THE PREDICTIVE VALUE OF MUSCULARIS MUCOSAE INVASION AND p53 OVER EXPRESSION ON PROGRESSION OF STAGE T1 BLADDER CARCINOMA

0022-5347/01/1651-0042/0 THE JOURNAL OF UROLOGY® Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 165, 42– 46, January 2001 Printed in U.S.A.

THE PREDICTIVE VALUE OF MUSCULARIS MUCOSAE INVASION AND p53 OVER EXPRESSION ON PROGRESSION OF STAGE T1 BLADDER CARCINOMA S. BERNARDINI,* C. BILLEREY, M. MARTIN, G. L. ADESSI, H. WALLERAND

AND

H. BITTARD

From the Departments of Molecular Oncology-Endocrinology, Anatomy and Pathology, and Urology, Centre Hospitalo-Universitaire, Besanc¸on, France

ABSTRACT

Purpose: We determine the significance of muscularis mucosae invasion and nuclear p53 over expression on the progression of stage T1 transitional cell bladder cancer. Materials and Methods: The pathological findings in 149 cases of T1 tumors diagnosed between 1973 and 1996 were reviewed. Diagnosis was stage T1 in 94 tumors in which the muscular layer was clearly identifiable and disease-free. Mean followup was 64.9 months (range 5 to 288). T1 bladder cancers were subclassified into 2 groups, with (T1b) or without (T1a) muscularis mucosae invasion. The p53 nuclear antibody immunoreactivity was determined with antibody D07 and a cutoff point at 15%. Results: T1 subclassification was possible in all 94 patients. Of all tumors 37.2% expressed p53 nuclear over expression. Univariate statistical analysis showed that p53 expression (p ⬍0.05) and tumor invasion depth (p ⬍0.001) significantly correlated with progression. However, on multivariate analysis only invasion depth (p ⬍0.0001) and associated carcinoma in situ (p ⬍0.03) remained independently significant as predictors of progression. Conclusions: In our study the depth of tumor invasion was a significant independent predictor of progression in patients with T1 bladder cancer. This result suggests that the depth of invasion in stage T1 should be included in the histopathological report. KEY WORDS: bladder neoplasms, protein p53, disease progression

sion did not appear to have a better predictive value than stage. However, available followup data were insufficient to draw a definitive conclusion from the prognostic role of p53 over expression, particularly in patients with T1 bladder tumors. Little information is available in regard to clinical usefulness of adopting routine pathological assessment of the level of lamina propria invasion in patients with stage T1 bladder cancer. However, it has been suggested that the level of lamina propria invasion could be a clinically relevant prognostic factor for progression of T1 bladder cancer.11–13 In the current study we evaluate p53 over expression using an immunohistochemical method that is highly sensitive and specific as a potential prognostic marker for progression in a cohort of 94 patients with T1 bladder cancer. We then determined the significance of muscularis mucosae invasion on T1 bladder cancer progression and whether this morphological feature can be potentially useful in daily clinical practice.

Many parameters have been evaluated to find prognostic factors for patients with T1 bladder tumors. Previous studies have shown a correlation between p53 over expression, determined by immunohistochemistry, and disease progression in T1 bladder tumors.1, 2 However, in other studies alterations of p53 were reported as having no prognostic value in T1 bladder tumors.3, 4 These discrepancies may be due to the fact that some authors interpret the result as positive if more than 10% of the cell nuclei is immunostained,5, 6 whereas others define positive as more than 20% of stained nuclei.1, 2 Another methodological problem is the various technical parameters used in different reports for the immunohistochemical assessment of p53 protein.7 When p53 over expression is analyzed with different monoclonal antibodies, some variability in the immunostaining can be observed depending on the antibody used.8 Furthermore, the intensity of staining may influence interpretation of results.5 Finally, molecular genetic disparity among the tumor population may also explain the differences in these results.9 Consequently, it seems desirable that a highly sensitive and specific immunohistochemical method to detect p53 alterations be standardized to compare results from different reports. In a previous study we compared immunohistochemical staining using the antibody D07 with DNA sequencing in 104 transitional cell carcinomas of the bladder, and a highly significant association between the presence of p53 mutation and accumulation of the mutated p53 protein was found.10 The higher values for sensitivity and specificity of immunohistochemical determination relative to p53 mutation were obtained with a cutoff of 15%. In our study p53 over expres-

MATERIALS AND METHODS

Patient population. A total of 149 primary stage T1 tumors diagnosed from 1973 to 1996 were reviewed to determine invasion level and p53 protein accumulation. Diagnosis was stage T1 disease in 94 cases when the muscular layer was clearly visible and disease-free. The muscular layer was not clearly identifiable in 55 patients and they were excluded from study. Mean patient age was 68.9 years (range 42 to 90), and mean followup was 64.9 months (range 5 to 288). All patients were followed at the Urological Department of the Medical School of Besanc¸on, France. Tissues were obtained from endoscopic resection in all cases and from partial or total cystectomy in 18. Specimens were immediately fixed in 10% formalin and subsequently embedded in paraffin. All specimens were reviewed by the

Accepted for publication July 21, 2000. Supported by grants from INSERM, PHRC (Ministe`re de l’Emploi et de la Solidarite´) and Ligue Nationale contre le Cancer. * Requests for reprints: Service d’Urologie, Hoˆpital Saint Jacques, 2 place Saint Jacques, 25000 Besanc¸on, France. 42

43

EFFECT OF MUSCULARIS MUCOSAE AND p53 ON STAGE T1 BLADDER CANCER

same pathologist who was unaware of final outcomes. Tumors with invasion of the connective tissue superficial to the level of the muscularis mucosae were classified as T1a, and those with invasion of the muscularis mucosae were classified as T1b. Tumors were evaluated using WHO grades 1, 2 and 3. Vascular invasion and carcinoma in situ were systematically sought in the tumor, deep muscle and mucosal biopsies of 93 and 94 tumors, respectively. Among the 94 tumors 72 were analyzed with flow cytometry after mechanical disintegration to measure DNA ploidy.14 Mitotic activity was measured in 92 tumors using the M/V INDEX method, which was originally described by Haapasalo et al.15 The cutoff point was 10 mm. or less.2 Treatment and evaluation. Before 1993 at our urological department transurethral resection was the only standardized treatment for stage T1 bladder cancer. In addition to transurethral resection, radical cystectomy was used in cases with large (greater than 5 cm.) or multifocal tumors of high grade (3) and carcinoma in situ. Patients with a large (2 to 5 cm.), grade 3 unifocal tumor in a mobile part of the bladder but no carcinoma in situ were treated with partial cystectomy and interstitial 192iridium radiotherapy in addition to transurethral resection. After 1993 in addition to transurethral resection bacillus Calmette-Guerin (BCG) therapy was systematically administered to patients with grade 3 tumor and carcinoma in situ. Thus, among the 94 patients transurethral resection was only performed in 68, while 8 also received intravesical instillations of BCG, 4 underwent total cystectomy and 14 underwent partial cystectomy with interstitial 192iridium radiotherapy. Every patient was evaluated at the end of followup. Disease-free survival was defined as the absence of bladder tumor and metastases, and disease progression was defined as progression to a higher tumor stage, metastases or death from bladder tumor. Immunohistochemical staining for p53. Immunohistochemical staining was performed as described previously.5 For each paraffin embedded specimen 5 ␮m. sections were deparaffinized and rehydrated. Pretreatment microwave heating in 10 mM. citrate buffer, pH 6.0, was performed before incubation with the primary antibody. The D07 antibody was applied at 1/50 dilution for 45 minutes at room temperature and revealed with the alkaline phosphatase, anti-alkaline phosphatase method. A complementary amplification was performed using the antigen retrieval method with steps 2 and 3 lasting 10 minutes each. The chromogen was fast red, and sections were lightly counterstained in hematoxylin. Negative controls consisted of tumor sections with voluntary omission of the primary antibody and positive controls were strongly positive slides of urothelial carcinoma. The percentage of tumor cells with red staining of the nuclei was evaluated by 1 member of the team blinded to final outcomes. Based on our previous report tumors were classified as immunohistochemically positive for p53 protein over expression if more than 15% of neoplastic cells contained the reaction product in the nuclei.10 Statistical analysis. Correlations among immunohistochemical detection of p53 protein over expression, invasion depth, grade, vascular invasion, carcinoma in situ, DNA ploidy and mitotic index were evaluated by the Fisher exact test. Progression curves according to the Kaplan-Meier method were used for univariate analysis of prognostic parameters. The Kaplan-Meier method was used to derive the progression-free function, while the log-rank test was used to compare curves for 2 or more groups. The Cox proportional hazards model was used for multivariate analysis of prognostic parameters and to estimate relative risk. In particular, stepwise regression (maximum partial likelihood ratio method) was used to select independent prognostic factors.

RESULTS

Incidence of p53 protein over expression. Of the 94 tumors 35 (37.2%) were positive and 59 (62.7%) negative. No cytoplasmic staining was observed in tumor cells. In all analyzed samples normal urothelial cells, fibroblasts, vessel endothelium and smooth muscles, and inflammatory elements, including lymphocytes, were unreactive to p53 antibody. Depth of tumor invasion. Invasion depth was identified in all 94 patients, including 54 with stage T1a and 40 with stage T1b (42.5%) tumor (57.4%). A correlation was observed in all patients treated with radical (4) or partial (14) cystectomy between the clinical staging at transurethral resection and final pathological finding. The frequency by which the muscularis mucosae was identified either partially or totally was about 70%. When the level of muscularis mucosae was not identified, large caliber arteries in the lamina propria running parallel to the surface were used as the landmark instead of muscularis mucosae as previously reported.16, 17 Relationship between p53 over expression, and standard clinical and pathological variables. The relationship between p53 over expression, and standard and pathological variables is reported in table 1. Over expression of p53 nuclear was identified in 35 patients, with a higher prevalence in stage T1b occurring in 21 of 40 cases (52.5%) compared with stage T1a in 14 of 54 cases (25.9%) (p ⬍0.02). It was observed in 50% of grade 3 versus 22.7% of grades 1 and 2 tumors, which was statistically significant (p ⬍0.01). Over expression of p53 protein was also associated with DNA aneuploidy (p ⬍0.05), carcinoma in situ (p ⬍0.001) and mitotic index (p ⬍0.01). No significant difference was observed between p53 protein over expression and vascular invasion. Relationship between invasion depth, and standard clinical and pathological variables. The relationship between invasion depth, and standard clinical and pathological variables is reported in table 2. The association between the level of lamina propria invasion and high tumor grade was statistically significant (54% grade 3 versus 29.5% grades 1 and 2, p ⬍0.03). The level of lamina propria invasion also correlated with DNA aneuploidy (p ⬍0.05). However, no significant difference was observed between the level of invasion and carcinoma in situ, and vascular invasion and mitotic index. Univariate analysis. At the end of followup 41 of the 94 patients had disease progression, 35 died of cancer and 50 were disease-free. When considered individually, invasion

TABLE 1. Clinicopathological profile of bladder carcinoma and p53 alterations Total No.

No. p53 Over Expression* 15% or Less

Invasion depth: pT1A 54 pT1B 40 Grade: 1 6 2 38 3 50 Ploidy: Diploid 24 Aneuploid 48 Not done 22 Mitotic index: 10 or Less 11 Greater than 10 81 Not done 2 Vascular invasion: Absent 81 Present 12 Not done 1 Ca in situ: Absent 41 Present 53 * Positive if more than 15% nuclei were

Greater Than 15%

40 19

14 21

4 30 25

2 8 25

20 28

4 20

11 48

0 33

54 4

27 8

34 25 stained.

7 28

44

EFFECT OF MUSCULARIS MUCOSAE AND p53 ON STAGE T1 BLADDER CANCER

TABLE 2. Clinicopathological profile of bladder carcinoma and invasion depth Total No. p53 Over expression:* p53⫺ p53⫹ Grade: 1 2 3 Ploidy: Diploid Aneuploid Not done Mitotic index: 10 or Less Greater than 10 Not done Vascular invasion: Absent Present Not done Ca in situ: Absent Present * Positive if more than 15% nuclei

Invasion Depth pT1A

pT1B

59 35

40 14

19 21

6 38 50

4 27 23

2 11 27

24 48 22

18 24

6 24

11 81 2

9 45

2 36

81 12 1

50 4

31 8

41 53 were stained.

26 28

15 25

depth (p ⬍0.0001), p53 alterations (p ⬍0.03), grade (p ⬍0.05), vascular invasion (p ⬍0.05) and associated carcinoma in situ (p ⬍0.02) were discriminative in predicting progression. However, DNA aneuploidy and mitotic index were not significantly associated with progression. An additional analysis was performed on a patient subgroup treated with transurethral resection only, and invasion depth (p ⬍0.0001), p53 alterations (p ⬍0.0002), grade (p ⬍0.004), vascular invasion (p ⬍0.03) and associated carcinoma in situ (p ⬍0.0003) were also discriminative in predicting progression (see figure). Of 8 patients, including 4 with T1a and 4 with T1b disease, treated with BCG after transurethral resection 1 with stage T1b had progression to deep muscle invasion after 19 months. All 4 patients treated with radical cystectomy, including 1 with T1a and 3 with T1b disease, were alive without evidence of disease after a mean followup of 63.2 months (range 35 to 84). Of the 14 patients, including 9 with T1a and 5 with T1b disease, treated with partial cystectomy and interstitial 192iridium radiotherapy 4 with stage T1b and 2 with stage T1a had progression to deep muscle invasion after a mean followup of 84.6 months (range 6 to 88). Multivariate analysis. To determine the most informative combination of independent factor prognosis, variables identified as statistically significant in predicting progression, including level of invasion, p53 status, grade, vascular invasion and associated carcinoma in situ, were subjected to multivariate analysis using the stepwise Cox proportional hazards regression. On multivariate analysis only level of

invasion (p ⬍0.0001) and associated carcinoma in situ (p ⬍0.03) remained independently significant as progression predictors. In the patient subgroup treated with transurethral resection only level of invasion (p ⬍0.0001) and associated carcinoma in situ (p ⬍0.0001) also remained independently significant as progression predictors (table 3). Based on the multivariate analysis performed on this subgroup, patients with stage T1b and carcinoma in situ had a risk of progression increased by a factor of 7.5 compared with those with stage T1a and carcinoma in situ (p ⬍0.0001). DISCUSSION

We evaluated p53 over expression in a cohort of 94 patients with T1 bladder cancer using an immunohistochemical method that is highly sensitive and specific to detect p53 alterations. Over expression of p53 protein occurred in 62.7% of patients. This high ratio is close to the rate observed by some authors of positive staining D07 in T1 tumors.3, 6 These results confirm our previous observation in regard to standard clinicopathological parameters in which p53 over expression frequently occurs in high grade disease and carcinoma in situ.10 Nuclear p53 over expression was also identified with a higher prevalence in stage T1b compared with T1a disease, as previously reported by Hermann et al.12 In our present study p53 over expression was a significant predictor of tumor progression on univariate analysis consistent with previous reports1, 2 but lost its significance on multivariate analysis. The existence of a muscularis mucosae in the human bladder was described by Dixon and Gosling in 1983.16 Muscularis mucosae varies from a continuous identifiable layer to a dispersion of thin wisps of smooth muscle cells, which is the most common form. Several groups have studied the value of assessing the level of invasion in stage T1 bladder cancer.18 –20 It has been proposed that the muscularis mucosae could be used to differentiate levels of invasion by T1a— invasion of connective tissue superficial to the level of the muscularis mucosae, T1b—invasion to the level of the muscularis mucosae and T1c—invasion through the level of the muscularis mucosae but superficial to the muscularis propriae.18 Angulo et al proposed that only 2 populations should be differentiated. Tumors superficial to the muscularis mucosae or reaching the muscularis mucosae but not passing through it were classified as T1a, and tumors passing through the muscularis mucosae and invading the submucosae were considered to be T1b.19 In our study T1 tumors were subclassified as with (T1b) or without (T1a) muscularis mucosae invasion. The muscularis mucosae was either partial or total in approximately 70% of biopsy specimens. This proportion appears to agree with what has been previously described in the literature.16, 18, 20 In other cases large arteries in the lamina propria running parallel to the surface were used as the landmark instead of

Kaplan-Meier, progression-free interval according to p53 status (A) and stage (B) in 68 patients with pT1 tumor treated with resection only. All observed differences were significant (log-rank p ⫽ 0.0002 for p53 and p ⬍0.0001 for stage).

EFFECT OF MUSCULARIS MUCOSAE AND p53 ON STAGE T1 BLADDER CANCER TABLE 3. Multivariate, progression-free interval analysis of 68 patients with pT1 bladder cancer treated with only resection Tested Variable

p Value*

␤ ⫾ SE†

Stage ⬍0.0001 2,012 ⫾ 411 Ca in situ ⬍0.0001 1,686 ⫾ 429 * Variables with p ⬍0.05 were independent predictors of progression-free interval, and grade, vascular invasion and p53 status were not significant parameters. † Estimated regression coefficient that indicates the relative effect of each variable of progression function.

muscularis mucosae as previously reported.16, 17 Angulo et al performed a similar retrospective analysis and found that categorization into T1a or T1b could be performed only in 98 of 170 (58%) specimens from transurethral resection.19 In our present study it was possible to determine whether cancer invasion to the level of the muscularis mucosae was present or absent in all cases. This high ratio is in accordance with what has been reported by Holmäng et al11 and more recently by Hermann et al.12 These authors found that depth of invasion could be evaluated in 94% and 100%, respectively, of the cases analyzed. It appears that the muscularis mucosae can be identified in the majority of cases by careful examination and minimal cauterizing artifacts when transurethral resection is performed. Pathological under staging by transurethral resection has been reported to occur in a significant number of patients with T1 bladder cancer.21, 22 Hermann et al12 found muscle invasive disease in 25% of patients with T1 bladder cancer, including a higher rate in tumors that invaded the muscularis mucosae. We excluded from study tumors diagnosed initially as T1 in which the muscular layer was not clearly identifiable. Moreover, we observed a correlation in the 18 patients treated with radical or partial cystectomy between the clinical staging at transurethral resection and the final pathological finding. Thus, we believe that in our study the under staging error was minimal. On multivariate analysis muscularis mucosae invasion was the most significant factor correlated with prognosis. The influence of the depth of lamina propria invasion on prognosis has been studied previously, and most found that the extent of lamina propria invasion is a clinically relevant prognostic factor for progression.11–13, 18, 20, 23 In a study of 70 patients with stage T1 bladder cancers Platz et al found no difference in 10-year survival rates between those with stage T1b and T1a disease.24 An explanation for this discrepancy may be that tissue had been obtained by various urologists and processed at multiple laboratories, resulting in noticeable variation in quality of material. To our knowledge only 1 report in the literature has assessed the relationship between nuclear accumulation of p53 and depth of lamina propria invasion on prognosis.12 The level of lamina propria was the only independent predictor of survival and p53 over expression did not have independent predictive value, although it did correlate with survival on univariate analysis. Our results support these observations. In addition, carcinoma in situ provided independent significant prognostic information, which supports the findings of others in regard to the prognostic value of carcinoma in situ associated with T1 bladder cancer.13, 25 On univariate analysis histological grade was related to progression but lost its prognostic impact on multivariate analysis. In our study patients with stage T1b and carcinoma in situ treated only with transurethral resection had a risk of progression increased by a factor of 7.5 compared to those with stage T1a and carcinoma in situ. This result suggests that patients with stage T1b should be treated more aggressively and supports adoption of a routine assessment of this substaging in patients with stage T1 bladder cancer. Currently, there is no consensus as to whether patients with stage T1b bladder

45

cancer should be treated differently than those with stage T1a disease. The number of patients in our study that underwent complementary treatment after transurethral resection was too small for statistical comparisons. However, prospective studies with a larger number of patients are needed to define the optimal therapy for patients with stage T1b tumors. CONCLUSIONS

T1 bladder cancer invading the muscularis mucosae has a high risk of disease progression after transurethral resection and is a more significant predictor of progression than p53 over expression. Consequently, our results suggest that histopathological reports should specify the level of infiltration of muscularis mucosae in T1 bladder tumors in view of the prognostic and therapeutic implications. Further studies are needed to determine optimal therapy for patients with stage T1b tumors. REFERENCES

1. Sarkis, A. S., Dalbagni, G., Cordon-Cardo, C. et al: Nuclear overexpression of p53 protein in transitional cell carcinoma: a marker for disease progression. J Natl Cancer Inst, 85: 53, 1993 2. Serth, J., Kuczyk, M. A., Bokemeyer, C. et al: p53 immunohistochemistry as an independent prognostic factor for superficial transitional cell carcinoma of the bladder. Br J Cancer, 71: 201, 1995 3. Gardiner, R. A., Walsh, M. D., Allen, V. et al: Immunohistological expression of p53 in primary pT1 transitional cell bladder cancer in relation to tumour progression. Br J Urol, 73: 526, 1994 4. Vatne, V., Maartmann-Moe, H. and Hoestmark, J.: Flow cytometric DNA and p53 analysis in superficially infiltrating bladder carcinoma. Anticancer Res, 14: 2735, 1994 5. Esrig, D., Spruck, C. H., Nichols, P. W. et al: p53 nuclear protein accumulation correlates with mutations in the p53 gene, tumor grade, and stage in bladder cancer. Am J Pathol, 143: 1389, 1993 6. Pages, F., Flam, T. A., Vieillefond, A. et al: p53 status does not predict initial clinical response to bacillus Calmette-Guerin intravesical therapy in T1 bladder tumors. J Urol, 159: 1079, 1998 7. Hall, P. A. and Lane, D. P.: p53 in tumour pathology: can we trust immunohistochemistry? Revisited!. J Pathol, 172: 1, 1994 8. Jacquemier, J. L., Molès, J. P., Penault-Llorca, F. et al: p53 immunohistochemical analysis in breast cancer with four monoclonal antibodies: comparison of staining and PCR-SSCP results. Br J Cancer, 69: 846, 1994 9. Greenblatt, M. S., Bennett, W. P., Hollstein, M. et al: Mutations in the p53 tumor suppressor gene: clues to cancer etiology and molecular pathogenesis. Cancer Res, 54: 4855, 1994 10. Bernardini, S., Adessi, G.-L., Billerey, C. et al: Immunohistochemical detection of p53 protein overexpression versus gene sequencing in urinary bladder carcinomas. J Urol, 162: 1496, 1999 11. Holma¨ng, S., Hedelin, H., Anderstro¨m, C. et al: The importance of the depth of invasion in stage T1 bladder carcinoma: a prospective cohort study. J Urol, 157: 800, 1997 12. Hermann, G. G., Horn, T. and Steven, K.: The influence of the level of lamina propria invasion and the prevalence of p53 nuclear accumulation on survival in stage T1 transitional cell bladder cancer. J Urol, 159: 91, 1998 13. Smits, G., Schaafsma, E., Kiemeney, L. et al: Microstaging of pT1 transitional cell carcinoma of the bladder: identification of subgroups with distinct risks of progression. Urology, 52: 1009, 1998 14. Bittard, H., Lamy, B. and Billerey, C.: Clinical evaluation of cell deoxyribonucleic acid content measured by flow cytometry in bladder cancer. J Urol, 155: 1887, 1996 15. Haapasalo, H., Pesonen, E. and Collan, Y.: Volume corrected mitotic index (M/V INDEX). The standard of mitotic activity in neoplasms. Pathol Res Pract, 185: 551, 1989 16. Dixon, J. S. and Gosling, J. A.: Histology and fine structure of the

46

17. 18.

19. 20. 21.

22. 23. 24. 25.

EFFECT OF MUSCULARIS MUCOSAE AND p53 ON STAGE T1 BLADDER CANCER muscularis mucosae of the human urinary bladder. J Anat, 136: 265, 1983 Ro, J. Y., Ayalla, A. G. and El-Naggar, A.: Muscularis mucosa of urinary bladder. Importance for staging and treatment. Am J Surg Pathol, 11: 668, 1987 Younes, M., Sussman, J. and True, L. D.: The usefulness of the level of the muscularis mucosae in the staging of invasive transitional cell carcinoma of the urinary bladder. Cancer, 66: 543, 1990 Angulo, J. C., Lopez, J. I., Grignon, D. J. et al: Muscularis mucosa differentiates two populations with different prognosis in stage T1 bladder cancer. Urology, 45: 47, 1995 Hasui, Y., Osada, Y., Kitada, S. et al: Significance of invasion to the muscularis mucosae on the progression of superficial bladder cancer. Urology, 43: 782, 1994 Pagano, F., Bassi, P., Galetti, T. P. et al: Results of contemporary radical cystectomy for invasive bladder cancer: a clinicopathological study with emphasis on the inadequacy of the tumor, nodes and metastases classification. J Urol, 145: 45, 1991 Amling, C. L., Thrasher, J. B., Frazier, H. A. et al: Radical cystectomy for stage Ta, Tis and T1 transitional cell carcinoma of the bladder. J Urol, 151: 31, 1994 Cheng, L., Neumann, R. M., Weaver, A. L. et al: Predicting cancer progression in patients with stage T1 bladder carcinoma. J Clin Oncol, 17: 3182, 1999 Platz, C. E., Cohen, M. B., Jones, M. P. et al: Is microstaging of early invasive cancer of the urinary bladder possible or useful?. Mod Pathol, 9: 1035, 1996 Milla´n-Rodrı´guez, F., Che´chile-Toniolo, G., Salvador-Bayarri, J. et al: Multivariate analysis of the prognostic factors of primary superficial bladder cancer. J Urol, 163: 73, 2000 EDITORIAL COMMENT

How best to make treatment decisions for T1 bladder cancer? The stakes are high. Conservative management in the face of biologically aggressive disease may result in death from disease, and radical cystectomy for superficial disease with a benign biological phenotype is surgical overkill. Thus, clinicians grasp eagerly at the straws of biological markers that could provide a rational guide to management. The p53 data at this point are conflicting. While early reports of p53 over expression reflecting mutations in T1 cancer suggested that it was profoundly predictive of progression, subsequent reports have been mixed. Of 7 reports in the last 2 years 4 have been positive and 3 negative,1–7 and 3 additional reports evaluating the value of p53 in predicting BCG response have been mixed, with 1 positive and 2 negative (reference 6 in article).8, 9 Interestingly, in the studies in which multivariate analysis comparing p53 to a number of other pathological markers is used p53 tends to wash out as a predictor. Bernardini et al report a retrospective immunohistological study of 94 patients with T1 bladder cancer in whom the predictive value of p53 expression and muscularis mucosae invasion were evaluated. The study showed that muscularis mucosal invasion was highly predictive and p53 expression did not have independent predictive value, although it did correlate with progression. What is the explanation for this variability in results? It may be related to technical considerations. These authors used a cutoff of 15% immunoreactivity, while others have used a cutoff of 25% or higher. In addition, immunohistochemistry requires interpretation, and this may vary between centers. Mutations of p53 may be present in the absence of p53 accumulation, and functionally normal p53

may accumulate. Tumor heterogeneity may account for a small percentage of truly p53 positive cells and these tumors would be considered negative using immunohistochemistry. In contrast, temperature gradient gel electrophoresis has been used to identify specific p53 mutations in bladder cancer, and mutations have been detected in as few as 4% of cells in a single tumor.10 Finally, patient cohorts and definitions of progression may vary. For example, in this study 26 of 94 patients received intravesical therapy with BCG, total cystectomy or partial cystectomy and interstitial irradiation. The impact of BCG on progression or the cytoreduction associated with radiotherapy represents potentially confounding variables. The value of muscularis mucosal invasion has been more consistently demonstrated to be useful, although in a limited number of studies. The major problem is that this layer is only identifiable in about half (40% to 70%) of patients. The valid conclusion of this report is that invasion of the muscularis mucosae, if present, should be considered as an independent manifestation of a more aggressive phenotype. The clinical role of p53 remains to be definitively elucidated. Laurence Klotz Department of Surgery University of Toronto Sunnybrook Health Sciences Centre Toronto, Canada 1. Llopis, J., Alcaraz, A., Ribal, M. J. et al: p53 expression predicts progression and poor survival in T1 bladder tumours. Eur Urol, 37: 644, 2000 2. Steiner, G., Bierhoff, E., Schmidt, D. et al: p53 immunoreactivity in biopsy specimens of T1G3 transitional cell carcinoma of the bladder—a helpful parameter in guiding the decision for or against cystectomy? Eur J Cancer, 36: 610, 2000 3. Toktas, G., Turkeri, L. N., Unluer, E. et al: Prognostic significance of p53 protein accumulation in stage pT1 transitional cell carcinoma of the bladder. Int Urol Nephrol, 31: 437, 1999 4. Liukkonen, T., Rajala, P., Raitanen, M. et al: Prognostic value of MIB-1 score, p53, EGFr, mitotic index and papillary status in primary superficial (Stage pTa/T1) bladder cancer: a prospective comparative study. The Finnbladder Group. Eur Urol, 36: 393, 1999 5. Atug, F., Turkeri, L., Ozyurek, M. et al: Bcl-2 and p53 overexpression as associated risk factors in transitional cell carcinoma of the bladder. Int Urol Nephrol, 30: 455, 1998 6. Grossman, H. B., Liebert, M., Antelo, M. et al: p53 and RB expression predict progression in T1 bladder cancer. Clin Cancer Res, 4: 829, 1998 7. Korkolopoulou, P., Christodoulou, P., Kapralos, P. et al: The role of p53, MDM2 and c-erb B-2 oncoproteins, epidermal growth factor receptor and proliferation markers in the prognosis of urinary bladder cancer. Pathol Res Pract, 193: 767, 1997 8. Lee, E., Park, I. and Lee, C.: Prognostic markers of intravesical bacillus Calmette-Guerin therapy for multiple, high-grade, stage T1 bladder cancers. Int J Urol, 4: 552, 1997 9. Lebret, T., Becette, V., Barbagelatta, M. et al: Correlation between p53 over expression and response to bacillus CalmetteGuerin therapy in a high risk select population of patients with T1G3 bladder cancer. J Urol, 159: 788, 1998 10. Schlechte, H. H., Schnorr, D., Loning, T. et al: Mutation of the tumor suppressor gene p53 in human prostate and bladder cancers—investigation by temperature gradient gel electrophoresis (TGGE). J Urol, 157: 1049, 1997