- Email: [email protected]
LOW CLINICAL STAGE RENAL CELL CARCINOMA: RELEVANCE OF MICROVASCULAR TUMOR INVASION AS A PROGNOSTIC PARAMETER
0022-5347/04/1722-0470/0 THE JOURNAL OF UROLOGY® Copyright © 2004 by AMERICAN UROLOGICAL ASSOCIATION
Vol. 172, 470 – 474, August 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000130582.31467.30
LOW CLINICAL STAGE RENAL CELL CARCINOMA: RELEVANCE OF MICROVASCULAR TUMOR INVASION AS A PROGNOSTIC PARAMETER ˜ O GONC PIERRE DAMIA ¸ ALVES, MIGUEL SROUGI,* MARCOS FRANCISCO DALL’OGLIO, ´ TIA RAMOS MOREIRA LEITE, VALDEMAR ORTIZ AND FLA ´ VIO HERING KA From the Division of Urology, Federal University of Sa˜o Paulo and Division of Surgical Pathology, Hospital Sı´rio Libaneˆs, Sa˜o Paulo, Brazil
ABSTRACT
Purpose: Renal cell carcinoma is a tumor with unpredictable behavior and defining reliable prognostic factors would be extremely valuable in the clinical setting. Tumor stage, nuclear grade and tumor cell type are the main prognostic clinical parameters available. In this study we evaluated the role of microvascular involvement in the primary lesion for predicting tumor behavior in patients with low stage clinical disease. Materials and Methods: A total of 95 patients with clinically localized renal cell carcinoma (stages T1-T2 Nx M0) underwent radical nephrectomy and/or nephron sparing surgery, and were followed for a median of 45 months. The impact of microvascular tumor invasion on disease progression and its correlation with known pathological outcomes (tumor size, nuclear grade and cell type) were studied. Results: Microvascular tumor invasion was observed in 24 patients (25%), of whom 50% had disease recurrence. Of the 71 patients without microvascular invasion only 4 (6%) showed tumor recurrence. When microvascular invasion was correlated with other histological parameters, a significant statistical association was noted with tumor diameter, perirenal fat invasion, macroscopic extension to the renal vein, nuclear grade, lymph node metastasis and sarcomatous elements in the tumor. Multivariate analysis showed that microvascular invasion and the involvement of regional lymph nodes were independent predictors of disease recurrence. Concerning cancer specific survival, microvascular invasion and perirenal fat infiltration were the only factors related to death. Conclusions: Microvascular invasion is an independent and relevant clinical prognostic parameter for low clinical stage renal cell carcinoma. KEY WORDS: kidney; carcinoma, renal cell; prognosis; neoplasm invasiveness; neoplasms, vascular tissue
Radical nephrectomy remains the gold standard treatment for localized renal cell carcinoma (RCC), providing 5-year survival rates between 60% and 90% when disease is confined to the organ.1–3 However, in some apparently favorable cases disease recurs up to 10 years after surgery. This fact has stimulated research on the clinical behavior of the disease to identify the pathological parameters of a worse prognosis and define groups of patients with greater chances of recurrent disease. Pathological stage is considered the main prognostic factor in RCC since patients with disease confined to the kidney have a favorable clinical outcome compared with patients with locally advanced tumors or those with distant dissemination.1, 2, 4, 5 Other factors, such as vascular invasion, renal pelvis involvement, nuclear grade and cell type, have been considered parameters for prognostic assessment but their true role has not been completely established. The significance of vascular invasion to the prognosis of renal adenocarcinoma has elicited great discussion. Some groups have observed that invasion of the renal vein by neoplasm has a negative effect on patient survival.1, 6, 7 With this in mind attempts have been made to correlate the cephalic growth of the tumor thrombus with shorter patient life expectancy.8, 9 Other studies indicate that vascular invasion is not an independent predictor if there is no perirenal fat involvement, lymphatic disease or distant metastasis.2, 3, 10
Research was also done to establish whether microvascular invasion, defined as neoplastic cells in tumor microvessels, could be considered a factor for a worse prognosis. Two studies showed that microvascular invasion compared with macroscopic invasion had no influence on patient outcome.10, 11 On the other hand, other series defined microvascular invasion as an important parameter with an unfavorable influence on patient survival.12–14 We evaluated the influence of microvascular invasion on the prognosis in patients with low clinical stage RCC who underwent surgical treatment. We compared this specific pathological parameter with other established factors related to the disease prognosis. MATERIALS AND METHODS
We retrospectively analyzed the records of 110 patients with RCC who underwent radical nephrectomy and/or renal conservative surgery (partial nephrectomy or tumor enucleation) from 1988 to 1999. Disease was clinically confined to the kidney (stages T1-T2 Nx M0), as shown on preoperative imaging, including abdominal ultrasonography, computerized tomography, magnetic resonance imaging and chest x-rays. Of the potential subjects 15 were excluded. In 11 cases tissue review was insufficient to provide adequate data. Three patients were lost to followup and 1 retrospectively showed a pulmonary node consistent with metastatic disease. A total of 95 patients fulfilled the criteria for inclusion and were evaluated in this analysis, including there were 72 males and 23 females with a median age of 60 years (range 9
Accepted for publication March 19, 2004. * Correspondence: Rua Peixoto Gomide, 2055/81, Sa˜o Paulo-SP, Brazil, Cep 01409-003 (telephone: 55-11-3257.8002; FAX: 55-113257.9006; e-mail: [email protected]). For another article on a related topic see page 718. 470
471
MICROVASCULAR INVASION OF RENAL CELL CANCER
to 81). Median followup was 45 months (range 14 to 132). At initial clinical presentation the tumor diagnosis was incidental in 45 patients (47%) and hematuria was the main finding in the 25 symptomatic patients (26%). All paraffin blocs obtained from surgical specimens were reviewed by a single pathologist (KRML), who paid special attention to tumor diameter, tumor in perirenal fat, tumor cell type, macroscopic (renal vein and/or vena cava) and microvascular invasion, neoplastic lymph node involvement and nuclear grade. The lesions were divided into 2 groups, namely diameter 7 cm or less and larger than 7 cm. Nuclear tumor grade was also analyzed according to the criteria of Fuhrman et al,15 which classify RCC into 4 grades, considering the features of the cellular nucleus and nucleolus. The Rochester classification was used for cell typing.16 It divides renal cell tumors into 5 types, namely conventional or clear cell, papillary, chromophobe, collecting duct and unclassified carcinoma. Tumors were also analyzed for sarcomatous degeneration, defined as spindle cells with high grade pleomorphism and/or giant cells similar to those seen in sarcomas. Microvascular tumor invasion was indicated by neoplastic cells invading the vessel walls or by neoplastic emboli in the vessel lumen (fig. 1). The main end points were the correlation of microvascular invasion and other pathological findings, and the correlation of disease recurrence and cancer specific survival rates with pathological outcomes. Statistical analysis used the chi-square test to evaluate associations among listed variables. The Kaplan-Meier method was used to develop survival curves and differences among them were analyzed by the log rank test. For parameters with significant association Cox regression analysis identified those that independently could explain disease occurrence and progression to death with p ⬍0.05 considered statistically significant. RESULTS
Microvascular invasion was seen in 24 tumors (25%), while 10 (10%) had macrovascular involvement. When tumor diameter was analyzed, 29 lesions (30%) were larger than 7 cm. Perirenal fat infiltration was seen in 18 tumors (19%). Clear cell carcinoma was the most frequent histological type, found in 56 patients (59%), while 21 (22%) had papillary tumors, 10 (11%) had the chromophobe type and 8 (8%) showed a cell predominant pattern consistent with a sarcomatous subtype. As to nuclear grade, 22 lesions (23%) were grade I, 41 (43%) were grade II, 23 (24%) were grade III and 9 (9%) were grade IV. The correlation between microvascular invasion and disease recurrence showed that 12 of 24 patients (50%) with microvascular invasion had recurrence, whereas of 71 without microvascular invasion disease progressed in 4 (6%) (p ⬍0.001, table 1). We observed a statistically significant association of micro-
FIG. 1. Microscopic vascular tumor invasion in primary lesion. H & E, reduced from ⫻400.
TABLE 1. Risk of disease recurrence and studied pathological parameters Parameter
No. Pts/No. Recurrence (%)
Microvascular invasion: No Yes Lymph node metastasis: No Yes Fat invasion: No Yes Macrovascular involvement: No Yes Size (cm): 7 or Less Greater than 7 Nuclear grade: I II III IV Sarcomatous type: No Yes p ⬍0.001.
71/4 (6) 24/12 (50) 87/9 (10) 8/7 (88) 77/6 (8) 18/10 (56) 85/9 (11) 10/7 (70) 66/4 (6) 29/12 (41) 22/2 (9) 41/3 (7) 23/5 (22) 9/6 (67) 87/12 (14) 8/4 (50)
vascular invasion with tumor diameter, perirenal fat infiltration, macrovascular involvement and lymph node metastasis (table 2). When analyzing the association between microvascular invasion and nuclear grade, we noted that 21 of the 24 tumors (87%) with microvascular invasion were nuclear grades III and IV. On the other hand, only 11 of the 71 tumors (16%) without microvascular invasion were nuclear grade III or IV. Concerning the correlation between microvascular invasion and cell type, there was a statistical association only between invasion and the sarcomatous type cell pattern. Univariate analysis showed a significant correlation between the risk of disease recurrence and microvascular invasion, lymph node metastasis, perirenal fat infiltration, macrovascular involvement, tumor diameter and sarcomatous pattern tumor (table 1). There was also a greater chance of progression in nuclear grade IV tumors. Figure 2 shows the impact of the mentioned parameters on cancer specific survival. On multivariate analysis we attempted to identify which parameters might independently influence the rates of
TABLE 2. Microvascular invasion and studied pathological parameters Parameter Tumor diameter (cm): 7 or Less Greater than 7 Perirenal fat infiltration: Absent Present Macrovascular involvement: Absent Present Nuclear grade: I II III IV Lymph node metastasis: Absent Present Cell type: Clear Chromophile Chromophobe Sarcomatous
No. Microvascular Invasion (%) Absent
Present
p Value
56 (79) 15 (21)
10 (42) 14 (58)
⬍0.001
67 (94) 4 (6)
10 (42) 14 (58)
⬍0.001
70 (99) 1 (1)
15 (62) 9 (37)
⬍0.001
22 (31) 38 (53) 9 (13) 2 (3)
0 3 (12) 14 (58) 7 (29)
⬍0.001
71 (100) 0
16 (67) 8 (33)
⬍0.001
46 (68) 16 (22) 8 (11) 1 (1)
10 (42) 5 (21) 2 (8) 7 (29)
Not significant Not significant Not significant ⬍0.001
472
MICROVASCULAR INVASION OF RENAL CELL CANCER
FIG. 2. Kaplan-Meier analysis of cancer specific survival. A, microvascular (mv) invasion. B, tumor diameter. C, nuclear grade. D, perirenal fat invasion. E, macrovascular (Mv) involvement. F, lymph node metastasis. G, cell type.
473
MICROVASCULAR INVASION OF RENAL CELL CANCER TABLE 3. Significance of each covariable that might affect disease recurrence and death rate Recurrence
Parameter Microvascular invasion Perirenal fat infiltration Nuclear grade Tumor diameter Macrovascular involvement Lymph node metastasis Sarcomatous elements * Statistically significant.
Death
RR
(95% CI)
RR
(95% CI)
6.78* 2.10 1.82 1.20 2.65 2.88* 1.16
(1.80–25.53) (0.51–8.62) (0.63–5.28) (0.29–5.03) (0.67–10.43) (0.88–9.49) (0.22–6.16)
15.72* 7.16* 0.63 10.10 2.85 0.23 8.39
(1.57–157.68) (1.21–42.34) (0.11–3.53) (0.58–175.11) (0.58–14.01) (0.02–2.32) (0.50–141.82)
disease recurrence or death (table 3). Of all patients 16 (17%) showed recurrence during followup but microvascular invasion and lymph node metastasis were the only prognostic parameters with independent statistical significance. Cancer related death occurred in 10 patients (10%) during followup. Again, microvascular invasion was an independent prognostic parameters that, together with perirenal fat infiltration, was associated with the death rate (table 3). DISCUSSION
Pathological stage has been described as the main prognostic parameter in RCC with a direct influence on patient survival.17 However, patients treated with radical nephrectomy with similar pathological stages can present different clinical behaviors. This finding has generated great interest among researchers, who are now attempting to identify additional parameters that might better predict disease outcome. Intratumoral microvascular neoplastic invasion has been evaluated as a prognostic indicator in RCC. Early studies suggested that patients with this histological finding should be considered carriers of micrometastases with a greater chance of disease progression.12 On the other hand, some studies indicated no relevance of microvascular invasion concerning patient prognosis.7, 10, 11 In our series we noted a strong association of microvascular invasion with other known prognostic indexes, and with disease recurrence and death. We observed microvascular invasion in 25% of our patients. Van Poppel et al studied 180 patients and found microvascular invasion in 28%.14 Similar rates were seen by Samma12 and Mrstik13 et al. In terms of the association between microvascular invasion and disease extension, several studies showed that only 21% to 47% of patients with microvascular invasion had disease confined to the kidney compared to 78% without microvascular invasion.13, 14 When analyzing the association between microvascular invasion and nuclear grade, we found a positive association with grades III and IV. In the series of Van Poppel et al this association, although present, was not as strong, since 17% of tumors with microvascular invasion were grade IV, a finding observed in about 5% of those without microvascular invasion.14 Mrstik et al had results similar to ours.13 They observed that 28% of tumors with microvascular invasion were grade IV as opposed to those without microvascular invasion (only 7%). At a median followup of 45 months we were able to see a statistically significant association between microvascular invasion and disease recurrence. Patients with tumors without microvascular invasion had an estimated 5-year diseasefree survival of 83%, whereas the rate was 37% in patients with microvascular invasion. These results are comparable to those of Van Poppel et al, who observed disease-free survival in 61% of patients with and in 94% without microvascular invasion at a mean followup of 52 months.14 They also described cancer specific survival rates of 75% and 97% in patients with and without microvascular invasion, respectively. The higher rates of progression and death found in our series might be related to a higher number of patients with macrovascular involvement and lymphatic dissemination.
Other groups could not prove the negative prognostic role of microvascular invasion in RCC.7, 10, 11 However, these investigators did not analyze microvascular invasion exclusively and there is a lack of uniformity among the definitions of macrovascular and microvascular involvement. In our study the concept of microscopic invasion was well defined, and patients with macroscopic and microscopic vascular involvement were identified separately. Macrovascular involvement has the same unfavorable predictive value as microvascular invasion (table 1). As a matter of fact, 9 of 10 patients with macrovascular invasion also had tumor thrombus in the microvasculature. Since macrovascular involvement can only occur following invasion of the small vessels within the tumor, the latter phenomenon is probably the determining factor for a worse patient outcome. One patient in our series was found to have macrovascular involvement without any lesion at microvascular level, which probably can be explained by inadequate random tissue sampling for microscopic examination. Our multivariate analysis clearly demonstrates that microvascular invasion is an independent prognostic factor able to predict progression and death due to disease. Two other parameters, lymph node involvement and perirenal fat infiltration, are also independent predictors for disease progression and death, respectively. Our study confirms and validates previous studies showing that microvascular invasion is the most important prognostic parameter in patients with clinically localized RCC, more important than pathological tumor stage, nuclear grade and size.14 Our findings also corroborated those of Kinouchi et al, who studied patients with stage I renal carcinoma and concluded that tumor diameter (smaller or larger than 5.5 cm) and microvascular invasion were independent parameters of prognosis in these patients.18 Currently available data justify randomized studies to confirm definitely the impact of microvascular invasion in cases of localized RCC and justify adjuvant therapy in unfavorable cases. CONCLUSIONS
According to our study microvascular invasion has a statistically significant association with disease extent, tumor nuclear grade and sarcomatous elements in patients with RCC. Furthermore, patients with microvascular invasion have a worse prognosis with a higher risk of disease recurrence and lower cancer specific survival. REFERENCES
1. Robson, C. J., Churchill, B. M. and Anderson, W.: The results of radical nephrectomy for renal cell carcinoma. J Urol, 101: 297, 1969 2. Skinner, D. G., Colvin, R. B., Vermillion, C. D., Pfister, R. C. and Leadbetter, W. F.: Diagnosis and management of renal cell carcinoma. A clinical and pathologic study of 309 cases. Cancer, 28: 1165, 1971 3. McNichols, D. W., Segura, J. W. and Weerd, J. H.: Renal cell carcinoma: long-term survival and late recurrence. J Urol, 126: 17, 1981 4. Medeiros, L. J., Gelb, A. B. and Weiss, L. M.: Renal cell carci-
474
5.
6. 7. 8.
9.
10. 11.
MICROVASCULAR INVASION OF RENAL CELL CANCER
noma. Prognostic significance of morphologic parameters in 121 cases. Cancer, 61: 1639, 1988 Gilberti, C., Oneto, F., Martorana, G., Rovida, S. and Carmignani, G.: Radical nephrectomy for renal cell carcinoma: long-term results and prognostic factors on a series of 328 cases. Eur Urol, 31: 40, 1997 Myers, G. H., Jr., Fehrenbaker, L. G. and Kelalis, P. P.: Prognostic significance of renal vein invasion by hypernephroma. J Urol, 100: 420, 1968 Ljungberg, B., Stenling, R., Osterdahl, B., Farrelly, E., Aberg, T. and Roos, G.: Vein invasion in renal cell carcinoma: impact on metastatic behavior and survival. J Urol, 154: 1681, 1995 Sosa, R. E., Muecke, E. C., Vaughan, E. D., Jr. and McCarron, J. P., Jr.: Renal cell carcinoma extending into the vena cava: the prognostic significance of the level of vena caval involvement. J Urol, 132: 1097, 1984 Skinner, D. G., Pritchett, T. R., Lieskovsky, G., Boyd, S. D. and Stiles, Q. R.: Vena caval involvement by renal cell carcinoma. Surgical resection provides meaningful long-term survival. Ann Surg, 210: 387, 1989 Golimbu, M., Joshi, P., Sperber, A., Tessler, A., Al-Askari, S. and Morales, P.: Renal cell carcinoma: survival and prognostic factors. Urology, 27: 291, 1986 Hoehn, W. and Hermanek, P.: Invasion of veins in renal cell carcinoma—frequency, correlation and prognosis. Eur Urol, 9: 276, 1983
12. Samma, S., Yoshida, K., Ozono, S., Ohara, S., Hayashi, Y., Tabata, S. et al: Tumor thrombus and microvascular invasion as prognostic factors in renal cell carcinoma. Jpn J Clin Oncol, 21: 340, 1991 13. Mrstik, C. H., Salamon, J., Weber, R. and Stogermayer, F.: Microscopic venous infiltration as predictor of relapse in renal cell carcinoma. J Urol, 148: 271, 1992 14. Van Poppel, H., Vandendriessche, H., Boel, K., Mertens, V., Goethuys, H., Haustermans, K. et al: Microscopic vascular invasion is the most relevant prognosticator after radical nephrectomy for clinically nonmetastatic renal cell carcinoma. J Urol, 158: 45, 1997 15. 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 16. Storkel, S., Eble, J. N., Adlakha, K., Amin, M., Blute, M. L., Bostwick, D. J. et al: Classification of renal cell carcinoma. Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 987, 1997 17. Thrasher, J. B. and Paulson, D. F.: Prognostic factors in renal cancer. Urol Clin North Am, 20: 247, 1993 18. 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
Vol. 172, 470 – 474, August 2004 Printed in U.S.A.
DOI: 10.1097/01.ju.0000130582.31467.30
LOW CLINICAL STAGE RENAL CELL CARCINOMA: RELEVANCE OF MICROVASCULAR TUMOR INVASION AS A PROGNOSTIC PARAMETER ˜ O GONC PIERRE DAMIA ¸ ALVES, MIGUEL SROUGI,* MARCOS FRANCISCO DALL’OGLIO, ´ TIA RAMOS MOREIRA LEITE, VALDEMAR ORTIZ AND FLA ´ VIO HERING KA From the Division of Urology, Federal University of Sa˜o Paulo and Division of Surgical Pathology, Hospital Sı´rio Libaneˆs, Sa˜o Paulo, Brazil
ABSTRACT
Purpose: Renal cell carcinoma is a tumor with unpredictable behavior and defining reliable prognostic factors would be extremely valuable in the clinical setting. Tumor stage, nuclear grade and tumor cell type are the main prognostic clinical parameters available. In this study we evaluated the role of microvascular involvement in the primary lesion for predicting tumor behavior in patients with low stage clinical disease. Materials and Methods: A total of 95 patients with clinically localized renal cell carcinoma (stages T1-T2 Nx M0) underwent radical nephrectomy and/or nephron sparing surgery, and were followed for a median of 45 months. The impact of microvascular tumor invasion on disease progression and its correlation with known pathological outcomes (tumor size, nuclear grade and cell type) were studied. Results: Microvascular tumor invasion was observed in 24 patients (25%), of whom 50% had disease recurrence. Of the 71 patients without microvascular invasion only 4 (6%) showed tumor recurrence. When microvascular invasion was correlated with other histological parameters, a significant statistical association was noted with tumor diameter, perirenal fat invasion, macroscopic extension to the renal vein, nuclear grade, lymph node metastasis and sarcomatous elements in the tumor. Multivariate analysis showed that microvascular invasion and the involvement of regional lymph nodes were independent predictors of disease recurrence. Concerning cancer specific survival, microvascular invasion and perirenal fat infiltration were the only factors related to death. Conclusions: Microvascular invasion is an independent and relevant clinical prognostic parameter for low clinical stage renal cell carcinoma. KEY WORDS: kidney; carcinoma, renal cell; prognosis; neoplasm invasiveness; neoplasms, vascular tissue
Radical nephrectomy remains the gold standard treatment for localized renal cell carcinoma (RCC), providing 5-year survival rates between 60% and 90% when disease is confined to the organ.1–3 However, in some apparently favorable cases disease recurs up to 10 years after surgery. This fact has stimulated research on the clinical behavior of the disease to identify the pathological parameters of a worse prognosis and define groups of patients with greater chances of recurrent disease. Pathological stage is considered the main prognostic factor in RCC since patients with disease confined to the kidney have a favorable clinical outcome compared with patients with locally advanced tumors or those with distant dissemination.1, 2, 4, 5 Other factors, such as vascular invasion, renal pelvis involvement, nuclear grade and cell type, have been considered parameters for prognostic assessment but their true role has not been completely established. The significance of vascular invasion to the prognosis of renal adenocarcinoma has elicited great discussion. Some groups have observed that invasion of the renal vein by neoplasm has a negative effect on patient survival.1, 6, 7 With this in mind attempts have been made to correlate the cephalic growth of the tumor thrombus with shorter patient life expectancy.8, 9 Other studies indicate that vascular invasion is not an independent predictor if there is no perirenal fat involvement, lymphatic disease or distant metastasis.2, 3, 10
Research was also done to establish whether microvascular invasion, defined as neoplastic cells in tumor microvessels, could be considered a factor for a worse prognosis. Two studies showed that microvascular invasion compared with macroscopic invasion had no influence on patient outcome.10, 11 On the other hand, other series defined microvascular invasion as an important parameter with an unfavorable influence on patient survival.12–14 We evaluated the influence of microvascular invasion on the prognosis in patients with low clinical stage RCC who underwent surgical treatment. We compared this specific pathological parameter with other established factors related to the disease prognosis. MATERIALS AND METHODS
We retrospectively analyzed the records of 110 patients with RCC who underwent radical nephrectomy and/or renal conservative surgery (partial nephrectomy or tumor enucleation) from 1988 to 1999. Disease was clinically confined to the kidney (stages T1-T2 Nx M0), as shown on preoperative imaging, including abdominal ultrasonography, computerized tomography, magnetic resonance imaging and chest x-rays. Of the potential subjects 15 were excluded. In 11 cases tissue review was insufficient to provide adequate data. Three patients were lost to followup and 1 retrospectively showed a pulmonary node consistent with metastatic disease. A total of 95 patients fulfilled the criteria for inclusion and were evaluated in this analysis, including there were 72 males and 23 females with a median age of 60 years (range 9
Accepted for publication March 19, 2004. * Correspondence: Rua Peixoto Gomide, 2055/81, Sa˜o Paulo-SP, Brazil, Cep 01409-003 (telephone: 55-11-3257.8002; FAX: 55-113257.9006; e-mail: [email protected]). For another article on a related topic see page 718. 470
471
MICROVASCULAR INVASION OF RENAL CELL CANCER
to 81). Median followup was 45 months (range 14 to 132). At initial clinical presentation the tumor diagnosis was incidental in 45 patients (47%) and hematuria was the main finding in the 25 symptomatic patients (26%). All paraffin blocs obtained from surgical specimens were reviewed by a single pathologist (KRML), who paid special attention to tumor diameter, tumor in perirenal fat, tumor cell type, macroscopic (renal vein and/or vena cava) and microvascular invasion, neoplastic lymph node involvement and nuclear grade. The lesions were divided into 2 groups, namely diameter 7 cm or less and larger than 7 cm. Nuclear tumor grade was also analyzed according to the criteria of Fuhrman et al,15 which classify RCC into 4 grades, considering the features of the cellular nucleus and nucleolus. The Rochester classification was used for cell typing.16 It divides renal cell tumors into 5 types, namely conventional or clear cell, papillary, chromophobe, collecting duct and unclassified carcinoma. Tumors were also analyzed for sarcomatous degeneration, defined as spindle cells with high grade pleomorphism and/or giant cells similar to those seen in sarcomas. Microvascular tumor invasion was indicated by neoplastic cells invading the vessel walls or by neoplastic emboli in the vessel lumen (fig. 1). The main end points were the correlation of microvascular invasion and other pathological findings, and the correlation of disease recurrence and cancer specific survival rates with pathological outcomes. Statistical analysis used the chi-square test to evaluate associations among listed variables. The Kaplan-Meier method was used to develop survival curves and differences among them were analyzed by the log rank test. For parameters with significant association Cox regression analysis identified those that independently could explain disease occurrence and progression to death with p ⬍0.05 considered statistically significant. RESULTS
Microvascular invasion was seen in 24 tumors (25%), while 10 (10%) had macrovascular involvement. When tumor diameter was analyzed, 29 lesions (30%) were larger than 7 cm. Perirenal fat infiltration was seen in 18 tumors (19%). Clear cell carcinoma was the most frequent histological type, found in 56 patients (59%), while 21 (22%) had papillary tumors, 10 (11%) had the chromophobe type and 8 (8%) showed a cell predominant pattern consistent with a sarcomatous subtype. As to nuclear grade, 22 lesions (23%) were grade I, 41 (43%) were grade II, 23 (24%) were grade III and 9 (9%) were grade IV. The correlation between microvascular invasion and disease recurrence showed that 12 of 24 patients (50%) with microvascular invasion had recurrence, whereas of 71 without microvascular invasion disease progressed in 4 (6%) (p ⬍0.001, table 1). We observed a statistically significant association of micro-
FIG. 1. Microscopic vascular tumor invasion in primary lesion. H & E, reduced from ⫻400.
TABLE 1. Risk of disease recurrence and studied pathological parameters Parameter
No. Pts/No. Recurrence (%)
Microvascular invasion: No Yes Lymph node metastasis: No Yes Fat invasion: No Yes Macrovascular involvement: No Yes Size (cm): 7 or Less Greater than 7 Nuclear grade: I II III IV Sarcomatous type: No Yes p ⬍0.001.
71/4 (6) 24/12 (50) 87/9 (10) 8/7 (88) 77/6 (8) 18/10 (56) 85/9 (11) 10/7 (70) 66/4 (6) 29/12 (41) 22/2 (9) 41/3 (7) 23/5 (22) 9/6 (67) 87/12 (14) 8/4 (50)
vascular invasion with tumor diameter, perirenal fat infiltration, macrovascular involvement and lymph node metastasis (table 2). When analyzing the association between microvascular invasion and nuclear grade, we noted that 21 of the 24 tumors (87%) with microvascular invasion were nuclear grades III and IV. On the other hand, only 11 of the 71 tumors (16%) without microvascular invasion were nuclear grade III or IV. Concerning the correlation between microvascular invasion and cell type, there was a statistical association only between invasion and the sarcomatous type cell pattern. Univariate analysis showed a significant correlation between the risk of disease recurrence and microvascular invasion, lymph node metastasis, perirenal fat infiltration, macrovascular involvement, tumor diameter and sarcomatous pattern tumor (table 1). There was also a greater chance of progression in nuclear grade IV tumors. Figure 2 shows the impact of the mentioned parameters on cancer specific survival. On multivariate analysis we attempted to identify which parameters might independently influence the rates of
TABLE 2. Microvascular invasion and studied pathological parameters Parameter Tumor diameter (cm): 7 or Less Greater than 7 Perirenal fat infiltration: Absent Present Macrovascular involvement: Absent Present Nuclear grade: I II III IV Lymph node metastasis: Absent Present Cell type: Clear Chromophile Chromophobe Sarcomatous
No. Microvascular Invasion (%) Absent
Present
p Value
56 (79) 15 (21)
10 (42) 14 (58)
⬍0.001
67 (94) 4 (6)
10 (42) 14 (58)
⬍0.001
70 (99) 1 (1)
15 (62) 9 (37)
⬍0.001
22 (31) 38 (53) 9 (13) 2 (3)
0 3 (12) 14 (58) 7 (29)
⬍0.001
71 (100) 0
16 (67) 8 (33)
⬍0.001
46 (68) 16 (22) 8 (11) 1 (1)
10 (42) 5 (21) 2 (8) 7 (29)
Not significant Not significant Not significant ⬍0.001
472
MICROVASCULAR INVASION OF RENAL CELL CANCER
FIG. 2. Kaplan-Meier analysis of cancer specific survival. A, microvascular (mv) invasion. B, tumor diameter. C, nuclear grade. D, perirenal fat invasion. E, macrovascular (Mv) involvement. F, lymph node metastasis. G, cell type.
473
MICROVASCULAR INVASION OF RENAL CELL CANCER TABLE 3. Significance of each covariable that might affect disease recurrence and death rate Recurrence
Parameter Microvascular invasion Perirenal fat infiltration Nuclear grade Tumor diameter Macrovascular involvement Lymph node metastasis Sarcomatous elements * Statistically significant.
Death
RR
(95% CI)
RR
(95% CI)
6.78* 2.10 1.82 1.20 2.65 2.88* 1.16
(1.80–25.53) (0.51–8.62) (0.63–5.28) (0.29–5.03) (0.67–10.43) (0.88–9.49) (0.22–6.16)
15.72* 7.16* 0.63 10.10 2.85 0.23 8.39
(1.57–157.68) (1.21–42.34) (0.11–3.53) (0.58–175.11) (0.58–14.01) (0.02–2.32) (0.50–141.82)
disease recurrence or death (table 3). Of all patients 16 (17%) showed recurrence during followup but microvascular invasion and lymph node metastasis were the only prognostic parameters with independent statistical significance. Cancer related death occurred in 10 patients (10%) during followup. Again, microvascular invasion was an independent prognostic parameters that, together with perirenal fat infiltration, was associated with the death rate (table 3). DISCUSSION
Pathological stage has been described as the main prognostic parameter in RCC with a direct influence on patient survival.17 However, patients treated with radical nephrectomy with similar pathological stages can present different clinical behaviors. This finding has generated great interest among researchers, who are now attempting to identify additional parameters that might better predict disease outcome. Intratumoral microvascular neoplastic invasion has been evaluated as a prognostic indicator in RCC. Early studies suggested that patients with this histological finding should be considered carriers of micrometastases with a greater chance of disease progression.12 On the other hand, some studies indicated no relevance of microvascular invasion concerning patient prognosis.7, 10, 11 In our series we noted a strong association of microvascular invasion with other known prognostic indexes, and with disease recurrence and death. We observed microvascular invasion in 25% of our patients. Van Poppel et al studied 180 patients and found microvascular invasion in 28%.14 Similar rates were seen by Samma12 and Mrstik13 et al. In terms of the association between microvascular invasion and disease extension, several studies showed that only 21% to 47% of patients with microvascular invasion had disease confined to the kidney compared to 78% without microvascular invasion.13, 14 When analyzing the association between microvascular invasion and nuclear grade, we found a positive association with grades III and IV. In the series of Van Poppel et al this association, although present, was not as strong, since 17% of tumors with microvascular invasion were grade IV, a finding observed in about 5% of those without microvascular invasion.14 Mrstik et al had results similar to ours.13 They observed that 28% of tumors with microvascular invasion were grade IV as opposed to those without microvascular invasion (only 7%). At a median followup of 45 months we were able to see a statistically significant association between microvascular invasion and disease recurrence. Patients with tumors without microvascular invasion had an estimated 5-year diseasefree survival of 83%, whereas the rate was 37% in patients with microvascular invasion. These results are comparable to those of Van Poppel et al, who observed disease-free survival in 61% of patients with and in 94% without microvascular invasion at a mean followup of 52 months.14 They also described cancer specific survival rates of 75% and 97% in patients with and without microvascular invasion, respectively. The higher rates of progression and death found in our series might be related to a higher number of patients with macrovascular involvement and lymphatic dissemination.
Other groups could not prove the negative prognostic role of microvascular invasion in RCC.7, 10, 11 However, these investigators did not analyze microvascular invasion exclusively and there is a lack of uniformity among the definitions of macrovascular and microvascular involvement. In our study the concept of microscopic invasion was well defined, and patients with macroscopic and microscopic vascular involvement were identified separately. Macrovascular involvement has the same unfavorable predictive value as microvascular invasion (table 1). As a matter of fact, 9 of 10 patients with macrovascular invasion also had tumor thrombus in the microvasculature. Since macrovascular involvement can only occur following invasion of the small vessels within the tumor, the latter phenomenon is probably the determining factor for a worse patient outcome. One patient in our series was found to have macrovascular involvement without any lesion at microvascular level, which probably can be explained by inadequate random tissue sampling for microscopic examination. Our multivariate analysis clearly demonstrates that microvascular invasion is an independent prognostic factor able to predict progression and death due to disease. Two other parameters, lymph node involvement and perirenal fat infiltration, are also independent predictors for disease progression and death, respectively. Our study confirms and validates previous studies showing that microvascular invasion is the most important prognostic parameter in patients with clinically localized RCC, more important than pathological tumor stage, nuclear grade and size.14 Our findings also corroborated those of Kinouchi et al, who studied patients with stage I renal carcinoma and concluded that tumor diameter (smaller or larger than 5.5 cm) and microvascular invasion were independent parameters of prognosis in these patients.18 Currently available data justify randomized studies to confirm definitely the impact of microvascular invasion in cases of localized RCC and justify adjuvant therapy in unfavorable cases. CONCLUSIONS
According to our study microvascular invasion has a statistically significant association with disease extent, tumor nuclear grade and sarcomatous elements in patients with RCC. Furthermore, patients with microvascular invasion have a worse prognosis with a higher risk of disease recurrence and lower cancer specific survival. REFERENCES
1. Robson, C. J., Churchill, B. M. and Anderson, W.: The results of radical nephrectomy for renal cell carcinoma. J Urol, 101: 297, 1969 2. Skinner, D. G., Colvin, R. B., Vermillion, C. D., Pfister, R. C. and Leadbetter, W. F.: Diagnosis and management of renal cell carcinoma. A clinical and pathologic study of 309 cases. Cancer, 28: 1165, 1971 3. McNichols, D. W., Segura, J. W. and Weerd, J. H.: Renal cell carcinoma: long-term survival and late recurrence. J Urol, 126: 17, 1981 4. Medeiros, L. J., Gelb, A. B. and Weiss, L. M.: Renal cell carci-
474
5.
6. 7. 8.
9.
10. 11.
MICROVASCULAR INVASION OF RENAL CELL CANCER
noma. Prognostic significance of morphologic parameters in 121 cases. Cancer, 61: 1639, 1988 Gilberti, C., Oneto, F., Martorana, G., Rovida, S. and Carmignani, G.: Radical nephrectomy for renal cell carcinoma: long-term results and prognostic factors on a series of 328 cases. Eur Urol, 31: 40, 1997 Myers, G. H., Jr., Fehrenbaker, L. G. and Kelalis, P. P.: Prognostic significance of renal vein invasion by hypernephroma. J Urol, 100: 420, 1968 Ljungberg, B., Stenling, R., Osterdahl, B., Farrelly, E., Aberg, T. and Roos, G.: Vein invasion in renal cell carcinoma: impact on metastatic behavior and survival. J Urol, 154: 1681, 1995 Sosa, R. E., Muecke, E. C., Vaughan, E. D., Jr. and McCarron, J. P., Jr.: Renal cell carcinoma extending into the vena cava: the prognostic significance of the level of vena caval involvement. J Urol, 132: 1097, 1984 Skinner, D. G., Pritchett, T. R., Lieskovsky, G., Boyd, S. D. and Stiles, Q. R.: Vena caval involvement by renal cell carcinoma. Surgical resection provides meaningful long-term survival. Ann Surg, 210: 387, 1989 Golimbu, M., Joshi, P., Sperber, A., Tessler, A., Al-Askari, S. and Morales, P.: Renal cell carcinoma: survival and prognostic factors. Urology, 27: 291, 1986 Hoehn, W. and Hermanek, P.: Invasion of veins in renal cell carcinoma—frequency, correlation and prognosis. Eur Urol, 9: 276, 1983
12. Samma, S., Yoshida, K., Ozono, S., Ohara, S., Hayashi, Y., Tabata, S. et al: Tumor thrombus and microvascular invasion as prognostic factors in renal cell carcinoma. Jpn J Clin Oncol, 21: 340, 1991 13. Mrstik, C. H., Salamon, J., Weber, R. and Stogermayer, F.: Microscopic venous infiltration as predictor of relapse in renal cell carcinoma. J Urol, 148: 271, 1992 14. Van Poppel, H., Vandendriessche, H., Boel, K., Mertens, V., Goethuys, H., Haustermans, K. et al: Microscopic vascular invasion is the most relevant prognosticator after radical nephrectomy for clinically nonmetastatic renal cell carcinoma. J Urol, 158: 45, 1997 15. 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 16. Storkel, S., Eble, J. N., Adlakha, K., Amin, M., Blute, M. L., Bostwick, D. J. et al: Classification of renal cell carcinoma. Workgroup No. 1. Union Internationale Contre le Cancer (UICC) and the American Joint Committee on Cancer (AJCC). Cancer, 80: 987, 1997 17. Thrasher, J. B. and Paulson, D. F.: Prognostic factors in renal cancer. Urol Clin North Am, 20: 247, 1993 18. 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