- Email: [email protected]
NOMOGRAM PREDICTING RENAL CANCER-SPECIFIC SURVIVAL IN SURGICALLY TREATED PATIENTS WITH METASTATIC RENAL CELL CARCINOMA
843 NOMOGRAM PREDICTING RENAL CANCER-SPECIFIC SURVIVAL IN SURGICALLY TREATED PATIENTS WITH METASTATIC RENAL CELL CARCINOMA Karakiewicz P.I.1, Hutterer G.C.2, Suardi N.1, Chromecki T.2, Jeldres C.1, KampelKettner K.2, Imamovic A.2, Zigeuner R.2, Bensaleh K., Avakian R., Shariat S.F.1, Montorsi F.4, Perrotte P.1, Patard J.J.
844 EXTERNAL VALIDATION A PRE-TREATMENT NOMOGRAM FOR PREDICTION OF RENAL CANCER-SPECIFIC MORTALITY Jeldres C.1, Suardi N.1, Patard J.J.2, Ficarra V., Cindolo L.4, De La Taille A.5, Salomon L.5, Tostain J.6, Mulders P.F.7, Zigeuner R.8, Prayer-Galetti T., Chautard D.9, Valeri A.10, Lechevallier E.10, Descotes J.L.10, Lang H.10, Avakian R.2, Bensaleh K.2, Mejean A.10, Bertini R.11, Montorsi F.11, Karakiewicz P.I.1 1
1
Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Centre, Dept. of Urology, Montreal, Canada, 2Graz Medical University, Dept. of Urology, Graz, Austria, Rennes University Hospital, Dept. of Urology, Rennes, France, 49LWD6DOXWH 8QLYHUVLW\ 6FLHQWLᚏF ,QVWLWXWH6DQ5DᚎDHOH'HSWRI8URORJ\0LODQ,WDO\ Introduction & Objectives: The survival of patients with metastatic renal cell carcinoma (MRCC) who are treated with nephrectomy is quite variable and may extend from few months to several years. Several variables may predict the natural history of surgically treated MRCC. These may include the local tumour characteristics, as well as the characteristics of the metastatic disease. We performed a detailed analysis of local and metastatic tumour characteristics to predict the natural history of surgically treated MRCC. Material & Methods: 7KHFOLQLFDODQGSDWKRORJLFDOGDWDZHUHSURVSHFWLYHO\UHFRUGHGIURP patients treated with cytoreductive nephrectomy. The detailed description of the distribution RI PHWDVWDWLF GLVHDVH ZDV XQDYDLODEOH LQ ZKLFK UHVXOWHG LQ HYDOXDEOH SDWLHQWV &R[ UHJUHVVLRQ PRGHOV DGGUHVVHG WKH WLPH WR FDXVHVSHFLᚏF PRUWDOLW\ 5HJUHVVLRQ FRHᚑFLHQWV ZHUH XVHG WR GHYHORS QRPRJUDPV SUHGLFWLQJ WKH SUREDELOLW\ RI FDXVHVSHFLᚏF VXUYLYDO DIWHU cytoreductive nephrectomy. The nomograms were internally validated with 200 bootstrap UHVDPSOHಬVWRUHGXFHRYHUᚏWELDV Results: The sites of MRCC were distributed as follows: 186 had lung, 107 bone, 71 nonUHJLRQDOO\PSKQRGHVEUDLQOLYHUDGUHQDODQGRWKHUVLWHV7KHPHGLDQVXUYLYDO ZDVPRQWKVZLWKDUDQJHXSWRPRQWKV7HVWHGSUHGLFWRUVRIFDXVHVSHFLᚏFVXUYLYDO included: age, gender, tumour size, T-stage, tumour size, time to metastatic progression and the site of metastatic disease. The location of metastases predicted survival. For example patients with brain metastases had substantially worse survival than the rest of the cohort (median 10 months). Conversely, patients with adrenal metastases had substantially better survival than the rest of the cohort (median 42 months). The most informative multivariable predictors of causeVSHFLᚏF VXUYLYDO ZHUH DJH DW QHSKUHFWRP\ SULPDU\ WXPRXU VL]H DQG WKH SUHVHQFH RI EUDLQ adrenal and other site metastases. The combined accuracy of these predictors was 64%. Conclusions: There is a tremendous amount of variability in the survival rates of patients with PHWDVWDWLF5&&&XUUHQWO\WKHDFFXUDF\RIH[LVWLQJULVNVWUDWLᚏFDWLRQVFKHPHVLVXQNQRZQDQG there are no models for prediction of survival in European patients. The current nomogram is simple, as it only requires 6 readily available variables and user friendly. Despite its intermediate accuracy, it represents a valuable tool and exceeds the value of quoting the median survival.
Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Centre, Dept. of Urology, Montreal, Canada, 2Rennes University Hospital, Dept. of Urology, Rennes, France, University of Padua, Dept. of Urology, Padua, Italy, 4”G. Rummo” Hospital, Dept. of Urology, Benevento, Italy, 5Henri Mondor University Hospital, Dept. of Urology, Creteil, France, 6St. Etienne University Hospital, Dept. of Urology, St. Etienne, France, 7Radboud University Nijmegen Medical Centre, Dept. of Urology, Nijmegen, The Netherlands, 8Graz Medical University, Dept. of Urology, Graz, Austria, 9Angers University Hospital, Dept. of Urology, Angers, France, 10&RPLW«GH&DQF«URORJLHGHOಬ$VVRFLDWLRQ)UDQ©DLVHGಬ8URORJLH'HSW of Urology, Paris, France, 119LWD6DOXWH 8QLYHUVLW\ 6FLHQWLᚏF ,QVWLWXWH 6DQ 5DᚎDHOH 'HSW RI 8URORJ\ Milan, Italy Introduction & Objectives: To develop a pre-treatment nomogram for the prediction of renal cancerVSHFLᚏFPRUWDOLW\DIWHUQHSKUHFWRP\ Material & Methods: Two cohorts of patients treated with either radical or partial nephrectomy were used: one (n = 2,474) for nomogram development, the second (n=1,972) for external validation. The QRPRJUDP SUHGLFWHG IUHHGRP IURP UHQDO FDQFHUVSHFLᚏF PRUWDOLW\ ZDV EDVHG RQ &R[ SURSRUWLRQDO hazards regression models, which used 2002 T and M stages, tumour size, symptoms at presentation, age and gender. Results: 0HGLDQIROORZXSLQSDWLHQWVZKRGLGQRWGLHRIUHQDOFDQFHUVSHFLᚏFGHDWKZDV\HDUVLQWKH GHYHORSPHQWYDOLGDWLRQFRKRUW&DQFHUVSHFLᚏFPRUWDOLW\ZDVREVHUYHGLQ SDWLHQWVZKHUHDV GLHGDVDUHVXOWRIRWKHUFDXVHV7KHDQG\HDUFDQFHUVSHFLᚏFVXUYLYDOZDVUHVSHFWLYHO\ DQG $OO SUHGLFWRUV H[FHSW IRU JHQGHU DFKLHYHG LQGHSHQGHQW SUHGLFWRU VWDWXV DW &R[ UHJUHVVLRQPXOWLYDULDEOHDQDO\VHV7KHVLJQLᚏFDQFHRIWKHSUHGLFWRUYDULDEOHVZDVFRQᚏUPHGLQFRPSHWLQJ risks regression models, which account for other cause mortality. All the variables were included in a QRPRJUDPIRUSUHGLFWLRQRIIUHHGRPIURPUHQDOFDQFHUVSHFLᚏFPRUWDOLW\,QWKHH[WHUQDOYDOLGDWLRQFRKRUW the nomogram predictions were 88.1%, 86.8%, 86.8%, and 84.2% accurate at respectively 1, 2, 5, and 10 years. Conclusions:The nomogrambased predictions may be used as benchmark data for pre-treatment management decision-making in patients with various stage of renal cell carcinoma.
845 INFLUENCE OF HOSPITAL AND SURGEON VOLUME ON OPERATIVE TIME, BLOOD LOSS, AND PERIOPERATIVE COMPLICATIONS IN RADICAL NEPHRECTOMY Yanaihara H.1, Deguchi N.1, Fuji K.2, Yasunaga H., Matsuyama Y.4, Ohe K.5 Saitama Medical University, Dept. of Urology, Saitama, Japan, 2Showa University, Dept. of Urology, Tokyo, Japan, Graduate School of Medicine, the University of Tokyo, Dept. of Health Management and Policy, Tokyo, Japan, 4School of Public Health, the University of Tokyo, Dept. of Biostatics, Tokyo, Japan, 5Graduate School of Medicine, the University of Tokyo, Dept. of Medical Informatics and Economics, Tokyo, Japan 1
Introduction & Objectives: Most of the prior volume-outcome researches for radical nephrectomy have just mentioned the relationship between hospital volume and mortality. We conducted a nationwide multi-centre survey using medical recordbased data to investigate the relationship between hospital/surgeon volumes and various outcomes including operative time, volume of blood loss, and incidence of perioperative complications in radical nephrectomy for renal cell carcinoma. Material & Methods: We collected a total of 1,704 patients undergoing radical nephrectomy at 461 hospitals in Japan between November 2006 and February 2007. In multi-variate analyses, the association between hospital/surgeon volumes and operative time, volume of blood loss, or incidence of perioperative complications were independently analyzed, regressing against age, gender, operation site, cancer stage, serum creatinine level, comorbid conditions, and surgical technique (open surgery or minimally invasive surgery).
846 PROGNOSTIC ROLE OF SSIGN SCORE IN RENAL CELL CARCINOMA Palazzo S., Lucarelli G., Palella G., Impedovo S., Ditonno P., Battaglia M., Selvaggi F.P. University of Bari, Dept. of Emergency and Organ Transplantation, Bari, Italy Introduction & Objectives: Currently outcome prediction for patients with renal cell carcinoma (RCC) is based on pathological stage and tumor grade. In 2002 the Mayo Clinic developed a predictive model which integrates the prognostic features of pathological stage, primary tumor size, nuclear grade and tumor necrosis into a single score: the SSIGN (stage, size, grade and necrosis) score. In this study we evaluated the predictive ability of SSIGN score in a cohort of patients surgically treated for RCC. Material & Methods: :H LGHQWLᚏHG SDWLHQWV WUHDWHG ZLWK UDGLFDO QHSKUHFWRP\ RU nephron-sparing surgery for RCC between 1980 and 2007. Clinical features: 218 F DQG 0 PHGLDQDJHDWVXUJHU\ \HDUV UDQJH WR PHGLDQ IROORZXS PRQWKV UDQJH WR SDWLHQWVXQGHUZHQWQHSKURQVSDULQJ VXUJHU\WKHUHPDLQLQJ XQGHUZHQWUDGLFDOQHSKUHFWRP\ ZHUH FOHDU FHOO W\SH DQG ZHUH QRW FOHDU FHOO &DQFHU VSHFLᚏF VXUYLYDO ZDV estimated for the following prognostic factors: age, sex, 1997 TNM stage, Fuhrman grade, histological subtype, tumor necrosis and SSIGN score. The median SSIGN score LQWKHSDWLHQWVVWXGLHGZDV UDQJHWR 7KHVWDWLVWLFDOPHWKRGVXVHGZHUHWKH Kaplan-Meier method and the multivariate Cox model.
Results: 1HLWKHUKRVSLWDOYROXPHQRUVXUJHRQYROXPHZDVDVLJQLᚏFDQWSUHGLFWRURI operative time or volume of blood loss. Any association between hospital volume DQGSHULRSHUDWLYHFRPSOLFDWLRQVZDVQRWLGHQWLᚏHG$VIRUVXUJHRQYROXPHVKLJK volume (>100) surgeons were relatively unlikely to have perioperative complications FRPSDUHGWRORZYROXPH VXUJHRQVEXWWKHUHODWLRQZDVQRWVLJQLᚏFDQW RGG UDWLR FRQᚏGHQFHLQWHUYDOV S
Results: 90 patients (15.5%) died of progressive disease. Mean time from surgery to FDQFHUVSHFLᚏFGHDWKZDVPRQWKV(VWLPDWHG\HDUFDQFHUVSHFLᚏFVXUYLYDOZDV $JH S VH[ S 7VWDJH S 1VWDJH S 0VWDJH (p=0.0001), nuclear grade (p=0.0001), tumor size (p=0.0002), tumor necrosis (p=0.02) DQG66,*1VFRUH S ZHUHDOOVLJQLᚏFDQWO\DVVRFLDWHGZLWKGHDWKIURP5&&LQD univariate setting. We showed also that SSIGN score maintained its predictive ability in non-clear cell RCC (p=0.0001). 4-year survival for patients with SSIGN scores of 0-1, 2, ZDV DQGUHVSHFWLYHO\$OO IHDWXUHVH[FHSWWXPRUVL]HDQGQHFURVLVZHUHVLJQLᚏFDQWO\DVVRFLDWHGZLWKGHDWKIURP RCC in a multivariate setting while for nuclear grade was showed a positive trend.
Conclusions: 5HVXOWVLQGLFDWHWKDWYROXPHRXWFRPHUHODWLRQVKLSZDVQRWFRQᚏUPHG LQ UDGLFDO QHSKUHFWRP\ $FFRUGLQJ WR RXU VWXG\ WKH MXVWLᚏFDWLRQ RI UHJLRQDOL]LQJ radical nephrectomy to centre of excellence is not clear.
Conclusions: 2XUVWXG\FRQᚏUPVWKHSURJQRVWLFDFFXUDF\RI0D\R&OLQLF66,*1VFRUH IRUFRQYHQWLRQDO5&&:HDOVRSURYLGHIRUWKHᚏUVWWLPHWKHYDOLGDWLRQRIWKLVPRGHOIRU non-clear cell RCC.
Eur Urol Suppl 2008;7(3):282
844 EXTERNAL VALIDATION A PRE-TREATMENT NOMOGRAM FOR PREDICTION OF RENAL CANCER-SPECIFIC MORTALITY Jeldres C.1, Suardi N.1, Patard J.J.2, Ficarra V., Cindolo L.4, De La Taille A.5, Salomon L.5, Tostain J.6, Mulders P.F.7, Zigeuner R.8, Prayer-Galetti T., Chautard D.9, Valeri A.10, Lechevallier E.10, Descotes J.L.10, Lang H.10, Avakian R.2, Bensaleh K.2, Mejean A.10, Bertini R.11, Montorsi F.11, Karakiewicz P.I.1 1
1
Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Centre, Dept. of Urology, Montreal, Canada, 2Graz Medical University, Dept. of Urology, Graz, Austria, Rennes University Hospital, Dept. of Urology, Rennes, France, 49LWD6DOXWH 8QLYHUVLW\ 6FLHQWLᚏF ,QVWLWXWH6DQ5DᚎDHOH'HSWRI8URORJ\0LODQ,WDO\ Introduction & Objectives: The survival of patients with metastatic renal cell carcinoma (MRCC) who are treated with nephrectomy is quite variable and may extend from few months to several years. Several variables may predict the natural history of surgically treated MRCC. These may include the local tumour characteristics, as well as the characteristics of the metastatic disease. We performed a detailed analysis of local and metastatic tumour characteristics to predict the natural history of surgically treated MRCC. Material & Methods: 7KHFOLQLFDODQGSDWKRORJLFDOGDWDZHUHSURVSHFWLYHO\UHFRUGHGIURP patients treated with cytoreductive nephrectomy. The detailed description of the distribution RI PHWDVWDWLF GLVHDVH ZDV XQDYDLODEOH LQ ZKLFK UHVXOWHG LQ HYDOXDEOH SDWLHQWV &R[ UHJUHVVLRQ PRGHOV DGGUHVVHG WKH WLPH WR FDXVHVSHFLᚏF PRUWDOLW\ 5HJUHVVLRQ FRHᚑFLHQWV ZHUH XVHG WR GHYHORS QRPRJUDPV SUHGLFWLQJ WKH SUREDELOLW\ RI FDXVHVSHFLᚏF VXUYLYDO DIWHU cytoreductive nephrectomy. The nomograms were internally validated with 200 bootstrap UHVDPSOHಬVWRUHGXFHRYHUᚏWELDV Results: The sites of MRCC were distributed as follows: 186 had lung, 107 bone, 71 nonUHJLRQDOO\PSKQRGHVEUDLQOLYHUDGUHQDODQGRWKHUVLWHV7KHPHGLDQVXUYLYDO ZDVPRQWKVZLWKDUDQJHXSWRPRQWKV7HVWHGSUHGLFWRUVRIFDXVHVSHFLᚏFVXUYLYDO included: age, gender, tumour size, T-stage, tumour size, time to metastatic progression and the site of metastatic disease. The location of metastases predicted survival. For example patients with brain metastases had substantially worse survival than the rest of the cohort (median 10 months). Conversely, patients with adrenal metastases had substantially better survival than the rest of the cohort (median 42 months). The most informative multivariable predictors of causeVSHFLᚏF VXUYLYDO ZHUH DJH DW QHSKUHFWRP\ SULPDU\ WXPRXU VL]H DQG WKH SUHVHQFH RI EUDLQ adrenal and other site metastases. The combined accuracy of these predictors was 64%. Conclusions: There is a tremendous amount of variability in the survival rates of patients with PHWDVWDWLF5&&&XUUHQWO\WKHDFFXUDF\RIH[LVWLQJULVNVWUDWLᚏFDWLRQVFKHPHVLVXQNQRZQDQG there are no models for prediction of survival in European patients. The current nomogram is simple, as it only requires 6 readily available variables and user friendly. Despite its intermediate accuracy, it represents a valuable tool and exceeds the value of quoting the median survival.
Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Centre, Dept. of Urology, Montreal, Canada, 2Rennes University Hospital, Dept. of Urology, Rennes, France, University of Padua, Dept. of Urology, Padua, Italy, 4”G. Rummo” Hospital, Dept. of Urology, Benevento, Italy, 5Henri Mondor University Hospital, Dept. of Urology, Creteil, France, 6St. Etienne University Hospital, Dept. of Urology, St. Etienne, France, 7Radboud University Nijmegen Medical Centre, Dept. of Urology, Nijmegen, The Netherlands, 8Graz Medical University, Dept. of Urology, Graz, Austria, 9Angers University Hospital, Dept. of Urology, Angers, France, 10&RPLW«GH&DQF«URORJLHGHOಬ$VVRFLDWLRQ)UDQ©DLVHGಬ8URORJLH'HSW of Urology, Paris, France, 119LWD6DOXWH 8QLYHUVLW\ 6FLHQWLᚏF ,QVWLWXWH 6DQ 5DᚎDHOH 'HSW RI 8URORJ\ Milan, Italy Introduction & Objectives: To develop a pre-treatment nomogram for the prediction of renal cancerVSHFLᚏFPRUWDOLW\DIWHUQHSKUHFWRP\ Material & Methods: Two cohorts of patients treated with either radical or partial nephrectomy were used: one (n = 2,474) for nomogram development, the second (n=1,972) for external validation. The QRPRJUDP SUHGLFWHG IUHHGRP IURP UHQDO FDQFHUVSHFLᚏF PRUWDOLW\ ZDV EDVHG RQ &R[ SURSRUWLRQDO hazards regression models, which used 2002 T and M stages, tumour size, symptoms at presentation, age and gender. Results: 0HGLDQIROORZXSLQSDWLHQWVZKRGLGQRWGLHRIUHQDOFDQFHUVSHFLᚏFGHDWKZDV\HDUVLQWKH GHYHORSPHQWYDOLGDWLRQFRKRUW&DQFHUVSHFLᚏFPRUWDOLW\ZDVREVHUYHGLQ SDWLHQWVZKHUHDV GLHGDVDUHVXOWRIRWKHUFDXVHV7KHDQG\HDUFDQFHUVSHFLᚏFVXUYLYDOZDVUHVSHFWLYHO\ DQG $OO SUHGLFWRUV H[FHSW IRU JHQGHU DFKLHYHG LQGHSHQGHQW SUHGLFWRU VWDWXV DW &R[ UHJUHVVLRQPXOWLYDULDEOHDQDO\VHV7KHVLJQLᚏFDQFHRIWKHSUHGLFWRUYDULDEOHVZDVFRQᚏUPHGLQFRPSHWLQJ risks regression models, which account for other cause mortality. All the variables were included in a QRPRJUDPIRUSUHGLFWLRQRIIUHHGRPIURPUHQDOFDQFHUVSHFLᚏFPRUWDOLW\,QWKHH[WHUQDOYDOLGDWLRQFRKRUW the nomogram predictions were 88.1%, 86.8%, 86.8%, and 84.2% accurate at respectively 1, 2, 5, and 10 years. Conclusions:The nomogrambased predictions may be used as benchmark data for pre-treatment management decision-making in patients with various stage of renal cell carcinoma.
845 INFLUENCE OF HOSPITAL AND SURGEON VOLUME ON OPERATIVE TIME, BLOOD LOSS, AND PERIOPERATIVE COMPLICATIONS IN RADICAL NEPHRECTOMY Yanaihara H.1, Deguchi N.1, Fuji K.2, Yasunaga H., Matsuyama Y.4, Ohe K.5 Saitama Medical University, Dept. of Urology, Saitama, Japan, 2Showa University, Dept. of Urology, Tokyo, Japan, Graduate School of Medicine, the University of Tokyo, Dept. of Health Management and Policy, Tokyo, Japan, 4School of Public Health, the University of Tokyo, Dept. of Biostatics, Tokyo, Japan, 5Graduate School of Medicine, the University of Tokyo, Dept. of Medical Informatics and Economics, Tokyo, Japan 1
Introduction & Objectives: Most of the prior volume-outcome researches for radical nephrectomy have just mentioned the relationship between hospital volume and mortality. We conducted a nationwide multi-centre survey using medical recordbased data to investigate the relationship between hospital/surgeon volumes and various outcomes including operative time, volume of blood loss, and incidence of perioperative complications in radical nephrectomy for renal cell carcinoma. Material & Methods: We collected a total of 1,704 patients undergoing radical nephrectomy at 461 hospitals in Japan between November 2006 and February 2007. In multi-variate analyses, the association between hospital/surgeon volumes and operative time, volume of blood loss, or incidence of perioperative complications were independently analyzed, regressing against age, gender, operation site, cancer stage, serum creatinine level, comorbid conditions, and surgical technique (open surgery or minimally invasive surgery).
846 PROGNOSTIC ROLE OF SSIGN SCORE IN RENAL CELL CARCINOMA Palazzo S., Lucarelli G., Palella G., Impedovo S., Ditonno P., Battaglia M., Selvaggi F.P. University of Bari, Dept. of Emergency and Organ Transplantation, Bari, Italy Introduction & Objectives: Currently outcome prediction for patients with renal cell carcinoma (RCC) is based on pathological stage and tumor grade. In 2002 the Mayo Clinic developed a predictive model which integrates the prognostic features of pathological stage, primary tumor size, nuclear grade and tumor necrosis into a single score: the SSIGN (stage, size, grade and necrosis) score. In this study we evaluated the predictive ability of SSIGN score in a cohort of patients surgically treated for RCC. Material & Methods: :H LGHQWLᚏHG SDWLHQWV WUHDWHG ZLWK UDGLFDO QHSKUHFWRP\ RU nephron-sparing surgery for RCC between 1980 and 2007. Clinical features: 218 F DQG 0 PHGLDQDJHDWVXUJHU\ \HDUV UDQJH WR PHGLDQ IROORZXS PRQWKV UDQJH WR SDWLHQWVXQGHUZHQWQHSKURQVSDULQJ VXUJHU\WKHUHPDLQLQJ XQGHUZHQWUDGLFDOQHSKUHFWRP\ ZHUH FOHDU FHOO W\SH DQG ZHUH QRW FOHDU FHOO &DQFHU VSHFLᚏF VXUYLYDO ZDV estimated for the following prognostic factors: age, sex, 1997 TNM stage, Fuhrman grade, histological subtype, tumor necrosis and SSIGN score. The median SSIGN score LQWKHSDWLHQWVVWXGLHGZDV UDQJHWR 7KHVWDWLVWLFDOPHWKRGVXVHGZHUHWKH Kaplan-Meier method and the multivariate Cox model.
Results: 1HLWKHUKRVSLWDOYROXPHQRUVXUJHRQYROXPHZDVDVLJQLᚏFDQWSUHGLFWRURI operative time or volume of blood loss. Any association between hospital volume DQGSHULRSHUDWLYHFRPSOLFDWLRQVZDVQRWLGHQWLᚏHG$VIRUVXUJHRQYROXPHVKLJK volume (>100) surgeons were relatively unlikely to have perioperative complications FRPSDUHGWRORZYROXPH VXUJHRQVEXWWKHUHODWLRQZDVQRWVLJQLᚏFDQW RGG UDWLR FRQᚏGHQFHLQWHUYDOV S
Results: 90 patients (15.5%) died of progressive disease. Mean time from surgery to FDQFHUVSHFLᚏFGHDWKZDVPRQWKV(VWLPDWHG\HDUFDQFHUVSHFLᚏFVXUYLYDOZDV $JH S VH[ S 7VWDJH S 1VWDJH S 0VWDJH (p=0.0001), nuclear grade (p=0.0001), tumor size (p=0.0002), tumor necrosis (p=0.02) DQG66,*1VFRUH S ZHUHDOOVLJQLᚏFDQWO\DVVRFLDWHGZLWKGHDWKIURP5&&LQD univariate setting. We showed also that SSIGN score maintained its predictive ability in non-clear cell RCC (p=0.0001). 4-year survival for patients with SSIGN scores of 0-1, 2, ZDV DQGUHVSHFWLYHO\$OO IHDWXUHVH[FHSWWXPRUVL]HDQGQHFURVLVZHUHVLJQLᚏFDQWO\DVVRFLDWHGZLWKGHDWKIURP RCC in a multivariate setting while for nuclear grade was showed a positive trend.
Conclusions: 5HVXOWVLQGLFDWHWKDWYROXPHRXWFRPHUHODWLRQVKLSZDVQRWFRQᚏUPHG LQ UDGLFDO QHSKUHFWRP\ $FFRUGLQJ WR RXU VWXG\ WKH MXVWLᚏFDWLRQ RI UHJLRQDOL]LQJ radical nephrectomy to centre of excellence is not clear.
Conclusions: 2XUVWXG\FRQᚏUPVWKHSURJQRVWLFDFFXUDF\RI0D\R&OLQLF66,*1VFRUH IRUFRQYHQWLRQDO5&&:HDOVRSURYLGHIRUWKHᚏUVWWLPHWKHYDOLGDWLRQRIWKLVPRGHOIRU non-clear cell RCC.
Eur Urol Suppl 2008;7(3):282