The addition of MDT to HT induced systemic T-cell activation and expansion, which was not observed in the HT-only arm. This systemic immune response was independently associated with improved PFS. In addition to cytoreduction of macroscopic disease, MDT-induced immune education may be an important complementary mechanism of micrometastatic control in oligometastatic prostate cancer.

250 Background: 20-30% of prostate cancers respond poorly to androgen receptor (AR)-directed therapies and have an atypical, virulent course. The aggressive variant prostate cancers ( AVPC) encompass virulent androgen indifferent tumors and are characterized by combined defects in ≥ 2/3 of TP53, RB1, and PTEN (AVPC-m). Post-hoc analysis in a phase I/II trial indicated a benefit of carboplatin+cabazitaxel in men with AVPCm+ metastatic castrate resistant prostate cancer (mCRPC). We assess the feasibility of detecting the AVPCm in CLIA-certified environments. Methods: Men with progression of mCRPC undergoing biopsies for standard molecular profiling at MD Anderson Cancer Center (MDACC) were eligible. TP53, RB1, and PTEN expression was assessed by immunohistochemistry (IHC) and read by 2 pathologists independently. Next generation sequencing of solid tumor DNA (stDNA) and circulating tumor DNA (ctDNA) was performed on the MDACC Molecular Diagnostics Laboratory and the University of Washington OncoPlex ctDNA platforms respectively . Results: Of 64 eligible patients, 49 (77%), 37 (58%), and 54 (84%) had results reported for IHC, stDNA, and ctDNA respectively. Concordance of IHC reads between pathologists was variable: Kappa 0.78, 0.67, and 0.91 for TP53, RB1, and PTEN respectively. 36 (66.7%) ctDNA samples had low tumor content, such that false negatives could not be excluded. AVPC-m was detected in 14 (29%) of patients by IHC per the original reading pathologist, 2 (5%) by stDNA NGS, and 7 (39%) by ctDNA NGS. Median turnaround times for IHC, stDNA and ctDNA were 12 (2-35), 28 (12-56) and 16 (11-56) days respectively. Conclusions: We show the operational characteristics of the AVPCm detection using CLIA certified assays. IHC detected a greater proportion of patients with AVPCm+ tumors than stDNA, likely due, partly, to the challenges of copy number loss detection. ctDNA detected the greatest proportion of AVPCm+ tumors but was not evaluable in two thirds of patients. These data serve to inform the design of clinical trials using AVPCm as a criterion for patient selection or stratification. [Table: see text]

522 Background: mSCUC is an aggressive cancer with historically poor outcomes. Methods: We performed a retrospective analysis of 216 pts with mSCUC treated at MD Anderson from 1985-2020. We captured baseline demographic and clinical characteristics, metastatic sites, systemic treatments, and survival. Median overall survival (mOS) was calculated from start of frontline (1L) therapy for metastatic disease. Results: Median age was 67 years, 84% male sex, 91% primary tumor in bladder, and 88% white race. Histology included pure SCUC (30%), predominant SCUC (44%), focal SCUC (25%) and large cell (1%). Metastatic sites were lymph nodes (54%), bone/peritoneum (39%), liver (38%), brain (31%) and lung (26%). Brain metastases frequency varied across histology (pure SCUC 48%, predominant SCUC 24.5%, focal SCUC 24%, p=0.06). Treatment regimens and mOS by frontline regimen are listed (Table). mOS was 11.91 months for treated pts with sufficient follow up (n=155). mOS by 1L treatment: EP (9.51 months), IA/EP (14.10 months), other platinum chemotherapy (13.26 months), nonplatinum chemotherapy (9.74 months), immune checkpoint inhibitor (ICI) (6.51 months), EP+ICI (undefined). Of 140 pts treated with 1L chemotherapy, 43 (31%) were able to receive second line (2L) therapy. mOS among those who received 2L chemotherapy (n=27) was 13.65 months vs those who received 2L ICI (n=16) was 22.53 months (HR=0.53, 95% CI=0.26-1.10; p=0.11). Conclusions: mSCUC is a devastating malignancy with high degree of brain involvement, and poor prognosis. 1L ICI therapy alone has suboptimal survival outcomes and a chemotherapy backbone remains necessary. A deeper understanding of SCUC’s driver biology is essential for developing better therapies. [Table: see text]

TPS273 Background: Prostate cancer is a disease of older age, and there is significant overlap with age-related illnesses. Androgens play a critical role in prostate cancer but also have been linked to cardiovascular (CV) disease, metabolic syndrome, bone health, and frailty. ADT is associated with an increased risk of CV events primarily due to increased rates of diabetes, dyslipidemia, and adverse changes in body composition. We assume that prostate cancer and age-related illnesses share biology with therapeutic implications in a subset of men. The ProTrio trial will test the hypothesis that a 16-week digital risk factor modification program will improve 10-year atherosclerotic cardiovascular disease (ASCVD) risk score as compared to usual care in men with potentially lethal prostate cancer receiving ADT. Methods: This single center, open-label, randomized phase II trial is evaluating a 16-week digital risk factor modification program, which includes an exercise prescription (150 minutes of titrated cardiovascular exercise and resistance exercise 2 times a week) with support via a smart phone application (Moving Analytics) and Fitbit, in patients with prostate cancer who are receiving ADT compared to usual care with Fitbit monitoring. Eligible patients are men receiving ADT +/- an androgen signaling inhibitor for at least 3 months prior to enrollment. The study will enroll to two cohorts: (1) distant metastatic disease and (2) biochemically recurrent or regional lymph node disease. Patients will be stratified according to their baseline ASCVD risk score. The primary endpoint is the difference in ASCVD risk score post-intervention. Secondary endpoints include difference in metabolic severity index z-score, PSA velocity, 6-minute walk, 5X Sit-to-Stand, body composition, physical activity measured by Fitbit, and health-related quality of life. Up to 100 patients will be enrolled in each cohort. A clinically relevant improvement in ASCVD score reduction is 2.5%. With 44 men per arm, after 12% dropout, we can detect an improvement of 2.5% vs. 0% at a 2-sided 5% significance level with 80% power assuming a standard deviation (SD) of 4.125%. A sample size adjustment is planned based on the SD of the first cohort to reach 44 patients. Each cohort will undergo futility analysis once half the men are enrolled. The study is enrolling at MD Anderson Cancer Center. Clinical trial information: NCT05054296 .

583 Background: TEM has level 1 evidence in aRCC with poor-risk disease. No trial compared a VEGFR-TKI with TEM as first-line (1L) therapy in this disease state. Methods: We randomly assigned (1:1) treatment-naïve pts with aCCRCC and > = 3 risk factors (as per Hudes et al., NEJM 2007) to receive PAZ 800 mg po qd or TEM 25 mg iv qw. Pts were offered to receive the alternative agent at disease progression (PD). The primary endpoint was progression-free survival (PFS), and the secondary endpoints were overall survival (OS), objective response rate (ORR) and safety. A blinded radiologist assessed the radiographic response using RECIST v1.1. A sample size of 90 pts was based on an assumption of improved median PFS from 3.8 mo with TEM to 6.1 mo with PAZ. Pts were stratified by prior nephrectomy (Nx). The Kaplan-Meier method was used for PFS and OS analysis, and the Fisher’s exact test was used for comparison of ORR between PAZ and TEM. Results: The study was closed to new patient enrollment when the results of the CheckMate 214 and CABOSUN studies were presented at ESMO 2017. A total of 69 pts were eligible and evaluable (median age 61, 52 males [75%], 44 [64%] had poor-risk by IMDC criteria). Thirty pts [43%] had prior Nx. Thirty-five pts received PAZ (intermediate-risk 13, poor-risk 22) and 34 pts received TEM (intermediate-risk 12, poor-risk 22). Of the 69 pts, 67 had PD or died. The median PFS was 5.2 mo (95% CI: 3.6 –7.4) for PAZ and 2.6 mo (95% CI: 1.9 –4.2) for TEM (p = 0.16). In 1 pt, no date of death was available. Of the remaining 68 pts, 58 (85.3%) have died. The median OS was 12.0 mo (95% CI: 8.3–20.1) for PAZ and 7.4 mo (95% CI: 5.3–17.4) for TEM (p = 0.61). Sixty-eight pts were evaluable for response: 9/35 pts (26%) who received PAZ and 2/33 pts (6%) who received TEM had partial response (p = 0.046). Adverse events (AEs) were consistent with the known safety profiles of PAZ and TEM. Only 2 pts in each arm discontinued treatment due to AEs. Conclusions: PAZ extended PFS and OS and yielded a significantly higher ORR than TEM as 1L therapy in pts with aCCRCC and intermediate/poor-risk disease. Clinical trial information: NCT01392183.

Castration-resistant prostate cancer (CRPC) is a lethal disease. A subset of these patients present with atypical clinical characteristics and aggressive disease behavior. These patients, classified as Aggressive Variant Prostate Cancer (AVPC), may derive benefit from platinum-based chemotherapy. However, the identification of AVPC patients is often challenging due to heterogeneity in clinical presentations and potentially in biological drivers, which may not be fully reflected in a biopsy obtained from a single tumor site. To address these challenges, we developed an unbiased genome sequencing method called PEGASUS (Plasma Exome and Genome Analysis by Size-Selection and Unbiased Sequencing) that enables the simultaneous detection of copy number aberrations and exome mutations from circulating-tumor DNA (ctDNA). We applied this method to plasma specimens obtained from 160 CRPC patients with and without AVPC participating in a randomized trial of cabazitaxel alone or in combination with carboplatin. We were able to successfully isolate ctDNA (>2 ng) for genomic profiling using PEGASUS in 91/160 (57%) CRPC patients. Among these 91 CRPC patients, we detected common prostate cancer mutations as well as genomic copy number changes in genes such as RB1, PTEN and TP53 and several other genes associated with DNA repair. One of the most salient features that distinguished the patients’ outcomes was whether they had diploid or aneuploid genomes detected by whole genome profiling of the ctDNA. Patients with aneuploid ctDNA had worse progression-free and overall survival. Overall, our data shows the feasibility of performing whole-genome and exome profiling of ctDNA in CRPC patients to characterize the molecular features associated with distinct clinical presentations. Candidate markers associated to benefit from platinum-based chemotherapy are being evaluated. These results pave the way for future clinical applications in biomarker discovery to assist treatment decisions in prostate cancer patients. Citation Format: Naveen Ramesh, Emi Sei, Pei Ching Tsai, Christopher Logothetis, Paul Corn, Ana Aparicio, Amado J. Zurita, Nicholas E. Navin. Plasma genome sequencing identifies prostate cancer patients that are sensitive to platinum-based therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 446.

259 Introduction: RA-223 prolongs OS in BM-CRPC pts but predictive markers are lacking. We hypothesized that molecular profiling of plasma-derived extracellular vesicles (EVs) from BM-CRPC pts treated with RA-223 would predict efficacy and inform development of combinations. Methods: Paired baseline (BL) and end of treatment (EoT) blood samples were collected from 25 pts treated with RA-223. Pts were stratified by median OS into 3 groups: 1) 6.3 months (range 3.3-8.5, n=5 pts, UnFav OS), 2) 18.2 months (range 12.6-19.4, n=7 pts, Middle OS) and 3) 27.6 months (range 23.1-33.8, n=13 pts, Fav OS). EVs were isolated from plasma and analyzed for specific proteins and RNA transcripts. Marker profiles by domain and analytic method were determined in pts from the stratified OS groups. Results: At EoT, there is enrichment of RNA transcripts and proteins involved in regulating immune responses, DNA Damage Response (DDR) and bone metabolism (Table). By benchmarking molecular signatures in pts with UnFav Os and Fav OS, we could analyze the molecular profiles of pts in the Middle OS group and stratify them to UnFav OS or Fav OS groups. Conclusion: Plasma-based molecular markers are predictive for RA-223 efficacy. An association between OS and immune and DDR profiles in men with BM-CRPC may indicate mechanism of action and identify subsets most likely to benefit. These preliminary results provide rationale for combination therapy with RA-223 and immune-checkpoint blockade or PARP inhibition. Clinical trial information: NCT02135484. [Table: see text]

11 Background: Combined external beam radiotherapy (RT) and androgen deprivation therapy (ADT) improves survival over RT alone for high risk prostate cancer (PC). Long-term ADT use, currently recommended for high risk PC, also increases toxicity. Recent data suggests synergistic efficacy with the addition of abiraterone acetate plus prednisone (AAP) to RT/ADT. Potent androgen blockade may provide biochemical control with short-term ADT course in men with aggressive but localized PC. Methods: This was a two center prospective phase 2 single arm clinical trial within the Department of Defense PCCTC (NCT01717053). Eligibility included 2+ intermediate or 1 high NCCN risk factors and no metastatic disease. Men received 6 months of ADT concurrently with 1000mg AA/5mg P daily and 78 Gy RT to prostate/SV. Primary endpoint was PSA < 0.1 ng/ml at 1 year. Secondary objectives included BPFS, PSA nadir, testosterone recovery, toxicity, and patient-reported QOL. Results: We enrolled 37 men (82% white, 18% black) with intermediate to high risk localized PC; 33 completed course of treatment (4 patients halted early for personal preference (N = 2), planned prostatectomy, or renal artery stenosis). Median age was 66 years; 46% Gleason 8-10, 40% Gleason 4+3 = 7, 62% T1c. Median follow-up is 23 months. Regimen was well tolerated with 12 (32%) G3 toxicities (10 hypertension, 2 hyperglycemia, 1 hypokalemia); no G4-5 or unexpected toxicities were observed. At 12 months from enrollment, PSA remained at undetectable levels in 52% of men. Testosterone recovery to normal lab value occurred in 12 (62%) at 12 months. In those patients, 20 (95%) and 21 (100%) remained with PSA under 0.5 and 1.0 ng/ml, respectively. No patient has failed by Phoenix definition to date. 1 year EPIC QOL had median summary scores above 90 for incontinence, urinary, bowel, hormonal, and satisfaction. Sexual summary score fell from median of 46 at baseline to 26 at 1 year. Conclusions: In men with high risk intermediate or limited high risk PC, utilizing short-term ADT/AAP with definitive RT shows 1) high rate of testosterone recovery and good quality of life and 2) excellent PSA control at 1 and 2 years. Clinical trial information: NCT01717053.

682 Background: Little data exists on objective response rates (ORR), progression-free survival (PFS), and safety of 2L VEGFR-TKI after 1L ICI therapy in pts with mCCRCC. Methods: This is a retrospective study of pts with mCCRCC who received 2L VEGFR-TKI after progressive disease (PD) with 1L ICI. Tumor response was assessed by a blinded radiologist using RECIST 1.1. Descriptive statistics, the Fisher’s test, and Kaplan-Meier method were used. Results: We report on 43 pts who were treated at MDACC from 2015 till present. Median age at mCCRCC diagnosis was 59 years (range: 43-72). 33 pts had lung mets, 20 had LN mets, 14 had bone mets, and 3 had liver mets. As 1L therapy, 20 pts received nivolumab + ipilimumab, 14 received nivolumab + bevacizumab, and 9 received nivolumab. Median time on ICI therapy was 29.4 weeks. All pts had resolution of Grade 3/4 AEs from ICI and PD before initiation of VEGFR-TKI. One patient (2%) had CR, 17 pts (40%) had PR, and 25 pts (58%) had SD, leading to 100% disease control rate (DCR) as best response to 2L VEGFR-TKI. Median PFS was 10.0 months (95% CI: 7.4, NA). Estimated 1-yr overall survival (OS) was 87.5% (95% CI: 74.6 - 100). Seven pts (16%) discontinued VEGFR-TKI therapy because of AEs: Gr 3 transaminitis (3 on pazopanib), Gr 3 hand-foot skin reaction (1 on axitinib), impaired wound healing (1 on axitinib), and Gr 3 pancreatitis (1 on pazopanib, 1 on axitinib). Conclusions: In this retrospective study, we observed a 42% ORR, a 10-month median PFS, and a 100% DCR in pts with mCCRCC who received VEGFR-TKI after PD with ICI. These results inform the design of trials with 2L VEGFR-TKI after failure of ICI therapy. [Table: see text]

Studies of combination ADT and RT have shown benefit to long-term ADT in high-risk prostate cancer; however years of ADT carry significant morbidity. Recent randomized trials show improved survival in advanced prostate cancer by adding abiraterone acetate/prednisone (AAP) to ADT to provide a potent complete androgen blockade. We hypothesized that such a combination of androgen inhibition in men with aggressive but localized PC would synergize with definitive RT to provide biochemical control with a 6 month ADT course. This prospective phase 2 single arm clinical trial was performed at two sites within the Department of Defense PCCTC (NCT01717053). Eligibility included 2+ intermediate or 1 high NCCN risk factors and no metastatic disease. Men received 6 months of ADT concurrently with 1000mg abiraterone acetate and 5mg prednisone daily. RT to prostate/SV (78 Gy) was initiated at week 9. Primary endpoint was PSA <0.1 ng/ml at 1 year. Secondary objectives included biochemical PFS, PSA nadir, testosterone recovery, toxicity, and patient-reported QOL. A total of 37 men were enrolled, with 33 completing the full course of treatment (4 patients halted within first 4 weeks of therapy for personal preference (N=3) or renal artery stenosis). Median age was 65 years. 66% were cT1, 46% Gleason 8-10, 42% Gleason 4+3=7, and baseline PSA was 7.3ng/ml (range 1.4-18.5). Median follow-up is 24 months. Acute grade 3 toxicities (32%) were all related to abiraterone acetate (hypertension, lab abnormalities) and managed with medication hiatus (32%) and medical therapy; no G4+ or unexpected toxicities were observed. 78% of patients nadired to =0.1. No patient has failed by Phoenix definition to date. The majority of patients have recovered testosterone to normal lab value (1 year 62%, 2 year 81%). PSA remained at undetectable levels in 52% of men at 1 year and 38% at 2 years. In patients with testosterone recovery, 20 (95%) and 21 (100%) remained with 1 year PSA under 0.5 and 1.0 ng/ml, respectively. Of the 17 men with testosterone recovery at 2 years, 82% and 100% had PSA < 0.5 and 1.0 ng/ml. Patients had excellent quality of life scores for urinary, bowel, hormonal, and overall satisfaction (EPIC summary >90). In men with aggressive prostate cancer (2+ intermediate or 1 high risk factors), utilizing short-term ADT/abiraterone acetate/prednisone with definitive RT promotes a high rate of testosterone recovery with excellent PSA control at 1 and 2 years.

Background: ICI therapy is an established strategy in mRCC after progressive disease on VEGFR-TKI. Little data exists in the reverse sequence on response rate, progression-free survival (PFS), safety, and tolerability of TKI after 1L ICI therapy. Methods: This is a retrospective analysis of patients with mRCC treated from 2015 to present with 1L ICI, followed by 2L TKI. Response assessment was provided by a blinded radiologist using RECIST 1.1. Descriptive statistics, Fisher's test and Wilcoxon rank sum test were used. Results: We report on 27 clear-cell mRCC patients with follow-up of at least 8 weeks on TKI post 1L ICI. Median age at diagnosis was 58 years. 78% of patients had lung, 37% bone, 37% lymph node, and 7% liver metastasis. As 1L therapy, 7 patients received nivolumab, 17 received nivolumab + ipilimumab, and 3 received nivolumab + bevacizumab. All 27 patients had resolution of Grade 3/4 toxicities from ICI and progressive disease at the time of TKI initiation. Median time from discontinuation of ICI to initiation of TKI was 4.1 weeks (range 0-23.3 weeks). 11 patients (41%) had PR (8 of whom had ≥40% tumor reduction), and 16 (59%) had SD as best response to TKI. Median PFS was 10.0 months (95% CI 6.8, not applicable). 9 patients discontinued 2L TKI after a median of 26.3 weeks (range 4.6-44 weeks), 8 patients because of PD and 1 because of toxicity. 2 patients developed Grade 3 transaminitis and 3 patients Grade 3 hand-foot skin reaction. Age, sex, IMDC risk score, nephrectomy status, and TKI agent did not predict PR or SD.Table891PVariableTotal n (%)PR (n)SD (n)P valueMale19 (70)7120.68Female8 (30)44Localized at presentation9 (33)720.01Metastatic at Presentation18 (67)414IMDC good risk4 (15)310.47IMDC intermediate risk19 (70)712IMDC poor risk4 (15)13Nephrectomy21 (78)10110.35Primary in-situ6 (22)15Pazopanib8 (30)440.37Axitinib12 (44)39Cabozantinib7 (26)43 Open table in a new tab Conclusions: In this small retrospective study, we observed a high response rate (41%), median PFS 10 months, and manageable toxicity in patients with mRCC treated with TKI after ICI. No patients had outright PD on 2L TKI after ICI. Legal entity responsible for the study: MD Anderson Cancer Center Dept of Genitourinary Medical Oncology Funding: None Disclosure: G. Jianjun: Travel/Honoraria & Consulting: AstraZeneca. M.T. Campbell: Consulting/advisory role: AstraZeneca, Eisai. A. Zurita Saavedra: Research support: Pfizer. E. Jonasch: Research funding: Exelixis, Pfizer, Novartis Honoraria: Bristol-Myers Squib, Eisai, Exelixis, Novartis, Pfizer. P. Sharma: Stock: Jounce, Kite Pharma, Evelo, Neon & Constellation Consulting/advisory role: Bristol-Myers Squib, GSK, AstraZeneca, Amgen, Constellation, Jounce, Kite Pharma, Evelo, Neon & EMC Serono Patents (spouse): Jounce, Merck, Bristol-Myers Squib. N. Tannir: Travel/Honoraria & Consulting: Bristol-Myers Squib, Exelixis, Nektar, Novartis, Pfizer, Argos, Calithera Research: Bristol-Myers Squib, Exelixis, Epizyme, Novartis, Miranti. All other authors have declared no conflicts of interest.

Src and Vegf, targeted by D and S respectively, have been implicated in CRPC progression. Moreover the unmet need for predictive markers has become pressing. We conducted a study to determine if the addition of either agents could prolong time to progression (TTP) on AA and to test a prespecified molecular signature. This is a phase II study of AA in bone mCRPC patients randomized upon progression to combination with D or S, comparing the two treatment strategy. Endpoints include PFS, safety, survival, assessment of aprespecified signaling signature in pts with benefit vs primary resistance to AA (progression ≤ 4 months). Pts had pretreatment bone biopsy. Tumor Markers included, Androgen Receptor-N terminal (AR-N), AR-C terminal (AR-C), CYP17, Ki67, GR, ERG, Ki67 (%), pSrc, vegf, DNA repair genes by IHC and steroids by LCMS. Study (03/2011-02/2015) accrued 179 bone mCRPC pts and median follow up 27ms (9-57). Medians: Age, 68 ys (range 45-87), PS ECOG 1 (range 0-1) baseline PSA 20.6 (range 0.5-1655). 27 (16%) pts have visceral mets, 40 (22%) prior chemo. Diagnostic Gleason Score was ≥8 in 71%. Upon progression 128/179 pts were randomized: 64 to D (AD) and 64 to S (AS). Of these 61 crossed over and 30 are still on treatment. Fifty Five pts had primary resistance to AA. Thirteen pts discontinued due to adverse events (2 AA, 6AD and 5 AS) Median (CI 95%) TTP and Overall Survival (OS) for the cohort are 12.85 (11.08, 14.98 ) and 26.26 ( 23.21, 31.93 ) ms respectively. There is no difference for drug sequence. On multivariate analysis, primary resistance to AA (p <0.0001) and prior chemo (0.0004) associate with poor prognosis. 52/55 primary resistant pts die within a year. Fifty pts had baseline tumor infiltrated biopsies. Testing of prespecified signature: AR-N terminal and CYP17 overexpression coupled with a ratio of AR-C terminal / AR-N terminal expression ≥ 0.8 has a predictive value (p value <0.0001) for AA benefit. ERG (p 0.05) is associated with benefit and ARV7 presence (p 0.04) with primary resistance to AA D or S do not improve AA efficacy or overcome primary resistance. Validation of prespecified androgen signaling signature predictive of AA response is planned in a multi-institutional study.

5020 Background: Clinically defined AVPC is associated with platinum-sensitivity and a molecular profile of combined (≥2) alterations in PTEN, Tp53 and/or RB1 (Aparicio et al. Clin Cancer Res 2013;19 and Clin Cancer Res 2015;1). Based on these findings we conducted a validation study of cabazitaxel (CAB) plus or minus carboplatin (CARB) in men with CRPC stratified for the presence or absence of AVPC criteria (Corn et al. ASCO 2015; Funding: Sanofi). Methods: Clinical data were updated and available specimens were assembled for analysis. Immunohistochemistry was performed for AR, Ki67, RB1 and PTEN. DNA was extracted from remaining tumor cells and subject to next-generation sequencing to include Tp53. Results: At a median follow up of 21.6 months (mo), median progression free survival (mPFS) is 4.6 mo (95%CI 3.5-5.8) with CAB vs 7.4 mo (95% CI 5.6-8.3) with CAB/CARB (p = 0.004). In men meeting AVPC criteria, mPFS is 3.8 mo (95% CI 2.8-5.7) with CAB vs 5.6 mo (95% CI 4.4-8.0) with CAB/CARB (p = 0.012). After 86 deaths, median overall survival (mOS) is 17.4 mo (95% CI 12.6-25.9) with CAB vs 19.2 (95% CI 17.0-29.3) with CAB/CARB (p=0.489). 53% (42/79) of CAB patients and 23% (19/81) of CAB/CARB patients received additional salvage platinum-based therapy after progression. We identified 124 samples from 92 patients, of which 95 contained sufficient tumor for analysis. Of 54 analyzed to date, 38 belonged to men receiving CAB/CARB (30 AVPC) and 16 to men receiving CAB (10 AVPC). ≤10% of cells stained for AR, RB1 and PTEN in 6 (15%), 17 (42.5%) and 5 (12.5%) of 40 AVPC tumors and 0 (0%), 5 (35.7%) and 2 (14.3%) of non-AVPC tumors respectively. Tp53 sequencing is ongoing. Conclusions: Adding CARB to CAB is safe and significantly improves mPFS and response rates in men with mCRPC with a trend towards improved mOS. Men meeting AVPC clinical criteria benefit most from the CAB/CARB combination. We observed a high frequency of tumor suppressor defects in the characterized samples. Further analysis of the rate of their combination, association with AVPC criteria and benefit from CARB will be informative. Clinical trial information: NCT01505868.

BACKGROUND:Pazopanib is associated with increased progression-free survival (PFS) in clear-cell renal cell carcinoma (RCC) and has become a standard of care in this disease. The drug is used in metastatic non-clear-cell RCC, but data on outcomes in this setting are limited.PATIENTS AND METHODS:We conducted a retrospective data analysis of records of consecutive metastatic non-clear-cell RCC patients who received pazopanib in front-line and salvage settings between November 2009 and November 2012. Tumor response rate was assessed by a blinded radiologist using Response Evaluation Criteria in Solid Tumors version 1.1. PFS and overall survival (OS) times were estimated using Kaplan-Meier methods.RESULTS:Twenty-nine patients were identified with non-clear-cell metastatic RCC, 9 received pazopanib in the front-line setting, 20 in the salvage setting after progression of disease with other targeted therapies. Seven patients (24%) had papillary RCC, 4 (14%) had chromophobe, 5 (17%) had unclassified histopathology, and 13 (45%) had other subtypes including collecting duct, translocation Xp11.2, and various subtypes with sarcomatoid differentiation. All patients discontinued pazopanib before analysis. Median PFS was 8.1 months (95% CI, 5.7-NA [not available]) in the front-line group, and 4 months (95% CI, 2.1-9.9) in the salvage group. Median OS was 31 months (95% CI, 9.2-NA) in the front-line group, and 13.6 months (95% CI, 6.4-NA) in the salvage group.CONCLUSION:Pazopanib showed efficacy in patients with metastatic non-clear-cell RCC in the front-line and salvage settings. Toxicity was mild to moderate and manageable. Further studies are needed to evaluate pazopanib's role in non-clear-cell RCC in terms of efficacy and safety.

Objective To characterise the incidence, onset, management, predictors, and clinical impact of mammalian target of rapamycin (mTOR) inhibitor‐associated non‐infectious pneumonitis (NIP) on patients with metastatic renal cell carcinoma (mRCC). Patients and Methods Retrospective review of 310 patients with mRCC who received temsirolimus and/or everolimus between June 2007 and October 2010. Clinical correlations were made with serial radiological imaging. Fisher's exact, Wilcoxon rank‐sum, and logistic regression analyses were used to evaluate the association of NIP with demographic or clinical factors. Log‐rank and Cox proportional hazards regression analyses were used for the time‐to‐event analysis. Results NIP occurred in 6% of temsirolimus‐treated and 23% of everolimus‐treated patients. Symptoms included cough, dyspnoea, and fever (median of two and three symptoms per patient, respectively). The median National Cancer Institute Common Toxicity Criteria for Adverse Events pneumonitis grade was 2 for both groups. Older age and everolimus treatment were predictive of NIP. Patients who developed NIP had a significantly longer time on treatment (median 4.1 vs 2 months) and overall survival (OS) (median 15.4 vs 7.4 months). NIP was a predictor of improved OS by multivariate analysis. Conclusions There was an increased incidence of NIP in everolimus‐treated patients. Improved OS in patients who developed NIP is an intriguing finding and should be further investigated. Given the incidence, morbidity, and outcomes seen in patients on everolimus who develop NIP, management should include proactive monitoring and treatment of NIP with the goal of preserving mTOR inhibitor therapy.

4513 Background: Cabozantinib (cabo) inhibits MET and VEGFR2. High rates of bone scan resolution, pain relief and overall disease control, independent of PSA changes, were previously reported in a phase II study in mCRPC patients (pts). This is a NRE cohort in docetaxel (D)-pretreated pts with a novel primary endpoint of bone scan response based on computer-aided quantitative assessment of bone scan lesion area (BSLA) and a double-reader, independent, blinded review (Nucl Med Commun, in press). Methods: D-pretreated (≥225 mg/m2) CRPC pts with bone metastasis were required to have progressed in soft-tissue or bone within 6 months of last dose of D. Pts received 100 mg cabo qd. Tumor response was assessed q6 wks. Bone scan response (BSR) was defined by a ≥30% decline in BSLA. Pain intensity (worst pain over the past 24 hrs; BPI scale 0-10) and interference with sleep and daily activity were prospectively assessed using an IVR system. Analgesic use was collected by diary. Bone turnover markers and CTCs were assessed. Results: 93 D-pretreated pts were enrolled (89 evaluable with ≥6 wks f/u). Median age was 67, 46% received cabazitaxel and/or abiraterone, 32% had visceral disease, 51% had fatigue, and 18% had anemia. 44% had worst pain ≥4 of which 95% were taking narcotics. Median CTC count was 49 and 80% had ≥5. Median f/u was 125 days (range, 23-305). Of 85 pts evaluable for BSR, 51 (60%) had a PR, 24 (28%) SD, 5 (6%) PD and 5 (6%) d/c’d prior to f/u scan. 21/30 pts (70%) had reduction of measurable disease.16/33 pts (49%) with BPI ≥4 and ≥12 wks f/u had pain reduction durable for ≥6 wks; 46% had decreased narcotic use, including 27% who discontinued use. Sleep and daily activity were improved in pts with pain relief. Among pts with elevated serum levels, 74%, 67% and 47% had declines on treatment of ≥30% in CTx, NTx and bALP, respectively. In 59 pts with CTCs ≥5, 92% had a decrease of ≥30% and 39% converted to <5 CTCs at weeks 6 or 12. 12% discontinued cabo due to AEs. Most common Gr 3/4 AEs were fatigue (19%), nausea (10%) and anemia (10%). Conclusions: Cabo treatment resulted in high rates of bone scan response, durable pain relief, and reductions in bone turnover markers and CTCs in D-pre-treated CRPC pts with bone metastases.

332 Background: Pneumonitis is a known adverse effect (AE) of mammalian target of rapamycin-inhibitors, with a literature reported incidence for everolimus ranging from 4 to 45%. The goal of this review was to characterize the incidence, timing, management, and outcomes related to everolimus-associated pneumonitis (EAP). Methods: Retrospective review of 86 mRCC patients (pts) with complete, evaluable records, given everolimus (E) between 4/2009 and 3/2010. We assessed baseline patient (pt) characteristics, previous therapies, time on E therapy, pt symptoms, physician management of AE, NCI-CTC pneumonitis grading, and survival outcomes. Radiologic CT indicated ground glass, inflammatory, and/or parenchymal opacities. Results: (See table.) EAP occurred in 28% of pts on E therapy, confirmed radiologically. 8% of EAP patients reported no symptoms. In EAP pts, 58% reported cough, 75% dyspnea and/or SOB, 17% fever, 71% fatigue. The median number of symptoms/patient was 3. 46% of pts received steroids (median 21 days (3-120)), 38% received antibiotics, 25% received pulmonary consultation, and 8% required oxygen. In pts who developed EAP, providers discontinued E in 75%, held and dose reduced E in 8%, and continued E in 17%. The median NCI-CTC pneumonitis grade was 2 (1-3); there were no treatment-related deaths. The median time to EAP onset was 67 days (8-442). There was no statistically significant difference in outcomes between EAP pts and non-EAP pts. Conclusions: EAP occurs often in mRCC pts treated with E. It is an important AE that can negatively affect pt symptoms, but did not adversely impact pt outcomes in our single-center experience. [Table: see text] [Table: see text]

351 Background: Pazopanib, a multityrosine kinase inhibitor (TKI), prolongs PFS compared to placebo in treatment-naive and cytokine refractory metastatic renal cell carcinoma (mRCC). No data has been reported about pazopanib salvage therapy after treatment with other targeted agents. METHODS We retrospectively reviewed the records of 88 consecutive patients (pts) with mRCC (median age 62.7, M:F 63/25, 84% clear cell) who received salvage pazopanib between 11/09-8/10. All pts failed previous treatment with one or more targeted agents (median number of prior targeted agents was 2, range 1-5; median time on previous treatments was 632 days). 78% of pts had progressed on sunitinib, 40% on sorafenib, 20% on temsirolimus, 51% on everolimus, and 26% on bevacizumab. 26% received previous chemotherapy and 16% received previous cytokines in addition to targeted therapies. 58% failed both TKI/VEGF inhibitors and mTOR inhibitors. 57% had intermediate-risk disease and 43% had poor-risk disease by MSKCC criteria. All pts had follow-up at least every 3 months after initiating pazopanib. RESULTS Median time to last follow-up was 114 days (range 30-278 days). 42% continued pazopanib at last follow- up. 25% had partial response (PR) by treating physician assessment. 50% failing 1-2 previous targeted therapies remained on pazopanib at last follow-up, compared to 27% of those failing more than 2 targeted therapies (p=0.04). 56% of pts with intermediate-risk disease by MSKCC criteria continued pazopanib at follow-up compared to 27% with poor-risk disease (p=0.002). 42% of those failing 1 prior targeted therapy achieved PR compared to 18% failing >1 prior targeted therapy (p=0.02). 35% discontinued pazopanib due to progressive disease (PD) (median time to PD 71 days, range 36-198 days), 10% discontinued due to adverse drug events, and 10% died of PD on treatment. There were no treatment related deaths. CONCLUSIONS In this retrospective study, pazopanib demonstrated clinically relevant activity in mRCC following PD with other targeted therapies. Mature survival data will be provided with final presentation. No significant financial relationships to disclose.

4516 Background: Cabozantinib (Cabo) is an inhibitor of MET and VEGFR2. MET signaling promotes tumor growth, invasion and metastasis. Methods: mCRPC patients (pts) with progressive measurable disease (mRECIST) received Cabo at 100 mg qd PO over a 12 week (wk) lead-in stage. Response was assessed q6 wks. Treatment ≥ wk 12 was based on response: pts with PR continued open-label Cabo, pts with SD were randomized to Cabo vs placebo, and pts with PD discontinued. Primary endpoint was objective response rate (ORR) per mRECIST in the lead-in stage. Up to 200 pts could be enrolled to target 70 randomizations. Bone scans (b-scans) were independently reviewed. Results: Accrual was halted at 168 pts based on an observed high rate of clinical activity. 100 pts are currently evaluable for the lead-in stage; median age 68, 47% with visceral disease, 78% with bone metastasis, and 47% docetaxel (D) pretreated. Median f/u was 4 months (range, 1-15); median PFS not yet reached. Most common related Grade 3/4 AEs were fatigue (11%), HTN (7%), and hand-foot syndrome (5%); no related Grade 5 AEs reported. Dose reductions for AEs occurred in 51% of pts, and discontinuations in 10%. Bone effects: 86% (56/65 pts evaluable by b-scan) had complete or partial resolution of lesions on b-scan as early as wk 6. Eight pts (12%) had SD and 1 pt (2%) had PD. In 28 pts receiving narcotics for bone pain, 64% had improved pain and 46% decreased or halted narcotics, per investigator. Median maximum rise in hemoglobin in anemic pts (Hb < 11 g/dL) was 2.2 g/dL (range, 0.6-3.5). Osteoclast and osteoblast effects were observed: 55% had declines of ≥50% in plasma C-Telopeptide; 56% of pts with elevated tALP had declines of ≥50%. Soft tissue effects: Objective tumor shrinkage occurred in 84% of pts. ORR at wk 12 was 5%; 3 additional PRs await confirmation. PSA changes were independent of clinical activity. Overall, wk 12 disease control rate (PR+SD) was 71%. Randomization was halted and pts unblinded due to high rates of b-scan resolution and pain relief. Conclusions: Cabo showed clinical activity regardless of prior D in mCPRC pts, particularly in pts with bone disease, as reflected by high rates of b-scan resolution and pain relief, in addition to improvements in Hb and tumor regression.

4649 Background: Bone is the most common site of progression in prostate cancer and the development of bone-targeted therapy may benefit such patients The aim of this phase II randomized controlled trial was to determine the progression-free survival of hormonal ablative therapy, combined with zoledronic acid and doxorubicin, with or without strontium-89, for patients with androgen-dependent prostate cancer and bone metastases. Methods: A total of 80 patients were enrolled on this study; 79 patients were randomized: 40 to the arm without and 39 to the arm with Sr-89 treatment. Eligible patients also received after randomization LHRH agonists of choice AND doxorubicin (20 mg/m2 on days 1, 8, and 15 q 4 weeks x2 cycles) AND zoledronic acid (4 mg IV q 4 weeks x6). Kaplan-Meier methodology was used to evaluate the primary outcome, which was progression-free survival (PFS). In addition, multivariate Cox proportional hazards regression was used to evaluate the effects of strontium-89 after controlling for the number of bone metastases. Results: The median follow-up time for the 44 of 79 randomized patients that were alive at the last follow-up, was 44.5 (range: 0.07–72.8) months. There was no statistical significance for PFS between the Sr-89 arm and no Sr-89 arm. Thirty-four of 72 (45.2%) patients died, 17 of 35 (48.6%) patients on the no Sr-89 treatment arm and 17 of 37 (45.9%) patients in the Sr-89. Median time to death was 46.5 months (95% CI: 32.3, not attained) in the no Sr-89 arm compared to 49.5 months (95% CI: 35.9, not attained). However, the Kaplan-Meier curves of overall survival for the evaluable population did not show any statistical significance. Unplanned subgroup analysis suggests that bone-targeted therapy may be more efficacious for patients with greater extent of bone involvement (ie, >6 vs <6 bone metastases on the bone scan). Conclusions: These preliminary data suggest that bone-targeted therapy using one dose of Sr-89 combined with chemohormonal ablative therapy did not improve the PFS of patients with androgen-dependent prostate cancer with bone metastases. Whether this bone-targeted therapy provided a favorable outcome for patients with greater tumor burden in the bone requires further investigation.