Predictive Value of Combined Immunohistochemical Markers in Patients With pt1 Urothelial Carcinoma at Radical Cystectomy

Transcription

1 Predictive Value of Combined Immunohistochemical in Patients With pt1 Urothelial Carcinoma at Radical Cystectomy Shahrokh F. Shariat,* Christian Bolenz, Guilherme Godoy, Yves Fradet, Raheela Ashfaq, Pierre I. Karakiewicz, Hendrik Isbarn, Claudio Jeldres, Jérôme Rigaud, Arthur I. Sagalowsky and Yair Lotan From the Departments of Urology (SFS, CB, AIS, YL) and Pathology (RA), University of Texas Southwestern Medical Center, Dallas, Texas, Division of Urology, Sidney Kimmel Center for Prostate and Urologic Cancer, Memorial Sloan-Kettering Cancer Center, New York, New York (SFS, GG), Centre de recherche en cancérologie de l Université Laval, L Hôtel-Dieu de Québec, Le Centre Hospitalier Universitaire de Québec (YF, JR), and Cancer Prognostics and Health Outcomes Unit, University of Montreal (PIK, HI, CJ), Montreal, Quebec, Canada Abbreviations and Acronyms RC radical cystectomy UCB urothelial carcinoma of the bladder Submitted for publication November 6, Study received local Human Investigations Committee approval and institutional review board approval. Supported by National Institutes of Health training grant T32CA * Correspondence: Division of Urology; Sidney Kimmel Center for Prostate and Urologic Cancer, Memorial Sloan-Kettering Cancer Center, 1275 York Ave., Box 27, New York, New York (telephone: ; sfshariat@gmail.com). Purpose: pt1 urothelial carcinoma of the bladder is a potentially aggressive cancer diathesis with heterogeneous clinical behaviors. We tested whether the combination of immunohistochemical markers could risk stratify cases of pt1 urothelial carcinoma of the bladder at radical cystectomy. Materials and Methods: p53, p21, prb, p27, survivin and Ki-67 immunohistochemical staining was performed on representative urothelial carcinoma of the bladder specimens of 80 patients with pt1 urothelial carcinoma of the bladder treated with radical cystectomy and bilateral pelvic lymphadenectomy (median followup 61.6 months). Results: p53 expression was altered in 25% of patients, p21 in 46%, prb in 39%, p27 in 35%, survivin in 49% and Ki-67 in 34%. On multivariable analyses p53, p27 and Ki-67 were independently associated with urothelial carcinoma of the bladder recurrence (HR 3.66, p 0.033; HR 3.76, p and HR 3.96, p 0.021, respectively) and disease specific mortality (HR 5.25, p 0.016; HR 3.68, p and HR 6.23, p 0.009, respectively). The combination of these 3 biomarkers stratified cases into statistically significantly different risk groups for disease recurrence (p 0.001) and disease specific mortality (p 0.001). The addition of the number of altered markers increased the concordance indices of the base model that included grade, lymph node status, lymphovascular invasion and concomitant carcinoma in situ for disease recurrence and disease specific survival from 54.7% to 71.7% and from 64.3% to 77.5%, respectively. Conclusions: Assessment of p53, p27 and Ki-67 in urothelial carcinoma of the bladder specimens improves the prediction of recurrence-free and urothelial carcinoma of the bladder specific survival in patients with pt1 disease at radical cystectomy. These markers may help stratify the heterogeneous population of patients with pt1 disease into risk groups that can be used to guide clinical decision making regarding observation vs adjuvant therapy. Key Words: immunohistochemistry, recurrence, survival, urinary bladder neoplasms, cystectomy RADICAL cystectomy with bilateral pelvic lymphadenectomy is the gold standard for intended cure of muscle invasive UCB and selected cases of nonmuscle invasive UCB refractory to bladder preserving therapy. Although most patients with pt1 disease are cured with RC alone, more /09/ /0 Vol. 182, 78-84, July 2009 THE JOURNAL OF UROLOGY Printed in U.S.A. Copyright 2009 by AMERICAN UROLOGICAL ASSOCIATION DOI: /j.juro

2 BIOMARKER EXPRESSION IN PT1 BLADDER CANCER 79 than 20% experience disease recurrence and eventual UCB specific death. 1 3 To date none of the conventional clinical and pathological methods have successfully helped distinguish patients with pt1 disease at increased risk for disease progression from those who are cured with RC alone. Molecular biomarkers may provide a better understanding of the biology of an individual tumor with the promise of accurate prognosis/prediction resulting in improved clinical decision making. Candidate biomarkers involved in cell cycle regulation (p53, prb, p21 and p27), cell proliferation (Ki-67) and apoptosis (survivin) are indispensable for keeping urothelial growth in balance and the urothelial genome stable. We and others have shown that structural and functional defects of these candidate biomarkers are common in human UCB and are associated with poor oncological outcomes However, these studies were limited by their design (ie small, single center) and analysis (ie suboptimal statistical analysis). Moreover, none of the previous studies focused on pt1 tumors alone. Therefore, we assessed whether the tissue expression of p53, p21, prb, p27, survivin and Ki-67 improves prognostication in the heterogeneous population of patients with pt1 UCB after RC. In addition, we tested whether the combination of multiple biomarkers adds clinically useful incremental information compared to each biomarker alone. MATERIALS AND METHODS Patient Population The study was performed after approval by a local Human Investigations Committee and in accordance with the Department of Health and Human Services. Informed consent was obtained from each subject. The study cohort was composed of a convenience sample of 80 patients with pt1 disease at RC performed between July 1995 and July 2006 at University of Texas Southwestern Medical Center, Dallas, Texas or Université Laval, Québec, Canada. None of the patients received neoadjuvant or adjuvant therapy. Histology, tumor grade, tumor stage and presence of carcinoma in situ were confirmed by blinded reevaluation of the original pathology slides according to strict criteria. For each patient comprehensive clinical and pathological data elements were collected and entered into an institutional review board approved database. Staff pathologists with expertise in genitourinary pathology examined all specimens according to previously published protocols. 2 The 2002 TNM classification was used for pathological staging and the 2002 WHO classification was used for pathological grading. The followup protocol was described elsewhere. 2,18 Patients were seen postoperatively at least every 3 to 4 months for the first year, semiannually for the second year and annually thereafter. Cancers detected in the ureter and/or the urethra were coded as second primaries rather than as local or distant recurrence. Immunohistochemistry and Scoring Immunostaining for the testing cohort was performed at a single laboratory (Veripath Laboratories, UT Southwestern, Dallas, Texas) on tissue microarray slides. Immunostaining was performed on the Dako Autostainer (Carpinteria, California). Staining and scoring protocols for p53 (monoclonal mouse anti-human antibody Pab1801; Oncogene, Cambridge, Massachusetts; dilution 1:200), p21 (monoclonal mouse anti-human, SX118, Dako, dilution 1:200), prb (monoclonal mouse antihuman, RB1, Dako, dilution 1:75), p27 (monoclonal mouse anti-human, SX53G8, Dako, dilution 1:400), survivin (polyclonal rabbit antibody; Novus Biologicals, Littleton, Colorado; 1:100 dilution) and Ki-67 (monoclonal mouse anti-human, clone MIB-1, Dako, 1:500 dilution) have been previously described. 5,12,14,15,17,19,20 Multiple known positive control sections were included in each run. Tumor sections with the primary antibodies substituted with rabbit immunoglobulin fraction (Normal) and/or IgG 1 monoclonal were used as negative controls. We used bright field microscopy imaging coupled with advanced color detection software (Automated Cellular Imaging System, Clarient, Inc., Aliso Viejo, California). We obtained the mean, maximum, range and standard deviation of staining intensity and percent positive nuclei/area measurements by using 10 random hot spots within each specimen. The mean was calculated for data analysis. As ideal cutoffs and definitions for these biomarkers we used the same positivity criteria previously explored and validated in the literature Nuclear p53 immunoreactivity was considered altered when samples demonstrated at least 10% nuclear reactivity. p21 immunoreactivity was considered altered when samples demonstrated no detectable or only low levels of p21 nuclear staining. Tumors with no prb expression and those with a strong homogeneous staining pattern (more than 50%) were categorized as having altered prb status. Nuclear p27 immunoreactivity was considered altered when samples demonstrated less than 30% nuclear reactivity. The percentages of cells expressing survivin were classified as normal (no reactivity or few focally positive cells) vs altered (more than 10% cells expressing survivin). The Ki-67 labeling index was considered high when samples demonstrated 20% or greater reactivity. Statistical Analyses Outcomes were measured by time to disease recurrence or to UCB specific mortality. The cause of death was determined by the treating physicians, by chart review corroborated by death certificates, or by death certificates alone. To reduce bias in attribution of cause of death only subjects who had UCB listed in part I of the death certificate were considered to have died of UCB in this study. Perioperative mortality (any death within 30 days of surgery or before discharge home) was censored at time of death for UCB specific survival analyses. Univariable and multivariable Cox regression models addressed time to recurrence and UCB specific mortality after RC. In all models proportional hazards assumptions were systematically verified using the Grambsch-

3 80 BIOMARKER EXPRESSION IN PT1 BLADDER CANCER Therneau residual based test. 21 The change in predictive accuracy resulting from the addition of biomarkers to standard predictor variables was quantified with Harrell s concordance index. 22 The predictive accuracy describes whether the predicted values match the actual values of the target field within the incertitude due to statistical fluctuations and noise in the input data values. Internal validation was performed using 200-bootstrap resamples. 22 Disease recurrence and cancer specific survival estimates were graphically represented using the Kaplan-Meier method and compared using the log rank test. All reported p values are 2-sided and statistical significance was set at All statistical tests were performed with SPSS (version 13.0) or S-Plus Professional. RESULTS Table 1. Descriptive features of 80 patients Median pt age (range) 64.6 (34 91) No. gender (%): M 61 (76.3) F 19 (23.8) No. pathological grade (%): Low 27 (33.8) High 53 (66.3) No. concomitant Ca in situ (%): Neg 26 (32.5) Pos 54 (67.5) No. lymphovascular invasion (%): Absent 74 (92.5) Present 6 (7.5) No. lymph node status (%): pn0 74 (92.5) pn1-2 6 (7.5) Median lymph nodes removed (range) 17 (5 40) No. p53 status (%): Normal 60 (75) Altered 20 (25) No. p21 status (%): Normal 43 (53.8) Altered 37 (46.3) No. prb status (%): Normal 49 (61.3) Altered 31 (38.8) No. p27 status (%): Normal 52 (65.0) Altered 28 (35.0) No. survivin status (%): Normal 41 (51.3) Altered 39 (48.8) No. Ki-67 status (%): Normal 53 (66.3) Altered 27 (33.8) Descriptive Characteristics Table 1 summarizes clinicopathological characteristics and marker expressions of the 80 patients. None of the clinicopathological characteristics was associated with any of the markers. Disease recurred in 13 patients (16.3%) and 21 (26.3%) were dead at analysis. Of these 80 patients 11 died of metastatic UCB (13.8%) and 10 died of other causes without evidence of UCB progression (12.5%). The 2, 5 and 8-year recurrence-free survival estimates were 87% (standard error 4), 76% (6) and 76% (6), respectively. The 2, 5 and 8-year UCB specific survival probability estimates were 93% (3), 82% (6) and 76% (6), respectively. Median followup was 61.6 months (range 0.1 to 195.4) for those patients alive at analysis. Association of p53, prb, p21, p27, Survivin and Ki-67 as Individual Variables With Outcomes Tables 2 and 3 show the univariate and multivariate Cox regression models for the prediction of disease recurrence after RC. On univariate analyses the status of p53, p27 and Ki-67 was associated with disease recurrence (p 0.005, p and p 0.046, respectively) and UCB specific mortality (p 0.006, p and p 0.019, respectively). On multivariable analyses adjusted for the effects of lymphovascular invasion, lymph node status and pathological grade, p53, p27 and Ki-67 retained their association with disease recurrence (p 0.033, p and p 0.021, respectively) and UCB specific mortality (p 0.016, p and p 0.009, respectively). The concordance indices of p53, prb, p21, p27, survivin and Ki-67 for disease recurrence were 65%, 61.9%, 50.2%, 66.3%, 56.4% and 61.5%, respectively. The concordance indices of p53, prb, p21, p27, survivin and Ki-67 for cancer specific survival were 68.3%, 62.5%, 55.1%, 64.5%, 54.8% and 70.7%, respectively. The concordance index of a base model that included tumor grade, lymph node status, lymphovascular invasion and concomitant carcinoma in situ for disease recurrence and UCB specific mortality was 54.7% and 64.3%, respectively. Addition of p53 increased the concordance indices of the base model for disease recurrence and UCB specific mortality to 60% and 67.9%, respectively. Addition of p27 increased the concordance indices of the base model for disease recurrence and UCB specific mortality to 60.8% and 66.8%, respectively. Addition of Ki-67 increased the concordance indices of the base model for disease recurrence and UCB specific mortality to 59.4% and 69.3%, respectively. Association of p53, p27 and Ki-67 as Combined Variables With Clinical Outcomes We repeated the analyses creating 4 categories using p53, p27 and Ki-67, with all 3 normal, 1 altered, 2 altered and all 3 altered. Of the 80 patients 29 (36.3%) exhibited normal status of all 3 markers, 1 marker was altered in 32 of 80 (40%) tumors, 2 were altered in 14 of 80 (17.5%) and all 3 were altered in 5 of 80 (6.3%). The association of combined biomarkers with clinical outcomes is shown in tables 2 and 3. On univariable analysis the risk of disease recurrence (p for trend 0.001) and UCB specific mortality (p for trend 0.001) increased with the number of altered biomark-

4 BIOMARKER EXPRESSION IN PT1 BLADDER CANCER 81 Table 2. Univariate and multivariate Cox regression analyses predicting disease recurrence Univariate Analysis Multivariate Analysis With Individual Multivariate Analysis With Combined Hazard Ratio 95% CI p Value Hazard Ratio 95% CI p Value Hazard Ratio 95% CI p Value Pathological grade Lymphovascular invasion Lymph node metastasis Concomitant Ca in situ p53 Status p21 Status prb Status p27 Status Survivin status Ki-67 status Combination of p53, p27 Ki-67: None altered 1.00 Referent 1.00 Referent 1 Altered Altered Altered ers. On multivariable Cox proportional hazards regression analyses that adjusted for the effects of lymph node status, lymphovascular invasion and pathological grade, an increasing number of the 3 altered biomarkers was independently associated with greater risk of disease recurrence (p for trend 0.001) and UCB specific mortality (p for trend 0.001). The concordance index of a base model that included tumor grade, lymph node status, lymphovascular invasion and concomitant carcinoma in situ for disease recurrence and UCB specific mortality was 54.7% and 64.3%, respectively. The concordance indices of the combination of p53, p27 and Ki-67 into number of altered markers for prediction of disease recurrence and UCB specific mortality were 67.4% and 72.8%. Addition of the number of altered markers increased the concordance indices of the base model for disease recurrence and UCB specific mortality to 71.7% and 77.5%, respectively (figs. 1 and 2). DISCUSSION Patients with pt1 disease who have negative surgical margins and negative lymph nodes at RC are generally considered cured of disease. We found that 16% of our patients with pt1 experienced disease recurrence despite the fact that only 7.5% had positive nodes. This recurrence rate is in concordance with those reported in other series. 1 3 While recommending multimodal therapy does not seem reasonable with such high cure rates after RC there is a nonnegligible number of patients who experience disease progression and eventual death despite apparently adequate radical cystec- Table 3. Univariate and multivariate Cox regression analyses predicting cancer specific survival Univariate Analysis Multivariate Analysis With Individual Multivariate Analysis With Combined Hazard Ratio 95% CI p Value Hazard Ratio 95% CI p Value Hazard Ratio 95% CI p Value Pathological grade Lymphovascular invasion Lymph node metastasis Concomitant Ca in situ p53 Status p21 Status prb Status p27 Status Survivin status Ki-67 status Combination of p53, p27 Ki-67: None Altered 1.00 Referent 1.00 Referent 1 Altered Altered Altered

5 82 BIOMARKER EXPRESSION IN PT1 BLADDER CANCER Figure 1. Kaplan-Meier plots representing estimated recurrence-free survival rates for combinations of biomarkers p53, p27 and Ki-67 in 80 patients with pt1 UCB treated with RC and bilateral pelvic lymphadenectomy. tomy and bilateral lymphadenectomy with curative intent. Currently patients with pt1 UCB are not considered appropriate candidates for adjuvant therapy based on the relatively low risk of disease progression after RC. Dissecting the bioclinical heterogeneity of patients with pt1 disease will lead to more accurate attribution of individual risk, thereby allowing identification of those who may benefit from intensified therapy. Indeed the early identification of patients with pt1 UCB at increased risk for disease progression after radical cystectomy and bilateral lymphadenectomy represents one of the main challenges in the management of UCB. Since conventional clinical and pathological features fail to adequately capture the individual clinical prognosis of patients with pt1 disease there is a strong need to integrate prognostic biomarkers in the management of UCB. In the present study we investigated the usefulness of 6 biomarkers in the prediction of disease recurrence and cancer specific death in 80 patients with pt1 UCB followed for a median of 5 years. The most informative biomarkers for the prediction of oncological outcomes were tissue expression of p53, p27 and Ki-67. Patients with alteration of all 3 of these biomarkers were at 24 and 29 times increased risk for disease recurrence and cancer specific death compared to those without alterations of any of these biomarkers. In addition, the combination of biomarkers improved the risk prediction of disease recurrence and cancer specific death based on conventional clinical and pathological risk factors as evidenced by a 17% and 13% gain in concordance index, respectively. Furthermore, the combination of these 3 biomarkers was a stronger predictor of disease recurrence and death than any clinical and pathological features, suggesting a more powerful association with the biological and clinical potential of UCB. These findings indicate that the combination of p53, p27 and Ki-67 expression patterns can help identify patients with pt1 UCB who should be considered for early aggressive treatment following RC such as adjuvant chemotherapy. If prospective studies confirm the strong predictive power of p53, p27 and Ki-67 expression patterns one could consider adjuvant chemotherapy for this small subset of patients with alteration of these biomarkers. We confirmed previous findings that a combined assessment of molecular biomarkers provides optimized prognostic power compared to a single marker. 5,8,11 13,19,23 The combination of independent, complementary markers provided a more accurate prediction of outcome due to the association with different paths for cancer progression. The tumor suppressor p53 inhibits phase specific Figure 2. Kaplan-Meier plots representing estimated cancer specific survival rates for combinations of biomarkers p53, p27 and Ki-67 in 80 patients with pt1 UCB treated with RC and bilateral pelvic lymphadenectomy.

6 BIOMARKER EXPRESSION IN PT1 BLADDER CANCER 83 cell cycle progression (G1-S) and mediates control through the transcriptional activation of p21 WAF1/CIP1. The multiple functions of p53 include the inhibition of abnormal growth of cells, induction of programmed cell death, 24 regulation of DNA repair 25 and inhibition of angiogenesis. 26 p27 is also a key regulator of progression from G1 to S-phase where it induces a block in the cell cycle. Defective regulation at the p27 checkpoint can result in uncontrolled cellular proliferation. Ki-67 is an established marker of cell proliferation, present during the G1, S, G2 and M stages of the cell cycle. 27 Clinical studies have revealed that alterations of the protein expression of these biomarkers are associated with worse outcomes in UCB. 4,6,8,11 13,17,28 30 Our study is limited by its retrospective nature, the use of immunohistochemical techniques and the small sample size, potentially precluding the detection of a prognostic value attributed to clinical and pathological parameters. To overcome problems associated with application of immunohistochemistry such as choice of antibody, variability in scoring and inconsistency in specimen handling we used standardized technical procedures such as tissue microarrays, an automated autostainer and automated scoring systems based on bright field microscopy imaging coupled with advanced color detection software as described and applied previously. CONCLUSIONS Assessment of p53, p27 and Ki-67 in the pt1 cystectomy specimen improves our current prediction of UCB recurrence and survival. An increasing number of these altered markers was strongly associated with disease recurrence and survival following RC. A combination of biomarkers improves prediction compared to single biomarkers likely due to a more comprehensive biological profile. Immunohistochemical staining for these biomarkers may aid in the decision making process in terms of the prospect of adjuvant chemotherapy in patients with pt1 disease after RC. After further validation our findings may improve the identification of high risk patients among those supposed to have a favorable outcome and a high chance of cure. ACKNOWLEDGMENTS Mr. Starbucks provided logistic support. REFERENCES 1. Madersbacher S, Hochreiter W, Burkhard F, Thalmann GN, Danuser H, Markwalder R et al: Radical cystectomy for bladder cancer today a homogeneous series without neoadjuvant therapy. J Clin Oncol 2003; 21: Shariat SF, Karakiewicz PI, Palapattu GS, Lotan Y, Rogers CG, Amiel GE et al: Outcomes of radical cystectomy for transitional cell carcinoma of the bladder: a contemporary series from the Bladder Cancer Research Consortium. J Urol 2006; 176: Stein JP, Lieskovsky G, Cote R, Groshen S, Feng AC, Boyd S et al: Radical cystectomy in the treatment of invasive bladder cancer: longterm results in 1,054 patients. J Clin Oncol 2001; 19: Cote RJ, Dunn MD, Chatterjee SJ, Stein JP, Shi SR, Tran QC et al: Elevated and absent prb expression is associated with bladder cancer progression and has cooperative effects with p53. Cancer Res 1998; 58: Shariat SF, Tokunaga H, Zhou J, Kim J, Ayala GE, Benedict WF et al: p53, p21, prb, and p16 expression predict clinical outcome in cystectomy with bladder cancer. J Clin Oncol 2004; 22: Stein JP, Ginsberg DA, Grossfeld GD, Chatterjee SJ, Esrig D, Dickinson MG et al: Effect of p21waf1/cip1 expression on tumor progression in bladder cancer. J Natl Cancer Inst 1998; 90: Zlotta AR, Noel JC, Fayt I, Drowart A, Van Vooren JP, Huygen K et al: Correlation and prognostic significance of p53, p21waf1/cip1 and Ki-67 expression in patients with superficial bladder tumors treated with bacillus Calmette-Guerin intravesical therapy. J Urol 1999; 161: Chatterjee SJ, Datar R, Youssefzadeh D, George B, Goebell PJ, Stein JP et al: Combined effects of p53, p21, and prb expression in the progression of bladder transitional cell carcinoma. J Clin Oncol 2004; 22: Korkolopoulou P, Christodoulou P, Konstantinidou AE, Thomas-Tsagli E, Kapralos P and Davaris P: Cell cycle regulators in bladder cancer: a multivariate survival study with emphasis on p27kip1. Hum Pathol 2000; 31: Migaldi M, Sgambato A, Garagnani L, Ardito R, Ferrari P, De Gaetani C et al: Loss of p21waf1 expression is a strong predictor of reduced survival in primary superficial bladder cancers. Clin Cancer Res 2000; 6: Shariat SF, Zlotta AR, Ashfaq R, Sagalowsky AI and Lotan Y: Cooperative effect of cell-cycle regulators expression on bladder cancer development and biologic aggressiveness. Mod Pathol 2007; 20: Shariat SF, Ashfaq R, Sagalowsky AI and Lotan Y: Predictive value of cell cycle biomarkers in nonmuscle invasive bladder transitional cell carcinoma. J Urol 2007; 177: Shariat SF, Karakiewicz PI, Ashfaq R, Lerner SP, Palapattu GS, Cote RJ et al: Multiple biomarkers improve prediction of bladder cancer recurrence and mortality in patients undergoing cystectomy. Cancer 2008; 112: Shariat SF, Ashfaq R, Karakiewicz PI, Saeedi O, Sagalowsky AI and Lotan Y: Survivin expression is associated with bladder cancer presence, stage, progression, and mortality. Cancer 2007; 109: Karam JA, Lotan Y, Karakiewicz PI, Ashfaq R, Sagalowsky AI, Roehrborn CG et al: Use of combined apoptosis biomarkers for prediction of bladder cancer recurrence and mortality after radical cystectomy. Lancet Oncol 2007; 8: Lopez-Knowles E, Hernandez S, Kogevinas M, Lloreta J, Amoros A, Tardon A et al: The p53 pathway and outcome among patients with T1G3 bladder tumors. Clin Cancer Res 2006; 12: Margulis V, Shariat SF, Ashfaq R, Sagalowsky AI and Lotan Y: Ki-67 is an independent predictor of bladder cancer outcome in patients treated with radical cystectomy for organ-confined disease. Clin Cancer Res 2006; 12: 7369.

7 84 BIOMARKER EXPRESSION IN PT1 BLADDER CANCER 18. Shariat SF, Palapattu GS, Karakiewicz PI, Rogers CG, Vazina A, Bastian PJ et al: Discrepancy between clinical and pathologic stage: impact on prognosis after radical cystectomy. Eur Urol 2007; 51: Shariat SF, Kim J, Raptidis G, Ayala GE and Lerner SP: Association of p53 and p21 expression with clinical outcome in patients with carcinoma in situ of the urinary bladder. Urology 2003; 61: Shariat SF, Weizer AZ, Green A, Laucirica R, Frolov A, Wheeler TM et al: Prognostic value of P53 nuclear accumulation and histopathologic features in T1 transitional cell carcinoma of the urinary bladder. Urology 2000; 56: Grambsch PM and Therneau TM: Proportional hazards tests and diagnostics based on weighted residuals. Biometrika 1994; 81: Harrell FE Jr, Califf RM, Pryor DB, Lee KL and Rosati RA: Evaluating the yield of medical tests. JAMA 1982; 247: Grossman HB, Liebert M, Antelo M, Dinney CP, Hu SX, Palmer JL et al: p53 and RB expression predict progression in T1 bladder cancer. Clin Cancer Res 1998; 4: Heinrichs S and Deppert W: Apoptosis or growth arrest: modulation of the cellular response to p53 by proliferative signals. Oncogene 2003; 22: Albrechtsen N, Dornreiter I, Grosse F, Kim E, Wiesmuller L and Deppert W: Maintenance of genomic integrity by p53: complementary roles for activated and non-activated p53. Oncogene 1999; 18: Vogelstein B, Lane D and Levine AJ: Surfing the p53 network. Nature 2000; 408: Cattoretti G, Becker MH, Key G, Duchrow M, Schluter C, Galle J et al: Monoclonal antibodies against recombinant parts of the Ki-67 antigen (MIB 1 and MIB 3) detect proliferating cells in microwave-processed formalin-fixed paraffin sections. J Pathol 1992; 168: Sarkis AS, Dalbagni G, Cordon-Cardo C, Zhang ZF, Sheinfeld J, Fair WR et al: Nuclear overexpression of p53 protein in transitional cell bladder carcinoma: a marker for disease progression. J Natl Cancer Inst 1993; 85: Shariat SF, Ashfaq R, Sagalowsky AI and Lotan Y: Correlation of cyclin D1 and E1 expression with bladder cancer presence, invasion, progression, and metastasis. Hum Pathol 2006; 37: Esrig D, Elmajian D, Groshen S, Freeman JA, Stein JP, Chen SC et al: Accumulation of nuclear p53 and tumor progression in bladder cancer. N Engl J Med 1994; 331: EDITORIAL COMMENT Shariat et al evaluated the predictive value of pathological data and several immunohistochemical markers in 80 patients with high grade stage T1 urothelial bladder carcinoma who were treated with radical cystectomy. In addition to the well established pathological markers they analyzed several immunohistochemical markers associated with the cell cycle, apoptosis and proliferation (p53, p21, p27, prb, survivin, Ki-67). Using univariate and multivariate analyses the authors showed that patients with p53, p27 and Ki-67 immunoreactivity alterations had 24 times greater risk of recurrence and 29 times greater risk of cancer specific mortality than those without alterations in any of these markers. The authors should be commended for their continued efforts in evaluating and identifying markers that predict patients at high risk for recurrence for pathologically localized bladder cancer that has been treated by radical cystectomy and pelvic lymphadenectomy. Despite our best surgical attempts approximately 20% of patients with negative surgical margins and negative lymph nodes experience disease recurrence and eventually die of advanced metastatic bladder cancer. Therefore, identifying markers associated with disease recurrence in patients who are presumably disease-free after radical cystectomy may help institute adjuvant therapies, which may improve cancer specific survival. It is clear that from this study and others that a single molecular marker will not adequately stratify patients who are risk of disease recurrence of high grade localized bladder carcinoma. The study is a step forward but until larger cohorts of patients are evaluated in multi-institutional studies and immunohistochemical studies are standardized, it will be difficult to imagine that immunomarkers will have a significant role in determining which cohort of patients with localized bladder cancer may benefit from more aggressive therapies. The authors have attempted to overcome some of these shortcomings by using automated autostainer and automated scoring systems to standardize their techniques. Aria F. Olumi Massachusetts General Hospital Harvard Medical School Boston, Massachusetts