Eur Urol Suppl 2011;10(2):272 organ conined disease and lymph node metastasis, respectively. Mean follow- up period was 45 months (1-162) and 143 (24.3%) patients had biochemical recurrence at a median duration of 20 months. Preoperative MR imaging was 83.4% sensitive to predict organ conined disease but its speciicity was low (41.3%). The sensitivity and speciicity for detecting the seminal vesicle invasion or lymph node metastasis were 40.1% and 91.5%, 16.1% and 95.1%, respectively. The c-index from multivariate model incorporating conventional parameters only to predict biochemical recurrence was 0.628. The addition of MR imaging in this model yielded the c-index of 0.631. Conclusions: Our indings suggest that preoperative MR imaging prior to RP for clinically high risk prostate cancer had limited value to predict pathologic outcomes and no incremental beneit for risk stratiication after RP. 862 the value of CoMPuted toMograPhy in deteCting Prostate CanCer lyMPh node Metastasis is negligible even in ConteMPorary Patients with very high risk of nodal involveMent Briganti A. 1 , Gallina A. 1 , Bianchi M. 1 , Tutolo M. 1 , Di Trapani E. 1 , Di Trapani D. 1 , Passoni N. 1 , Camerota T. 1 , Cozzarini C. 2 , Corti S. 1 , Colombo R. 1 1 Urological Research Institute, Vita-Salute San Raffaele University, Dept. of Urology, Milan, Italy, 2 Urological Research Institute, Vita-Salute San Raffaele University, Dept. of Radiotherapy, Milan, Italy introduction & objectives: Previous studies have shown a limited ability of Computed Tomography (CT) to predict lymph node invasion (LNI) in prostate cancer (PCa). According to available guidelines, CT scanning may only be warranted in patients with a very high risk of harbouring nodal metastases. However, the performance characteristics of CT scan has never been tested according to prostate cancer characteristics of patients treated with extended pelvic lymph node dissection (ePLND). Materials & Methods: The study included 1541 consecutive patients with clinically localized PCa treated with radical prostatectomy and ePLND between 2002 and 2009 at a single referral center. Extended PLND was deined as removal of obturator, hypogastric and external iliac nodes. All patients underwent pre- operative CT scan for staging purposes. Suspicious CT scan was deined as the presence of at least 1 pelvic enlarged node (≥1 cm).Results were compared to the histological presence of LNI at ePLND. The sensitivity, speciicity and accuracy of CT scan in detecting LNI was calculated according to risk groups: low (PSA<10 ng/ml and biopsy Gleason scum≤6 and cT1), high (PSA>20 ng/ml or biopsy Gleason sum≥8 or cT3) and intermediate (all the remaining patients).Moreover, the performance characteristics of CT scan in detecting LNI was assessed according to most-informative risk cut-offs for LNI calculated using a nomogram developed on ePLND series. results: Overall, a CT scan suspicious for nodal involvement was found in 73 patients (4.7%). Of these, 24 (32.9%) had LNI at ePLND. Pre-operative risk groups included 471 (30.6%), 689 (44.7%) and 381 (24.7%) patients in the low, intermediate and high risk group, respectively. Overall (n=1541), sensitivity, speciicity and accuracy was 13, 96 and 54%, respectively. In the low risk group (n=471), sensitivity, speciicity and accuracy was 8.3, 96 and 52%, respectively. In the intermediate risk group (n=689), sensitivity, speciicity and accuracy was 4, 97 and 51%, respectively. In the high risk group (n=471), sensitivity, speciicity and accuracy was 18, 94 and 56%, respectively. After applying the most-informative cut-offs derived from the nomogram (namely, 36 and 50% risk of LNI), the discriminative power of CT scanning for LNI detection remained modest (sensitivity, speciicity and accuracy was 23.5, 93 and 58.3% vs. 28.9, 87.5 and 56.4% for patients with a calculated LNI risk>36 and >50%, respectively). Conclusions: We demonstrated that the sensitivity and accuracy of CT scan in detecting positive lymph nodes at RP and ePLND are low even in the presence of very high risk of nodal involvement. Sensitivity of CT scanning in detecting LNI did never exceed 30% even when the calculated risk of LNI was high. Therefore, optimal treatment planning cannot be reliably based on the results of pre-operative CT scan even in high risk disease. Poster session 72 radiCal CysteCtoMy 2 Monday, 21 March, 12.15-13.45, hall C 863 liMited lyMPh node disseCtion during radiCal CysteCtoMy underMine CanCer Control: a 12-yr single Centre exPerienCe Simone G. 1 , Papalia R. 1 , Guaglianone S. 1 , Ferriero M. 1 , Buscarini M. 2 , Gallucci M. 1 1 Regina Elena National Cancer Institute, Dept. of Urology, Rome, Italy, 2 Campus Biomedico University, Dept. of Urology, Rome, Italy introduction & objectives: Lymph node dissection (LND) should be considered a key step of radical cystectomy(RC) but there is not yet consensus on its optimal template and its therapeutic role. Materials & Methods: Between January 1998 and December 2009 data of 850 consecutive RCs performed at our Institute were collected in a prospectively- maintained database. Out of 850 patients, 81 patients where lost at follow-up and 595 with non-metastatic pure urothelial carcinoma not undergone neoadjuvant treatments were selected. All patients underwent RC and pelvic lymph-node dissection(LND). The template was chosen by surgeon and templates were recorded as standard when obturator, internal, external, common iliac and presacral lymph nodes were sent to pathologist and as limited if one or more of these lymph node sites where not removed. The prognostic role of LND extension was analyzed with univariate and multivariate Cox regression analyses together with common prognostic factors. results: Statistically signiicant variables at univariate analysis were lymph node density (p<0.001), pN (p=0.024), pT (p=0.003), lymph node count (p=0.001) and LND extension (p<0.001). Statistically signiicant different were found within all pN subgroups but pN3 where the survival beneit of a standard LND did not reach statistical threshold. (Fig 1) At stepwise Cox regression analysis only lymph node density (p<0.001) and lymph node dissection extension (p<0.001) proved to have independent role on CSS. Conclusions: A LND up to aortic bifurcation improves cancer speciic survival and should be integral part of RC. 864 is an extended bilateral PelviC lyMPhadeneCtoMy (Plnd) still needed in striCtly unilateral invasive bladder CanCer? Roth B. 1 , Zehnder P. 1 , Birkhaeuser F.D. 1 , Thalmann G.N. 1 , Krause T.M. 2 , Studer U.E. 1 1 University Hospital Berne, Dept. of Urology, Berne, Switzerland, 2 University Hospital Berne, Dept. of Nuclear Medicine, Berne, Switzerland introduction & objectives: PLND is performed bilaterally even in strictly unilaterally sited bladder cancer although a contralateral lymphatic drainage in all the endopelvic regions has never been demonstrated. The aim of this prospective trial was to map the primary lymphatic landing sites of the lateral bladder wall using a dual modality imaging approach of single photon emission computed tomography (SPECT) fused with computed tomography (CT) and intraoperative veriication by gamma probe. Materials & Methods: 40 patients scheduled for cystectomy due to bladder cancer (≤cT3, cN0, cM0) were included. One day prior to surgery we injected Tc-99m nanocolloid into the non-tumor-bearing lateral bladder wall (left: n=21; right: n=19) in local anaesthesia using a lexible cystoscope. We then performed SPECT-CT 3 and 6 h after injection of the radiopharmaceutical with a bladder lushing catheter in place. The SPECT-CT detected radioactive lymph nodes (LNs) were conirmed intraoperatively with a gamma probe at the time of PLND. Radioactive LNs were removed separately. A backup extended PLND was performed to preclude missed primary lymphatic landing sites. The SPECT-CT and intraoperative indings were used to generate a three-dimensional projection of each LN site. results: A total of 228 radioactive LNs (median 6 LNs per patient; range: 1-17) were detected. Bilateral backup extended PLND along the major vessels removed another 1427 non-radioactive LNs. 193 radioactive LNs (85%) were located on the ipsilateral side of the injection and 35 (15%) on the contralateral side (external iliac 6%, obturator fossa 5%, common iliac 4%), but none in the internal iliac region. Only 11% of radioactive LNs were located cephalad to the uretero-iliac crossing. All patients had at least 1 radioactive LN on the ipsilateral side and 40% at least 1 additional radioactive LN on the contralateral side. Conclusions: Cross-over is a common phenomenon and a bilateral extended PLND including the common iliac region up to the uretero-iliac crossing is mandatory even in strictly laterally sited bladder tumors. Still, we could not detect any radioactive LNs in the contralateral internal iliac region. Therefore, contralateral PLND may be limited to the obturator fossa, external iliac and common iliac region in select patients in which nerve sparing is attempted. 865 stage-sPeCifiC iMPaCt of PelviC lyMPh node disseCtion (Plnd) on survival in Patients with non-MetastatiC bladder CanCer treated with radiCal CysteCtoMy (rC) Abdollah F. 1 , Schmitges J. 2 , Sun M. 3 , Jeldres C. 3 , Liberman D. 4 , Tian Z. 3 , Karakiewicz P. 3 1 Vita Salute San Raffaele University, Dept. of Urology, Milan, Italy, 2 Martini-Clinic, Prostate Cancer Center Hamburg-Eppendorf, Hamburg, Germany, 3 University of Montreal Health Center, Cancer Prognostics and Health Outcomes Unit, Montreal, Canada, 4 University of Montreal Health Center, Dept. of Urology, Montreal, Canada introduction & objectives: At RC, surgeons may omit PLND more frequently in patients with favorable disease characteristics. However, this may have a negative impact on cancer control outcomes. We set to examine the effect of PLND on stage-speciic cancer-speciic- (CSM) and overall mortality (OM) rates in patients treated with RC for bladder cancer. Materials & Methods: We examined 11183 patients who underwent RC, within the Surveillance, Epidemiology and End Results database. Univariable and multivariable Cox regression analyses tested the effect of PLND on CSM and OM