Search

PURPOSE The first interim analysis of the phase III, randomized, placebo-controlled TITAN study showed that apalutamide significantly improved overall survival (OS) and radiographic progression-free survival in patients with metastatic castration-sensitive prostate cancer (mCSPC) receiving ongoing androgen deprivation therapy (ADT). Herein, we report final efficacy and safety results after unblinding and placebo-to-apalutamide crossover. METHODS Patients with mCSPC (N = 1,052) were randomly assigned 1:1 to receive apalutamide (240 mg QD) or placebo plus ADT. After unblinding in January 2019, placebo-treated patients were allowed to receive apalutamide. Efficacy end points were updated using the Kaplan-Meier method and Cox proportional-hazards model without formal statistical retesting and adjustment for multiplicity. Change from baseline in Functional Assessment of Cancer Therapy-Prostate total score was assessed. RESULTS With a median follow-up of 44.0 months, 405 OS events had occurred and 208 placebo-treated patients (39.5%) had crossed over to apalutamide. The median treatment duration was 39.3 (apalutamide), 20.2 (placebo), and 15.4 months (crossover). Compared with placebo, apalutamide plus ADT significantly reduced the risk of death by 35% (median OS not reached v 52.2 months; hazard ratio, 0.65; 95% CI, 0.53 to 0.79; P < .0001) and by 48% after adjustment for crossover (hazard ratio, 0.52; 95% CI, 0.42 to 0.64; P < .0001). Apalutamide plus ADT delayed second progression-free survival and castration resistance ( P < .0001 for both). Health-related quality of life, per total Functional Assessment of Cancer Therapy-Prostate, in both groups was maintained through the study. Safety was consistent with previous reports. CONCLUSION The final analysis of TITAN confirmed that, despite crossover, apalutamide plus ADT improved OS, delayed castration resistance, maintained health-related quality of life, and had a consistent safety profile in a broad population of patients with mCSPC.

5066 Background: TITAN, a phase 3 PBO-controlled study in patients (pts) with mCSPC, showed APA + ADT improved radiographic progression-free survival and OS vs PBO + ADT. In this exploratory analysis, we report the relationships between biomarkers and OS in TITAN. Methods: Circulating tumor (ct)DNA and genomic aberrations in 17 PC-related genes, including androgen receptor (AR), were assessed at baseline (BL; 114 pts) and end of study treatment (EOST; 129 pts) using next-generation sequencing. ctDNA was assessed qualitatively; genomic aberrations were assessed within ctDNA-positive samples as inactivation (heterozygous/homozygous deletion or single nucleotide variant [SNV]) or activation (amplification or SNV). Associations of detected ctDNA/aberrations at BL or EOST with OS, and biomarkers at EOST with OS on subsequent therapies, were evaluated using univariate or multivariate analyses and Cox proportional hazards model; results were stratified by treatment arm. Results: Among pts from both treatment groups, 36% had detectable ctDNA, of which 27% had any genomic AR aberration and 24% had AR gene amplification at BL; prevalence of these biomarkers increased significantly from BL to EOST (Table). The most prevalent non-AR aberrations at BL ( TP53, homologous recombination repair [HRR] pathway, PTEN, RB1, and PIK3CA genes) increased at EOST but not significantly (Table). Among assessed aberrations, presence of ctDNA or any AR genomic aberrations at BL (HR, 1.9 or 6.7; all p < 0.05) and any AR genomic aberrations or PI3K pathway activation at EOST (1.7, p < 0.05 or 2.2, p < 0.001) was significantly associated with poor OS in multivariate analyses from both treatment groups. In univariate analyses of pts who received subsequent therapy (chemotherapy: 106; hormonal therapy: 161), worse OS was associated only with PIK3CA activation, PI3K pathway activation, or TP53 inactivation at EOST (3.7, p < 0.05; 2.4, p < 0.05; 3.0, p < 0.01, respectively) in chemo-treated pts. A small sample size in some biomarker subgroups limits interpretation. Conclusions: These hypothesis-generating data from TITAN show that presence of ctDNA or select AR and non-AR biomarkers at BL or EOST were associated with poor OS. The predictive value of these biomarkers for survival in mCSPC needs further confirmation. Clinical trial information: NCT02489318. [Table: see text]

89 Background: Addition of androgen receptor signaling inhibitors to ADT + DOC has been shown to improve clinical outcomes in pts with mCSPC. TITAN, a placebo-controlled phase 3 study, showed that APA + ADT improved overall survival (OS) and other clinical outcomes in mCSPC (Chi, J Clin Oncol 2021). This post hoc analysis of TITAN evaluated outcomes in pts who had received DOC prior to treatment with APA + ADT versus those who did not. Methods: In TITAN, 1052 pts were randomized 1:1 to APA (240 mg QD) or placebo added to ongoing ADT. We assessed radiographic progression-free survival (rPFS), OS, and time to prostate-specific antigen (PSA) progression in pts receiving DOC and ADT prior to adding APA vs those receiving only ADT plus APA. Outcomes by prior DOC were also assessed in pts with high- or low-volume disease at randomization (baseline [BL]) per adapted CHAARTED criteria, or those with matched BL characteristics. A Cox proportional hazards model was used to derive hazard ratios (HRs) and p values. rPFS was assessed using the first interim analysis cutoff (23 mo median follow-up); OS and time to PSA progression were assessed using the final analysis cutoff (44 mo median follow-up). Results: A total of 58/525 (11%) pts from the APA + ADT group had received DOC prior to randomization: 76% (n = 44) had high-volume disease, 62% (n = 36) had bone-only metastases, 16% (n = 9) had visceral metastases, and 59% (n = 34) had > 10 bone lesions. In the overall APA-treated population and in the subset of pts with high-volume disease, OS, rPFS, and time to PSA progression were similar in those who received prior DOC and those who did not (Table). Pts with low-volume disease also had similar results, although the number of pts was small. Clinical outcomes in pts with matched BL characteristics (including PSA and time from initial diagnosis to randomization, among others) were similar regardless of prior use of DOC (Table). The safety profile of APA was not substantially different between pts with or without prior DOC. Limitations of this analysis include lack of data on tumor volume and other disease characteristics at the initiation of prior DOC treatment; interpretation was based on small number of pts with prior DOC (only 11% of TITAN pts), most notably in the rPFS analysis. Conclusions: Prior use of DOC in pts with mCSPC did not further improve clinical benefits of APA + ADT in TITAN. Clinical trial information: NCT02489318. [Table: see text]

15 Background: PRESIDE (NCT02288247) evaluated the benefit of continued ENZA + androgen deprivation therapy (ADT) with DOC + PDN in men with mCRPC who progressed on ENZA + ADT. Methods: PRESIDE (Dec 2014–Apr 2020) enrolled chemotherapy-naïve men with mCRPC and disease progression while on a luteinizing hormone-releasing hormone (LHRH) agonist/antagonist (ADT) or after bilateral orchiectomy. Pts received open-label ENZA (160 mg) + ADT in Period 1 (P1). Those with a prostate-specific antigen (PSA) response of ≥50% change from baseline to week (wk) 13 and later progression were eligible for Period 2 (P2). P2 pts received DOC (75 mg/m2), PDN (10 mg), and ADT, and were randomized to ENZA (160 mg) or PBO. The primary endpoint was progression-free survival (PFS) in P2 (from randomization to radiographic/clinical progression or death). Secondary endpoints included time to PSA progression (TTPP) [≥25% increase; absolute increase ≥2 ng/mL] and PSA response in P2. Hazard ratios (HRs) were from a Cox proportional hazards model with covariates for treatment and P1 progression. Adverse events (AEs) were recorded to assess safety. Results: 687 pts received ENZA in P1; 273 pts were randomized and 271 were treated in P2. Baseline demographics and characteristics were balanced between P2 arms. Median ENZA exposure was 62.6 wks in P1 and 36.1 and 30.1 wks in P2 with ENZA and PBO, respectively. At P2 data cut-off (Apr 30, 2020), 269 (99.3%) pts had discontinued therapy; 93 pts in each arm (ENZA, 74.4%; PBO, 75.6%) had progression. PFS was significantly improved with ENZA (HR 0.72; 95% confidence interval [CI] 0.53, 0.96; p = 0.027), with a higher median PFS with ENZA (9.53 months; 95% CI 8.25, 10.87) than with PBO (8.28 months; 95% CI 6.28, 8.71). ENZA also significantly delayed TTPP (8.44 vs. 6.24 months with PBO; HR 0.58; 95% CI 0.41, 0.82; p = 0.002) and improved PSA response at any time (ENZA, n = 76 [55.9%]; PBO, n = 50 [37.0%]). There were 46 (6.7%) deaths in P1 and 20 (ENZA, n = 13 [9.6%]; PBO, n = 7 [5.2%]) in P2. In P2, 264 (97.4%) pts had a treatment-emergent AE (TEAE) [ENZA, n = 133 (97.8%); PBO, n = 131 (97.0%)]. Grade 3/4 TEAEs were reported by 84 (61.8%) pts on ENZA and 84 (62.2%) on PBO, and 12 (8.8%) and 9 (6.7%) pts, respectively, had TEAEs leading to discontinuation. Neutropenia (ENZA, 16.9%; PBO, 20.7%) was the most common grade 4 TEAE. Drug-related TEAEs (ENZA, 46.3%; PBO, 41.5%) were similar, and 90.4% of pts in each arm reported DOC-related TEAEs. Conclusions: Continued ENZA therapy in men with mCRPC who progressed on ENZA + ADT and received post-progression DOC + PDN significantly improved PFS compared to PBO. Treatment was well tolerated and ENZA AEs were consistent with its known safety profile. Clinical trial information: NCT02288247.

5068 Background: The phase 3 TITAN study evaluated APA vs PBO in pts with mCSPC receiving ADT. At primary analysis with 22.7 mo median follow-up, APA significantly improved overall survival (OS) and radiographic progression-free survival vs PBO (Chi NEJM 2019) while preserving HRQoL (Agarwal Lancet Oncol 2019). The study was unblinded; pts on PBO were allowed to cross over to APA. At final analysis with 44 mo median follow-up, APA significantly improved OS vs PBO, reducing risk of death by 35% despite crossover (Chi ASCO GU 2021). We evaluated HRQoL and treatment bother at final analysis. Methods: mCSPC pts (N = 1052) were randomized 1:1 to APA (240 mg QD; n = 525) or PBO (n = 527). All pts received ADT. Patient-reported outcomes were assessed using Brief Pain Inventory-Short Form (BPI-SF) and Functional Assessment of Cancer Therapy-Prostate (FACT-P). BPI was completed for 7d consecutively (Days -6 to 1 of each 28-d cycle [C]) through end of treatment (EOT). FACT-P was completed at baseline (BL), C2-C7, then every other C through EOT. Mean scores were reported by treatment group and over time. Time to deterioration on BPI and FACT-P scores was calculated by Kaplan-Meier methods and compared between groups by fitting proportional hazards regression models. Results: Of eligible pts per C, > 62% completed BPI through C32 and > 50% completed FACT-P through C31. Pts were relatively asymptomatic with good BL HRQoL: on 0-10 worst pain severity scale (BPI), median scores were 1.1 (APA) and 1.0 (PBO); on 0-156 HRQoL scale (FACT-P total; higher score = better HRQoL), median scores were 113.0 (APA) and 113.3 (PBO). Low BL BPI scores remained stable over time in both groups. On average, favorable BL FACT-P scores did not notably worsen over time in APA or PBO groups. There were no significant differences between groups in median time to deterioration in any BPI or FACT-P scores (Table). At each C at least 86% (APA) and 85% (PBO) of pts were either “not at all” or “a little bit” bothered by side effects. At BL, 76% (APA) and 72% (PBO) had favorable energy levels (reporting lack of energy “not at all” or “a little bit”). Energy levels remained stable or improved at each C for > 78% (APA) and > 71% (PBO) of pts. Conclusions: In the final analysis of TITAN, survival benefit with addition of APA to ADT was achieved without significant patient-reported side effect burden or reduced HRQoL compared with PBO in pts with mCSPC. Clinical trial information: NCT02489318. [Table: see text]

11 Background: TITAN evaluated APA or PBO added to ADT in pts with mCSPC. Pts with high- and low-volume disease, prior docetaxel, prior treatment for localized disease, and prior ADT (≤ 6 mos) were eligible. At the first interim analysis, with 22.7 mos median follow-up, APA significantly improved dual primary end points of overall survival (OS) (hazard ratio [HR] 0.67) and radiographic progression-free survival (rPFS) (HR 0.48) compared with PBO (Chi et al. NEJM. 2019). At that time, OS analysis was first planned interim while rPFS was final. TITAN was unblinded, allowing pts without progression who were still receiving PBO to cross over to APA. Herein, we report the final analysis of efficacy and safety results from TITAN. Methods: 1052 mCSPC pts were randomized 1:1 to receive APA (240 mg QD) or PBO plus ADT. Time-to-event end points were analyzed by Kaplan-Meier method and Cox proportional hazards model. A preplanned sensitivity analysis for OS, accounting for crossover using inverse probability censoring weighted (IPCW) log-rank test, was conducted. No formal statistical retesting was performed; nominal p values were reported without multiplicity adjustment. Change from baseline in Functional Assessment of Cancer Therapy-Prostate (FACT-P) total score was assessed using a mixed-effect repeated-measures model. Results: With 44 mos median follow-up, 405 OS events had occurred. After unblinding, 208 PBO pts (39.5%) crossed over to APA. Median treatment duration was 39.3 mos for the APA group, 20.2 mos for the entire PBO group, and 15.4 mos for the PBO→APA crossover group. OS was superior in the APA group compared with the PBO group despite crossover (Table). 48-mo survival rates were 65% (APA) vs 52% (PBO). Other end points also favored APA vs PBO (Table). Health-related quality of life (HRQoL), per total FACT-P, was maintained in the APA group through the study and was not different from the PBO group. Safety was consistent with previous reports. Conclusions: With close to 4 yrs of follow-up, the final analysis of TITAN demonstrated that in a broad population of pts with mCSPC, APA plus ADT provides an improvement in OS with a 35% reduction in risk of death, which increased to 48% reduction after adjusting for pts who crossed over from PBO to APA. In addition, there was consistent benefit with APA in other end points, including delaying castration resistance, and HRQoL continued to be maintained with an acceptable safety profile. Clinical trial information: NCT02489318. [Table: see text]

TPS190 Background: Cabozantinib inhibits multiple tyrosine kinases, including MET, VEGFR, RET, and TAM kinases (Tyro3, AXL, MER), involved in tumor growth and angiogenesis, and whose mutations and expression are associated with prostate cancer aggressiveness and poor prognosis. Targeting these kinases with cabozantinib may promote an immune permissive tumor environment and may enhance response to immune checkpoint inhibitors. In the ongoing phase 1b COSMIC-021 study of pts with solid tumors, cabozantinib plus the PD-L1 inhibitor atezolizumab, showed preliminary meaningful clinical activity in soft tissue disease and a tolerable safety profile for 44 pts with mCRPC (Agarwal et al., ASCO 2020; abstract 5564). We present the study design of a phase 3 trial of cabozantinib plus atezolizumab versus second NHT in pts with mCRPC. Methods: This randomized, open-label, controlled phase 3 study (NCT04446117) evaluates the efficacy and safety of cabozantinib plus atezolizumab versus second NHT (abiraterone or enzalutamide) in pts with mCRPC who previously received one NHT to treat metastatic castration-sensitive PC (mCSPC), non-metastatic CRPC (M0 CRPC), or mCRPC. Additional eligibility criteria include histologically or cytologically confirmed adenocarcinoma of the prostate, measurable visceral disease or measurable extrapelvic adenopathy per RECIST 1.1 by investigator, prostate specific antigen progression and/or soft-tissue disease progression, ECOG 0 or 1, and age ≥18 years. Key exclusion criteria include prior nonhormonal therapy for mCRPC and uncontrolled significant illness. Eligible pts (N = 580) are randomized 1:1 to receive cabozantinib (40 mg PO QD) + atezolizumab (1200 mg IV Q3W) vs abiraterone (1000 mg PO QD) + prednisone (5 mg PO BID) or enzalutamide (160 mg PO QD). Designated NHT will differ from previous NHT taken. Randomization is stratified by: liver metastasis (yes, no), prior docetaxel treatment for mCSPC (yes, no), and disease stage for which the first NHT was given (mCSPC, M0 CRPC, mCRPC). Treatment will continue until there is no longer clinical benefit as determined by the treating investigator, unacceptable toxicity, or consent withdrawal. The multiple primary endpoints are progression-free survival per RECIST 1.1 by blinded independent radiology committee (BIRC) and overall survival. Additional endpoints include objective response rate per RECIST 1.1 by BIRC, safety, correlation of biomarkers with outcomes, quality of life and pharmacokinetics. Patient enrollment is ongoing. Clinical trial information: NCT04446117.

80 Background: The poly(ADP-ribose) polymerase (PARP) inhibitor rucaparib was granted accelerated approval by the US Food and Drug Administration for patients with BRCA+ mCRPC based on results from the phase 2 TRITON2 study (NCT02952534). The TP53 tumor suppressor gene is among the most frequently mutated genes in human cancers, including mCRPC, and alterations in TP53, PTEN, and RB1 are associated with poor prognosis in patients with prostate cancer and other tumor types. We present data on co-occurring alterations in patients with BRCA+ mCRPC treated with rucaparib in TRITON2. Methods: Patients had progressed on 1–2 lines of androgen receptor-directed therapy and 1 taxane-based chemotherapy and were treated with rucaparib 600 mg BID. Tissue and/or cell-free DNA extracted from plasma samples were profiled comprehensively for genomic alterations using Foundation Medicine, Inc., next-generation sequencing assays. Objective response rate (ORR) was assessed per modified Response Evaluation Criteria in Solid Tumors and Prostate Cancer Working Group 3 criteria by independent radiologic review of patients with measurable disease. Prostate-specific antigen (PSA) response rate (≥50% decrease from baseline) was assessed in all patients. Results: Tissue and/or plasma samples were available for 114/115 patients with BRCA+ mCRPC (visit cutoff date: Dec. 23, 2019). Among patients with BRCA+ mCRPC who had samples available for comprehensive genomic profiling, 36.8% (42/114) had a co-occurring alteration in TP53. Deleterious alterations in PTEN were observed in 34.2% (39/114) of patients, 44% (17/39) of which were homozygous deletions of PTEN. RB1 loss was observed in 12.3% (14/114) of patients and was seen more frequently in patients with measurable disease (18.0%, 11/61) than in patients with non-measurable disease (5.7%, 3/53). Although patients with and without TP53 mutations had generally similar baseline demographics and disease characteristics, visceral disease was more prevalent in patients with TP53 mutations (54.8%; 23/42) than in those without them (29.2%; 21/72). Similar ORR and PSA response rates were seen in patients with BRCA+ mCRPC with or without TP53 mutation, with a non-significant trend towards lower response rates in patients with co-occurring TP53 alterations. Conclusions: Results from TRITON2 showed antitumor activity for rucaparib in patients with BRCA+ mCRPC associated with or without co-occurring alterations in TP53. Demographics and additional efficacy analyses in genomic subgroups with co-occurring alterations in TP53, PTEN, and RB1 will be reported. Clinical trial information: NCT02952534. [Table: see text]

181 Background: Biallelic loss of CDK12 has recently been identified as a novel subtype of prostate cancer (PCa). CDK12 altered PCa associates with elevated neoantigen burden and thus may be suitable for checkpoint inhibition. Up to now, data about CDK12 refer to its genetic alterations in PCa while its characterization on protein level and its association with tumor infiltrating T-cells are lacking. Methods: Immunohistochemistry (IHC) for CDK12 was performed on a PCa cohort including 74 benigns, 391 primary tumors from 222 patients, 63 locally advanced tumors, 92 lymph node (LN) metastases, and 56 distant metastases. CDK12 was categorized into negative, weak, moderate and high expression. Density of tumor associated T-cells per tumor area was assessed by IHC for CD3 and graduated into negative (50%). Results: CDK12 significantly increases during PCa progression showing highest levels in LN and distant metastases while benign samples harbor no or weak CDK12 expression (ANOVA p

492 Background: Atezo is an approved therapy for mUC based on the IMvigor210 (IM210) and IMvigor211 (IM211) phase 2 and 3 trials. The single-arm phase 3b SAUL study in a broader patient (pt) population showed consistent efficacy and safety. The value of complete response to immunotherapy is well recognized, but less is known about the benefit of partial response (PR) or stable disease (SD). We performed post hoc analyses of outcomes in pts with PR/SD in atezo trials. Methods: Pts with mUC received atezo 1200 mg q3w until disease progression (PD) or unacceptable toxicity. Baseline characteristics, efficacy and safety were analyzed in pts achieving PR or SD in IM210 cohort 2 (2nd line), IM211 (atezo and chemotherapy [CT] arms analyzed separately) and SAUL. Data cutoffs were 5/5/15, 3/13/17 and 9/16/18, respectively. Results: The analysis population included 229 PR and 583 SD pts (183 and 421, respectively, treated with atezo); ~40% of pts had 0 Bellmunt risk factors. Baseline characteristics were generally similar between trials. PD-L1 IC 2/3 was more common in PR than SD pts in atezo-treated (41% vs 26%) and CT-treated (37% vs 23%) pts. Median time to best response (PR or SD) was 2.1 mo across trials and treatments. In PR pts, disease control (DC) and overall survival (OS) durations were longer with atezo than CT (Table). Durations of DC and OS were shorter in SD than PR pts; OS was numerically longer with atezo than with CT in the SD population. Presence of more Bellmunt risk factors was associated with worse OS with CT but showed only a weak (SD pts) or no (PR pts) association in atezo-treated pts. Conclusions: Pts achieving a PR or SD have meaningful OS expectancy; pts with PR on atezo have sustained DC. Clinical trial information: NCT02108652 ; NCT02302807 ; NCT02928406 . [Table: see text]

488 Background: UTUC is rarer than bladder UC and typically responds poorly to standard chemotherapy. Analysis of 220 biomarker-evaluable atezo-treated patients (pts) in phase II/III trials suggested worse outcomes in UT vs lower tract UC [Galsky, ESMO 2018]. We explored clinical outcomes in pts with UTUC (renal pelvis or ureter) treated with atezo in the SAUL study. Methods: The single-arm SAUL study (NCT02928406) [Sternberg, Eur Urol 2019] enrolled a broader pt population, including pts with poor clinical characteristics and/or immune-mediated conditions, more representative of real-world practice than typically enrolled in randomized phase III immunotherapy trials. Pts with urinary tract carcinoma received atezo 1200 mg q3w until disease progression/unacceptable toxicity. Baseline characteristics, safety and efficacy were analyzed in subgroups of pts with UTUC (subdivided into renal pelvis or ureter UC) vs bladder UC. Results: Baseline characteristics in the 4 subgroups were generally similar, except for a numerically lower proportion of pts with 0 prior lines of therapy for metastatic disease in the UTUC vs bladder UC subgroup (30% vs 41%). Treatment exposure, safety and efficacy are shown below. Conclusions: These exploratory analyses of SAUL showed very similar efficacy and safety in UT vs bladder UC. This provides reassurance that atezo is active and has an acceptable safety profile in pts with UTUC, who are generally expected to have worse outcomes than bladder UC pts. Clinical trial information: NCT02928406 . [Table: see text]

5066 Background: Somatic DRD deficiency in 15-30% of mCRPC pts have been observed and inhibition of enzyme poly ADP ribose polymerase (PARP) could prove beneficial. We aimed to define DRD status using plasma from pts treated on AA and evaluate associations with prospectively-collected outcome measures. Methods: Plasma DNA samples (128 baseline [BL], 134 cycle 2 day 1 [C2D1], 46 progression [PROG]) from chemotherapy-naïve mCRPC pts in a phase 2 study (NCT01867710) evaluating AA+P/D were subjected to custom target-capture next-generation sequencing. DRD assay was optimized and validated to detect pathogenic point mutations, small insertions/deletions, and copy number alternations (DRD+) in 8 DRD genes: BRCA1, BRCA2, FANCA, ATM, CHEK2, HDAC2, BRIP1, and PALB2. Analysis for genomic aberrations was a secondary exploratory objective. Associations with overall survival (OS), progression-free survival (PFS), and radiographic PFS (rPFS) were assessed using Cox regression models and Kaplan Meier analyses. Results: 11.7% of BL and 17.4% of PROG were DRD+. Bi-allelic was observed in 73.3% of BL DRD+ samples. Shorter PFS was observed in BL DRD+ vs DRD- (5.3 vs 15.5 mo; HR: 2.32; 95%CI:1.39-4.28; P < 0.002). Median PFS for BL DRD biallelic + vs DRD biallelic- was 5.1 vs 15.4 mo (HR: 2.49; 95% CI: 1.23-4.38; P < 0.0095). For multivariate analysis using DRD+, ALP, and LDH as covariates, DRD+ (HR: 2.1; 95% CI: 1.18-3.75; P < 0.012) and high ALP (HR: 1.66; 95% CI: 1.08-2.56; P < 0.021) were strongly associated with worse PFS. Median OS for BL DRD+ vs DRD- was 28.8 vs 41.3 mo (HR: 1.67; 95%CI:0.88-3.18; P = 0.116). Median rPFS for BL DRD+ vs DRD- was 16.2 vs 20.9 mo (HR: 1.64; 95%CI:0.83-3.21; P = 0.152). Of 39 Pts with BL, C2D1 and PROG samples, 3 were DRD+ (7.7%) at all 3 timepoints, 3 (7.7%) only at BL, 3 (7.7%) only at PROG (bi-allelic), 2 (5.1%) had extra deletion at PROG. Conclusions: Patients with mCRPC harboring DRD+ have worse outcomes with AA and represent a population with an unmet medical need.

5006 Background: TITAN was designed to determine whether APA, a selective next-generation androgen receptor inhibitor, plus ADT improves radiographic progression-free survival (rPFS) and overall survival (OS) compared with PBO plus ADT in pts with mCSPC. Methods: In this randomized, double-blind phase 3 study, pts with mCSPC regardless of extent of disease were randomized (1:1) to APA (240 mg/d) or PBO, added to ADT, in 28-day cycles. Pts with prior treatment (tx) for localized disease or prior docetaxel for mCSPC were allowed. All pts received continuous ADT. Dual primary end points were rPFS and OS. Secondary end points were time to a) initiation of cytotoxic chemotherapy, b) pain progression, c) chronic opioid use, d) skeletal-related event. Time-to-event end points were estimated by Kaplan-Meier and Cox proportional hazards methods. This first planned OS interim analysis took place after ~50% of expected events. Results: 525 pts were randomized to APA and 527 to PBO. Median age was 68 y; 8% had prior tx for localized disease; 11% had prior docetaxel. 63% and 37% had high- or low-volume disease, respectively. At median 22.6 mo follow-up, 66% APA and 46% PBO pts remained on tx. APA significantly improved rPFS (HR, 0.48; 95% CI, 0.39-0.60; p < 0.0001), with a 52% reduction in risk of death or radiographic progression; benefit was observed across all subgroups analyzed. Median rPFS was not reached in the APA group and 22.1 mos in the PBO group. APA also significantly improved OS (HR, 0.67; 95% CI, 0.51-0.89; p = 0.0053), with a 33% reduction in risk of death. Median OS was not reached in the APA or PBO group. Time to initiation of cytotoxic chemotherapy was significantly improved with APA (HR, 0.39; 95% CI, 0.27-0.56; p < 0.0001). Based on these results, the independent data monitoring committee recommended unblinding to allow crossover of PBO pts to receive APA. Rates of grade 3/4 adverse events (AEs) (42% APA, 41% PBO) were similar, and discontinuations due to AEs (8% APA, 5% PBO) were low. Conclusions: In the TITAN study in pts with mCSPC, including pts with high- and low-volume disease and prior docetaxel, addition of APA to ADT significantly improved rPFS and OS, and the safety profile was tolerable. These results support the addition of APA to ADT for tx of pts with mCSPC. Clinical trial information: NCT02489318.

4519 Background: Atezo, a monoclonal antibody targeting PD-L1, is an approved therapy for locally advanced/metastatic UC based on IMvigor210 and IMvigor211 phase II and III trials. The single-arm SAUL study (NCT02928406) with a broader patient (pt) population demonstrated median overall survival (OS) of 8.7 months and a safety profile consistent with previous atezo trials. Methods: Pts with locally advanced/metastatic UC or non-UC of the urinary tract received atezo 1200 mg every 3 weeks until disease progression or unacceptable toxicity. Populations excluded from IMvigor211 (renal impairment, ECOG PS 2, treated asymptomatic CNS metastases, stable controlled autoimmune disease, concomitant steroids, HIV positive, non-UC) were eligible. The primary endpoint was safety; OS and overall response rate (ORR) were secondary endpoints. Predefined subgroup analyses included outcomes according to PD-L1 status (VENTANA SP142) and age in the overall population (and the IMvigor211-like subgroup for PD-L1). Results: Between Nov 2016 and Mar 2018, 1004 pts were enrolled; 997 received atezo. Efficacy is summarized below. Incidences of grade ≥3 treatment-related adverse events were similar irrespective of PD-L1 status (overall IC 0/1 vs 2/3: 11% vs 16%; IMvigor211-like IC 0/1 vs 2/3: 11% vs 15%) or age (≥65 y: 13%; ≥75 y: 12%; ≥80 y: 10%). Conclusions: OS and ORR appear more favorable in IC 2/3 vs IC 0/1 subgroups (overall and in the IMvigor211-like population). Atezo was effective and well tolerated across subgroups including elderly pts. Clinical trial information: NCT02928406. [Table: see text]

6 Background: Stratifying prostate cancer (PCa) patients into risk groups at time of initial diagnosis for individual treatment decision is still a major clinical challenge. PSMA expression has emerged as a promising prognostic biomarker since its overexpression in radical prostatectomy specimens (RP) has been linked to disease recurrence. Aim of our study was to assess the prognostic value of PSMA on prostate biopsies (Bx) thus improving risk stratification at time of initial diagnosis. Methods: Immunohistochemistry for PSMA expression was performed on 294 Bx with corresponding RP, 621 primary tumor foci from 242 RP, 43 locally advanced or recurrent tumors, 34 lymph node metastases, 78 distant metastases, and 52 benign prostatic samples. Grade group, PSA, TNM-, and R-status were assessed as clinico-pathologic features. Primary endpoint was recurrence-free survival (RFS). Chi-square, ANOVA-analyses, Cox regression and log rank tests were performed for statistical analyses. Results: PSMA expression significantly associates with grade group and initial PSA level. Elevated PSMA expression on both RP and Bx significantly correlates with an increased risk of disease recurrence after curative surgery. 5-year RFS rates are 88.2%, 74.2%, 67.7%, and 26.8% for patients with no, low, medium, or high PSMA expression on Bx, respectively. Elevated PSMA level on Bx predict a 4-fold increased risk of disease recurrence independently from initial PSA and grade group on Bx. PSMA expression significantly increases during PCa progression. Conclusions: PSMA qualifies as an independent prognostic biomarker on Bx at time of initial diagnosis in addition to the established markers PSA and grade group. PSMA predicts disease recurrence following curative surgery and potentially improves the discrimination indolent vs. aggressive disease. We propose the routine assessment of PSMA expression on Bx for outcome prediction and risk stratification at time of initial diagnosis prior to treatment decision.

461 Background: Identification of molecular drivers that promote a muscle invasive phenotype and chemoresistance in bladder cancer, as well as an understanding of their mechanism of action should help in designing potential prognostic markers and effective targeted treatments for patients with advanced bladder cancer. β-arrestins (ARRBs) are classic attenuators of G-Protein coupled receptor signaling. However, they have multiple roles in cellular physiology, including carcinogenesis. Methods: β-arrestin-1 (ARRB1) and β-arrestin-2 (ARRB2) mRNA levels were measured by quantitative RT-PCR in two clinical specimen cohorts (n = 63; n = 43). The role of ARRBs in regulating a stem cell-like phenotype and response to chemotherapy treatments was investigated using CRISPR-Cas9 mediated gene-knockout of ARRB1. The consequence of forced expression of ARRBs on tumor growth and response to gemcitabine in vivo were investigated using bladder tumor xenografts in nude mice. Results: ARRB1 levels were significantly elevated and ARRB2 levels downregulated in cancer tissues compared to normal tissues. In multivariate analysis only ARRB2 was an independent predictor of metastasis, disease-specific-mortality and failure to Gemcitabine + Cisplatin (G+C) chemotherapy; ~ 80% sensitivity and specificity to predict clinical outcome. ARRBs were found to regulate stem cell characteristics in bladder cancer cells. Depletion of ARRB2 resulted in increased cancer stem cell markers but ARRB2 overexpression reduced expression of stem cell markers (CD44, ALDH2, and BMI-1), and increased sensitivity towards gemcitabine. Overexpression of ARRB2 resulted in reduced tumor growth and increased response to Gemcitabine in tumor xenografts. Conclusions: β-Arrestins regulate response towards gemcitabine and are therefore potential targets and predictive markers for chemotherapy failure in bladder cancer. The present work shows for the first time that β-arrestins have prognostic significance for predicting metastasis and response to chemotherapy in bladder cancer.

152 Background: The mediator complex plays a pivotal role in the regulation of gene transcription. The subunits CDK8 and CDK19 are overexpressed in advanced prostate cancer (PCa) and promote migration and invasion, as shown previously by our group. The aim of this study was to analyze if CDK8/CDK19 inhibition can impede PCa progression. Methods: Immunohistochemistry for CDK8 and CDK19 was performed on a large PCa cohort (376 radical prostatectomy (RP) specimens, 39 locally advanced tumors, 32 lymph node metastases, 24 distant metastases, 73 benign prostatic specimens). PCa cell lines (DU145, PC3) were treated with CDK8/CDK19 small molecule inhibitors followed by MTT assay, wound healing assay, and Western Blot for epithelial/mesenchymal markers. Results: CDK19 is higher expressed in primary PCa vs. benign specimens and locally advanced PCa/distant metastases vs. primary PCa. CDK8 and CDK19 are higher expressed in castration-resistant specimens vs. hormone-naive specimens. CDK19 significantly correlates with grade group in RP specimens. Biochemical recurrence free survival rates are lower for patients with high vs. medium vs. no/low CDK19 expression in the tumor. CDK19 predicts recurrence free survival independently from grade group and PSA. CDK8/CDK19 inhibition results in decreased migration, while there is no effect on proliferation. Mesenchymal markers are reduced while epithelial markers are induced following CDK8/CDK19 inhibition. Conclusions: CDK19 qualifies as a prognostic marker predicting biochemical recurrence following RP. CDK8/CDK19 inhibition leads to decreased migratory potential and mesenchymal phenotype. CDK8/19 expression increases during progression to castration-resistance. Collectively, CDK8/CDK19 represents a new target for treatment of advanced and/or castration-resistant PCa.

5067Background: pAR gene aberrations in mCRPC patients (pts) may be associated with worse outcome on AA+P/D. Methods: This was an international, multi-institutional, open-label phase 2 study of AA ...

5039 Background: Carboplatin plus docetaxel (DC) may be effective in mDRPC. Platinum(II)-complexes interfere with steroid biosynthesis lowering testosterone levels by inhibiting the cholesterol side chain cleavage enzyme (CYP11A1), 3β-hydroxysteroid dehydrogenase (HSD3B1,2) and 17α hydroxylase/C17,20-lyase (CYP17A1). Methods: Docetaxel failure/resistance was defined according to the Prostate Cancer Working Group (PCWG2 2007) criteria. Treatment consisted of at least two cycles of carboplatin AUC5 iv for 30 min on day 1 every 4 weeks (q4w), docetaxel at a dose of 35 mg/m2 iv for one hour on days 1, 8, (15) plus prednisone 2x5mg/day orally after receiving informed consent until disease progression or occurrence of intolerable adverse effects. Efficacy measures were done following PCWG2 recommendations. Free testosterone levels were measured before (n = 77) and during DC chemotherapy (n = 69). Results: Of the 100 pts. treated since February 2005, 96% had bone metastases, 45% had lymph node, 27% liver and 21% lung involvement. At the time of the current analysis, the median follow-up was 13.6 months, 93 pts. had died and 97 had progressive disease. The objective response rate was 36.5% in the 63 pts. with measureable disease. Response of prostate-specific antigen (≥50%) was observed in 50% of patients. Median progression-free survival (PFS) for all patients was 6.9 months (CI 95% 5.5, 8.3) and median OS was 15.4 months (CI 95% 11.5, 19.4). The most common reversible grade 3/4 toxicity was leukopenia/ neutropenia (40/32%). Median free testosterone levels were 0.61 pg/ml before and < 0.18 pg/ml during carboplatin/docetaxel treatment (nadir levels, p < 0.001; detection limit < 0.18 pg/ml). Median serum androgene levels (T+DHT) were 0.1 ng/ml before and below the detection limit of < 0.05 ng/ml during DC treatment. In multivariate analyses, LDH, PSA response, free testosterone nadir levels below the detection limit ( < 0.18 pg/mL) during DC treatment were associated with longer OS (p < 0.05). Conclusions: These data suggest that carboplatin plus weekly docetaxel may be an important salvage treatment option for DRPC patients.

318 Background: We had previously shown that a significant number of patients after radical cystectomy (RC) suffer from changes in bowel habits and defecation. Reports addressing long-term bowel disorders following RC are rare. This cross sectional study evaluates long-term bowel issues in a large cohort with the help of an issue-tailored questionnaire. Methods: For this purpose a questionnaire assessing changes in bowel function and its impact on daily life was developed and distributed among members of the German bladder cancer self-help group. A total of 431 patients after RC were evaluated. Symptoms such as diarrhea, constipation, urge to defecate, sensation of incomplete defecation, flatulence, and impact on quality of life (QoL) were assessed. Results: A total of 324 patients were followed ≥1 year, 43% patients of those reported current bowel disorders, 40% life restriction and 60% dissatisfaction. Most frequent bowel symptoms were flatulence (49%), followed by diarrhea (30%) and the sensation of incomplete defecation (23%). The highest prevalence rate of diarrhea is reported in year 3 after surgery

351 Background: Muscle-invasive urothelial carcinoma is usually treated with radical cystectomy (RC). However, patients face a relevant risk of postoperative complications on variable factors. Recent reports described that sarcopenia, which is defined by a loss of sceletal muscle body composition, may enhance risk stratification. Methods: Our retrospective study includes 200 patients (154 men, 46 women, mean age: 67.9 years) which were treated with RC between 2005 and 2014. Median follow-up time was 26 months. Computed tomography (CT) scans were performed within 60 days prior to surgery. Lumbar skeletal muscle index (SMI) were measured with MeVisLab 2.7 by manually segmentation of preoperative CTs. The fat mass index (FMI) was calculated based on the same CT slides. Cut off points were developed for men and women separately. Additionally, short-, mid- and long-term complications were assessed using the Clavien-Dindo classification. Results: SMI (cm2/m2), FMI (kg/m2) and BMI (kg/m2) for men were calculated with mean scores of 52.8, 9.6 and 26.7, respectively. The corresponding calculated means for women were 38.9, 10.0 and 26.1. The SMI for short- and mid-term complications was statistically significant (p < 0.01) when applied on patients with a BMI over 25 kg/m2. Using Kaplan-Meier plots none of the other markers (BMI and FMI) correlated with complications or overall survival (OS). Conclusions: The presence of sarcopenia assessed by CT may improve risk stratification in patients undergoing radical cystectomy. BMI and FMI may not be used in risk stratification models, even though further validation for subgroups is needed. Future research is needed to validate the current findings. Noninvasive and semiautomatic calculation of SMI in preoperative CT scans may improve risk stratification in patients undergoing radical cystectomy and thus could reduce morbidity and mortality.

5029Background: Carboplatin plus docetaxel may be effective in mDRPC. Platinum(II)-complexes have been shown to interfere with steroid biosynthesis lowering testosterone levels by inhibiting the ch...