signatures predict outcome in non-muscle-invasive bladder carcinoma: a multicenter validation study.
        Clin Cancer Res. 2007; 13: 3545-3551
        • Blaveri E.
        • Simko J.P.
        • Korkola J.E.
        • Brewer J.L.
        • Baehner F.
        • Mehta K.
        • Devries S.
        • Koppie T.
        • Pejavar S.
        • Carroll P.
        • Waldman F.M.
        Bladder cancer outcome and subtype classification by gene expression.
        Clin Cancer Res. 2005; 11: 4044-4055
        • Dyrskjot L.
        • Zieger K.
        • Kruhoffer M.
        • Thykjaer T.
        • Jensen J.L.
        • Primdahl H.
        • Aziz N.
        • Marcussen N.
        • Moller K.
        • Orntoft T.F.
        A molecular signature in superficial bladder carcinoma predicts clinical outcome.
        Clin Cancer Res. 2005; 11: 4029-4036
        • Orntoft T.F.
        • Thykjaer T.
        • Waldman F.M.
        • Wolf H.
        • Celis J.E.
        Genome-wide study of gene copy numbers, transcripts, and protein levels in pairs of non-invasive and invasive human transitional cell carcinomas.
        Mol Cell Proteomics. 2002; 1: 37-45
        • Kononen J.
        • Bubendorf L.
        • Kallioniemi A.
        • Barlund M.
        • Schraml P.
        • Leighton S.
        • Torhorst J.
        • Mihatsch M.J.
        • Sauter G.
        • Kallioniemi O.P.
        Tissue microarrays for high-throughput molecular profiling of tumor specimens.
        Nat Med. 1998; 4: 844-847
        • Heeboll S.
        • Borre M.
        • Ottosen P.D.
        • Andersen C.L.
        • Mansilla F.
        • Dyrskjot L.
        • Orntoft T.F.
        • Torring N.
        SMARCC1 expression is upregulated in prostate cancer and positively correlated with tumour recurrence and dedifferentiation.
        Histol Histopathol. 2008; 23: 1069-1076
        • Soreide K.
        Receiver-operating characteristic curve analysis in diagnostic, prognostic and predictive biomarker research.
        J Clin Pathol. 2009; 62: 1-5
        • Als A.B.
        • Dyrskjot L.
        • von der Masse H.
        • Koed K.
        • Mansilla F.
        • Toldbod H.E.
        • Jensen J.L.
        • Ulhoi B.P.
        • Sengelov L.
        • Jensen K.M.
        • Orntoft T.F.
        Emmprin and survivin predict response and survival following cisplatin-containing chemotherapy in patients with advanced bladder cancer.
        Clin Cancer Res. 2007; 13: 4407-4414
        • Mansilla F.
        • Birkenkamp-Demtroder K.
        • Kruhoffer M.
        • Sorensen F.B.
        • Andersen C.L.
        • Laiho P.
        • Aaltonen L.A.
        • Verspaget H.W.
        • Orntoft T.F.
        Differential expression of DHHC9 in microsatellite stable and instable human colorectal cancer subgroups.
        Br J Cancer. 2007; 96: 1896-1903
        • Harrell Jr., F.E.
        • Califf R.M.
        • Pryor D.B.
        • Lee K.L.
        • Rosati R.A.
        Evaluating the yield of medical tests.
        JAMA. 1982; 247: 2543-2546
        • Fernandez-Gomez J.
        • Madero R.
        • Solsona E.
        • Unda M.
        • Martinez-Pineiro L.
        • Ojea A.
        • Portillo J.
        • Montesinos M.
        • Gonzalez M.
        • Pertusa C.
        • Rodriguez-Molina J.
        • Camacho J.E.
        • Rabadan M.
        • Astobieta A.
        • Isorna S.
        • Muntanola P.
        • Gimeno A.
        • Blas M.
        • Martinez-Pineiro J.A.
        The EORTC tables overestimate the risk of recurrence and progression in patients with non-muscle-invasive bladder cancer treated with bacillus Calmette-Guerin: external validation of the EORTC risk tables.
        Eur Urol. 2011; 60: 423-430
        • Pepe M.S.
        • Feng Z.
        • Janes H.
        • Bossuyt P.M.
        • Potter J.D.
        Pivotal evaluation of the accuracy of a biomarker used for classification or prediction: standards for study design.
        J Natl Cancer Inst. 2008; 100: 1432-1438
        • Kawakubo T.
        • Okamoto K.
        • Iwata J.
        • Shin M.
        • Okamoto Y.
        • Yasukochi A.
        • Nakayama K.I.
        • Kadowaki T.
        • Tsukuba T.
        • Yamamoto K.
        Cathepsin E prevents tumor growth and metastasis by catalyzing the proteolytic release of soluble TRAIL from tumor cell surface.
        Cancer Res. 2007; 67: 10869-10878
        • Shin M.
        • Kadowaki T.
        • Iwata J.
        • Kawakubo T.
        • Yamaguchi N.
        • Takii R.
        • Tsukuba T.
        • Yamamoto K.
        Association of cathepsin E with tumor growth arrest through angiogenesis inhibition and enhanced immune responses.
        Biol Chem. 2007; 388: 1173-1181
        • Thykjaer T.
        • Workman C.
        • Kruhoffer M.
        • Demtroder K.
        • Wolf H.
        • Andersen L.D.
        • Frederiksen C.M.
        • Knudsen S.
        • Orntoft T.F.
        Identification of gene expression patterns in superficial and invasive human bladder cancer.
        Cancer Res. 2001; 61: 2492-2499
        • Wild P.J.
        • Herr A.
        • Wissmann C.
        • Stoehr R.
        • Rosenthal A.
        • Zaak D.
        • Simon R.
        • Knuechel R.
        • Pilarsky C.
        • Hartmann A.
        Gene expression profiling of progressive papillary noninvasive carcinomas of the urinary bladder.
        Clin Cancer Res. 2005; 11: 4415-4429
        • Zou Z.
        • Anisowicz A.
        • Hendrix M.J.
        • Thor A.
        • Neveu M.
        • Sheng S.
        • Rafidi K.
        • Seftor E.
        • Sager R.
        Maspin, a serpin with tumor-suppressing activity in human mammary epithelial cells.
        Science. 1994; 263: 526-529
        • Pemberton P.A.
        • Tipton A.R.
        • Pavloff N.
        • Smith J.
        • Erickson J.R.
        • Mouchabeck Z.M.
        • Kiefer M.C.
        Maspin is an intracellular serpin that partitions into secretory vesicles and is present at the cell surface.
        J Histochem Cytochem. 1997; 45: 1697-1706
        • Maass N.
        • Teffner M.
        • Rosel F.
        • Pawaresch R.
        • Jonat W.
        • Nagasaki K.
        • Rudolph P.
        Decline in the expression of the serine proteinase inhibitor maspin is associated with tumour progression in ductal carcinomas of the breast.
        J Pathol. 2001; 195: 321-326
        • Machtens S.
        • Serth J.
        • Bokemeyer C.
        • Bathke W.
        • Minssen A.
        • Kollmannsberger C.
        • Hartmann J.
        • Knuchel R.
        • Kondo M.
        • Jonas U.
        • Kuczyk M.
        Expression of the p53 and Maspin protein in primary prostate cancer: correlation with clinical features.
        Int J Cancer. 2001; 95: 337-342
        • Mohsin S.K.
        • Zhang M.
        • Clark G.M.
        • Craig A.D.
        Maspin expression in invasive breast cancer: association with other prognostic factors.
        J Pathol. 2003; 199: 432-435
        • Blandamura S.
        • D'Alessandro E.
        • Giacomelli L.
        • Guzzardo V.
        • Battanello W.
        • Repele M.
        • Ninfo V.
        Expression of maspin in papillary Ta/T1 bladder neoplasms.
        Anticancer Res. 2008; 28: 471-478
        • Friedrich M.G.
        • Toma M.I.
        • Petri S.
        • Cheng J.C.
        • Hammerer P.
        • Erbersdobler A.
        • Huland H.
        Expression of Maspin in non-muscle invasive bladder carcinoma: correlation with tumor angiogenesis and prognosis.
        Eur Urol. 2004; 45: 737-743
        • Sugimoto S.
        • Maass N.
        • Takimoto Y.
        • Sato K.
        • Minei S.
        • Zhang M.
        • Hoshikawa Y.
        • Junemann K.P.
        • Jonat W.
        • Nagasaki K.
        Expression and regulation of tumor suppressor gene maspin in human bladder cancer.
        Cancer Lett. 2004; 203: 209-215
        • Duffy M.J.
        • O'Donovan N.
        • Brennan D.J.
        • Gallagher W.M.
        • Ryan B.M.
        Survivin: a promising tumor biomarker.
        Cancer Lett. 2007; 249: 49-60
        • Yin W.
        • Chen N.
        • Zhang Y.
        • Zeng H.
        • Chen X.
        • He Y.
        • Wang X.
        • Zhou Q.
        Survivin nuclear labeling index: a superior biomarker in superficial urothelial carcinoma of human urinary bladder.
        Mod Pathol. 2006; 19: 1487-1497
        • Ku J.H.
        • Kwak C.
        • Lee H.S.
        • Park H.K.
        • Lee E.
        • Lee S.E.
        Expression of survivin, a novel inhibitor of apoptosis, in superficial transitional cell carcinoma of the bladder.
        J Urol. 2004; 171: 631-635
        • Karam J.A.
        • Lotan Y.
        • Ashfaq R.
        • Sagalowsky A.I.
        • Shariat S.F.
        Survivin expression in patients with non-muscle-invasive urothelial cell carcinoma of the bladder.
        Urology. 2007; 70: 482-486
        • van Vugt M.A.
        • Medema R.H.
        Getting in and out of mitosis with Polo-like kinase-1.
        Oncogene. 2005; 24: 2844-2859
        • Ando K.
        • Ozaki T.
        • Yamamoto H.
        • Furuya K.
        • Hosoda M.
        • Hayashi S.
        • Fukuzawa M.
        • Nakagawara A.
        Polo-like kinase 1 (Plk1) inhibits p53 function by physical interaction and phosphorylation.
        J Biol Chem. 2004; 279: 25549-25561
        • Smith M.R.
        • Wilson M.L.
        • Hamanaka R.
        • Chase D.
        • Kung H.
        • Longo D.L.
        • Ferris D.K.
        Malignant transformation of mammalian cells initiated by constitutive expression of the polo-like kinase.
        Biochem Biophys Res Commun. 1997; 234: 397-405
        • Rodel F.
        • Keppner S.
        • Capalbo G.
        • Bashary R.
        • Kaufmann M.
        • Rodel C.
        • Strebhardt K.
        • Spankuch B.
        Polo-like kinase 1 as predictive marker and therapeutic target for radiotherapy in rectal cancer.
        Am J Pathol. 2010; 177: 918-929
        • Gleixner K.V.
        • Ferenc V.
        • Peter B.
        • Gruze A.
        • Meyer R.A.
        • Hadzijusufovic E.
        • Cerny-Reiterer S.
        • Mayerhofer M.
        • Pickl W.F.
        • Sillaber C.
        • Valent P.
        Polo-like kinase 1 (Plk1) as a novel drug target in chronic myeloid leukemia: overriding imatinib resistance with the Plk1 inhibitor BI 2536.
        Cancer Res. 2010; 70: 1513-1523
        • Seth S.
        • Matsui Y.
        • Fosnaugh K.
        • Liu Y.
        • Vaish N.
        • Adami R.
        • Harvie P.
        • Johns R.
        • Severson G.
        • Brown T.
        • Takagi A.
        • Bell S.
        • Chen Y.
        • Chen F.
        • Zhu T.
        • Fam R.
        • Maciagiewicz I.
        • Kwang E.
        • McCutcheon M.
        • Farber K.
        • Charmley P.
        • Houston Jr., M.E.
        • So A.
        • Templin M.V.
        • Polisky B.
        RNAi-based therapeutics targeting survivin and PLK1 for treatment of bladder cancer.
        Mol Ther. 2011; 19: 928-935
        • Yamamoto Y.
        • Matsuyama H.
        • Kawauchi S.
        • Matsumoto H.
        • Nagao K.
        • Ohmi C.
        • Sakano S.
        • Furuya T.
        • Oga A.
        • Naito K.
        • Sasaki K.
        Overexpression of polo-like kinase 1 (PLK1) and chromosomal instability in bladder cancer.
        Oncology. 2006; 70: 231-237