Search

PURPOSE In this phase II response-adaptive trial, we investigated the rational application of immune checkpoint blockade in renal cell carcinoma (RCC; ClinicalTrials.gov identifier: NCT03203473 ). METHODS We enrolled patients with metastatic RCC with no prior checkpoint inhibitor exposure. All patients received nivolumab alone with subsequent arm allocation based on response. Patients with a confirmed partial response (PR) or complete response (CR) within 6 months discontinued nivolumab and were observed (arm A). Patients with stable disease or progressive disease (PD) after no more than 6 months of nivolumab received two doses of ipilimumab (arm B). The primary endpoints were the proportion of patients with PR/CR at 1 year after nivolumab discontinuation (arm A) and proportion of nivolumab nonresponders who converted to PR/CR after ipilimumab (arm B). RESULTS Overall, 83 patients initiated treatment, of whom 96% had clear-cell histology, 51% were treatment naïve, and 67% had intermediate/poor-risk disease. Median follow-up was 19.5 months. Within 6 months, induction nivolumab resulted in a confirmed PR in 12% of patients (n = 10). Fourteen patients were not allocated to a study arm (seven because of toxicity, seven because of PD). Twelve patients (14%) were allocated to arm A and discontinued nivolumab, of whom five (42%; 90% CI, 18% to 68%) remained off nivolumab at ≥ 1 year. Of 57 patients (69%) allocated to arm B, two patients converted to a confirmed PR (4%; 90% CI, 1% to 11%), and no CRs were observed. CONCLUSION In this study, nivolumab followed by two doses of ipilimumab resulted in no CRs and a low PR/CR conversion. The number of patients evaluated for nivolumab discontinuation was too small to assess the value of this approach. Currently, our data do not support a response-adaptive strategy for checkpoint blockade in advanced RCC.

2501 Background: The immunoregulatory cytokine IL-27 upregulates inhibitory immune checkpoint receptors (eg, PD-L1, TIGIT) and downregulates proinflammatory cytokines (eg, IFNγ, TNFα). SRF388 is a fully human IgG1 blocking antibody to IL-27 with potential to promote immune activation in the tumor microenvironment. A phase 1 study was conducted to establish the preliminary safety of SRF388 and to identify recommended phase 2 doses (RP2D) suitable for monotherapy and combination expansions (NCT04374877). Methods: The dose-escalation study (accelerated single patient followed by standard 3+3) enrolled patients (pts) with advanced treatment-refractory solid tumors. Upon RP2D selection, monotherapy and combination expansions opened for treatment-refractory clear cell renal cell cancer (ccRCC), hepatocellular cancer (HCC), and non-small cell lung cancer. SRF388 was administered IV every 4 weeks (wks) as monotherapy and every 3 wks with pembrolizumab. Tumor response was assessed by RECIST1.1. Results: The monotherapy dose-escalation enrolled 29 pts with doses ranging from 0.003 to 20 mg/kg. Median age was 64 years. Most pts were female (62%) with ECOG PS of 1 (72%). Approximately 80% had prior PD-(L)1 blockade, and 48% had ≥4 prior therapies. Treatment-related adverse events (TRAEs) occurred in 21%, and all were low grade. Fatigue was the only TRAE reported in ≥10% (n = 3). No dose-limiting toxicities (DLTs) or Grade ≥3 TRAEs were observed. Median time on study was 9 wks (range 0–59). One patient with highly treatment-refractory NSCLC experienced a confirmed partial response (PR) at 8 wks that was durable for 20 wks. Nine pts (31%) experienced disease stabilization at 8 wks, with 6 of 9 exhibiting durable disease control at 6 months. Of the 7 pts with ccRCC in the dose-escalation portion of the trial, 3 (43%) experienced durable disease control for ≥20 wks (range: 20-32). With doses up to 20 mg/kg, SRF388 PK remain linear with an estimated T1/2 of 10-12 days. PK characteristics and safety profile support dosing every 3 or 4 wks. Based on safety, tolerability, PK, peripheral pSTAT1 inhibition, and preliminary efficacy, 10 mg/kg was selected as the RP2D. Both the pembrolizumab safety cohort (n = 10) and Stage 1 of the ccRCC monotherapy expansion (n = 17) have fully enrolled. Of the 10 evaluable pts with ccRCC, 1 confirmed monotherapy PR has been reported, enabling Stage 2 initiation. Changes in several serum cytokines and chemokines were observed after SRF388 administration, including expected increase in circulating IL-27 levels. Conclusions: Results of IL-27 pathway blockade with a first-in-class therapeutic demonstrates that SRF388 has good tolerability with encouraging preliminary antitumor activity as a monotherapy. Updated data, including safety and clinical outcomes as well as correlative biomarker analyses, will be presented. Clinical trial information: NCT04374877.

Purpose FGFR3-TACC3 (fibroblast growth factor receptor 3–transforming acidic coiled coil-containing protein 3) fusions have recently been identified as driver mutations that lead to the activation of FGFR3 in bladder cancer and other tumor types and are associated with sensitivity to tyrosine kinase inhibitors. We examined the clinical and molecular characteristics of patients with FGFR3-TACC3 fusions and hypothesized that they are enriched in a subset of patients with bladder cancer. Materials and Methods We correlated somatic FGFR3-TACC3 fusions with clinical and molecular features in two cohorts of patients with bladder cancer. The first cohort consisted of the muscle-invasive bladder cancer (MIBC) data set (n = 412) from The Cancer Genome Atlas. The second cohort consisted of patients with MIBC or high-grade non-MIBC at the Dana-Farber Cancer Institute that had targeted capture sequencing of a selected panel of cancer genes (n = 356). All statistical tests were two sided. The clinical response of one patient with FGFR3-TACC3 bladder cancer to an FGFR3 inhibitor was investigated. Results Overall, 751 patients with high-grade bladder cancer without FGFR3-TACC3 fusions and 17 with FGFR3-TACC3 fusions were identified in the pooled analysis of the data sets from The Cancer Genome Atlas and the Dana-Farber Cancer Institute. FGFR3-TACC3 fusions were enriched in patients age ≤ 50 years versus age 51 to 65 years versus those older than 65 years (pooled, P = .002), and were observed in four (12%) of 33 patients age ≤ 50 years in the pooled analysis. Similarly, FGFR3-TACC3 fusions were significantly more common in Asians (13%) compared with African Americans (4%) and whites (2%; pooled, P < .001), as well as in never smokers (5.6%) compared with ever smokers (1.1%; pooled, P < .001). One patient with the fusion who was treated with an FGFR3 inhibitor achieved complete remission for 10 months. Conclusion Clinical testing to identify FGFR3 fusions should be prioritized for patients with bladder cancer who are younger, never smokers, and/or Asian.

2551 Background: IL-27 is an immunosuppressive cytokine, consisting of two subunits p28 and EBI3, that upregulates immune checkpoint receptors (eg, PD-L1, TIGIT) and downregulates proinflammatory cytokines such as IFNγ, TNFα, and IL-17. SRF388 is a first-in-class, fully human IgG1 antibody to IL-27 that blocks the interaction between IL-27 and its receptor, thereby promoting immune activation in the tumor microenvironment. The IL-27 pathway is activated in hepatocellular carcinoma (HCC) and renal cell carcinoma (RCC), and high circulating levels of EBI3 are associated with inferior outcomes in both. Circulating EBI3 levels may serve as a predictive biomarker of SRF388 activity. Methods: Patients with advanced solid tumors refractory to standard therapy were enrolled in a phase 1 dose-escalation study (accelerated single patient followed by standard 3+3) to establish the preliminary safety of SRF388 as a monotherapy and to identify a dose suitable for expansion (NCT04374877). SRF388 was administered intravenously every 4 weeks. Tumor response was assessed by RECIST v1.1. SRF388 pharmacokinetic (PK) and pharmacodynamic (PD) [phospho-STAT (pSTAT) inhibition] analyses were performed. Results: As of January 26, 2021, 12 patients have received SRF388 at doses ranging from 0.003 to 10 mg/kg with 2 patients undergoing intra-patient dose escalation. Median age was 68 years, 67% were female, and ECOG PS was 0/1 (42%/58%). Median number of prior therapies was 2 (range 1–9), and 75% were anti-PD-(L)1 experienced (n = 9). The only treatment-related adverse events observed across dose levels were low-grade fatigue (n = 1, 8%), nausea (n = 1, 8%) and excess salivation (n = 1, 8%). No dose-limiting toxicities (DLTs) or ≥ Grade 3 related toxicity have occurred. Mean time on study is 12.5 weeks (range 4–40). One patient with RCC who received prior anti-PD-1 has prolonged stable disease for > 9 months. SRF388 PK are linear with estimated T1/2 ranging from 6–19 days. There is evidence of accumulation and no anti-drug antibody development to date. Maximal inhibition of the IL-27 signaling pathway as measured by > 90% pSTAT inhibition in whole blood was achieved starting at 0.3 mg/kg. Given combined evidence of near-complete pathway inhibition and preclinical human equivalent dose modeling projecting biologically active doses, additional slots were opened for RCC and HCC starting at 1 mg/kg. Conclusions: Preliminary results of IL-27 pathway blockade with a first-in-class therapeutic demonstrates that SRF388 is well tolerated at doses that achieve maximal inhibition of downstream pSTAT signaling through the dosing period. Expansions are planned in HCC and RCC. Updated data including the recommended phase 2 dose, clinical outcomes, PK/PD and correlative analyses will be presented. Clinical trial information: NCT04374877.

98 Background: Treatment (tx) options for patients (pts) with metastatic castration-resistant prostate cancer (mCRPC) to bone are limited. Radium-223 (R223) has demonstrated overall survival (OS) benefit, but objective clinical responses to R223 or the anti-PD1 checkpoint inhibitor (CPI) pembrolizumab (pem) are infrequent. As R223 may increase immunogenicity of mCRPC to bone and increase activity of CPI, we undertook a Phase 2 study to assess safety of the combination and differences in immune cell infiltrate in bone biopsies (bx) and preliminary clinical activity of R223 + pem vs. R223 alone. Methods: Eligibility required mCRPC to bone with no visceral metastases (mets) or lymph nodes > 2 cm, ECOG PS 0 or 1, Hgb ≥ 9 g/dL, and no prior R223 or CPI. Pts underwent bone bx at screening and at 8 wks. Pts were stratified by alkaline phosphatase ≥220 vs. < 220 U/L and high vs. low volume bony mets (CHAARTED criteria) and randomized 2:1 to receive R223 55 kBq/kg q4wks + pem 200 mg q3wks (Arm A) or R223 55 kBq/kg q4wks alone (Arm B). If restaging after 3 doses R223 showed at least stable disease, pts in Arm A continued pem alone until progressive disease (PD). Upon PD, R223 was resumed if no new visceral mets. Pts continued tx until clinical/radiologic PD, unacceptable toxicity or completion of 6 R223 doses. The primary endpoint was difference in CD4+ and CD8+ T-cell infiltrate in 8 wk vs. baseline bx; secondary endpoints were safety/tolerability, radiographic progression-free survival (rPFS) and OS. Exploratory endpoints included PSA response and rate of symptomatic skeletal events (SSEs). Results: Of 45 pts enrolled, 42 received study tx (29 Arm A, 13 Arm B) and were eligible for analysis. 21 pts in Arm A and 5 in Arm B had evaluable paired bone bx. Median fold-change of proportion of CD4+ T-cells/total cell count from baseline to 8 wks was 0.90 (range 0.0-26.6) in Arm A and 0.40 (0.0-13.0) in Arm B (P = 0.87); for CD8+ cells, median 0.67 (0.0-40.4) in Arm A and 0.40 (0.1-28.8) in Arm B (P = 0.77). Grade 3 treatment-related non-hematologic adverse events (AEs) occurred in 3 pts (10%) in Arm A (pneumonitis, diarrhea, AST increased); none in Arm B. Median rPFS was 6.7 mo (95% CI 2.7-11.0 mo) in Arm A and 5.7 mo (2.6-NR) in Arm B. Median OS was 16.9 mo (12.7-NR) in Arm A and 16.0 mo (9.0-NR) in Arm B. 3 pts (10%) in Arm A and 0 in Arm B had PSA reduction of ≥ 50%. SSE rate was 38% in Arm A and 54% in Arm B, with pathologic fractures in 0% of pts in Arm A and 23% in Arm B. Conclusions: In the 62% of treated pts with evaluable paired bx at baseline and after 8 wks, there was no evidence of increased CD4+ or CD8+ T-cell infiltration with R223 + pem. Additional biomarker analyses will be presented. This study revealed that R233 + pem did not result in unexpected AEs, but did not lead to prolonged rPFS or OS compared to R223 alone to support this two-drug combination in a biomarker-unselected population in this setting. Clinical trial information: NCT03093428.

306 Background: Approved rapalogs inhibit mTORC1 and have limited activity in mRCC, possibly due to compensatory feedback loops. Sapanisertib addresses the incomplete inhibition of the mTOR pathway through targeting of both mTORC1 and mTORC2 with antitumour activity demonstrated in patients with mRCC. In this multicenter, single arm phase II trial, we evaluated the efficacy of sapanisertib in patients with mRCC progressing on standard therapies (NCT03097328). Methods: Eligible mRCC patients had an ECOG performance status of 0-2 and had progressed on standard therapies. Prior therapy with rapalogs (everolimus, temsirolimus) and variant RCC histologies were permitted. Patients had a baseline biopsy and received treatment with sapanisertib 30 mg by mouth weekly until unacceptable toxicity or disease progression. The primary endpoint was overall response rate (ORR) by RECIST 1.1. Tissue biomarkers of mTOR pathway activation were explored. Results: We enrolled 38 mRCC patients (clear cell = 28; variant histology = 10) between August 2017 and November 2019. The majority had intermediate (76%) or poor risk (11%) by IMDC criteria. Twenty (53%) had received ≥ 3 lines of therapy; 13 (34%) patients received prior rapalogs. Median follow-up was 10.4 months (range 1-27.4) and median duration of therapy was 1.6 (range 0.3-13.8) months. ORR by central review was 2 of 38 (5.3% 90%CI: 1%-15.6%). 31.6% of all patients and 30.7% of those with prior rapalog exposure had some tumor shrinkage during course of treatment. Median progression free survival (PFS) was 2.5 months (95% CI 1.8,3.7). Twelve patients (32%) developed treatment-related grade 3 adverse events (AEs) with no grade 4 or 5 toxicity reported; 6 patients (16%) required dose reduction and 4 (11%) discontinued therapy for AEs. Oncopanel tumor sequencing identified alterations in the mTOR pathway in 6 of 29 patients ( MTOR n = 2, PTEN n = 3, TSC1 n = 1.) Reduced PTEN expression by immunohistochemistry was seen in 7 of 19 patients. There was no association between mTOR pathway mutations or PTEN loss and response to sapanisertib. Conclusions: In this study we demonstrate minimal activity of sapanisertib in patients with treatment refractory mRCC with no clear benefit among patients with mTOR/PTEN pathway alterations. Additional treatment strategies are needed for patients with refractory mRCC.

310 Background: Cabo shows robust clinical activity in mRCC. Patients (pts) with BM have been underrepresented in clinical trials and effective systemic therapy is lacking. We retrospectively characterized the clinical activity and toxicity of cabo in pts with BM from RCC. Methods: Consecutive medical records from mRCC pts with BM treated with cabo monotherapy across 15 institutions were reviewed. Pts were grouped by radiologic presence (cohort 1) or absence (cohort 2) of progressing intracranial metastases. Brain-directed local therapy was allowed but radiological confirmation of intracranial progression at cabo start was required in cohort 1. Radiological response rate was investigator-assessed by modified RECIST 1.1 for intracranial and RECIST 1.1 for extracranial responses. Time to treatment failure (TTF) and overall survival (OS) were estimated by Kaplan-Meier. Results: We identified 69 pts with BM from RCC, 25 (36%) in cohort 1 and 44 (64%) in cohort 2. Majority were IMDC intermediate/poor (87%) and received cabo as ≥2nd line (75%). Median time from mRCC diagnosis to BM was 19.1 months (mos) (IQR 4.4-39.5). Overall, median number of BM was 3 (range 1-27) and median size of largest lesion was 1.2 cm (range 0.2-6.6) with frontal (62%) and parietal (48%) as the most frequent localizations. Prior brain directed therapy was used in 65% and 93% of pts in cohort 1 and 2 respectively. Median follow-up after cabo initiation was 11 mos (range 4-72). Twenty three percent of pts remained on therapy while 52% discontinued for progression and 9% for toxicity. Intracranial response rate was 61% (95%CI 39%-80%), with 3 complete responses, for cohort 1 and 57% (95%CI 41%-72%) for cohort 2. Only 10% (n = 7) had intracranial progression as best response. For cohort 1, extracranial response was 52% (95%CI 31%-72%), median TTF was 9.9 mos (95%CI 5.9-14.0) and OS was 14.7 mos (95%CI 7.7-23.0). For cohort 2, extracranial response was 41% (95%CI 26%-57%), TTF was 9.0 mos (95%CI 4.6-11.4) and OS was 14.1 mos (95%CI 11.0-22.0). Most common adverse events were fatigue (77%) and diarrhea (46%). Eight pts received concomitant brain-directed treatment during cabo therapy without neurological toxicities. Conclusions: Cabo shows significant intracranial activity and acceptable safety profile in pts with BM from RCC. [Table: see text]

459 Background: First-line platinum-based combination chemotherapy (fPBC) is standard of care for fit patients with mUC. Further lines of therapy include immuno-oncology agents, erdafitinib, and enfortumab vedotin, but patients ineligible for these therapies or who subsequently progress may be considered for further chemotherapy. As the choice of chemotherapy regimen is unclear for these patients, we compared the efficacy of subsequent platinum-based chemotherapy (sPBC) and subsequent non-platinum-based chemotherapy (sNPBC) in patients with mUC. Methods: Data was analyzed from the Retrospective International Study of Cancers of the Urothelium (RISC), comprising patients from 28 international centers treated 2005-2012. Inclusion criteria were diagnosis of mUC, receipt of fPBC for mUC, and receipt of ≥2 cycles of subsequent chemotherapy. Patients who had received prior platinum-based chemotherapy in the non-metastatic setting were excluded. A multivariate Cox proportional hazards model was used to compare overall survival (OS), while χ2 and student’s t-test were used for univariate analyses. A two-sided p value of

TPS4596 Background: There is no standard adjuvant systemic therapy that increases overall survival (OS) over surgery alone for non-metastatic RCC. Anti-PD-1 nivolumab (nivo) improves OS in metastatic RCC and is well tolerated. In mouse models, priming the immune system prior to surgery with anti-PD-1 results in superior OS compared to adjuvant dosing. Remarkable pathologic responses have been seen with neoadjuvant PD-1 in multiple ph 2 studies in bladder, lung and breast cancers. Phase 2 neoadjuvant RCC trials of nivo show preliminary feasibility and safety with no surgical delays. PROSPER RCC seeks to improve clinical outcomes by priming the immune system with neoadjuvant nivo prior to nephrectomy followed by continued immune system engagement with adjuvant blockade in patients (pts) with high risk RCC compared to standard of care surgery alone. Methods: This global, unblinded, phase 3 National Clinical Trials Network study is accruing pts with clinical stage ≥T2 or TanyN+ RCC of any histology planned for radical or partial nephrectomy. Select oligometastatic disease is permitted if the pt can be rendered ‘no evidence of disease’ within 12 weeks of nephrectomy (≤3 metastases; no brain, bone or liver). In the investigational arm, nivo is administered 480mg IV q4 weeks with 1 dose prior to surgery followed by 9 adjuvant doses. The control arm is nephrectomy followed by standard of care surveillance. There is no placebo. Baseline tumor biopsy is required only in the nivo arm but encouraged in both. Randomized pts are stratified by clinical T stage, node positivity, and M stage. 805 pts provide 84.2% power to detect a 14.4% absolute benefit in recurrence-free survival at 5 years assuming the ASSURE historical control of ̃56% to 70% (HR = 0.70). The study is powered to evaluate a significant increase in OS (HR 0.67). Critical perioperative therapy considerations such as safety, feasibility, and quality of life metrics are integrated. PROSPER RCC embeds a wealth of translational studies to examine the contribution of the baseline immune milieu and neoadjuvant priming with anti-PD-1 on clinical outcomes. As of February 10, 2021, 704 patients have been enrolled (N = 805). Clinical trial information: NCT03055013.

5005 Background: Nivo + Ipi is an established first-line treatment (tx) for advanced RCC. We hypothesized that the addition of CTLA-4 blockade may not be required for all patients (pts). Furthermore, the optimal duration of Nivo maintenance in responding pts is unknown. In this phase II response-adaptive trial, we investigate the sequential addition of 2 doses of Ipi to induce response in Nivo non-responders (NR) and duration of Nivo in responding pts (NCT03203473). Methods: We enrolled pts with advanced RCC with no prior checkpoint inhibitor exposure. All pts received Nivo alone with subsequent arm allocation based on RECISTv1.1 response within 6 months (mos) of tx. Pts with a confirmed partial response (PR) or complete response (CR) within 6 months (mos) discontinued Nivo and were observed (Arm A). Arm A pts reinitiated Nivo if they developed progressive disease (PD); Ipi was added to Nivo if PD persisted or recurred. Pts with stable disease (SD) or PD after no more than 6 mos of Nivo alone received 2 doses of Ipi (Arm B). The primary endpoints were the proportion with PR/CR at 1-year (yr) after Nivo discontinuation (Arm A) and proportion of Nivo NR who convert to PR/CR after adding Ipi (Arm B). Results: 83 pts initiated tx of whom 99% had ECOG 0-1, 96% clear cell RCC, 51% tx-naïve, and 69% IMDC intermediate/poor risk. Median follow-up was 17.0 mos. 15 pts were not allocated to an arm [7 withdrew for PD, 7 withdrew for toxicity, 1 still on tx with unconfirmed PR (uPR)]. At 6 mos, induction Nivo resulted in a confirmed PR in 11% of pts (n=9/83): 12% (n=5/42) tx-naïve, 10% (4/41) prior tx, 8% (n=1/13) favorable risk, 11% (n=8/70) intermediate/poor risk (Table). 11 pts (13%: 9 PR, 1 uPR, 1 SD) were allocated to Arm A, of whom 5 (45%, 90% CI 20-73%) remained off Nivo at ≥ 1 yr. Of 57 pts (69%) allocated to Arm B, 2 pts converted to a PR (4%, 90% CI 1-11%), both of whom had prior tx and PD as best response to Nivo alone. Grade 3-4 treatment related adverse events (TrAE) occurred in 7% (n=6/83) on induction Nivo and in 23% (n=13/57) on Arm B (Nivo + Ipi). Conclusions: We cannot currently recommend a strategy of Nivo followed by response-based addition of Ipi due to the absence of CR and low PR/CR conversion rate (4%). Though a subset of pts treated with Nivo alone can maintain durable responses off tx at 1-yr, early Nivo discontinuation in the absence of toxicity cannot currently be recommended. Investigation into biomarkers to guide tx is ongoing. Clinical trial information: NCT03203473 . [Table: see text]

e17005 Background: The optimal choice of first-line chemotherapy for urothelial carcinoma (UC) patients who recur after previous cisplatin-based chemotherapy for locally-advanced disease is unclear. Our objective is to compare the efficacy of platinum (PBC) versus non-platinum (NPBC) based first-line chemotherapy regimens for such patients after metastatic recurrence. Methods: Data was collected from the Retrospective International Study of Cancers of the Urothelial Tract (RISC), a database consisting of patients with muscle-invasive or advanced UC from 28 centers between 2005 and 2012. Patient inclusion criteria included: 1) UC with no initial metastases (cT2-4, cN0-N3, and cM0), 3) receipt of cisplatin in the locally advanced setting, and 4) receipt of chemotherapy in the first-line metastatic setting. Overall survival (OS) was the primary endpoint. Secondary endpoints included progression-free survival (PFS) and response to chemotherapy. Kaplan-Meier and Cox regression models estimated OS, PFS, and response, adjusting for age, gender, Eastern Cooperative Oncology Group (ECOG-PS), Charlson comorbidity index (CCI), surgery, T and N stage, albumin, creatinine clearance, number of initial cisplatin cycles, and time from last chemotherapy. Results: 152 patients with metastatic UC (88 PBC and 64 NPBC) were analyzed. Twelve (7.9%) patients received local definitive radiation and 7 of these 12 also underwent cystectomy. The most common NPBC regimens included taxanes, gemcitabine, or pemetrexed. The median OS was 8.70 (95% CI: 7.53 to 11.16) and 10.27 months (95% CI: 7.37 to 13.10) for PBC and NPBC (HR: 1.04, 95% CI: .67 – 1.61, p = 0.86), respectively. Multivariable analysis showed an independent prognostic effect on OS for number of previous chemotherapy cycles (3-4 vs. 1-2) (HR: 0.44, 95% CI 0.20 – 0.96, P = 0.03) and whether surgery was performed (HR: 0.44, 95% CI 0.26 – 0.75, P = 0.003). Time from last chemotherapy was not prognostic for OS (HR: 0.99, 95% CI: 0.99 – 1.00, p = 0.19). There were no significant differences for both investigator-designated PFS (HR: 0.84, 95% CI: 0.57 – 1.24, p = .39) and response to chemotherapy between PBC and NPBC (p = 0.57). Conclusions: There is no significant outcome difference between PBC vs. NPBC in patients with metastatic-recurrent urothelial carcinoma who previously received cisplatin-based chemotherapy for locally advanced disease. Those who previously underwent radical surgery or who received 3-4 cycles of cisplatin had better OS with PBC.

5542 Background: Ra-223, a targeted alpha therapy, showed a survival benefit and favorable safety profile over 3 years’ (yrs) follow-up in mCRPC pts (ALSYMPCA trial). REASSURE (NCT02141438) is a global, prospective, single-arm, observational study of long-term Ra-223 safety in routine clinical practice in mCRPC pts (planned 7-yr follow-up). Methods: This analysis, based on the second prespecified interim analysis (data cutoff 3-20-2019) of REASSURE (N = 1465), evaluated safety/OS in the pt subset that was chemotherapy-naïve at Ra-223 administration but received subsequent taxane therapy any time after Ra-223 completion. Results: 182 pts received taxane therapy after Ra-223. Most (58%) had unresected primary tumors, 69% had ≥6 metastases, 99% received prior systemic anticancer therapy (Table). 143 (79%) completed 5 or 6 Ra-223 injections. Subsequent anticancer therapies included docetaxel (95%), enzalutamide (25%), cabazitaxel (24%), abiraterone (12%), lutetium-177-prostate-specific membrane antigen (4%), and sipuleucel-T (1%). During/up to 30 days after taxane therapy, 15 pts (8%) had grade 3/4 hematologic adverse events: anemia (erythropenia) (n = 11, 6%), neutropenia (n = 3, 2%), and thrombocytopenia (n = 2, 1%). Median OS was 24.3 (95% CI: 20.9–27.5) months from Ra-223 initiation and 11.8 (95% CI: 10.6–14.1) months from subsequent taxane initiation. Conclusions: In this cohort where Ra-223 was integrated prior to taxane therapy, most pts received multiple subsequent anticancer therapies. It appears that sequencing of multiple treatment modalities with different mechanisms of action may contribute to improved OS. Taxane therapy in routine clinical practice in pts previously treated with Ra-223 had acceptable hematologic safety/tolerability profiles. Clinical trial information: NCT02141438 . [Table: see text]

698 Background: Several ICIs are approved for use in first and subsequent lines of therapy in mRCC, either alone or in combination with another ICI or targeted therapy. There is limited data on whether patients who receive an ICI derive benefit from a subsequent line of ICI-based therapy. Methods: We reviewed mRCC patients at our institution who were treated with an ICI between 2009 and 2018 having previously received ICI-based therapy. The primary outcomes of interest were radiologic response and time to treatment failure (TTF) on ICI retreatment. Immune-related adverse events (irAEs) were graded using CTCAE v5.0. Results: A total of 31 patients were retreated with an ICI, either alone (n=15) or in combination with another ICI (n=8), tyrosine kinase inhibitor (n=2) or investigational agent (n=6); reasons for discontinuation of prior ICI were disease progression (n=23), toxicity (n=7) and study completion (n=1). Median age at the start of first-line therapy was 62 years and the majority of patients (n=30, 90%) had International Metastatic RCC Database Consortium intermediate- or poor-risk disease; median follow-up was 3.4 years (95% CI 2.5-4.6). The median number of lines of therapy received prior to ICI retreatment was 3 (range 1-7). Of 27 evaluable patients, 3 (11%) had a partial response (PR) to ICI retreatment (2 with ICI-ICI combination therapy and 1 with single-agent ICI), 13 (48%) had stable disease, and 11 (41%) had progressive disease (PD) as their best response; of the 3 with a PR, 2 had a PR and 1 had SD to prior ICI therapy. In comparison, 13 of 31 patients (42%) had a PR or better to first ICI-based therapy, with 4 (13%) having PD. Median TTF on ICI retreatment was 3.2 months (95% CI 1.8-5.2), compared to 8.2 months (3.3-10.0) on prior ICI. 19 patients experienced an irAE with ICI retreatment, with 8 (26%) suffering a grade 3 or higher irAE. Conclusions: Retreatment with ICIs after prior therapy with an ICI was relatively safe but demonstrated limited efficacy. Additional data, ideally from prospective studies, assessing outcomes of retreating patients with ICIs are needed to determine whether using different ICIs in sequence has a role in treatment of mRCC.

30 Background: BM in patients (pts) with CRPC are associated with shorter overall survival (OS) and higher costs. BRI zoledronic acid and denosumab are frequently used to prevent skeletal-related events (SRE) in pts with CRPC and BM. AA is the most common 1st line therapy for men with metastatic CRPC. We aimed to assess the impact of BRI on OS and time to first SRE (ttSRE) of pts receiving 1st line treatment AA for CRPC with BM. Methods: A retrospective cohort of pts starting AA as 1st line therapy for CRPC with BM between 2013-2016 was identified through 8 hospitals’ IRB approved registries. Pts were classified by use of concomitant BRI and subgrouped by volume of disease (per E3805 definition) at AA start. Kaplan-Meier method and Cox models were used to assess OS and ttSRE with hazard ratio (HR) estimates (95% CI). Results: Of the 745 pts included (543 deaths), 529 (71.0%) had AA alone and 216 (29.0%) AA+BRI. Median follow-up was 23.5 months. Pts receiving concomitant BRI showed a significantly longer OS and a 35% reduced risk of death compared to AA alone (HR=0.65; 95% CI, 0.54-0.79; P5, PS 0 vs. ≥1, and PSA are independently associated with longer OS. Conclusions: The addition of BRI to 1st line AA for CRPC men with BM was associated with improved OS, particularly in HV, and worsened ttSRE, more evident in LV. These data suggest a potentially different impact of concomitant BRI on HV vs. LV, which could affect clinical decision making.[Table: see text]

171 Background: IL-8 (CXCL-8) is a cytokine produced by leukocytes and tumor cells, which promotes inflammation, angiogenesis, anti-apoptosis, and metastasis. Our published training set identified serum IL-8 levels drawn within 28 days of androgen deprivation therapy (ADT) initiation as prognostic for shorter overall survival (OS). We sought to validate this finding using samples from CHAARTED patients treated with ADT +/- docetaxel for mHSPC. Methods: We assayed serum samples drawn ≤28 days from ADT initiation from 233 patients using the same Mesoscale multiplex ELISA assay as the training set. Multivariable Cox proportional hazards models adjusted for ECOG performance status, disease volume, and prior local therapy (radiation/prostatectomy vs. none) with IL-8 as continuous and binary variables. The training median 9.3 pg/mL was the a priori binary cutpoint. Fixed-effects meta-analysis of the training and validation sets was performed. Results: Higher IL-8 levels were prognostic for shorter OS and time to CRPC for ADT alone and ADT + docetaxel (Table) and were independent of disease volume or docetaxel use. Meta-analysis of binary IL-8 levels >9.3pg/mL from patients treated with ADT alone (training + validation cohorts) was prognostic for poorer OS (HR 1.8, 95% CI: 1.2-2.7, p= 0.007) and shorter time to CRPC (HR: 1.4, 95%CI: 0.99-1.9, p=0.057). The validation cohort’s baseline median IL-8 level was similar at 10 pg/mL. Conclusions: Higher IL-8 is prognostic for shorter time to CRPC and OS independent of docetaxel use, disease volume and presentation with metachronous or de novo metastatic disease. These findings support targeting IL-8 as a strategy to improve mHSPC outcomes.[Table: see text]

681 Background: Patients (pts) with RCC and oligometastatic pancreas metastases are treated with pancreatectomy, stereotactic body radiation therapy (SBRT), or systemic therapy. The optimal approach is not clear. We aimed to evaluate the comparative efficacy of the modalities in terms of progression-free survival (PFS) and overall survival (OS). Methods: This IRB-approved, multi-institutional, retrospective study evaluated pts with pancreatic-only RCC metastasis without concurrent metastases elsewhere. Data on pt demographics, tumor characteristics, treatment, and outcomes were collected. PFS and OS in pts treated with pancreatectomy vs. systemic therapy were compared by log rank tests. Results: Fifty-one pts from 9 institutions were included. All had clear cell RCC; 50 pts had nephrectomy; 30 pts (58.8%) and 18 pts (35.3%) had IMDC favorable and intermediate risk, respectively. Median time from RCC diagnosis to oligometastatic disease was 120 months (mo) (range: 0, 175). As initial treatment, 23 (45%) pts had pancreatectomy (mostly partial); 25 (49%) had systemic therapy (VEGFR TKI and/or immunotherapy); 1 had SBRT; 2 had other treatments. Too few pts had SBRT for comparison. With a median follow-up of 25 mo (2, 68), median PFS for the population was 25 mo (17, 42 95% CI). Median PFS was 36 mo (8, 43 95% CI) for surgery pts and 22 mo (17, NR 95% CI) for systemic therapy pts; this was not statistically significant (NS), p = 0.3. Median OS for the population was 121 mo (100, NR 95% CI). With a median follow-up of 51 mo (2, 217), OS was 121 mo (100, NR 95% CI) for surgery pts and not reached (64, NR 95% CI) for systemic therapy pts; NS, p = 0.52. Conclusions: In this retrospective series, RCC pts with oligometastatic pancreatic-only disease had similar PFS and OS outcomes from initial pancreatectomy or systemic therapy. RCC pts with pancreas-only metastases represent a unique patient population and studies informing the underlying biology are needed to optimize clinical management.

452 Background: Prior research suggests that systemic antibiotic (ABX) exposure may impact gut microbiome and potentially result in suboptimal immune checkpoint inhibitor (ICI) treatment outcomes. A recent real-world analysis demonstrated that cumulative ABX exposure was associated with poorer outcomes across multiple tumor types independent of known prognostic clinical factors. Given the paucity of real-world data, we set out to evaluate the association of prior ABX exposure with ICI treatment outcomes among patients with metastatic urothelial cancer (mUC). Methods: This was a retrospective analysis using Truven Health MarketScan Commercial, Medicare Supplemental, and Coordination of Benefits (Medicare) databases. Patients with mUC, ≥18 years old, who received first line (1L) ICI therapy between 1/1/2016 and 12/31/2018 were analyzed. Prior ABX exposure was defined as any use in the 90 days prior to 1L ICI therapy initiation. Time to treatment discontinuation was used as a proxy to quantify ICI treatment outcomes. Results: Among the 304 ICI treated patients, 128 (42%) had ABX exposure within 90 days prior to ICI initiation. Statistically significant differences in baseline co-morbidities and urinary tract infections (36% vs. 17%; p

129 Background: Adenosine, generated by the ectonucleotidase CD73, mediates immunosuppression within the tumor microenvironment by triggering adenosine 2A receptors (A2AR) on immune cells. Tumor CD73 expression may be prognostic in prostate cancer. We evaluated the clinical activity of adenosine blockade using A2AR antagonist, ciforadenant, with or without the anti-PDL-1 antibody, atezolizumab (atezo), in advanced mCRPC patients (pts) in an ongoing phase 1 trial. Methods: Eligibility required measurable disease and up to 5 prior systemic therapies. Prior anti- PD-(L)1 was allowed. Ciforadenant was administered orally BID as monotherapy at 50-200mg or 100mg in combination with atezo 840mg IV Q 2 weeks (wks). Safety and efficacy were evaluated by CTCAE4, RECIST1.1 and PCWG2. Serum was obtained for measurement of cytokines. Results: Of 33 enrolled pts, 10 received ciforadenant monotherapy and 23 in combination with atezo. As of 10/21/19, 14 pts are evaluable for response and described further. Median prior therapies is 3 (range 2-6) with median follow-up 10.8 (4-33) wks. Metastatic burden included 4 node only, 2 bone plus node, and 8 visceral metastases. Five pts experienced tumor regression: 2 ciforadenant monotherapy (tumor reductions 12%, 17%); and 3 combination (tumor reductions 4%, 27%, 42%). The pt with a partial response had PSA decline from 98 ng/mL to

740 Background: Metastatic vRCC are aggressive tumors with poor prognosis. Our phase 2 trial of AB in vRCC showed a response rate of 33%. We investigated on-therapy changes in circulating immune cells and cytokines and their association with outcomes. Methods: Blood was collected at baseline (C1D1) and on-therapy (C3D1). Peripheral blood mononuclear cells were analyzed for cell type, expression of immune checkpoints, markers of activation, proliferation and function using flow cytometry; circulating cytokines by multiplex immunoassay. Relationship with progression-free (PFS) and overall survival (OS) was assessed by cox regression models. Results: Baseline and on-therapy samples were collected from all 60 patients. High baseline levels of immunosuppressive cytokines IL1α, IL6, CCL4 and IL13, as well as high baseline levels of CD4+ lymphocytes expressing CD69, were associated with inferior PFS and OS (Table). However, a decline in these markers on-therapy was not predictive of outcomes. On-therapy assessments showed an increase in the IFN-γ inducible cytokine CXCL10 (p

4571 Background: RCC and UC pts with clinically significant AD were generally excluded from CPI trials due to potential AD exacerbation. Thus, the safety and clinical activity of CPI in AD pts are not well characterized. Methods: We retrospectively collected data from RCC and UC pts with AD treated with CPI at 9 centers. Adverse events (AEs) were assessed using CTCAEv5 criteria. Objective response rate (ORR) was assessed by RECIST principles. Overall survival (OS) was estimated by Kaplan Meier. Results: Of 103 pts (57 RCC & 46 UC) with a broad spectrum of AD such as psoriasis (22%), thyroiditis (20%), rheumatoid arthritis (13%), polymyalgia rheumatica (8%), inflammatory bowel disease (6%), multiple sclerosis (3%) and lupus (3%), most received CPI as 1st or 2nd line (77% RCC, 93% UC) and anti-PD-1/L1 monotherapy (65% RCC, 98% UC). At CPI start, 36 had clinically active AD (all grade 1-2) and 4(11%) requiring systemic immunosuppression. AD exacerbations occurred in 37% (n = 38); most frequent: arthritis (12% RCC, 24% UC), rash (11% RCC, 9% UC). New onset immune related (ir) AEs occurred in 36% (n = 37); most frequent: colitis (12% RCC, 4% UC), rash (11% RCC, 9% UC), hypothyroid (each 7%), nephritis (7% UC). Table details timing and management. Median followup was 12.5 (1-52) mos for RCC and 14.5 (1-53) mos for UC. Median time on CPI was 6 mos (1-36) RCC and 4 mos (0.5-40) UC. At data cutoff, 39 RCC & 36 UC had discontinued CPI; 16% for toxicity. ORR was 31% for RCC and 35% for UC. 1 yr-OS rate was 74% (95%CI 58-84) for RCC and 60% (95%CI 43-74) for UC. Conclusions: AD exacerbations or new irAEs occurred in 37% and 36% respectively and were generally manageable. Only a minority required CPI cessation due to toxicity. [Table: see text]

e16038 Background: In other cancers such as metastatic lung and melanoma, pts who experience irAEs may have a greater degree of clinical benefit. We sought to evaluate whether the development of irAEs correlates with clinical benefits in mUC pts. Methods: We identified mUC pts who received PD-1/L1 inhibitors at the Dana-Farber Cancer Institute (DFCI). The severity of irAEs was graded using CTCAE v.5.0. Clinical benefit was defined as any objective reduction (complete- or partial- response) in tumor burden. Fisher’s exact test was used to evaluate for differences in the proportion of pts experiencing clinical benefit between pts experiencing an irAEs within 90 days after starting therapy and those who did not experience an irAEs. The log-rank test assessed differences in progression-free survival (PFS) and overall survival (OS) between cohorts. Results: A total of 199 mUC pts were enrolled between July 2013 and October 2018 [median (range) age, 69.6 (26.6-89.0) years; 141 men (70.9%), 58 women (29.1%)]. 114 (57.3%) pts were treated with anti-PD-1 and 85 (42.7%) with anti-PD-L1. irAEs were observed in 67 pts (33.7%), of which 34 (17.1%) < 90 days from start of therapy. Common irAEs included 20 (29.9%) hypothyroidism, 17 (25.4%) colitis, 12 (17.9%) rash/pruritus, and 10 (14.9%) transaminitis. Grade ≥3 irAEs were observed in 14 pts (20.9%).13 (40.6%) pts with irAE < 90 days experienced clinical benefit compared with 21 (17.8%) of pts with no irAE (p-value = 0.008). No difference (p = 0.26 and 0.18) was observed for either PFS (6-month PFS = 19.2% vs 35.3% for no irAE and irAE < 90 days) or OS (1-year OS = 41.8% vs 57.7% for no irAE vs irAE < 90 days). Conclusions: The development of irAEs within 90 days from starting therapy in mUC pts may herald clinical benefit in pts with mUC. Further evaluation of this potential relationship in a large prospective study is warranted.

TPS4597 Background: The anti-PD-1 antibody nivo improves overall survival (OS) in metastatic RCC and is well tolerated. There is no standard adjuvant (adjuv) systemic therapy that increases OS over surgery alone for non-metastatic RCC. Priming the immune system prior to surgery with anti-PD-1 has shown an OS benefit compared to a pure adjuv approach in mouse solid tumor models. Multiple ph 2 studies in bladder, lung and breast cancers have shown remarkable pathologic responses with neoadjuvant (neoadj) PD-1 blockade. Two ongoing ph 2 studies of perioperative nivo in M0 RCC patients are showing preliminary feasibility and safety with no surgical delays (NCT02575222; NCT02595918). PROSPER RCC (NCT03055013) aims to improve clinical outcomes by priming the immune system prior to nephrectomy with neoadj nivo and continued engagement with adjuv blockade in patients with high risk RCC compared to surgery alone. Methods: This global, unblinded, phase 3 National Clinical Trials Network study is currently accruing patients with clinical stage ≥T2 or TanyN+ RCC of any histology planned for nephrectomy. Oligometastases are permitted if can be rendered NED. We amended the study to enhance accrual and patient quality of life by changing nivo dosing to 480mg q4 wks and requiring baseline tumor biopsy only in the nivo arm. The investigational arm receives 1 dose of nivo prior to surgery followed by 9 adjuv doses. The control arm undergoes standard nephrectomy followed by observation. Randomized patients are stratified by clinical T stage, node positivity, and M stage. Accrual of 805 patients provides 84.2% power to detect a 14.4% absolute benefit in recurrence-free survival (RFS) at 5 years assuming the ASSURE historical control of ~56% to 70% (HR = 0.70). The study is powered to evaluate a significant increase in OS (HR 0.67). Critical perioperative therapy considerations such as safety, feasibility, and quality of life endpoints have been integrated. PROSPER RCC embeds a wealth of translational work aimed at investigating the impact of the baseline immune milieu, the changes induced by neoadjuvant anti-PD-1 priming, and how both may predict clinical outcomes. Clinical trial information: NCT03055013.

4583 Background: NccRCC and ccRCCsd are aggressive tumors associated with poor prognosis and response to therapy. Combination strategies co-targeting VEGF signaling and inhibitory immune checkpoints are highly active in clear-cell renal cell carcinoma, but data is lacking in NccRCC and ccRCCsd. We conducted a multicenter, open-label, single arm phase II trial of atezolizumab plus bevacizumab in NccRCC and ccRCCsd. Methods: Patients with NccRCC and ccRCCsd ( > 20% sarcomatoid differentiation), and ECOG performance status of 0-2 were eligible. Prior systemic treatment was allowed with the exception of prior PD-1/PD-L1-directed therapy. Atezolizumab 1200mg and bevacizumab 15mg/kg were administered every 3 weeks until progression, unacceptable toxicity, or patient withdrawal. Primary endpoint was objective response rate (ORR) per RECIST 1.1. Exploratory biomarker analyses included PD-L1 expression on tumor (TC) and immune cells (IC), and spatial analysis of the immune infiltrate. Results: Sixty patients received at least 1 cycle of treatment, among whom 56 were evaluable for response (17 ccRCCsd and 39 NccRCC). ORR was 34% in the overall population, 53% in ccRCCsd and 26% in NccRCC. Median progression-free survival was 8.4 months (95%CI, 6.9-16.5). Baseline tumor tissue was available for 36 patients. TC PD-L1 expression ≥1% was associated with improved ORR (9/14, 64%) compared to patients with PD-L1 expression < 1% (4/20, 20%). Patients with TC PD-L1 expression ≥1% who experienced progressive disease as best response had shorter average distance between tumor cells and nearest neighboring immune cells at baseline. Further analysis of the immune tumor microenvironment on an expanded cohort, including IC PD-L1 expression and correlation with clinical outcomes, is ongoing and will be updated. Conclusions: The combination of atezolizumab plus bevacizumab is active in NccRCC and ccRCCsd. Candidate predictive biomarkers include PD-L1 expression in TC and topological analysis of the immune infiltrate. Clinical trial information: NCT02724878.

4548 Background: Checkpoint inhibitors have improved outcomes in UC. However, response rates are low and additional mechanisms of immune evasion need to be ascertained. CD73 (encoded by NT5E) converts extracellular AMP to adenosine, which exerts an immunosuppressive effect in the tumor microenvironment by inhibiting infiltrating T and NK cells. Utilizing The Cancer Genome Atlas (TCGA) bladder cancer dataset, we evaluated correlations between NT5E expression and the immune milieu in UC. Methods: RNA-seq data from 411 primary UC tumor samples were obtained from the TCGA. Patients were split into low, intermediate, and high NT5E expression groups (≤ -1, -1 to 1 and ≥1 standard deviation from the overall mean). A tumor inflammation signature (TIS) reflecting an inflamed tumor phenotype was calculated based on the averaged expression of 18 previously validated genes (Ayers et al, 2017). NT5E expression was compared between tumors with high and low TIS scores and among the TCGA molecular subtypes. Abundance of infiltrating immune cell subsets was estimated based on expression of previously identified 782 immune metagenes and compared between NT5E expression groups (Charoentong et al, 2017). The Mann-Whitney U test assessed statistical significance, and the Bonferroni correction was used to control for false discovery rate. Results: NT5E expression was significantly higher in tumors with a high TIS score compared to those with low TIS score (P

4582 Background: CD73 or ecto-5'-nucleotidase mediates the de-phosphorylation of adenosine monophosphate to adenosine which promotes immunosuppression in the tumor microenvironment. Its prognostic significance in localized RCC has not been well characterized. Methods: We assessed CD73 protein expression using immunohistochemistry (Cell Signaling Technology, D7F9A, 1:25) on tissue microarrays (TMAs) containing tumor tissue from patients with pT1-4,N0-1, M0 RCC who underwent nephrectomy.CD73 expression was quantified using a combined score (CS: intensity x percentage of cells staining positive). CD73 positivity was defined as any CD73 expression on tumor cells (CS > 0). Patients were categorized into CD73 negative (CS = 0), low ( < median CS) and high (≥ median CS) groups. Clinical data was collected retrospectively.Baseline patient characteristics were compared between CD73 negative and positive (high + low) groups using the Cochran-Armitage trend test. Multivariable Cox regression evaluated associations of CD73 expression with disease-free and overall survival (DFS, OS) after adjusting for other baseline prognostic variables. Results: Of the 112 patients included, clear cell was the most common histology (70%) followed by chromophobe (12.5%), and papillary (12%) RCC. CD73 expression (CS > 0) was noted in 22% (n = 25) of tumors and was associated with more advanced stage (T3-4/N+: 37.5% vs. 25%; p = 0.02) and a trend towards higher nuclear grade (≥3: 48% vs. 35%; p = 0.07). Median follow-up from nephrectomy was 9.7 yrs. In multivariable analysis adjusting for nuclear grade ( < 3 vs. ≥3) and stage (T1-2, N0 vs. T3-4/N+), tumors with high CD73 expression had significantly worse DFS and OS (Table). Conclusions: CD73 expression was found in 22% of RCC patients and was associated with adverse pathologic features and poor prognosis independent of tumor stage and histologic grade. Our results provide strong rationale for further investigation of the CD73/adenosine pathway as a therapeutic target in RCC. [Table: see text]

4514 Background: S/R mRCC are poorly characterized rapidly progressing tumors associated with poor prognosis. Although conventional therapies are less effective for these tumors, emerging data suggests that ICIs may be especially effective. Our aim was to characterize the genomic alterations (GA) in S/R mRCC tumors and evaluate their response to ICIs. Methods: We retrospectively compared the activity of first-line ICIs to non-ICI-based therapies for S/R mRCC patients (pts) treated at DFCI and analyzed sequencing data from an NGS panel (275-447 genes) on a subset of these patients (matched by histology to non-S/R mRCC). For S/R mRCC pts treated with ICI vs non-ICI therapies, overall survival (OS) and time to treatment failure (TTF) were compared by Cox regression and objective response rate (ORR) by logistic regression. GA frequencies were compared by Fisher’s test and tumor mutational burden (TMB) by Mann Whitney U between S/R and non-S/R mRCC. Results were considered statistically significant if p < 0.05 or q < 0.10. Results: 125 S/R mRCC pts were included (88 S, 23 R, 14 S&R) among which 103 were clear cell and 48 had sequencing data. GA in BAP1 were significantly more frequent in S/R vs non-S/R (25% vs 4.3%; q = 0.096) while other GA had similar frequencies and TMB (median [IQR]) was similar (7.2 [5.2-8.4] vs 6.8 [5.3-9.1] mut/Mb; p = 0.98). Median follow-up was 35.4 (95% CI = 24.9 – 46.0) months (m). On multivariable analysis, S/R mRCC pts treated with ICI had significantly better clinical outcomes (Table). Conclusions: Pts with S/R mRCC have a higher frequency of BAP1 GA and better outcomes on ICIs compared to non-ICI-based therapies. Future studies should determine the molecular mechanisms underlying the improved response to ICIs in S/R mRCC. [Table: see text]

e16508 Background: BM in patients (pts) with CRPC correlate with higher mortality and costs. BRI zoledronic acid and denosumab are frequently used for the prevention of skeletal-related events (SRE) in pts with CRPC and BM. AA is the most common 1st line treatment for men with metastatic CRPC. We sought to evaluate the impact of BRI on time to first SRE (ttSRE) and OS of pts receiving 1st line therapy AA for CRPC with BM. Methods: We identified a cohort of men starting AA as 1st line therapy for CRPC with BM between 2013-2015 from 7 hospitals’ IRB approved registries. Pts were grouped by use of concomitant BRI and subgrouped by volume of disease (per E3805 definition) at AA start. The endpoints were OS, defined as time from AA start to death or last follow-up visit, and ttSRE. Results: Of the 338 pts included, 256 (76%) received AA alone and 82 (24%) AA+BRI. ECOG PS (PS) was ≥1 for 178 pts (52.7%). No statistically significant difference in ttSRE was found between the 2 cohorts [see Table]. Median follow-up for OS was 25.6 months. Pts receiving concomitant BRI showed a significantly longer OS and a 36% decreased risk of death compared to AA alone (HR = 0.64; 95% CI, 0.64 0.46-0.91; p = 0.012). Notably, OS in the AA alone group was shorter than commonly described. The OS benefit with BRI was greater for men with high volume disease (HV) (HR = 0.42; 95% CI, 0.25-0.71; p = 0.001). On MVA, BRI vs. no BRI, low volume of disease vs. HV, PS 0 vs. ≥1, baseline VAS pain ≤3 vs. > 5, and baseline PSA are independently associated with longer OS. Conclusions: Using a multicenter database, the addition of BRI to 1st line AA for CRPC men with BM and poor prognostic factors did not improve prevention of SRE. However, concomitant use of BRI and AA was associated with a significantly improved OS, particularly in HV. Further research to determine the driving factors is needed. [Table: see text]

154 Background: Immune checkpoint therapy has changed the treatment landscape for many malignancies. As these agents are rapidly being incorporated into clinical practice, a subset of patients appear to have an accelerated tumor growth rate (TGR) after treatment initiation, a phenomenon termed hyperprogressive disease (HPD). Methods: A systematic search of original articles published between January 2000 and June 2018 was performed using PubMed, Embase, and Web of Science. Studies were eligible for inclusion if they reported on HPD in adult patients with advanced solid malignancies treated with checkpoint inhibitors and provided a specific HPD definition. Abstracts from ASCO, ESMO, and IASLC meetings were also included. Prespecified outcomes of interest included criteria used to define HPD, incidence of HPD, and association between HPD with baseline clinical or molecular characteristics and survival outcomes. Results: Of 700 studies screened, 12 retrospective studies were eligible, 3 of which evaluated patients treated in the context of clinical trials. Lung cancer and melanoma were the most common of the diverse histologies represented. The most common treatment was anti-PD1/PD-L1 monotherapy. A variety of criteria were utilized to define HPD, including TGR, RECIST progression, and time to treatment failure. HPD incidence ranged from 2% to 29%, and the only clinical factor significantly associated with HPD in more than one study was increased age. MDM2/4 and EGFR alterations appear to be associated with a higher incidence of HPD. HPD was significantly associated with poor overall survival in 3 studies. Conclusions: It remains unclear whether HPD reflects worsening disease caused by immunotherapy or is simply a manifestation of unfavorable, treatment unresponsive disease biology. There is a need to establish a consensus definition for HPD, and to examine HPD in prospective cohorts with appropriate comparison arms.

66 Background: Patients with mRCC with S/R components tend to have a poor prognosis with few therapeutic options available. Recent data suggest that immune checkpoint inhibitor (ICI)-based therapies may be especially effective for these patients. Our aim was to evaluate the efficacy of ICI-based therapies in patients with S/R mRCC. Methods: We retrospectively assessed objective response rate (ORR), progression free survival (PFS) & overall survival (OS) of patients with S/R mRCC treated at our institution with first-line ICI-based therapies and compared these to those of patients treated with first-line non-ICI-based therapies. Univariable and multivariable (adjusted for IMDC group) Cox and logistic regressions were performed. Results: 92 patients (70 S, 9 R, and 13 S&R) patients were included, of which 74 with a clear-cell component. For all patients (regardless of therapy), 74 (80.4%) patients experienced a PFS event (progression or death) and 52 (56.5%) died at 25.3 months (m) median follow-up. Overall median PFS was 5.3 m (95% CI= 3.4–7.2) and 24 m OS rate was 39.5% (27.4–51.7). Out of 78 patients in whom response was evaluable, ORR was 30.8% (20.4–41.2). Patients treated with ICI-based therapies had significantly better ORR, PFS, and OS on multivariable analysis (table). Conclusions: mRCC patients with S/R components have significantly better ORR, PFS, and OS with first-line ICI-based compared to non-ICI-based therapies. These data support the use of ICI-based therapies for patients with S/R mRCC. [Table: see text]

548 Background: NccRCC and sccRCC have historically been underrepresented in clinical trials. Even with targeted therapy, most patients have inferior survival compared to clear cell renal cell carcinoma. The combination of atezolizumab and bevacizumab has demonstrated safety and efficacy in ccRCC. In this multicenter, phase II, open-label, single arm trial we evaluate the efficacy of atezolizumab and bevacizumab in patients with nccRCC and sccRCC with >20% sarcomatoid differentiation. Methods: Eligible patients had an ECOG performance status of 0-2 and may have received prior therapy. Prior PD-1/PD-L1 therapy was not allowed. Patients underwent a mandatory baseline biopsy and subsequently received atezolizumab 120 mg and bevacizumab 15 mg/kg intravenously every 3 weeks. Patients remained on therapy until radiographic progression, unacceptable adverse events, or withdrawal. The primary end point was overall response rate (ORR) as determined by RECIST version 1.1. Results: 65 patients were enrolled of whom 52 had ≥1 response assessment and were included in this analysis. 36 patients had nccRCC (papillary n=14, chromophobe n=8, unclassified RCC n=3, collecting duct n=3, translocation n=3, other n=5), and 16 patients had sccRCC. 17 patients received prior systemic therapy, 16 of whom had nccRCC. The ORR was 31% in the overall cohort (Table 1). 10 patients (19%) developed grade 3 treatment-related adverse events (AEs), half of which were immune-related. There were no grade 4-5 AEs. Conclusions: In this study, we show that therapy with atezolizumab and bevacizumab was safe and demonstrated anti-tumor activity in nccRCC and sccRCC. Further analyses will report ORR by histologic subtype and PD-L1 expression status. Analysis of tissue and blood-based biomarkers of response are ongoing. Clinical trial information: NCT02724878. [Table: see text]

566 Background: An elevated BMI is associated with improved survival in mRCC patients treated with oral targeted therapies (TT); however, this relationship in the contemporary treatment landscape is unknown. We investigated the effect of BMI on outcomes in mRCC patients treated with PD-1/PD-L1 ICB. Methods: We analyzed 147 patients with mRCC who received ICB alone or in combination with VEGF or other therapies. The association of BMI with objective response rate (ORR), progression-free survival (PFS), and overall survival (OS) was evaluated using logistic and Cox regression models, adjusted for known prognostic factors including International Metastatic RCC Database Consortium (IMDC) risk groups, line of therapy, ECOG (0 vs ≥1), and histology. Results: Median follow up was 25.5 (4.3-78.6) months (mos). Median time on ICB was 5.1 ( < 0.1-69.7) mos. Overall, most patients were male (71%), had clear cell histology (85%), and were intermediate risk (60%). 43% received first-line ICB and 45% received ICB in combination therapy (37% with VEGF inhibition, 8% with other therapies). At ICB initiation, 46 (31%) patients were considered underweight/normal weight (BMI < 25 kg/m2), 56 (38%) overweight (BMI 25-30 kg/m2), and 45 (31%) obese (BMI > 30 kg/m2). Patients with high BMI (≥ 25) had improved OS (median 34.3 vs 16.7 mos, 2-yr OS 61 vs 42%, p = 0.016) compared to those with low BMI ( < 25), with an adjusted hazard ratio (HR) = 0.74 (0.44-1.26). ORR (33 vs 28%, p = 0.7) and PFS (median 8.2 vs 5.9 mos, p = 0.4) did not statistically differ. Patients who experienced a BMI change from ≥ 25 at start of first-line TT to < 25 at start of subsequent-line ICB displayed shorter OS (adjusted HR = 2.25 (0.94-5.35)) compared to those with no change. Conclusions: High BMI appears to be associated with improved OS in mRCC patients treated with ICB. This contemporary data is consistent with the “obesity paradox” demonstrated in the TT era. Validation with the IMDC database and examination of underlying mechanisms are ongoing.

435 Background: The impact of concurrent medications (meds) on outcomes with PD1/PD-L1 inhibitors in metastatic urothelial carcinoma (mUC) is unclear. We investigated whether candidate concurrent meds (NSAIDS [N], metformin [M], antibiotics [A], statins [S] and corticosteroids [C]) have an association with outcomes in mUC patients (pts) receiving a PD1/PD-L1 inhibitor. We hypothesized that A and C compromise outcomes, while N, M and S improve outcomes. Methods: Data from mUC pts who received PD1/PD-L1 inhibitors at the Dana-Farber Cancer Institute (DFCI) was obtained. The concurrent medication was required to be administered within 1 month before starting to anytime during PD1/PD-L1 inhibitor therapy. A Cox regression analysis was done to study the association of variables with response and survival. Results: Data was available for 101 pts with mUC who received atezolizumab [n = 52], pembrolizumab [n = 39], nivolumab [n = 9] and durvalumab [n = 1]. Prior platinum had been administered in 74 pts (73.2%), 25 were chemonaive (24.8%) and prior therapy status was unknown in 2 pts (2%). The concurrent meds were N (n = 30), M (n = 7), A (n = 26), S (n = 33) and C (n = 12). The median survival was 57.9 weeks. Response was seen in 26 pts [25.7%]. A was associated with a lower probability of response (11.5%) than those not on A (30.7%), and worse survival (HR = 1.93, 95% CI 1.93 – 3.42, P = 0.024). Pts who received neither A nor C, one of them or both had a response rate (RR) of 30.6%, 20% and 0%, and median survival of 65.3, 53.1 and 14.9 weeks, respectively (HR = 3.02, 95% CI = 1.34-6.83, p = 0.027). Pts who did not receive N, M and S (n = 52) exhibited a median OS of 39.6 weeks, while those who received ≥1 of these meds (n = 49) exhibited a median survival of 160.3 weeks (p = NS). The study is limited by the retrospective design and modest sample size. Conclusions: In this hypothesis-generating study, concurrent antibiotics or corticosteroids compromised outcomes in mUC pts receiving a PD1/PD-L1 inhibitor and receiving both further compromised outcomes. The numerically higher survival with concurrent N, M or S did not attain statistical significance, but requires further study in larger datasets.

652 Background: CPI can induce a range of irAEs with various degrees of severity. Clinical experience with retreatment following clinically significant irAEs is growing. Methods: We retrospectively reviewed mRCC patients treated with CPI-based regimens who had >1 week therapy delay for irAEs at Dana Farber and Emory to characterize the safety and outcomes of retreatment. Toxicity was graded by CTCAEv5. Patients were divided into retreatment (R) and discontinuation (D) cohorts. Patient characteristics were compared by Fisher’s exact test or Wilcoxon’s rank sum test. ORR, OS and PFS were assessed descriptively. Results: Of 339 treated with CPI, 53 (16%) had irAEs warranting treatment interruption: 24 (45%) in R and 29 (55%) in D. Median (med) time to drug interruption was 2.0 (40 mg: 29% vs 79%). The most common irAEs in R were transaminitis, pancreatitis and dermatitis with 7 gr 3/4 events (pancreatitis, colitis, adrenal insufficiency and dermatitis). Median drug break before retreatment was 1.2 (0.3-4.8) mo. Subsequent irAEs occurred in 42% (n=10) after restarting (8 new, 2 recurrent) with 2 discontinuations; 6/10 were gr 3 with no gr 4/5 events. Median interval to irAE recurrence after retreatment was 2.1 (0.5-3.3) mos. Retreatment resulted in 4 (17%) PRs in patients whose disease had not responded prior to irAE (med time to PR from R = 2 mos). Conclusions: Despite a significant rate of irAE recurrence, most patients with prior clinically significant irAEs can be safely retreated. Retreatment can induce responses in an appreciable proportion of patients with no response prior to irAE. Larger studies are warranted to confirm these findings. [Table: see text]

664 Background: Multiple clinical risk scores and gene expression models have predicted recurrence in localized ccRCC. However, few studies explored genomic alterations (GA) predicting recurrence. Methods: We assessed genomic and clinical correlates of disease-free survival (DFS) in surgically treated localized ccRCC using a targeted next generation sequencing (NGS) platform (Oncopanel/PROFILE) and publicly available NGS and clinical data from TCGA. Univariable and stepwise multivariable Cox regression models (stratified by database) were performed. Results: 478 patients (123 patients from our institution and 355 patients from TCGA) were included. 150 (31.4%) patients experienced a DFS event (recurrence or death) and 94 (19.7%) died at 3.1 years (yrs) of median follow-up. Median DFS was 6.3 (5.4-7.2) yrs and the 5-yr overall survival rate was 70.8% (64.9-76.7). On multivariable analysis, 4 clinical factors and mutations in 3 genes were significantly associated with recurrence (Table). Conclusions: Our study suggests that PTEN, BAP1 and KDM5C GA may improve on clinical factors for prediction of localized ccRCC recurrence. Further work is needed to determine if these GA could improve existing validated risk models. [Table: see text]

TPS339 Background: PC is a radiosensitive disease. PSMA is selectively overexpressed in advanced PC with upregulation by androgen receptor (AR) pathway dysregulation; limited expression exists in other organs. Prior studies of beta-emitting radiolabeled anti-PSMA antibody J591 demonstrated accurate targeting, efficacy with dose-response effect, and safety with predictable dose-limiting myelosuppression. Recent prospective trials have shown efficacy and safety of 177Lu-PSMA-617 in mCRPC. Studies demonstrate different binding sites of J591 and PSMA-617 and co-administration leads to non-competitive additive binding and delivery of payloads to PSMA+ cells. As the pharmacokinetics and biodistribution of the 2 agents is mostly non-overlapping (other than tumor) and given our prior dose-response data, we hypothesize that delivery of the combination will yield higher tumor delivery with less off-target toxicity (i.e. better responses without increased toxicity). Methods: Men with progressive mCRPC following at least 1 potent AR-targeted agent (e.g. abi/enza) and docetaxel (or unfit/refuse chemo) with metastatic disease on CT/MRI or bone scan without limit of # prior therapies (excluding bone-targeted beta-emitting radioisotopes) provided adequate organ function will be enrolled. Pre-treatment 88Ga-PSMA-11 will be performed, but results are not used for eligibility. Treatment includes fractionated 177Lu-J591 at a fixed moderate dose with escalating doses of 177Lu-PSMA-617 (7.4 – 18.5 GBq per cycle, fractionated D1 and D15) in a 3+3 dose-escalation study. Dose-limiting toxicity (DLT) is defined as attributable grade > 3 heme toxicity or grade > 2 non-heme toxicity. Following determination of recommended phase 2 dose (RP2D), the phase 2 portion will enroll in a 2-stage design. Primary endpoints: DLT and RP2D (ph 1) and PSA decline proportion (ph 2). Secondary endpoints include toxicity, radiographic response, PFS, rPFS, OS, CTC count changes. Correlatives include baseline/follow up PSMA imaging, whole body distribution of 177Lu, tissue and circulating genomic assessment, immunologic assessment, and patient reported outcomes (FACT-P and BPI-SF). Clinical trial information: NCT03545165.

653 Background: The pivotal clinical trials of CPI generally excluded pts with AD given concern for serious exacerbations. There is limited data on the safety and efficacy of CPI in this population. Methods: We conducted a retrospective single center analysis of RCC and UC pts treated with CPI. Pts were grouped by presence or absence of AD. We catalogued the incidence of new irAEs (CTCAEv4), AD exacerbations, objective response rate (ORR, RECIST 1.1), and overall survival (OS, Kaplan Meier). Competing risk models estimated cumulative incidences of irAEs and CPI discontinuation using 6 months (mos) as a benchmark. Results: 271 RCC and 220 UC pts were identified. Median followup was 21 mos for RCC and 13 mos for UC. 25 (9%) RCC and 27 (12%) UC had pre-existing AD; most commonly dermatologic and rheumatologic. A minority (3% RCC/4% UC) had clinically active AD requiring concurrent immunomodulators. AD exacerbations occurred in 8 (32%) RCC and 12 (44%) UC; median time to exacerbation was 86 (25-270) and 27 (8-300) days, respectively. Cumulative incidence of new irAEs at 6 mos was numerically higher in RCC AD pts (50% vs 37%). CPI discontinuation and clinical outcomes were similar among AD and non-AD pts (Table). Conclusions: CPI are active in advanced RCC and UC pts with pre-existing AD. Exacerbations and new irAEs should be expected, but AD pts can experience similar outcomes and rates of drug discontinuation. Larger scale investigations are warranted. [Table: see text]

Purpose Previous studies suggest that some common medications alter prostate-specific antigen (PSA) levels. It remains unclear whether these reported medication effects are due to clinicodemographic factors or concurrent use of other medications. We investigated the impact of individual and combinations of common medications on PSA in a large cross-sectional study of the United States population. Patients and Methods The study included men ≥ 40 years old without prostate cancer from the 2003 to 2004 and 2005 to 2006 cycles of the National Health and Nutrition Examination Survey (NHANES). Men with recent prostate manipulation, prostatitis, and those on hormone therapy were excluded. Weighted multivariate linear regression was performed on log-transformed total PSA to determine the effect of the 10 most commonly prescribed medication classes, adjusting for potential confounders including demographics, clinical characteristics, physical examination, laboratory studies, and duration of medication use. Results In total, 1,864 men met inclusion criteria. Nonsteroidal anti-inflammatory drug (NSAID; P = .03), statin (P = .01), and thiazide diuretic (P = .025) intake was inversely related to PSA levels. Five years of NSAID, statin, and thiazide diuretic use was associated with PSA levels lower by 6%, 13%, and 26%, respectively. The combination of statins and thiazide diuretics showed the greatest reduction in PSA levels: 36% after 5 years. Concurrent calcium channel blocker use minimizes or negates the inverse relationship of statin use and PSA level. Conclusion We found that men using NSAIDs, statins, and thiazide diuretics have reduced PSA levels by clinically relevant amounts. The impact of regularly consuming these common medications on prostate cancer screening is unknown.

4579Background: Cabo shows robust clinical activity in advanced clear cell RCC. nccRCC, a heterogeneous mix of diseases, have been underrepresented in clinical trials and effective systemic therapy...

5046Background: Recent data has highlighted the high frequency with which metastatic prostate cancer can progress either radiographically or symptomatically without a rise in PSA, so called PSA-Cli...

e17041Background: Sipuleucel-T (Provenge) received FDA approval in April 2010 for pts with asymptomatic or minimally symptomatic mCRPC based on prolongation of overall survival. However, its adopti...

TPS4600Background: CheckMate 025 established NIVO as a standard of care (SOC) option for advanced clear cell RCC (ccRCC) after prior antiangiogenic therapy. The combination of IPI plus NIVO is anti...

4539Background: Tumor mutational burden is associated with response to PD1/PD-L1 inhibitors in advanced UC. We explored the ability of copy-number variants (CNVs) and specific genomic alterations t...

4527Background: While the genetic characteristics of muscle-invasive bladder cancer have been studied in detail, there are few reports using the same genetic analysis platform to investigate differ...

233 Background: Ra-223, a targeted alpha therapy, extended survival and had a favorable safety profile at 3 years’ follow up in pts with mCRPC in the pivotal phase 3 ALSYMPCA trial. The maturing global, prospective, single-arm, observational REASSURE study, designed to evaluate long-term safety at 7 years’ follow up, enrolled pts with mCRPC with bone metastases planned to receive Ra-223. Methods: We performed a descriptive analysis of safety and BL characteristics of US pts according to prior or concomitant abiraterone/enzalutamide (abi/enza) or prior docetaxel/cabazitaxel (chemo) using data from the first planned interim analysis (pts receiving ≥1 Ra-223 dose; median follow up 8 mo). Results: 244 US pts were included; 80% had no prior chemo. Prior abi/enza and/or chemo pts had higher median BL PSA and were less likely to complete 5-6 doses (Table). Subgroups had similar median ALP, LDH and Hb. Overall, drug-related tx-emergent AEs occurred in 71 pts (29%) and serious AEs in 9 (3.7%). Most common AEs were diarrhea, fatigue and anemia. AE incidence was numerically higher in pts who received prior chemo and/or abi/enza. Clinical trial information: NCT02141438. Conclusions: To date, REASSURE has not revealed any new safety findings and most pts complete 5-6 Ra-223 doses in routine US clinical practice. Pts with prior tx lines had lower Ra-223 tx completion and higher AE incidence, likely reflecting greater disease burden, as evidenced by higher median BL PSA.[Table: see text]

195 Background: Quantification of circulating free DNA (cfDNA) has been proposed both as a prognostic biomarker and as a biomarker of response and resistance to therapy. However, dynamics of cfDNA derived from both tumor and non-cancerous tissues through dense longitudinal monitoring over the course of treatment with multiple therapeutic modalities have not been well described. Methods: CfDNA isolated from banked and prospectively collected plasma samples from pts with metastatic castration-resistant prostate cancer (mCRPC) was subject to ultra-low pass (~0.1x coverage) whole genome sequencing (ULP-WGS). The fraction of cfDNA derived from tumor rather than non-cancerous tissues (i.e. the tumor fraction, or TFx) in each sample was estimated using a novel computational tool called ichorCNA, and was correlated with clinical features and response to therapy. Results: 663 plasma samples from 140 pts were included (median 3 samples/pt; range 1-20). TFx was correlated with the number of bone metastases (median TFx = 0.014 with no bone mets, 0.047 with 1-3, 0.190 for 4+), and with the presence of distant visceral mets (p < 0.0001). In multi-variate analysis, TFx was positively correlated with Alk Phos (p = 0.0227) and negatively correlated with Hgb (p < 0.001), but was not correlated with PSA (p = 0.75). All new treatment starts leading to > = 30% PSA decline at > = 6 weeks were associated with a decline in TFx from baseline when baseline TFx was > 7% (median TFx change -92.2%, range -70.3% to -100%); however, TFx in patients being subsequently maintained on secondary hormonal therapy without radiographic progression was quite dynamic. The total yield of cfDNA derived from both tumor and non-cancerous tissues tracked closely with TFx, and did not markedly change with initiation of therapies expected to lead to generalized tissue damage, such as chemotherapy and radiation, over the time scales examined. Conclusions: Decline in TFx is a promising biomarker for initial response to therapy, but subsequent temporary rises in TFx do not necessarily indicate the cessation of clinical benefit from secondary hormonal therapies. Analysis of cfDNA may allow a window into short-term tumor dynamics not easily assayed through other methods.

TPS399 Background: PC is a radiosensitive disease. PSMA is overexpressed in advanced PC with upregulation by androgen receptor (AR) pathway dysregulation; limited expression exists in other organs. A series of sequential studies of beta-emitting radiolabeled anti-PSMA monoclonal antibody (mAb) J591 have demonstrated accurate targeting, efficacy with dose-response effect, and safety with predictable, reversible dose-limiting myelosuppression. Alpha emitters are significantly more potent with a shorter range than beta emitters. Though there is no direct tumor-targeting, the bone-targeting alpha emitter Ra223 is approved. Anecdotal reports of PSMA small molecule targeted alpha emitters have hinted at efficacy but are limited by xerostomia and in mouse models may lead to long-term renal damage. Intact mAb J591 has comparatively no to minimal distribution in salivary glands and kidneys. Preclinical studies demonstrated purity, immunoreactivity, and stability with efficacy in a xenograft model [AACR 2017]. Methods: Men with progressive mCRPC following at least 1 potent AR-targeted agent (e.g. abi/enza) and docetaxel (or unfit/refuse chemo) without limit of # prior therapies (excluding beta-emitting bone-targeted radioisotopes) provided adequate organ function will undergo imaging with 68Ga-PSMA-11 PET/CT followed by a single dose of 225Ac-J591. Single-subject cohorts will be enrolled until grade > 1 attributable toxicity, then transition to 3+3 design. Cohort 1 = 13.3 KBq/kg with planned escalation up to 93.3 KBq/kg of 225Ac with fixed 20 mg J591. Dose-limiting toxicity (DLT) is defined as attributable grade 4 heme toxicity or grade 3/4 non-heme toxicity. Planned cohort expansion will occur at recommended phase 2 dose (RP2D) in a 2-stage design. The primary endpoint is determination of DLT and RP2D. Secondary endpoints include toxicity, PSA decline rate, RECIST response, PFS, rPFS, OS, and patient reported outcomes (FACT-P and BPI-SF). Correlatives include baseline/follow up PSMA imaging, CTC count (CellSearch) changes, tissue and circulating genomic assessment, and immune studies. Enrollment began in October, 2017. Clinical trial information: NCT03276572.

643 Background: Agents targeting the PD-1 pathway have improved outcomes in RCC. CD73 may be an additional mechanism which tumors can exploit for immune evasion. It is regulated by hypoxia inducible factor (HIF) and converts AMP to adenosine. The resulting increase in extracellular adenosine can inhibit T-cell effector function. We evaluated CD73 expression in primary and metastatic tumor samples in patients with RCC. Methods: A commercial TMA (US Biomax) consisting of 31 primary clear cell RCC samples (2 with sarcomatoid features) with 8 matched and 1 unmatched metastases was used to assess CD73 expression. A genitourinary pathologist (OH) confirmed pathology and grade. Immunohistochemistry (IHC) was performed using a monoclonal anti-CD73 antibody (Cell Signaling; D7F9A). A combined score (CS: % of cells positive x intensity) was employed to quantify CD73 expression. Expression levels between matched primary and metastatic tissue was compared using Wilcoxon signed-rank test. Correlation of CD73 expression (CS > 0) in primary tumor with baseline clinical and pathologic characteristics was assessed using Kruskal-Wallis and Fisher's exact tests. Overall survival (OS) was estimated using the Kaplan-Meier method. Comparison between CD73 expression groups used log rank test. Results: CD73 expression was seen in 19% (n = 6) of primary and 66.7% (n = 6) of metastatic samples. Among matched primary and metastatic samples (n = 8 each), median CS was 25 (Q1-Q3:0-105) in metastatic samples while none of the corresponding primary samples demonstrated CD73 expression (median CS = 0, p = 0.062). CD73 expression in primary tumor samples did not correlate with baseline clinical or pathologic features. Five-year OS was 50% in patients who expressed CD73 in primary tumors compared to 84% in those who did not (p = 0.26). Conclusions: We observed a numerical increase in CD73 expression in metastatic tissue compared to primary RCC samples and worse 5 year OS. The small number of samples has limited statistical significance. Further studies are needed to examine the impact of CD73 expression on outcomes in RCC patients treated with currently approved checkpoint inhibitors and the investigational CD73 antagonists that are in early phases of development.

79 Background: Patients with high-risk PC have an increased risk of recurrence and mortality despite therapy. Abiraterone, a CYP17 inhibitor, and enzalutamide, a next generation anti-androgen, have demonstrated improved overall survival in metastatic PC. In this multicenter randomized phase II trial, we evaluate the impact of second generation hormone therapy on RP pathologic outcomes. Methods: Eligible patients had biopsy Gleason score ≥4+3=7, PSA >20 ng/mL or cT3 disease (by prostate MRI). Lymph node were require to be 20 ng/mL n=17 (23%)]. All patients completed 6 cycles followed by RP. Median PSA nadir was 0.03 and 0.02 ng/mL and time to nadir was 3.7 and 4.6 months in the APEL and EL arms, respectively. The combined pCR or MRD rate was 30% (n=15/50) in the APEL arm and 16% (n=4/25) in the EL arm. The response difference was 14% (80% CI -3%-30%, p=0.263). 15 patients (14 in APEL; 1 in EL) had grade 3 adverse events (AEs). The most common grade 3 AEs were hypertension (n=7) and ALT increase (n=5). No grade 4-5 AEs occurred. Conclusions: Neoadjuvant hormone therapy plus RP in men with high-risk PC resulted in favorable pathologic responses (≤5 mm residual tumor) in 16-30% with a trend towards improved pathologic outcomes with APEL and acceptable safety profile. Follow-up is necessary to evaluate the impact of therapy on recurrence rates. Clinical trial information: NCT02268175. [Table: see text]

465 Background: Bladder cancer patients (pts) are typically elderly with median age ~70 years. We hypothesized that younger pts (≤age 50) with muscle-invasive bladder cancer (MIBC) have distinct molecular features, including potential driver mutations that could serve as therapeutic targets. Methods: We used the MIBC TCGA cohort (n = 412, Cell 2017) analyzed by whole exome sequencing as a discovery cohort, and then confirmed observations using 356 pts with MIBC and high grade (HG) non-MIBC analyzed at Dana Farber Cancer Institute (DFCI) by the Oncopanel assay. Oncopanel assesses 447 somatic cancer genes and 191 regions across 60 genes for rearrangement detection by massively parallel sequencing. We examined associations between age (≤50, 51-65, and > 65) and molecular features, using the χ2 test for discrete data, and the Wilcoxon Rank Sum Test for quantitative data. Nominal p values were obtained, and the FDR correction was employed to obtain q values. Results: The following DNA alterations were significantly enriched in the TCGA in ≤50 vs 51-65 vs > 65 age groups respectively: focal deletion in CREBBP (3/25 [12%] vs 2/137 [1.5%] vs 1/250 [0.4%], p < 0.0001, q = 0.0126); microRNA Cluster III (13/25 [52%] vs 53/137 [38.7%] vs 55/250 [22%], p = 0.0005, q = 0.0252); fusion in FGFR3-TACC3 (3/25 [12%] vs 2/137 [1.5%] vs 5/250 [2%], p = 0.0055, q = 0.1386). Given that FGFR3-TACC3 fusions are potential therapeutic targets, we examined the association between FGFR3-TACC3 fusions and clinical features in a pooled analysis combining the TCGA and DFCI cohorts totaling 768 pts. FGFR3-TACC3 fusions were enriched in: ≤50 age group vs 51-65 vs > 65 (4/33 [12.1%] vs 7/232 [3%] vs 6/503 [1.2%] respectively, p = 0.0001). Fusions were also significantly more common in Asians vs. Blacks vs. Whites (6/47 [12.8%] vs 1/28 [3.6%] vs 10/666 [1.5%] respectively, p < 0.0001) and in never-smokers vs. ever smokers (12/194 [6.2%] vs 5/545 [0.9%] respectively, p < 0.0001). Conclusions: FGFR3-TACC3 fusion, a known potentially actionable genomic alteration in MIBC, is significantly enriched in young pts, never-smokers and those of Asian ethnicity. Clinical testing to detect FGFR3-TACC3 fusion should be prioritized for MIBC and HG-non-MIBC pts meeting these criteria.