4543 Background: VHL inactivation occurs in most clear cell RCCs and results in expression of the HIF hypoxia response program and tumor angiogenesis. Antiangiogenic therapies are central to therapy for mRCC, but resistance develops in all patients (pts). We hypothesized that a dual inhibition strategy focused on HIF would be active in the refractory setting. CRLX101 (Cerulean Pharma, Inc., Cambridge, MA, USA), an NDC with a camptothecin payload, has been shown in preclinical models to durably inhibit both HIF1a and HIF2a. Synergy has been observed in the pre-clinical setting between this NDC and antiangiogenic agents including bev. Methods: Pts with refractory mRCC were treated every 2 weeks with bev (10 mg/kg) and escalating doses of CRLX101 (12 mg/m2, 15 mg/m2) in a 3+3 phase I design. A phase IIa expansion cohort of 10 pts was treated at the RP2D. Pts were treated until progressive disease (PD) or prohibitive toxicity. Adverse events (AEs) were assessed using CTCAE v 4.0 and clinical outcome using RECIST v1.1. Correlative assays, designed to translate clinical results in terms of HIF expression and inhibition, were performed in parallel with the clinical investigation on baseline tumor material and on pre- and post-treatment bone marrow and plasma. Results: 22 response-evaluable pts were enrolled on study at two AMCs (12 clear cell, 5 papillary, 3 unclassified, 2 chromophobe). Pts had a median of 2 prior therapies, and all had at least 1 prior standard (std) therapy. No dose-limiting toxicities were observed. CRLX101 at its RP2D (15 mg/m2) combined safely with std bev. Grade 3 treatment-related AEs were observed - non-infectious cystitis (4) and hypertension (4). All other AEs were grade 1 or 2. The median PFS was 9.9 months. Overall response rate (ORR) was 23%, and 85% experienced either a PR or SD as best response. The 5 pts with a PR included 3 with clear cell and 2 with papillary RCC. The primary efficacy endpoint of this study was met. Conclusions: CRLX101 can be safely combined with std bev in mRCC. This two-drug combination demonstrated provocative PFS and ORR signals. A randomized phase II clinical trial in mRCC is enrolling pts. Clinical trial information: NCT01625936.

e15519 Background: There is no standard second line therapy for metastatic urothelial cancer (mUC) following platinum chemotherapy (CT). Taxanes have shown clinical activity in this setting. Cabazitaxel is a taxane exhibiting preclinical anti-tumor activity in docetaxel-sensitive and -resistant tumors, and survival benefit in advanced prostate cancer. This two stage phase II trial was designed to assess the efficacy and tolerability of cabazitaxel in patients (pts) with mUC and progression following platinum CT. Methods: Pts with mUC and measurable progression following platinum CT, CrCl >50, and ECOG performance status (PS) < 2 were eligible. Cabazitaxel was administered IV every 3 weeks, initiated at 20 mg/m2 in cycle 1 with inter-patient escalation to 25 mg/m2in cycle 2 at investigator discretion if the patient had no grade 3/4 toxicity during cycle 1. Pegfilgrastim 6 mg SQ was administered day 2 or 3 in all cycles. Radiographic assessment was performed every three cycles. Primary endpoint was overall response proportion. Results: Fourteen pts (12 male, 2 female) enrolled at 3 sites over 17 months with a median age of 60 (range 41-85) and median PS 0. A median of 2 cycles were administered (range 1-6). Most frequent adverse events included, grade 1-2: fatigue (8), diarrhea (6), anemia (5), nausea (4), neuropathy, increased creatinine, lightheadedness (3); grade 3 fatigue (1), anemia (1), diarrhea (1) and hypophosphatemia (1). One patient had grade 4 sepsis and refractory diarrhea precluding further cycles, and one died of refractory hypoxia, both in cycle 1, and were not evaluable for response. One patient had stable disease for 6 cycles. No CR or PR were observed in the initial 14 patients, and per protocol pre-specified stopping rule, the study did not proceed to the second stage due to lack of efficacy. Conclusions: Cabazitaxel showed limited clinical activity in platinum refractory mUC and toxicity with this dosing schema. Patients with platinum refractory mUC remain a challenging population to treat. Clinical trial information: NCT01437488.

412 Background: VHL is inactivated in most clear cell RCC (ccRCC) tumors giving rise to the HIF hypoxia response program and tumor angiogenesis. Antiangiogenic therapies are active in ccRCC, but resistance develops in all pts. Dual inhibition strategies may be needed to overcome resistance. CRLX101, a novel nano pharmaceutical formulation of camptothecin, has been shown in pre-clinical models to target HIF. We combined CRLX101 with bev in the treatment of pts with refractory mRCC in order to determine the safety and the recommended phase 2 dose (RP2D) of the combination, and to determine the therapeutic activity of treatment. Methods: Pts with mRCC refractory to conventional antiangiogenic therapies (all subtypes) were treated every 2 weeks with bev 10 mg/kg and escalating doses of CRLX101 (12 mg/m2, 15 mg/m2) in a standard 3+3 design with an expansion at the RP2D. Pts were treated until progression, death, or prohibitive toxicity. Adverse events were classified using CTCAE v 4.0. Radiographic results were assessed by RECIST v1.1. Results: 10 pts were enrolled on study – 6 clear cell, 2 papillary, 2 chromophobe. Median prior number of therapies was 2. No dose limiting toxicities were observed. CRLX101 at its single-agent RP2D (15 mg/m2) was safely combined with standard bev. 2 episodes of grade 3 non-infectious cystitis and 1 episode of grade 3 hypertension were observed. All other toxicities were grade 1 or 2. The median progression free survival (PFS) was 7.6 months which exceeded our pre-specified threshold of 16 weeks and has triggered 2nd stage accrual. 3 of 9 evaluable pts (33%) achieved a confirmed partial response (PR) including 1 patient with papillary mRCC; 4 pts (44%) displayed stable disease, and 2 had progressive disease as their best overall response. Conclusions: Our results demonstrate that CRLX101 can be safely combined with bev in pts with refractory mRCC. The RP2D is 15 mg/m2 q2w of CRLX101 with standard bev. Preliminary outcomes to date suggest that clinical benefit, manifest as prolonged PFS, is conferred to pts with all histologic subtypes of mRCC. Clinical trial information: NCT01625936.

e15611 Background: VHL is inactivated in most clear cell RCC (ccRCC) giving rise to the HIF hypoxia response program and tumor angiogenesis. Antiangiogenic therapies are active in ccRCC, but resistance develops in all patients (pts). Dual inhibition strategies may be needed to overcome resistance. CRLX101 (Cerulean Pharma, Inc., Cambridge, MA, USA), a nano pharmaceutical formulation of camptothecin, has been shown in pre-clinical models to inhibit HIF. We combined CRLX101 with bev in the treatment of pts with refractory mRCC to determine the safety and the recommended phase 2 dose (RP2D) of the combination, and to determine the therapeutic activity of treatment. Methods: Pts with refractory mRCC (all subtypes) were treated every 2 weeks with bev 10 mg/kg and escalating doses of CRLX101 (12 mg/m2, 15 mg/m2) in a standard 3+3 design with an expansion at the RP2D. Pts were treated until progression, death, or prohibitive toxicity. Adverse events were classified using CTCAE v 4.0. Radiographic results were assessed by RECIST v1.1. Results: 12 pts were enrolled on study – 8 clear cell, 2 papillary, 2 chromophobe. Median prior number of therapies was 2. No dose limiting toxicities were observed. CRLX101 at its single-agent RP2D (15 mg/m2) was safely combined with standard bev. 2 episodes of grade 3 non-infectious cystitis and 1 episode of grade 3 hypertension were observed. All other toxicities were grade 1 or 2. The median progression free survival (PFS) was 7.6 months which exceeded our pre-specified threshold and has triggered 2nd stage accrual. 3 of 9 response-evaluable pts (33%) achieved a confirmed partial response (PR) including 1 patient with papillary mRCC; 4 pts (44%) displayed stable disease as their best overall response. Conclusions: Our results demonstrate that CRLX101 can be safely combined with bev in pts with refractory mRCC. The RP2D is 15 mg/m2 q2w of CRLX101 with standard bev. Preliminary outcomes to date suggest that clinical benefit, manifest as prolonged PFS, is conferred to pts with all histologic subtypes of mRCC. Clinical trial information: NCT01625936.

406 Background: Carbonic anhydrase IX (CAIX), a transmembrane glycoprotein with an intracellular enzymatic domain, is overexpressed in ccRCC as a result of VHL loss. Chimeric antibody girentuximab is a monoclonal antibody that binds an extracellular epitope of CAIX. Iodine radiolabeled (I-124) girentuximab can be used as a PET radiotracer and has been shown to identify ccRCC primary tumors with high sensitivity and specificity. This study was designed to test the feasibility of using I-124 girentuximab PET/CT in the m ccRCC setting and to produce preliminary data regarding the operating characteristics of these scans in m ccRCC. Methods: Patients (pts) with ccRCC refractory to standard antiangiogenic therapy were treated with brivanib (Bristol-Myers Squibb, Co.), a tyrosine kinase inhibitor of VEGFR2 and FGFR. Baseline I-124 girentuximab scans were obtained. For 1 pt, a second scan was performed after 8 weeks. Clinical outcome was measured radiographically by RECIST 1.0 and standard cross-sectional imaging. Results: 5 pts had 6 I-124 girentuximab PET/CT scans. No adverse events occurred. 17 RECIST target lesions were identified at baseline. PET showed heterogeneous expression of CAIX in all scans: overall, 9 of 17 (53%) metastatic target lesions expressed CAIX. Progressive disease occurred in 5/6 pts within 4 months on treatment, and progression occurred in all 8 of the CAIX non-avid lesions (100%) in these pts. One pt achieved a RECIST partial response (PR) with treatment – the pt for whom paired studies were obtained. For this pt, a PET PR was observed 2 cycles earlier than with conventional imaging. Conclusions: I-124 girentuximab PET/CT was safe and feasible in m ccRCC. We found that metastases were detected. More importantly, for pts with ccRCC refractory to conventional antiangiogenic therapies, CAIX expression was heterogeneous, and progression occurred in CAIX non-avid lesions. Taken together, these data suggest that CAIX expression identified with this functional imaging modality represents a biological prerequisite for continued antiangiogenic therapeutic activity, and the heterogeneous CAIX expression observed helps to explain why brivanib was not useful in this setting. Clinical trial information: NCT01253668.

Background/Aim: Electron-beam computerized tomography (EBCT) is able to noninvasively quantify coronary artery calcification (CAC). Chronic kidney disease (CKD) patients frequently have CAC, and clinicians are puzzled regarding the clinical significance of this finding and the diagnostic accuracy of coronary EBCT in CKD. The aim of this study was to determine the correlation in CKD patients between CAC measured by EBCT and 50% stenosis determined by coronary angiography (CA), the gold standard to identify atherosclerotic lesions. Method: We recruited 37 patients with CKD from a single institution and compared their coronary EBCT and CA results using standard statistical analysis. Results: Patients with at least one vessel with ≧50% stenosis by CA had higher mean CAC scores [2,407.9 ± (SD) 3,165.3 vs. 227 ± 443.4; p < 0.001] and higher median CAC scores (1,052 vs. 25.8; p < 0.001) as compared with those having no stenosis ≧50%. The sensitivity was 85.7%, and the specificity 82.6% using 50% stenosis as the definition for coronary artery disease and using a CAC score of 400 as a cutoff value for the EBCT results. The area under the receiver operating characteristic curve was 0.84. The diagnostic accuracy (proportion of correct results) was 83.8%. The negative predictive value was 90.5%. The receiver operating characteristic curve suggests that the optimal cutoff value for CAC scores in our cohort is 315.9, increasing the area under the receiver operating characteristic curve to 0.91. The total coronary artery stenosis was significantly associated with the CAC score (p = 0.01). Conclusions: EBCT has a very good predictive value for obstructive coronary artery disease. EBCT could be used as a screening tool in CKD patients with a low-to-intermediate risk for coronary artery disease.