Japanese Journal of Clinical Oncology 31:536-540 (2001)
© 2001 Foundation for Promotion of Cancer Research
Treated Natural History of Superficial Bladder Cancer
1Department of Urology, Nara Medical University, Kashihara, 2Faculty of Medicine, Pharmacoepidemiology, The University of Tokyo, Tokyo, 3School of Health Sciences and Nursing, Biostatistics, The University of Tokyo, Tokyo, 4Department of Urology, Institute of Clinical Medicine, University of Tsukuba, Tsukuba and 5Department of Pathology, School of Medicine, Osaka Municipal University, Osaka, Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
Background: The present study was conducted to examine the natural history of superficial bladder cancer.
Methods: One hundred and forty-four patients with superficial bladder cancer who had been treated with transurethral resection of bladder tumor (TURBt) alone were analyzed.
Results: The non-recurrence rate was 64.8% at 36 months and 61.2% at 60 months after TURBt. When the non-recurrence rate after TURBt was analyzed by background variables, the rate differed significantly between the solitary tumor group and the multiple tumor group. The tumor recurrence hazard curves for the entire population had one high peak before 500 days and another slight peak around 1500 days after TURBt.
Conclusions: These results will provide basic information useful when evaluating new regimens of intravesical instillation therapy for prophylaxis of superficial bladder cancer after our complete TURBt in the Nara Uro-Oncology Research Group.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
Transurethral resection of bladder tumor (TURBt) is the usual approach to treat superficial bladder cancer. Intravesical instillations of anti-cancer drugs are administered for prophylaxis treatment for tumor recurrence. Despite this preventive therapy, however, the incidence of tumor recurrence within 1–2 years after TURBt is as high as 50% or more (1–5). Furthermore, recurrent tumors undergo progression in 10–20% of all cases. Recent analyses of hazard curves revealed a peak of early phase recurrence in addition to a peak of late phase recurrence (6). It is desirable to develop a prophylaxis therapy to prevent tumor recurrence, focusing on these two peaks.
The present study was conducted to examine the pattern of tumor recurrence among patients who had received no intravesical instillation therapy after TURBt for superficial bladder cancer in the Nara Uro-Oncology Research Group (NUORG) (see Appendix) and to analyze the natural history of superficial bladder cancer.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
The subjects of the present study were 144 patients with primary superficial bladder cancer (59 pTa and 85 pT1 cases, 33 G1 and 111 G2 cases) who had served as the control group in our previous randomized controlled trials (RCTs) on superficial bladder cancer, SBT-002 (thiotepa versus UFT versus control) (7,8) and 003 (epirubicin versus control) (9), carried out by the Nara Uro-Oncology Research Group (Table 1). The SBT-002 and 003 studies included those patients who satisfied all of the following inclusion criteria: (1) newly diagnosed patients with superficial bladder cancer who underwent TURBt; (2) patients whose tumor was histopathologically diagnosed as pTa-1, G1-2 transitional cell carcinoma (TCC); and (3) patients who had given fully informed consent to the study (7–9).
|
As a rule, complete TURBt in our institutions was performed as follows: (1) all visible tumors were resected; (2) normal-looking bladder mucosa on the border of the tumor were resected and confirmed as histopathologically normal; (3) one more resection of basement layer of the tumor was added and confirmed as histopathologically normal; (4) no malignant cells were observed histopathologically in the randomly biopsied specimens of the apparently normal urothelium from trigone, post-trigone, bilateral walls, dome and anterior wall of the bladder and both sides of the prostatic urethra; and (5) 1–2 weeks after TURBt, no remaining tumor was confirmed by cystoscopic and urinary cytological examination before enrollment in the study. Therefore, histopathological confirmation of the apparently normal tissues surrounding the tumor(s), such as both borders and basement of the tumor(s) and randomly biopsied specimens, was important information for consideration of prophylactic treatment of the superficial bladder cancer (10). Pathological evaluation was performed by a single pathologist (S.F.) in these two trials.
During the follow-up period, cystoscopic and urinary cytological examinations were carried out every 3 months for the first 3 years after TURBt and every 6 months for the fourth and fifth years and once a year from the fifth year on. However, patients whose recurrent tumor had progressed to G3 or above pT2 and patients with recurrence in the renal pelvis, ureter or urethra were allowed to receive other appropriate therapy.
The non-recurrence rate after TURBt was calculated using the Kaplan–Meier method and the significance of differences was tested using the generalized Wilcoxon test. A multivariate analysis was carried out using Cox’s proportional hazards regression model influential for tumor recurrence.
All tumors were categorized as grade (G1/G2), stage (pTa/pT1) and multiplicity (solitary/multiple) and the hazard function was estimated for all cases and each of the six subgroups (G1, G2, pTa, pT1, solitary and multiple). Smoothing was carried out by using a kernel function method (11) (width: 90 days). The validity of combining the data from two studies for estimating the hazard function was checked by testing the heterogeneity among the hazards of the studies based on log-lank statistics. Statistical analysis was performed by using SAS software (version 6.12) (SAS Institute, Cary, NC) and Microsoft Access software (version 7) (Microsoft, Redmond, WA) was employed for smoothing and graphic presentation.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
The follow-up periods ranged from 1 to 118 months (mean 43.3 ± 29.2 months, median 30 months). The non-recurrence rate was 64.8% at 36 months and 61.2% at 60 months after TURBt (Fig. 1a). The recurrence rate by the person–years method, calculated as total number of recurrence/total follow-up period in months, was 0.0126. When the non-recurrence rate was analyzed in relation to background variables, the rate did not depend on the tumor grade or stage (Fig. 1b and c). As can be seen in Fig. 1d, however, the rate differed significantly between the solitary and multiple tumor cases (p < 0.01).
|
A test was performed to determine whether there was heterogeneity among the SBT-002 and 003 studies for the tumor recurrence rate. The results of the test indicated that, for each of the factors, there was no heterogeneity among the studies. Smoothing of the hazard function was performed for combined untreated cases and the plots of the hazard rate were drawn up to 2000 days after TUR-Bt.
When the tumor recurrence hazard curves for all cases were analyzed, two peaks, one high peak before 500 days and another slight peak around 1500 days after TURBt, were observed (Fig. 2a). When the hazard curves were analyzed in relation to tumor grades, a high peak was observed before 500 days for both G1 and G2 groups and another slight peak was observed around 1500 days for G2 group (Fig. 2b). When analyzed by stages, the curves peaked high before 500 days for the pTa and pT1 groups and peaked again slightly around 1500 days (Fig. 2c). When analyzed by the tumor multiplicity, the curves peaked high before 500 days and peaked slightly around 1500 days for the solitary tumor group; the curves for the multiple tumor group had a high and early peak at 0–500 days and a slight peak at 500–1000 days after TURBt (Fig. 2d).
|
A multivariate analysis of the background variables involved in recurrence, using Cox’s proportional hazards regression model, revealed that the degree of involvement of background variables in tumor recurrence descends in the following order: multiplicity > age > gender > stage > grade. Only multiplicity was found to have statistically significant effects on tumor recurrence (p < 0.05) (Table 2).
|
Tumor progression to G3 or above pT2 or recurrence in the renal pelvis, ureter or urethra were observed in 10 cases (6.9%). The therapy was switched to intravesical BCG instillation in five CIS, G3 cases and three T1, G3 cases. One patient (T4, G3) died of cancer despite systemic chemotherapy using M-VAC. The one remaining patient underwent total nephroureterectomy to treat secondary ureteral cancer (T1, G2) (Table 3). Of all the patients, six died of disease other than bladder cancer, including one patient who died of myocardial infarction, two patients who died of pneumonia and two patients who died of double cancer, lung cancer in one case and rectal cancer in the other.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
Recently, several reports have described predictable factors involved in the recurrence or progression of superficial bladder cancer (1–5). The present study was conducted to examine the natural history of superficial bladder cancer treated with TURBt alone by following the clinical course of this cancer, in patients extracted from the control group of our previous RCT of superficial bladder cancer (7–9).
When the non-recurrence rate after TURBt was analyzed for all subjects and for groups of subjects divided by each background variable, we found no significant difference in this rate depending on tumor grade or stage. However, the non-recurrence rate differed significantly between the solitary tumor group and the multiple tumor group. This result was confirmed by the multivariate analysis of background variables involved in recurrence (carried out using Cox’s proportional hazards regression model), which revealed that the multiplicity of tumors was the greatest contributor to recurrence and that the non-recurrence rate differed significantly depending on multiplicity of tumor, but did not differ significantly depending on any other background variable. In this connection, Kobashi also reported that tumor multiplicity was found to be associated most closely with the risk of recurrence by a multivariate analysis of 336 cases with superficial bladder cancer (12).
When hazard curves, which indicate the tumor recurrence rate at each point in time, were analyzed, the curves for the entire population had one high peak before 500 days and another slight peak around 1500 days after TURBt. We interpreted these two peaks as representing elevated incidences of early recurrence and late recurrence, respectively. In addition, when the hazard curves were analyzed by background variables, we obtained interesting findings that even low-risk cases, pTa and solitary tumor cases, had two peaks before 500 days and around 1500 days; the G1 cases, however, had only one peak before 500 days, which probably reflected elevated incidences of early recurrence and late recurrence among pTa and solitary tumor cases. In general, primary, G1, pTa and solitary tumors represented an extremely low risk of tumor recurrence and constituted 18 out of 144 patients. Four of these 18 patients had tumor recurrence, although our complete TURBt was performed, which was considered to indicate that early recurrence might be brought about by the remaining invisible microscopic lesions (i.e. multicentricity) or tumor seeding immediately after TURBt. On the other hand, high-risk cases such as G2 and pT1 also showed two peaks before 500 days and around 1500 days after TURBt. The detection of a high peak before 500 days and another slight peak around 1500 days for the solitary tumor cases and the detection of a high peak at 0–500 days and a slight peak at 500–1000 days for the multiple tumor cases suggest that the number of tumors affects the risk of tumor recurrence in a complicated manner, probably involving some additional factors.
Hinotsu et al. (6) reported a similar investigation to estimate the natural history of superficial bladder cancer using a large consecutive series of 604 patients who were entered as the control group from five RCTs of the Japanese Urological Cancer Research Group. These 604 patients included primary and recurrent cases, whereas our study included only newly diagnosed patients, i.e. primary cases. Similar hazard curves were observed in our study and the primary/solitary and primary/multiple subgroups in Hinotsu et al.’s study. However, the peak hazard rate of early recurrence in the primary/multiple subgroups in Hinotsu et al.’s study was around 0.0035 and was also fairly high compared with that of multiple cases in the present study, <0.002. It is difficult to compare the absolute values of the hazard among two or more different sized populations. However, this difference in the hazards might be due to some qualitative difference such as the operational procedure. In addition, similar hazard curves, one high peak in the early phase and another slight peak in the late phase after TURBt, were observed in the two studies, suggesting that the former may depend on the remaining invisible microscopic lesions (i.e. multicentricity) or tumor seeding immediately after TURBt and the latter may result in a new occurrence (i.e. a second primary tumor).
In the present study, tumor progression to grade G3 or above stage pT2 or recurrence in the renal pelvis, ureter or urethra were observed in 10 cases (6.9%) and one patient (0.69%) died of cancer after surgery; however, factors determining tumor progression were not analyzed in detail. Takashi et al. (13) analyzed 223 cases and reported tumor progression in 12 cases (5.4%) and death from cancer in five cases (2.2%). They further reported that the degree of involvement in tumor progression descends in the order irritable symptom > grade > stage > tumor growth pattern and that tumor progression was determined significantly by irritable symptom (p < 0.001) and grade (p < 0.05).
The results obtained in this present study will provide basic information useful when evaluating new regimens of intravesical instillation therapy for prophylaxis of superficial bladder cancer after our complete TURBt in NUORG.
|
APPENDIX |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
The members of the Nara Uro-Oncology Research Group (NUORG) are:
Eigoro Okajima (Nara Medical University), Yoshihiko Hirao (Nara Medical University), Shoji Samma (Nara Prefectural Nara Hospital), Yoshiteru Kaneko (Nara Prefectural Mimuro Hospital), Shoichi Tabata (Nara Prefectural Gojo Hospital), Masashi Yamamoto (Nara National Hospital), Osamu Natsume (Nara Prefectural Rehabilitation Center), Kojiro Yoshida (Yamatotakada City Hospital), Yoshihiro Tani (Haibara General Hospital), Hideo Aoyama (Saiseikai Nara Hospital ), Shuji Watanabe (Saiseikai Chuwa Hospital), Shigehiro Tsujimoto (Saiseikai Gose Hospital), Kenichi Tsumatani (Kokuho Chuo Hospital), Hisashi Matsuki (Takanohara Chuo Hospital), Shoki Kimura (Nishinara Chuo Hospital), Kazuya Hirao (Hirao Hospital), Nobumichi Tani (Takai Hospital), Noboru Morita (Yukokai Hospital), Hitoshi Momose (Hoshigaoka Koseinenkin Hospital), Tsutomu Shiomi (Bobath Memorial Hospital), Tadashi Hiramatsu (Nissei Hospital), Yoriaki Kagebayashi (Osaka Kaisei Hospital), Yoshiki Hayashi (Tane General Hospital), Akio Iwai (Ishinkai Yao General Hospital), Akira Moriya (Asakayama Hospital ), Yoshio Maruyama (Matsusaka Chuo Hospital), Motoyoshi Yoshikawa (Okanami General Hospital).
|
FOOTNOTES |
|---|
+ For reprints and all correspondence: Seiichiro Ozono, Department of Urology, Nara Medical University, 840 Shijo-cho, Kashihara, Nara 634-8522, Japan. E-mail: ozn-kkr@nmu-gw.naramed-u.ac.jp
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONAPPENDIXREFERENCES |
|---|
1 Rübben H, Lutzeyer W, Fischer N, Deutz F, Lagrange W, Giani G, Members of the Registry for Urinary Tract Tumors, Rheinisch Westfälische Technische Hochschule, Aachen. Natural history and treatment of low and high risk superficial bladder tumors. J Urol 1988;139:283–5.[Web of Science][Medline]
2 Cookson MS, Herr HW, Zhang Z-F, Soloway S, Sogani PC, Fair WR. The treated natural history of high risk superficial bladder cancer: 15-year outcome. J Urol 1997;158:62–7.[Web of Science][Medline]
3 Herr HW. Natural history of superficial bladder tumors: 10- to 20-year follow-up of treated patients. World J Urol 1997;15:84–8.[Web of Science][Medline]
4 Lee C, Park MS. Prophylactic treatment of superficial bladder tumor. Int J Urol 1998;5:511–20.[Medline]
5 Haukaas S, Dæhlin L, Maartmann-Moe H, Ulvik NM. The long-term outcome in patients with superficial transitional cell carcinoma of the bladder: a single-institutional experience. BJU Int 1999;83:957–63.[Medline]
6 Hinotsu S, Akaza H, Ohashi Y, Kotake T. Intravesical chemotherapy for maximum prophylaxis of new early phase superficial bladder carcinoma treated by transurethral resection. A combined analysis of trials by the Japanese Urological Cancer Research Group using smoothed hazard function. Cancer 1999;86:1818–26.[Medline]
7 Hirao Y, Okajima E, Ozono S, Samma S, Sasaki K, Hiramatsu T, et al. A prospective randomized study of prophylaxis of tumor recurrence following transurethral resection of superficial bladder cancer-intravesical thio-TEPA versus oral UFT. Cancer Chemother Pharmacol 1992;30:26–30.
8 Tabata S, Ozono S, Hirao Y, Takashima K, Sasaki K, Watanabe S, et al. Long-term prophylactic effects of intravesical instillation of thiotepa and oral administration of UFT on the recurrence of superficial bladder cancer after transurethral resection. Int J Clin Oncol 1999;4:224–9.
9 Hirao Y, Ozono S, Momose H, Okajima E, Hiramatsu T, Yoshida K, et al. Prospective randomized study of prophylaxis of superficial bladder cancer with epirubicin: the role of a central pathology laboratory. Cancer Chemother Pharmacol 1994;35:36–40.
10 Okajima E. Transurethral resection of bladder tumor. In: Yoshida O, Okuyama A, editors. Modern Clinical Points. Urology: Kenehara 1993;142–3 (in Japanese).
11 Gray RJ. Some diagnostic methods for Cox regression models through hazard smoothing. Biometrics 1990;46:93–102.[Medline]
12 Kobashi K. Statistical analysis of recurrent factors in superficial bladder cancer. Natural history of superficial bladder cancer. Nippon Hinyokika Gakkai Zasshi 1993;84:1188–96 (in Japanese).[Medline]
13 Takashi M, Sakata T, Murase T, Takagi Y, Nagai T, Sahashi M, et al. Statistical evaluation of determinants of progression of superficial bladder cancer by proportional hazards model. Nippon Hinyokika Gakkai Zasshi 1990;81:116–21 (in Japanese).[Medline]
Received May 18, 2001; accepted August 1, 2001.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||