Castle, Erik P.

536 Background: The management of malignant mesothelioma of the tunica vaginalis of the testes (MMTVT) is not clearly defined, though retroperitoneal lymph node dissection has been reported as a diagnostic and potentially therapeutic option. Herein we present our experience with robot assisted retroperitoneal lymph node dissection (RARPLND) in a series of patients with MMTVT. Methods: The Mayo Clinic cancer registry was queried from 1972-present for all patients who had a diagnosis of MMTVT. Six patients were identified, five of whom were treated with RPLND, where four underwent RARPLND. Data was collected on pertinent demographic, preoperative, perioperative, postoperative outcomes and follow-up oncological data. Results: In five patients who underwent RPLND, the median age was 50 years (IQR 34-51), median BMI was 26.9 (IQR 26.4-30.6). Four patients originally presented with right sided symptomatic hydroceles, while one presented with right sided chronic epididymitis. Orchiectomy (one simple, two inguinal radical) was performed in three patients and hydrocelectomy in the other two patients prior to presentation. Preoperative cross-sectional imaging, including PET-CT scan in three patients, was negative for lymphadenopathy or metastasis. RARPLND was performed in 4/5 (80%) cases and concomitant hemiscrotectomy in 4/5 (80%) cases. Full bilateral template was performed in three patients and right modified template was performed in the remaining two. Median lymph node yield was 29 (IQR 22-32) and median blood loss was 275 cc (IQR 200-300). Positive retroperitoneal lymph nodes were found in 3/5 (60%) cases. Final margins of hemiscrotectomy were negative in all four cases. All patients who underwent RARPLND were discharged home on postoperative day one. One patient received four cycles of adjuvant Cisplatin and Pemetrexed. Mean follow-up was 27 months (range 2-47). No patients recurred. Conclusions: MMTVT is a very rare malignancy with high potential for retroperitoneal lymph node dissemination. Regardless of the approach, RPLND may provide both a diagnostic and therapeutic benefit, with the robotic approach affording a potentially expedited recovery.

492 Background: There is scant information about intermediate and long-term comparative outcomes between RARC and ORC, with the exception of a recent, small randomized-trial. We present our experience with RARC and ORC in managing bladder cancer patients who require cystectomy. Methods: A query of all patients who underwent radical cystectomy for a primary bladder tumor between 01/2007 and 6/2017 at our institution yield 469 patients. Data was collected on patient demographics, pre-operative information, operative details, pathology, and follow-up. Most RARCs were performed by a high-volume robotic surgeon who preferentially used the robotic approach at the start of all cases. Statistical analyses were generated using SPSS 22.0. Any open conversions were analyzed in the RARC cohort. Results: In 469 total patients, 197 (42.0%) and 272 (58.0%) underwent RARC and ORC, respectively. There were 163 (82.7%) and 224 (82.4%) males in each group and the mean ages (SD) were 71.5 (8.6) and 70.2 (10.5) years. Mean follow-up (SD) was 75 (37.7) and 61 (35.7) months, respectively. There were 130 (66.0%) and 172 (63.2%) patients who presented with T2 or greater disease (p=0.605). Soft-tissue margin rates were 4.1% for RARC and 6.7% for ORC patients (p=0.232). Neoadjuvant chemotherapy was used in 56.9% of RARC and 45.9% of ORC patients with ≥ cT2 disease (p=0.058). The 5-year overall survival was 60.0% vs 57.0% and the 10-year overall survival was 40.3% vs 44.8% for RARC vs. ORC patients, respectively (p=0.283). The 5-year recurrence-free survival was survival (RFS) was 71.7% vs 65.7% and the 10-year RFS was 70.5% vs 62.3% for the RARC vs. open groups, respectively (p=0.078). Aggressive histological variants, which were associated with greater likelihood for recurrence, were more common in the ORC group (13.7 vs. 22.4% p=0.013). Atypical recurrences (i.e. peritoneal, extrapelvic nodes) occurred in 6 (3.0%) and 14 (5.1%) RARC and ORC patients, respectively (p=0.266). Conclusions: In experienced hands, intermediate and long-term outcomes of RARC appear equivalent to ORC with regards to recurrence free survival, overall survival, and risk of atypical recurrences.

464 Background: Although the von Hippel-Lindau (VHL) tumor suppressor gene is mutated in 60% of ccRCC, deletion of VHL in mice is insufficient for tumorigenesis. Sequencing of ccRCC tumors identified mutations in SETD2, a histone H3 lysine 36 (H3K36) trimethyltransferase. We hypothesize that loss of SETD2 methyltransferase activity alters the genome wide pattern of H3K36 trimethylation (H3K36me3) in ccRCC, and contributes to the cancer phenotype. Methods: To generate a genome-wide profile of H3K36me3 in frozen nephrectomy samples and RCC cell lines, we optimized a chromatin immunoprecipitation (ChIP) protocol for the isolation of DNA associated with H3K36me3. H3K36me3 is associated with open chromatin and an H3K36me3-specific antibody was used for immunoprecipitation of endogenous H3K36me3-bound DNA. ChIP PCR primers were optimized for active genes, such as actin, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and a “gene desert” on chromosome 12 (negative control). ChIP libraries were then generated from 3 paired uninvolved kidney and RCC and 2 RCC cell lines. In order to identify H3K36Me3 upregulated regions in uninvolved kidney and RCC, reads from the ChIP sequencing were mapped to the human genome using Burrows-Wheeler Aligner and SICER algorithms. Results: Using ChIP PCR, we found that active genomic regions were enriched 15-30 fold over the negative controls indicating that the quality and yield of immunoprecipitated DNA/chromatin complexes from frozen tissue was sufficient for ChIP sequencing. A preliminary ChIP sequencing analysis of RCC cell lines and frozen ccRCC tissue indicates that H3K36me3 enriched DNA sequences were mapped to exons (31.3%) compared to introns (13.5%, p<0.001), consistent with the role of H3K36me3 in transcription. Conclusions: Genomic regions enriched for H3K36Me3 binding were identified from patient-derived tissue and RCC cell lines. Current efforts are focused on comparing the H3K36me3 profiles between matched tumor and uninvolved kidney ChIP libraries to generate a genome wide map of dysregulated H3K36me3 modifications.

287 Background: Our preliminary immunohistochemical studies of bladder cancer patients have shown that the inflammatory/immune responses associated with bladder cancer appear to be Th2-polarized as judged by the preponderance of tumor infiltrating GATA-3+ (Th2) vs. T-bet+ (Th1) T cells, as well as by evidence of tumor-associated eosinophil infiltration and activation. The standard-of-care treatment for Tis bladder cancer is intravesical administration of BCG, a Th1-polarizing immunomodulator. The aim of the present study was to assess the Th2/Th1 ratio correlation with response to BCG therapy in patients with Tis bladder cancer. Methods: Two groups of patients were studied. Group A included 20 patients that responded to BCG, and Group B included 20 patients that did not respond to therapy. Response to BCG was determined as presence/absence of disease at 6 months post-BCG therapy. The initial diagnostic biopsies of these patients were subjected to immunohistochemical analysis using commercially available antibodies specific for Th2 (GATA-3+) and Th1 (T-bet+) lymphocytes, and our unique monoclonal antibody specific for eosinophil peroxidase (EPX-mAb). Ten random high powered (400X) fields at the maximum focus of mononuclear infiltration were examined and the number of GATA 3+ and T-bet+ tumor infiltrating T cells were counted to define the G/T ratio for each patient sample. Eosinophils and their degranulation activity were counted in the same fields. Results: Patients with response to BCG therapy had a G/T ratio of ≥ 3 and patients with non-response to BCG therapy had a G/T ratio of ≤ 1.5. Correlation of these scores with subsequent patient outcomes following BCG immune therapy demonstrated that patient responsiveness (i.e., tumor elimination at 6 months) trended with the higher G/T ratio patients. In addition, initial biopsies of BCG responsive patients display a higher number of eosinophils and degranulation relative to BCG non-responsive patients. Conclusions: Evaluations of initial biopsies revealed a strong correlation between Th1 and Th2 immune response with BCG therapy outcomes. The development of specific and novel tools to identify/measure Th2 immune responses in Tis bladder cancer provides a unique opportunity to predict outcomes at the time of diagnosis.

214 Background: In our preliminary in vitro studies, we have demonstrated evidence of enhanced apoptosis and inhibition of cellular proliferation in both hormone sensitive and castrate resistant prostate cancer (PCa) cell lines using a combination of an antiandrogen (Bicalutamide) and a selective estrogen receptor modulator (Raloxifene). The aim of this study was to study the effect of the administration of these two drugs in in vivo models of castrate resistant PCa. Methods: In vivo model consisted on NCr Nude: Mice bearing s.c. human prostate (PC3 cell line) xenografts. Based on the treatment received, mice were divided into 5 groups as follows: Group 1: No drugs (control); Group 2: Bicalutamide 50mg; Group 3: Raloxifene 60 mg; Group 4: Combined administration of Bicalutamide 50 mg and Raloxifene 60 mg; Group 5 Combined administration of Bicalutamide 150 mg and Raloxifene 120 mg. A total of 10 mice where included in each group. All drugs dosages were converted to their equivalent in the mice. Drugs were administered by gavage technique to the mice once per day for a total of 14 days. Results: As expected, Bicalutamide administered alone causes minimal inhibition without reaching statistical significance (Group 2: 0.34 g Vs Group 1: 0.40 g; p=0.073). Although Raloxifene causes some marked growth inhibition, its effect is not statistically significant (Group 3: 0.31 Vs Group 1: 0.40 g; p=0.062). Bicalutamide and Raloxifene, when administered in combination, induced prominent growth inhibition in PC3 tumors when compared to the control group (Group 4: 0.26 g Vs Group 1: 0.40 g; p=0.038). Growth inhibition is significantly more evident when the drugs dosages are increased (Group 5: 0.17 g Vs Group 1: 0.40 g; p=0.024). Conclusions: The simultaneous administration of Bicalutamide and Raloxifene appears to have a synergistic effect on tumor growth inhibition in PC3 xenografts. The pathway(s) responsible for this observation may be independent of the androgen receptor as PC3 cells are AR negative and still affected by the combination over the drugs administered alone. Research is warranted to identify these potential pathways.

624 Background: The risk of intravenous contrast induced nephropathy (CIN) is not well described in post-nephrectomy or metastatic kidney cancer. Patients with nephrectomies and tumors of the kidney frequently have CKD and often require serial contrast imaging placing them at risk for CIN. Methods: This is a retrospective review of 292 patients metastatic RCC and/or with partial or radical nephrectomies (PN or RN) for kidney tumors at Mayo Clinic Arizona or Florida from 2010 through 2014. Results: We found 207 patients received at least one contrast scan, of whom 99 had metastatic RCC, 104 had local RCC and 4 had oncocytoma. At baseline (pre-nephrectomy) 41% (46/113) and 15% (11/76) of RN and PN respectively had CKD stages 3-5 and with at least 3 months follow up after both nephrectomy and contrast exposure 61% (68/113) and 33% (25/76) RN and PN developed CKD stages 3-5 respectively. The median contrast exposure scans were 4 (range 1-30 scans) with a median follow up of 26 months (range 1-96 months). In all patients with at least one contrast exposure, the median change in creatinine and GFR from baseline was 0.2 mg/dL and 8 ml/min respectively. Excluding those with bilateral nephrectomy there were 8 (8/207 = 4%) CIN events, 6/113 had a radical nephrectomy and 2/76 with 1 partial nephrectomy and 1 aborted nephrectomy (P value 0.37). Of these patients 5 (5/99 = 5%) had metastatic RCC and 3 (3/104 = 3%) local RCC. The median contrast exposure scans for patients with CIN were 8 (range 3-12) and median follow up was 31 months (range 4-47 months). Three patients all with metastatic RCC developed acute kidney injury from CIN requiring dialysis and died of complications in part related to CIN during that hospital stay or soon after. Two other patients developed CKD stage III after CIN events and 3 patients (including the one with a partial nephrectomy) showed no significant change in GFR after CIN. Conclusions: In this retrospective analysis, CIN after nephrectomy or metastatic RCC contributed to acute renal failure with 3 deaths and 2 patients developing CKD stage III. Further analysis is required to determine what risk factors for CIN in this population.