Objective This study aimed at the analysis of the immunohistochemical expression of some biological markers, namely CD 44, Cytokeratin (CK) 20 and Cyclin D1 in transitional cell carcinoma (TCC) of the bladder and at a trial to correlate the dif-ferent results of their expression in order to establish clinico-pathologic prognostic fac-tors.
Material and Methods A total of 80 bladder specimens were included in the study. Seventy specimens were taken from patients with newly diagnosed primary transitional cell carcinoma of the bladder, and the other 10 bladder specimens were obtained from patients suffering from other non-malignant urological diseases and used as a control. The specimens were immunostained by using CD 44, CK 20 and Cyclin D1 antibodies, and statistical rela-tionships to tumor stage, tumor grade, configuration and nodal and/or distant metastases were assessed.
Results 35% of T1 tumors, 82% of mucle-invasive (T2-T4) tumors showed either complete or focal loss of the normal pattern of CD 44 expression (P<0.001). Also, 43% of low-grade tumors and 73% of high-grade tumors showed this loss of pattern of CD 44 (P<0.05) which also significantly correlated to positive metastases (p<0.01) but not to tumor configuration (p>0.1). 38% of low-grade tumors showed CK 20 positivity, while 78% of high-grade tumors were CK 20-positive (P<0.001). Also, 42% of T1

tumors were CK 20-positive compared to 81% of muscle invasive tumors (T2-T4) (P<0.001). CK 20 expression did not corre-late to tumor configuration (P>0.1) or to metastasis (P>0.5). A significant inverse relationship in staining patterns of CD 44 and CK 20 was strikingly evident in corresponding areas of some tumors. A Cyclin D1-positive expression was seen in 54% of T1 and in 23% of muscle-invasive (T2-T4) tumors (P<0.01). Cyclin D1 was positive in 71% of low-grade tumors and in 18% of high-grade tumors (P<0.001). A Cyclin D1-positive expression was also noted in 49% of papillary tumors but in only 14% of solid tumors (P<0.01). On the other hand, Cyclin D1 positivity did not correlate with metastasis (P>0.1).
Conclusion From this study we conclude that a loss of CD 44 expression, an increased CK 20 expression and a decreased Cyclin D1 expression are significantly associated with high-grade invasive transitional cell carcinoma of the bladder denoting a poor prognosis. Although histopathology will re-main the gold standard of tumor diagnosis, the immunohistochemical expression of these biological markers may be a useful adjunct for predicting the behaviour of a given tumor and for the stratification of risk factors in bladder neoplasm giving the clue to the management of these cases.
Key Words bladder carcinoma, immunohisto-chemistry, CD 44, CK 20, Cyclin D1

INTRODUCTION
Bladder cancer accounts for approximately 5% of all human cancers and represents 95% of all urothelial tumors1. Transitional cell car-cinoma (TCC) of the urinary bladder presents a spectrum of tumors ranging from indolent non-
progress to invasive cancer3. Since tumor inva-sion and metastasis formation are major obstacles to successful cancer treatment, the identification of TCCs with a high probability of developing recurrence and an aggressive course at the time of initial diagnosis is im-portant for the management of the patients4.

The natural history of bladder cancer can be correlated with several prognostic factors. These include histological grade, the depth of penetration into the bladder wall (stage), the appearance of vascular / lymphatic invasion and the presence of carcinoma in situ (CIS). Although these prognostic indicators are widely recognized, the biologic basis of the subsets of disease within each of these categories is incompletely understood. Some tumors appear to have a favourable histology but progress and express aggressive activity despite theore-tically curative treatments, whereas other tu-mors appear to have an unfavourable histology and, yet, are less aggressive or at least more amenable to cure5. Although clinically useful, a histological assessment is not sufficiently sen-sitive and discriminatory in determining the specific biological potential of a particular can-cer. Therefore, an assessment of the genetics and molecular biology of a tumor and of the entire process of carcinogenesis is warranted as novel molecular tools may reveal prognostic factors5.

It is widely accepted that deregulation of cell proliferation is a step of vital importance in cancer evolution, therefore, the definition of key events that underlie the control of this mechanism is likely to contribute not only to our understanding of cancer development and progression but also to the identification of novel biological landmarks for diagnostic and prognostic purposes6. A large number of biolo-gical markers have been extensively investi-gated as potential predictors of bladder tumor recurrence, progression or response to therapy7.

The term CD 44 designates a family of widely distributed cell surface glycoproteins with several functions. It has been implicated in cell-cell adhesion and cell matrix cytoskeleton reaction, lymphocyte activation, recirculation and homing and tumor invasion8. These pro-teins can be expressed as several isoforms generated by the alternative splicing of at least 11 of the 21 exons of the CD 44 gene9. There has been interest in the expression of CD 44 as a marker of tumor aggressiveness and

Table 1: Demographic Data of Patients with TCC of the Bladder

* 27 cases were Nx and/or Mx

Table 2: Details of Antibodies Used

tion, excretory urography, bimanual examina-tion under anaesthesia, cystoscopy, ultrasono-graphy, CT and/or MRI together with histolo-gical examination of the removed specimens. The cases were assigned into two grade categories (low and high) according to the cri-teria of the recently revised WHO classifica-tion17. When tumor cells of different grades coexisted, the grade that predominated was scored. In each case the tumor configuration (papillary vs. flat or solid) was also recorded according to the predominant pattern. In 27 cases the nodal and/or distant metastases could not be assessed (NX and/or MX). Patients with nodal and/or distant metastasis were categorized as metastasis-positive group, while patients with N0, M0 were categorized as

invasive to diffuse, invasive, highly aggressive bladder cancer2. The majority of TCCs are either localized non-invasive (pTa) or minimally invasive (pT1) tumors at the time of initial dia-gnosis. More than 50% of patients presenting with superficial papillary urothelial tumors have recurrences, and about 10-30% of these will
metastatic potential in human breast cancer8, gastric cancer10 and urothelial carcinoma2,9. However, the precise function of CD 44 in the metastatic process and the degree of involve-ment in bladder cancer has not yet been fully established9.

CK 20 belongs to the epithelial subgroup of cytoskeleton-associated intermediate fila-ments11. Because of its restricted expression in certain epithelial tissues (gastrointestinal mu-cosa, urothelium and Merkel cells), CK 20 has been suggested as a useful marker in the dia-gnosis of metastatic urothelial tumors12.

Cyclin D1 is a candidate oncogene located on chromosome 11q13. The Cyclin D1 gene product has a function in the regulation of G1/S phase transition of the cell13. Cyclin-D1 over-expression has been shown in a number of human tumors and cell lines14. Recent studies have indicated that Cyclin D1 overexpression occurs in bladder cancer15.

In this study we aim at the analysis of the different immunohistochemical expression forms of some biological markers, namely CD 44, CK 20 and Cyclin D1 in transitional cell carcinoma of the bladder and at a trial to cor-relate the different results of their expression to established clinico-pathologic prognostic fac-tors of TCC of the bladder, such as stage, grade and pattern of growth (configuration) of the tumor and associated metastases.

MATERIAL AND METHODS

A total of 80 bladder specimens were included in the study. Seventy specimens were taken from patients with newly diagnosed primary transitional cell carcinoma of the bladder.presenting to the urology department of Zagazig University Hospitals These spe-cimens were obtained by transurethral resection (n=44) or en-block radical cystic-tomy and pelvic lymphadenectomy (n=26). The other 10 bladder specimens were obtained from patients complaining of other non-malig-nant urological diseases (3 patients with bladder stones, 4 with benign prostatic hyper-plasia, 2 with polypoid cystitis and 1 with accidental bladder trauma), and they were used as control cases. Diagnosis and clinical staging of the bladder tumors were done according to the 1987 TNM staging system16 based on a combination of clinical and histopa-thological data retrieved by clinical examina-
metastasis-negative group. The demographic data of the patients and tumors included in the study are shown in Table 1.

For the immunohistochemical staining 4- µm-thick sections were cut from paraffin blocks and mounted on silane-coated slides (Super Frost/plus microscope slides, Menzel, Ger-many). The slides were dewaxed in xylene and rehydrated in descending grades of alcohol ending in distilled water. The endogenous per-oxidase activity was blocked using 0.6% H2O2 in absolute methanol for 20 minutes at room temperature. Antigen retrieval was accom-plished by immersion of the slides in a 10 mM/l concentration of citrate buffer (pH, 6.0) and heating on high power in a microwave oven (Panasonic, NN-4241) for two cycles of 5 minutes each. The slides were allowed to cool at room temperature for 20 minutes and then washed with distilled water and transferred to Tris buffered saline (TBS) buffer (pH, 7.4). The sections were incubated with normal horse serum (Vectastain, Super ABC kit, Vector labs., Burlingame, California, USA) for 20 minutes as a protein block for non-specific staining. After tapping off the serum, the sections were covered with either of the following monoclonal antibodies (Table 2): a 1µg/ml dilution of CD 44 antibodies (clone: MBH, S1V, NeoMarkers, Vermont, USA), or 3 µg/ml dilution of Cyclin D1 (clone: DC6, NeoMarkers, Vermont, USA), or 1/30 dilution of anti-Cytokeratin 20 (clone: k.r.s20, Vector Labs., Burlingame, California, USA). The monoclonal antibodies were incubated at room temperature for 2 hours except for Cyclin D1 which is incubated overnight at 4° C. After incubation with monoclonal antibodies, they were followed by biotinylated rabbit anti-mouse IgG secondary immunoglobulin for 30 minutes at room temperature. The bound antibodies were visualized with avidin-biotin immuno-peroxidase conjugate for 30 minutes at room temperature. Careful rinses with several changes of TBS (ph, 7.4) were performed between each step of the procedure. The reaction was developed with a solution of 3?,3? diaminobenzidine tetrahydro-chloride (DAB, NeoMarkers, Vermont, USA) for 6 minutes at room temperature in the dark. The sections were rinsed, counterstained with Harris hematoxylin and mounted. Incubation with an isotype-matched control antibody in the place of the primary antibodies was used as a negative control.

* P compares T1 versus T2+T3+T4; NS= not significant

Fig. 1: CD 44 immunohistological staining. A: Diffuse and strong membrane staining in papillary T1 low-grade tumor. B: Focal staining (arrows) in T2 high-grade tumor. C & D: Most cells are non-reactive with occasional positive cells detectable (arrows) in invasive T4 high-grade tumor (original magnification: A: x 100, B & C: x 200, D: x 400)

RESULTS

CD 44

In all 10 cases of normal urothelium, CD 44 showed consistent but patchy membranous staining of only the basal layers. With regard to tumors, there was a statistically significant correlation between a CD 44-loss pattern of staining and an increase of the stage and grade of the tumor (Fig. 1 & 2), as 35% of T1 tumors and 82% of muscle-invasive (T2-T4) tumors showed this loss pattern (P<0.001). Also, 43% of low-grade tumors and 73% of high-grade tumors showed a loss pattern of
Cyclin D1

CD 44 (P<0.05). This loss-pattern significantly correlated to positive metastases (p<0.01) but not to tumor configuration (p> 0.1) (Table 3).

CK 20

In the 10 control cases with a normal urothelium CK 20 demonstrated patchy cyto-plasmic staining of only the superficial (umbrella) cell layer. In the tumor cells, a higher frequency of CK 20 positivity was ob-served with increasing grade and stage (Fig. 2 & 3). 38% of low-grade tumors showed CK 20 positivity, while 78% of high-grade tumors were CK 20-positive (P<0.001). Also, 42% of T1 tumors were CK 20-positive compared to 81% of muscle-invasive tumors (T2-T4) (P<0.001). CK 20 expression did not correlate to the pattern of growth (P>0.1), or to metastases (P>0.5) (Table 3).

We observed a significant inverse relation-ship in staining patterns of CD 44 and CK 20. This was strikingly evident in corresponding areas of same tumors.
Cyclin D1- positivity was not detected in the 10 normal urothelium samples from patients without cancer history. Twenty-four tumors

DISCUSSION

this study, in agreement with previously published data on CD 44 expression2, 9, 23, we found a statistically significant correlation between progressive loss of CD 44 immuno-reactivity and increasing tumor grade, stage and positive metastases. Lipponen et al.23 found a positive correlation of CD 44 to nodal metastases but not to distant metastases. Others have recently shown a significant asso-ciation between CD 44 immunoreactivity and clinical outcome2, 23. The correlation of CD 44 expression with stage and grade can be explained on the grounds that CD 44 plays a definite role in cell-cell and cell-matrix inter-actions8. Thus, its down-regulation would facili-tate loss of cell-cell cohesion, detachment from the basement membrane and subsequent infiltration of the underlying tissues18. It has been stated that the initial rise and subsequent fall of CD 44 expression during tumor pro-gression, which is documented in the present study and has been reported by others9,18, 23, indicates that even in more advanced deeply invasive carcinomas the regulation of this gene can still be affected by the microenvironment surrounding the cell9.

The progressive decrease in CD 44 expres-sion, culminating in its almost complete eli-mination in deeply invasive carcinomas, may prove to be of interest in relation to understanding mechanisms involved in the tumor progression and in the regulation of synthesis of cell surface molecules2. On the other hand, the overexpression of CD 44 in initial stages of tumor progression confirmed in the current study continues to be of practical clinical importance as it may help and facilitate early diagnosis in exfoliated cells resulting in a timely treatment9.

Cytokeratins are widely used as biochemi-cal markers of epithelial differentiation11. The findings of the present study concur with those
recurrence in superficial bladder carcinoma. Am J Pathol 1999, 155:1427-1432.

3. Hukaas S, Daehlin L, Maartman-Moe H, Ulvik NM. The long term outcome in patients with superficial transitional cell carcinoma of the bladder: a single-institutional experience. BJU 1999, 83:957-963.

4. Hong RL, Pu YS, Hsieh TS, Chu J, Lee W. Expression of E-cadherin and exon V6-containing isoforms of CD 44 and their prognostic values in human transitional cell carcinoma. J Urol 1995, 153:2025-2028.

5. Lee R, Droller MJ. The natural history of bladder cancer: implications of therapy. Urol Clin North Am 2000, 27:1-13.

6. Korkolopoulou P, Christodoulou P, Lazaris A et al. Prognostic implications in p16/p Rb pathway in urothelial bladder carcinoma: A multivariate analysis including p53 expression and proliferation markers. Eur Urol 2001, 39:167-177.

7. Knowles MA. Molecular genetics of bladder cancer. Br J Urol 1995, 75:57-66.

8. Bourguignon LY, Gunja-Smith Z, Lida N, Zhu HB, Young LJ, Muller WJ. CD 44v (3,8-10) is involved in cytoskeleton-mediated tumor cell migration and matrix metalloproteinase (MMP-9) association in metastatic breast cancer cells. J Cell Physio 1998, 176:206-215.

9. Sugino T, Gorham H, Yoshida K et al. Progressive loss of CD44 gene expression in invasive bladder cancer. Am J Pathol 1996, 149:873-882.

10. Washington K, Gottfried MR, Telen MJ. Expres-sion of the cell adhesion molecule CD 44 in gastric adenocarcinomas. Hum Pathol 1994, 25:1043-1049.

11. Moll R, Lowe A, Laufer J, Franke WW. Cytokeratin 20 in human carcinomas. A new histodiagnostic marker detected by monoclonal antibodies. Am J Pathol 1992, 140:427-447.

12. Chu P, Wu E, Weiss LM. Cytokeratin 7 and Cyto-keratin 20 expression in epithelial neoplasms. A survey of 435 cases. Mod Pathol 2000, 13:962-972.

13. Inaba T, Matsushime H, Valentine M, Roussel MF, Sherr CJ, Look AT. Genomic organization, chromosomal localization and independent expres-sion of human Cyclin D genes. Genomics 1992, 13:565-574.

14. Resnitzky D, Gossen M, Bujard H, Reed SI. Acceleration of the G1/S phase transition by expression of cyclins D1 and E with an inducible system. Mol Cell Biol 1994, 14:1669-1679.

15. Lee CC, Yamamoto S, Morimura K. Significance of Cyclin D1 overexpression in transitional cell car-cinoma of urinary bladder and its correlation with histopathologic features. Cancer 1997, 79:780.

16. Hermanek P, Sobin LH. UICC-International Union Against Cancer, TNM Classification of Malignant Tumors. 4th ed. Heidelberg:Springer-Verlag, p. 135, 1987.
30. Likkonen T, Lipponen P, Raitanen M et al. Evaluation of p21 and cyclin D1 expression in the progression of superficial bladder cancer. Urol Res 2000, 28:285-292.

All correspondence to be sent to:

Ashraf M.S. Shahin, M.D.
Urology Department
Faculty of Medicine
Zagazig University
Zagazig
Egypt

Ashrf1959@hotmail.com

Patients with non-invasive urothelial neo-plasms are most often treated conservatively and have a protracted clinical course that requires a close clinical follow-up. It is, thus, useful to identify reliable prognostic para-meters that would stratify patients at a higher risk of recurrence and/or progression of the disease18. On the other hand, patients with muscle-invasive disease may benefit from the use of markers that predict a tumor's meta-static potentials and the outcome of therapy21.

Histopathologic staging alone is now reco-gnized as inadequate for the precise charac-
of others18, 24 who have demonstrated that in a normal urothelium, in contrast to CD 44 expression which shows intense membrane staining only of the basal layers, CK 20 is occasionally expressed only in terminally differentiated umbrella cells. Also in this study we observed that CK 20 showed stronger and consistent immunoreactivity in high-grade, less differentiated invasive tumors, thus confirming the results of others24 -26.

In the current study we observed that the inverse staining pattern of CD 44 and CK 20 in the normal urothelium was also maintained in neoplastic lesions. This may be explained on the grounds that CD 44 cytoplasmic tail interacts with underlying cytoskeleton pro-teins27, however, the complete molecular rela-tionship between both markers needs further study at the molecular level that is beyond the scope of the present study.

In this study, the expression of Cyclin D1 was inversely related to tumor stage, grade and positive metastasis, which is at variance with the findings of Osman et al.28, but in accordance with the results of Bringuier et al.29, Lee et al.15 and Liukkonen30. It is worth to be noticed that the Osman et al.28 cohort in-cluded squamous cell carcinoma with bilhar-zias as an aetiological factor and it consisted mainly of muscle-infiltrating tumors. Our results support the results of Wagner et al.31 that sug-gest that the low frequency of Cyclin D1 positivity in advanced stage and high-grade tumors is largely driven by a strong association between Cyclin D1 expression with papillary tumor growth. The significance of this observa-tion in correlation to survival needs to be fur-ther confirmed by others in an extensive work.

It is worth to mention that the discrepancies observed in the results of different studies of biological markers used can be correlated to patient selection bias and lack of standard immunohistochemical assays including differ-ent properties of the antibodies used23.

REFERENCES

1. Kubota Y, Miyamoto H, Noguchi S et al. The loss of retinoblastoma gene in association with c-myc and transforming growth factor-B1 gene expression in human bladder cancer. J Urol 1995, 154:371-374.

2. Toma V, Hauri D, Schmid U et al. Focal loss of CD 44 variant protein expression is related to

17. Epstein JI, Amin MB, Reuter VR, Mostofi FK. Bladder Consensus Conference Committee: The World Health Organization (International Society of Urological Pathology) Consensus Classification of Urothelial (Transitional Cell) Neoplasms of the Urinary Bladder. Am J Surg Pathol 1998, 22:1435-1448.

18. Desai S, Lim SD, Jimenez RE et al. Relationship of cytokeratin 20 and CD 44 protein expression with WHO/ISUP grade in pTa and pT1 papillary urothelial neoplasia. Mod Pathol 2000, 13:1315-1323.

19. Gillet CE, Smith P, Gregory WM et al. Cyclin D1 and prognosis in breast cancer. Int J Cancer 1996, 69:92-99/

20. Suwa Y, Takano Y, Iki M, Asakura T, Noguchi S, Masuda M. Cyclin D1 protein overexpression is related to tumor differentiation, but not to tumor progression or proliferation activity in transitional cell carcinoma of the bladder. J Urol 1998, 160:897-900.

21. Witjess IA, Umbas R, Debruyne FM, Schalken JA. Expression of markers for transitional cell carcinoma in normal bladder mucosa in patients with bladder cancer. J Urol 1995, 154:2185-2189.

22. Sauter G, Mihatsch M. Pussycats and baby tigers: non-invasive (pTa) and minimally invasive (pT1) bladder carcinoma are not the same! J Pathol 1998, 185:339-341.

23. Lipponen P, Aaltoma S, Kosma VM, Ala-Opas M, Eskelinen M. Expression of CD 44 standard and variant-v6 proteins in transitional cell bladder tumours and their relation to prognosis during a long-term follow-up. J Pathol 1998, 186:157-164.

24. McKenney JK, Desai S, Cohen C, Amin M. Discriminatory immunohistochemical staining of urothelial carcinoma in situ and non-neoplastic urothelium. Am J Surg Pathol 2001, 25:1074-1078.

25. Harnden P, Eardley I, Joyce AD, Southgate J. Cytokeratin 20 as an objective marker of urothelial dysplasia. BJU 1996, 78:870-875.

26. Buchuemensky V, Klein A, Zemer R, Kessler OJ, Zimlichman S, Nissenkorn I. Cytokeratin 20: a new marker for early detection of bladder cell carcinoma. J Urol 1998, 160:1971-1974.

27. Lokeshwar VB, Bourguignon LY. Post-translation protein modification and expression of ankyrin-binding site (s) in GP85 (Pgp-1/CD44) and its bio-synthetic precursors during T-lymphoma mem-brane biosynthesis. J Biol Chem 1991, 266:17983-17989.

28. Osman I, Scher H, Zhang Z et al. Expression of cyclin D1, but not cyclins E and A, is related to progression in bilharzial bladder cancer. Clin Cancer Res 1997, 3:2247-2251.

29. Bringuier PP, Tamimi Y, Schuuring E, Schalken J. Expression of cyclin D1 and ENS1 in bladder tumors: relationship with 11q13 amplification. Oncogene 1996, 12:1747-1753.
31. Wagner UR, Suess K, Luginbuhl T, Schmid U, Ackermann D, Jordan P, Sauter G. Cyclin D1 overexpression lacks prognostic significance in superficial urinary bladder cancer. J Pathol 1999, 188:44-50.