significant minority of these patients showed development of protein- uria. These contradictory findings may reflect a loss of muscle mass in these patients causing serum creatinine to remain stable despite renal deterioration, and suggests that other means of estimating GFR not primarily based on serum creatinine and age may be needed in these patients. Source of Funding: None 713 EXTENDED PELVIC LYMPHADENECTOMY DURING ROBOT-ASSSTED LAPAROSCOPIC RADICAL PROSTATECTOMY FOR HIGH RISK PROSTATE CANCER Jae Hung Jung*, Wonju, Korea, Republic of; Won Sik Jang, Hyung Ho Lee, Francis Raymond Arkoncel, Ho Song Yu, Sung Joon Hong, Koon Ho Rha, Seoul, Korea, Republic of INTRODUCTION AND OBJECTIVES: We aim to compare the outcomes of D'Amico high risk prostate cancer patients who underwent extended pelvic lymphadenectomy (ePLND) versus standard pelvic lymphadenectomy (sPLND). METHODS: We retrospectively reviewed the clinical records of 224 patients who underwent robot-assisted laparoscopic radical pros- tatectomy (RALP) for patients with high risk prostate cancer between July 2005 to July 2010. 162 patients (72.3%) had sPLND (Group 1), 17 (7.3%) had a ipsilateral ePLND and contralateral sPLND according to tumor location (Group 2), and 45 (20.1%) had a bilateral ePLND (Group 3). The number and location of lymph node metastasis were analysed between three groups. RESULTS: Patient age, body mass index, prostate gland weight, and prostate specific antigen level were similar in three groups. 10 patients (43.5%) had lymph node metastasis in the group 1, 3 patients (13.0%) in group 2, and 10 patients (43.5%) in group 3. Positive lymph node was significantly higher in patients with bilateral ePLND (15.1 vs 20.9 vs 24.1, p0.001). Bilateral ePLND group was associated with a longer operation time (29.5 vs 40.3 vs 48.5 minutes, p0.001). Blood loss, hospital stay, transfusion and com- plication rates were similar in the three groups. Half of the lymph node metastasis (4/8) was found contralateral to the side of the tumor identified by prostate needle biopsy and 37.5% (3/8) by using magnetic resonance imaging. Furthermore, lymph node metastasis was seen beyond the sPLND template (3 patients in the common iliac nodes and 4 patients in the internal iliac nodes). CONCLUSIONS: Half of the lymph node metastasis was found contralateral to the side of the tumor and was also seen beyond the sPLND template. We recommend that all patients with high risk pros- tate cancer undergo bilateral ePLND. Source of Funding: None 714 OUTCOMES IN PATIENTS WITH GLEASON SCORE 8 –10 PROSTATE CANCER: RELATION TO PRE-OPERATIVE PSA LEVEL Barry B. McGuire*, Daniel C. O’Brien, Chicago, IL; Stacy Loeb, Baltimore, MD; Cheng Li, Chicago, IL; Kimberly A. Delli-Zotti, St. Louis, MO; Brian T. Helfand, William J. Catalona, Chicago, IL INTRODUCTION AND OBJECTIVES: Among prostate cancer patients, both higher preoperative prostate-specific antigen (PSA) lev- els and Gleason scores have been reported to be associated with a poorer prognosis after radical prostatectomy. However, very poorly differentiated tumors often secrete less PSA per cell than moderately- or well-differentiated tumors. In patients with Gleason 8 –10 PCa, we assessed the outcomes after radical prostatectomy in relation to their preoperative PSA levels. METHODS: From 1983 and 2003, 3,478 patients in our study group underwent radical prostatectomy. Of these, 230 had Gleason score of 8 –10 in the prostatectomy specimen. We stratified patients according to preoperative PSA into groups: 2.5 ng/ml (N=22); 2.6 – 4 ng/ml (N=14); 4.1– 10 ng/ml (N=111), and 10 ng/ml (n=83) and assessed the outcomes of biochemical progression-free sur- vival, metastasis-free survival, and cancer- specific survival. RESULTS: The median age was 63 years and median fol- low-up was 77 months. The table shows that patients with PSA levels  2.5 ng/ml had a proportionately higher recurrence rate (7 year progression free survival 35%) than those with PSA 2.6 – 4.0 and 4.1–10 (49% and 44%, respectively), but not than those with PSA 10 (31%). The 7-year metastasis free survival also was lower in patients with PSA 2.5 ng/ml (75% vs. 86%, 88%, and 87%, respectively). The cancer specific survival also was lowest among men with PSA  2.5 ng/ml (79% vs. 100%, 92%, and 90%, respec- tively). CONCLUSIONS: Patients with Gleason score 8 –10 PCa with a preoperative PSA  2.5ng/ml generally have proportionately worse outcomes after radical prostatectomy than those with intermediate PSA values. PSA ng/ml 7-year PFS (%) 7-year MFS (%) 7-year CSS (%) PSA 2.5 35 75 79 PSA 2.6–4 49 86 100 PSA 4.1–10 44 88 92 PSA 10 31 87 90 p-value* 0.07 0.61 0.36 7 year outcomes for men with high grade PCa stratified according to PSA. PFS = Biochemical Progression Free Survival, MFS = Metastasis Free Survival, CSS = Cancer Specific Survival. *log rank. Source of Funding: Supported in part by the Urological Research Foundation, Prostate SPORE grant (P50 CA90386- 05S2) and the Robert H. Lurie Comprehensive Cancer Center Grant (P30 CA60553) 715 GENETIC RISK VARIANTS ON 8Q24 ASSOCIATED WITH PROSTATE CANCER AGGRESSIVENESS Brian T. Helfand, Matthias D. Hofer*, Qiaoyan Hu, Barry B. McGuire, Chicago, IL; Daniel C. O’Brien, Chicato, IL; William J. Catalona, Chicago, IL INTRODUCTION AND OBJECTIVES: Genome wide associa- tion studies have identified that a striking set of prostate cancer (CaP) risk alleles are clustered within a large gene desert on chromosome 8q24. To date, more than 8 different risk alleles have been mapped to this region. Our prior studies conducted in relatively small populations of men have suggested that a few of these alleles are associated with aggressive pathology (e.g. Helfand et al. J Urol, 2008; 179: 2197). However, the collective influence of many of the 8q24 alleles on aggressiveness in large populations remains to be determined. METHODS: 1376 Caucasian men underwent radical prostatec- tomy from March 2003 to September 2009 at our institution and were genotyped for 5 different risk alleles located on chromosome 8q24. The associations between allele frequency, carrier status and adverse clinico-pathologic features were compared. Aggressive disease was defined as pathologic Gleason score 4+3 and/or pathologic tumor stage 3b. Possible clinically insignificant disease was defined using Ohori criteria: organ-confined, tumor volume 0.5cc, and no Gleason grade 4 (Goto et al. J Urol, 1996; 156: 1059). RESULTS: 3 of the 5 8q24 risk alleles were present at higher frequencies in men with aggressive disease (Table 1). There was a significantly higher proportion of carriers of the 8q24 risk allele, SNP rs16902094, with aggressive disease compared to non-aggressive disease (44.2% vs. 28.4%, p 0.001). In addition, there was a signif- icantly lower proportion of carriers of this allele with “insignificant” disease compared to “significant” disease (31.4% vs. 16.9%; p=0.01). CONCLUSIONS: Risk alleles on chromosome 8q24 are asso- ciated with PCa susceptibility and aggressiveness. Specifically, the rs16902094 allele significantly increase the risk of aggressive disease. Future confirmatory studies in other populations are warranted. Vol. 185, No. 4S, Supplement, Monday, May 16, 2011 THE JOURNAL OF UROLOGY e287