- Email: [email protected]
RENAL CELL AND TRANSITIONAL CELL CARCINOMA IN A JAPANESE POPULATION UNDERGOING MAINTENANCE DIALYSIS
0022-5347/05/1745-1749/0 THE JOURNAL OF UROLOGY® Copyright © 2005 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 174, 1749 –1753, November 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000177489.98031.54
Oncology: Adrenal/Renal/Upper Tract/Bladder RENAL CELL AND TRANSITIONAL CELL CARCINOMA IN A JAPANESE POPULATION UNDERGOING MAINTENANCE DIALYSIS SHIGERU SATOH,* NORIHIKO TSUCHIYA, TOMONORI HABUCHI, TAKESHI ISHIYAMA, KOJI SEIMO AND TETSURO KATO From the Department of Urology, Akita University School of Medicine and Akita Nephrology Dialysis Transplantation Association, Akita, and San-ai Hospital, Morioka, Japan
ABSTRACT
Purpose: We verified differences in the incidence, clinical characteristics and outcomes between patients on chronic dialysis for end stage renal disease with renal cell carcinoma (RCC) and those with transitional cell carcinoma (TCC). Materials and Methods: Data regarding RCC and TCC were reviewed in the medical records of 6,201 patients with end stage renal disease who underwent chronic dialysis between January 1990 and June 2003 in our 38 affiliated dialysis centers, and data were compared with those reported in Australia and New Zealand. Results: Among the patients RCC developed in 38 (0.61%) and TCC developed in 16 (0.26%) during maintenance dialysis. The primary renal disease was chronic glomerulonephritis in patients with RCC (68.4%) and diabetic nephropathy in patients with TCC (43.8%, p⫽0.002). Mean patient age at initiation of dialysis was 45 years for those with RCC and 63 for those with TCC (p ⬍0.001). Mean interval from dialysis induction to tumor diagnosis was 143 months for patients with RCC and 54 months for patients with TCC (p ⬍0.001). Of 38 RCCs 23 (60.5%) were incidentally detected by regular abdominal imaging examinations while painless gross hematuria was the cardinal symptom in 13 (81.2%) of 16 TCCs. Overall and cancer specific survivals after tumor diagnosis were significantly superior in patients with RCC compared to those with TCC (p⫽0.0001 and p⫽0.0003, respectively), and the cancer specific 5-year survival was 88.9% for RCC and 29.5% for TCC. In both cancers tumor stage significantly increased the risk of cancer specific death. Compared with patients from Australia and New Zealand, the incidence of RCC was higher and that of TCC was lower in our patients (p ⬍0.001). Conclusions: In the Japanese population on dialysis RCC is more common than TCC. Since long-term dialysis is a risk factor for RCC, regular imaging examinations may have contributed to the favorable outcome of our patients on dialysis with RCC. In contrast, the unfavorable outcome of TCC suggests the need for effective diagnostic measures for early detection of TCC in patients on dialysis. KEY WORDS: kidney failure, dialysis, kidney neoplasms, urologic neoplasms, survival rate
Patients with end stage renal disease (ESRD) have been recognized as a high risk group for cancer.1 National registries for patients with cancer on dialysis have been established in the United States (United States Renal Data System, [USRDS]), Europe (European Dialysis and Transplant Association [EDTA]), and Australia and New Zealand (Australia and New Zealand Dialysis and Transplant Registry [ANZDATA]).2, 3 Reviewing a cohort of 831,804 patients from the USRDS, EDTA and ANZDATA data sets, Maisonneuve et al showed that the incidence of cancer, especially in the kidney and urinary tract, was higher in patients on maintenance dialysis compared to the general population.2 They also mentioned that such an increased risk was related to the primary renal disease and that cancer of the kidney was Submitted for publication January 19, 2005. * Correspondence: Department of Urology, Akita University School of Medicine, 1–1-1 Hondo, Akita 010-8543, Japan (telephone: 81–18-834– 1111; FAX: 81–18-836–2619; e-mail: [email protected]). See Editorial on page 1723.
associated with the duration of dialysis.3 However, there was no more detailed information on the tumor stage at diagnosis or the outcome. Almost all cancers in the renal parenchyma are renal cell carcinoma (RCC) while those in the urinary tract (ie the renal pelvis, ureter, bladder and urethra) are transitional cell carcinoma (TCC). Both disease entities are different in terms of carcinogenesis and biological behavior. However, the USRDS, EDTA and ANZDATA have classified renal and urinary tract tumors according to the ninth edition of the International Classification of Disease (ICD-9),2, 3 in which cancers in the renal parenchyma and urinary tract, except for the bladder, are categorized as a single code of kidney cancer (ICD-9 189), and only bladder cancer is categorized as another code (ICD-9 188). Therefore, the differences in clinical characteristics and outcome between RCC and TCC cannot be identified in those data. RCC has been recognized as the most common renal and urinary tract carcinoma in patients on dialysis in Western
1749
1750
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS
countries, while TCC is the most common in Asian populations.4 There might be regional variations in the incidence of kidney and urinary tract cancers in patients on dialysis due to differences in dialysis duration, underlying renal disease, renal transplant rate and other factors in individual countries. In Japan, despite the annual statistical survey of patients on dialysis conducted by the Japanese Society for Dialysis Therapy, there is no available information on cancer incidence in those patients.5 In the present study we investigated the occurrence, clinical characteristics and outcome of RCC and TCC in patients on dialysis without renal transplantation in Japan. We also compared the incidence of both carcinomas in our patients with that extracted from the ANZDATA data. METHODS
Subjects. The charts of patients who underwent dialysis for ESRD between January 1, 1990 and June 30, 2003 at our 38 affiliated dialysis centers were reviewed for the occurrence of renal or urinary tract cancers along with dialysis related information. For the patients with these cancers, cardinal symptoms, diagnostics, treatments, pathological findings, outcome and, where appropriate, cause of death were investigated. Patients with a history of renal or urinary tract cancer before dialysis induction and those undergoing renal transplantation during the study period were excluded from analysis. Classification of kidney and urinary tract cancer. We classified only cancers of the renal parenchyma as kidney cancers, while cancers of the renal pelvis, ureter, bladder, or urethra were classified as urinary tract cancers. All of the tumors were diagnosed histologically using surgical, autopsy, biopsy and/or cytological specimens. Clinical tumor stage was based on imaging examinations including computerized tomography (CT) and bone scintigraphy, and classified according to the Robson system for RCC and the UICC staging system for TCC.6, 7 Data from the ANZDATA. Among the previous ANZDATA, USRDS and EDTA data sets, only the ANZDATA recorded the histological type of renal and upper urinary tract cancers, thus allowing us to distinguish the incidence of RCC from that of TCC. We extracted the data of RCC and TCC from the ANZDATA data consisting of 13,497 patients on dialysis for ESRD between 1980 and 1994 in Australia and New Zealand.2, 3 Statistical analyses. Data analysis was done with SPSS® 12.0 for Windows software. Group differences for qualitative variables were tested with the Pearson chi-square analysis. Comparisons of data between the 2 groups were performed by unpaired Student’s t test and the Mann-Whitney U test. Patient survival rates were stratified by the Kaplan-Meier method and tested with the log rank statistic. A Cox proportional hazards model was used to evaluate univariate relative risks for cancer specific death. Statistical significance was defined as p ⬍0.05. RESULTS
Characteristics of the cohort. According to our inclusion criteria, we assembled a retrospective cohort of 6,201 paTABLE 1. Characteristics of the study population No. pts % Male pts % Type of dialysis (hemodialysis/ peritoneal dialysis) Mean age at first dialysis (range) Mean mos dialysis (range) % Primary renal disease: Glomerulonephritis Diabetic nephropathy Miscellaneous conditions
6,201 59.1 95.9/4.1 59.9 (4–96) 82.5 (1–397) 42.0 27.3 30.7
tients on dialysis. Clinical characteristics of these patients are shown in table 1. Mean age at the start of dialysis was 59.9 years, mean duration of dialysis was 82.5 months and 95.9% of the patients were on hemodialysis. Primary glomerulonephritis including IgA nephropathy was the major cause of ESRD (42.0%), followed by diabetic nephropathy (27.3%). Miscellaneous conditions (30.7%) included nephrosclerosis due to hypertension, infective nephropathies, congenital disease such as polycystic kidney, toxic nephropathies, collagen diseases and uncertain causes. Characteristics of the patients with cancer. Of the 6,201 patients 54 (0.87%) were identified as having tumors in the kidney or the urinary tract during the study period. Cytohistological examination of the 54 tumors confirmed that 38 (0.61%) were RCC and 16 (0.26%) were TCC. Of the 16 TCCs the tumor sites were renal pelvis or ureter in 5 patients, upper urinary tract and bladder in 5 and bladder only in 6. There were no other histological tumor types in our patients. Characteristics of patients with RCC or TCC are shown in table 2. Males comprised 78.9% of patients with RCC and 62.5% of patients with TCC. In those with RCC compared to TCC, mean age at the beginning of dialysis and at cancer diagnosis was younger (44.8 vs 62.6 years, p ⬍0.001, and 56.5 vs 67.1 years, p⫽0.010) and mean duration of dialysis until cancer diagnosis was longer (143.2 vs 53.5 months, p ⬍0.001), respectively. The most common cause of ESRD was primary glomerulonephritis in patients with RCC (68.4%) and diabetic nephropathy in those with TCC (43.8%). Cardinal symptoms and diagnostics for RCC and TCC. The cardinal symptoms of RCC were hematuria in 11 patients (29.0%) and pain or fever in 4 (10.5%), while 23 remained asymptomatic (60.5%). Regular abdominal CT, ultrasonography or magnetic resonance imaging was performed every 6 months or 1 year in 24 of the 38 dialysis centers. Consequently the tumors of the 23 patients with RCC without symptoms were diagnosed incidentally by regular imaging examinations. Clinical tumor stage of the 38 cases of RCC was stage I in 34 (89.5%, low stage) and stage II to IV in 4 (high stage). On the other hand 13 (81.3%) of the 16 patients with TCC had gross hematuria and bloody urethral discharge, 3 (18.7%) had pain or fever, and none were diagnosed incidentally on regular examination. A combination of diagnostics such as CT and endoscopy was needed to establish the diagnosis in all TCC cases. Clinical tumor stage of the 10 upper urinary tract TCCs was stage 0a-I (Ta-T1) in 5 cases (50%, low stage) and stage II-IV (T2-T4) in 5 (high stage), and that of the 6 bladder TCC was stage 0a-I in 4 (66.7%, low stage) and stage II-IV in 2 (high stage). Treatments for RCC or TCC. The main treatment for RCC was nephrectomy in 33 patients (86.9%), embolization in 1 (2.6%), and no cancer specific treatment in 4 (10.5%). The main treatment for TCC was irradiation in 1 patient, transurethral resection of bladder tumor in 4, total cystectomy with ureter ligature in 3, nephroureterectomy in 4 and no treatment in 4. Outcome. Of the 38 patients with RCC 4 died of cancer and 7 died of other causes during the study period. Of the 16 patients with TCC 8 died of cancer and 3 died of other causes. The overall and cancer specific survival was significantly superior in patients with RCC than in those with TCC (p⫽0.0001 and p⫽0.003, respectively, figs. 1 and 2). The overall 5-year survival after tumor diagnosis was 78.6% for RCC and 21.1% for TCC, and cancer specific 5-year survival was 88.9% for RCC and 29.5% for TCC. We then evaluated the relative risk (RR) for cancer specific death associated with each cancer type using the Cox proportional hazards model. Patients with high stage tumors had a significantly increased risk of cancer specific death compared to those with low stage tumors in both tumor types (RR⫽21.21, p⫽0.013 in RCC and RR⫽12.16, p⫽0.021 in TCC, table 3).
1751
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS TABLE 2. Characteristics of patients on dialysis with RCC or TCC % Male pts % Hemodialysis Mean age at first dialysis ⫾ SD Mean age at Ca diagnosis ⫾ SD Mean mos dialysis until Ca diagnosis ⫾ SD % Primary renal disease: Glomerulonephritis Diabetic nephropathy Miscellaneous conditions
RCC
TCC
p Value
78.9 92.1 44.8 ⫾ 12.1 56.5 ⫾ 10.0 143.2 ⫾ 79.2
62.5 100 62.6 ⫾ 11.5 67.1 ⫾ 9.6 53.5 ⫾ 59.6
0.308 0.547 ⬍0.001 0.010 ⬍0.001 0.002
68.4 7.9 23.7
25.0 43.8 31.2
TABLE 3. Univariate relative risk of cancer specific death for RCC and TCC RR
FIG. 1. Overall survival rate after cancer diagnosis in patients on dialysis with RCC in kidney or TCC in urinary tract. In patients with RCC 5-year survival was 78.6% and in those with TCC it was 21.1%.
FIG. 2. Cancer specific survival rate after cancer diagnosis in patients on dialysis with RCC in kidney or TCC in urinary tract. In patients with RCC 5-year survival was 88.9% and in those with TCC it was 29.5%.
Regional differences. Mean dialysis duration of all patients, cancer incidence and main underlying disease of ESRD in patients with RCC or TCC in the ANZDATA and our series are shown in table 4. Compared with ANZDATA series, it was noted that in our series the dialysis duration of all patients was significantly longer, and the occurrence of RCC was significantly higher while that of TCC was lower. There was a remarkable difference in the main primary renal diseases between our patients and ANZDATA patients with RCC or TCC. Toxic nephropathies were the main renal diseases in the ANZDATA patients with RCC or TCC, especially TCC, while there were no patients with toxic nephropathies in our patients on dialysis. DISCUSSION
The risk and pathogenesis of RCC and TCC in patients on dialysis may differ from those in the general population. The
95% CI
p Value
Sex male vs female: RCC 1.35 0.34–5.44 Not TCC 3.14 0.75–13.27 Not Diabetic nephropathy yes/no: RCC 0.77 0.06–9.37 Not TCC 1.43 0.33–6.16 Not Age at dialysis induction: RCC 1.05 0.96–1.15 Not TCC 0.99 0.92–1.05 Not Age at Ca diagnosis: RCC 1.08 0.95–1.23 Not TCC 1.01 0.94–1.09 Not Mos dialysis until Ca diagnosis: RCC 1.00 0.98–1.01 Not TCC 1.02 1.01–1.03 Stage high vs low: RCC 21.21 1.91–235.90 TCC 12.16 1.45–102.00 Site with/without upper urinary 6.11 0.70–3.54 Not tract (TCC)
significant significant significant significant significant significant significant significant significant 0.041 0.013 0.021 significant
kidney of ESRD shows tubular atrophy, interstitial inflammation and fibrosis, and arterial and glomerular sclerosis, resulting in acquired cysts due to the loss of structural integrity.3 Of these histopathological degenerations, acquired renal cystic disease alone has been implicated as the risk factor for malignant transformation.8, 9 The wall of acquired cysts is lined with hyperplastic epithelial cells, and the resultant papillary hyperplasia of cystic epithelium has a specific role in the development of RCC in patients on chronic dialysis. Acquired renal cystic disease appears in patients on hemodialysis and peritoneal dialysis independent of patient age or primary renal disease,10 and the prevalence increases with the duration of dialysis.11 Thus, RCC from the acquired cysts is associated with duration of dialysis. Also, the cysts may develop more rapidly in men than in women.11 On the other hand, uremic immune suppression, dialysis procedure and primary kidney diseases have been mentioned as the pathogenic conditions for the development of TCC in patients on dialysis.4 Although real pathogenesis still remains unclear, the toxic nephropathies account for the excess risk of TCC.2, 3 We identified 38 patients with RCC (0.61%) and 16 patients with TCC (0.26%) in the present review of 6,201 patients on dialysis. The patients with RCC compared to TCC were younger, had longer dialysis duration, and included more males and a larger number of patients with chronic glomerulonephritis. Of 38 patients with RCC 34 (89.5%) had acquired renal cysts, a finding similar to that of a previous report in which more than 80% of patients with RCC on dialysis had acquired renal cysts.9 Toxic nephropathies, a disease entity highly associated with carcinogenesis of urothelium,3, 12 are not a rare cause of primary renal diseases underlying ESRD in the USRDS, EDTA and ANZDATA, but are extremely rare in Japan.2, 3, 5 The major primary renal disease in our patients with TCC
1752
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS
TABLE 4. Comparison of mean dialysis duration of all patients, cancer frequencies and main primary renal disease of patients with RCC or TCC between the present study and the ANZDATA data Present Study Total No. pts Mean mos dialysis for all patients No. Ca incidence (%): RCC TCC No. main primary renal disease in RCC (%): Glomerulonephritis Toxic nephropathies No. main primary renal disease in TCC (%): Diabetic nephropathy Glomerulonephritis Toxic nephropathies
6,201 82.5
ANZDATA 13,497 30.6
p Value ⬍0.001 ⬍0.001
38 (0.61) 16 (0.26)
34 (0.25) 87 (0.64)
26 (68.4) 0 (0)
14 (41.2) 6 (17.6)
0.008
7 (43.8) 4 (25.0) 0 (0)
2 (2.3) 14 (16.1) 51 (58.6)
⬍0.001
was diabetic nephropathy (43.8%). This may reflect the fact that the major underlying disease among the newly initiated patients on dialysis is diabetic nephropathy (36.6%) in Japan.5 There is no report suggesting that TCC in patients on dialysis is related to diabetic nephropathy. Cigarette smoking and chemical exposure were other risk factors contributing to TCC,4 however, there was no information of smoking history in the USRDS, EDTA or ANZDATA data sets. Although we did not survey the smoking history in our cohort, 1 study showed that tumor stage had the highest correlation and smoking history had a low correlation with bladder cancer specific survival in Japanese patients not on dialysis.13 In the present series the mean dialysis duration until the detection of TCC was significantly shorter than that for RCC (53.5 vs 143.2 months), and 7 (43.8%) of 16 TCCs were diagnosed within 1 year of dialysis (data not shown). Stewart et al also described an excess of urinary tract cancers in the first year of dialysis.3 This may have come from the diagnostic evaluation at the induction of dialysis.3 Previous studies reported that the 5-year survival rate of patients with ESRD with kidney cancer was approximately 35%, similar to that of patients not on dialysis,9 and suggested that cancer screening was an inefficient allocation of resources, adding little to the life expectancy of these patients.14 However, regular screening of patients on dialysis may increase survival by encouraging radical treatments for kidney cancers at earlier tumor stage.15 In fact, the present study of patients on dialysis showed that 89.5% of those with RCC had low stage tumors, and that the overall and cancer specific 5-year survival of patients with RCC was 78.6% and 88.9%, respectively, thus indicating that the tumor stage was a significant risk factor for cancer death (RR⫽21.21). This favorable outcome of patients with RCC can be attributed to the regular imaging screening applied by most of our institutions. In contrast, only 56.3% of our patients with TCC had low stage tumors, and the overall and cancer specific 5-year survival of patients with TCC was 21.1% and 29.5%, respectively, again with tumor stage being a significant risk factor for cancer death (RR⫽12.6). A small study suggested that urinary cytology was a good screening method for the early detection of TCC in patients on dialysis with analgesic nephropathy.16 However, the screening urinary cytology was performed in only 1 of our affiliated dialysis centers, and no TCC was detected by screening urinary cytology or regular imaging. In addition to the delay in tumor diagnosis, older patient age and more aggressive underlying disease of diabetic nephropathy may be responsible for the poor outcome of patients with TCC. Since tumor stage is a significant risk factor for cancer specific death, effective diagnostic measures for early detection of TCC in patients on dialysis will be needed. In the general population the age standardized incidence rates of RCC per 100,000 of the Japanese population for men
and women were 4.9 and 1.8, respectively. The incidence rates of RCC in Japan were approximately the same as among Japanese in Los Angeles, and lower than North American and European countries.17 There was no nationwide survey for TCC in Japan or Australia and New Zealand. One regional report from Japan showed that the incidence rates of bladder carcinoma per 100,000 men and women were 5.3 and 1.7, respectively.18 Bladder cancer specific mortality in Japan was lower than in North America, European counties, and Australia and New Zealand.19 In patients on dialysis the incidence of TCC was significantly lower in our series compared to the ANZDATA data series, which comes from the difference in primary underlying renal diseases. Toxic nephropathies, a highly responsible disease for urothelial carcinoma (TCC),12 is more common as a primary underlying renal disease in the USRDS, EDTA and ANZDATA regions than in Japan.2, 3, 5 On the other hand, the excess rate of RCC in our patients may reflect the Japanese feature of renal replacement therapy. In Japan with the total number of patients on dialysis of 206,134 in 2000,5 only 744 renal transplantations were performed, a frequency approximated to one- twentieth of that in the United States.20 Consequently most patients with ESRD in Japan are treated with long-term dialysis which will accelerate the development of the acquired renal cyst and then RCC. Wu et al mentioned that RCC was common and TCC was rare in patients on dialysis in Western countries, while TCC is the most common carcinoma in Taiwanese and Japanese patients on dialysis.4 However, the present study, and the data of ANZDATA, EDTA and USRDA revealed that RCC is common and TCC is rare in patients on dialysis in Japan relative to Western countries.2, 3 The frequency and types of screening studies were other factors contributing the cancer detection rate and outcome in patients on dialysis, but there was no information on diagnostics, cancer stage or prognosis for RCC and TCC in the USRDS, EDTA or ANZDATA data series.
CONCLUSIONS
A total of 54 renal and urinary tract cancers (0.87%) were identified among 6,201 patients on dialysis in a Japanese cohort. The RCC incidence of 0.61% was significantly higher than that of 0.25% in the ANZDATA data, suggesting a long-term dialysis in our patients. Early detection by routine abdominal imaging examination may have contributed to the favorable RCC specific 5-year survival of 88.9% in our cohort. In contrast the TCC incidence of 0.26% was lower than that of 0.64% in the ANZDATA data, probably resulting from the lack of toxic nephropathies as an underlying renal disease in our patients. The significantly worse outcome of TCC indicated the need to develop early detection measures for TCC in patients on dialysis.
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS REFERENCES
1. Vamvakas, S., Bahner, U. and Heidland, A.: Cancer in end-stage renal disease: potential factors involved - editorial-. Am J Nephrol, 18: 89, 1998 2. Maisonneuve, P., Agodoa, L., Gellert, R., Stewart, J. H., Buccianti, G., Lowenfels, A. B. et al: Cancers in patients on dialysis for end-stage renal disease: an international collaborative study. Lancet, 354: 93, 1999 3. Stewart, J. H., Buccianti, G., Agodoa, L., Gellert, R., McCredie, M. R., Lowenfels, A. B. et al: Cancers of the kidney and urinary tract in patients on dialysis for end-stage renal disease: analysis of data from the United States, Europe, and Australia and New Zealand. J Am Soc Nephrol, 14: 197, 2003 4. Wu, C.-F., Shee, J.-J., Ho, D.-R., Chen, W.-C. and Chen, C.-S.: Different treatment strategies for end stage renal disease in patients with transitional cell carcinoma. J Urol, 171: 126, 2004 5. Nakai, S., Shinzato, T., Sanaka, T., Kikuchi, K., Kitaoka, T., Shinoda, T. et al: The current state of chronic dialysis treatment in Japan. J Jpn Soc Dial Ther, 35: 1155, 2002 6. Novick, A. C. and Campbell, S. C.: Renal tumors. In: Campbell’s Urology, 8th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. and A. J. Wein. Philadelphia: W. B. Saunders Co., vol. 4, sect. X, chapt. 75, pp. 2672–2731, 2002 7. Messing, E. M.: Urothelial tumors of the urinary tract. In: Campbell’s Urology, 8th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. and A. J. Wein. Philadelphia: W. B. Saunders Co., vol. 4, sect. X, pp. 2732–2784, 2002 8. Hughson, M. D., Buchwald, D. and Fox, M.: Renal neoplasia and acquired cystic kidney disease in patients receiving long-term dialysis. Arch Pathol Lab Med, 110: 592, 1986 9. Matson, M. A. and Cohen, E. P.: Acquired cystic kidney disease: occurrence, prevalence, and renal cancers. Medicine (Baltimore), 69: 217, 1990 10. Park, J. H., Kim, Y. O., Park, J. H., Kim, B. S., Yoon, S. A., Yang, C. W. et al: Comparison of acquired cystic kidney disease
11. 12. 13. 14. 15. 16.
17. 18.
19. 20.
1753
between hemodialysis and continuous ambulatory peritoneal dialysis. Korean J Intern Med, 15: 51, 2000 Ishikawa, I., Saito, Y., Nakamura, M., Takada, K., Ishii, H., Nakazawa, T. et al: Fifteen-year follow-up of acquired renal cystic disease –a gender difference. Nephron, 75: 315, 1997 Blohme, I. and Johansson, S.: Renal pelvic neoplasms and atypical urothelium in patients with end-stage analgesic nephropathy. Kidney Int, 20: 671, 1981 Takashi, M., Murase, T., Mizuno, S., Hamajima, N. and Ohno, Y.: Multivariate evaluation of prognostic determinants in bladder cancer patients. Urol Int, 42: 368, 1987 Chertow, G. M., Paltiel, A. D., Owen, W. F., Jr. and Lazarus, J. M.: Cost-effectiveness of cancer screening in end-stage renal disease. Arch Intern Med, 156: 1345, 1996 Li, M. K., Choy, D. K. and Yip, S. K.: Renal cell carcinoma in patients with chronic renal failure. Ann Acad Med Singapore, 28: 512, 1999 Veltman, G. A., Bosch, F. H., van der Plas-Cats, M. B. and van Leusen, R.: Urinary cytology as a screening method for transitional-cell carcinoma in dialysis patients with analgesic nephropathy. Nephrol Dial Transplant, 6: 346, 1991 Marumo, K., Satomi, Y., Miyao, N., Hasegawa, M., Tomita, Y., Igarashi, T. et al: The prevalence of renal cell carcinoma: a nation-wide survey in Japan in 1997. Int J Urol, 8: 359, 2001 The Research Group for Population-Based Cancer Registration in Japan: Cancer Incidence in Japan, 1975-Cancer Registry Statistics. Cancer Mortality and Morbidity Statistics. GANN Monograph on Cancer Research, No. 26, 1981 Kurihara, M., Aoki, K. and Hisamichi, S.: Cancer Mortality Statistics in the World. Nagoya: The University of Nagoya Press, 1989 The Japanese Society for Clinical Renal Transplantation and the Japanese Society for Transplantation: Annual progress report from the Japanese renal transplant registry. The number of the renal transplantation in 2000. Jpn J Transplant, 36: 87, 2001
Vol. 174, 1749 –1753, November 2005 Printed in U.S.A.
DOI: 10.1097/01.ju.0000177489.98031.54
Oncology: Adrenal/Renal/Upper Tract/Bladder RENAL CELL AND TRANSITIONAL CELL CARCINOMA IN A JAPANESE POPULATION UNDERGOING MAINTENANCE DIALYSIS SHIGERU SATOH,* NORIHIKO TSUCHIYA, TOMONORI HABUCHI, TAKESHI ISHIYAMA, KOJI SEIMO AND TETSURO KATO From the Department of Urology, Akita University School of Medicine and Akita Nephrology Dialysis Transplantation Association, Akita, and San-ai Hospital, Morioka, Japan
ABSTRACT
Purpose: We verified differences in the incidence, clinical characteristics and outcomes between patients on chronic dialysis for end stage renal disease with renal cell carcinoma (RCC) and those with transitional cell carcinoma (TCC). Materials and Methods: Data regarding RCC and TCC were reviewed in the medical records of 6,201 patients with end stage renal disease who underwent chronic dialysis between January 1990 and June 2003 in our 38 affiliated dialysis centers, and data were compared with those reported in Australia and New Zealand. Results: Among the patients RCC developed in 38 (0.61%) and TCC developed in 16 (0.26%) during maintenance dialysis. The primary renal disease was chronic glomerulonephritis in patients with RCC (68.4%) and diabetic nephropathy in patients with TCC (43.8%, p⫽0.002). Mean patient age at initiation of dialysis was 45 years for those with RCC and 63 for those with TCC (p ⬍0.001). Mean interval from dialysis induction to tumor diagnosis was 143 months for patients with RCC and 54 months for patients with TCC (p ⬍0.001). Of 38 RCCs 23 (60.5%) were incidentally detected by regular abdominal imaging examinations while painless gross hematuria was the cardinal symptom in 13 (81.2%) of 16 TCCs. Overall and cancer specific survivals after tumor diagnosis were significantly superior in patients with RCC compared to those with TCC (p⫽0.0001 and p⫽0.0003, respectively), and the cancer specific 5-year survival was 88.9% for RCC and 29.5% for TCC. In both cancers tumor stage significantly increased the risk of cancer specific death. Compared with patients from Australia and New Zealand, the incidence of RCC was higher and that of TCC was lower in our patients (p ⬍0.001). Conclusions: In the Japanese population on dialysis RCC is more common than TCC. Since long-term dialysis is a risk factor for RCC, regular imaging examinations may have contributed to the favorable outcome of our patients on dialysis with RCC. In contrast, the unfavorable outcome of TCC suggests the need for effective diagnostic measures for early detection of TCC in patients on dialysis. KEY WORDS: kidney failure, dialysis, kidney neoplasms, urologic neoplasms, survival rate
Patients with end stage renal disease (ESRD) have been recognized as a high risk group for cancer.1 National registries for patients with cancer on dialysis have been established in the United States (United States Renal Data System, [USRDS]), Europe (European Dialysis and Transplant Association [EDTA]), and Australia and New Zealand (Australia and New Zealand Dialysis and Transplant Registry [ANZDATA]).2, 3 Reviewing a cohort of 831,804 patients from the USRDS, EDTA and ANZDATA data sets, Maisonneuve et al showed that the incidence of cancer, especially in the kidney and urinary tract, was higher in patients on maintenance dialysis compared to the general population.2 They also mentioned that such an increased risk was related to the primary renal disease and that cancer of the kidney was Submitted for publication January 19, 2005. * Correspondence: Department of Urology, Akita University School of Medicine, 1–1-1 Hondo, Akita 010-8543, Japan (telephone: 81–18-834– 1111; FAX: 81–18-836–2619; e-mail: [email protected]). See Editorial on page 1723.
associated with the duration of dialysis.3 However, there was no more detailed information on the tumor stage at diagnosis or the outcome. Almost all cancers in the renal parenchyma are renal cell carcinoma (RCC) while those in the urinary tract (ie the renal pelvis, ureter, bladder and urethra) are transitional cell carcinoma (TCC). Both disease entities are different in terms of carcinogenesis and biological behavior. However, the USRDS, EDTA and ANZDATA have classified renal and urinary tract tumors according to the ninth edition of the International Classification of Disease (ICD-9),2, 3 in which cancers in the renal parenchyma and urinary tract, except for the bladder, are categorized as a single code of kidney cancer (ICD-9 189), and only bladder cancer is categorized as another code (ICD-9 188). Therefore, the differences in clinical characteristics and outcome between RCC and TCC cannot be identified in those data. RCC has been recognized as the most common renal and urinary tract carcinoma in patients on dialysis in Western
1749
1750
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS
countries, while TCC is the most common in Asian populations.4 There might be regional variations in the incidence of kidney and urinary tract cancers in patients on dialysis due to differences in dialysis duration, underlying renal disease, renal transplant rate and other factors in individual countries. In Japan, despite the annual statistical survey of patients on dialysis conducted by the Japanese Society for Dialysis Therapy, there is no available information on cancer incidence in those patients.5 In the present study we investigated the occurrence, clinical characteristics and outcome of RCC and TCC in patients on dialysis without renal transplantation in Japan. We also compared the incidence of both carcinomas in our patients with that extracted from the ANZDATA data. METHODS
Subjects. The charts of patients who underwent dialysis for ESRD between January 1, 1990 and June 30, 2003 at our 38 affiliated dialysis centers were reviewed for the occurrence of renal or urinary tract cancers along with dialysis related information. For the patients with these cancers, cardinal symptoms, diagnostics, treatments, pathological findings, outcome and, where appropriate, cause of death were investigated. Patients with a history of renal or urinary tract cancer before dialysis induction and those undergoing renal transplantation during the study period were excluded from analysis. Classification of kidney and urinary tract cancer. We classified only cancers of the renal parenchyma as kidney cancers, while cancers of the renal pelvis, ureter, bladder, or urethra were classified as urinary tract cancers. All of the tumors were diagnosed histologically using surgical, autopsy, biopsy and/or cytological specimens. Clinical tumor stage was based on imaging examinations including computerized tomography (CT) and bone scintigraphy, and classified according to the Robson system for RCC and the UICC staging system for TCC.6, 7 Data from the ANZDATA. Among the previous ANZDATA, USRDS and EDTA data sets, only the ANZDATA recorded the histological type of renal and upper urinary tract cancers, thus allowing us to distinguish the incidence of RCC from that of TCC. We extracted the data of RCC and TCC from the ANZDATA data consisting of 13,497 patients on dialysis for ESRD between 1980 and 1994 in Australia and New Zealand.2, 3 Statistical analyses. Data analysis was done with SPSS® 12.0 for Windows software. Group differences for qualitative variables were tested with the Pearson chi-square analysis. Comparisons of data between the 2 groups were performed by unpaired Student’s t test and the Mann-Whitney U test. Patient survival rates were stratified by the Kaplan-Meier method and tested with the log rank statistic. A Cox proportional hazards model was used to evaluate univariate relative risks for cancer specific death. Statistical significance was defined as p ⬍0.05. RESULTS
Characteristics of the cohort. According to our inclusion criteria, we assembled a retrospective cohort of 6,201 paTABLE 1. Characteristics of the study population No. pts % Male pts % Type of dialysis (hemodialysis/ peritoneal dialysis) Mean age at first dialysis (range) Mean mos dialysis (range) % Primary renal disease: Glomerulonephritis Diabetic nephropathy Miscellaneous conditions
6,201 59.1 95.9/4.1 59.9 (4–96) 82.5 (1–397) 42.0 27.3 30.7
tients on dialysis. Clinical characteristics of these patients are shown in table 1. Mean age at the start of dialysis was 59.9 years, mean duration of dialysis was 82.5 months and 95.9% of the patients were on hemodialysis. Primary glomerulonephritis including IgA nephropathy was the major cause of ESRD (42.0%), followed by diabetic nephropathy (27.3%). Miscellaneous conditions (30.7%) included nephrosclerosis due to hypertension, infective nephropathies, congenital disease such as polycystic kidney, toxic nephropathies, collagen diseases and uncertain causes. Characteristics of the patients with cancer. Of the 6,201 patients 54 (0.87%) were identified as having tumors in the kidney or the urinary tract during the study period. Cytohistological examination of the 54 tumors confirmed that 38 (0.61%) were RCC and 16 (0.26%) were TCC. Of the 16 TCCs the tumor sites were renal pelvis or ureter in 5 patients, upper urinary tract and bladder in 5 and bladder only in 6. There were no other histological tumor types in our patients. Characteristics of patients with RCC or TCC are shown in table 2. Males comprised 78.9% of patients with RCC and 62.5% of patients with TCC. In those with RCC compared to TCC, mean age at the beginning of dialysis and at cancer diagnosis was younger (44.8 vs 62.6 years, p ⬍0.001, and 56.5 vs 67.1 years, p⫽0.010) and mean duration of dialysis until cancer diagnosis was longer (143.2 vs 53.5 months, p ⬍0.001), respectively. The most common cause of ESRD was primary glomerulonephritis in patients with RCC (68.4%) and diabetic nephropathy in those with TCC (43.8%). Cardinal symptoms and diagnostics for RCC and TCC. The cardinal symptoms of RCC were hematuria in 11 patients (29.0%) and pain or fever in 4 (10.5%), while 23 remained asymptomatic (60.5%). Regular abdominal CT, ultrasonography or magnetic resonance imaging was performed every 6 months or 1 year in 24 of the 38 dialysis centers. Consequently the tumors of the 23 patients with RCC without symptoms were diagnosed incidentally by regular imaging examinations. Clinical tumor stage of the 38 cases of RCC was stage I in 34 (89.5%, low stage) and stage II to IV in 4 (high stage). On the other hand 13 (81.3%) of the 16 patients with TCC had gross hematuria and bloody urethral discharge, 3 (18.7%) had pain or fever, and none were diagnosed incidentally on regular examination. A combination of diagnostics such as CT and endoscopy was needed to establish the diagnosis in all TCC cases. Clinical tumor stage of the 10 upper urinary tract TCCs was stage 0a-I (Ta-T1) in 5 cases (50%, low stage) and stage II-IV (T2-T4) in 5 (high stage), and that of the 6 bladder TCC was stage 0a-I in 4 (66.7%, low stage) and stage II-IV in 2 (high stage). Treatments for RCC or TCC. The main treatment for RCC was nephrectomy in 33 patients (86.9%), embolization in 1 (2.6%), and no cancer specific treatment in 4 (10.5%). The main treatment for TCC was irradiation in 1 patient, transurethral resection of bladder tumor in 4, total cystectomy with ureter ligature in 3, nephroureterectomy in 4 and no treatment in 4. Outcome. Of the 38 patients with RCC 4 died of cancer and 7 died of other causes during the study period. Of the 16 patients with TCC 8 died of cancer and 3 died of other causes. The overall and cancer specific survival was significantly superior in patients with RCC than in those with TCC (p⫽0.0001 and p⫽0.003, respectively, figs. 1 and 2). The overall 5-year survival after tumor diagnosis was 78.6% for RCC and 21.1% for TCC, and cancer specific 5-year survival was 88.9% for RCC and 29.5% for TCC. We then evaluated the relative risk (RR) for cancer specific death associated with each cancer type using the Cox proportional hazards model. Patients with high stage tumors had a significantly increased risk of cancer specific death compared to those with low stage tumors in both tumor types (RR⫽21.21, p⫽0.013 in RCC and RR⫽12.16, p⫽0.021 in TCC, table 3).
1751
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS TABLE 2. Characteristics of patients on dialysis with RCC or TCC % Male pts % Hemodialysis Mean age at first dialysis ⫾ SD Mean age at Ca diagnosis ⫾ SD Mean mos dialysis until Ca diagnosis ⫾ SD % Primary renal disease: Glomerulonephritis Diabetic nephropathy Miscellaneous conditions
RCC
TCC
p Value
78.9 92.1 44.8 ⫾ 12.1 56.5 ⫾ 10.0 143.2 ⫾ 79.2
62.5 100 62.6 ⫾ 11.5 67.1 ⫾ 9.6 53.5 ⫾ 59.6
0.308 0.547 ⬍0.001 0.010 ⬍0.001 0.002
68.4 7.9 23.7
25.0 43.8 31.2
TABLE 3. Univariate relative risk of cancer specific death for RCC and TCC RR
FIG. 1. Overall survival rate after cancer diagnosis in patients on dialysis with RCC in kidney or TCC in urinary tract. In patients with RCC 5-year survival was 78.6% and in those with TCC it was 21.1%.
FIG. 2. Cancer specific survival rate after cancer diagnosis in patients on dialysis with RCC in kidney or TCC in urinary tract. In patients with RCC 5-year survival was 88.9% and in those with TCC it was 29.5%.
Regional differences. Mean dialysis duration of all patients, cancer incidence and main underlying disease of ESRD in patients with RCC or TCC in the ANZDATA and our series are shown in table 4. Compared with ANZDATA series, it was noted that in our series the dialysis duration of all patients was significantly longer, and the occurrence of RCC was significantly higher while that of TCC was lower. There was a remarkable difference in the main primary renal diseases between our patients and ANZDATA patients with RCC or TCC. Toxic nephropathies were the main renal diseases in the ANZDATA patients with RCC or TCC, especially TCC, while there were no patients with toxic nephropathies in our patients on dialysis. DISCUSSION
The risk and pathogenesis of RCC and TCC in patients on dialysis may differ from those in the general population. The
95% CI
p Value
Sex male vs female: RCC 1.35 0.34–5.44 Not TCC 3.14 0.75–13.27 Not Diabetic nephropathy yes/no: RCC 0.77 0.06–9.37 Not TCC 1.43 0.33–6.16 Not Age at dialysis induction: RCC 1.05 0.96–1.15 Not TCC 0.99 0.92–1.05 Not Age at Ca diagnosis: RCC 1.08 0.95–1.23 Not TCC 1.01 0.94–1.09 Not Mos dialysis until Ca diagnosis: RCC 1.00 0.98–1.01 Not TCC 1.02 1.01–1.03 Stage high vs low: RCC 21.21 1.91–235.90 TCC 12.16 1.45–102.00 Site with/without upper urinary 6.11 0.70–3.54 Not tract (TCC)
significant significant significant significant significant significant significant significant significant 0.041 0.013 0.021 significant
kidney of ESRD shows tubular atrophy, interstitial inflammation and fibrosis, and arterial and glomerular sclerosis, resulting in acquired cysts due to the loss of structural integrity.3 Of these histopathological degenerations, acquired renal cystic disease alone has been implicated as the risk factor for malignant transformation.8, 9 The wall of acquired cysts is lined with hyperplastic epithelial cells, and the resultant papillary hyperplasia of cystic epithelium has a specific role in the development of RCC in patients on chronic dialysis. Acquired renal cystic disease appears in patients on hemodialysis and peritoneal dialysis independent of patient age or primary renal disease,10 and the prevalence increases with the duration of dialysis.11 Thus, RCC from the acquired cysts is associated with duration of dialysis. Also, the cysts may develop more rapidly in men than in women.11 On the other hand, uremic immune suppression, dialysis procedure and primary kidney diseases have been mentioned as the pathogenic conditions for the development of TCC in patients on dialysis.4 Although real pathogenesis still remains unclear, the toxic nephropathies account for the excess risk of TCC.2, 3 We identified 38 patients with RCC (0.61%) and 16 patients with TCC (0.26%) in the present review of 6,201 patients on dialysis. The patients with RCC compared to TCC were younger, had longer dialysis duration, and included more males and a larger number of patients with chronic glomerulonephritis. Of 38 patients with RCC 34 (89.5%) had acquired renal cysts, a finding similar to that of a previous report in which more than 80% of patients with RCC on dialysis had acquired renal cysts.9 Toxic nephropathies, a disease entity highly associated with carcinogenesis of urothelium,3, 12 are not a rare cause of primary renal diseases underlying ESRD in the USRDS, EDTA and ANZDATA, but are extremely rare in Japan.2, 3, 5 The major primary renal disease in our patients with TCC
1752
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS
TABLE 4. Comparison of mean dialysis duration of all patients, cancer frequencies and main primary renal disease of patients with RCC or TCC between the present study and the ANZDATA data Present Study Total No. pts Mean mos dialysis for all patients No. Ca incidence (%): RCC TCC No. main primary renal disease in RCC (%): Glomerulonephritis Toxic nephropathies No. main primary renal disease in TCC (%): Diabetic nephropathy Glomerulonephritis Toxic nephropathies
6,201 82.5
ANZDATA 13,497 30.6
p Value ⬍0.001 ⬍0.001
38 (0.61) 16 (0.26)
34 (0.25) 87 (0.64)
26 (68.4) 0 (0)
14 (41.2) 6 (17.6)
0.008
7 (43.8) 4 (25.0) 0 (0)
2 (2.3) 14 (16.1) 51 (58.6)
⬍0.001
was diabetic nephropathy (43.8%). This may reflect the fact that the major underlying disease among the newly initiated patients on dialysis is diabetic nephropathy (36.6%) in Japan.5 There is no report suggesting that TCC in patients on dialysis is related to diabetic nephropathy. Cigarette smoking and chemical exposure were other risk factors contributing to TCC,4 however, there was no information of smoking history in the USRDS, EDTA or ANZDATA data sets. Although we did not survey the smoking history in our cohort, 1 study showed that tumor stage had the highest correlation and smoking history had a low correlation with bladder cancer specific survival in Japanese patients not on dialysis.13 In the present series the mean dialysis duration until the detection of TCC was significantly shorter than that for RCC (53.5 vs 143.2 months), and 7 (43.8%) of 16 TCCs were diagnosed within 1 year of dialysis (data not shown). Stewart et al also described an excess of urinary tract cancers in the first year of dialysis.3 This may have come from the diagnostic evaluation at the induction of dialysis.3 Previous studies reported that the 5-year survival rate of patients with ESRD with kidney cancer was approximately 35%, similar to that of patients not on dialysis,9 and suggested that cancer screening was an inefficient allocation of resources, adding little to the life expectancy of these patients.14 However, regular screening of patients on dialysis may increase survival by encouraging radical treatments for kidney cancers at earlier tumor stage.15 In fact, the present study of patients on dialysis showed that 89.5% of those with RCC had low stage tumors, and that the overall and cancer specific 5-year survival of patients with RCC was 78.6% and 88.9%, respectively, thus indicating that the tumor stage was a significant risk factor for cancer death (RR⫽21.21). This favorable outcome of patients with RCC can be attributed to the regular imaging screening applied by most of our institutions. In contrast, only 56.3% of our patients with TCC had low stage tumors, and the overall and cancer specific 5-year survival of patients with TCC was 21.1% and 29.5%, respectively, again with tumor stage being a significant risk factor for cancer death (RR⫽12.6). A small study suggested that urinary cytology was a good screening method for the early detection of TCC in patients on dialysis with analgesic nephropathy.16 However, the screening urinary cytology was performed in only 1 of our affiliated dialysis centers, and no TCC was detected by screening urinary cytology or regular imaging. In addition to the delay in tumor diagnosis, older patient age and more aggressive underlying disease of diabetic nephropathy may be responsible for the poor outcome of patients with TCC. Since tumor stage is a significant risk factor for cancer specific death, effective diagnostic measures for early detection of TCC in patients on dialysis will be needed. In the general population the age standardized incidence rates of RCC per 100,000 of the Japanese population for men
and women were 4.9 and 1.8, respectively. The incidence rates of RCC in Japan were approximately the same as among Japanese in Los Angeles, and lower than North American and European countries.17 There was no nationwide survey for TCC in Japan or Australia and New Zealand. One regional report from Japan showed that the incidence rates of bladder carcinoma per 100,000 men and women were 5.3 and 1.7, respectively.18 Bladder cancer specific mortality in Japan was lower than in North America, European counties, and Australia and New Zealand.19 In patients on dialysis the incidence of TCC was significantly lower in our series compared to the ANZDATA data series, which comes from the difference in primary underlying renal diseases. Toxic nephropathies, a highly responsible disease for urothelial carcinoma (TCC),12 is more common as a primary underlying renal disease in the USRDS, EDTA and ANZDATA regions than in Japan.2, 3, 5 On the other hand, the excess rate of RCC in our patients may reflect the Japanese feature of renal replacement therapy. In Japan with the total number of patients on dialysis of 206,134 in 2000,5 only 744 renal transplantations were performed, a frequency approximated to one- twentieth of that in the United States.20 Consequently most patients with ESRD in Japan are treated with long-term dialysis which will accelerate the development of the acquired renal cyst and then RCC. Wu et al mentioned that RCC was common and TCC was rare in patients on dialysis in Western countries, while TCC is the most common carcinoma in Taiwanese and Japanese patients on dialysis.4 However, the present study, and the data of ANZDATA, EDTA and USRDA revealed that RCC is common and TCC is rare in patients on dialysis in Japan relative to Western countries.2, 3 The frequency and types of screening studies were other factors contributing the cancer detection rate and outcome in patients on dialysis, but there was no information on diagnostics, cancer stage or prognosis for RCC and TCC in the USRDS, EDTA or ANZDATA data series.
CONCLUSIONS
A total of 54 renal and urinary tract cancers (0.87%) were identified among 6,201 patients on dialysis in a Japanese cohort. The RCC incidence of 0.61% was significantly higher than that of 0.25% in the ANZDATA data, suggesting a long-term dialysis in our patients. Early detection by routine abdominal imaging examination may have contributed to the favorable RCC specific 5-year survival of 88.9% in our cohort. In contrast the TCC incidence of 0.26% was lower than that of 0.64% in the ANZDATA data, probably resulting from the lack of toxic nephropathies as an underlying renal disease in our patients. The significantly worse outcome of TCC indicated the need to develop early detection measures for TCC in patients on dialysis.
RENAL AND UROTHELIAL CARCINOMAS IN PATIENTS ON DIALYSIS REFERENCES
1. Vamvakas, S., Bahner, U. and Heidland, A.: Cancer in end-stage renal disease: potential factors involved - editorial-. Am J Nephrol, 18: 89, 1998 2. Maisonneuve, P., Agodoa, L., Gellert, R., Stewart, J. H., Buccianti, G., Lowenfels, A. B. et al: Cancers in patients on dialysis for end-stage renal disease: an international collaborative study. Lancet, 354: 93, 1999 3. Stewart, J. H., Buccianti, G., Agodoa, L., Gellert, R., McCredie, M. R., Lowenfels, A. B. et al: Cancers of the kidney and urinary tract in patients on dialysis for end-stage renal disease: analysis of data from the United States, Europe, and Australia and New Zealand. J Am Soc Nephrol, 14: 197, 2003 4. Wu, C.-F., Shee, J.-J., Ho, D.-R., Chen, W.-C. and Chen, C.-S.: Different treatment strategies for end stage renal disease in patients with transitional cell carcinoma. J Urol, 171: 126, 2004 5. Nakai, S., Shinzato, T., Sanaka, T., Kikuchi, K., Kitaoka, T., Shinoda, T. et al: The current state of chronic dialysis treatment in Japan. J Jpn Soc Dial Ther, 35: 1155, 2002 6. Novick, A. C. and Campbell, S. C.: Renal tumors. In: Campbell’s Urology, 8th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. and A. J. Wein. Philadelphia: W. B. Saunders Co., vol. 4, sect. X, chapt. 75, pp. 2672–2731, 2002 7. Messing, E. M.: Urothelial tumors of the urinary tract. In: Campbell’s Urology, 8th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. and A. J. Wein. Philadelphia: W. B. Saunders Co., vol. 4, sect. X, pp. 2732–2784, 2002 8. Hughson, M. D., Buchwald, D. and Fox, M.: Renal neoplasia and acquired cystic kidney disease in patients receiving long-term dialysis. Arch Pathol Lab Med, 110: 592, 1986 9. Matson, M. A. and Cohen, E. P.: Acquired cystic kidney disease: occurrence, prevalence, and renal cancers. Medicine (Baltimore), 69: 217, 1990 10. Park, J. H., Kim, Y. O., Park, J. H., Kim, B. S., Yoon, S. A., Yang, C. W. et al: Comparison of acquired cystic kidney disease
11. 12. 13. 14. 15. 16.
17. 18.
19. 20.
1753
between hemodialysis and continuous ambulatory peritoneal dialysis. Korean J Intern Med, 15: 51, 2000 Ishikawa, I., Saito, Y., Nakamura, M., Takada, K., Ishii, H., Nakazawa, T. et al: Fifteen-year follow-up of acquired renal cystic disease –a gender difference. Nephron, 75: 315, 1997 Blohme, I. and Johansson, S.: Renal pelvic neoplasms and atypical urothelium in patients with end-stage analgesic nephropathy. Kidney Int, 20: 671, 1981 Takashi, M., Murase, T., Mizuno, S., Hamajima, N. and Ohno, Y.: Multivariate evaluation of prognostic determinants in bladder cancer patients. Urol Int, 42: 368, 1987 Chertow, G. M., Paltiel, A. D., Owen, W. F., Jr. and Lazarus, J. M.: Cost-effectiveness of cancer screening in end-stage renal disease. Arch Intern Med, 156: 1345, 1996 Li, M. K., Choy, D. K. and Yip, S. K.: Renal cell carcinoma in patients with chronic renal failure. Ann Acad Med Singapore, 28: 512, 1999 Veltman, G. A., Bosch, F. H., van der Plas-Cats, M. B. and van Leusen, R.: Urinary cytology as a screening method for transitional-cell carcinoma in dialysis patients with analgesic nephropathy. Nephrol Dial Transplant, 6: 346, 1991 Marumo, K., Satomi, Y., Miyao, N., Hasegawa, M., Tomita, Y., Igarashi, T. et al: The prevalence of renal cell carcinoma: a nation-wide survey in Japan in 1997. Int J Urol, 8: 359, 2001 The Research Group for Population-Based Cancer Registration in Japan: Cancer Incidence in Japan, 1975-Cancer Registry Statistics. Cancer Mortality and Morbidity Statistics. GANN Monograph on Cancer Research, No. 26, 1981 Kurihara, M., Aoki, K. and Hisamichi, S.: Cancer Mortality Statistics in the World. Nagoya: The University of Nagoya Press, 1989 The Japanese Society for Clinical Renal Transplantation and the Japanese Society for Transplantation: Annual progress report from the Japanese renal transplant registry. The number of the renal transplantation in 2000. Jpn J Transplant, 36: 87, 2001