- Email: [email protected]
Prognostic value of angiogenesis in schistosoma-associated squamous cell carcinoma of the urinary bladder
ADULT UROLOGY
PROGNOSTIC VALUE OF ANGIOGENESIS IN SCHISTOSOMA-ASSOCIATED SQUAMOUS CELL CARCINOMA OF THE URINARY BLADDER EMAD ELSOBKY, MAHMOUD EL-BAZ, MOHAMED GOMHA, HASSAN ABOL-ENEIN, ATALLAH A. SHAABAN
AND
ABSTRACT Objectives. To evaluate angiogenesis as a prognostic marker in squamous cell carcinoma of the urinary bladder in 154 patients who underwent radical cystectomy. Methods. The tumors from 98 men and 56 women (mean age 46.3 ⫾ 8.4 years) were examined. Vessels were stained using an antibody to the platelet endothelial cell adhesion molecule CD31. Microvessels were counted in active areas of angiogenesis within the tumors. Microvessel density (MVD) was quantified using the mean of three counts. Age, sex, tumor grade and stage, DNA ploidy, and MVD were evaluated in relation to outcome. Univariate and multivariate analyses of survival were performed. Results. The median follow-up period was 63 months. The overall 5-year survival rate was 56 ⫾ 4.1. Tumor grade, tumor stage, DNA ploidy, and MVD had a significant impact on the survival of patients in univariate analysis. The 5-year survival rate in patients with a low MVD (11 or less) was 68.1% compared with 50.4% for those with a high MVD (greater than 11; P ⬍0.01). Men had more vascular tumors than did women. Also, high-grade tumors had significantly higher vascular counts. In a Cox proportional hazard model, tumor angiogenesis sustained its significant impact on survival of the patients in addition to tumor stage and DNA ploidy. Conclusions. These findings suggest that angiogenesis and DNA ploidy are independent additional prognostic factors in patients with squamous cell carcinoma of the urinary bladder. UROLOGY 60: 69–73, 2002. © 2002, Elsevier Science Inc.
P
rimary squamous cell carcinoma is a relatively rare tumor comprising 1% to 7% of all bladder cancers in the Western world.1,2 Bladder carcinoma is the most common solid tumor in men in Egypt,3 where squamous cell carcinoma represents two thirds of the cases in cystectomy specimens.4 Such a high incidence is due to associated schistosomiasis.5 Other potential etiologies include chronic infection, vesical stones, chronic indwelling catheters, and bladder diverticula.6,7 Almost all squamous cell carcinomas are already advanced and muscle invasive at diagnosis. Transurethral resection, partial cystectomy, radiotherapy, and chemotherapy are not effective.3 Radical From the Departments of Urology and Pathology, Urology and Nephrology Center, Mansoura University, Mansoura, Egypt Reprint requests: Atallah A. Shaaban, M.D. Department of Urology, Urology and Nephrology Center, Mansoura University, Mansoura, Egypt Submitted: December 6, 2001, accepted (with revisions): February 18, 2002 © 2002, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
cystectomy with urinary diversion remains the standard of care. Tumor stage and grade have a significant impact on survival.4 Recently, tumor angiogenesis was considered as a prognostic factor in many solid tumors. Also, urologic malignancies depend on angiogenesis for growth and metastasis.8 The rationale of the present study was to evaluate tumor angiogenesis as a possibly better prognostic marker in squamous cell carcinoma of the bladder and to assess its relationship to different prognostic factors. MATERIAL AND METHODS The pathologic materials of 154 patients (98 men and 56 women) with Schistosoma-associated squamous cell carcinoma of urinary bladder were studied. Patient age ranged from 28 to 67 years (mean 46.3 ⫾ 8.4). They underwent radical cystectomy and urinary diversion between 1985 and 1991. Tumors were evaluated using the World Health Organization grading system9 and were staged according to the TNM system of the International Union Against Cancer.10 Flow cytometric DNA analysis was performed using the 0090-4295/02/$22.00 PII S0090-4295(02)01669-2 69
technique previously described.11 The DNA profile was divided into two groups: diploid and aneuploid. Cell populations demonstrating the normal histographic findings with the distinct G1 peak were ranked as diploid. Those with a distinct peak, in addition to the G1 diploid peak, and found in a position deviating more than 10% from the diploid internal standard were ranked as aneuploid.
IMMUNOHISTOCHEMICAL STAINING FOR TUMOR ANGIOGENESIS
Immunohistochemical staining was performed on 4-m sections of paraffin-embedded tumor specimens. Unmasking of the antigen sites using a microwave antigen retrieval technique12 was performed before incubation with the primary antibody. The slides were incubated with antibody to the platelet endothelial cell adhesion molecule CD31 as the primary antibody (Serotec, UK). It was applied at a 1/50 dilution for 60 minutes at room temperature. Slides were stained according to the standard avidin-biotin method. The immunoreaction was visualized using diaminobenzidine (Sigma Chemical, UK) as the chromogen. The slides were counterstained with hematoxylin. Vessel counting was performed in areas with maximal neovascularization within the tumors away from any areas of artifact, necrosis, or inflammation and without prior knowledge of patient outcome. Low-power light microscopy at ⫻40 magnification was used to scan the often heterogeneous tumor sections for these areas. At ⫻250 magnification, counts were made of all distinct brown staining endothelial cells. The microvessel density (MVD) was defined as the mean of the highest three counts. To be defined as an individual vessel, cell nests showing immunostaining had to be clearly separate from adjacent microvessels, tumor cells, or connective tissue elements. Vessel lumina or intraluminal red blood cells were not necessary to define microvessels. Larger muscular walled vessels were excluded from the assessment.
FOLLOW-UP Regular follow-up of patients included history taking, clinical examination, and radiologic evaluation every month during the first 6 months and bimonthly thereafter. They were examined for treatment failure using clinical, radiologic, or histopathologic findings. Treatment failure was defined as death from cancer or development of local pelvic recurrence or distant metastasis. The overall median follow-up period was 63 months (range 0.7 to 189.7). The median follow-up period for the treatment failure group was 10.4 months (range 0.7 to 173); it was 126.9 months (range 4.2 to 189.7) for the group living free of disease.
PROGNOSTIC VARIABLES Age, sex, tumor grade, tumor stage, DNA ploidy, and MVD were evaluated in relation to the outcome of the patients.
STATISTICAL ANALYSIS Statistical analysis of outcome was performed with KaplanMeier survival curves.13 Univariate analysis was performed with the log-rank test. As it was not known for which value tumor vascularity would provide effective prognostic groups and as the prognostic factors are usually considered dichotomized discontinuous variables, a cutpoint analysis was performed.14,15 A considerable range of cutoff values was tried, and, of these, the optimal cutoff point was 11 units. It divided the study population into a low-vascular group (MVD 11 or less) and a high-vascular group (MVD greater than 11). The relationship between angiogenesis and different clinical and pathologic factors was estimated by the chi-square test. Mul70
TABLE I. Clinicopathologic characteristics of the study group and their relation to the tumor angiogenesis Variable
MVD P <11 >11 Value*
Total (n)
Age (yr) ⱕ50 109 (70.8) ⬎50 45 (29.2) Sex Male 98 (63.6) Female 56 (36.4) Grade G1 82 (53.2) G2 55 (35.7) G3 17 (11.1) Stage Organ-confined 118 (76.6) Non-organ-confined 36 (23.4) DNA ploidy Diploid 102 (66.2) Aneuploid 52 (33.8)
0.22 35 10
74 35
20 25
78 31
31 14 —
51 41 17
36 9
82 27
34 11
68 41
0.002
0.006
0.52
0.116
KEY: MVD ⫽ microvessel density. Numbers in parentheses are percentages. * Chi-square test.
tivariate analysis was performed with Cox’s proportional hazard model.16 The statistical analysis was performed using the Statistical Package for the Social Sciences software package, version 6.1 (SPSS Inc., Chicago, Ill).
RESULTS The patient and tumor characteristics are summarized in Table I. TUMOR VASCULARITY A total of 45 patients had tumors with low MVD (11 or less), and 109 had tumors with high MVD (greater than 11). The MVD ranged from 8 to 26 (mean 14.4 ⫾ 4; Fig. 1). ANGIOGENESIS AND CLINICOPATHOLOGIC CHARACTERISTICS No significant relationship was found between MVD and age, stage, or DNA ploidy. Sex and tumor grade had a significant association with the MVD. High-grade tumors had a significantly higher MVD than did low-grade tumors (P ⫽ 0.0058). Also, men had tumors of higher vascularity than did women (P ⫽ 0.0015). SURVIVAL Overall. The 5-year survival rate was 56% ⫾ 4.1%. Treatment failure was high during the first 2 years, then declined thereafter. Age and sex had no impact on survival (Fig. 2A). Grade. A significant relationship was observed between tumor grade and survival (P ⫽ 0.043). UROLOGY 60 (1), 2002
FIGURE 1. Squamous cell carcinoma of the urinary bladder. (A) Low-vascular tumor and (B) high-vascular tumor, both stained with anti-CD31 monoclonal antibody and avidin-biotin immunoperoxidase. Original magnification ⫻250.
The 5-year survival rate for patients with grade 1, 2, and 3 disease was 62.2%, 52%, and 35.3%, respectively. Stage. Tumors were divided into two groups, organ confined and non-organ confined. Organ-confined tumors (n ⫽ 118) included Stage pT1 and pT2 with negative lymph nodes. Non-organ-confined tumors (n ⫽ 36) included Stage pT3 and pT4 and any stage with regional lymph node metastasis. The 5-year survival rate for organ-confined tumors was 66.9% compared with 19.4% for non-organconfined disease (P ⬍0.0001) (Fig. 2B). DNA Ploidy. The DNA profile was diploid in 102 cases and aneuploid in 52. Aneuploidy was associated with a poor prognosis. Patients with diploid tumors had a 67% 5-year survival rate versus 32.5% for those with aneuploid tumors (P ⬍0.0001). Tumor Angiogenesis. Tumor angiogenesis had a significant impact on survival. The 5-year survival rate of patients with a low MVD was 68.1% comUROLOGY 60 (1), 2002
FIGURE 2. (A) Overall survival of 154 patients with squamous cell carcinoma of the urinary bladder. (B) Survival of 154 patients with squamous cell carcinoma of the urinary bladder in relation to the stage. (C) Survival of 154 patients with squamous cell carcinoma of the urinary bladder in relation to MVD.
pared with 50.4% for those with a high MVD (P ⫽ 0.0096; Fig. 2C). MULTIVARIATE ANALYSIS Multivariate analysis was performed for the significant variables in the univariate analysis (grade, stage, DNA, and MVD). The grade could not sustain its significance, but stage, DNA ploidy, and MVD proved to be independent variables for prognosis and sustained their significance (Table II). 71
TABLE II. Results of proportional hazard analysis (Cox regression) of disease-free survival
MVD ⱕ11 ⬎11 DNA ploidy Diploid Aneuploid Stage Organ-confined Non-organ-confined
Regression Estimate
Standard Error
Relative Risk (95% CI)
P Value
0.6590
0.2985
1 1.9328 (1.0768, 3.4694)
0.0272
0.7295
0.2433
1 2.0740 (1.2875, 3.3409)
0.0027
1.2379
0.2459
1 3.4482 (2.1296, 5.5833)
0.0000
KEY: CI ⫽ confidence interval; MVD ⫽ microvessel density.
COMMENT
CONCLUSIONS
In this report, tumor grade and stage were significant prognostic indicators of survival in squamous cell carcinoma of the urinary bladder, as previously reported.4 However, the inaccuracy of tumor staging11 and grading17 dictated the need for additional objective information about the biologic behavior of such tumors. Flow cytometric DNA analysis and tumor angiogenesis represent reliable means that have proven validity in the assessment of bladder tumors.11,14 In this series, DNA ploidy had a considerable impact on patient survival. The 5-year survival rate was 67% for those with diploid tumors versus 32.5% for those with aneuploid tumors. Similar results have been previously reported.18,19 Angiogenesis has been proposed as a prognostic factor in tumors in which it is a prerequisite for tumor growth.20 Angiogenic activity correlated with a higher incidence of lymph node metastasis and a poor prognosis for patients with transitional cell carcinoma and adenocarcinoma of the bladder.14,21,22 In this study, a significant relationship was observed between angiogenesis and sex, as well as tumor grade. Men might have a higher MVD because of the angiogenic stimulatory activity of androgens.23,24 Furthermore, the high-grade tumors might be associated with increased MVD because of increased activity of angiogenic stimulatory factors such as thymidine phosphorylase25,26 and acidic fibroblast growth factor.27 Interestingly, a significant relationship was noted between tumor angiogenesis and patient outcome. Patients with a MVD of 11 or lower had a 5-year survival rate of 68.1%. In contrast, patients with a high MVD (greater than 11) had a 5-year survival rate of 50.4%. Furthermore, on multivariate analysis, tumor stage, DNA ploidy, and tumor angiogenesis sustained their significant impact on patient survival, but, as Dalbagni et al.28 recently reported, the histologic grade did not.
Tumor angiogenesis is an independent prognostic factor in squamous cell carcinoma in addition to tumor stage and DNA ploidy. These factors together could provide a more objective tool for better judgment on patient survival and might help in the choice of the more effective therapy for the individual patient.
72
REFERENCES 1. Costello AT, Tiptaft RC, England HR, et al: Squamous cell carcinoma of the bladder. Urology 23: 234 –236, 1984. 2. Lynch CF, and Cohen MB: Urinary system. Cancer 75(suppl): 316 –329, 1995. 3. Ghoneim MA: Nontransitional cell bladder cancer, in Krane RJ, Siroky MB, and Fitzpatrick JM (Eds): Clinical Urology. Philadelphia, JB Lippincott, 1994, pp 679 – 687. 4. Ghoneim MA, El-Mekresh MM, El-Baz MA, et al: Radical cystectomy for carcinoma of the bladder: critical evaluation of the results in 1,026 cases. J Urol 158: 393–399, 1997. 5. El-Bolkainy MN, Tawfik NH, and Kamel IA: Histopathologic classification of carcinomas in schistosomal bladder, in El-Bolkainy MN, and Chu Ew (Eds): Detection of Bladder Cancer Associated with Schistosomiasis. Cairo, AlAhram Press, 1982, pp 107–123. 6. Bejany DE, Lockhart JL, and Rhamy RK: Malignant tumors following spinal cord injury. J Urol 138: 1390 –1392, 1987. 7. Broecker BH, Klein FA, and Hackler RH: Cancer of the bladder in spinal cord injury patients. J Urol 125: 196 –197, 1981. 8. Campbell SC: Advances in angiogenesis research: Relevance to urological oncology. J Urol 158: 1663–1674, 1997. 9. Mostofi FK, Sobin LH, and Torloni H: Histological Typing of Urinary Bladder Tumors. International Classification of Tumors, No. 10. Geneva, World Health Organization, 1973. 10. Fleming ID, Cooper JS, Henson DE et al (Eds): AJCC Cancer Staging Manual, 5th ed. Philadelphia, LippincottRaven, 1997, pp 241–243. 11. Shaaban AA, Tribukait B, El-Bedeiwy A, et al: Characterization of squamous cell bladder tumors by flow cytometric deoxyribonucleic acid analysis: a report of 100 cases. J Urol 144: 879 – 883, 1990. 12. Kok L, and Boon M: Microwave Cookbook for Microscopists: Art and Science of Visualization, 3rd ed. Leiden, Coulomb Press, 1992. 13. Kaplan EL, and Meier P: Nonparametric estimation UROLOGY 60 (1), 2002
from incomplete observations. J Am Stat Assoc 53: 457– 481, 1958. 14. Dickinson AJ, Fox SB, Persad RA, et al: Quantification of angiogenesis as an independent predictor of prognosis in invasive bladder carcinomas. Br J Urol 74: 762–766, 1994. 15. Tandon A, Clark GM, Chamness GC, et al: Cathepsin D and prognosis in breast cancer. N Engl J Med 322: 297–302, 1990. 16. Cox DR: Regression models and life tables. J Roy Stat Soc Ser B 34: 187–220, 1972. 17. Ooms ECM, Anderson WAD, Alons CL, et al: Analysis of the performance of pathologists in the grading of bladder tumors. Hum Pathol 14: 140 –143, 1983. 18. Winkler HZ, Nativ O, Hosaka Y, et al: Nuclear DNA ploidy in squamous cell bladder cancer. J Urol 141: 297–302, 1989. 19. Shaaban AA, Javadpour N, Tribukait B, et al: Prognostic significance of flow-DNA analysis and cell surface antigens in carcinoma of bilharzial bladder. Urology 39: 207–210, 1992. 20. Folkman J, and Shing Y: Angiogenesis. J Biol Chem 267: 10931–10934, 1992. 21. Bochner BH, Cote RJ, Weidner N, et al: Angiogenesis in bladder cancer: relationship between microvessel density and tumor prognosis. J Natl Cancer Inst 87: 1603–1612, 1995. 22. Elsobky E, Gomha M, El-Baz M, et al: Prognostic sig-
UROLOGY 60 (1), 2002
nificance of tumour angiogenesis in Schistosoma-associated adenocarcinoma of the urinary bladder. BJU Int 89: 126 –132, 2002. 23. Buttyan R, Ghafar MA, and Shabsigh A: The effects of androgen deprivation on the prostate gland: cell death mediated by vascular regression. Curr Opin Urol 10: 415– 420, 2000. 24. Franck-Lissbrant I, Ha¨ gstro¨ m S, Damber J-E, et al: Testosterone stimulates angiogenesis and vascular regrowth in the ventral prostate in castrated rats. Endocrinology 139: 451– 456, 1998. 25. O’Brien TS, Fox SB, Dickinson AJ, et al: Expression of the angiogenic factor thymidine phosphorylase/platelet-derived endothelial cell growth factor in primary bladder cancers. Cancer Res 56: 4799 – 4804, 1996. 26. Arima JI, Imazono Y, Takebayashi Y, et al: Expression of thymidine phosphorylase as an indicator of poor prognosis for patients with transitional cell carcinoma of the bladder. Cancer 88: 1131–1138, 2000. 27. Chopin DK, Caruelle JP, Colombel M, et al: Increased immunodetection of acidic fibroblast growth factor in bladder cancer, detectable in urine. J Urol 150: 1126 –1130, 1993. 28. Dalbagni G, Genega E, Hashibe M, et al: Cystectomy for bladder cancer: a contemporary series. J Urol 165: 1111–1116, 2001.
73
PROGNOSTIC VALUE OF ANGIOGENESIS IN SCHISTOSOMA-ASSOCIATED SQUAMOUS CELL CARCINOMA OF THE URINARY BLADDER EMAD ELSOBKY, MAHMOUD EL-BAZ, MOHAMED GOMHA, HASSAN ABOL-ENEIN, ATALLAH A. SHAABAN
AND
ABSTRACT Objectives. To evaluate angiogenesis as a prognostic marker in squamous cell carcinoma of the urinary bladder in 154 patients who underwent radical cystectomy. Methods. The tumors from 98 men and 56 women (mean age 46.3 ⫾ 8.4 years) were examined. Vessels were stained using an antibody to the platelet endothelial cell adhesion molecule CD31. Microvessels were counted in active areas of angiogenesis within the tumors. Microvessel density (MVD) was quantified using the mean of three counts. Age, sex, tumor grade and stage, DNA ploidy, and MVD were evaluated in relation to outcome. Univariate and multivariate analyses of survival were performed. Results. The median follow-up period was 63 months. The overall 5-year survival rate was 56 ⫾ 4.1. Tumor grade, tumor stage, DNA ploidy, and MVD had a significant impact on the survival of patients in univariate analysis. The 5-year survival rate in patients with a low MVD (11 or less) was 68.1% compared with 50.4% for those with a high MVD (greater than 11; P ⬍0.01). Men had more vascular tumors than did women. Also, high-grade tumors had significantly higher vascular counts. In a Cox proportional hazard model, tumor angiogenesis sustained its significant impact on survival of the patients in addition to tumor stage and DNA ploidy. Conclusions. These findings suggest that angiogenesis and DNA ploidy are independent additional prognostic factors in patients with squamous cell carcinoma of the urinary bladder. UROLOGY 60: 69–73, 2002. © 2002, Elsevier Science Inc.
P
rimary squamous cell carcinoma is a relatively rare tumor comprising 1% to 7% of all bladder cancers in the Western world.1,2 Bladder carcinoma is the most common solid tumor in men in Egypt,3 where squamous cell carcinoma represents two thirds of the cases in cystectomy specimens.4 Such a high incidence is due to associated schistosomiasis.5 Other potential etiologies include chronic infection, vesical stones, chronic indwelling catheters, and bladder diverticula.6,7 Almost all squamous cell carcinomas are already advanced and muscle invasive at diagnosis. Transurethral resection, partial cystectomy, radiotherapy, and chemotherapy are not effective.3 Radical From the Departments of Urology and Pathology, Urology and Nephrology Center, Mansoura University, Mansoura, Egypt Reprint requests: Atallah A. Shaaban, M.D. Department of Urology, Urology and Nephrology Center, Mansoura University, Mansoura, Egypt Submitted: December 6, 2001, accepted (with revisions): February 18, 2002 © 2002, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED
cystectomy with urinary diversion remains the standard of care. Tumor stage and grade have a significant impact on survival.4 Recently, tumor angiogenesis was considered as a prognostic factor in many solid tumors. Also, urologic malignancies depend on angiogenesis for growth and metastasis.8 The rationale of the present study was to evaluate tumor angiogenesis as a possibly better prognostic marker in squamous cell carcinoma of the bladder and to assess its relationship to different prognostic factors. MATERIAL AND METHODS The pathologic materials of 154 patients (98 men and 56 women) with Schistosoma-associated squamous cell carcinoma of urinary bladder were studied. Patient age ranged from 28 to 67 years (mean 46.3 ⫾ 8.4). They underwent radical cystectomy and urinary diversion between 1985 and 1991. Tumors were evaluated using the World Health Organization grading system9 and were staged according to the TNM system of the International Union Against Cancer.10 Flow cytometric DNA analysis was performed using the 0090-4295/02/$22.00 PII S0090-4295(02)01669-2 69
technique previously described.11 The DNA profile was divided into two groups: diploid and aneuploid. Cell populations demonstrating the normal histographic findings with the distinct G1 peak were ranked as diploid. Those with a distinct peak, in addition to the G1 diploid peak, and found in a position deviating more than 10% from the diploid internal standard were ranked as aneuploid.
IMMUNOHISTOCHEMICAL STAINING FOR TUMOR ANGIOGENESIS
Immunohistochemical staining was performed on 4-m sections of paraffin-embedded tumor specimens. Unmasking of the antigen sites using a microwave antigen retrieval technique12 was performed before incubation with the primary antibody. The slides were incubated with antibody to the platelet endothelial cell adhesion molecule CD31 as the primary antibody (Serotec, UK). It was applied at a 1/50 dilution for 60 minutes at room temperature. Slides were stained according to the standard avidin-biotin method. The immunoreaction was visualized using diaminobenzidine (Sigma Chemical, UK) as the chromogen. The slides were counterstained with hematoxylin. Vessel counting was performed in areas with maximal neovascularization within the tumors away from any areas of artifact, necrosis, or inflammation and without prior knowledge of patient outcome. Low-power light microscopy at ⫻40 magnification was used to scan the often heterogeneous tumor sections for these areas. At ⫻250 magnification, counts were made of all distinct brown staining endothelial cells. The microvessel density (MVD) was defined as the mean of the highest three counts. To be defined as an individual vessel, cell nests showing immunostaining had to be clearly separate from adjacent microvessels, tumor cells, or connective tissue elements. Vessel lumina or intraluminal red blood cells were not necessary to define microvessels. Larger muscular walled vessels were excluded from the assessment.
FOLLOW-UP Regular follow-up of patients included history taking, clinical examination, and radiologic evaluation every month during the first 6 months and bimonthly thereafter. They were examined for treatment failure using clinical, radiologic, or histopathologic findings. Treatment failure was defined as death from cancer or development of local pelvic recurrence or distant metastasis. The overall median follow-up period was 63 months (range 0.7 to 189.7). The median follow-up period for the treatment failure group was 10.4 months (range 0.7 to 173); it was 126.9 months (range 4.2 to 189.7) for the group living free of disease.
PROGNOSTIC VARIABLES Age, sex, tumor grade, tumor stage, DNA ploidy, and MVD were evaluated in relation to the outcome of the patients.
STATISTICAL ANALYSIS Statistical analysis of outcome was performed with KaplanMeier survival curves.13 Univariate analysis was performed with the log-rank test. As it was not known for which value tumor vascularity would provide effective prognostic groups and as the prognostic factors are usually considered dichotomized discontinuous variables, a cutpoint analysis was performed.14,15 A considerable range of cutoff values was tried, and, of these, the optimal cutoff point was 11 units. It divided the study population into a low-vascular group (MVD 11 or less) and a high-vascular group (MVD greater than 11). The relationship between angiogenesis and different clinical and pathologic factors was estimated by the chi-square test. Mul70
TABLE I. Clinicopathologic characteristics of the study group and their relation to the tumor angiogenesis Variable
MVD P <11 >11 Value*
Total (n)
Age (yr) ⱕ50 109 (70.8) ⬎50 45 (29.2) Sex Male 98 (63.6) Female 56 (36.4) Grade G1 82 (53.2) G2 55 (35.7) G3 17 (11.1) Stage Organ-confined 118 (76.6) Non-organ-confined 36 (23.4) DNA ploidy Diploid 102 (66.2) Aneuploid 52 (33.8)
0.22 35 10
74 35
20 25
78 31
31 14 —
51 41 17
36 9
82 27
34 11
68 41
0.002
0.006
0.52
0.116
KEY: MVD ⫽ microvessel density. Numbers in parentheses are percentages. * Chi-square test.
tivariate analysis was performed with Cox’s proportional hazard model.16 The statistical analysis was performed using the Statistical Package for the Social Sciences software package, version 6.1 (SPSS Inc., Chicago, Ill).
RESULTS The patient and tumor characteristics are summarized in Table I. TUMOR VASCULARITY A total of 45 patients had tumors with low MVD (11 or less), and 109 had tumors with high MVD (greater than 11). The MVD ranged from 8 to 26 (mean 14.4 ⫾ 4; Fig. 1). ANGIOGENESIS AND CLINICOPATHOLOGIC CHARACTERISTICS No significant relationship was found between MVD and age, stage, or DNA ploidy. Sex and tumor grade had a significant association with the MVD. High-grade tumors had a significantly higher MVD than did low-grade tumors (P ⫽ 0.0058). Also, men had tumors of higher vascularity than did women (P ⫽ 0.0015). SURVIVAL Overall. The 5-year survival rate was 56% ⫾ 4.1%. Treatment failure was high during the first 2 years, then declined thereafter. Age and sex had no impact on survival (Fig. 2A). Grade. A significant relationship was observed between tumor grade and survival (P ⫽ 0.043). UROLOGY 60 (1), 2002
FIGURE 1. Squamous cell carcinoma of the urinary bladder. (A) Low-vascular tumor and (B) high-vascular tumor, both stained with anti-CD31 monoclonal antibody and avidin-biotin immunoperoxidase. Original magnification ⫻250.
The 5-year survival rate for patients with grade 1, 2, and 3 disease was 62.2%, 52%, and 35.3%, respectively. Stage. Tumors were divided into two groups, organ confined and non-organ confined. Organ-confined tumors (n ⫽ 118) included Stage pT1 and pT2 with negative lymph nodes. Non-organ-confined tumors (n ⫽ 36) included Stage pT3 and pT4 and any stage with regional lymph node metastasis. The 5-year survival rate for organ-confined tumors was 66.9% compared with 19.4% for non-organconfined disease (P ⬍0.0001) (Fig. 2B). DNA Ploidy. The DNA profile was diploid in 102 cases and aneuploid in 52. Aneuploidy was associated with a poor prognosis. Patients with diploid tumors had a 67% 5-year survival rate versus 32.5% for those with aneuploid tumors (P ⬍0.0001). Tumor Angiogenesis. Tumor angiogenesis had a significant impact on survival. The 5-year survival rate of patients with a low MVD was 68.1% comUROLOGY 60 (1), 2002
FIGURE 2. (A) Overall survival of 154 patients with squamous cell carcinoma of the urinary bladder. (B) Survival of 154 patients with squamous cell carcinoma of the urinary bladder in relation to the stage. (C) Survival of 154 patients with squamous cell carcinoma of the urinary bladder in relation to MVD.
pared with 50.4% for those with a high MVD (P ⫽ 0.0096; Fig. 2C). MULTIVARIATE ANALYSIS Multivariate analysis was performed for the significant variables in the univariate analysis (grade, stage, DNA, and MVD). The grade could not sustain its significance, but stage, DNA ploidy, and MVD proved to be independent variables for prognosis and sustained their significance (Table II). 71
TABLE II. Results of proportional hazard analysis (Cox regression) of disease-free survival
MVD ⱕ11 ⬎11 DNA ploidy Diploid Aneuploid Stage Organ-confined Non-organ-confined
Regression Estimate
Standard Error
Relative Risk (95% CI)
P Value
0.6590
0.2985
1 1.9328 (1.0768, 3.4694)
0.0272
0.7295
0.2433
1 2.0740 (1.2875, 3.3409)
0.0027
1.2379
0.2459
1 3.4482 (2.1296, 5.5833)
0.0000
KEY: CI ⫽ confidence interval; MVD ⫽ microvessel density.
COMMENT
CONCLUSIONS
In this report, tumor grade and stage were significant prognostic indicators of survival in squamous cell carcinoma of the urinary bladder, as previously reported.4 However, the inaccuracy of tumor staging11 and grading17 dictated the need for additional objective information about the biologic behavior of such tumors. Flow cytometric DNA analysis and tumor angiogenesis represent reliable means that have proven validity in the assessment of bladder tumors.11,14 In this series, DNA ploidy had a considerable impact on patient survival. The 5-year survival rate was 67% for those with diploid tumors versus 32.5% for those with aneuploid tumors. Similar results have been previously reported.18,19 Angiogenesis has been proposed as a prognostic factor in tumors in which it is a prerequisite for tumor growth.20 Angiogenic activity correlated with a higher incidence of lymph node metastasis and a poor prognosis for patients with transitional cell carcinoma and adenocarcinoma of the bladder.14,21,22 In this study, a significant relationship was observed between angiogenesis and sex, as well as tumor grade. Men might have a higher MVD because of the angiogenic stimulatory activity of androgens.23,24 Furthermore, the high-grade tumors might be associated with increased MVD because of increased activity of angiogenic stimulatory factors such as thymidine phosphorylase25,26 and acidic fibroblast growth factor.27 Interestingly, a significant relationship was noted between tumor angiogenesis and patient outcome. Patients with a MVD of 11 or lower had a 5-year survival rate of 68.1%. In contrast, patients with a high MVD (greater than 11) had a 5-year survival rate of 50.4%. Furthermore, on multivariate analysis, tumor stage, DNA ploidy, and tumor angiogenesis sustained their significant impact on patient survival, but, as Dalbagni et al.28 recently reported, the histologic grade did not.
Tumor angiogenesis is an independent prognostic factor in squamous cell carcinoma in addition to tumor stage and DNA ploidy. These factors together could provide a more objective tool for better judgment on patient survival and might help in the choice of the more effective therapy for the individual patient.
72
REFERENCES 1. Costello AT, Tiptaft RC, England HR, et al: Squamous cell carcinoma of the bladder. Urology 23: 234 –236, 1984. 2. Lynch CF, and Cohen MB: Urinary system. Cancer 75(suppl): 316 –329, 1995. 3. Ghoneim MA: Nontransitional cell bladder cancer, in Krane RJ, Siroky MB, and Fitzpatrick JM (Eds): Clinical Urology. Philadelphia, JB Lippincott, 1994, pp 679 – 687. 4. Ghoneim MA, El-Mekresh MM, El-Baz MA, et al: Radical cystectomy for carcinoma of the bladder: critical evaluation of the results in 1,026 cases. J Urol 158: 393–399, 1997. 5. El-Bolkainy MN, Tawfik NH, and Kamel IA: Histopathologic classification of carcinomas in schistosomal bladder, in El-Bolkainy MN, and Chu Ew (Eds): Detection of Bladder Cancer Associated with Schistosomiasis. Cairo, AlAhram Press, 1982, pp 107–123. 6. Bejany DE, Lockhart JL, and Rhamy RK: Malignant tumors following spinal cord injury. J Urol 138: 1390 –1392, 1987. 7. Broecker BH, Klein FA, and Hackler RH: Cancer of the bladder in spinal cord injury patients. J Urol 125: 196 –197, 1981. 8. Campbell SC: Advances in angiogenesis research: Relevance to urological oncology. J Urol 158: 1663–1674, 1997. 9. Mostofi FK, Sobin LH, and Torloni H: Histological Typing of Urinary Bladder Tumors. International Classification of Tumors, No. 10. Geneva, World Health Organization, 1973. 10. Fleming ID, Cooper JS, Henson DE et al (Eds): AJCC Cancer Staging Manual, 5th ed. Philadelphia, LippincottRaven, 1997, pp 241–243. 11. Shaaban AA, Tribukait B, El-Bedeiwy A, et al: Characterization of squamous cell bladder tumors by flow cytometric deoxyribonucleic acid analysis: a report of 100 cases. J Urol 144: 879 – 883, 1990. 12. Kok L, and Boon M: Microwave Cookbook for Microscopists: Art and Science of Visualization, 3rd ed. Leiden, Coulomb Press, 1992. 13. Kaplan EL, and Meier P: Nonparametric estimation UROLOGY 60 (1), 2002
from incomplete observations. J Am Stat Assoc 53: 457– 481, 1958. 14. Dickinson AJ, Fox SB, Persad RA, et al: Quantification of angiogenesis as an independent predictor of prognosis in invasive bladder carcinomas. Br J Urol 74: 762–766, 1994. 15. Tandon A, Clark GM, Chamness GC, et al: Cathepsin D and prognosis in breast cancer. N Engl J Med 322: 297–302, 1990. 16. Cox DR: Regression models and life tables. J Roy Stat Soc Ser B 34: 187–220, 1972. 17. Ooms ECM, Anderson WAD, Alons CL, et al: Analysis of the performance of pathologists in the grading of bladder tumors. Hum Pathol 14: 140 –143, 1983. 18. Winkler HZ, Nativ O, Hosaka Y, et al: Nuclear DNA ploidy in squamous cell bladder cancer. J Urol 141: 297–302, 1989. 19. Shaaban AA, Javadpour N, Tribukait B, et al: Prognostic significance of flow-DNA analysis and cell surface antigens in carcinoma of bilharzial bladder. Urology 39: 207–210, 1992. 20. Folkman J, and Shing Y: Angiogenesis. J Biol Chem 267: 10931–10934, 1992. 21. Bochner BH, Cote RJ, Weidner N, et al: Angiogenesis in bladder cancer: relationship between microvessel density and tumor prognosis. J Natl Cancer Inst 87: 1603–1612, 1995. 22. Elsobky E, Gomha M, El-Baz M, et al: Prognostic sig-
UROLOGY 60 (1), 2002
nificance of tumour angiogenesis in Schistosoma-associated adenocarcinoma of the urinary bladder. BJU Int 89: 126 –132, 2002. 23. Buttyan R, Ghafar MA, and Shabsigh A: The effects of androgen deprivation on the prostate gland: cell death mediated by vascular regression. Curr Opin Urol 10: 415– 420, 2000. 24. Franck-Lissbrant I, Ha¨ gstro¨ m S, Damber J-E, et al: Testosterone stimulates angiogenesis and vascular regrowth in the ventral prostate in castrated rats. Endocrinology 139: 451– 456, 1998. 25. O’Brien TS, Fox SB, Dickinson AJ, et al: Expression of the angiogenic factor thymidine phosphorylase/platelet-derived endothelial cell growth factor in primary bladder cancers. Cancer Res 56: 4799 – 4804, 1996. 26. Arima JI, Imazono Y, Takebayashi Y, et al: Expression of thymidine phosphorylase as an indicator of poor prognosis for patients with transitional cell carcinoma of the bladder. Cancer 88: 1131–1138, 2000. 27. Chopin DK, Caruelle JP, Colombel M, et al: Increased immunodetection of acidic fibroblast growth factor in bladder cancer, detectable in urine. J Urol 150: 1126 –1130, 1993. 28. Dalbagni G, Genega E, Hashibe M, et al: Cystectomy for bladder cancer: a contemporary series. J Urol 165: 1111–1116, 2001.
73