Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cancer Epidemiol Biomarkers Prev. Author manuscript; available in PMC 2013 September 1.
Published in final edited form as:
PMCID: PMC3480660

Cell-Cycle Control in Urothelial Carcinoma: Large-scale Tissue Array Analysis of Tumor Tissue from Maine and Vermont



Cell-cycle proteins are important predictive markers in urothelial carcinoma but may also exhibit exposure-specific heterogeneity.


Tumor tissue from 491 bladder cancer cases enrolled in the Maine and Vermont component of the New England Bladder Cancer Study was assembled as tissue microarrays and examined for aberrant expression of p53, p63, p16, cyclin D1, Rb, and Ki-67. The association between expression and histopathology, demographics, and cigarette smoking was examined using χ2 tests, multivariable Poisson, and multinomial regression models.


We found that overexpression of p53 and Ki-67 was associated with high-stage/grade tumors [relative risk (RR), 1.26; Ptrend = 0.003; and RR, 3.21; Ptrend < 0.0001, respectively], whereas expression of p63 and p16 was decreased in high-stage/grade tumors (RR, 0.52; Ptrend < 0.0001; and RR, 0.88; Ptrend = 0.04, respectively). No significant aberrations of cell-cycle proteins were identified using various smoking variables and multiple statistical models.


The results of this population-based study of histologically confirmed urothelial carcinomas show significant aberration of cell-cycle proteins p53, p63, p16, and Ki-67, but not Rb or cyclin D1. p53 showed the most significant heterogeneity with respect to tumor stage and grade, especially when stratified for different staining intensities using novel digital image analysis techniques. Our findings do not support that smoking modifies expression of cell-cycle proteins.


Our study shows significant heterogeneity in the expression of key cell-cycle proteins that are associated with disease progression in bladder cancer. Further studies may lead to the identification of biomarkers and their multiplexed interactions as useful prognostic and therapeutic targets.


Malignancies of the urinary bladder pose a significant public health burden. Approximately, 69,000 new cases of bladder cancer are diagnosed annually in the United States (1). Bladder cancer occurs more often in individuals older than 50 years and is more common in men than in women (1). Bladder cancer is the fourth leading cancer in men in the United States. Clinically, bladder cancer is characterized by frequent recurrences with possible progression to more aggressive lesions and by poor prognosis for advanced/high-stage disease (2).

On the other hand, tumors of similar histologic grade and stage may exhibit different clinical outcomes and often display distinct molecular profiles, suggesting a critical role for genetic alterations in urothelial carcinogenesis (35). Insights into the pathogenesis of bladder cancer have improved as new technologies allow for the identification of key molecular markers that are altered early in the carcinogenic process. A variety of molecular genetic alterations have been identified in bladder cancer including activation of oncogenes, inactivation of tumor suppressor genes (p53, Rb, p16), and alterations in telomerase and DNA methylation patterns, several of which have proven to be potentially useful prognostic and therapeutic targets (6). Molecular alterations that affect cell cycling are important in both the etiology and subsequent progression of urothelial carcinomas (7). Disruptions in cell-cycle control not only result in increased cellular proliferation but also affect cellular checkpoints that respond to DNA damage.

The most common genetic damage/alterations in urothelial carcinoma are p53 mutations and partial loss of chromosome 9, which is the site of several tumor suppressor genes (p16, p21, and p14; ref. 8). p53 functions in the inhibition of cell growth, response to DNA damage, and as a gatekeeper to apoptotic pathways. Alterations of p53, either due to mutation or protein dysregulation, have been reported to be more common among invasive bladder cancer cases (9). p63 is a recently discovered homologue of p53 that is thought to control cellular growth through similar pathways. p63 has been shown to play a role in the differentiation of the bladder epithelium and is expressed in normal urothelium but downregulated in invasive bladder cancer (10). A major target in the pathogenesis of human urinary bladder cancer is the 9p21 locus harboring the CDKN2A/ARF gene, which encodes 2 cell-cycle regulatory proteins, p16INK4a and p14ARF. Immunohistochemical studies of bladder cancer suggest that functional reduction of the tumor suppressor gene p16 may be associated with more aggressive clinical behavior (1113). p53, p63, and p16 all inhibit cyclin D1, a protein with oncogenic properties that regulates cell-cycle progression by activating cyclin-dependent kinases 4 and 6, which then phosphorylate the retinoblastoma protein (Rb) leading to progression through the G1–S checkpoint (14). Overexpression of cyclin D1 has been implicated in various cancers, including carcinomas of the bladder (15). Increased cell proliferation can be measured by the Ki-67 protein, which is expressed during all phases of the cell cycle except G0. Expression of Ki-67 protein indicates loss of cell-cycle regulation and increased cellular proliferation of tumor tissue (16).

Whereas the role of aberrant cell-cycle control in bladder cancer is well established, the possible correlation with case characteristics and environmental exposures, particularly smoking-–a well-documented bladder carcinogen—is not yet understood (17). In this analysis, we measured the expression of a panel of key cell-cycle proteins—p53, p63, p16, cyclin D1, Rb, and Ki-67—using high throughput tissue microarrays (TMA) and novel automated scoring for a large case series of histologically confirmed bladder cancer cases from a multicenter case–control study in New England. We investigated the association between the expression of cell-cycle proteins and demographic and histopathologic characteristics to further elucidate the role of cell-cycle control in bladder cancer etiology and, in particular, the possible effects of cigarette smoking on the expression of these key cell-cycle proteins.

Materials and Methods

Study population

The New England Bladder Cancer Study (NEBCS) is a population-based case–control study that includes all patients with a histologically confirmed carcinoma of the urinary bladder newly diagnosed between September 1, 2001, and October 31, 2004 (ME and VT) or between January 1, 2002, and July 31, 2004 (NH), among residents of these 3 states aged 30 to 79 years. Patient ascertainment in each state during the study period was conducted through hospital pathology departments, hospital cancer registries, and the state cancer registries (18). Of a total of 810 eligible interviewed cases of incident bladder cancer from ME (n = 591) and VT (n = 219), 782 cases were diagnosed as urothelial carcinoma. Consent to obtain pathology material was obtained from 766 cases and material was received for 734 (95.8%) of these cases. Adequate quantity of tumor tissue was available from 649 (88.4%) of the received cases, which were assembled into tissue arrays. After scoring the tumor tissue cores on the arrays, a total of 491 cases (75.7%) had sufficient quality tumor tissue for statistical analysis. The study, including tissue collection, was approved by the NIH Institutional Review Board and the boards of the participating institutions.

Smoking assessment

Information on smoking was obtained from detailed computer-assisted personal interviews. “Never smokers” were defined as subjects who smoked less than 100 cigarettes over their lifetime. Subjects who smoked more than one cigarette per day for at least 6 months were categorized as “current smokers” if they were smoking at the time of interview or quit within 1 year of the reference date and “former smokers” if they quit smoking 1 year or more before the diagnosis date for cases or selection date for controls (18). Occasional smokers defined as subjects who smoked more than 100 cigarettes but never consumed cigarettes regularly (i.e., at least 1 cigarette per day for at least 6 months) were excluded from the study, resulting in 764 subjects with smoking status data.

Pathology review and TMA construction

Histopathologic characteristics including morphology, tumor grade, and tumor stage were obtained from the ME and VT state cancer registries and through independent assessment by the study pathologist (A. R. Schned). Formalin-fixed, paraffin-embedded blocks from 626 bladder cancer cases were collected from the pathology archives of the medical facilities where the patients were diagnosed. The specimens were assembled into TMAs using a Beecher MTA-1 (Beecher Instruments). Each case was represented by a single 1-mm tissue core. Five-micrometer tissue sections were cut, placed on glass slides using a tape transfer method, and UV cross-linked (Instrumedics).


Immunohistochemical staining of the TMAs was conducted with standard avidin–biotin peroxidase methods using the Dako Autostainer (Dako). Staining conditions for p53, p63, p16, cyclin D1, Rb, and Ki-67 are described in Table 1. Briefly, slides were deparaffinized according to standard protocols, treated with 3% hydrogen peroxide to block endogenous peroxidase activity, and treated with antigen retrieval solution (Dako Target Retrieval Solution) for 20 minutes in a steamer before application of the primary antibody. The reactions were developed in 3,3′-diaminobenzidine (DAB), and the slides were counterstained with hematoxylin, dehydrated, and mounted using standard procedures. Representative immunohistochemical staining for p53 is shown in Fig. 1 and for Ki-67, cyclin D1, p16, and Rb in Fig. 2.

Figure 1
Representative immunohistochemical staining for p53 in bladder tumors (×400 magnification). A, tumor with predominantly strong p53 staining. B, digital image analysis of A: red, strong (3+); orange, moderate (2+); yellow, weak (1+); blue, negative. ...
Figure 2
Representative immunohistochemical staining for Ki67 (A), cyclin D1 (B), p16 (C), and Rb (D) in bladder tumors (×400 magnification).
Table 1
Antibodies used in immunohistochemical analysis

Immunohistochemistry scoring

Digital images were created by scanning whole slides using an Aperio CS instrument (Aperio Technologies) with a ×20 objective. On each core, tumor tissue was selected for image analysis by positive or negative annotation. A nuclear staining algorithm (Aperio Technologies) was used to develop quantitative scoring models to compute the percentage of positive cells for each marker. Representative examples of quantitative p53 scoring are shown in Fig. 1 (strong p53 staining shown in A and B, predominantly negative to weak p53 in C and D).

Statistical analysis

Protein expression levels were analyzed as positive versus negative based on predetermined cutoff points for the total percentage of positive cells in the core. The cutoff values used to define negative and positive staining were based on the distribution of the percentage of positive cells for each marker for all cases combined and were set at the 25th percentile, resulting in the following cutoff values: 5% for p53, p63, and Ki-67; 0.5% for cyclin D1, and 30% for p16 and Rb.

Pearson χ2 statistic was used to assess differences in the proportion of cases positive for each marker by subject and tumor characteristics. Spearman rank correlation coefficients were calculated to examine relationships between cell-cycle markers. For p63, Rb protein, Ki-67, cyclin D1, and p16, Poisson regression (PROC GENMOD, SAS 9.1) was used to estimate relative risks (RR) and 95% confidence intervals (CI) for the association between the mean count of positive cells for markers with tumor grade, stage/grade, and smoking variables. The combined stage/grade variable was stratified into 3 groups: low-stage/grade comprised all Ta cases with histology grade I; intermediate stage/grade comprised Ta cases with histology grade II or Tis/T1/T2 cases with histology grades I/II; and high-stage/grade comprised all cases with either stages T3/T4 or histology grade III. Smoking was analyzed as smoking status (never/former/current), duration of smoking in quartiles, and pack-years in quartiles for all smokers or separately in former and current smokers. Models were adjusted for age at diagnosis in quartiles, sex, race (white/other), and state (ME/VT). The logarithm of the total number of cells analyzed per core was included as an offset in the models. We allowed for over- or underdispersion in the data. For p53 staining, cells were analyzed categorically as weak, moderate, or strong expression using multinomial regression models to estimate association parameters between the variables listed above (PROC GENMOD, SAS 9.1, with the cumulative logistic link). In a sensitivity analysis, we also evaluated the number of moderately or strongly stained cells for p53 combined using a Poisson model. All statistical analyses were conducted using SAS version 9.1 statistical software. Results were considered statistically significant if P values were less than 0.05 for a 2-sided test.


Characteristics of the 491 cases with sufficient quality tissue for statistical analysis are presented in Table 2. The cohort is largely Caucasian (93.9%) and included more males than females (78.6% vs. 21.4%). Approximately, half of the cases had grade I tumors (47.3%) and were in the low combined stage/grade category (45.5%). Cases from ME and VT did not differ with respect to age, gender, smoking, histologic grade, or the combined stage/grade variable; however, VT had a more diverse population in regard to race. When compared with excluded cases, those included in this analysis did not differ in any of the case characteristics but had a larger proportion of cases from ME than VT (P < 0.0001), which can be explained by a lower tissue recovery rate from VT (73.7%) than ME (86.6%). Automated image analysis was used to quantify nuclear expression of the markers studied. Images of the TMAs were manually annotated by a pathologist (P. Lenz) to identify regions of malignant epithelium, and then these regions were analyzed with a nuclear algorithm (Fig. 1).

Table 2
Characteristics of bladder cancer cases in the NEBCS, cases included versus cases excluded from the analysis


Distribution of p53 analyzed as a binary variable (≤5%, >5% cells having positive nuclear staining) in univariate analysis is given in Table 3. Overall p53 positivity showed no association with any personal or tumor characteristics, but a greater proportion of patients from VT had p53-positive tumors than patients from ME (85.1% vs. 74.4%, P = 0.02). p53 protein expression was also evaluated in 2 intensity categories: weak (1+) versus moderate to strong (2+/3+), based on the percentage of positive bladder tumor cells at a given intensity divided by the total number of bladder tumor cells analyzed per core. Moderate to strong p53 expression (2+/3+), which is indicative of p53 inactivation, was significantly higher in tumors with high histologic grade (P = 0.006) and the combined stage/grade variable (P = 0.01). These associations became even more significant when analyzing for strong p53 expression only (P for both grade and stage/grade < 0.0001, data not shown). In contrast, weak p53 expression (1+), also considered to represent normal p53 expression (19), was significantly higher in low-grade tumors and in tumors with a low combined stage/grade variable (P < 0.0001, data not shown). In multinominal regression models that used p53 expression in 4 categories (negative, weak, moderate, strong; Table 4), p53 expression significantly increased with higher stage/grade tumors (RR for high-stage/grade tumors, 1.26; Ptrend = 0.003). Several statistical models were applied to conduct a detailed investigation of the effect of smoking on p53 expression, including smoking status (never/ever/current), duration of smoking, pack-years smoked, and pack-years stratified for current and former smokers. No significant associations were found between p53 expression and any of the smoking variables tested. Poisson models that analyzed the count of combined number of moderate to strong p53 cells gave very similar results (data not shown).

Table 3
Univariate analysis of p53 expression
Table 4
RRs of cell-cycle protein expression and case characteristics


p63 percent positivity was assessed as binary variable (≤5%, >5%) and using the number of positive cells in a Poisson model that included the total number of cells in the core as an offset. p63 expression was significantly lower in grade III tumors (P < 0.0001) and tumors with high combined stage/grade variable (P < 0.0001) compared with cases with grade I/II tumors and low-to-intermediate stage/grade tumors (data not shown). These results were confirmed in multivariable Poisson models (Table 4), which showed that the RR of p63 expression was inversely associated with increasing tumor stage/grade (RR for high-stage/grade tumors, 0.52; Ptrend < 0.0001). Interestingly, p63 expression was lower in males than in females (P = 0.03).


In Poisson analysis, the RR of p16 expression decreased with increasing tumor stage/grade (RR for high-stage/ grade tumors, 0.88; Ptrend = 0.04; Table 4). It was also lower in ME than in VT (RR, 1.16; Ptrend = 0.01).

Cyclin D1

In univariate analysis, the proportion of cyclin D1–positive cases decreased significantly with increasing tumor grade (P = 0.002) and stage/grade (P = 0.0007; data not shown). However, in adjusted Poisson models, cyclin D1 expression was higher in females than in males (RR, 1.45; P = 0.02) but was not associated with stage/ grade (Table 4).


In Poisson analysis, the RR of Rb expression was higher in VT than in ME (RR, 1.15; Ptrend = 0.004; Table 4).


Multivariable analysis (Table 4), using Poisson regression showed that Ki-67 expression increased significantly with increasing stage/grade (RR for high-stage/grade tumors as compared with low-stage/grade tumors, 3.21; Ptrend < 0.0001).


In this large population-based study of bladder cancer tumor heterogeneity, we examined changes in cell cycling as a function of tumor grade, stage, and patient characteristics including cigarette smoking to elucidate the specific role of aberrant cell cycling in urothelial carcinomas. To our knowledge, this is the largest study of well-characterized incident cases of bladder cancer that analyzed a comprehensive set of key cell-cycle proteins using novel automated scoring methods.

Our results show that aberrant protein expression of p53, p63, p16, and Ki-67, but not Rb or cyclin D1, correlates with tumor stage and grade. Of the cell-cycle proteins examined in this study, p53 showed the most heterogenic expression pattern with respect to tumor stage and grade, especially when stratified for different staining intensities using novel digital image analysis techniques. Whereas overall expression of p53 was not associated with tumor stage/grade as described in previous smaller studies (2022), we showed a significant relationship with staining intensity. One smaller study by Kelsey and colleagues suggested that the histologic character of p53 staining, especially the intensity, is more indicative of bladder cancer aggressiveness than the percentage of p53 positive nuclei (19). Applying novel, automated, digital image analysis, we were able to stratify p53 expression for staining intensity and showed that weak p53 expression was significantly higher in low-stage/grade tumors (P < 0.0001), whereas strong p53 and moderate-to-strong expression were higher in high-stage/grade tumors (P < 0.0001 and P = 0.01, respectively). These findings underscore the need for the development of p53 analysis protocols that take into account p53 levels and intensity.

Other cell-cycle molecules associated with stage and grade were p63, p16, and Ki-67. We found an inverse association between p63 expression and tumor stage/ grade similar to the findings of previous studies (2326). Although the exact targets of p63 are unknown, it is thought to regulate the cell cycle through p53-dependent pathways (27). Several studies have suggested a role for tumor suppressor genes Rb and p16 in bladder cancer progression, by showing that functional inactivation of Rb and p16 is associated with progression of bladder cancer to more aggressive disease(12), where as expression of p16 is associated with a better outcome in low-grade/ stage bladder cancers (13). Consistent with these data, we found that p16 expression significantly decreased with increasing tumor stage/grade. However, we identified no significant association between Rb expression and any of the histopathologic factors tested. Previous univariate studies have suggested that cyclin D1 may be used as an inverse indicator for the level of invasiveness in bladder cancer, but multivariate analysis showed that cyclin D1 alone was not an independent prognostic marker (15, 28, 29). Our results are in agreement with these previous studies in that cyclin D1 expression decreased withtumorstage/gradeinunivariate(datanotshown)but not multivariate models. The cell-cycle molecule with the highest RR associated with high tumor stage/grade analyzed in our study was Ki-67, a known predictor of tumor stage and grade in bladder cancer (16, 30, 31). Together, these results show significant aberration of cell cycling in the progression of urothelial carcinoma.

Cigarette smoking is a well-established risk factor for bladder cancer and resulted in a 2- to 3-fold increased risk compared with nonsmokers (18). However, it remains unclear to date whether smoking modifies cell cycling. A number of previous studies have examined the association between smoking and p53 protein alterations, but results have been inconsistent. Two studies found direct relationships between p53 nuclear accumulation and smoking status and smoking intensity (32, 33), whereas more recent studies have found no association (9, 20). Such inconsistencies may be attributed to insufficient power and to difficulty in differentiating staining intensities, which was improved by our use of an automated image analysis approach. Taking advantage of our large study population with detailed, historical smoking information, we found that smoking, perhaps the most important bladder cancer risk factor in Western populations, did not modify the expression of any of the key cell-cycle molecules examined. We observed that protein expression was virtually similar between smokers and never-smokers and across additional exposure indices presented in this analysis. These results indicate that smoking does not modify protein expression of key cell-cycle molecules that are documented to play important roles in bladder cancer progression. Therefore, the increased risk of bladder cancer in smokers identified in epidemiologic studies could not be accounted for by aberrant expression of cell-cycle proteins alone. p53 alterations observed in other studies may be induced by a combination of several endogenous and exogenous carcinogens and/or other molecular targets may be affected by the carcinogens in the cigarette smoke.

There were significantly more cases excluded from VT than from ME (P < 0.001), largely due to differences in tissue retrieval rates. However, this difference in the proportion of cases from the 2 states did not bias the distribution of demographic and tumor characteristics or the smoking variables (Table 2). Although differences by region were not observed for p16 and Rb protein expression in the univariate analysis, they were found in the multivariant model adjusted for age, gender, race, smoking status, and stage/grade. This finding may be due to additional differences in case characteristics that we were unable to evaluate in this study, such as occupational and environmental exposures or level of education.

In summary, the findings from our large and well-characterized population-based study of histologically confirmed urothelial carcinomas showed significant heterogeneity in protein expression of cell-cycle proteins that are predictive of disease progression. Further studies are needed to assess the molecular mechanisms in bladder cancer associated with various environmental exposures and to further elucidate their role in bladder tumorigenesis.


Grant Support

This project has been funded, in whole or in part, with federal funds from the National Cancer Institute, NIH, under Contract No. HHSN261200800001E. Mention of trade names, commercial products, or organizations does not imply endorsement by the U.S. Government. This research was supported by the Intramural Research Program of the NIH, Division of Cancer Epidemiology and Genetics (contract number N02-CP-01037).


Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Authors’ Contributions

Conception and design: P. Lenz, D. Baris, M. Jones, N. Rothman, D.T. Silverman, S.M. Hewitt, L.E. Moore

Development of methodology: P. Lenz, D. Baris, A.R. Schned, N. Roth-man, D.T. Silverman, S.M. Hewitt, L.E. Moore

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): P. Lenz, D. Baris, A.R. Schned, M. Takikita, M. Schwenn, A. Johnson, M. Kida, N. Rothman, D.T. Silverman, S.M. Hewitt, L.E. Moore

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): P. Lenz, R. Pfeiffer, D. Baris, M. Jones, N. Rothman, S.M. Hewitt, L.E. Moore

Writing, review, and/or revision of the manuscript: P. Lenz, R. Pfeiffer, A. R. Schned, A. Johnson, M. Jones, K.P. Cantor, N. Rothman, D.T. Silverman, S.M. Hewitt, L.E. Moore

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): P. Lenz, D. Baris, A.R. Schned, M.C. Poscablo, S.M. Hewitt

Study supervision: A. Johnson, D.T. Silverman, S.M. Hewitt, L.E. Moore


1. Silverman DT, Hartge P, Morrison AS, Devesa SS. Epidemiology of bladder cancer. Hematol Oncol Clin North Am. 1992;6:1–30. [PubMed]
2. Soloway MS, Bruck DS, Kim SS. Expectant management of small, recurrent, noninvasive papillary bladder tumors. J Urol. 2003;170:438–41. [PubMed]
3. Zhang ZF, Cordon-Cardo C, Rothman N, Freedman AN, Taylor JA. Methodological issues in the use of tumour markers in cancer epidemiology. IARC Sci Publ. 1997:201–13. [PubMed]
4. Schulz WA. Understanding urothelial carcinoma through cancer pathways. Int J Cancer. 2006;119:1513–8. [PubMed]
5. Cordon-Cardo C. Molecular alterations associated with bladder cancer initiation and progression. Scand J Urol Nephrol Suppl. 2008:154–65. [PubMed]
6. Baffa R, Letko J, McClung C, LeNoir J, Vecchione A, Gomella LG. Molecular genetics of bladder cancer: targets for diagnosis and therapy. J Exp Clin Cancer Res. 2006;25:145–60. [PubMed]
7. Grossfeld GD, Muscheck M, Stein JP, Chew K, Cote RJ, Ginsberg DA, et al. Cellular proliferation and cell-cell cycle regulatory proteins as prognostic markers for transitional cell carcinoma of the bladder. Adv Exp Med Biol. 1999;462:425–35. [PubMed]
8. Luis NM, Lopez-Knowles E, Real FX. Molecular biology of bladder cancer. Clin Transl Oncol. 2007;9:5–12. [PubMed]
9. Kelsey KT, Hirao T, Hirao S, Devi-Ashok T, Nelson HH, Andrew A, et al. TP53 alterations and patterns of carcinogen exposure in a U.S. population-based study of bladder cancer. Int J Cancer. 2005;117:370–5. [PubMed]
10. Urist MJ, Di Como CJ, Lu ML, Charytonowicz E, Verbel D, Crum CP, et al. Loss of p63 expression is associated with tumor progression in bladder cancer. Am J Pathol. 2002;161:1199–206. [PubMed]
11. Mhawech P, Greloz V, Oppikofer C, Szalay-Quinodoz I, Herrmann F. Expression of cell cycle proteins in T1a and T1b urothelial bladder carcinoma and their value in predicting tumor progression. Cancer. 2004;100:2367–75. [PubMed]
12. Quentin T, Henke C, Korabiowska M, Schlott T, Zimmerman B, Kunze E. Altered mRNA expression of the Rb and p16 tumor suppressor genes and of CDK4 in transitional cell carcinomas of the urinary bladder associated with tumor progression. Anticancer Res. 2004;24:1011–23. [PubMed]
13. Brunner A, Verdorfer I, Prelog M, Mayerl C, Mikuz G, Tzankov A. Large-scale analysis of cell cycle regulators in urothelial bladder cancer identifies p16 and p27 as potentially useful prognostic markers. Pathobiology. 2008;75:25–33. [PubMed]
14. Mitra AP, Datar RH, Cote RJ. Molecular pathways in invasive bladder cancer: new insights into mechanisms, progression, and target identification. J Clin Oncol. 2006;24:5552–64. [PubMed]
15. Shariat SF, Ashfaq R, Sagalowsky AI, Lotan Y. Association of cyclin D1 and E1 expression with disease progression and biomarkers in patients with nonmuscle-invasive urothelial cell carcinoma of the bladder. Urol Oncol. 2007;25:468–75. [PubMed]
16. Mulder AH, Van Hootegem JC, Sylvester R, ten Kate FJ, Kurth KH, Ooms EC, et al. Prognostic factors in bladder carcinoma: histologic parameters and expression of a cell cycle-related nuclear antigen (Ki-67) J Pathol. 1992;166:37–43. [PubMed]
17. Rossman TG. Mechanism of arsenic carcinogenesis: an integrated approach. Mutat Res. 2003;533:37–65. [PubMed]
18. Baris D, Karagas MR, Verrill C, Johnson A, Andrew AS, Marsit CJ, et al. A case-control study of smoking and bladder cancer risk: emergent patterns over time. J Natl Cancer Inst. 2009;101:1553–61. [PMC free article] [PubMed]
19. Kelsey KT, Hirao T, Schned A, Hirao S, Devi-Ashok T, Nelson HH, et al. A population-based study of immunohistochemical detection of p53 alteration in bladder cancer. Br J Cancer. 2004;90:1572–6. [PMC free article] [PubMed]
20. Moore LE, Smith AH, Eng C, DeVries S, Kalman D, Bhargava V, et al. P53 alterations in bladder tumors from arsenic and tobacco exposed patients. Carcinogenesis. 2003;24:1785–91. [PubMed]
21. Yeniyol CO, Suelozgen T, Vardar E, Sural S, Postaci H, Ayder AR. The relation of mutant p53 accumulation in transitional cell carcinoma of bladder with pathological stage, grade, recurrence and survival. Int Urol Nephrol. 2001;33:473–8. [PubMed]
22. Toktas G, Turkeri LN, Unluer E, Caliskan M, Aksoy B, Akdas A. Clinical significance of nuclear p53 protein accumulation in bladder cancer. Int Urol Nephrol. 1999;31:327–34. [PubMed]
23. Wu X, Hildebrandt MA, Chang DW. Genome-wide association studies of bladder cancer risk: a field synopsis of progress and potential applications. Cancer Metastasis Rev. 2009;28:269–80. [PubMed]
24. Comperat E, Bieche I, Dargere D, Ferlicot S, Laurendeau I, Benoit G, et al. p63 gene expression study and early bladder carcinogenesis. Urology. 2007;70:459–62. [PubMed]
25. Zigeuner R, Tsybrovskyy O, Ratschek M, Rehak P, Lipsky K, Langner C. Prognostic impact of p63 and p53 expression in upper urinary tract transitional cell carcinoma. Urology. 2004;63:1079–83. [PubMed]
26. Koga F, Kawakami S, Fujii Y, Saito K, Ohtsuka Y, Iwai A, et al. Impaired p63 expression associates with poor prognosis and uroplakin III expression in invasive urothelial carcinoma of the bladder. Clin Cancer Res. 2003;9:5501–7. [PubMed]
27. Melino G, Lu X, Gasco M, Crook T, Knight RA. Functional regulation of p73 and p63: development and cancer. Trends Biochem Sci. 2003;28:663–70. [PubMed]
28. Takagi Y, Takashi M, Koshikawa T, Sakata T, Ohshima S. Immunohistochemical demonstration of cyclin D1 in bladder cancers as an inverse indicator of invasiveness but not an independent prognostic factor. Int J Urol. 2000;7:366–72. [PubMed]
29. Shariat SF, Ashfaq R, Sagalowsky AI, Lotan Y. Correlation of cyclin D1 and E1 expression with bladder cancer presence, invasion, progression, and metastasis. Hum Pathol. 2006;37:1568–76. [PubMed]
30. Bol MG, Baak JP, van Diermen B, Buhr-Wildhagen S, Janssen EA, Kjellevold KH, et al. Proliferation markers and DNA content analysis in urinary bladder TaT1 urothelial cell carcinomas: identification of subgroups with low and high stage progression risks. J Clin Pathol. 2003;56:447–52. [PMC free article] [PubMed]
31. Margulis V, Lotan Y, Karakiewicz PI, Fradet Y, Ashfaq R, Capitanio U, et al. Multi-institutional validation of the predictive value of Ki-67 labeling index in patients with urinary bladder cancer. J Natl Cancer Inst. 2009;101:114–9. [PubMed]
32. Zhang ZF, Sarkis AS, Cordon-Cardo C, Dalbagni G, Melamed J, Aprikian A, et al. Tobacco smoking, occupation, and p53 nuclear overexpression in early stage bladder cancer. Cancer Epidemiol Bio-markers Prev. 1994;3:19–24. [PubMed]
33. Curigliano G, Zhang YJ, Wang LY, Flamini G, Alcini A, Ratto C, et al. Immunohistochemical quantitation of 4-aminobiphenyl-DNA adducts and p53 nuclear overexpression in T1 bladder cancer of smokers and nonsmokers. Carcinogenesis. 1996;17:911–6. [PubMed]