focal salvage. Optimal patient selection is essential to ensure those pa- tients with metastatic disease are ruled-out. This study therefore aims to develop and internally validate a prediction model for biochemical failure (BF) following focal salvage high intensity focused ultrasound (HIFU). METHODS: A comprehensive focal therapy registry was used to identify cases (Nov 2006-Sept 2014). Recurrences were assessed with multi-parametric MRI in combination with template prostate mapping bi- opsies or systematic transrectal ultrasound guided biopsies as well as PET/ CT and a bone-scan. Focal salvage HIFU was performed as quadrant ablation or hemi-ablation. Multivariable Cox proportional-hazards regression was used to quantify the effect of determinants related to BF (Phoenix-defi- nition). In multivariable analysis, factors with a p-value0.25 were retained. Multiple imputation was used for missing data. The C-statistic of the final model was calculated. Internal validation was performed using bootstrap resampling (500 datasets) after which the C-statistic and hazard ratios could be adjusted (shrinkage). Goodness-of-fit of the final model was evaluated with calibration plots. A risk score was created to aid clinical application. RESULTS: 139 prospective focal salvage HIFU patients were identified. Patients were primarily treated with external beam radio- therapy (EBRT, n¼134) or EBRT with a high dose rate brachytherapy boost (n¼5). Mean follow-up was 37 months (sd 21 months). BF occurred in 71, resulting in 50% biochemical disease-free survival (bDFS) at 35 months. After multivariable analysis, disease-free survival interval (DFSI) after primary radiotherapy, pre-salvage PSA, PSA doubling time (PSADT), prostatic volume and T-stage (both MRI-based) were independent predictors of BF. The adjusted C-statistic was 0.69. Calibration was accurate up to 36 months. Our risk score consisted of 4 groups, highly predictive of bDFS at 3 years (88% [95%CI: 75-100%], 68% [51-81%], 33% [18-48%] and 20% [0-40%]. CONCLUSIONS: Improved patient selection using the DFSI after primary therapy, pre-salvage PSA, PSADT, prostatic volume en T- stage before focal salvage can increase recurrence free rates from 50% to almost 90% at 3 years. Source of Funding: None MP18-03 PREDICTORS OF LOCAL DISEASE CONTROL AFTER SALVAGE WHOLE GLAND PROSTATE CRYOABLATION Yaw Nyame, Ahmed Elshafei*, Hans Aurora, Cleveland, OH; Robert W. Given, Virginia Beach, VA; Thomas J. Polascik, Durham, NC; Ashley E. Ross, Baltimore, MD; Vladimir B. Mouraviev, Celebration, FL; J. Stephen Jones, Cleveland, OH INTRODUCTION AND OBJECTIVES: The objective of this study was to assess factors that affect prostate biopsy results following salvage whole gland cryoablation METHODS: A retrospective analysis of 174 patients who under- went transrectal ultrasound-guided biopsy following salvage whole gland cryoablation of the prostate was performed using the Cryo OnLine Data- base (COLD). Covariates analyzed included age, ethnicity, initial prostate specific antigen (PSA), clinical stage, Gleason score, D Amico risk strati- fication, neoadjuvant therapy, PSA nadir, time to nadir, and achievement of undetectable PSA level (<0.2ng/ml). Univariate was performed using Wilcoxon rank-sum test for continuous variables and c2 test for categorical variables. A multivariate logistic regression analysis was also performed. RESULTS: 52/174 (29.8%) patients who underwent prostate biopsy had a positive result following salvage cryoablation with a me- dian time of 11 months to biopsy. On univariate analysis, patients with positive biopsy were more likely to be in the D Amico high-risk category (69.2% vs. 45.9%), more likely to have a higher median PSA nadir (0.7 vs. 0.1, P¼0.0001), and less likely to have an undetectable PSA prior to PSA (25.4% vs.57.02%, P¼ 0.0002). On multivariate analysis, PSA nadir (OR 1.08, P¼0.0132) was the only independent factor of post- operative biopsy outcome (Table 1). CONCLUSIONS: PSA nadir level following salvage whole gland cryoablation was the only factor that was independently associ- ated with prostate biopsy outcomes on multivariate analysis. This may guide the need for prostate biopsy when local failure following salvage whole gland cryoablation is suspected Source of Funding: NONE MP18-04 WHOLE GLAND SALVAGE CRYOABLATION OF PROSTATE: LONG-TERM ONCOLOGICAL OUTCOMES Khurram Siddiqui*, Michele Billia, London, Canada; Philippe Violette, Woodstock, Canada; Andrew Arifin, KimChi Tran, Jonathan Izawa, Joseph Chin, London, Canada INTRODUCTION AND OBJECTIVES: Salvage treatment op- tions for localized radio-recurrent prostate cancer (rrPC) are not stan- dardized due to the paucity of long term data on the oncologic control and toxicity of various treatment modalities. The objective of this study is to analyze the long term oncological outcomes and morbidity of salvage cryoablation (sCRYO) for rrPC. METHODS: All patients undergoing sCRYO for biopsy proven localized rrPC from 1995 to 2004 were included. The preoperative characteristic, perioperative morbidity and postoperative data including PSA, biopsy results, androgen deprivation therapy (ADT) use, metas- tasis and death were reviewed from a prospectively maintained data- base or via telephonic contact with the patient. The primary outcome was overall survival (OS). Secondary outcomes were metastasis-free survival (MFS), biochemical disease-free survival (bDFS) defined by the Phoenix criterion (PSA> nadir + 2ng/ml), time to initiation of ADT and morbidity. Survival analysis was performed by the Kaplan-Meier’s method and Cox proportional regression analysis. RESULTS: 197 whole gland salvage cryoablation procedures were performed on 187 patients from 1994 to 2004. One hundred and fifty seven (84%) had records available for follow up. The median follow-up was 117 months (IQR 55-154 months). 75 patients were followed for >10 years. The mean pre cryoablation Age was 69.45.8 years and mean pre salvage PSA was 6.6 5.7 ng/ml. 183 patients had EBRT, 3 Brachy- therapy and 1 patient had a combination of the two as the initial radio- therapy. Complete information on the oncologic status/survival of the patients was available for 126 patients at 5 year, 97 patients at 10 years and 43 patients at 15 years. During this period, 50 patients died, 14 due to prostate cancer and 36 from other causes. The median overall survival was 173 months. At 10 year, OS was 76%. Progression to metastatic prostate cancer occurred in 36 patients. The 10 year metastasis free survival was 74%. The median biochemical disease free survival (bDSF) was 56 months. The bDFS at 10 and 15 years was 34% and 22.6%. ADT was initiated in 71 patients and the mean time from cryoablation to ADT initiation was 45 months. At 10 years, 51% patients did not require any adjuvant treatment. ADT was initiated solely at the discretion of the treating urologist; however it was never prescribed before documentation of disease recurrence. On multivariate Cox regression analysis we found age at cryo (p¼0.014, HR 1.08 and CI 1.017-1.163) and pre-salvage Gleason score (p¼0.025, HR 1.29, CI 1.032-1.65) to be significant pre- dictors of overall survival. Similarly, pre salvage PSA was a significant predictor of bDFS (p¼0.007, HR1.05 and CI 1.013-1.089). When con- trolling for pre cryo PSA, Gleason score and age, nadir PSA of >1ng/ml was predictor of poor overall survival (p¼0.026, HR 1.28, CI 1.03-1.612) Nadir PSA of <1ng/m was also a predictor of early institution of ADT. A Vol. 195, No. 4S, Supplement, Friday, May 6, 2016 THE JOURNAL OF UROLOGY â e193