570 AN ANALYSIS OF PATIENTS WITH T2 RENAL CELL CARCINOMA ACCORDING TO TUMOR SIZE: A POPULATION-BASED ANALYSIS

Vol. 187, No. 4S, Supplement, Sunday, May 20, 2012

570 AN ANALYSIS OF PATIENTS WITH T2 RENAL CELL CARCINOMA ACCORDING TO TUMOR SIZE: A POPULATIONBASED ANALYSIS Maxine Sun*, Montreal, Canada; Marco Bianchi, Milan, Italy; QuocDien Trinh, Detroit, MI; Jens Hansen, Hamburg, Germany; Firas Abdollah, Milan, Italy; Nawar Hanna, Montreal, Canada; Shahrokh Shariat, New York, NY; Paul Perrotte, Pierre Karakiewicz, Montreal, Canada INTRODUCTION AND OBJECTIVES: The most contemporary version of the TNM staging for renal cell carcinoma (RCC) subclassifies patients with T2 lesions between T2a and T2b, according to a cut-off of 10 cm. In the current study, we examined this and other cut-offs within a large North-American cohort. METHODS: Between years 1988 –2006, within the Surveillance, Epidemiology, and End Results database, patients with T2 N0-2 M0-1 RCC treated with a nephrectomy were abstracted. Tumor size was evaluated according to several cut-offs: ⱖ8 cm, ⱖ9 cm, ⱖ10 cm, ⱖ11 cm, and ⱖ12 cm. Kaplan-Meier and life tables for cancer-specific mortality (CSM) were computed. Several Cox regression modes were fitted for prediction of CSM, using different cut-offs. The predictive accuracy of various cut-offs was compared using the area under the curve and methods of calibration. RESULTS: A total of 4963 patients were identified. KaplanMeier analyses revealed statistically significant CSM-free survival differences between all examined cut-offs. In multivariable Cox-regression models, all tested tumor size cut-offs emerged as independent predictors of CSM. Of all cut-offs, the value of 9 cm (55.0%) and 11 cm (55.0%) achieved the highest discrimination in univariable analysis, followed by 10 cm (53.9%), 12 cm (53.9%), and 8 cm (53.1%). When the cut-offs were combined with all other variables, the 11 cm (68.8%) achieved the highest discrimination. CONCLUSIONS: The discriminant properties of the cut-off of 10 cm are adequate. Nonetheless, several other cut-offs (9 and 11 cm) showed very similar discriminant properties. In consequence, the 10 cm cut-off is not clearly superior to neighboring cut-off values. Source of Funding: None

571 TUMOR LOCATION AS AN INDEPENDENT PROGNOSTIC FACTOR FOR CANCER SPECIFIC SURVIVAL AND RECURRENCE FREE SURVIVAL IN PATIENTS WITH UPPER TRACT UROTHELIAL CARCINOMA Sejun Park*, Jongwon Kim, Wansuk Kim, Seoul, Korea, Republic of; Jinsung Park, Daejun, Korea, Republic of; Cheryn Song, Bumsik Hong, Jun Hyuk Hong, Choung-soo Kim, Hanjong Ahn, Seoul, Korea, Republic of INTRODUCTION AND OBJECTIVES: Pathological stage and tumor grade have been established as prognostic factors for upper tract urothelial carcinoma (UTUC), whereas the tumor location as an independent prognostic factor has been still debated probably because of small sample size and the difference in tumor characteristics among the studies. We aimed to evaluate the prognostic impact of tumor location on oncologic outcomes of radical nephroureterectomy in patients with UTUC. METHODS: We reviewed the medical records of 392 consecutive patients who underwent radical nephroureterectomy for UTUC between 1991 and 2010. Five-year cancer specific survival (CSS) and 5-year recurrence free survival (RFS) were compared according to tumor location. Cox regression analysis was used to determine the prognostic factors for CSS and RFS. Mean follow up period was 52.6 months. RESULTS: Overall, 188 patients had pelvic tumors (47.9%), 171 patients had ureteral tumors (43.6%) and 33 patients had multifocal tumors (8.5%). Uretereral tumors were associated with a higher local recurrence rate than pelvic tumors (12.3% vs. 4.3%, p⫽0.001), whereas distant metastasis rate was not different between pelvic tumors and ureteral tumor. Five-year CSS was 85.2% for pelvic tumors, 78.1 % for

THE JOURNAL OF UROLOGY姞

e233

uretereral tumors, and 78.7% for multifocal tumors respectively (p⬍0.002). Five-year RFS was 83.1% for pelvic tumors, 70.0% for ureteral tumors, and 75.5% for multifocal tumors respectively (p⫽0.009). Independent prognostic factors of 5-year CSS and RFS were pT stage, pN stage and tumor location on multivariate analysis (all, p⬍0.05), whereas tumor grade, lymphovascular invasion, surgical margin status and microscopic tumor architecture were not associated with oncologic outcomes. CONCLUSIONS: Tumor location, pT stage and pN stage were independent prognostic factors for CSS and RFS in patients with UTUC after radical nephroureterectomy. Uretereral tumors had a worse prognosis than pelvic tumors and were prone to recur locally probably due to limited anatomical barrier to tumor spread. Source of Funding: None

572 CLINICAL, MOLECULAR AND GENETIC CORRELATES OF LYMPHATIC SPREAD IN CLEAR CELL RENAL CELL CARCINOMA Nils Kroeger*, David Seligson, Tobias Klatte, Edward Rampersaud, Fre´de´deric Birkha¨user, Nagesh Rao, John Leppert, Nazy Zomorodian, Fairooz Kabbinavar, Arie Belldegrun, Allan Pantuck, Los Angeles, CA INTRODUCTION AND OBJECTIVES: While it is well known that clear cell renal cell carcinoma (ccRCC) that presents with lymphatic spread is associated with an extremely poor prognosis, its molecular and genetic biology is poorly understood. Therefore, we aimed to define the unique clinicopathological, molecular, and genetic characteristics of ccRCC having positive lymph nodes. METHODS: We retrospectively examined clinicopathological features for their association with lymphatic spread. These features were used as co-variates to further adjust the results of molecular and genetic analyses. Patients with positive lymph nodes were compared with localized patients. ROC analyses as well as uni- and multivariable regression models were carried out to test correlations with lymphatic spread. RESULTS: In clinical analyses, smoking history (OR1.791; 95% CI: 1.028 - 3.122; p⫽0.040), T stage (OR 3.732; 95% CI: 2.591 5.376; p⬍0.0001), Fuhrman grade (OR 2.012; 95% CI: 1.383 - 2.927; p⬍0.0001), ECOG PS (OR 1.828; 95% CI: 1.143 - 2.924; p⬍0.0001), and MVI (OR 6.359; 95% CI: 2.739 - 14.763; p⬍0.0001) were independently associated with lymphatic spread. After adjustment for these clinical variables, low CAIX (OR 3.362; 95% CI: 1.037 - 10.906; p⫽0.043) and high epithelial VEGFR-2 (OR 4.821; 95% CI: 1.137 20.450; p⫽0.033) protein expression were independently related to lymphatic spread. Furthermore, loss of chromosome 3p (OR 0.091; 95% CI: 0.026 - 0.321; p⬍0.0001) was independently associated with a lower risk of lymph node involvement. CONCLUSIONS: Low CAIX and high epithelial VEGFR-2 protein expression, as well as low rate of chromosome 3p loss were associated with metastatic spread into lymph nodes, suggesting that RCCs that spread via the lymphatics develop independently of VHL inactivation. These results provide a genetic and molecular basis for the poor prognosis and lack of response to IL-2 in lymph node positive ccRCC. Source of Funding: None

573 THE RELATIONSHIP OF LYMPH NODE DISSECTION WITH RECURRENCE AND SURVIVAL FOR PATIENTS TREATED WITH NEPHRECTOMY FOR HIGH-RISK RENAL CELL CARCINOMA Simon Kim*, R. Houston Thompson, Christopher Weight, John Cheville, Christine Lohse, Stephen Boorjian, Bradley Leibovich, Rochester, MN INTRODUCTION AND OBJECTIVES: The role of lymph node dissection (LND) in the surgical management of renal cell carcinoma (RCC) is controversial. Although LND improves staging accuracy and