Predictive Value of N-Acetylglucosaminyltransferase-V for Superficial Bladder Cancer Recurrence

Predictive Value of N-Acetylglucosaminyltransferase-V for Superficial Bladder Cancer Recurrence Toshiko Takahashi, Shigeru Hagisawa, Kazuyuki Yoshikawa, Fumiaki Tezuka, Mitsuo Kaku and Chikara Ohyama* From the Department of Clinical Laboratory, Tohoku University Hospital, Departments of Urology and Clinical Laboratory, Sendai Medical Center and Department of Infection Control and Laboratory Diagnostic Internal Medicine, Tohoku University Graduate School of Medicine, Sendai and Department of Urology, Hirosaki University School of Medicine, Hirosaki, Japan,

Purpose: GnT-V is an enzyme that catalyzes ␤1– 6 branching of N-acetylglucosamine on asparagine (N)-linked oligosaccharides of cell proteins. GnT-V expression has been closely related to malignant potentials in colon cancer, brain cancer and hepatocellular carcinoma. We determined whether GnT-V expression is predictive of superficial bladder cancer recurrence. Materials and Methods: The cohort comprised 60 consecutive patients with first time superficial bladder cancer treated with transurethral resection. None of the patients received prophylactic intravesical therapy until recurrence. Paraffin embedded tumor specimens were immunohistochemically examined by the avidin-biotin peroxidase method using monoclonal antibody against GnT-V. Kaplan-Meier survival curves were generated to determine disease-free survival. Univariate and multivariate analyses were done to compare GnT-V expression to other clinical and pathological variables. Results: GnT-V expression correlated inversely with tumor grade and stage. The positive incidence of GnT-V in G1 to G3 tumors was 7 of 9 (78%), 21 of 43 (49%) and 3 of 8 (38%), respectively. GnT-V was positive in 26 of 44 cases of pTa (60%) and in 5 of 16 of pT1 (31%) disease. The 31 patients with positive GnT-V expression had significantly higher disease-free survival than the 29 with negative GnT-V expression (log rank test p ⫽ 0.0034). Multivariate analysis revealed that patient age, pT, grade and negative GnT-V expression were independent predictors of recurrence (p ⫽ 0.015, 0.001, 0.019 and 0.011, respectively). Conclusions: Immunohistochemical detection of GnT-V is an independent predictor of superficial bladder cancer recurrence. Key Words: bladder, bladder neoplasms, tumor marker, biological, recurrence, N-acetylglucosaminyltransferases

atients with superficial bladder cancer are at risk for recurrence and progression. Studies indicate that the recurrence rate is 50% to 80% and the risk of progression is 10% to 25%.1 Determining which bladder tumors will recur and progress is a problem of major clinical interest and it could be useful for designing clinical followup and treatment strategies. A number of recent investigations have been performed to determine whether new biological markers can help predict disease recurrence. Among the biological markers of aggressiveness the p53 tumor suppressor gene has been extensively studied.2 However, the relationship between p53 and superficial bladder cancer recurrence is contradictory.3,4 Despite extensive studies focused on p21WAF1,5 Ki-67 antigen6 and vascular endothelial growth factor7 their clinical evidence as a recurrence predictor is still equivocal. Aberrant glycosylation occurs in essentially all types of experimental and human cancers.8 However, glycosylation status in bladder cancer has not been studied extensively

except for the roles of ganglioside GM3 and sialyl Lewis X.9,10 GnT-V, a key enzyme in the formation of ␤1-6 branching of asparagine (N)-linked oligosaccharides, is the one most strongly linked to tumor metastasis in many glycosyltransferases.11 Figure 1 shows the structure of ␤1– 6 branching of asparagine (N)-linked oligosaccharides and synthetic pathway catalyzed by GnT-V. In colorectal cancer and brain tumors GnT-V expression significantly correlates with distant metastasis.12,13 In contrast, hepatocellular carcinoma cases with low or no GnT-V expression are more likely to show recurrence than cases with high expression.14 To our knowledge there have been no reports of the clinical or pathological significance of GnT-V in human bladder cancer. We report that, as defined by its antibody, GnT-V expression is an independent recurrence predictor of superficial bladder cancer, as defined by its antibody.

P

MATERIALS AND METHODS Patients. A total of 60 consecutive patients with first time superficial bladder cancer treated with TUR at the department of urology, Sendai Medical Center, Sendai, Japan between 1996 and 2002 comprised the study cohort. Paraffin specimens obtained by biopsy or TUR were used for immunohistochemistry. After TUR patients were followed by routine cystoscopy and urinary cytology at 3-month interval

Submitted for publication March 3, 2005. Supported by Grant-in-Aid for Scientific Research B-16390459 from the Japan Society for the Promotion of Science and CREST, Japan Science and Technology Agency. * Correspondence: Department of Urology, Hirosaki University School of Medicine, 5 Zaifucyou, Hirosaki 036-8562, Japan (telephone: ⫹81-17239-5092; FAX: ⫹81-172-39-5091; e-mail: [email protected]).

0022-5347/06/1751-0090/0 THE JOURNAL OF UROLOGY® Copyright © 2006 by AMERICAN UROLOGICAL ASSOCIATION

90

Vol. 175, 90-93, January 2006 Printed in U.S.A. DOI:10.1016/S0022-5347(05)00013-3

SUPERFICIAL BLADDER CANCER RECURRENCE

FIG. 1. Structure of ␤1– 6 branching of asparagine (Asn) (N)-linked oligosaccharides and reaction catalyzed by GnT-V. N-acetylglucosamine (GlcNAc) is attached to mannose (Man) residue of asparagine (N)-linked oligosaccharides by GnT-V. Resultant ␤1– 6 branching structure is associated with tumor malignancy. Arrow indicates reaction pathway catalyzed by GnT-V.

until recurrence. None of the patients received prophylactic intravesical therapy until recurrence. Table 1 lists patient clinical and pathological characteristics. Mean followup was 25.5 months (range 3 to 78). Written consent was obtained from all patients. The tumor staging system was based on the American Joint Commission on Cancer staging system.15 Immunohistochemistry. Paraffin embedded tumor specimens were immunohistochemically examined by the avidinbiotin peroxidase method using monoclonal antibody against GnT-V.12 Paraffin embedded samples were cut at 3 ␮m and subjected to hematoxylin and eosin staining, and immunohistochemistry. Immunohistochemical staining was performed as previously described using monoclonal antibody against human GnT-V.12–14 Briefly, deparaffinized specimens were reacted with anti-GnT-V monoclonal antibody as the primary antibody. Anti-mouse Ig antibody conjugated with horseradish peroxidase was used as the secondary antibody and peroxidase activity was visualized with aminoethylcarbazol solution (Nichirei, Tokyo, Japan). A control experiment was done by omitting the primary antibody from the staining procedure and no specific staining was found. Immunostaining results were evaluated while blinded to all clinical data. Based on the staining status of Golgi apparatus specimens with 10% or more positive cancer cells were considered GnT-V positive, as described previously.12–14 Statistical analysis. The chi-square test was used to assess the association of the GnT-V status with clinical and pathological parameters. Recurrence-free survival was evaluated by Kaplan-Meier curves. Differences between the groups were evaluated by the log rank test. We used Cox proportional hazards regression analysis to test the association of GnT-V expression with other clinical and patholog-

91

FIG. 2. Immunohistochemistry of bladder cancer using anti-GnT-V monoclonal antibody. GnT-V was positive in low grade bladder cancer (A) but negative in high grade invasive cancer (B). Since intracellular localization of GnT-V was done Golgi apparatus, GnT-V was clearly detected in Golgi pattern around nuclei. Reduced from ⫻200.

ical variables, including patient sex, age, pathological stage, tumor diameter, single or multiple lesions and tumor grade for the prediction of recurrence. We used the statistical program SPSS 12.0 (SPSS, Chicago, Illinois). RESULTS Figure 2 shows representative microphotographs of immunohistochemistry with anti-GnT-V monoclonal antibody. Since the intracellular localization of GnT-V is the Golgi apparatus, GnT-V was clearly detected in the Golgi pattern by immunohistochemistry. Table 2 lists GnT-V status and pathological parameters. GnT-V expression inversely correlated with tumor grade and stage. The incidence of positive GnT-V expression in bladder cancer was significantly higher in low grade/less invasive cancer than in higher grade/invasive cancer. The positive incidence of GnT-V in G1 to G3 tumors was 7 of 9 (78%), 21 of 43 (49%) and 3 of 8 (38%), respectively. GnT-V was positive in 26 of 44 cases of pTa (60%) and 5 of 16 of pT1 (31%) disease. The 31 patients with positive GnT-V expression had significantly higher disease-free survival than the 29 with negative GnT-V expression (log rank test p ⫽ 0.0034, fig. 3). Univariate analysis showed that male sex, tumor size less than 1.5 cm, grade, pT and negative GnT-V expression were significant predictors of recurrence (p ⫽ 0.009, 0.012, 0.016 and 0.005, respectively, table 3). Multivariate analysis revealed that patient age, pT, grade and negative GnT-V expression were independent predictors of recurrence (p ⫽ 0.015, 0.001, 0.019 and 0.011, respectively, table 3). DISCUSSION After TUR mandatory cystoscopy and urine cytology are required for the routine followup of superficial bladder canTABLE 2. GnT-V status and pathological parameters

TABLE 1. Patient clinical and pathological characteristics Mean age ⫾ SD No. men (%)/women No. tumor grade (%): G1 G2 G3 No. pathological stage (%): pTa pT1

67.7 ⫾ 10.7 47 (78)/13 (22) 9 (15) 43 (72) 8 (13) 44 (73) 16 (27)

No. GnT-V Pos/Total No. (%) Tumor grade: G1 G2 G3 p Value Pathological stage: pTa pT1 p Value

7/9 (78) 21/43 (49) 3/8 (38) 0.021 26/44 (60) 5/16 (31) 0.019

92

SUPERFICIAL BLADDER CANCER RECURRENCE TABLE 3. Cox proportional hazard model for predicting recurrence Analysis Univariate: Age Male sex Multiple diseases Size Pathological stage Grade GnT-V pos Multivariate: Age Male sex Multiple diseases Size Pathological stage Grade GnT-V pos

FIG. 3. Kaplan-Meier curve for recurrence-free survival according to GnT-V staining status. Cohort was divided into 2 groups according to GnT-V staining. Recurrence-free survival in GnT-V positive cases was significantly higher than in negative cases.

cer. Accurate prediction of bladder cancer recurrence could significantly decrease the number of cystoscopies during followup. However, the most important predictors of recurrence are still pathological stage, grade, multiplicity and tumor size.16,17 Although they are the most accurate prognostic factors in the evaluation of superficial bladder tumors, they cannot always predict the true tumor biological potential since superficial tumors of the same stage and grade may have completely different clinical courses. In the current study we observed that histology significantly correlated with GnT-V expression, in that high GnT-V expression was more frequently found in low grade/ pTa cancer than in high grade/pT1 cancer. Immunocytology with monoclonal antibody against Lewis X showed high sensitivity for detecting bladder cancer.18 Although the precise molecular background remains unknown, considering the biochemical function of GnT-V, which produces the precursor of Lewis X, negative GnT-V expression may have some role in bladder cancer recurrence.18 More importantly the current study demonstrates a significant role of GnT-V expression for predicting superficial bladder tumor recurrence. On univariate analysis male sex, tumor size less than 1.5 cm, grade, pT and negative GnT-V expression were significant predictors of recurrence. Multivariate analysis revealed that patient age, pT, grade and negative GnT-V expression were independent predictors of recurrence. Although our cohort was small, the current data suggest considerable value of GnT-V for predicting recurrence after TUR. In colon cancer and brain tumors GnT-V expression is closely related to invasion and metastatic potential.12,13 However, in bladder cancer GnT-V expression was associated with a low risk of recurrence in this study. In hepatocellular carcinoma14 and nonsmall cell lung cancer19 GnT-V expression is associated with a good prognosis, as in superficial bladder cancer. Since GnT-V expression is regulated in a tissue specific manner,20 its clinical implication in various cancers should be defined for each cancer. The biological importance of GnT-V expression for the maintenance of physiological function as well as for the development and progression of cancer may be different in each organ and tissue depending on the biological function of target sub-

HR

95% CI

p Value

1.006 0.351 1.605 1.722 6.693 2.374 2.848

0.974–1.038 0.161–0.767 0.803–3.205 1.129–2.626 3.113–14.389 1.175–4.797 1.372–5.914

0.733 0.009 0.180 0.012 0.020 0.016 0.005

1.053 0.644 2.055 1.410 5.464 2.944 0.322

1.010–1.098 0.278–1.493 0.943–4.481 0.793–2.507 2.477–12.056 1.198–7.232 0.135–0.768

0.015 0.305 0.070 0.242 ⬍0.001 0.019 0.011

strate glycoproteins, which can vary among organs and tissues. Decreased GnT-V expression may contribute to altered biological properties of bladder cancer by decreased the synthesis of the ␤1– 6 branching oligosaccharides of certain target glycoproteins, resulting in shorter survival in patients who have tumors with low GnT-V expression compared with those who have tumors with high GnT-V expression. To our knowledge target glycoproteins of GnT-V in the bladder mucosa remain to be determined. Another important clinical concern in prognostic feature of superficial bladder tumor is progression to invasive tumor. Our studies are in progress to evaluate whether GnT-V expression can predict the progression of superficial bladder tumors. CONCLUSIONS These data suggest that immunohistochemical detection of GnT-V can be a reliable marker for predicting superficial bladder cancer recurrence. It may be relevant for determining the followup schedule and planning adjuvant treatment. ACKNOWLEDGMENT Monoclonal antibody against GnT-V was provided by Dr. Naoyuki Taniguchi, Osaka University, Osaka, Japan.

Abbreviations and Acronyms GnT-V ⫽ N-acetylglucosaminyltransferase-V TUR ⫽ transurethral resection

REFERENCES 1. Heney, N. M., Ahmed, S., Flanagan, M. J., Frable, W., Corder, M. P., Hafermann, M. D. et al: Superficial bladder cancer progression and recurrence. J Urol, 130: 1083, 1983 2. Cordon-Cardo, C. and Reuter, V. E.: Alteration of tumor suppressor gene in bladder cancer. Semin Diagn Pathol, 14: 123, 1997 3. Esrig, D., Elmajian, D., Groshen, S., Freeman, J. A., Stein, J. P., Chen, S. C. et al: Accumulation of nuclear p53 and tumor progression in bladder cancer. N Engl J Med, 331: 1259, 1994

SUPERFICIAL BLADDER CANCER RECURRENCE 4. Zlotta, A. R., Noel, J.-C., Fayt, I. Drowart, A., Van Vooren, J.-P., Huygen, K. et al: Correlation and prognostic significance of p53, p21WAF/1CIP1 and Ki-67 expression in patients with superficial bladder tumors treated with bacillus CalmetteGuerin intravesical therapy. J Urol, 161: 792, 1999 5. Lipponen, P., Aaltomaa, S., Eskelinen, M., Ala-Opas, M. and Kosma, V. M.: Expression of p21(waf1/cip1) protein in transitional cell bladder tumours and its prognostic value. Eur Urol, 34: 237, 1998 6. Stavropoulos, N. E., Ioackim-Velogianni, E., Hastazeris, K., Kitsiou, E., Stefanaki, S. and Agnantis, N.: Growth fractions in bladder cancer defined by Ki67: association with cancer grade, category and recurrence rate of superficial lesions. Br J Urol, 72: 736, 1993 7. Crew, J. P., O’Brien, T., Bradburn, M., Fuggle, S., Bicknell, R., Cranston, D. et al: Vascular endothelial growth factor is a predictor of relapse and stage progression in superficial bladder cancer. Cancer Res, 57: 5281, 1997 8. Hakomori, S.: Glycosylation defining cancer malignancy new wine in an old bottle. Proc Natl Acad Sci USA, 99: 10231, 2002 9. Numahata, K., Satoh, M., Handa, K., Saito, S., Ohyama, C., Ito, A. et al: Sialosyl-Le(x) expression defines invasive and metastatic properties of bladder carcinoma. Cancer, 94: 673, 2002 10. Watanabe, R., Ohyama, C., Aoki, H., Takahashi, T., Satoh, M., Saito, S. et al: Ganglioside G(M3) overexpression induces apoptosis and reduces malignant potential in murine bladder cancer. Cancer Res, 62: 3850, 2002 11. Dennis, J. W., Laferte, S., Waghorne, C., Breitman, M. L. and Kerbel, R. S.: Beta 1– 6 branching of Asn-linked oligosaccharides is directly associated with metastasis. Science, 236: 582, 1987 12. Murata, K., Miyoshi, E., Kameyama, M., Ishikawa, O., Kabuto, T., Sasaki, Y. et al: Expression of N-acetylglucosaminyltransferase V in colorectal cancer correlates with metastasis and poor prognosis. Clin Cancer Res, 6: 1772, 2000 13. Yamamoto, H., Swoger, J., Greene, S., Saito, T., Hurh, J., Sweeley, C. et al: Beta1,6-N-acetylglucosamine-bearing N-glycans in human gliomas implications for a role in regulating invasivity. Cancer Res, 60: 134, 2000 14. Ito, Y., Miyoshi, E., Sakon, M., Takeda, T., Noda, K., Tsujimoto, M. et al: Elevated expression of UDP-N-acetylglucosamine alphamannoside beta1,6 N-acetylglucosaminyltransferase is an early event in hepatocarcinogenesis. Int J Cancer, 91: 631, 2001 15. Greene, F. L., Page, D. L., Fleming, I. D., Fritz, A., Balch, C. M., Haller, D. G. et al: AJCC Cancer Staging Manual, 6th ed. New York: Springer-Verlag, pp. 317-322, 2002

93

16. Millán-Rodríguez, F., Chéchille-Toniolo, G., Salvador-Bayarri, J., Palou, J. and Vicente-Rodríguez, J.: Multivariate analysis of the prognostic factors of primary superficial bladder cancer. J Urol, 163: 73, 2000 17. Holmang, S., Hedelin, H., Anderstrom, C., Holmberg, E., Busch, C. and Johansson, S. L.: Recurrence and progression in low grade papillary urothelial tumors. J Urol, 162: 702, 1999 18. Friedrich, M. G., Hellstern, A., Hautmann, S. H., Graefen, M., Conrad, S., Huland, E. et al: Clinical use of urinary markers for the detection and prognosis of bladder carcinoma: a comparison of immunocytology with monoclonal antibodies against Lewis X and 486p3/12 with the BTA STAT and NMP22 tests. J Urol, 168: 470, 2002 19. Dosaka-Akita, H., Miyoshi, E., Suzuki, O., Itoh, T., Katoh, H. and Taniguchi, N.: Expression of N-acetylglucosaminyltransferase v is associated with prognosis and histology in nonsmall cell lung cancers. Clin Cancer Res, 10: 1773, 2004 20. Perng, G. S., Shoreibah, M., Margitich, I., Pierce, M. and Fregien, N.: Expression of N-acetylglucosaminyltransferase V mRNA in mammalian tissues and cell lines. Glycobiology, 4: 867, 1994

EDITORIAL COMMENT Growth factor receptors, cell adhesion molecules and extracellular matrix components, which have a pivotal role in tumor growth and metastasis, are glycosylated. In glycoproteins, oligosaccharides may be N-linked (to asparagine) or O-linked (to serine/threonine) and they are necessary for glycoprotein function. The over expression of GnT-V, which is involved in the branching of N-linked oligosaccharides, correlates with distant metastasis in colorectal cancer and brain tumors. Contrary to these findings, the authors report that in bladder cancer, GnT-V expression inversely correlates with tumor grade and stage. Furthermore, positive GnT-V expression correlates with increased disease-free survival and its negative expression is an independent predictor of tumor recurrence. Although these results need confirmation using a larger number of tissues, especially low grade tumors, the study demonstrates that in bladder cancer GnT-V may increase the N-linked glycosylation of glycoproteins that function as negative regulators of tumor progression and recurrence. GnT-V may also be a prognostic indicator for bladder cancer recurrence and survival. Vinata B. Lokeshwar Department of Urology and Cell Biology and Anatomy University of Miami Miami, Florida