Highlights Bladder Cancer AUA 2010

Shahrokh F. Shariat, Thomas F. Chromecki

* Weill Medical College of Cornell University, 525 East 68th Street, Starr 900, New York, NY 10065, phone (212) 746-556, email sfs2002@med.cornell.edu




At this year annual meeting in San Francisco, one of the main focuses was on bladder cancer preclinical studies, diagnostics and prognostic factors.


Key words: bladder cancer, AUA annual meeting San Francisco 2010, prognostic factors





Nakanishi et al. (#776) evaluated the impact of gene deletion of membrane-associated prostaglandin synthase-1 (mPGES-1), the terminal synthase in the generation of PGE2, in the BBN mouse model. Two month old male C57Bl/6 mice with global homozygous deletion for Ptges (mPGES-1 KO) and wild-type (WT) were treated with and without BBN in drinking water (n=5/grp) for 23 weeks. Bladders were harvested, weighed, fixed, step sectioned and evaluated by H&E for pathologic stage and grade and percent involvement of tissue by tumor. Urine was collected prior to sacrifice and evaluated for PGE-2 metabolites (PGEM) by ELISA. WT bladders were significantly heavier than KO bladders (mean 0.18 vs. 0.04 g respectively; p=0.03). Four of 5 WT mice had pT3 and one pT2 disease. Three of 5 KO mice had pT1 and 2 had pT2 disease. The tissue involvement was significantly less in the KO animals as well (54 vs. 87%, p=0.03). Pathologic stage was dependent. Urine metabolites were reduced 50% in the KO compared to WT animals (mean 18.9 vs. 45.2 pg/ml respectively; p<0.01). Genetic deletion of mPGES-1 resulted in significant decreases in PGE-2 as evidenced by lower urinary PGEM and in pathologic stage and percent tissue involvement by tumor. Stage distribution was dependent on genotype. Increased bladder weights in the WT mice compared to KO mice were directly related to the increased tumor burden. Taken together, these results suggest that inhibitors of mPGES-1 may prove to be an alternative to COX inhibition and possibly prove useful in the management of high-grade bladder cancer.

Oncolytic herpes simplex viruses replicate selectively in cancer cells and are promising therapeutic agents for cancer. G47delta, a triple-mutated herpes simplex virus type 1 (HSV-1) with deletions in the gamma34.5 gene and the alpha47 gene and with inactivation of the ICP6 gene, exhibits high replication capabilities in tumor cells while preserving the safety. Fukuhara et al. (#778) developed an armed oncolytic HSV-1 generation system that allowed insertion of desired transgenes into the ICP6 locus of the G47delta backbone. Using this system, they generated T-mfIL12 that expresses fusion-type murine interleukin 12 (IL-12) and T-01, a control vector. Human bladder cancer lines RT4, T-24, SW800, EJ-1 and JON and murine bladder cancer lines MB49 and MBT-2 were used for in vitro evaluation. In C57BL/6 mice with subcutaneous murine MB49 tumors, mock, T-01 or T-mfIL12 was inoculated intraneoplastically and tumor size measured. In a lung metastases model, C57BL/6 mice received intravenous injections of MB49 cells, and viruses were repeatedly administered into the tail veins. In all cell lines studied, T-01 and T-mfIL12 showed comparable cytopathic activities. In MB49 subcutaneous tumor model, both T-01 and T-mfIL12 caused significant inhibition of tumor growth compared with mock. Furthermore, T-mfIL12 was significantly more efficacious than T-01. In MB49 lung metastases model, intravenous administration of T-01 or T-mfIL12 significantly prolonged the survival compared with mock treatment. T-mfIL12 was more efficacious than T-01, especially at lower doses. The authors conclude that oncolytic HSV-1 therapy can be a potent therapeutic approach for bladder cancer. Especially for lung metastases, intravenous administration may be a useful delivery route of oncolytic HSV-1. Moreover, antitumor efficacy of oncolytic HSV-1 can be augmented by expression of immunostimulatory transgenes such as IL-12.

Verma et al. (# 1153) investigated the effect of Mytomycin C (MMC) on levels of vascular endothelial growth factor and its receptors in bladder cancer cells and in bladders of rats intravesically instilled with MMC. They hypothesized that MMC increases the angiogenic potential of both cancer and normal cells. One day after MMC treatment (6-100 μg/ml), they measured proliferation, and VEGF-A, VEGFR-1 and VEGFR-2 levels in T-24 and RT4 bladder cancer cells. The effect of pretreatment of VEGF siRNA on MMC induced decreases in proliferation was measured. Rats were intravesically instilled with saline or MMC (200 μg/rat). Urinary VEGF levels and bladder levels of VEGFR-2 protein relative to actin and VEGF, VEGFR-1, and VEGFR-2 mRNA relative to GAPDH mRNA were measured. Although MMC treatment inhibited cell proliferation, it did not decrease VEGF protein expression in T-24 and RT4 cells. In fact, in T-24 cells, MMC (6-50 μg/ml) increased VEGF mRNA expression normalized to GAPDH 3-13 fold and increased VEGFR-2 mRNA and protein expression, 2-12 fold and 2-5 fold, respectively. Levels of VEGFR-1 were unchanged. In T-24 cells, MMC (25 μg/ml) reduced proliferation 40 ± 5%, while increasing VEGF mRNA levels 3.4 fold. However, pretreatment of MMC treated cells with VEGF siRNA blocked VEGF mRNA production and potentiated MMC induced reductions in proliferation by an additional 27 ± 4%. Similar results were seen with RT-4 cells. In rats intravesically instilled with MMC, urinary VEGF/creatinine was increased 80% and VEGFR-2 mRNA and protein levels were increased 80% and 150%, respectively, compared to levels in saline instilled rats (n = 4-6). MMC treatment increased levels of VEGF mRNA and VEGFR-2 mRNA and protein in bladder cancer cells. VEGF siRNA potentiated mitomycin inhibitory effects on proliferation in bladder cancer cells. Intravesical instillation of MMC increases urinary VEGF and bladder VEGFR-2 levels. Taken together, these results suggest that MMC increases the angiogenic potential of both cancer and normal cells. Combining MMC with agents that reduce VEGF levels may be of value in the treatment of urothelial carcinoma.

Gust et al. (# 774) investigated the Notch pathway which plays a critical role in growth and differentiation. Thirteen bladder cancer cell lines were screened for Notch-1 and Notch-2 expression by Western blot and real-time PCR. Immunohistochemistry was performed in 152 bladder cancer patients. Alterations in gene expression after Notch-2 silencing with siRNA were measured by expression microarray in 3 cell lines. Wound healing assays were performed after treatment with Notch-2 siRNA versus scrambled siRNA. Epithelial cell lines with low invasive potential expressed Notch-1 at high and Notch-2 at low relative levels. Highly invasive, mesenchymal cell lines expressed high levels of Notch-2 and low levels of Notch-1. Microarray analysis revealed altered expression of genes related to epithelial to mesenchymal transition, protein and endosome trafficking, and cell-cell connections, when Notch-2 was silenced with siRNA. Notch-2 silencing decreased the migration of UC3 cells in the wound-healing assay. These data suggest that Notch signaling is a promising target for therapy in bladder cancer.


Yoon et al. (# 971) explored synergistic anti-tumor effect of cisplatin and novel dual PI3K/mTOR inhibitor NVP-BEZ235 against T24R2 cell and related death receptor signaling. The cisplatin resistant invasive human bladder cancer cell line T24R2 was treated with various dose-combinations of cisplatin (5-15ug/ml) and NVP-BEZ235 (0.3-20uM) and cell viability was assessed with CCK-8 assay. C-FLIP, caspase 3, 8, 9, p-Akt and t-Akt expression was evaluated with immunoblotting and flow cytometric cell cycle analysis was performed. Suppression of death receptor signaling in T24R2 was evaluated by dose-response study with TRAIL. T24R2 showed significantly higher c-FLIP expression compared to other chemo-naive human bladder cancer cell lines (HTB5, HTB9, J82, SW1710, UMUC14, 253J, T24) and unlike other cell lines, c-FLIP expression in T24R2 was not suppressed by cisplatin only. Treatment with TRAIL markedly suppressed T24 proliferation while there was virtually no change in T24R2 proliferation by TRAIL treatment. NVP-BEZ235 single treatment did not cause any changes in c-FLIP expression in T24R2. On the contrary, combined treatment with NVP-BEZ235 and cisplatin completely abrogated c-FLIP expression in T24R2 and potentiated anti-tumor effect of each drug up to 80%. Combined treatment also activated caspase 3, 8, and 9 in T24R2 and suppressed p-Akt / t-Akt, resulting in significant increase in the sub-G1 apoptotic cell population. These results show that cisplatin resistance in T24R2 cell is closely related with c-FLIP expression and suppression of death receptor signaling, which can be reversed by combined treatment with NVP-BEZ235. Further research is necessary to test NVP-BEZ235 as a new combinational-targeted therapeutic agent to overcome cisplatin resistance in human bladder cancer cells.



Ploeg et al. (#1719) evaluated the impact of time from definitive therapy to recurrence with clinical outcomes. 1,409 patients from the Netherlands with muscle-invasive bladder cancer were included in this study. All patients underwent either radical cystectomy or radiation therapy with curative intent from 1989-2005. Of the 1,409 patients, 330 (23%) experienced disease recurrence. The median survival for those who had a recurrence was 4 months. 1 and 3 year survival after a subsequent recurrence was 22% and 9% respectively. On multivariable analysis, location of and treatment for recurrence were independent predictors of survival in patients who underwent cystectomy. Age at diagnosis of disease recurrence, location, and treatment for recurrence were independent predictors of survival for patients who underwent radiation therapy. The authors conclude that time to recurrence had no prognostic value in this study. Time from initial therapy to recurrence does not appear to impact survival of cystectomy patients. Patients who were treated after disease recurrence, regardless of initial treatment choice, did better than those who were not treated. While this could be a selection bias for lower tumor burden and healthier patients, there is a signal that even recurrences are heterogeneous in their biological and clinical behaviour. Larger datasets with more homogenous populations would allow to refine the management of patients with bladder cancer after definitive treatment.




In a study of 315 patients treated with radical cystectomy, Youssef et al. (#1155) found a significant difference in tumor stage, grade, and architecture and COX-2 alterations between patients with bilharzial related bladder cancer and those with non-bilharzial related bladder cancer (p<0.05). Bilharzial related bladder cancer presented with lower grade, higher stage, and non-papillary non-urothelial carcinoma. COX-2 overexpression was associated with pathological T stage, grade and lymphovascular invasion. COX-2 expression was an independent predictor of disease recurrence and cancer specific mortality only in bilharzial related bladder cancer but not in non-bilharzial related bladder cancer. Taken together, these findings suggest that COX-2 may be a good prognostic marker especially in inflammatory cancers like in bilharzial related bladder cancer. Further research should focus on the evaluation of COX-2 as well as COX-2-targeted prevention or therapies in bilharzial related bladder cancer.

Kauffman et al. (# 162) evaluated bladder cancer cells for BCG receptors to investigate why BCG can kill bladder cancer cells in the absence of immune cells. Traditionally, BCG is presumed to treat bladder cancer cells via stimulation of the immune system. However, there are toll-like receptors (TLR) 2 and 4 that are known receptors for BCG. These receptors are believed to be restricted to immune cells. The authors evaluated bladder cancer cells for the presence of TLR using a tissue microarray constructed from 41 patients subsequently treated with 6-course BCG. TLR2 and 4 was expressed in 7 bladder cancer cell lines with RTPCR techniques. Of the 41 patients, 20 were BCG-responsive and 21 were BCG-nonresponsive. TLR2 and TLR4 proteins both stained positive in most (>70%) bladder cancer specimens, with no significant correlation to cancer stage or grade. For TLR2, the mean %-tissue positivity score was significantly higher in BCG responsive tumors relative to BCG non-responsive tumors (1.8 vs. 1.0, p=0.027). Furthermore, TLR2 was expressed at some level in all BCG responsive tumor specimens compared to fewer than half (10/21) of BCG non-responsive tumors (p<0.001). Similarly, for TLR4, absent expression was nearly twice as common in the BCG-NR group, which had a lower mean %-tissue positivity score (1.0 vs. 1.3) and lower mean staining intensity score (0.65 vs. 0.80) than the BCG-R group, however these findings were not statistically significant (p=0.31, 0.30, respectively). Consistent with the immunohistochemistry data, TLR2 and TLR4 RNA transcripts were each detected in the majority of bladder cancer cell lines. The authors suggest that this study demonstrates that immune-cell BCG receptors, TLR2 and TLR4, are present in bladder cancer cells and may be potential markers for BCG sensitivity. This is an exciting avenue for further research.

Matsumura et al. (#161) studied the expression of the molecular transporter human equilibrative nucleoside transporter (hENT) that is necessary for gemcitabine transportation into the cells of tumors. The authors hypothesized that the expression level of hENT1 can predict survival in advanced bladder cancer treated by gemcitabine-based chemotherapy. 40 patients with advanced bladder cancer treated by GC (gemcitabine+cisplatin) regimen (17 patients) and GCT (GC+paclitaxel) regimen (23 patients) were evaluated. They studied the expression level of hENT1 by immunohistochemical staining of formalin-fixed, paraffin-embedded bladder cancer tissues. Of the 40 tissue samples, 20 samples revealed hENT1 high and low expression. Patients with hENT1 low expression had a median overall survival of 11.6 months from initiation of gemcitabine based chemotherapy. In contrast, patients with hENT1 high expression had a median survival of 17.3 months. Kaplan-Meier analysis revealed that patients who had low hENT1, were at significantly higher risk of death compared to patients with high hENT1 expression (p=0.0026). Moreover, in a multivariable analysis that adjusted for effects of extent of metastases only hENT1 expression was an independent predictor (p=0.003) of survival. The authors concluded that patients with high levels of hENT1 in their bladder cancer have significantly better overall survival than those with low expression. Immunohistochemical evaluation of hENT1 expression can be a promising biomarker for the prediction of clinical response and the prognosis of patients with advanced bladder cancer treated by gemcitabine-based chemotherapy. Validation in a larger cohort is necessary and establishment of set criteria for low as well as high expression is necessary.

Angiogenesis is critical to tumor growth by providing growth factor milieu including nutrients, oxygen, and extracellular structure to the tumor. Vascular endothelial growth factor (VEGF), fibroblast growth factor (BFGF), thrombospondin 1 (TSP-1) levels and microvessel density (MVD-CD31) have prognostic value in urothelial cancer in small studies. Shariat et al. (#1032) evaluated 204 radical cystectomy specimens with urothelial cancer for expression of the above named angiogenesis molecular markers and correlated this expression with cell cycle- (Cyclin E1, Cyclin D1, p53, p21, p27, pRB), proliferation- (Ki-67), and apoptosis-related (caspase-3, survivin and Bcl-2) molecular markers. All four angiogenesis markers were associated with established clinicopathologic features of aggressive bladder urothelial carcinoma such as stage, lymphovascular invasion, and lymph node metastasis and other molecular markers. After adjusting for the effects stage and grade of the tumor, BFGF and TSP-I were independent predictors of disease recurrence (HR 3.6, p=0.002 and HR=2.2, p=0.001, respectively) and cancer specific mortality (HR 2.8, p=0.02 and HR=2.3, p=0.003, respectively). When all four angiogenesis markers were included in the multivariable model, BFGF and TSP-1 retained their independent association with disease recurrence (HR=2.9, p=0.014 and HR=1.8, p=0.022, respectively) but only TSP-1 was independently associated with cancer specific mortality (HR=1.9, p=0.031). The authors concluded that angiogenesis molecular markers are altered in urothelial carcinoma and are associated with outcomes of radical cystectomy. In this initial study VEGF was commonly altered in UCB specimens, thereby serving as a possible target for therapy. Down-regulation of TSP-1 and up-regulation of BFGF are independent predictors of bladder cancer recurrence, and down-regulation of TSP-1 is also associated with bladder cancer specific mortality. The integrated understanding of the biology of angiogenesis related molecules in bladder cancer and its translational relevance may lead to the development of novel treatment approaches for this disease.

Schultz et al. (# 160) evaluated tissue microarrays of urothelial cancers for alterations in the mTOR pathway to identify prognostic markers. The mTOR pathway is critical in cell proliferation, angiogenesis, and cell migration. The tissue microarray included 144 radical cystectomies (14 pTis/pTa, 16 pT1, and 114 pT2-4 cases). On univariate analysis, PTEN, c-MYC, p27, phos Akt, phos S6 had lower expression in invasive urothelial carcinoma compared to benign urothelium. C-MYC, p27, and phos Akt expression significantly correlated with divergent aggressive cellular differentiation (sarcomatoid, squamous, and adenocarcinoma) and presence of invasion. Phos S6 was predictive of overall survival (p=0.01), disease-free survival (p=0.008) and progression (p=0.05) with higher expression levels predicting favorable outcome. On multivariable analysis phos S6 remained an independent predictor of disease-free survival (p=0.006) after adjusting for the effects of clinical stage, pathologic stage, invasive tissue microarray spot histology and divergent aggressive differentiation. The authors conclude that there is an overall downregulation of mTOR pathway members in advanced bladder cancers. Phos S6 was an independent predictor of disease-free survival on univariate and multivariate analysis. Further validation of these findings is necessary in larger advanced cohorts.

Understaging of bladder cancer occurs in 50% of patients resulting in non-appropriate care. Additionally, understanding the biological aggressiveness of the cancer is critical to counsel patients regarding the need for systemic therapy. Several investigators have evaluated biomarkers that individually have minimal additive benefit in predicting the biological potential of a cancer, but collectively may improve the predictive accuracy. Youssef et al. (#1021) prospectively assessed the predictive value of a panel of biomarkers (cyclin E1, p53, p21, p27 and pRB/Ki-67) for staging of patients with urothelial carcinoma of the bladder (UCB). 174 consecutive patients treated with transurethral resection (TUR) and/or radical cystectomy (RC) for UCB were prospectively included in the study starting in January 2007. A prognostic score (PS; favorable = 2 or less altered biomarkers; unfavorable = more than 2 altered biomarkers) was defined and correlated with clinical and pathological data. The study comprised of 102 RC patients and 72 TUR cases. Unfavorable PS was noted in 36.3% of RC cases and 16.6% of TUR cases. In RC patients, an unfavorable PS was significantly associated with advanced tumor stage (p=0.02), the presence of lymphovascular invasion (p=0.017) and disease recurrence (p=0.04). In the subgroup of patients who had biomarker evaluation at TUR and then underwent RC (n=32), 75% and 50% of the cases with unfavorable PS showed upstaging and LN involvement at RC versus 29% and 12.5% in cases with favorable PS. Unfavorable PS at TUR was associated with upstaging (p=0.022), lymphovascular invasion (p=0.001), LN metastasis (p=0.026) and recurrence (p=0.014). The preliminary analysis of this prospective trial strongly suggests that a panel of five biomarkers not only predicts poor outcome after RC but also improves the identification of patients at risk of upstaging at RC. An unfavorable prognostic score may identify patients who are most likely to benefit from neo-adjuvant and adjuvant chemotherapy. These findings need multicenter validation before wide application.

Lerner et al. (# 1715) A prospective multi-institutional clinical trial was conducted to test the hypothesis that p53 is prognostic and predictive in patients with pathologically organ-confined bladder cancer following radical cystectomy. The secondary objective was to evaluate cell cycle regulatory proteins p21, Rb, p16, p27 and their role in clinical outcomes. Patients with pT1-2N0M0 bladder carcinoma following radical cystectomy and bilateral pelvic lymph node dissection were eligible. Entry criteria required ≥ 15 dissected nodes or a post-operative abdominopelvic CT for patients with < 15 nodes. Immunohistochemistry was performed on representative TUR or cystectomy tumor specimens. As adjuvant chemotherapy did not impact recurrence free survival or overall survival, all patients were included regardless of adjuvant therapy. 499 eligible patients from 43 sites in US, Canada and Europe were registered between 8/97 and 1/06. Lymphovascular invasion was associated with decreased recurrence free survival and overall survival, and patient age was associated with overall survival. No single biomarker was associated with recurrence free survival or overall survival. However, the combined variables p53/p21, p53/p21/Rb, or 5 markers in combination were not associated with recurrence free survival or overall survival. These findings are in contradiction to a large body of evidence regarding the predictive/prognostic value of these markers.

Rhijn et al. (#1714) evaluated  tumor sub-stage, clinicopathologic parameters, and 4 molecular markers with regard to clinical outcomes in patients with pT1 urothelial carcinoma treated with BCG. Tumors were sub-staged into micro-invasive and extensive-invasive and also by invasion of the muscularis mucosae. Multivariable analysis was carried out for recurrence and progression with clinical, pathologic and molecular markers (FGFR3 mutation, MIB-1, p53, p27) as variables. European Organisation for Research and Treatment of Cancer (EORTC) risk-scores were also calculated. Actual progression was observed in 20% of patients. In multivariable analysis for recurrence, tumor multiplicity was the only significant variable (HR: 2.0, 95% CI: 1.4-3.0, p<0.001). In multivariable analysis for progression, female gender (HR: 2.7, 95% CI 1.3-5.8, p=0.014), sub-stage (HR: 2.8; 95%CI 1.5-5.7, p=0.003) and CIS (HR: 2.1, 95% CI: 1.2-4.0, p=0.015) were all significant predictors. Molecular markers did not appear to add significant benefit for prediction of recurrence and progression in this study. Additionally, the value of the EORTC risk score was also limited. The authors conclude that clinico-pathologic features at this point remain the most robust predictors of recurrence and progression in this heterogenous patient population.




Alkhateeb et al. (#160) As men and women age, their bladder wall thicknesses markedly change. Men develop a thicker bladder due to prostatic obstruction and women develop a thinner bladder, particularly after menopause. Proper sampling of detrusor muscle is essential for accurate pathological diagnosis. There is a general belief among urologists that females are at a higher risk for absence of detrusor muscle due to a perceived danger of increased risk of bladder perforation. The authors evaluated patients undergoing a TURBT between 1990 and 2009 for the presence of muscle in the biopsy specimen. A total of 921 patients with primary non muscle-invasive bladder cancer were included. Male to female ratio was 3.2:1, mean age was 65.9 years, pathological stage distribution was pTa in 55.7%, ≥ pT1 in 41.8% and pTis in 2.5% and detrusor muscle was present in 515 patients (63.2%). In a multivariable model, independent predictors for the presence of detrusor muscle were age ≥65 years (OR 1.48, 95% CI 1.10-1.99, p=0.02), pT1 tumors (OR 12.34, 95% CI 4.31-35.2, p=0.001), grade 3 tumors (OR 1.68, 95% CI 0.39-0.89, p=0.01), tumors ≥ 3cm in size (OR 3.2, 2.18-4.68, p=0.001), multifocality (OR 1.60, 95% CI 1.18-2.18, p=0.002) and concomitant CIS (OR 1.83, 95% CI 1.12-2.98, p=0.01). Gender did not predict for the presence of detrusor muscle (OR 1.21, 95% CI 0.86-1.71, p=0.26). The authors conclude that females are not at higher risk for absence of detrusor muscle in the first TURBT. However, younger patients and those with small and single tumors are at risk for not having detrusor muscle in the pathological specimen. Every effort should be made for proper sampling of detrusor muscle in order to have an accurate pathological diagnosis in order to administer risk-adjusted management.

For the complete article please refer to the pdf file.


Write your comment:

BCG Bastian T1G3
Original articles
Original articles have a 3,000 word limit. Review articles have a 5,000 word limit (excluding references). Articles must be divided into the sections Background, Methods, Results, Discussion and Concl
Case reports
Case reports have a 1,500 word limit (excluding references) and must be sectioned into Background, Case Presentation and Discussion. The number of authors and references is not limited. Every article
Poster Journal
The article has a 1,200 word limit (excluding references) and must be sectioned into Introduction, Methods, Results and Conclusions. Number of authors (10) and references (15) is limited. Every articl
Conference reports
Articles are written by invitation of the Editors-in-chief. Manuscript length and format are not limited. Every article must be preceded by an Abstract. At the authorís request, reports about German-s
Presentations of new diagnostic or therapeutic approaches and commentaries of experts on scientific research on bladder cancer by invitation of the Editors-in-chief. Articles must be sectioned into Ba