1 Tumori, 96: , 2010 Influence of stage discrepancy on outcome in patients treated with radical cystectomy Ja Hyeon Ku 1, Kyung Chul Moon 2, Cheol Kwak 1, and Hyeon Hoe Kim 1 1 Department of Urology, and 2 Department of Pathology, Seoul National University College of Medicine, Seoul, Korea ABSTRACT Aims and background. To evaluate the influence of stage discrepancy on clinical outcome in patients with bladder cancer who have undergone radical cystectomy. Methods and study design. We reviewed the records of 155 patients who had undergone radical cystectomy. Of the 155 patients (128 males, 27 females), 68 had clinical nonmuscle invasive disease and 87 had muscle invasive disease. Follow-up ranged from to months (median, 34.3). Results. There was no significant difference in overall survival according to clinical T stage (P = 83). However, the actuarial overall survival rate of the pathological muscle invasive disease group was significantly less than that in the pathological nonmuscle invasive disease group (P <01). Multivariate analysis with the Cox regression model revealed that lymphovascular invasion (P = 01, relative risk [RR] = 2.463) and pathological T stage (P = 03, RR = 3.148) were strongly associated with overall survival. There was no difference in cancer-specific survival according to clinical T stage (P = 55). However, cancer-specific survival rate of the pathological muscle invasive disease group was significantly less than that in the pathological nonmuscle invasive disease group (P <01). Multivariate Cox proportional hazards model analysis showed that lymphovascular invasion (P = 01, RR = 2.545) and pathological T stage (P = 02, RR = 3.823) were independent predictors of cancerspecific survival. Conclusions. Our findings indicate that clinical stage determined by transurethral resection is not predictive of clinical outcome after radical cystectomy in patients with bladder cancer. Free full text available at Introduction Accurate clinical staging is important, and the mainstay of clinical staging for muscle invasive bladder cancer (MIBC) is pathologic examination of the transurethral resection (TUR) specimen. The pathological evaluation of tumor stage provided by TUR is critical, since the information directs therapy and influences prognosis. However, the clinical stage of disease may be discordant with the final pathologic stage determined after radical cystectomy by incomplete resection of the tumor, and a single TUR leaves residual disease in a significant number of patients. The percentage of positive histologic findings at the second TUR has been reported to range from 15% to 75% 1,2. In addition, it has been reported that more than 30% of patients with T1 transitional cell carcinoma of the bladder who have undergone radical cystectomy may be under-staged, which reflects limitations in the current staging modalities 3-6. Furthermore, potential impact of stage discrepancy on prognosis has yet to be evaluated in patients treated with radical cystectomy. To evaluate the impact of the stage discrepancy on clinical outcome, we performed a retrospective analysis in patients with bladder cancer who had undergone radical cystectomy. Key words: bladder neoplasm, cystectomy, recurrence, stage, survival. Correspondence to: Hyeon Hoe Kim, MD, PhD, Department of Urology, Seoul National University Hospital, 28, Yongon Dong, Jongno Ku, Seoul, , Korea. Tel ; fax ; Received May 14, 2009; accepted October 16, 2009.
2 700 JH KU, KC MOON, C KWAK, HH KIM Material and methods Approval of the study was obtained from the Institutional Review Board. We retrospectively reviewed the records of 162 consecutive patients with transitional cell carcinoma of the bladder who had undergone radical cystectomy at our institution between January 1991 and December Before radical cystectomy, TUR was performed in all patients for diagnostic and therapeutic purposes by staff urologic oncologists at our institution. The TUR were fully standardized; the bladder was thoroughly examined endoscopically, and the location, number and size of tumors were indentified. TUR of all visible tumors was then performed down to the lamina propria, and several deep muscular samples were also taken from the tumor base. No second resection was performed. Pathological data were reviewed and clinical staging was assigned based on an interpretation of the initial TUR specimens. In the present study, clinical staging was assigned based on an interpretation of the TUR specimens, and we used the term pathologic stage based on an interpretation of radical cystectomy specimens. We then compared the clinical and final pathological stage after radical cystectomy. Patients who received either preoperative or postoperative chemotherapy were excluded. Thus, 155 patients were enrolled in the study. Clinical information was collected for all patients, including preoperative demographics and indication for radical cystectomy. The mean age at radical cystectomy was 6 years (range, 32-84). The indications for radical cystectomy in cases of nonmuscle invasive bladder cancer (NMIBC) were disease refractory to intravesical therapy and tumor characteristics associated with a high risk of recurrence or progression. Of the 155 patients (128 males, 27 females) studied, 68 had clinical NMIBC and 87 had MIBC. The mean number of previous TUR was 2.0 (range, 0-25), and 32 patients (2%) had a previous history of intravesical instillation of bacillus Calmette-Guerin (BCG). Follow-up ranged from to months (median, 34.3). Evaluation included the histopathological result of the radical cystectomy and clinical outcome. All grading was performed using the World Health Organization (WHO) consensus classification of urothelial neoplasms 7, whereas stage was determined according to the WHO/International Society of Urologic Pathology (ISUP) 1998 classification system 8. Pathologic vascular invasion was defined as tumor present in vessels with a vascular wall and red blood cells in the lumen. Only when the tumor had invaded vessels with a definite endothelial lining was it considered lymphatic invasion. Perineural invasion was defined as tumor invading the perineural sheath or endoneurium. To evaluate which risk factor was most highly associated with the pathological MIBC in radical cystectomy specimens with respect to clinical and histologic variables, odds ratios and the P values for trend were estimated by multivariate logistic regression analysis. Clinical characteristics were entered into the model individually as independent variables. The associations between these parameters and the pathological MIBC were described using maximum likelihood estimates of odds ratios and 95% confidence intervals (CI) based on the logistic regression model. CI were based on the standard errors of the coefficients and a normal approximation. Outcomes were measured by time to overall survival and cancer-specific survival. Time to overall survival was calculated as the time from cystectomy to the date of death for any cause or until last follow-up if the patient had not died. Time to cancer-specific survival was calculated as the time from cystectomy to the date of death attributable to bladder cancer or until last followup if the patient had not died of bladder cancer. Survival analyses were conducted according to the Kaplan- Meier method, and survival characteristics were compared using logrank tests 9. Multivariate analysis by Cox s proportional hazards model was used to determine the contribution of several clinicopathologic factors to the survival rates of patients 10. The statistical software package SPSS 13.0 (SPSS, Inc., Chicago, IL, USA) was used for all statistical analyses. All calculated P values were 2- sided, and a P value of <5 was considered statistically significant. Results Table 1 compares clinical parameters according to the clinical stage determined by TUR. No significant difference in terms of clinical stage was observed according to age, sex, American Society of Anesthesiologists (ASA) score, number of previous TUR, or the incidence of perineural invasion in pathologic specimens. However, patients with clinical NMIBC had more intravesical BCG instillation (P = 47), received more orthotopic bladder substitute (P = 17), had higher incidence of lymphovascular invasion (P = 20), and had a lower pathological stage (P = 05). The comparison of clinical and pathologic stages is shown in Table 2. Clinical and pathologic stages were identical in 27.7% (43 of 155) of patients. Pathologic down-staging and up-staging occurred in 2% (31 of 155) and 52.3% (81 of 155), respectively. For predicting pathological MIBC, sensitivity and specificity of clinical stage were 64.7% (66 of 102) and 6% (32 of 53), respectively. To evaluate the influence of clinical and histologic parameters on pathological MIBC, the parameters were included in the univariate logistic model. The univariate logistic regression analysis indicated that history of BCG instillation, lymphovascular invasion and clinical T
3 INFLUENCE OF STAGE DISCREPANCY IN BLADDER CANCER 701 Table 1 - Clinical and pathological characteristics Tumor stage at TUR 1 Variable P 2 Age (yr) 09 <60 26 (38.2) 39 (44.8) (61.8) 48 (55.2) Sex 22 Male 55 (80.9) 73 (83.9) Female 13 (19.1) 14 (16.1) ASA score (33.8) 27 (3) (66.2) 60 (69.0) No. of previous TUR (57.4) 63 (72.4) 1 29 (42.6) 24 (27.6) History of BCG instillation 47 Absent 49 (72.1) 74 (85.1) Present 19 (27.9) 13 (14.9) Procedure 17 Conduit 46 (67.6) 73 (83.9) Neobladder 22 (32.4) 14 (16.1) Lymphovascular invasion 20 Absent 52 (76.5) 51 (58.6) Present 16 (23.5) 36 (41.4) Perineural invasion Absent 64 (94.1) 78 (89.7) Present 4 (5.9) 9 (10.3) Pathologic stage 05 pt0 7 (10.3) 3 (3.4) CIS 1 (1.5) 5 (5.7) pta 4 (5.9) 6 (6.9) pt1 20 (29.4) 7 (8.0) pt2 15 (22.1) 24 (27.6) pt3 16 (23.5) 35 (4) pt4 5 (7.4) 7 (8.0) ASA, American Society of Anesthesiologists; BCG, bacillus Calmette- Guerin; TUR, transurethral resection. 1 Data presented are number and percentage of patients. 2 Chi-square test. Table 2 - Comparison of bladder tumor stage after initial transurethral resection (TUR) and subsequent radical cystectomy Clinical stage at TUR cta ct1 ct2 Pathologic stage at radical cystectomy pt0 2 (2) 5 (8.6) 3 (3.4) PTis 0 () 1 (1.7) 5 (5.7) PTa 2 (2) 2 (3.4) 6 (6.9) PT1 3 (3) 17 (29.3) 7 (8.0) PT2 2 (2) 13 (22.4) 24 (27.6) PT3 1 (1) 15 (25.9) 35 (4) pt4 0 () 5 (8.6) 7 (8.0) Stage discrepancy Down-staged 2 (2) 8 (13.8) 21 (24.1) Same staged 2 (2) 17 (29.3) 24 (27.6) Up-staged 6 (6) 33 (56.9) 42 (43.8) Number and percent of cases. stage were possible risk factors for pathological MIBC. These variables were selected for the multivariate logistic model to determine their association with pathological MIBC. In the multivariate model used, all three variables were independent risk factors: no history of BCG instillation was associated with a increased likelihood of pathological MIBC (odds ratio, 3.16; 95% CI, 1.32 to 7.58; P = 10); patients with lymphovascular invasion had a 4.3-fold higher risk (odds ratio, 4.29; 95% CI, 1.72 to 19; P = 02) than those without lymphovascular invasion; clinical T stage was associated with pathological MIBC (odds ratio, 2.10; 95% CI, 0 to 4.38; P = 49). The results are shown in Table 3. During the follow-up, 61 patients (39.4%) died, and median overall survival was 34.3 months (95% CI, 4.0 to 139.5). there was no significant difference in overall survival according to clinical T stage (Figure 1A, P = 83, logrank test). However, the actuarial overall survival rate of the pathological MIBC group was significantly less than that of the pathological NMIBC group (Figure 1B, P <01, logrank test). In addition, multivariate analysis with the Cox regression model revealed that lymphovas- Table 3 - Risk factors for pathological muscle invasive disease in patients who underwent radical cystectomy OR (95% CI) P Adjusted OR P (95% CI) Age (yr) < ( ) 45 Sex Male 00 Female 48 ( ) ASA score ( ) No. of previous TUR (73-84) 83 History of BCG instillation Absent ( ) < ( ) 10 Present Procedure Conduit 00 Neobladder 85 (26-39) 63 Lymphovascular invasion Absent Present ( ) < ( ) 02 Perineural invasion Absent 00 Present ( ) 54 Clinical T stage ( ) ( ) 49 ASA, American Society of Anesthesiologists; BCG, bacillus Calmette- Guerin; CI, confidence interval; OR, odds ratio; TUR, transurethral resection.
4 702 JH KU, KC MOON, C KWAK, HH KIM cular invasion (P = 01, relative risk [RR] = 2.463) and pathological T stage (P = 03, RR = 3.148) were strongly associated with overall survival, regardless of age, sex, ASA score, number of previous TUR, history of intravesical BCG instillation, type of procedure, perineural invasion or clinical T stage (not shown). There was no difference in cancer-specific survival according to clinical T stage determined by TUR (Figure 1C, P = 55, logrank test). However, cancer-specific survival rate of the pathological MIBC group was significantly less than that in the pathological NMIBC group (Figure 1D, P <01, logrank test). Multivariate Cox proportional hazards model analysis showed that lymphovascular invasion (P = 01, RR = 2.545) and pathological T stage (P = 02, RR = 3.823) were independent predictors of cancer-specific survival (not shown). Discussion In most cases, therapeutic decisions concerning bladder tumors are based on the results of an initial TUR of the bladder tumor. If adequately performed, TUR allows precise differentiation between NMIBC and MIBC. However, there are many factors that confound the adequacy of resection, which include the multiplicity of disease, the capability and perseverance of the resectionist, the qualities of specimens provided and the pathological analysis 2. The etiology of stage disparity has another possible sources, i.e., delay in the interval from TUR to radical cystectomy 11. Furthermore, in the case of TUR specimens, the pathologic material given for examination may be limited by the size, depth, quality, or location of the tumor 12. A B Overall survival Overall survival P = 83, log-rank test P <01, log-rank test C D Cancer specific survival Cancer specific survival P = 55, log-rank test P <01, log-rank test Figure 1 - A) Overall survival by clinical T stage determined by transurethral resection specimens (P = 83, logrank test). B) Overall survival by pathological T stage determined by radical cystectomy (P <01, logrank test). C) Cancer-specific survival by clinical T stage determined by transurethral resection specimens (P = 55, logrank test). D) Cancer-specific survival by pathological T stage determined by radical cystectomy (P <01, logrank test).
5 INFLUENCE OF STAGE DISCREPANCY IN BLADDER CANCER 703 The inaccuracies of a single TUR have been documented as inconsistencies between the clinical and pathologic stages. Soloway et al. 5 reported that the overall clinical staging error was 61.5%, with 41.5% of the cancers under-staged. Richie et al. 13 compared clinical and pathologic stages and found that 26% of patients studied had been up-staged and 40% under-staged 13. Herr 2 reported that 29% with NMIBC was up-staged to an MIBC on repeated resection. Freeman et al. 6 indicated that in patients with clinical NMIBC, pathologic upstaging to MIBC or metastatic tumor occurred in 34% of patients. Furthermore, whereas less than 20% of patients with Ta tumors were under-staged 6, up to 40% of patients with T1 tumors were up-staged to muscle invasion 3,6,14. Conceptually, a second TUR would allow more accurate staging by excluding patients whose disease stage is revised upward by repeated resection 15,16. However, since no second resection was performed in our study, clinical stage may have been under-staged, especially in T1 disease. In accordance with previous reports 6,17, we found that a number of patients with high-risk clinical NMIBC had muscle invasion in their cystectomy specimen. Thus, discrepancy between clinical and pathologic stage is common in patients who undergo radical cystectomy. These findings may support the need to consider early cystectomy in some of patients with high-risk clinical NMIBC. Furthermore, our findings indicate that clinical stage determined by TUR is not predictive of clinical outcome after radical cystectomy in patients with bladder cancer. Finally, the prognostic implications of pathologic features such as vascular invasion, lymphatic invasion, and perineural invasion have been drawing attention. Controversy still surrounds the prognostic implications of vascular, lymphatic, and perineural invasion in bladder cancer. Some investigators reported that these may be useful prognostic factors for bladder cancer 18,19. However, others have suggested that these pathologic features were not associated with patient survival 20 or not independent prognostic factors regarding survival 21,22. In the present study, lymphovascular invasion was an independent prognostic factor regarding overall survival and cancer-specific survival of patients with bladder cancer after curative radical cystectomy. Although further study is needed to verify these findings, our results suggest that extra effort should be made to detect such a feature during pathologic examination, and additional treatment and close follow-up should be considered for patients with such pathologic findings. References 1. Klan R, Loy V, Huland H: Residual tumor discovered in routine second transurethral resection in patients with stage T1 transitional cell carcinoma of the bladder. J Urol, 146: , Herr HW: The value of a second transurethral resection in evaluating patients with bladder tumors. J Urol, 162: 74-76, Paulson DF: Critical review of radical cystectomy and indicators of prognosis. Semin Urol, 11: , Amling CL, Thrasher JB, Frazier HA, Dodge RK, Robertson JE, Paulson DF: Radical cystectomy for stages Ta, Tis and T1 transitional cell carcinoma of the bladder. J Urol, 151: 31-35, Soloway MS, Lopez AE, Patel J, Lu Y: Results of radical cystectomy for transitional cell carcinoma of the bladder and the effect of chemotherapy. Cancer, 73: , Freeman JA, Esrig D, Stein JP, Simoneau AR, Skinner EC, Chen SC, Groshen S, Lieskovsky G, Boyd SD, Skinner DG: Radical cystectomy for high risk patients with superficial bladder cancer in the era of orthotopic urinary reconstruction. Cancer, 76: , Epstein JI, Amin MB, Reuter VR, Mostofi FK: The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Am J Surg Pathol, 22: , Reuter VE, Epstein JI, Amin MB, Mostofi FK: The WHO/ISUP Consensus Classification of Urothelial (Transitional Cell) Neoplasms : continued discussion. Hum Pathol, 30: , Kaplan E, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc, 53: , Cox DR: Regression models and life-tables. J R Stat Soc, 34: , Chang SS, Hassan JM, Cookson MS, Wells N, Smith JA Jr: Delaying radical cystectomy for muscle invasive bladder cancer results in worse pathological stage. J Urol, 170: , Coblentz TR, Mills SE, Theodorescu D: Impact of second opinion pathology in the definitive management of patients with bladder carcinoma. Cancer, 91: , Richie JP, Skinner DG, Kaufman JJ: Radical cystectomy for carcinoma of the bladder: 16 years of experience. J Urol, 113: , Dalbagni G, Herr HW, Reuter VE: Impact of the second transurethral resection on the staging of T1 bladder cancer. Urology, 60: , Schwaibold HE, Sivalingam S, May F, Hartung R: The value of a second transurethral resection for T1 bladder cancer. BJU Int, 97: , Divrik T, Yildirim U, Eroglu AS, Zorlu F, Ozen H: Is a second tranurethral resection necessary for newly diagnosed pt1 bladder cancer? J Urol, 175: , Stein JP: Indications for early cystectomy. Urology, 62: , Fossa SD, Reitan JB, Ous S, Odegaard A, Loeb M: Prediction of tumour progression in superficial bladder carcinoma. Eur Urol, 11: 1-5, Leissner J, Koeppen C, Wolf HK: Prognostic significance of vascular and perineural invasion in urothelial bladder cancer treated with radical cystectomy. J Urol, 169: , Platz CE, Cohen MB, Jones MP, Olson DB, Lynch CF: Is microstaging of early invasive cancer of the urinary bladder possible or useful? Mod Pathol, 9: , Bassi P, Ferrante GD, Piazza N, Spinadin R, Carando R, Pappagallo G, Pagano F: Prognostic factors of outcome after radical cystectomy for bladder cancer: a retrospective study of a homogeneous patient cohort. J Urol, 161: , Hara S, Miyake H, Fujisawa M, Okada H, Arakawa S, Kamidono S, Hara I: Prognostic variables in patients who have undergone radical cystectomy for transitional cell carcinoma of the bladder. Jpn J Clin Oncol, 31: , 2001.