Costello, Brian A.

5082 Background: We evaluated the association of germ-line variation in 10 candidate genes important in prostate cancer biology (NKX3-1, JAK2, SLCO2B1, STAT3, CYP19A1, HSD17B4, TRMT11, PRMT3, HSD17B12, NCOA4) with failure of AA in advanced prostate cancer patients. Methods: Patients (N=619) were enrolled for genotyping at the time of AA failure. Candidate genes were selected from the literature and previous pilot studies in smaller cohorts which had identified variation in TRMT11 and HSD17B12 to be associated with AA failure. Genes were tagged using single nucleotide polymorphisms (SNPs) from HapMap with minor allele frequency of >5% and r2≥0.8. DNA was extracted from peripheral blood and genotyped using Illumina Veracode platform. The primary endpoint was time to failure on AA, defined as time from initiating continuous AA to two serial PSA increases over the nadir or clinical or imaging based progression, whichever came first. Principal component analysis was used for gene-levels tests. Association with the primary endpoint was assessed using proportional hazards regression models at the gene-level and at the SNP level. For SNP level results we estimated per allele (ordinal model) hazard ratios (HR) and 95% confidence intervals (CI) using Cox regression, adjusted for Gleason score (GS). Results: We successfully genotyped 60 SNPs (from 10 genes) in 519 subjects. The median age of the cohort was 71 years (range 43, 92). The GS distribution was 51% subjects with GS≥8; 33% with GS=7 and 17% with GS<7. Median time to AA failure for the cohort was 2.3 years (IQ range: 1.0-4.5). In gene level analyses adjusting for GS, TRMT11 (p=0.004), HSD17B12 (p=0.016), and JAK2 (p=0.044) genes were associated with time to AA failure. Four of the 18 genotyped SNPs in JAK2 were associated with AA failure after adjustment for GS (Table). Conclusions: Variation in the JAK2 gene is significantly associated with time to AA failure. Validation is needed to develop these as predictive biomarkers for AA in advanced prostate cancer. [Table: see text]

4657 Background: While active surveillance (AS) is well recognized as an acceptable treatment strategy for low-risk prostate cancer (PC), the extent to which radiation oncologists and urologists perceive AS as effective and routinely recommend it to patients is unknown. Therefore, we sought to assess the attitudes and treatment recommendations for low-risk PC from a national survey of PC specialists. Methods: A mail survey was sent to a population-based sample of 1,439 physicians in the U.S. from late 2011 and early 2012. Physicians were queried about their attitudes regarding AS and treatment recommendations for patients diagnosed with low-risk PC (PSA<10 ng/dl; T1c; Gleason 6 in one of twelve cores). Pearson Chi-square and multivariate logistic regression were used to test for differences in attitudes and treatment recommendations by physician demographics, compensation structure, primary place of employment, and specialty. Results: Overall, 321 radiation oncologists and 322 urologists completed the survey for a 45% response rate. Most physicians reported that AS is effective for low-risk PC (71%) and stated that they were comfortable routinely recommending AS (67%). Urologists were more likely to agree that AS is effective (77% vs. 67%; p=0.005) and were comfortable recommending AS (74% vs. 61%; p=0.001) compared with radiation oncologists. Most physicians recommended radical prostatectomy (47%) or radiation therapy (32%), but fewer endorsed AS (21%) for low-risk disease. After adjusting for physician covariates, radiation oncologists were more likely to recommend radiation therapy (OR: 10.97; p<0.001), while urologists were more likely to recommend surgery (OR: 4.69; p<0.001) and AS (OR: 2.18; p=0.001) for low-risk PC. Conclusions: Although AS is widely viewed as effective by both radiation oncologists and urologists, most urologists continue to recommend surgery, while most radiation oncologists recommend radiation therapy. Our results may explain in part the relatively low contemporary use of AS in the U.S.

e13052 Background: The combination of GEM and CDDP has shown activity in a variety of cancers, including biliary tract/gallbladder. Preclinical testing shows a potential benefit to the addition of EV. Methods: Using a standard 3+3 design, the MTDs of GEM + EV (Cohort I) and GEM + CDDP + EV (Cohort II) were determined as shown in the table. The MTD was defined as the dose level below the lowest dose that induces a DLT in at least one-third of pts. At the MTD of Cohort II, 10 patients were enrolled with biliary tract/gallbladder cancer (Cohort III). A weekly CBC was obtained. Assessments occurred every 3 weeks and imaging for response at every other cycle. Results: In Cohort I (N=12), no DLT occurred at dose level 0, and in dose level 1, grade 3 thrombocytopenia was found in 2 of 6 pts. The MTD for Cohort I was determined to be dose level 0. Responses were seen in 3 pts: 2 CRs (primary peritoneal, pancreatic) and 1 PR (breast). In Cohort II (N=15) DLTs at dose level 0 were neutropenia and thrombocytopenia in 2 of 3 pts and at dose level -1, thrombocytopenia in 2 of 6 pts. At dose level -2, 1 of 6 pts experienced a DLT (grade 3 thrombocytopenia), establishing this dose level as the MTD. Responses were seen in 2 pts, both PRs (ampullary, pheochromocytoma). All 10 pts have been enrolled in Cohort III with 2 DLTs (neutropenia and thrombocytopenia). Stable disease seen in 5 of 7 evaluable pts. Conclusions: Gem + EV was well tolerated at dose level 0, though dose escalation was limited by thrombocytopenia. GEM + CDDP + EV had DLTs of neutropenia and thrombocytopenia leading to the MTD of dose level of -2. Cohort III is fully accrued and the 2 DLTs are hematologic. [Table: see text]

5077 Background: A standard of care (SOC) for the adjuvant treatment of ovarian and primary peritoneal cancer is a platinum-based intravenous (IV) chemotherapy doublet. Intraperitoneal (IP) chemotherapy is also a SOC, but the high incidence of grade 3/4 adverse events has limited its acceptance among clinicians. To identify genetic markers for completion of IP chemotherapy, we analyzed SNPs in four genes known to play a role in metabolism of platinum or taxane drugs. Methods: Patients diagnosed with primary or recurrent stage III or IV epithelial ovarian or primary peritoneal cancer who had primary or secondary debulking surgery at Mayo Clinic (Rochester, MN) between January 2007 and February 2009 followed by IP chemotherapy were included in this study. A cycle was defined as IV paclitaxel (135mg/m2) on day one, IP cisplatin (100mg/ m2) on day two, and IP paclitaxel (60mg/ m2) on day eight. With prior consent from the patient, peripheral blood was obtained before treatment for extraction of germline DNA. Using a custom Illumina BeadXpress 96-plex panel, SNPs in GSTM1 (N=7), ABCB1 (N=57), CYP3A4 (N=7), and CYP2C8 (N=25) were genotyped. The association between SNPs in these subsets of genes and completion of IP chemotherapy was analyzed using linear regression. Results: Thirty-seven patients were included in this study and 16 (43.2%) completed IP chemotherapy. Twenty-two out of the fifty-seven ABCB1 SNP’s were associated with the number of cycles (p<0.15). There were no significant associations for GSTM1, CYP3A4 and CYP2C8 SNP’s. The minor A allele at missense SNP rs2229109 in the ABCB1 gene was present in five (31.2%) who completed treatment and only one (4.8%) who did not (p<0.007). However, the later was due to catheter complication. There was no association between rs2229109 and overall or progression-free survival. Conclusions: Our findings suggest that SNP rs2229109 in ABCB1 gene is associated with completion of IP chemotherapy, and potentially, less toxicity from chemotherapy. Thus, these patients may be optimal candidates for IP chemotherapy. Further studies in a larger population are needed.

TPS586 Background: Chemoradiotherapy(CRT) is a SOC for patients with muscle invasive bladder cancer (MIBC) who refuse or are not fit for radical cystectomy. Radiotherapy and chemotherapy are known to increases the PD-L1 expression in bladder cancer. Based on these observations, we hypothesized that addition of atezolizumab to CRT will increase its efficacy. Methods: This is a randomized phase III trial testing CRT with and without atezolizumab for 6 months in 475 patients with MIBC. Pts will be stratified on performance status (0-1 vs. 2); clinical stage (T2 vs T3/T4a, chemotherapy(cisplatin vs 5FU+mitomycin vs gemcitabine); and radiation field (bladder only vs small pelvis). Patients will undergo biopsy 18 week from registration. If they have residual disease they will be taken off protocol treatment and can proceed with alternative SOC option including radical cystectomy. Patients will be followed for 5 years. The primary end point of the study is bladder intact event –free survival (BIEFS). The event is comprised of: recurrence or residual muscle-invasive bladder cancer at 18 weeks or later, clinical evidence of nodal or metastatic disease, radical cystectomy, or death due to any cause. This composite endpoint is reflective of the intent of bladder preservation strategy with radical cystectomy included as one of the outcomes. Secondary end points include overall survival, modified event free survival, pathologic response at 18 weeks, metastasis free survival, cancer specific survival, rate of salvage cystectomy, rate of adverse event and QOL & PRO end points. The expected 3 year BIEFS for the control arm is 52%. The study leadership concluded that a 12% improvement in this endpoint is meaningful for this patient population. With a sample size of 475 patients the study has 85% power to detect the improvement from 52% to 64% in the BIEFS at 3 years (hazard ratio=0.68). The study team will perform translational studies evaluating tumor tissue, whole blood and urine for molecular and immunologic markers of immune response and response to RT. Successful completion of this trial could lead to a new treatment paradigm for patients with muscle invasive bladder cancer. Clinical trial information: NCT03775265.

83 Background: PT-112, the first pyrophosphate conjugate in Phase I/II clinical development, induces robust immunogenic cell death and is osteotropic, prompting study in mCRPC during Phase I. We report safety and efficacy findings in the mCRPC sub-population given PT-112 monotherapy (NCT02266745) or in combination with avelumab (“PAVE”) (NCT03409458, ongoing). Methods: Patients (pts) received PT-112 days 1, 8, 15 of a 28d cycle; pts on the PAVE combination also received 800 mg avelumab days 1 and 15; all enrolled during dose escalation or at PT-112 doses previously deemed safe. Pts on therapy (Tx) for ≥2 cycles or who stopped Tx due to progressive disease or treatment-related adverse events (TRAEs) were evaluated for exploratory efficacy. Results: 10 pts on PT-112 (200-420 mg/m2) and 18 pts on PAVE (150-200 mg/m2 PT-112) were evaluable for safety. The most common PT-112 TRAEs were thrombocytopenia (70%, grade (Gr) ≤3) and fatigue (40%, Gr ≤2) with mono-Tx; nausea (50%, Gr ≤2) and fatigue (39%, Gr ≤2) with PAVE. Mono-Tx pts had 6.5 median prior lines of Tx; 2/7 pts with measurable disease (MD) had reductions in target lesions; prostate specific antigen (PSA) declined in 3/10 pts, 1 with ≥50% reduction; and serum alkaline phosphatase (ALP) reductions were seen in 9/10 pts. PAVE pts had 5.5 median prior lines of Tx; radiographic reductions were seen in 2/9 pts with MD; PSA declined in 6/14 pts, 3 with ≥50% decrease, with 1 responder progression free 11.3 months (microsatellite stable). An additional response is ongoing at 4 mos, with 93% PSA decrease and 48% reduction in target lesions (PIK3CBmut & PTEN loss). 13/14 pts with bone metastases had ALP reductions. Pain improvement, opioid cessation and improved performance status were noted in both cohorts. Conclusions: PT-112 was well-tolerated with evidence of efficacy in mCRPC as mono-Tx and in combination with avelumab in heavily pre-treated pts. Bone pain improvement and nearly universal observation of ALP reduction suggest marked therapeutic activity of PT-112 in bone metastases. Serologic / RECIST responses and prolonged disease control in multiple pts substantiate further development of PT-112 in mCRPC. Clinical trial information: NCT02266745, NCT03409458.

725 Background: The impact of surgical resection of renal tumors on peripheral immune related cells is not well characterized, and has potential implications as biomarkers for systemic immune therapy are being developed. We sought to assess the effect of surgical resection on circulating cytotoxic T-cells (CTLs) for patients with renal tumors. Methods: We prospectively enrolled 40 patients undergoing partial, radical, or cytoreductive nephrectomy (PN, RN, CN) for unilateral primary renal tumors between 2016 and 2018. Blood draws were performed pre-op, 1 day post-op, and 3 months post-op. Flow cytometry was performed on peripheral blood mononuclear cells (PBMCs). The % of PBMCs expressing CD11a/CD8 (CTLs) were assessed. The % of CTLs expressing PD-1, Bim (a downstream PD-1 pathway pro-apoptotic mediator), CX3CR1/GZMB (an effector memory T-cell phenotype), and Ki67 (a proliferation marker) were assessed. Median changes in % of CTLs were evaluated with the Wilcoxon signed rank test. Comparisons between aggressive (pT3-4, N1, M1, or aggressive histology (high-grade, coagulative necrosis, sarcomatoid dedifferentiation, or specific RCC-variant histologies)) versus indolent tumors were assessed using the Wilcoxon rank sum test. Results: Twenty, 12, and 8 patients underwent RN, PN, and CN, respectively. Thirty, 7, and 3 patients had clear-cell RCC, non-clear cell RCC, and oncocytoma, respectively. Twenty-three and 17 patients had aggressive and indolent tumors, respectively. While there was no significant change at 1 day, by 3 months there were significantly increased CTLs among PBMCs (+1.3%; p=0.004). There was significant decrease in CX3CR1+GZMB+ CTLs (-2.9%; p<0.001) and significant increase in Bim+ CTLs (+3.0%; p=0.03) at 1 day but not 3 months. Interestingly, Ki67+ CTLs increased at day 1 (+0.6%; p<0.001) but decreased by 3 months (-0.6%; p=0.001). There were no significant changes in PD-1+ CTLs at either time point. There were no significant changes in any CTL characteristics at 1 day or 3 months between aggressive and indolent tumors. Conclusions: These findings characterize changing CTL profiles over time with surgical resection of renal tumors which may help guide biomarker development.

709 Background: Several immunotherapy (IMT) combinations either as an IMT doublet or IMT in combination with TKIs are now available for first-line therapy of metastatic renal cell carcinoma (mRCC). In the absence of head-to-head clinical trials, we performed an indirect comparison of frontline treatment options to provide clinical guidance. Methods: Medline, Embase and Cochrane Library were searched to identify relevant trials. Hazard ratios (HR) and confidence interval (CI) for primary outcome of progression-free survival (PFS) and secondary outcome of overall survival (OS) were abstracted. Network meta-analysis was performed using a multivariate, consistency model, random-effects meta-regression. Data on Grade 3 or higher AEs was abstracted and meta-analyzed. Pre-specified subgroup analyses were performed based on risk categories (high risk vs low and intermediate vs), history of prior nephrectomy, PD-L1 positivity, age and sex. Risk for bias(RoB) was assessed using the Cochrane Collaboration’s tool. Results: Nine studies were included for PFS analysis, and 6 studies for OS analysis. Avelumab-axitinib (AA) (HR 0.69, 0.48-0.96), and Pembrolizumab-Axitinib (PA) (HR 0.89, 0.50-0.96) significantly improved PFS, while there was no significant PFS benefit with atezolizumab-bevacizumab (AB) (HR 0.83, 0.62-1.13) and nivolumab-ipilimumab (NI) (HR 0.85, 0.63-1.15) as compared to Sunitinib. PA (HR 0.53, 0.38-0.75) and NI (HR 0.63, 0.44-0.90) significantly improved OS, while AB (HR 0.93, 0.75-1.12) showed no significant OS benefit as compared to Sunitinib. NI and AB had significantly fewer grade ≥3 AE than sunitinib. No significant interaction was found by risk group, age, sex or prior nephrectomy. Significant interaction was found by PD-L1 expression(p<0.001), with significant PFS improvement in PD-L1 positive (HR 0.62, 0.53-0.73) but not in PD-L1 negative patients (HR 0.92, 0.81-1.05). Overall, RoB was low amongst included studies. Conclusions: AA improved PFS, NI improved OS, whereas PA improve both PFS and OS as compared to sunitinib monotherapy. However, no IMT combination was superior to other combinations. Cost effectiveness analysis will be reported separately.

696 Background: A number of treatments for metastatic renal cell carcinoma (RCC) have been FDA approved since December, 2005. Trials evaluating these agents excluded non-clear cell histologies and therefore the efficacy of these treatments remains unclear in patients (pts) with metastatic non-clear cell RCC (mncRCC). Methods: This is a single arm Phase II study designed to determine the efficacy of pazopanib in mncRCC. Treatment was pazopanib 800 mg by mouth daily until progression or intolerability of treatment. Dose reductions were allowed for toxicity with dose level -1 being 600 mg daily and dose level -2, 400 mg daily. Cycles were every 28 days. Main eligibility criteria included: (1) age ³ 18 years old; (2) histologic confirmation of ncRCC; (3) ECOG Performance Status (PS) 0, 1, or 2; (4) up to one prior treatment for mncRCC (5) measurable or non-measurable metastatic disease per RECIST criteria. The primary endpoint was overall survival rate at 12 months. Secondary endpoints were best tumor response rates after two cycles, PFS, OS and toxicity. Results: 38 pts were enrolled between May 16, 2013 and February 1, 2018. 35 pts were evaluable for primary endpoint. Median age 63 years old and 22/35 (62.9%) were male. Most common histology was papillary RCC, 14/35 (40%), and most common number of metastatic sites was 1. 29/35 (82.9%) had prior nephrectomy and 30/35 (85.7%) had no prior systemic therapy. Median number of cycles was 5. 12-month OS was 65.7% (90% CI, 50.5-78.9%). Best response in first 2 cycles: PR 11%, SD 60%, PD 9%, not evaluated 20%. Median PFS was 7.5 months (90% CI, 5.0-11.0); median OS 18.9 months (90% CI, 13.0-NE). Grade 3 / 4 AEs seen in 25 pts with hypertension, transaminitis and abdominal pain being most common (seen in >10%). Conclusions: There are limited data about the efficacy of available therapies for mncRCC. This study shows that pazopanib has similar efficacy compared to historical data in clinical trials using other TKIs in mncRCC. The safety profile of pazopanib in pts with mncRCC was found to be similar to that found in prior clinical trials studying pazopanib. Our findings warrant further investigation into the utility of pazopanib in metastatic ncRCC. Clinical trial information: NCT01767636.

610 Background: Personalized treatment for clinical T1 renal cortical masses (RCMs) should account for competing risks related to tumor and patient characteristics. Using a contemporary multi-institutional cohort, we developed treatment-specific prediction models for cancer-specific mortality (CSM), other-cause mortality (OCM), and 90-day complication rates for patients managed with surgery, thermal ablation (TA), and active surveillance (AS). Methods: Preoperative clinical and radiological features were collected for eligible patients aged 18-91 years treated at four academic centers from 2000-2016. Prediction models used competing risks regressions for CSM and OCM and logistic regressions for 90-day Clavien >3 complications, adjusting for tumor size as well as patient age, sex, ECOG performance status (PS), and Charlson comorbidity index (CCI). Predictions accounted for missing data using multiple imputation. Results: After excluding 25 patients with no follow-up, the cohort included 4995 patients treated with radical nephrectomy (RN, n=1270), partial nephrectomy (PN, n=2842), thermal ablation (n=479), or active surveillance (n=404). Median follow-up was 5.1 years (IQR 2.5-8.5). Predictions from the fitted model are shown in an online calculator ( https://rgulati.shinyapps.io/rcc-risk-calculator ). To illustrate the use of this calculator for a specific patient, a 70-year-old female with a 5.5 cm RCM, PS of 2, and CCI of 3 has a predicted 5-year CSM of 4-7% across treatments, 5-year OCM of 34-49%, and 90-day risk of Clavien ≥3 complications of 4%, 10%, and 6% for RN, PN, and TA respectively. Conclusions: Personalized treatment selection for cT1 RCM is challenging. We present a competing risk calculator that incorporates pretreatment features to quantify competing causes of mortality and treatment-associated complications. Pending validation, this tool may be used in clinical practice to provide patients with estimated individualized treatment-specific probabilities of competing causes of death and complication risks to facilitate shared decision-making.