- Email: [email protected]
Renal Cell Carcinoma Invading the Urinary Collecting System: Implications for Staging
0022-5347/02/1676-2392/0 THE JOURNAL OF UROLOGY® Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.®
Vol. 167, 2392–2396, June 2002 Printed in U.S.A.
RENAL CELL CARCINOMA INVADING THE URINARY COLLECTING SYSTEM: IMPLICATIONS FOR STAGING ROBERT G. UZZO,* EDWARD E. CHERULLO, JONATHAN MYLES
AND
ANDREW C. NOVICK
From the Urological Institute, Cleveland Clinic Foundation, Cleveland, Ohio
ABSTRACT
Purpose: Current TNM staging of renal cell carcinoma is based on the tumor propensity for local extension (T), nodal involvement (N) and metastatic spread (M). Locally advanced renal cell carcinoma may involve the perirenal fat, adrenal glands, renal vein, vena cava and/or urinary collecting system. The existing TNM classification does not reflect the ability of renal cell carcinoma to invade the urothelium. We evaluated the incidence and characteristics as well as overall and cancer specific survival of renal cell carcinoma invading the urinary collecting system. Methods and Materials: We reviewed pathological findings in 504 kidneys from 475 patients with renal cell carcinoma who presented to our institution in a 3-year period. Urothelial involvement required evidence of gross or histological invasion of the renal calices, infundibulum, pelvis or ureter. Demographic and survival data were obtained from medical records and an institutional cancer registry for tumors invading the urothelium. Stage specific survival data were then compared with tumors not involving the urinary collecting system. Results: Definitive urothelial involvement by the primary tumor was interpretable in 426 of 504 kidneys. Invasion of the collecting system was identified in 61 of 426 cases (14%). Mean diameter of the invading lesions was 10.2 cm. (range 3 to 26). The majority of cases showed clear cell and sarcomatoid histology. Invasion by a papillary lesion was rare. Involvement of the collecting system was most common at the renal poles. Of 61 lesions invading the collecting system 48 (79%) were stage pT3 or greater, while only 13 (21%) were pathologically localized stage pT2 or less. Vascular invasion was identified in 38 renal cell carcinoma cases (62%) with urothelial involvement. A total of 16 cases (26%) were associated with vena caval thrombus. Invading tumors were high Fuhrman grade III or IV in 43 cases (70%). Overall disease specific survival was poor with a median of 19 months. In patients with localized stage pT1 or pT2N0M0 disease and urothelial invasion median disease specific survival was 46 months. Conclusions: Renal cell carcinoma lesions involving the renal collecting system are characteristically large, high grade and high stage. Clear cell carcinoma most commonly invades, while invasion by papillary tumors is rare. Overall the prognosis for high stage lesions with urothelial involvement is poor and does not appear significantly different from the reported disease specific survival of patients with high stage lesions without urothelial invasion. Localized tumors 4 cm. or less, which are amenable to elective nephron sparing surgery, rarely invade the urothelium. However, when a low stage pT2 or less renal lesion involves the urinary space, survival appears worse than equivalently staged renal cell carcinoma without invasion. Including urothelial invasion into current TNM staging systems for renal cell carcinoma is unlikely to provide significant additional prognostic or therapeutic information. KEY WORDS: kidney; carcinoma, renal cell; urothelium; neoplasm invasiveness; neoplasm staging
Epithelial tumors of the kidney account for approximately 3% of all solid neoplasms with adenocarcinoma or renal cell carcinoma representing almost 85%.1 Recent clinical surveys indicate that the incidence of this tumor is increasing, which may be attributable to incidental detection through the widespread use of noninvasive imaging techniques.2, 3 Data from the National Cancer Institute Surveillance, Epidemiology and End Results program show a steady increase in renal cell carcinoma of 2% to 4% yearly between 1975 and 1995.4 Approximately 25% to 30% of patients with renal cell carcinoma present with advanced stage disease.5 A delayed diagnosis may result from the ability of space occupying lesions of the retroperitoneum to become large before causing local symptoms. In addition, manifestations of renal cell carci-
noma are protean and may give rise to a constellation of nonspecific symptoms, thereby, delaying detection. Local renal tumor growth and extension may involve the perirenal fat, adrenal glands, renal vein, inferior vena cava, urinary collecting system and/or adjacent retroperitoneal structures. Current methods of staging renal cell carcinoma reflect the propensity of these tumors to extend locally (T), involve regional lymph nodes (N) and/or initiate metastatic growth (M). The recent TNM classification of renal cell carcinoma designates a separate tumor (T) stage or substage for local involvement of the renal capsule (pT2), perirenal fat and/or adrenal gland (pT3a), renal vein or inferior vena cava below the diaphragm (pT3b), inferior vena cava above the diaphragm (pT3c) or adjacent retroperitoneal structures outside of Gerota’s fascia (pT4).6 To our knowledge no studies have characterized renal cell carcinoma with urothelial invasion or included invasion of the urinary collecting system as a
Accepted for publication January 25, 2002. * Current address: Urologic Oncology, Fox Chase Cancer Center, Temple University School of Medicine, 7701 Burholme Ave., Philadelphia, Pennsylvania 19111. 2392
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM 7
criterion for staging renal cell carcinoma. Urothelial involvement by renal cell carcinoma may have important prognostic implications as well as practical operative concerns during nephron sparing surgery. We investigated the incidence, characteristics and survival of renal cell carcinoma invading the collecting system to determine whether a separate clinical stage or substage may be helpful. MATERIALS AND METHODS
Pathological findings in all cases of renal cell carcinoma presenting to our institution between January 1995 and December 1997 treated with radical or partial nephrectomy were retrospectively reviewed for involvement of the urinary collecting system by the primary tumor. In patients with bilateral disease each renal unit was reviewed independently for pathological but not survival purposes. Involvement of the urinary collecting system required gross and/or histological evidence of invasion of the calices, infundibulum, pelvis or ureter. During prosection of the specimens the adjacent urothelium was sampled and analyzed. Cases in which tumor compressed or displaced the collecting system but did not directly invade it were excluded from analysis. Pathology reports were reviewed to determine stage according to the pTNM system proposed in 1997 by the International Union Against Cancer.8 Tumor size was determined at pathological examination of the specimen as the largest dimension of the lesion. All tumor histological analysis was performed by an experienced uropathologist (J. M.) and graded using the Fuhrman classification system.9 Tumors were then evaluated for the number of established criteria independent of collecting system involvement requiring an advanced TNM stage designation (pT3 ⫹ N1 ⫹ M.⫹). These criteria included involvement of the ipsilateral adrenal gland, extracapsular extension into the perirenal fat, venous involvement, contiguous spread into adjacent retroperitoneal structures, lymph node involvement or metastatic disease. Demographic and survival data were obtained from medical records and an institutional cancer registry. Kaplan-Meier estimates for overall and disease specific survival were generated according to T stage, grade, nodal status and metastases. For overall survival death from any cause was counted as an event, while for disease specific survival only deaths from cancer were counted as events. The log rank test was used to compare survival according to the variable of interest. RESULTS
Definitive histopathological involvement or sparing of the collecting system by the primary tumor was interpretable in 426 of 504 renal specimens. Invasion of any portion of the
2393
collecting system, including the calices, infundibulum or renal pelvis, was identified in 61 (14%) of these specimens (fig. 1). No cases of local involvement of the ureter by tumor were identified. The 61 tumors invading the urothelium were surgically resected in 56 radical and 5 partial nephrectomy specimens (table 1). Involvement of the collecting system was more common in men (70.5%) than in women (29.5%), most likely reflecting the higher propensity for renal cell carcinoma in men.10 There was no predilection regarding the side of collecting system involvement in right (49%) and left (51%) renal specimens. Mean diameter of the 61 tumors invading the collecting system was 10.2 cm. (range 3 to 26). Of these lesions 27 (44%) were 10 cm. or greater in the largest diameter. In addition, 60 lesions (98.4%) identified as invading the collecting system were classified as Fuhrman nuclear grade II or higher and 43 (70.5%) were assigned a nuclear grade of III or IV. Only 1 low Fuhrman grade I lesion invaded the urothelium (fig. 2). Renal sinus involvement was identified in 45 renal cell carcinoma specimens (74%) invading the collecting system. Renal capsular invasion was identified in 49 specimens (80.3%) and 14 (23%) were multifocal. A total of 38 (62.3%) specimens had concurrent vascular invasion, including 16 (26%) associated with inferior venal caval thrombus. Involvement of the collecting system was most common at the renal poles with the upper and lower poles involved in 49% and 41% of cases, respectively (table 1). Clear cell carcinoma in 62% of cases and sarcomatoid variants in 8% were the most common histopathological designation among renal cell carcinoma specimens invading the renal collecting system. Only 1 papillary lesion (2%) was noted to invade the urothelium. Mixed histology was present in 17 specimens (28%), most commonly with a predominant clear cell pattern (table 1). Since urothelial involvement by renal cell carcinoma is not an established criterion of advanced stage disease, we calculated the number of accepted criteria shown by each tumor invading the renal collecting system to be designated high stage pT3 to 4, N1 to 2 and/or M1, including involvement of the perirenal fat, adrenal, renal vein, inferior vena cava, contiguous retroperitoneal structures, lymph nodes or metastatic sites. Overall 48 of the 61 tumors (79%) involving the renal collecting system were high stage independent of renal collecting system involvement. Of these lesions 16% met 1, 31% met 2, 13% met 3 and 18% met 4 established high stage criteria. Cumulatively 62% of the lesions invading the renal collecting system met 2 or more established criteria for high stage disease.
FIG. 1. Involvement of urinary collecting system by renal cell carcinoma. A, gross specimen shows tumor invading renal calices, pelvis, renal sinus and vasculature. B, histological study reveals uninvolved side (bottom) of urinary space with normal transitional epithelium and underlying smooth muscle. Clear cell carcinoma invading full thickness of collecting system with attenuation of transitional epithelium on opposite side of urinary space (top).
2394
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM
TABLE 1. Characteristics of renal cell carcinoma involving the urinary collecting system in 61 patients No. gender (%): Male Female No. operation (%): Radical nephrectomy Nephron sparing surgery Median mos. Ca specific survival % 5-Yr Ca specific survival Mean mos. followup (range) Mean cm. size (range) No. side (%): Rt. Lt. No. pole location (%): Upper Mid Lower No. histological findings (%): Clear cell Ca Sarcomatoid Papillary Mixed histology No. renal sinus involvement (%) No. multifocal lesions (%) No. Fuhrman nuclear grade (%): I II III/IV No. pathological T stage:6 pT1 pT2 pT3a pT3b pT3c pT4 No. node pos. (%) No. metastasis (%) No. stage* (%): I II III IV
43 18
(70) (30)
56 (92) 5 (8) 18.6 35.3 22 (2–64) 10.2 (3–26) 30 31
(49) (51)
30 6 25
(49) (10) (41)
38 5 1 17 45 14
(62) (8) (2) (28) (74) (23)
1 17 43
(2) (28) (70)
4 9 10 17 11 10 10 5
(7) (15) (16) (28) (18) (16) (16) (8)
4 9 26 22
(6) (15) (43) (36)
In only 13 (21%) tumors involving the renal pelvis was disease limited by the renal capsule (stages pT1 to 2N0M0). No renal lesion 4 cm. or less confined to the renal capsule involved the collecting system. Using the current TNM classification system, in which a cutoff of 7 cm. is used to distinguish stage pT1 from pT2 lesions, 4 (7%) tumors involving the renal collecting system were pathological stage pT1. They would be designated stage pT1b lesions (size greater than 4 to 7 cm. or less ) under the proposed revision of the current TNM system for renal cell carcinoma.11 Therefore, no localized tumors amenable to elective nephron sparing surgery involved the urinary collecting system. Table 2 lists the characteristics of low stages pT1 to 2 localized lesions invading the urothelial tract. Of the tumors 77% were clear cell carcinoma of high nuclear grade. No tumor confined by the renal capsule yet invading the collecting system involved local lymph nodes or distant metastatic sites. Overall median disease specific survival in patients with renal cell carcinoma involving the urothelial tract was 19 months with an estimated 5-year disease specific survival of 35%. In these patients existing local pathological tumor (T) staging successfully predicted and discriminated disease specific survival based only on pathological T stage of the primary tumor (p ⫽ 0.016, fig. 3, A). Median disease specific survival in patients with stage I or II (pT1 or T2 N0M0) renal cell carcinoma and urothelial involvement was 46 months compared with 24 months in those with stage III (pT3N0 to 1) disease and 7 months in those with stage IV (pT4N0 to 1 M0 to 1) disease (fig. 3, B). This difference was statistically significant (p ⫽ 0.023), suggesting that the overall TNM staging system currently in use successfully stratified disease specific survival in patients with renal cell carcinoma involving the collecting system.
DISCUSSION
The incidence of renal cell carcinoma has slowly increased in the last several decades.4 Surgical extirpation remains the most widely accepted and potentially curative treatment option. When it is locally advanced or disseminated, the prognosis in patients with this disease is generally unfavorable given its relative resistance to adjuvant immunotherapy and chemotherapy.5 When planning optimal therapy in patients with renal cell carcinoma, information on the natural history and behavior of the tumor is essential. A number of patient and tumor related prognostic factors for renal cell carcinoma have been proposed, including weight loss, performance status, anemia, serum calcium and alkaline phosphatase, nuclear morphometry,12 flow cytometry and locally advanced or metastatic disease.13, 14 Yasunaga et al performed multivariate analysis that identified pathological stage and nuclear grade as the most significant predictors of renal cell carcinoma behavior and patient outcome.15 Tumor size alone was also implicated as a separate prognostic indicator of tumor behavior.16, 17 Most authors agree that pathological TNM stage remains the most useful determinant of patient outcome in renal cell carcinoma18, 19 pending the discovery and development of specific molecular markers.20 Given the importance of stage for predicting biological behavior and subsequent patient outcome, close consideration and analysis must be given to identify the most salient criteria for tumor staging. New imaging techniques, including 3-dimensional computerized tomography, have improved our ability to characterize local involvement by these lesions preoperatively.21 After tumor characteristics are identified clinical stage can be assigned and a treatment strategy can be implemented. In 1997 the International Union Against Cancer and American Joint Committee on Cancer revised the TNM staging system for renal cell carcinoma.6 The major revision from the previous TNM system was in the size delineation of stages pT1 and pT2 lesions, which is now set at 7 cm. Since then, it has been suggested that the T1 category should be subdivided into stages T1a (lesions 4 cm. or less) and T1b (lesions greater than 4 to 7 cm. or less).11, 22 It was believed that this subdivision would better reflect patient outcomes after treatment with nephron sparing surgery. To our knowledge there are no data to date on the incidence or implications of renal cell carcinoma invading the urothelial compartment of the kidney. We present the first study of renal cell carcinoma invading the collecting system that includes tumor characteristics and outcomes. Our data show that most renal cell carcinomas that invade the renal collecting system have associated features that independently characterize these lesions as high stage. The mean size of this type of lesion in our study was 10.2 cm. Of the lesions 79% met at least 1 and 62% met 2 or more criteria for high stage III or IV disease independent of renal collecting system involvement. In addition, histological evaluation revealed that 62% were clear cell type, 28% were mixed histology with a primary clear cell component and 8% were sarcomatoid lesions. Only 1 renal cell carcinoma lesions (2%) invading the renal urothelium was identified as papillary. In this series after considering all T stages, relatively few renal cell carcinomas invaded the upper tract urothelium in the absence of at least 1 other criterion of high stage disease. We believe that the association of urothelial invasion with other poor prognostic pathological findings reflects the overall aggressive biology of these tumors. Survival in patients with renal cell carcinoma decreases with increasing local tumor stage. Large published reports following radical nephrectomy demonstrate excellent prognostic stratification in regard to 5-year survival rates based on existing TNM staging criteria. Patients with early stage
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM
2395
FIG. 2. Tumor size and grade of renal cell carcinoma lesions involving urinary collecting system. Low Fuhrman grade I lesions rarely invaded renal pelvis or its components. Lesions invading collecting system tended to be moderate to high Fuhrman grades II to III/IV. Greatest mean diameter of these tumors was 10.2 cm. (range 3 to 26 cm.). Involvement by lesions 4 cm. or less was rare.
TABLE 2. Characteristics of stages pT1 and pT26 renal cell carcinoma involving the urinary collecting system in 13 patients No. gender (%): Male Female No. operation (%): Radical nephrectomy Nephron sparing surgery No. pole location (%): Upper Mid Lower Mean cm. size (range) No. cm. (%): Less than 4 Greater than 4–7 or less Greater than 7–10 or less Greater than 10–14 or less Greater than 14 No. histological findings (%): Clear cell Ca Mixed histogenesis No. Fuhrman nuclear grade (%): I II III/IV Node Positive (N⫹) Metastasis (M⫹) Median mos. disease specific survival
10 3
(77) (23)
10 3
(77) (23)
5 (38) 1 (8) 7 (54) 9.3 (5.5–14.5) 0 4 6 2 1
(31) (46) (15) (8)
10 3
(77) (23)
1 5 7 0 0 46
(8) (38) (54)
T1 disease have a reported 5-year disease specific survival of 80% to 100%. Those with stage T2 tumors can expect 70% to 80% 5-year disease specific survival, while the rate in those with stage T3 lesions is 50% to 60%. The prognosis in patients with locally invasive stage T4 lesions, lymph node metastasis or distal disease is poor with a 5-year disease specific survival of 5% to 10%.14, 19, 23, 24 Long-term stage specific data on nephron sparing surgery have been reported.25, 26 Recent 10-year followup data at our institution on patients who underwent partial nephrectomy included none with renal cell carcinoma involving the urothelial tract.25 In this cohort the 5-year cancer specific survival rates was identical to that published for radical nephrectomy at 95% to 100% for stages pT1 to 2 lesions, 85% for stage pT3a lesions and 59% for stage pT3b lesions. Our current data on renal cell carcinoma invading the urinary collecting system show equivalent median and 5-year disease specific survival rates in patients with high stage disease, suggesting that in the presence of other poor prognostic indicators associated with high stage disease urothelial involvement did not significantly worsen the prognosis of these biologically aggressive
FIG. 3. Kaplan Meier estimates of disease specific survival in patients with renal cell carcinoma invading urinary collecting system. A, according to pathological T stage. B, according to overall TNM stage.6
tumors. However, in patients with localized stages pT1 to 2N0M0 renal cell carcinoma invading the urinary space our median disease specific survival of 46 months is significantly worse than in published reports of radical nephrectomy or nephron sparing approaches for comparable low stage lesions and no urothelial involvement.14, 19, 25 Cases of low stage
2396
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM
localized renal cell carcinoma with urothelial involvement are uncommon, accounting for only 3% of all cases of renal cell carcinoma reviewed. However, urothelial involvement in low stage tumors should be noted due to its potential prognostic implications. Consideration should be given to adjuvant therapy in these patients. Future prospective studies corroborating these findings are warranted. When specifically evaluating tumor size, only 2 lesions 4 cm. or less invaded the renal collecting system. However, in each case these tumors were small and central, and showed other aggressive biological features, in that they invaded the renal vein or vena cava (stage pT3b). No tumors 4 cm. or less that were confined by the renal capsule invaded the urothelium. This finding has important implications when considering elective nephron sparing surgery for small solitary lesions. Under these circumstances the adjacent collecting system is rarely involved and surgical resection into the collecting system is typically unnecessary to provide a negative margin. The majority of renal tumors amenable to elective partial nephrectomy do not invade the renal collecting system. A potential shortcoming of our study is the retrospective nature of these data, which may limit interpretation. Prospective data collection is currently under way at our institution to corroborate these findings and determine the most significant clinical and biological prognostic factors that may help to clarify uncertainty in patients with this disease and select appropriate candidates for adjuvant therapy. CONCLUSIONS
Renal cell carcinoma invaded the urinary collecting system in 61 of the 426 patients (14%) in our cohort. Tumors invading the renal collecting system are characteristically large with high nuclear grade and predominantly high stage. Invasion of the renal collecting system by papillary tumors is rare, while invasion of the collecting system by clear cell carcinoma or tumors with sarcomatoid elements is more common. Involvement of the renal collecting system by renal cell carcinoma is most common at the renal poles, possibly due to the more peripheral location of the polar calices. Involvement of the urinary collecting system by renal cell carcinoma is rare in the absence of other poor prognostic features associated with the primary tumor. Although urothelial invasion by low stage localized renal cell carcinoma is uncommon, when it occurs, it portends a worse prognosis. In contrast, high stage renal cell carcinoma invading the urothelium does not appear to have a worse prognosis than equivalent pathological T stage lesions that do not invade the collecting system. In the majority of cases of renal cell carcinoma with urothelial involvement invasion is unlikely to impact management independently. For these reasons we believe that including urinary collecting system invasion into the TNM staging system is not warranted at this time. Tumors amenable to elective nephron sparing surgery are unlikely to involve the urothelium. Lisa A. Rybicki, Department of Biostatistics and Epidemiology assisted with the statistical analysis. REFERENCES
1. Landis, S. H., Murray, T., Bolden, S. and Wingo, P. A.: Cancer statistics. CA Cancer J Clin, 49: 8, 1999 2. Rodriguez, R., Fishman, E. K. and Marshall, F. F.: Differential diagnosis and evaluation of the incidentally discovered renal mass. Semin Urol Oncol, 13: 246, 1995 3. Vallancien, G., Torres, L. O., Gurfinkel, E., Veillon, B. and Brisset, J. M.: Incidental detection of renal tumours by abdominal ultrasonography. Eur Urol, 18: 94, 1990 4. Chow, W. H., Devesa, S. S., Warren, J. L. and Fraumeni, J. F., Jr.: Rising incidence of renal cell cancer in the United States. JAMA, 281: 1628, 1999 5. Motzer, R. J. and Russo, P.: Systemic therapy for renal cell
carcinoma. J Urol, 163: 408, 2000 6. Fleming, I. D., Cooper, J. S., Henson, D. E., Hutter, R. V., Kennedy, B. J., Murphy, G. P. et al: AJCC Cancer Staging Manual, 5th ed. Philadelphia: Lippincott Williams and Wilkins, pp. 215–217, 1997 7. Guinan, P., Sobin, L. H., Algaba, F., Badellino, F., Kameyama, S., MacLennan, G. et al: TNM staging of renal cell carcinoma. Workshop No. 3. Union International Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 992, 1997 8. Sobin, L. H. and Wittekind, C. H.: International Union Against Cancer: TNM Classification of Malignant Tumours, 5th ed. New York: Wiley-Liss, pp. 180 –182, 1997 9. Fuhrman, S. A., Lasky, L. C. and Limas, C.: Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol, 6: 655, 1982 10. Uzzo, R. G. and Novick, A. C.: Nephron sparing surgery for renal tumors: indications, techniques and outcomes. J Urol, 166: 6, 2001 11. Hafez, K. S., Fergany, A. F. and Novick, A. C.: Nephron sparing surgery for localized renal cell carcinoma: impact of tumor size on patient survival, tumor recurrence and TNM staging. J Urol, 162: 1930, 1999 12. Carducci, M. A., Piantadosi, S., Pound, C. R., Epstein, J. I., Simons, J. W., Marshall, F. F. et al: Nuclear morphometry adds significant prognostic information to stage and grade for renal cell carcinoma. Urology, 53: 44, 1999 13. Elson, P. J.: Prognostic factors in metastatic renal cell carcinoma. In: Renal Cell Carcinoma: Molecular Biology, Immunology, and Clinical Management. Edited by R. M. Bukowski and A. C. Novick. New Jersey: Humana Press, pp. 147–160, 2000 14. Jennings, S. B. and Linehan, W. M.: Renal, perirenal, and ureteral neoplasms. In: Adult and Pediatric Urology, 3rd ed. Edited by J. Y. Gillenwater, J. T. Grayhack, S. S. Howards and J. W. Duckett. St. Louis: Mosby Press, pp. 643– 694, 1996 15. Yasunaga, Y., Shin, M., Miki, T., Okuyama, A. and Aozasa, K.: Prognostic factors of renal cell carcinoma: a multivariate analysis. J Surg Oncol, 68: 11, 1998 16. Guinan, P. D., Vogelzang, N. J., Fremgen, A. M., Chmiel, J. S., Sylvester, J. L., Sener, S. F. et al: Renal cell carcinoma: tumor size, stage and survival. J Urol, 153: 901, 1995 17. Kinouchi, T., Saiki, S., Meguro, N., Maeda, O., Kuroda, M., Usami, M. et al: Impact of tumor size on the clinical outcomes of patients with Robson stage I renal cell carcinoma. Cancer, 85: 689, 1999 18. Ulchaker, J. C. and Klein, E. A.: Biology of metastasis and its clinical implications: renal cell cancer. World J Urol, 14: 175, 1996 19. Thrasher, J. B. and Paulson, D. F.: Prognostic factors in renal cancer. Urol Clin North Am, 20: 247, 1993 20. Hofmockel, G., Tsatalpas, P., Muller, H., Da¨ mmrich, J., Poot, M., Maurer-Schultze, B. et al: Significance of conventional and new prognostic factors for locally confined renal cell carcinoma. Cancer, 76: 296, 1995 21. Coll, D. M., Uzzo, R. G., Herts, B. R., Davros, W. J., Wirth, S. L. and Novick, A. C.: 3-dimensional volume rendered computerized tomography for preoperative evaluation and intraoperative treatment of patients undergoing nephron sparing surgery. J Urol, 161: 1097, 1999 22. Igarashi, T., Tobe, T., Nakatsu, H.-O., Suzuki, N., Murakami, S., Hamano, M. et al: The impact of a 4 cm. cutoff point for stratification of T1NoMo renal cell carcinoma after radical nephrectomy. J Urol, 165: 1103, 2001 23. Javidan, J., Stricker, H. J., Tamboli, P., Amin, M. B., Peabody, J. O., Deshpande, A. et al: Prognostic significance of the 1997 TNM classification of renal cell carcinoma. J Urol, 162: 1277, 1999 24. Moch, H., Gasser, T., Amin, M. B., Torhorst, J., Sauter, G. and Mihatsch M. J.: Prognostic utility of the recently recommended histologic classification and revised TNM staging system of renal cell carcinoma: a Swiss experience with 588 tumors. Cancer, 89: 604, 2000 25. Fergany, A. F., Hafez, K. S. and Novick, A. C.: Long-term results of nephron sparing surgery for localized renal cell carcinoma: 10-year followup. J Urol, 163: 442, 2000 26. Herr, H. W.: Partial nephrectomy for unilateral renal cell carcinoma and a normal contralateral kidney: 10-year followup. J Urol, 161: 33, 1999
Vol. 167, 2392–2396, June 2002 Printed in U.S.A.
RENAL CELL CARCINOMA INVADING THE URINARY COLLECTING SYSTEM: IMPLICATIONS FOR STAGING ROBERT G. UZZO,* EDWARD E. CHERULLO, JONATHAN MYLES
AND
ANDREW C. NOVICK
From the Urological Institute, Cleveland Clinic Foundation, Cleveland, Ohio
ABSTRACT
Purpose: Current TNM staging of renal cell carcinoma is based on the tumor propensity for local extension (T), nodal involvement (N) and metastatic spread (M). Locally advanced renal cell carcinoma may involve the perirenal fat, adrenal glands, renal vein, vena cava and/or urinary collecting system. The existing TNM classification does not reflect the ability of renal cell carcinoma to invade the urothelium. We evaluated the incidence and characteristics as well as overall and cancer specific survival of renal cell carcinoma invading the urinary collecting system. Methods and Materials: We reviewed pathological findings in 504 kidneys from 475 patients with renal cell carcinoma who presented to our institution in a 3-year period. Urothelial involvement required evidence of gross or histological invasion of the renal calices, infundibulum, pelvis or ureter. Demographic and survival data were obtained from medical records and an institutional cancer registry for tumors invading the urothelium. Stage specific survival data were then compared with tumors not involving the urinary collecting system. Results: Definitive urothelial involvement by the primary tumor was interpretable in 426 of 504 kidneys. Invasion of the collecting system was identified in 61 of 426 cases (14%). Mean diameter of the invading lesions was 10.2 cm. (range 3 to 26). The majority of cases showed clear cell and sarcomatoid histology. Invasion by a papillary lesion was rare. Involvement of the collecting system was most common at the renal poles. Of 61 lesions invading the collecting system 48 (79%) were stage pT3 or greater, while only 13 (21%) were pathologically localized stage pT2 or less. Vascular invasion was identified in 38 renal cell carcinoma cases (62%) with urothelial involvement. A total of 16 cases (26%) were associated with vena caval thrombus. Invading tumors were high Fuhrman grade III or IV in 43 cases (70%). Overall disease specific survival was poor with a median of 19 months. In patients with localized stage pT1 or pT2N0M0 disease and urothelial invasion median disease specific survival was 46 months. Conclusions: Renal cell carcinoma lesions involving the renal collecting system are characteristically large, high grade and high stage. Clear cell carcinoma most commonly invades, while invasion by papillary tumors is rare. Overall the prognosis for high stage lesions with urothelial involvement is poor and does not appear significantly different from the reported disease specific survival of patients with high stage lesions without urothelial invasion. Localized tumors 4 cm. or less, which are amenable to elective nephron sparing surgery, rarely invade the urothelium. However, when a low stage pT2 or less renal lesion involves the urinary space, survival appears worse than equivalently staged renal cell carcinoma without invasion. Including urothelial invasion into current TNM staging systems for renal cell carcinoma is unlikely to provide significant additional prognostic or therapeutic information. KEY WORDS: kidney; carcinoma, renal cell; urothelium; neoplasm invasiveness; neoplasm staging
Epithelial tumors of the kidney account for approximately 3% of all solid neoplasms with adenocarcinoma or renal cell carcinoma representing almost 85%.1 Recent clinical surveys indicate that the incidence of this tumor is increasing, which may be attributable to incidental detection through the widespread use of noninvasive imaging techniques.2, 3 Data from the National Cancer Institute Surveillance, Epidemiology and End Results program show a steady increase in renal cell carcinoma of 2% to 4% yearly between 1975 and 1995.4 Approximately 25% to 30% of patients with renal cell carcinoma present with advanced stage disease.5 A delayed diagnosis may result from the ability of space occupying lesions of the retroperitoneum to become large before causing local symptoms. In addition, manifestations of renal cell carci-
noma are protean and may give rise to a constellation of nonspecific symptoms, thereby, delaying detection. Local renal tumor growth and extension may involve the perirenal fat, adrenal glands, renal vein, inferior vena cava, urinary collecting system and/or adjacent retroperitoneal structures. Current methods of staging renal cell carcinoma reflect the propensity of these tumors to extend locally (T), involve regional lymph nodes (N) and/or initiate metastatic growth (M). The recent TNM classification of renal cell carcinoma designates a separate tumor (T) stage or substage for local involvement of the renal capsule (pT2), perirenal fat and/or adrenal gland (pT3a), renal vein or inferior vena cava below the diaphragm (pT3b), inferior vena cava above the diaphragm (pT3c) or adjacent retroperitoneal structures outside of Gerota’s fascia (pT4).6 To our knowledge no studies have characterized renal cell carcinoma with urothelial invasion or included invasion of the urinary collecting system as a
Accepted for publication January 25, 2002. * Current address: Urologic Oncology, Fox Chase Cancer Center, Temple University School of Medicine, 7701 Burholme Ave., Philadelphia, Pennsylvania 19111. 2392
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM 7
criterion for staging renal cell carcinoma. Urothelial involvement by renal cell carcinoma may have important prognostic implications as well as practical operative concerns during nephron sparing surgery. We investigated the incidence, characteristics and survival of renal cell carcinoma invading the collecting system to determine whether a separate clinical stage or substage may be helpful. MATERIALS AND METHODS
Pathological findings in all cases of renal cell carcinoma presenting to our institution between January 1995 and December 1997 treated with radical or partial nephrectomy were retrospectively reviewed for involvement of the urinary collecting system by the primary tumor. In patients with bilateral disease each renal unit was reviewed independently for pathological but not survival purposes. Involvement of the urinary collecting system required gross and/or histological evidence of invasion of the calices, infundibulum, pelvis or ureter. During prosection of the specimens the adjacent urothelium was sampled and analyzed. Cases in which tumor compressed or displaced the collecting system but did not directly invade it were excluded from analysis. Pathology reports were reviewed to determine stage according to the pTNM system proposed in 1997 by the International Union Against Cancer.8 Tumor size was determined at pathological examination of the specimen as the largest dimension of the lesion. All tumor histological analysis was performed by an experienced uropathologist (J. M.) and graded using the Fuhrman classification system.9 Tumors were then evaluated for the number of established criteria independent of collecting system involvement requiring an advanced TNM stage designation (pT3 ⫹ N1 ⫹ M.⫹). These criteria included involvement of the ipsilateral adrenal gland, extracapsular extension into the perirenal fat, venous involvement, contiguous spread into adjacent retroperitoneal structures, lymph node involvement or metastatic disease. Demographic and survival data were obtained from medical records and an institutional cancer registry. Kaplan-Meier estimates for overall and disease specific survival were generated according to T stage, grade, nodal status and metastases. For overall survival death from any cause was counted as an event, while for disease specific survival only deaths from cancer were counted as events. The log rank test was used to compare survival according to the variable of interest. RESULTS
Definitive histopathological involvement or sparing of the collecting system by the primary tumor was interpretable in 426 of 504 renal specimens. Invasion of any portion of the
2393
collecting system, including the calices, infundibulum or renal pelvis, was identified in 61 (14%) of these specimens (fig. 1). No cases of local involvement of the ureter by tumor were identified. The 61 tumors invading the urothelium were surgically resected in 56 radical and 5 partial nephrectomy specimens (table 1). Involvement of the collecting system was more common in men (70.5%) than in women (29.5%), most likely reflecting the higher propensity for renal cell carcinoma in men.10 There was no predilection regarding the side of collecting system involvement in right (49%) and left (51%) renal specimens. Mean diameter of the 61 tumors invading the collecting system was 10.2 cm. (range 3 to 26). Of these lesions 27 (44%) were 10 cm. or greater in the largest diameter. In addition, 60 lesions (98.4%) identified as invading the collecting system were classified as Fuhrman nuclear grade II or higher and 43 (70.5%) were assigned a nuclear grade of III or IV. Only 1 low Fuhrman grade I lesion invaded the urothelium (fig. 2). Renal sinus involvement was identified in 45 renal cell carcinoma specimens (74%) invading the collecting system. Renal capsular invasion was identified in 49 specimens (80.3%) and 14 (23%) were multifocal. A total of 38 (62.3%) specimens had concurrent vascular invasion, including 16 (26%) associated with inferior venal caval thrombus. Involvement of the collecting system was most common at the renal poles with the upper and lower poles involved in 49% and 41% of cases, respectively (table 1). Clear cell carcinoma in 62% of cases and sarcomatoid variants in 8% were the most common histopathological designation among renal cell carcinoma specimens invading the renal collecting system. Only 1 papillary lesion (2%) was noted to invade the urothelium. Mixed histology was present in 17 specimens (28%), most commonly with a predominant clear cell pattern (table 1). Since urothelial involvement by renal cell carcinoma is not an established criterion of advanced stage disease, we calculated the number of accepted criteria shown by each tumor invading the renal collecting system to be designated high stage pT3 to 4, N1 to 2 and/or M1, including involvement of the perirenal fat, adrenal, renal vein, inferior vena cava, contiguous retroperitoneal structures, lymph nodes or metastatic sites. Overall 48 of the 61 tumors (79%) involving the renal collecting system were high stage independent of renal collecting system involvement. Of these lesions 16% met 1, 31% met 2, 13% met 3 and 18% met 4 established high stage criteria. Cumulatively 62% of the lesions invading the renal collecting system met 2 or more established criteria for high stage disease.
FIG. 1. Involvement of urinary collecting system by renal cell carcinoma. A, gross specimen shows tumor invading renal calices, pelvis, renal sinus and vasculature. B, histological study reveals uninvolved side (bottom) of urinary space with normal transitional epithelium and underlying smooth muscle. Clear cell carcinoma invading full thickness of collecting system with attenuation of transitional epithelium on opposite side of urinary space (top).
2394
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM
TABLE 1. Characteristics of renal cell carcinoma involving the urinary collecting system in 61 patients No. gender (%): Male Female No. operation (%): Radical nephrectomy Nephron sparing surgery Median mos. Ca specific survival % 5-Yr Ca specific survival Mean mos. followup (range) Mean cm. size (range) No. side (%): Rt. Lt. No. pole location (%): Upper Mid Lower No. histological findings (%): Clear cell Ca Sarcomatoid Papillary Mixed histology No. renal sinus involvement (%) No. multifocal lesions (%) No. Fuhrman nuclear grade (%): I II III/IV No. pathological T stage:6 pT1 pT2 pT3a pT3b pT3c pT4 No. node pos. (%) No. metastasis (%) No. stage* (%): I II III IV
43 18
(70) (30)
56 (92) 5 (8) 18.6 35.3 22 (2–64) 10.2 (3–26) 30 31
(49) (51)
30 6 25
(49) (10) (41)
38 5 1 17 45 14
(62) (8) (2) (28) (74) (23)
1 17 43
(2) (28) (70)
4 9 10 17 11 10 10 5
(7) (15) (16) (28) (18) (16) (16) (8)
4 9 26 22
(6) (15) (43) (36)
In only 13 (21%) tumors involving the renal pelvis was disease limited by the renal capsule (stages pT1 to 2N0M0). No renal lesion 4 cm. or less confined to the renal capsule involved the collecting system. Using the current TNM classification system, in which a cutoff of 7 cm. is used to distinguish stage pT1 from pT2 lesions, 4 (7%) tumors involving the renal collecting system were pathological stage pT1. They would be designated stage pT1b lesions (size greater than 4 to 7 cm. or less ) under the proposed revision of the current TNM system for renal cell carcinoma.11 Therefore, no localized tumors amenable to elective nephron sparing surgery involved the urinary collecting system. Table 2 lists the characteristics of low stages pT1 to 2 localized lesions invading the urothelial tract. Of the tumors 77% were clear cell carcinoma of high nuclear grade. No tumor confined by the renal capsule yet invading the collecting system involved local lymph nodes or distant metastatic sites. Overall median disease specific survival in patients with renal cell carcinoma involving the urothelial tract was 19 months with an estimated 5-year disease specific survival of 35%. In these patients existing local pathological tumor (T) staging successfully predicted and discriminated disease specific survival based only on pathological T stage of the primary tumor (p ⫽ 0.016, fig. 3, A). Median disease specific survival in patients with stage I or II (pT1 or T2 N0M0) renal cell carcinoma and urothelial involvement was 46 months compared with 24 months in those with stage III (pT3N0 to 1) disease and 7 months in those with stage IV (pT4N0 to 1 M0 to 1) disease (fig. 3, B). This difference was statistically significant (p ⫽ 0.023), suggesting that the overall TNM staging system currently in use successfully stratified disease specific survival in patients with renal cell carcinoma involving the collecting system.
DISCUSSION
The incidence of renal cell carcinoma has slowly increased in the last several decades.4 Surgical extirpation remains the most widely accepted and potentially curative treatment option. When it is locally advanced or disseminated, the prognosis in patients with this disease is generally unfavorable given its relative resistance to adjuvant immunotherapy and chemotherapy.5 When planning optimal therapy in patients with renal cell carcinoma, information on the natural history and behavior of the tumor is essential. A number of patient and tumor related prognostic factors for renal cell carcinoma have been proposed, including weight loss, performance status, anemia, serum calcium and alkaline phosphatase, nuclear morphometry,12 flow cytometry and locally advanced or metastatic disease.13, 14 Yasunaga et al performed multivariate analysis that identified pathological stage and nuclear grade as the most significant predictors of renal cell carcinoma behavior and patient outcome.15 Tumor size alone was also implicated as a separate prognostic indicator of tumor behavior.16, 17 Most authors agree that pathological TNM stage remains the most useful determinant of patient outcome in renal cell carcinoma18, 19 pending the discovery and development of specific molecular markers.20 Given the importance of stage for predicting biological behavior and subsequent patient outcome, close consideration and analysis must be given to identify the most salient criteria for tumor staging. New imaging techniques, including 3-dimensional computerized tomography, have improved our ability to characterize local involvement by these lesions preoperatively.21 After tumor characteristics are identified clinical stage can be assigned and a treatment strategy can be implemented. In 1997 the International Union Against Cancer and American Joint Committee on Cancer revised the TNM staging system for renal cell carcinoma.6 The major revision from the previous TNM system was in the size delineation of stages pT1 and pT2 lesions, which is now set at 7 cm. Since then, it has been suggested that the T1 category should be subdivided into stages T1a (lesions 4 cm. or less) and T1b (lesions greater than 4 to 7 cm. or less).11, 22 It was believed that this subdivision would better reflect patient outcomes after treatment with nephron sparing surgery. To our knowledge there are no data to date on the incidence or implications of renal cell carcinoma invading the urothelial compartment of the kidney. We present the first study of renal cell carcinoma invading the collecting system that includes tumor characteristics and outcomes. Our data show that most renal cell carcinomas that invade the renal collecting system have associated features that independently characterize these lesions as high stage. The mean size of this type of lesion in our study was 10.2 cm. Of the lesions 79% met at least 1 and 62% met 2 or more criteria for high stage III or IV disease independent of renal collecting system involvement. In addition, histological evaluation revealed that 62% were clear cell type, 28% were mixed histology with a primary clear cell component and 8% were sarcomatoid lesions. Only 1 renal cell carcinoma lesions (2%) invading the renal urothelium was identified as papillary. In this series after considering all T stages, relatively few renal cell carcinomas invaded the upper tract urothelium in the absence of at least 1 other criterion of high stage disease. We believe that the association of urothelial invasion with other poor prognostic pathological findings reflects the overall aggressive biology of these tumors. Survival in patients with renal cell carcinoma decreases with increasing local tumor stage. Large published reports following radical nephrectomy demonstrate excellent prognostic stratification in regard to 5-year survival rates based on existing TNM staging criteria. Patients with early stage
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM
2395
FIG. 2. Tumor size and grade of renal cell carcinoma lesions involving urinary collecting system. Low Fuhrman grade I lesions rarely invaded renal pelvis or its components. Lesions invading collecting system tended to be moderate to high Fuhrman grades II to III/IV. Greatest mean diameter of these tumors was 10.2 cm. (range 3 to 26 cm.). Involvement by lesions 4 cm. or less was rare.
TABLE 2. Characteristics of stages pT1 and pT26 renal cell carcinoma involving the urinary collecting system in 13 patients No. gender (%): Male Female No. operation (%): Radical nephrectomy Nephron sparing surgery No. pole location (%): Upper Mid Lower Mean cm. size (range) No. cm. (%): Less than 4 Greater than 4–7 or less Greater than 7–10 or less Greater than 10–14 or less Greater than 14 No. histological findings (%): Clear cell Ca Mixed histogenesis No. Fuhrman nuclear grade (%): I II III/IV Node Positive (N⫹) Metastasis (M⫹) Median mos. disease specific survival
10 3
(77) (23)
10 3
(77) (23)
5 (38) 1 (8) 7 (54) 9.3 (5.5–14.5) 0 4 6 2 1
(31) (46) (15) (8)
10 3
(77) (23)
1 5 7 0 0 46
(8) (38) (54)
T1 disease have a reported 5-year disease specific survival of 80% to 100%. Those with stage T2 tumors can expect 70% to 80% 5-year disease specific survival, while the rate in those with stage T3 lesions is 50% to 60%. The prognosis in patients with locally invasive stage T4 lesions, lymph node metastasis or distal disease is poor with a 5-year disease specific survival of 5% to 10%.14, 19, 23, 24 Long-term stage specific data on nephron sparing surgery have been reported.25, 26 Recent 10-year followup data at our institution on patients who underwent partial nephrectomy included none with renal cell carcinoma involving the urothelial tract.25 In this cohort the 5-year cancer specific survival rates was identical to that published for radical nephrectomy at 95% to 100% for stages pT1 to 2 lesions, 85% for stage pT3a lesions and 59% for stage pT3b lesions. Our current data on renal cell carcinoma invading the urinary collecting system show equivalent median and 5-year disease specific survival rates in patients with high stage disease, suggesting that in the presence of other poor prognostic indicators associated with high stage disease urothelial involvement did not significantly worsen the prognosis of these biologically aggressive
FIG. 3. Kaplan Meier estimates of disease specific survival in patients with renal cell carcinoma invading urinary collecting system. A, according to pathological T stage. B, according to overall TNM stage.6
tumors. However, in patients with localized stages pT1 to 2N0M0 renal cell carcinoma invading the urinary space our median disease specific survival of 46 months is significantly worse than in published reports of radical nephrectomy or nephron sparing approaches for comparable low stage lesions and no urothelial involvement.14, 19, 25 Cases of low stage
2396
RENAL CELL CARCINOMA INVADING URINARY COLLECTING SYSTEM
localized renal cell carcinoma with urothelial involvement are uncommon, accounting for only 3% of all cases of renal cell carcinoma reviewed. However, urothelial involvement in low stage tumors should be noted due to its potential prognostic implications. Consideration should be given to adjuvant therapy in these patients. Future prospective studies corroborating these findings are warranted. When specifically evaluating tumor size, only 2 lesions 4 cm. or less invaded the renal collecting system. However, in each case these tumors were small and central, and showed other aggressive biological features, in that they invaded the renal vein or vena cava (stage pT3b). No tumors 4 cm. or less that were confined by the renal capsule invaded the urothelium. This finding has important implications when considering elective nephron sparing surgery for small solitary lesions. Under these circumstances the adjacent collecting system is rarely involved and surgical resection into the collecting system is typically unnecessary to provide a negative margin. The majority of renal tumors amenable to elective partial nephrectomy do not invade the renal collecting system. A potential shortcoming of our study is the retrospective nature of these data, which may limit interpretation. Prospective data collection is currently under way at our institution to corroborate these findings and determine the most significant clinical and biological prognostic factors that may help to clarify uncertainty in patients with this disease and select appropriate candidates for adjuvant therapy. CONCLUSIONS
Renal cell carcinoma invaded the urinary collecting system in 61 of the 426 patients (14%) in our cohort. Tumors invading the renal collecting system are characteristically large with high nuclear grade and predominantly high stage. Invasion of the renal collecting system by papillary tumors is rare, while invasion of the collecting system by clear cell carcinoma or tumors with sarcomatoid elements is more common. Involvement of the renal collecting system by renal cell carcinoma is most common at the renal poles, possibly due to the more peripheral location of the polar calices. Involvement of the urinary collecting system by renal cell carcinoma is rare in the absence of other poor prognostic features associated with the primary tumor. Although urothelial invasion by low stage localized renal cell carcinoma is uncommon, when it occurs, it portends a worse prognosis. In contrast, high stage renal cell carcinoma invading the urothelium does not appear to have a worse prognosis than equivalent pathological T stage lesions that do not invade the collecting system. In the majority of cases of renal cell carcinoma with urothelial involvement invasion is unlikely to impact management independently. For these reasons we believe that including urinary collecting system invasion into the TNM staging system is not warranted at this time. Tumors amenable to elective nephron sparing surgery are unlikely to involve the urothelium. Lisa A. Rybicki, Department of Biostatistics and Epidemiology assisted with the statistical analysis. REFERENCES
1. Landis, S. H., Murray, T., Bolden, S. and Wingo, P. A.: Cancer statistics. CA Cancer J Clin, 49: 8, 1999 2. Rodriguez, R., Fishman, E. K. and Marshall, F. F.: Differential diagnosis and evaluation of the incidentally discovered renal mass. Semin Urol Oncol, 13: 246, 1995 3. Vallancien, G., Torres, L. O., Gurfinkel, E., Veillon, B. and Brisset, J. M.: Incidental detection of renal tumours by abdominal ultrasonography. Eur Urol, 18: 94, 1990 4. Chow, W. H., Devesa, S. S., Warren, J. L. and Fraumeni, J. F., Jr.: Rising incidence of renal cell cancer in the United States. JAMA, 281: 1628, 1999 5. Motzer, R. J. and Russo, P.: Systemic therapy for renal cell
carcinoma. J Urol, 163: 408, 2000 6. Fleming, I. D., Cooper, J. S., Henson, D. E., Hutter, R. V., Kennedy, B. J., Murphy, G. P. et al: AJCC Cancer Staging Manual, 5th ed. Philadelphia: Lippincott Williams and Wilkins, pp. 215–217, 1997 7. Guinan, P., Sobin, L. H., Algaba, F., Badellino, F., Kameyama, S., MacLennan, G. et al: TNM staging of renal cell carcinoma. Workshop No. 3. Union International Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 992, 1997 8. Sobin, L. H. and Wittekind, C. H.: International Union Against Cancer: TNM Classification of Malignant Tumours, 5th ed. New York: Wiley-Liss, pp. 180 –182, 1997 9. Fuhrman, S. A., Lasky, L. C. and Limas, C.: Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol, 6: 655, 1982 10. Uzzo, R. G. and Novick, A. C.: Nephron sparing surgery for renal tumors: indications, techniques and outcomes. J Urol, 166: 6, 2001 11. Hafez, K. S., Fergany, A. F. and Novick, A. C.: Nephron sparing surgery for localized renal cell carcinoma: impact of tumor size on patient survival, tumor recurrence and TNM staging. J Urol, 162: 1930, 1999 12. Carducci, M. A., Piantadosi, S., Pound, C. R., Epstein, J. I., Simons, J. W., Marshall, F. F. et al: Nuclear morphometry adds significant prognostic information to stage and grade for renal cell carcinoma. Urology, 53: 44, 1999 13. Elson, P. J.: Prognostic factors in metastatic renal cell carcinoma. In: Renal Cell Carcinoma: Molecular Biology, Immunology, and Clinical Management. Edited by R. M. Bukowski and A. C. Novick. New Jersey: Humana Press, pp. 147–160, 2000 14. Jennings, S. B. and Linehan, W. M.: Renal, perirenal, and ureteral neoplasms. In: Adult and Pediatric Urology, 3rd ed. Edited by J. Y. Gillenwater, J. T. Grayhack, S. S. Howards and J. W. Duckett. St. Louis: Mosby Press, pp. 643– 694, 1996 15. Yasunaga, Y., Shin, M., Miki, T., Okuyama, A. and Aozasa, K.: Prognostic factors of renal cell carcinoma: a multivariate analysis. J Surg Oncol, 68: 11, 1998 16. Guinan, P. D., Vogelzang, N. J., Fremgen, A. M., Chmiel, J. S., Sylvester, J. L., Sener, S. F. et al: Renal cell carcinoma: tumor size, stage and survival. J Urol, 153: 901, 1995 17. Kinouchi, T., Saiki, S., Meguro, N., Maeda, O., Kuroda, M., Usami, M. et al: Impact of tumor size on the clinical outcomes of patients with Robson stage I renal cell carcinoma. Cancer, 85: 689, 1999 18. Ulchaker, J. C. and Klein, E. A.: Biology of metastasis and its clinical implications: renal cell cancer. World J Urol, 14: 175, 1996 19. Thrasher, J. B. and Paulson, D. F.: Prognostic factors in renal cancer. Urol Clin North Am, 20: 247, 1993 20. Hofmockel, G., Tsatalpas, P., Muller, H., Da¨ mmrich, J., Poot, M., Maurer-Schultze, B. et al: Significance of conventional and new prognostic factors for locally confined renal cell carcinoma. Cancer, 76: 296, 1995 21. Coll, D. M., Uzzo, R. G., Herts, B. R., Davros, W. J., Wirth, S. L. and Novick, A. C.: 3-dimensional volume rendered computerized tomography for preoperative evaluation and intraoperative treatment of patients undergoing nephron sparing surgery. J Urol, 161: 1097, 1999 22. Igarashi, T., Tobe, T., Nakatsu, H.-O., Suzuki, N., Murakami, S., Hamano, M. et al: The impact of a 4 cm. cutoff point for stratification of T1NoMo renal cell carcinoma after radical nephrectomy. J Urol, 165: 1103, 2001 23. Javidan, J., Stricker, H. J., Tamboli, P., Amin, M. B., Peabody, J. O., Deshpande, A. et al: Prognostic significance of the 1997 TNM classification of renal cell carcinoma. J Urol, 162: 1277, 1999 24. Moch, H., Gasser, T., Amin, M. B., Torhorst, J., Sauter, G. and Mihatsch M. J.: Prognostic utility of the recently recommended histologic classification and revised TNM staging system of renal cell carcinoma: a Swiss experience with 588 tumors. Cancer, 89: 604, 2000 25. Fergany, A. F., Hafez, K. S. and Novick, A. C.: Long-term results of nephron sparing surgery for localized renal cell carcinoma: 10-year followup. J Urol, 163: 442, 2000 26. Herr, H. W.: Partial nephrectomy for unilateral renal cell carcinoma and a normal contralateral kidney: 10-year followup. J Urol, 161: 33, 1999