Decreased DOC-2/DAB2 Expression in Urothelial Carcinoma of the Bladder Josea

Imaging, Diagnosis, Prognosis

Decreased DOC-2/DAB2 Expression in Urothelial Carcinoma of the Bladder JoseA. Karam,1Shahrokh F. Shariat,1Hong-Ying Huang,2 Rey-Chen Pong,1Raheela Ashfaq,3 Ellen Shapiro,2 Yair Lotan,1Arthur I. Sagalowsky,1Xue-Ru Wu,2,4 andJer-Tsong Hsieh1

Abstract Purpose: DOC-2/DAB2 (differentially expressed in ovarian carcinoma-2/disabled-2), a potential tumor suppressor gene, is underexpressed in several cancers. Little is known about the expression of this gene in urothelial carcinoma of the bladder (UCB).We profiled DOC-2/DAB2 expressioninmouse andhumannormalandneoplasticurothelia. Experimental Design: Immunohistochemical staining for DOC-2/DAB2 was carried out on tissuespecimensfromtwotransgenicmousemodelswithurothelium-specificmolecularalterations and on a tissue microarray containing cores from 9 normal controls, 44 patients who underwent transurethral resection of the bladder tumor (TURBT),195 patients who underwent radical cystectomy for UCB, and 39 lymphnodes withmetastatic UCB. Results: Normal mouse urothelium stained uniformly with DOC-2/DAB2. Weaker staining was observed in low-grade, superficial papillary bladder tumors from transgenic mice harboring constitutively active Ha-Ras, whereas carcinoma in situ ^ like lesions and high-grade bladder tumors from transgenic mice expressing a SV40 T antigen completely lacked DOC-2/DAB2 expression. In human tissues, DOC-2/DAB2 expression was decreased in 11% of normal bladder specimens, 59% of TURBT specimens, 65% of radical cystectomy specimens, and 77% of the metastatic lymph node specimens. Decreased DOC-2/DAB2 expression was associated with advanced pathologic stage (P = 0.023), lymph node metastases (P = 0.050), and lymphovascular invasion (P < 0.001). In univariable, but not in multivariable analysis, decreased DOC-2/ DAB2 was associated with an increased probability of bladder cancer recurrence (log-rank test, P = 0.020) and bladder cancer ^ specific mortality (log-rank test, P =0.023). Conclusions: Decreased DOC-2/DAB2 expression seems to occur early in bladder tumorigenesis and becomes more prominent in advanced stages of UCB.

The American Cancer Society estimated that bladder cancer bladder tumorigenesis (2). Alterations of chromosome 9 are will affect f67,000 persons, resulting in f13,000 deaths in thought to predispose cells to acquire more advanced chro- the United States in 2007 (1). Multiple chromosomal mosomal and genetic abnormalities, thus facilitating advanced abnormalities have been implicated in bladder tumorigenesis. stages of tumorigenesis (3). Alterations in chromosome 5 have Chromosome 9 abnormalities have been the most studied been recognized in bladder cancer specimens (4, 5), and more entity because these are typically present in the early stages of recent studies narrowed the abnormal area of interest to the 5p13 region (6–8). DOC-2/DAB2 (differentially expressed in ovarian carcinoma- 2/disabled-2) is a candidate tumor suppressor gene originally Authors’ Affiliations: Departments of 1Urology and 2Pathology, University of Texas Southwestern Medical Center at Dallas, Dallas, Texas; and 3Department identified via a search for differentially down-regulated tran- of Urology, New York University School of Medicine, 4Veterans Affairs Medical scripts in human ovarian carcinoma (9, 10). The DOC-2/DAB2 Center in Manhattan, NewYork, NewYork gene has been cloned and localized to chromosome 5p13 (11). Received 2/5/07; revised 5/29/07; accepted 6/7/07. Down-regulation of DOC-2/DAB2 has been reported in Grant support: J.A. Karam is supported in part by NIH K30-RR022269-08 multiple cancers, including prostate (12), bladder (13), breast (Clinical Research Curriculum Award: Training Curriculum in Patient-Oriented Research). X.R. Wu is supported by NIH (DK56903 and DK69688) and the (14), esophageal (15), pancreas (16), colon (17), and gesta- Veterans Administration (Merit Review Award). J.T. Hsieh is supported by NIH tional trophoblastic disease (18). (CA95730) and U.S. Army grant (DAMD17-03-2-0033). To our knowledge, DOC-2/DAB2 protein expression has only The costs of publication of this article were defrayed in part by the payment of page been reported in bladder cancer cell lines in vitro, but not in advertisement charges. This article must therefore be hereby marked in accordance animal or human bladder cancer specimens. The aims of the with 18 U.S.C. Section 1734 solely to indicate this fact. Note: Supplementary data for this article are available at Clinical Cancer Research present study are 2-fold: the first is to profile DOC-2/DAB2 Online (http://clincancerres.aacrjournals.org/). expression in several transgenic mouse models of bladder Requests for reprints: Jer-Tsong Hsieh, Department of Urology, University of cancer; the second is to determine the association of DOC-2/ Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX DAB2expression with the presence of human urothelial 75390-9110. Phone: 214-648-3988; Fax: 214-648-8786; E-mail: JT.Hsieh@ UTSouthwestern.edu. carcinoma of the bladder (UCB), clinical and pathologic F 2007 American Association for Cancer Research. characteristics, metastases, and clinical outcomes in UCB doi:10.1158/1078-0432.CCR-07-0287 patients with different disease stages. Toward the latter goal,

Clin Cancer Res 2007;13(15) August 1,2007 4400 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2007 American Association for Cancer Research. Decreased DOC-2/DAB2 Expression in Urothelial Carcinoma

Fig. 1. DOC-2/DAB2 expression in (A) normal urothelium of wild-type mice; (B) hyperplastic urothelium in transgenic mice expressing a constitutively active Ha-ras oncogene; low-grade, superficial papillary bladder tumors of (C) heterozygous and (D) homozygous ras-transgenic mice; (E)carcinomain situ ^ like lesions and (F) high-grade bladder tumors derived from transgenic mice expressing a SV40 Tantigen. Urothelium-specific transgenic mice were developed using a uroplakin II ^ controlled promoter.

www.aacrjournals.org 4401 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2007 American Association for Cancer Research. Imaging, Diagnosis, Prognosis

purpose, bladder tissues were procured from wild-type as well as transgenic mice specifically expressing a constitutively active Ha-ras oncogene (19) or a SV40 large T antigen (20) in the urothelium under the control of a uroplakin II promoter. Patient population. All human studies were undertaken with the approval and institutional oversight of the Institutional Review Board at the University of Texas Southwestern Medical Center, Dallas, Texas. The study comprised three patient cohorts: 9 patients who underwent cystectomy for causes other than malignancy without evidence of any genitourinary malignancy; 44 patients who underwent TURBT for treatment of Ta, Tis, and/or T1 UCB; and 206 patients who underwent radical cystectomy with bilateral lymphadenectomy. In the cohort of 44 patients that underwent TURBT, 21 patients (48%) underwent previous TURBTs. Twenty-one patients (48%) were treated with intravesical therapy such as Bacillus Calmette-Guerin and mitomycin C. None of these 44 patients underwent subsequent radical cystectomy at the time of last follow-up. All 383 consecutive patients admitted at our institution for treatment of UCB with radical cystectomy and bilateral lymphadenectomy during the period from 1987 to 2002 were potential candidates for these analyses. Our analysis was limited to patients who had UCB. Adequately DOC-2/DAB2–stained archival tissue was available for 206 of the 383 cystectomy patients. Histology, grade, stage, and presence of carcinoma in situ were confirmed by blinded review of the original pathology slides. The indications for radical cystectomy were tumor invasion into the muscularis propria or prostatic stroma or Ta, T1, or Tis (carcinoma in situ) refractory to TUR with intravesical chemotherapy and/or immunotherapy. No patient had distant metastatic disease at the time of cystectomy. For each patient, comprehensive clinical and pathologic data elements were collected and entered into an Institutional Review Board–approved database. Multiple data reviews and quality checks were done to assure the accuracy and completeness of data elements. Eleven patients that received neoadjuvant chemotherapy were excluded, leaving 195 patients diagnosed with UCB for final analysis. Median age at radical cystectomy was 66.5 years (range, 38.9-86.0). Tissue microarray. The tissue microarrays comprised (a) noncancer tissue from nine patients who were treated with cystectomy for other causes than malignancy without evidence of any cancer; (b) tumor tissue from the primary, index cancer from 44 patients who were treated with TURBT; (c) tumor tissue from the primary, index cancer from 206 patients who were treated with radical cystectomy; and (d) lymph node tissue involved with cancer from 43 of the 206 patients who were treated with radical cystectomy.

Table 1. DOC-2/DAB2 expression in patients with Fig. 2. DOC-2/DAB2 expression in (A) normal bladder tissue; (B)UCB;(C) and without UCB cancerous lymphoid tissue. Number DOC-2/DAB2 of patients expression we assessed the expression of DOC-2/DAB2 in (a) normal bladder urothelial tissue; (b) UCB tissue from patients who Normal, Decreased, n (%) n (%) underwent transurethral resection of the bladder tumor (TURBT) for Ta, T1, and/or Tis disease; (c) UCB tissue from Patients without bladder UCB who 9 8 (88.9) 1 (11.1) patients who underwent radical cystectomy and pelvic lympha- underwent simple cystectomy d Patients treated with TURBT 44 18 (40.9) 26 (59.1) denectomy for advanced disease; and ( ) UCB metastases in Patients treated with regional lymph nodes from patients treated with radical radical cystectomy cystectomy and pelvic lymphadenectomy. Primary, index UCB* 195 68 (34.9) 127 (65.1) UCB metastases to 39 9 (23.1) 30 (76.9) lymph nodesc Materials and Methods *After exclusion of 11 patients that received neoadjuvant chemotherapy. Mouse tissues. DOC-2/DAB2 expression in normal urothelium and cAfter exclusion of six patients that received neoadjuvant bladder tumors obtained from several transgenic mouse models was chemotherapy. initially assessed immunohistochemically as described below. For this

Clin Cancer Res 2007;13(15) August 1,2007 4402 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2007 American Association for Cancer Research. Decreased DOC-2/DAB2 Expression in Urothelial Carcinoma

Table 2. Association of DOC-2/DAB2 expression with clinicopathologic characteristics of 195 consecutive patients treated with radical cystectomy and bilateral lymphadenectomy for UCB

Number of patients (%) DOC-2/DAB2 expression Normal Decreased P

Total (%) 195 (100) 68 (34.9) 127 (65.1) Age (y), median (range) 66.5 (35.2-86.0) 63.3 (38.8-80.5) 68.9 (35.2-86.0) 0.001* Gender (%) Female 38 (19.5) 9 (23.7) 29 (76.3) Male 157 (80.5) 59 (37.6) 98 (62.4) 0.130c Final pathologic tumor stage (%) Ta, Tis, T1 38 (19.5) 15 (39.5) 23 (60.5) T2 58 (29.7) 28 (48.3) 30 (51.7) T3 69 (35.4) 19 (27.5) 50 (72.5) T4 30 (15.4) 6 (20.0) 24 (80.0) 0.023c Final pathologic grade (%) Grade 1 or 2 15 (7.7) 6 (40.0) 9 (60.0) Grade 3 180 (92.3) 62 (34.4) 118 (65.6) 0.779c Carcinoma in situ (%)b Negative 110 (56.7) 39 (35.5) 71 (64.5) Positive 84 (43.3) 28 (33.3) 56 (66.7) 0.879c Lymph node status (%)b N0 132 (68.0) 52 (39.4) 80 (60.6) N1, N2 62 (32.0) 15 (24.2) 47 (75.8) 0.050c Lymphovascular invasion (%)b Negative 101 (52.1) 46 (45.5) 55 (54.5) Positive 93 (47.9) 21 (22.6) 72 (77.4) 0.001c

*Mann Whitney U test. cm2 test. bLymph node status, lymphovascular status, and carcinoma in situ status were not available for one patient.

Tissue microarrays were built using a manual tissue arrayer (Beecher increments of cutoffs for DOC-2/DAB2 positivity with regard to bladder Instruments). The index tumor, defined as the largest and/or highest cancer characteristics and prognosis (data not shown). Patients with a tumor stage and grade, and areas of normal transitional cell not total score of 3 or less were at higher risk of bladder cancer recurrence adjacent to cancer cells were identified and circled on the H&E slides. and mortality compared with patients with a total score of more than Triplicate 0.6-mm cores were obtained from the circled areas of tumor 3. DOC-2/DAB2 immunoreactivity was therefore assigned to one of two and/or normal transitional cell and transferred to a recipient paraffin categories of staining in the tumor cells: decreased expression (total block. Multiple sausage internal controls were also placed with the score V3) or normal expression (total score >3). standard controls. Pathology. Staff pathologists with expertise in genitourinary Immunohistochemistry. Paraffin sections containing mouse as well pathology examined all specimens according to previously published as arrayed human bladder tissues were routinely deparaffinized, protocols. The 2002 tumor-node-metastasis classification was used hydrated, and treated with 3% hydrogen peroxide in methanol for for pathologic staging, and the 1973 WHO classification was used 15 min to quench the endogenous peroxidase activity. For antigen for pathologic grading. To ensure the validity of the data extraction, retrieval, the sections were microwave treated at maximum power for two clinicians read pathology reports of 162consecutive cystectomy 15 min in 0.01 mol/L citrate buffer (pH, 6.0). After an overnight patients, while blinded to patient clinical parameters and the finding incubation at 4jC with the anti–DOC-2/DAB2 antibody (1:100 of the other reviewer. Inter-reader reliability measured using the dilution of monoclonal DOC-2/p96 antibody from BD Transduction intraclass correlation coefficient was >0.95 for all pathologic Laboratories), a biotinylated goat anti-rabbit secondary antibody parameters. was applied. The slides were then incubated with avidin conjugated Follow-up. Patients generally were seen postoperatively at least with peroxidase and developed in a solution containing 3,3¶- every 3 to 4 months for the first year, semiannually for the second year, diaminobenzidine (DAB) and H2O2. All sections were then counter- and annually thereafter. When patients died, the cause of death was stained with hematoxylin, dehydrated, mounted, and observed by light determined by the treating physicians, by chart review corroborated by microscopy. death certificates, or by death certificates alone. Most patients who were The expression of DOC-2/DAB2 was scored by assigning a identified as having died of bladder cancer had progressive, widely proportion score and an intensity score according to Allred’s scoring disseminated, and often symptomatic metastases at the time of death. protocol (21). The proportion score was assigned, which represented Perioperative mortality (death within 30 days of surgery) was censored the estimated proportion of positive-staining cells (0, none; 1, <1/100; at time of death for bladder cancer–specific survival analyses. 2, 1/100 to 1/10; 3, 1/10 to 1/3; 4, 1/3; to 2/3; and 5, >2/3). The Statistical analysis. The m2 test was used to evaluate the association intensity score was assigned, which represented the average intensity of between DOC-2/DAB2 and clinicopathologic parameters. Differences positive cells (0, none; 1, weak; 2, intermediate; and 3, strong). The in variables with a continuous distribution across dichotomous proportion and intensity scores were combined to obtain a total score, categories were assessed using the Mann-Whitney U test. The Kaplan- which ranged from 0 to 8. All slides were scored independently by two Meier method was used to calculate survival functions, and differences investigators who were blinded to patient clinical information. In a were assessed with the log-rank statistic. Multivariable survival analyses preliminary study, we assessed the discriminative value of DOC-2/ were done using the Cox proportional hazards regression model. The DAB2as a continuous variable and as a categorical variable with serial lowest quartile was used as the reference category when calculating the

www.aacrjournals.org 4403 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2007 American Association for Cancer Research. Imaging, Diagnosis, Prognosis hazards ratios. Statistical significance in this study was set as P < 0.050. DAB2expression in different phenotypic variants of bladder All reported P values are two sided. All analyses were done with SPSS tumors and set a stage for human studies. version 13.0 (SPSS Inc.). Expression of DOC-2/DAB2 in human normal bladder urothelial cells, UCB, and cancerous lymph node tissues. DOC-2/DAB2 Results was predominantly expressed in the superficially located umbrella cells of normal urothelium. The staining was pri- Expression of DOC-2/DAB2 in normal and neoplastic urothe- marily cytoplasmic, with occasional weak nuclear labeling lium of transgenic mice. Preliminary evaluation of DOC-2/ (Fig. 2). Table 1 shows DOC-2/DAB2 expression in subjects DAB2expression in normal urothelium and bladder tumors with and without bladder urothelial carcinoma. DOC-2/DAB2 was carried out using transgenic mouse models specifically expression was decreased in 1:9 (11.1%) of noncancerous expressing various transgenes in the urothelium. Normal bladder specimens, 26:44 (59.1%) of the specimens from urothelium was labeled uniformly throughout basal, interme- patients who underwent exclusively TURBT, 127:195 (65.1%) diate, and superficial layers, with mostly cytoplasmic staining of the specimens from patients who underwent radical (Fig. 1A). Significantly weaker DOC-2/DAB2 staining was cystectomy, and 30:39 (76.9%) of the metastatic lymph node found in hyperplastic urothelium in transgenic mice expressing specimens. a constitutively active Ha-Ras oncogene (ras transgenic; Fig. 1B). Association of DOC-2/DAB2 expression with clinicopathologic Weaker membranous and cytoplasmic staining was observed in characteristics of patients who underwent TURBT for UCB. Median low-grade, superficial papillary bladder tumors of heterozygous age at TURBT was 64.5 years (range, 41.1 to 89.3). Seven patients (Fig. 1C) and homozygous ras-transgenic hosts (Fig. 1D). In were of female gender (15.9%). Fifteen (34.1%) had Ta UCB, nine contrast, carcinoma in situ –like lesions and high-grade bladder (20.5%) had Tis UCB, and 20 (45.5%) had T1 UCB. Five (11.4%) tumors derived from transgenic mice expressing a SV40 T had grade 1 disease, 21 (47.7%) had grade 2 disease, and 18 antigen completely lacked DOC-2/DAB2 expression (Fig. 1E (40.9%) had grade 3 disease. None of the patients underwent and F). These data reveal a heterogeneous pattern of DOC-2/ radical cystectomy within a median follow-up of 40.4 months (range, 0.3 to 124.6). DOC-2/DAB2 expression was decreased in the primary index cancer in 26 of 44 (59.1%) of the patients who underwent TURBT for bladder Ta, T1, and/or Tis. Expression of DOC-2/ DAB2was not associated with gender, tumor grade, stage, previous TURBT, or previous intravesical therapy (P values z0.164). Association of DOC-2/DAB2 expression with clinicopathologic characteristics and clinical outcomes of patients who underwent radical cystectomy for UCB. DOC-2/DAB2 expression was decreased in the primary index cancer in 127 of 195 patients (65.1%) who were treated with radical cystectomy and bilateral lymphadenectomy. Pathologic characteristics of the 195 cystectomy patients and association with DOC-2/DAB2 expression are shown in Table 2. A total of 56 of the 195 patients (28.7%) underwent adjuvant chemotherapy, and 14:195 (7.1%) underwent adjuvant radiation therapy for positive lymph node status and/or extravesical extension. Exactly 23 of 38 patients (60.5%) with low-stage (Ta, Tis, and T1) UCB and 9 of 15 patients (60.0%) with low-grade (grades 1 and 2) UCB had decreased expression of DOC-2/DAB2. Decreased expression of DOC-2/DAB2 was associated with advanced pathologic stage (P = 0.023), metastases to lymph nodes (P = 0.050), and lymphovascular invasion (P = 0.001). Disease recurred in 87:195 patients (44.6%), and 89:195 (45.6%) patients were dead at the time of analysis. Of these 89 patients, 71 patients died of metastatic bladder cancer, and 18 patients died of other causes without evidence of disease progression. The mean follow-up was 49.3 months (median, 33.4; range, 1.3 to 183.4) for patients alive at the time of analysis. Kaplan-Meier plots showed that a decreased expression of DOC-2/DAB2 was associated with an increased probability of tumor recurrence (log-rank test, P = 0.020; Fig. 3A) and bladder cancer-specific mortality (log-rank test, P = 0.023; Fig. 3B). In multivariable Cox proportional hazards regression analyses (Table 3), DOC-2/DAB2 expression Fig. 3. Probabilities of (A) recurrence-free survival and (B) disease-specific survival in 195 patients who underwent radical cystectomy with bilateral was not independently associated with disease recurrence lymphadenectomy stratified according to DOC-2/DAB2 expression. (P = 0.383) or disease-specific survival (P = 0.607).

Clin Cancer Res 2007;13(15) August 1,2007 4404 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2007 American Association for Cancer Research. Decreased DOC-2/DAB2 Expression in Urothelial Carcinoma

Table 3. Multivariable Cox regression analyses of pathologic features, adjuvant chemotherapy, and DOC-2/ DAB2 expression for the prediction of disease recurrence and disease-specific survival of 195 patients treated with radical cystectomy and bilateral lymphadenectomy for UCB

Recurrence Disease-specific survival Hazards ratio 95% CI P Hazards ratio 95% CI P

Pathologic tumor grade* 1.13 0.31-4.02 0.848 2.62 0.33-20.72 0.361 Pathologic tumor stage Ta, Tis, T1 1.00 Reference — 1.00 Reference — T2 0.95 0.35-2.58 0.929 1.22 0.36-4.13 0.740 T3 2.05 0.78-5.41 0.145 3.04 0.92-10.05 0.068 T4 3.80 1.38-10.47 0.010 6.41 1.88-21.86 0.003 Trend 0.002 <0.001 Lymphovascular invasion 1.80 1.04-3.11 0.033 1.76 0.97-3.21 0.061 Lymph node metastases 1.89 1.14-3.14 0.014 1.71 0.98-2.97 0.055 Adjuvant chemotherapy 1.31 0.81-2.13 0.262 0.83 0.48-1.41 0.492 Decreased DOC-2/DAB2 expressionc 0.80 0.49-1.30 0.383 0.86 0.50-1.49 0.607

Abbreviation: 95% CI, 95% confidence interval. *Pathologic tumor grade was analyzed as dichotomous variable (grade 3 versus grade 1 and 2). cDOC-2/DAB2 expression was analyzed as a dichotomous variable (score>3 versus V3).

Discussion carcinoma, loss of DOC-2/DAB2 occurs as early as in hyperplasia (15). Taken together, these findings suggest a UCB is a heterogeneous disease that present in two major potential role for DOC-2/DAB2 in early tumorigenesis. phenotypes (22). About 80% of UCB present as low-grade DOC-2/DAB2 expression was more frequently decreased in papillary tumors, with limited potential to metastasize, whereas patients with higher pathologic stage, lymph node metastasis, 20% present with high-grade, invasive tumors, with up to 50% and lymphovascular invasion. DOC-2/DAB2 expression was of patients dying of metastases within 2years. Alterations in not decreased significantly in patients with higher grades, chromosome 5 have been recognized in bladder cancer possibly because only 7.7% of the patients had grades 1 and 2 specimens (4, 5), and more recent studies narrowed the disease, which may have limited the power of this comparison. abnormal area of interest to the 5p13 region (6–8). As DOC- On multivariable analysis that adjusted for standard clinico- 2/DAB2 gene is located on chromosome 5p13 (11), we decided pathologic parameters, only pathologic stage and lymph node to evaluate its association with bladder tumorigenesis in metastases, but not decreased DOC-2/DAB2, were independent transgenic mice and human UCB specimens. predictors of disease recurrence and disease-specific survival. DOC-2/DAB2 expression was strong in normal urothelium Because decreased DOC-2/DAB2 expression occurs early in of wild-type mice, but decreased in urothelial hyperplasia and bladder tumorigenesis, it is possible that this marker is not low-grade papillary urothelial carcinoma in urothelium-specific useful to independently predict outcomes such as recurrence, Ras-transgenic mice. Moreover, DOC-2/DAB2 expression was bladder cancer–specific survival, and overall survival. absent in high-grade and invasive urothelial carcinoma in DOC-2/DAB2 is a phosphoprotein involved in signal urothelium-specific SV40 T antigen-transgenic mice. This transduction, containing multiple SH3 binding motifs at the suggests that the loss of DOC-2/DAB2 expression is an early COOH terminus (10). These motifs are conserved in the mouse event in bladder cancer tumorigenesis, as well as an important homologue of DOC-2/DAB2, p96/Dab2, suggesting important factor in bladder cancer progression in a murine model. In biological functions associated with these motifs (25, 26). The humans, we also found that only 11.1% of patients without SH3 binding motifs of the mouse p96/Dab2protein interact UCB had decreased DOC-2/DAB2 expression. Conversely, with the SH3 domains of growth factor receptor binding 59.1% of TURBT bladder specimens, 65.1% of radical protein 2(Grb2)to reduce the binding between Grb2and Sos cystectomy bladder specimens, and 76.9% of lymph node (25). DOC-2/DAB2 inhibits Ras-mediated mitogenic stimula- specimens from patients with UCB had decreased DOC-2/ tion by binding to Grb2 (27, 28) and negatively regulates Src in DAB2expression. Interestingly, similar to the results we found the prostate (29). DOC-2/DAB2 is also a negative regulator of in transgenic mice, DOC-2/DAB2 expression seems to be Wnt signaling via stabilization of h-catenin degradation com- decreased early in human UCB: 60.5% of patients with low- plex (30). We have recently shown, in urothelial carcinoma cell stage UCB and 60.0% with low-grade UCB had decreased lines, that DOC-2/DAB transcription regulation is mediated expression of DOC-2/DAB2. In one study of ovarian carcino- by the epigenetic modulation through histone acetylation and ma, DOC-2/DAB2 was decreased in 74% of tumors studied by specific transcription factors such as GATA6 (13), but not by (23). The loss of DOC-2/DAB2 expression in ovarian epithelial DNA methylation (see Supplementary Data). Similarly, in tumors did not correlate with tumor grade; subsequently, it was esophageal cancer, DNA methylation was infrequently associ- postulated that the loss of DOC-2/DAB2 is an early event in ated with DOC-2/DAB2 gene promoter region (15). Taken ovarian tumorigenesis (24). Similarly, no relation was found together, these findings suggest that DOC-2/DAB2 acts as a between the level of DOC-2/DAB2 expression and stage of homeostatic factor that prevents mitogen-activated signaling colon cancer (17). Moreover, in esophageal squamous cell pathways from overactivation and seems to be a strong

www.aacrjournals.org 4405 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2007 American Association for Cancer Research. Imaging, Diagnosis, Prognosis suppressor of cancer cell growth. In bladder tumors with Potential limitations to our study should be noted. The use normal expression of DOC-2/DAB2 (f35%), it is possible that of immunohistochemistry has its inherent limitations with another member of the DOC-2/DAB2 signaling pathway is regards to choice of antibody, specimen handling, staining altered, but not DOC-2/DAB2 per se, therefore contributing to techniques, and scoring protocols. To minimize variability due early bladder cancer pathogenesis. to these factors, we used a tissue microarray and a standardized In addition to its role as a signaling molecule, DOC-2/ scoring protocol. In addition, the staining patterns were read DAB2plays a role in cell-cell adhesion. The presence of a separately by two pathologists who were blinded to clinical basement membrane in breast and ovarian carcinomas outcomes, resulting in good reproducibility. The small number correlates with the expression of DOC-2/DAB2, and plating of patients and relatively short duration of follow-up might DOC-2/DAB2–transfected tumor cells on a basement mem- have limited the ability to detect significant associations with brane rescued the cells from death (31). Moreover, re- clinical outcomes. expression of DOC-2/DAB2 in DOC-2/DAB2–negative ovarian and breast cancer cells results in cell death (31). These Conclusions results indicate that DOC-2/DAB2 acts as a tumor suppressor gene by controlling the position of epithelial cells to the DOC-2/DAB2, a phosphoprotein involved in modulating basement membrane. Loss of DOC-2/DAB2 may contribute to signal transduction, is frequently decreased in patients with the basement membrane–independent, disorganized prolifer- UCB and seems to be an early event in bladder tumorigenesis. ation of tumor cells in ovarian and breast carcinomas (31). Further studies are needed to elucidate the molecular However, little is known whether DOC2/DAB2, in this consequences of DOC-2/DAB2–decreased expression in the functional capacity, has any impact on the urothelium and pathogenesis of UCB and the potential of DOC-2/DAB2– its predilection to develop malignancy. targeted therapy.

References 1. JemalA,SiegelR,WardE,MurrayT,XuJ,ThunMJ. tion of rat DOC-2 gene during castration-induced response to adjuvant endocrine therapy in breast Cancer statistics, 2007. CA Cancer J Clin 2007;57: rat ventral prostate degeneration and its growth inhi- cancer. J Clin Oncol 1999;17:1474 ^ 81. 43^66. bitory function in human prostatic carcinoma cells. 22. Koss LG. Bladder cancer from a perspective of 2. Spruck CH, Ohneseit PF, Gonzalez-Zulueta M, et al. Endocrinol 1998;139:3542 ^ 53. 40 years. J Cell Biochem Suppl 1992;16I:23 ^9. Two molecular pathways to transitional cell carcinoma 13. Zhou J, Hernandez G, Tu SW, et al. Synergistic 23. Yang DH, Smith ER, Cohen C, et al. Molecular of the bladder. Cancer Res 1994;54:784 ^ 8. induction of DOC-2/DAB2 gene expression in transi- events associated with dysplastic morphologic 3. Wu XR. Urothelial tumorigenesis: a tale of divergent tional cell carcinoma in the presence of GATA6 and transformation and initiation of ovarian tumorigenicity. pathways. Nat Rev Cancer 2005;5:713^ 25. histone deacetylase inhibitor. Cancer Res 2005;65: Cancer 2002;94:2380 ^ 92. 4. Gibas Z, Prout GR, ConnollyJG, PontesJE, Sandberg 6089^96. 24. Fazili Z, Sun W, Mittelstaedt S, Cohen C, Xu XX. AA. Nonrandom chromosomal changes in transitional 14. Wang SC, Makino K, Xia W, et al. DOC-2/hDab-2 Disabled-2 inactivation is an early step in ovarian cell carcinoma of the bladder. Cancer Res 1984;44: inhibits ILK activity and induces anoikis in breast tumorigenicity. Oncogene 1992;18:3104 ^ 13. 12 57 ^ 6 4. cancer cells through an Akt-independent pathway. 25. Xu XX, Yi T, Tang B, Lambeth JD. Disabled-2 5. Gibas Z, Prout GR, PontesJE, ConnollyJG, Sandberg Oncogene 2001;20:6960 ^ 4. (Dab2) is an SH3 domain-binding partner of Grb2. AA. A possible specific chromosome change in tran- 15. Anupam K, Tusharkant C, Gupta SD, Ranju R. Loss Oncogene1998;16:1561^9. sitional cell carcinoma of the bladder. Cancer Genet of disabled-2 expression is an early event in esopha- 26. XuXX,YangW,JackowskiS,RockCO.Cloningofa Cytogenet 1986;19:229 ^ 38. geal squamous tumorigenesis. World J Gastroenterol novel phosphoprotein regulated by colony-stimulating 6. Bohm M, Kirch H, Otto T, Rubben H, Wieland I. De- 2006;12:6041^5. factor 1shares a domain with the Drosophila disabled letion analysis at the DEL-27, APC and MTS1 loci in 16. HuangY,FriessH,KleeffJ,etal.Doc-2/hDab2 gene product. J Biol Chem 1995;270:14184 ^ 91. bladder cancer: LOH at the DEL-27 locus on 5p13^ expression is up-regulated in primary pancreatic 27. Zhou J, Hsieh JT. The inhibitory role of DOC-2/ 12 is a prognostic marker of tumor progression. Int cancer but reduced in metastasis. Lab Invest 2001; DAB2 in growth factor receptor-mediated signal J Cancer 1997;74:291 ^ 5. 81:863 ^ 73. cascade. DOC-2/DAB2-mediated inhibition of ERK 7. Bohm M, Kleine-Besten R,Wieland I. Loss of hetero- 17. KleeffJ, HuangY,Mok SC, Zimmermann A, Friess H, phosphorylation via binding to Grb2. J Biol Chem zygosity analysis on chromosome 5p defines 5p13^ Buchler MW. Down-regulation of DOC-2 in colorectal 2001;276:27793^ 8. 12 as the critical region involved in tumor progression cancer points to its role as a tumor suppressor in this 28. Tseng CP, Ely BD, Pong RC, Wang Z, Zhou J, Hsieh of bladder carcinomas. Int J Cancer 2000;89:194 ^ 7. malignancy. Dis Colon Rectum 2002;45:1242 ^ 8. JT.The role of DOC-2/DAB2 protein phosphorylation 8. Bohm M,Wieland I, Schmidt C, Rubben H, Allhoff EP. 18. Fulop V, Colitti CV, Genest D, et al. DOC-2/hDab2, in the inhibition of AP-1 activity. An underlying Loss of heterozygosity on chromosome 5p13^ 12 a candidate tumor suppressor gene involved in the mechanism of its tumor-suppressive function in predicts adverse prognosis in advanced bladder development of gestational trophoblastic diseases. prostate cancer. J Biol Chem 1999;274:31981 ^ 6. cancer independent of tumor stage and grade. J Urol Oncogene 1998;17:419 ^ 24. 29. Zhou J, Scholes J, Hsieh JT. Characterization of a 2002;168:2655 ^ 8. 19. Zhang ZT, Pak J, Huang HY, et al. Role of Ha-ras novel negative regulator (DOC-2/DAB2) of c-Src in 9. Mok SC,Wong KK, Chan RK, et al. Molecular cloning activation in superficial papillary pathway of urothelial normal prostatic epithelium and cancer. J Biol Chem of differentially expressed genes in human epithelial tumor formation. Oncogene 2001;20:1973 ^80. 2003;278:6936 ^ 41. ovarian cancer. Gynecol Oncol 1994;52:247 ^ 52. 20. Zhang ZT, Pak J, Shapiro E, SunTT,Wu XR. Urothe- 30. Hocevar BA, Mou F, Rennolds JL, Morris SM, 10. Mok SC, Chan WY,Wong KK, et al. DOC-2, a candi- lium-specific expression of an oncogene in transgenic CooperJA, Howe PH. Regulation of theWnt signaling date tumor suppressor gene in human epithelial mice induced the formation of carcinoma in situ and pathway by disabled-2 (Dab2). EMBO J 2003;22: ovarian cancer. Oncogene 1998;16:2381 ^ 7. invasive transitional cell carcinoma. Cancer Res 1999; 3084^94. 11. Albertsen HM, Smith SA, Melis R, et al. Sequence, 59:3512^7. 31. Sheng Z, Sun W, Smith E, Cohen C, Sheng Z, Xu genomic structure, and chromosomal assignment of 21. Harvey JM, Clark GM, Osborne CK, Allred DC. XX. Restoration of positioning control following human DOC-2. Genomics 1996;33:207^13. Estrogen receptor status by immunohistochemistry is Disabled-2 expression in ovarian and breast tumor 12. Tseng CP, Ely BD, Li Y, Pong RC, Hsieh JT. Regula- superior to the ligand-binding assay for predicting cells. Oncogene 2000;19:4847 ^ 54.

Clin Cancer Res 2007;13(15) August 1,2007 4406 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 24, 2021. © 2007 American Association for Cancer Research. Decreased DOC-2/DAB2 Expression in Urothelial Carcinoma of the Bladder

Jose A. Karam, Shahrokh F. Shariat, Hong-Ying Huang, et al.

Clin Cancer Res 2007;13:4400-4406.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/13/15/4400

Cited articles This article cites 31 articles, 10 of which you can access for free at: http://clincancerres.aacrjournals.org/content/13/15/4400.full#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/13/15/4400.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/13/15/4400. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.