Nivolumab plus cabozantinib versus sunitinib for first-line treatment of advanced renal cell carcinoma: extended follow-up from the phase III randomised CheckMate 9ER trial

BACKGROUND

Nivolumab plus cabozantinib (NIVO + CABO) was approved for first-line treatment of advanced renal cell carcinoma (aRCC) based on superiority versus sunitinib (SUN) in the phase III CheckMate 9ER trial (18.1 months median survival follow-up per database lock date); efficacy benefit was maintained with an extended 32.9 months of median survival follow-up. We report updated efficacy and safety after 44.0 months of median survival follow-up in intent-to-treat (ITT) patients and additional subgroup analyses, including outcomes by International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) prognostic risk score.

PATIENTS AND METHODS

Patients with treatment-naïve aRCC received NIVO 240 mg every 2 weeks plus CABO 40 mg once daily or SUN 50 mg for 4 weeks (6-week cycles), until disease progression/unacceptable toxicity (maximum NIVO treatment, 2 years). Primary endpoint was progression-free survival (PFS) per blinded independent central review (BICR). Secondary endpoints were overall survival (OS), objective response rate (ORR) per BICR, and safety and tolerability.

RESULTS

Overall, 323 patients were randomised to NIVO + CABO and 328 to SUN. Median PFS was improved with NIVO + CABO versus SUN [16.6 versus 8.4 months; hazard ratio (HR) 0.59; 95% confidence interval (CI) 0.49-0.71]; median OS favoured NIVO + CABO versus SUN (49.5 versus 35.5 months; HR 0.70; 95% CI 0.56-0.87). ORR (95% CI) was higher with NIVO + CABO versus SUN [56% (50% to 62%) versus 28% (23% to 33%)]; 13% versus 5% of patients achieved complete response, and median duration of response was 22.1 months versus 16.1 months, respectively. PFS and OS favoured NIVO + CABO over SUN across intermediate, poor and intermediate/poor IMDC risk subgroups; higher ORR and complete response rates were seen with NIVO + CABO versus SUN regardless of IMDC risk subgroup. Any-grade (grade ≥3) treatment-related adverse events occurred in 97% (67%) versus 93% (55%) of patients treated with NIVO + CABO versus SUN.

CONCLUSIONS

After extended follow-up, NIVO + CABO maintained survival and response benefits; safety remained consistent with previous follow-ups. These results continue to support NIVO + CABO as a first-line treatment for aRCC.

TRIAL REGISTRATION

ClinicalTrials.gov, NCT03141177.

Phase Ib/II trial of gemcitabine, cisplatin, plus lenalidomide as first-line therapy for patients with metastatic urothelial carcinoma

321

Background: Cisplatin-based chemotherapy is standard in patients (pts) with metastatic urothelial carcinoma (MUC), though outcomes remain poor and novel approaches are needed. Lenalidomide, a potent thalidomide analog with antiangiogenic and immunomodulatory properties, has been shown to enhance the antiproliferative properties of standard chemotherapy in nonclinical studies of urothelial cancer (Apolo, ASCO 2011). In this phase Ib/II clinical trial we aimed to determine the safety and efficacy of gemcitabine, cisplatin, plus lenalidomide (GCL) in pts with MUC. Methods: Pts with chemotherapy-naïve MUC received gemcitabine 1000mg/m2 days 1 and 8 and cisplatin 70mg/m2 day 1, of a 21-day cycle. In the phase Ib portion, the dose of lenalidomide was to be dose escalated in successive cohorts starting at 10 mg PO daily, on days 1-14, using standard “3+3” dose escalation rules. The primary objective of the phase II portion was to determine the progression-free survival at one year. Results: A total of 7 pts received GCL (lenalidomide 10 mg dose level) in the phase 1b portion. A single pt experienced a dose limiting toxicity (DLT), grade 4 neutropenia. A decision was made to not dose escalate lenalidomide further due to the frequent need for dose delays and reductions of GC due to cytopenias. An additional 2 pts were enrolled in the phase II portion but the study was terminated due to frequent dose delays/reductions and slow accrual. Among the 9 pts enrolled, 5 required dose reductions of G (4 requiring >1), 4 required dose reductions of C (1 requiring >1), and 6 required at least one dose delay. The most frequent grade ≥ 3 adverse events were: neutropenia (n=7), anemia (n=3), thrombocytopenia (n=3), and diarrhea (n=2). Tumor responses included: partial response (n=3), stable disease (n=3), progressive disease (n=2), and unevaluable (n=1). Conclusions: Chronic administration of the combination of GCL was compromised by overlapping myelosuppression and the need for frequent dose reductions/delays of GC due to cytopenias. Despite promising nonclinical data, combining cytotoxic chemotherapy with “targeted” small molecules faces practical challenges. Clinical trial information: NCT01342172.

A phase II study of cabozantinib in patients (pts) with relapsed or refractory metastatic urothelial carcinoma (mUC)

307

Background: Translational studies have shown that shed MET levels in serum and urine of pts with UC correlate with stage and visceral metastases and that cabozantinib reverses HGF-driven UC cell growth and invasion. These data support the evaluation of cabozantinib in pts with mUC. Methods: In this phase II study, pts receive cabozantinib 60 mg daily in 28-day cycles. There are 3 study cohorts (1) mUC, (2) bone only mUC (3) metastatic rare bladder histology. The primary objective is to determine the response rate (RR) by RECIST. Response is assessed every 2 cycles. Tissue, blood, and urine were collected on all pts to test for MET/HGF and immune subsets. Results: 26 out of 55 pts have enrolled (19 M, 7 F): median age 62 (42-82) and median KPS of 80%. Primary sites include bladder (77%) and upper tract UC (23%). Prior therapy includes 30% pts with 1 regimen, 39% with 2, 15% with 3, 8% with 4 and 8% with 6. 81% of pts had visceral metastases (lung, liver and bone) and 19% lymph node only metastases. 23 pts (19 in cohort 1, 3 in cohort 2 and 1 in cohort 3) were evaluable for response after completing at least 4 weeks of therapy. In cohort 1, 2 pts achieved PR (1 remained on study for 10 months and 1 remains on study after >12 months of therapy); 7 pts had SD for at least 16 weeks (1 remained on study for 11 months); 10 had PD; 1 is too early to assess for response; 1 was removed before restaging for toxicity and 1 was removed for not meeting eligibility. The objective RR is 11% and SD 37% for a clinical benefit of 48%. In cohort 2, 1 of 3 pts had a near resolution of bone lesions on NaF PET/CT for 11 months. In cohort 3, only pt enrolled (squamous cell carcinoma of the bladder) achieved SD for 16 weeks. Mixed responses with regression of lung, bone or lymph nodes were observed in 30% of pts with PD. Grade 3/4 toxicities included: fatigue (8%), hyponatremia (8%), hypophosphatemia (8%) diarrhea (4%), thromboembolism (4%), transaminitis (4%), hypothyroidism (4%), thrombocytopenia (4%), dysphonia (4%), hypomagnesemia (4%), creatinine increase (4%) and proteinuria (4%). Conclusions: Cabozantinib has clinical activity in pts with relapsed or refractory mUC with manageable toxicities. Further studies are underway to correlate response to therapy with MET expression. Clinical trial information: NCT01688999.

A safety study of trebananib (AMG 386) and abiraterone in metastatic castration-resistant prostate cancer

218

Background: Trebananib is an angiopoietin1/2 antagonist peptibody. Androgens stimulate expression of VEGF via activation of hypoxia inducible factor-a (HIFa). Androgen deprivation therapy (ADT) is associated with lower HIF1a gene expression in prostate cancer tissue. Dual targeting of the androgen and angiogenic axis represents a potential synergistic anti-angiogenic therapeutic approach in metastatic castration resistant prostate cancer (mCRPC). In this preliminary safety study we hypothesize that trebananib in combination with abiraterone will have a favorable tolerability and efficacy profile. Methods: Patients with mCRPC were treated with abiraterone 1000mg daily and prednisone 5 mg twice daily. Trebananib was administered intravenously every week, in escalating doses from 15mg/kg to 30mg/kg on days 1, 8, 15, and 22 every 28-days. Results: A total of 9 patients were enrolled. Three of nine patients had prior chemotherapy. The median age was 63.8 (63-71yrs). No dose limiting toxicities were observed. The most common grade ≥ 2 toxicities included limb edema (3/9), hyperglycemia (1/9), gastrointestinal (2/9), fatigue (2/9), hypertension (1/9), confusion (1/9), weight gain (1/9) and insomnia (2/9). 5/9 of patients had an overall PSA decline of >30%. 8/9 patients were evaluable for response. Prior chemotherapy patients were on study for 1 and 3 months. No prior chemotherapy patients were treated for 1, 6, 9, 10, 10, and 17 months. Conclusions: Trebananib in combination with abiraterone is well tolerated and displayed an acceptable safety profile in patients with mCRPC. Based on this safety data a randomized phase II study randomizing chemotherapy-naïve mCRPC patients to either abiraterone/prednisone plus AMG 386 at 30mg/kg or abiraterone/prednisone is currently accruing at the NCI. Clinical trial information: NCIT01553188.

Assessment of bone metastases in patients (pts) with urothelial carcinoma using 18F-sodium fluoride PET/CT (18F-NaF) versus 18F-fluorodeoxyglucose PET/CT (18F-FDG)

329

Background: 18F-NaF has shown improved sensitivity for bone imaging when compared to conventional planar imaging or SPECT/CT using 99mTc-MDP. We compared the number of bone lesions detected on 18F-NaF versus 18F-FDG in urothelial cancer pts with known bone metastases undergoing treatment. Methods: Pts enrolled in a prospective single-arm phase II study of cabozantinib underwent 18F-NaF and 18F-FDG scans at baseline and at 8 weeks of therapy. In a lesion-based analysis independently confirmed by a nuclear medicine physician, abnormal foci of radiotracer uptake were categorized by location (skull, spine, pelvis, thorax, or long bones) and by disease state (benign, malignant, or indeterminate). A patient-based analysis was performed to determine if findings indicated disease progression, stable disease, or improvement of disease, based on the number of lesions and standardized uptake values (SUVs). Results: 294 total bone lesions were identified at baseline in 10 pts (8 male and 2 female, ages 44-73). 18F-NaF identified more lesions than 18F-FDG at baseline, 294 vs. 119. In a paired analysis, the median difference was 11.5 more lesions detected per patient on 18F-NaF vs. 18F-FDG (by Wilcoxon signed-rank test, p = 0.023). More total thoracic bone lesions at baseline, 100 vs. 23, were also detected on 18F-NaF vs. 18F-FDG, median 6.5 vs. 1.0 with a median difference of 6 more lesions per patient on 18F-NaF (p = 0.016). 18F-NaF also detected more skull lesions at baseline, 19 vs. 1, which was clinically but not statistically significant (p = 0.250). There was general concordance in the patient-based analysis; only 1 18F-NaF scan demonstrated progressive disease while its corresponding 18F-FDG scan showed stable disease. Conclusions:18F-NaF identified more lesions than 18F-FDG at baseline, making it a good staging exam. However, there was agreement between 18F-NaF and 18F-FDG in terms of tumor response in almost all the follow-up scans. Therefore, although a greater number of bone lesions are seen in 18F-NaF compared with 18F-FDG, the clinical significance in assessing treatment response remains to be determined. Clinical trial information: NCT01688999.

The high incidence of vascular thromboembolic events (VTE) in advanced urothelial cancer (UC) patients (pts) treated with carboplatin (Cb): Analysis of treatment with gemcitabine (G)/cb (GCb), gcb/bevacizumab (GCbBev), or gemcitabine/cisplatin (GCis)

316

Background: VTE occur in ~13% of pts receiving Cis-based therapy for UC. Cb-based therapy is considered less thrombogenic but no definitive data exist to support this conclusion. While Bev added to chemotherapy (Ch) increases VTE in other tumors, VTE impact when added to UC Ch is unknown. This study evaluated the incidence of and etiologic factors for VTE in UC pts treated with GCb, GCbBev and GCis. Methods: UC pts treated from 6/2006 to 6/2010 on a GCbBev protocol were analyzed prospectively. Planned therapy was 6 cycles of GCbBev then Bev alone; pts who had ≥ 3 cycles were included. Similar UC pts treated with GCb or GCis during the same time were retrospectively evaluated. Type of VTE and potential contributing clinical factors were collected. VTE was defined as pulmonary embolism, vascular thrombosis, myocardial infarction or cerebral vascular accident. Associations with Ch regimen were tested using Fisher’s exact test or linear regression. Factors associated with VTE were analyzed using conditional logistic regression stratified by Ch regimen. Results: 198 pts were analyzed. VTE occurred in 13/51 (26%) GCbBev pts, 22/92 (24%) GCb pts, and 8/55 (15%) GCis pts. Age (≤65 vs >65; p<0.001), having had a prior cystectomy (p<0.001), mass near pelvic vessels (p=0.027), Khorana risk group (p=0.025) and anti-platelet therapy (p=0.036) were significantly associated with the Ch regimen reflecting cohort-specific differences. Type of Ch was not associated with any VTE (p=0.3) or type of VTE (arterial vs venous) (p=0.11). Having had a prior cystectomy was associated with increased risk of VTE (OR 2.2, 95% CI 1.0-5.0, p=0.047). Conclusions: This is the largest series reporting VTE in Cb-treated UC pts; the VTE rate of 24% (95% CI 17, 32%) is higher than expected. Bev does not appear to increase this risk. VTE in Cis-treated pts (15%) was similar to prior reports. The finding that pts with advanced UC are at high risk of VTE regardless of the specific platinum agent warrants further study of Cb therapy-related risk and analysis for contributing factors including prior pelvic surgery.

A phase I study of the multikinase inhibitor cabozantinib (C) plus docetaxel (D) and prednisone (P) in metastatic castrate-resistant prostate cancer (mCRPC)

108

Background: Cabozantinib (C) is a multikinase inhibitor of c-Met, vascular endothelial growth factor receptor two and RET. C has shown activity in metastatic castrate resistant prostate cancer (mCRPC), with resolution of bone lesions on bone scan (BS), regression of soft tissue/visceral disease (STD), reductions in circulating tumor cells and bone biomarkers. Combining docetaxel (D) with other agents, without overlapping toxicities, can target different cellular signaling pathways necessary for tumor survival. Methods: Patients (pts), with no prior D for CRPC, receive a fixed dose of D (75 mg/m2 IV day one of each 21 day cycle) and prednisone (P) (5 mg po q12 hours) with C at three escalating dose levels: 20 mg, 40 mg, or 60 mg (all po daily). Using a standard three-plus-three design, three to six pts are treated at each dose level until the maximum tolerated dose (MTD) has been defined. Results: Thirteen pts have been accrued; four on dose level one, six on dose level two, and three on dose level three. Median age 69 (45 to 84). Four pts have an Eastern Cooperative Oncology Group Performance Status (ECOG PS) of zero and nine pts have a PS of one. Median Gleason score is nine (7 to 10). Median on-study prostate-specific antigen (PSA) is 129.2 ng/mL (0.01-508.5 ng/mL). Median cycles is six (1 to 17). Grade 1 adverse events (AEs), possibly related to C; dysgeusia (4/12), oral mucositis (4/12), increased ALT (3/12), and epistaxis (3/12). Grade 2 AEs; nausea (2/12), hand/foot syndrome (2/12), fatigue (2/12), dysgeusia (2/12), oral mucositis (2/12), hypophosphatemia (2/12), and anemia (2/12). Grade 3 AE is hypophosphatemia (2/12). Grade 4 AE is neutropenia (1/12). MTD of C is 60 mg. Of nine evaluable pts, six have bone only disease. Of these six, three pts have PSA declines of less than 30% with improvement on BS (two pts) or stable BS (one pt). The other three pts have PSA declines of greater than 30% and bone scan improvement. Three pts have STD and bone disease; one patient had a PSA decline of greater than 30% with improvement on BS and SD by CT scan. One patient had an increase in PSA of less than 30% with improvement on BS and CT. The third pt had PD by CT and an increase in PSA equal to 30%. PFS probability at six months is 90.0% and is 67.5% at eight months and beyond. Conclusions: The addition of C to D and P, has an acceptable toxicity profile. CT scan and BS improvements did not correlate with PSA declines in all pts. An expansion cohort will combine D plus P with C at the MTD (60 mg) to determine clinical benefit. Clinical trial information: NCT01683994.

Tumor-intrinsic and tumor-extrinsic factors impacting hsp90- targeted therapy

In 1994 the first heat shock protein 90 (Hsp90) inhibitor was identified and Hsp90 was reported to be a target for anticancer therapeutics. In the past 18 years there have been 17 distinct Hsp90 inhibitors entered into clinical trial, and the small molecule Hsp90 inhibitors have been highly valuable as probes of the role of Hsp90 and its client proteins in cancer. Although no Hsp90 inhibitor has achieved regulatory approval, recently there has been significant progress in Hsp90 inhibitor clinical development, and in the past year RECIST responses have been documented in HER2-positive breast cancer and EML4-ALK-positive non-small cell lung cancer. All of the clinical Hsp90 inhibitors studied to date are specific in their target, i.e. they bind exclusively to Hsp90 and two related heat shock proteins. However, Hsp90 inhibitors are markedly pleiotropic, causing degradation of over 200 client proteins and impacting critical multiprotein complexes. Furthermore, it has only recently been appreciated that Hsp90 inhibitors can, paradoxically, cause transient activation of the protein kinase clients they are chaperoning, resulting in initiation of signal transduction and significant physiological events in both tumor and tumor microenvironment. An additional area of recent progress in Hsp90 research is in studies of the posttranslational modifications of Hsp90 itself and Hsp90 co-chaperone proteins. Together, a picture is emerging in which the impact of Hsp90 inhibitors is shaped by the tumor intracellular and extracellular milieu, and in which Hsp90 inhibitors impact tumor and host on a microenvironmental and systems level. Here we review the tumor intrinsic and extrinsic factors that impact the efficacy of small molecules engaging the Hsp90 chaperone machine.

Use of supportive measures to improve outcome and decrease toxicity in docetaxel-based antiangiogenesis combinations in metastatic castrate resistant prostate cancer (mCRPC)

e16017

Background: We have completed accrual of 63 patients (pts) to our study combining lenalidomide (L), with bevacizumab (B), docetaxel (D), and prednisone (P) (ART-P) in mCRPC. Due to the lack of improved survival and the increased toxicity of anti-angiogenic docetaxel combinations in the MAINSAIL and CALGB 90401 trials, we attempted to compare and contrast our studies with these failed phase III trials. Methods: Among the first 52 pts on ART-P, 3 received L 15 mg daily, 3 received 20 mg daily, and the others received 25 mg daily for 14 days of every 21−day cycle (C). We then enrolled 11 pts at L 15 mg. All pts received D 75 mg/m2 and B 15 mg/kg on day 1 with P 10 mg and enoxaparin daily. Pegfilgrastim was given on day 2. Patients on CALGB 90401 received D 75 mg/m2 and B 15 mg/kg on day 1, with P 10 mg. On MAINSAIL, pts received D 75 mg/m2, L 25 mg daily for 14 days of every 21−day cycle with daily P. Patients on CALGB 90401 and MAINSAIL did not receive enoxaparin or pegfilgrastim prophylactically. Results: The median number of Cs on ART-P is 18 (1-52). Median PFS is 19.1 months. Twenty-seven pts had a PR, and one pt with measurable disease had a CR. Two patients (3%) had deep vein thromboses. Of 1,334 Cs given, 14 cycles were complicated by febrile neutropenia (FN) (1%). There were no treatment related deaths. In comparison, median number of Cs in MAINSAIL L+DP arm was 6, with a PFS of 45 weeks and an OS of 77 weeks. Thirty-four pts (6.5%) developed pulmonary emboli and there were 2 deaths due to toxicity in the experimental arm. Nearly 12% of Cs were complicated by FN. In the experimental arm of CALGB 90401 trial, median OS was 22.6 months with median PFS of 9.9 months. The median number of Cs were 8 and 19 pts developed thrombosis/emboli (3.6%). In addition, 37 patients developed FN and treatment related deaths were reported at 4%. Conclusions: The use of supportive care allowed longer treatment duration with the ART-P combination as compared to D+L (MAINSAIL) and D+B (CALGB 90401), potentiating a longer PFS, RR and possibly OS with an improved safety profile. This data demonstrates the potential importance of supportive measures and is hypothesis generating for future combination studies. Clinical trial information: NCT00942578.

A phase I study of cabozantinib (Cabo) plus docetaxel (D) and prednisone (P) in metastatic castrate resistant prostate cancer (mCRPC)

TPS5095

Background: In mCRPC, two randomized trials demonstrated an overall survival (OS) benefit with the chemotherapeutic agent D. However, the survival improvement is modest and new strategies are needed to enhance clinical response. D-based combinations have been evaluated as one alternative strategy. Cabo targets multiple tyrosine kinases including c-Met, vascular endothelial growth factor receptor 2 (VEGFR2) and RET. Cabo has shown activity in mCRPC, with resolution of bone lesions on bone scan, regression of soft tissue/visceral disease, and reductions in circulating tumor cells and bone biomarkers (Smith, et al, J Clin Oncol 30, 2012 [suppl; abstr 4513]). We hypothesize the addition of Cabo to D and P, in patients (pts) with mCRPC, will have an acceptable toxicity profile and could lead to improved survival by targeting different cellular pathways simultaneously. This combination therapy may represent a safe and effective strategy to improve the outcome of mCRPC pts treated with D-based chemotherapy. Methods: This is a phase I trial to determine the safety profile and the recommended phase II dose of Cabo in combination with D and P. Pts receive a fixed dose of D (75 mg/m2 IV day 1 of each 21 day cycle) and P (5 mg po q12 hours) in combination with Cabo at three escalating doses: dose level 1 is 20 mg, level 2 is 40 mg, and level 3 is 60 mg (all po qdaily). Using a standard 3 + 3 design, three patients will initially be treated at each dose level until the maximum tolerated dose (MTD) has been defined. An expansion cohort will then be enrolled at the MTD. The accrual ceiling for the study, including both the dose escalation and the expansion phases, is set at 24 pts. Secondary objectives include assessments of pharmacokinetics of each agent, evaluation of antitumor activity of the combination therapy, and assessment of changes in molecular biomarkers for receptor tyrosine kinase and angiogenesis pathways, as well as biomarkers for bone metabolism. Restaging with bone and CT scan will be undertaken every 3 cycles. Enrollment at dose level 1 has been completed without dose-limiting toxicity. Accrual is ongoing at the second dose level. Clinical trial information: NCT01683994.

Preclinical and correlative studies of cabozantinib (XL184) in urothelial cancer (UC)

4543

Background: Mounting evidence supports Met as a therapeutic target in urothelial cancer (UC). Activated Met can promote angiogenesis and tumor growth by upregulating VEGF and may play a role in UC pathogenesis. Cabozantinib inhibits VEGFR2 and Met pathways. In this study, we assessed shed Met (sMet) levels in the urine and serum of UC patients (pts) and cabozantinib’s effects on HGF-driven UC cell growth and invasion. Methods: sMet levels in serum and urine samples from 31 pts with UC (23 metastatic, 8 muscle-invasive) were correlated with stage, presence of visceral metastases and urinary source. The effects of cabozantinib on 4 human UC-derived cell lines were studied in vitro. Intact RT4, TCC-SUP, T24M2 and T24M3 cells at 80% confluence were serum deprived 16 h, then left untreated or treated with hepatocyte growth factor (HGF) and/or cabozantinib prior to analysis of Met, phospho- (p)Met, pAkt, Akt, pMAPK and MAPK by immunoassay or immunoblotting. Cabozantinib effects on basal and HGF-induced UC cell invasion, proliferation and soft agar growth were measured. Results: Median serum Met levels were modestly higher in pts with metastatic versus muscle-invasive disease. Urinary Met levels were clearly higher in pts with visceral metastasis (P=0.0111) and in urine from ileal conduits and neobladders compared to normally voided urine, regardless of stage (P=0.0489). Met content in UC cell lines was low in RT4 and higher in T24M2, T24M3 and TCC-SUP. Basal pMet content was universally low, increased significantly by HGF and this was reversed by cabozantinib. HGF-driven increases in pAkt/Akt and pMAPK/MAPK in all 4 cell lines were reversed by cabozantinib, as were HGF-enhanced UC cell invasion, proliferation and anchorage independent growth. Conclusions: Median urinary sMet is significantly higher in pts with visceral metastasis and in specimens from ileal conduits and neobladders relative to normally voided urine. UC cell Met content in culture increased with disease grade; HGF stimulated activation of Met and known effectors, and enhanced invasion, growth rate and anchorage-independent growth; cabozantinib effectively reversed these HGF-driven effects. These data support evaluation of cabozantinib in pts with metastatic UC.

A phase II study of trebananib (AMG 386) and abiraterone in metastatic castration resistant prostate cancer

TPS5102

Background: Preclinical studies support the use of an antiangiogenic approach in the treatment of prostate cancer. Trebananib is a novel peptide-Fc fusion protein that sequesters angiopoeitin 1 and angiopoeitin 2, thereby preventing their interaction with their common receptor Tie2, and inhibiting tumor endothelial cell proliferation and tumor growth. Trebananib is currently in Phase 3 trials for the treatment of ovarian carcinoma and has been shown to have clinical activity in multiple tumor types. Previous studies have demonstrated that in vivo alterations of testosterone levels regulate the expression of vascular endothelial growth factor, fibroblast growth factor, and angiopoietin associated factors. Dual inhibition of the androgen and angiogenic axis represents a novel strategy of combined targeted therapy for patients with metastatic castration-resistant prostate cancer (mCRPC). We hypothesize that the addition of trebananib to CYP17 inhibitor abiraterone and prednisone will increase the median progression free survival (PFS) in chemotherapy-naïve mCRPC. Methods: This phase 2 study will evaluate the treatment effect as measured by progression free survival in patients treated with trebananib plus abiraterone/prednisone relative to abiraterone/prednisone alone. 72 patients with progressive, mCRPC will be randomized 1:1 to either study arm. Trebananib is administered intravenously every week, on days 1, 8, 15 and 22 of each 28-day cycle. Abiraterone acetate is taken once daily with prednisone 5 mg twice daily. We have completed the initial run-in phase of trebananib at 15mg/kg and 30mg/kg. The randomized phase of the study will use the 30 mg/kg dose of trebananib with the standard dose (1000 mg) of abiraterone. The primary end point is radiographic PFS. Secondary end points include overall survival, changes in genetic biomarkers related to the androgen and angiogenesis signaling axis, molecular markers of angiogenesis, circulating tumor cells and androgen receptor signaling status in circulating tumor cells before and after treatment. This combination of angiogenesis inhibition and abiraterone has the potential to improve clinical outcomes in front-line therapy for mCRPC. Clinical trial information: NCT01553188.

A phase II study of cabozantinib (XL184) in patients with advanced/metastatic urothelial carcinoma

TPS4589

Background: Accumulating evidence supports MET as a therapeutic target in urothelial carcinoma. Activated MET can promote angiogenesis and tumor growth by upregulating VEGF and may play a role in urothelial carcinoma pathogenesis. Cabozantinib inhibits primarily VEGFR2 and MET pathways. Cabozantinib has been approved by the FDA for the treatment of progressive metastatic medullary thyroid cancer, is in Phase 3 trials for metastatic castration-resistant prostate cancer and has demonstrated clinical activity in multiple solid tumors. We previously reported that shed MET levels in serum and urine of patients with urothelial carcinoma correlate with stage, presence of visceral metastases and urinary source and that cabozantinib is effective in reversing HGF-driven urothelial carcinoma cell growth and invasion. These data support the evaluation of cabozantinib in patients with metastatic urothelial carcinoma. Methods: This is a phase II study of oral cabozantinib 60mg daily given continuously in 28-day cycles. There are three study cohorts: [1] metastatic urothelial carcinoma [2] bone only metastatic urothelial carcinoma [3] metastatic non-urothelial carcinoma of the bladder, urethra, ureter, or renal pelvis. A maximum of 55 subjects will be enrolled. Up to 45 patients will be accrued to cohort 1.The remainder will be enrolled on exploratory cohorts 2 & 3. A two-stage single-arm phase II design will be employed. The primary objective is to determine the objective response rate in patients with metastatic urothelial carcinoma who have progressed on prior chemotherapy. Secondary objectives include progression free survival, safety and toxicity, and overall survival. Exploratory objectives include tumor tissue Met expression, shed MET levels in serum and urine, immune subsets, genetic biomarkers, molecular markers of angiogenesis and circulating tumor cells, correlation with clinical response parameters. Finally we will explore treatment evaluation with FDG and NaF PET/CT compared to standard imaging. This study is supported by the Cancer Therapy Evaluation Program (CTEP). NCT01688999 Clinical trial information: NCT01688999.

Preliminary results of a prospective study of 18F-NAF PET/CT in prostate cancer

103

Background: We performed a prospective study of 18F-NaF PET/CT bone scan (NaF) in the detection of bone metastases in men with prostate cancer. We previously reported that NaF identified more malignant lesions than Technetium-99m MDP bone scan (TcBS) (ASCO 2012 10589). This study evaluates the ability of NaF to detect bone metastasis in men with normal TcBS and also explores the change in NaF over 6 and 12 months compared to PSA changes. Methods: In a prospective 2-arm study, 60 men with prostate cancer (30 with and 30 without bone metastases by TcBS) were studied (ages 51-79). All had NaF and TcBS at baseline, followed by repeat NaF at 6 and 12 months. TcBS and NaF were reviewed by experienced nuclear medicine physicians. Abnormal foci of uptake on TcBS and NaF were classified as benign, malignant or indeterminate. Malignant uptake on NaF was confirmed by characteristic osteoblastic features on CT. Scan results were categorized as “positive” if any malignant lesion was present. In the 6 and 12 months follow up NaF, results were categorized as progression of disease (PD) = any new lesions or SUV increase > 30% in known lesions; stable disease (SD) = no new lesions or SUV changes < 30% in known lesions; and improvement of disease (ID) = resolution of known lesions or decrease SUV > 30% in known lesions. Results: 60 men have enrolled on study, 58 and 34 completed 6 and 12 month follow-up respectively. At baseline, 14 of 30 (47%) men with negative TcBS showed evidence of bone metastases in NaF (PSA mean 45); 7/14 had 2 baseline NaF and showed positive results in both, demonstrating reproducibility; 13/14 and 7/14 had follow up NaF at 6 and 12 months, respectively, all of which remained positive. In follow-up, 13/58 men at 6 months and 8/34 men at 12 months had PD from baseline on NaF, of whom 5/13 (38%) at 6 months and 5/8 (63%) at 12 months also had a PSA increase > 50%. All men who had PD on NaF at 6 months and had a follow-up scan at 12 months remained positive. 15 men at 6 months and 7 men at 12 months had ID on follow-up NaF, of which 11/15 (73%) and 6/7 (86%) had PSA decrease > 50% at 6 and 12 months, respectively. Conclusions: Early results of this ongoing NaF study are encouraging and suggest NaF identifies metastatic bone disease earlier than TcBS and correlates with changes in PSA. Clinical trial information: NCT01240551.

Preclinical and correlative studies of cabozantinib (XL184) in urothelial cancer (UC)

314

Background: Mounting evidence supports Met as a target in urothelial cancer (UC). Activated Met can promote angiogenesis and tumor growth by upregulating VEGF and may play a role in UC pathogenesis. Cabozantinib inhibits both VEGFR2 and Met pathways. In this study, we assessed shed Met (sMet) levels in the urine and serum of UC patients (pts) and cabozantinib’s effects on HGF-driven UC cell growth and invasion. Methods: sMet levels in serum and urine samples from 31 pts with UC (23 metastatic, 8 muscle-invasive) were correlated with stage, presence of visceral metastases and urinary source. The effects of cabozantinib on 4 human UC-derived cell lines were studied in vitro. Intact RT4, TCC-SUP, T24M2 and T24M3 cells at 80% confluence were serum deprived 16 h, then left untreated or treated with hepatocyte growth factor (HGF) and/or cabozantinib prior to analysis of Met, phospho- (p)Met, pAkt, Akt, pMAPK and MAPK by immunoassay or immunoblotting. Cabozantinib effects on basal and HGF-induced UC cell invasion, proliferation and soft agar growth were measured. Results: Median serum Met levels were modestly higher in pts with metastatic versus muscle-invasive disease. Urinary Met levels were clearly higher in pts with visceral metastasis (p=0.0111) and in urine from ileal conduits and neobladders compared to normally voided urine, regardless of stage (p=0.0489). Met content in UC cell lines was low in RT4 and higher in T24M2, T24M3 and TCC-SUP. Basal pMet content was universally low, increased significantly by HGF and this was reversed by cabozantinib. HGF-driven increases in pAkt/Akt and pMAPK/MAPK in all 4 cell lines were reversed by cabozantinib, as were HGF-enhanced UC cell invasion, proliferation and anchorage independent growth. Conclusions: Median urinary sMet is significantly higher in pts with visceral metastasis and in specimens from ileal conduits and neobladders relative to normally voided urine. UC cell Met content in culture increased with disease grade; HGF stimulated activation of Met and known effectors, and enhanced invasion, growth rate and anchorage-independent growth; cabozantinib effectively reversed these HGF-driven effects. These data support evaluation of cabozantinib in pts with metastatic UC.

Use of supportive measures to improve outcome and decrease toxicity in docetaxel-based antiangiogenesis combinations

128

Background: We have recently completed accrual of 63 patients (pts) to our study combining lenalidomide (L), with bevacizumab (B), docetaxel (D), and prednisone (P) (ART-P). Due to the lack of improved survival and the increased toxicity of anti-angiogenic docetaxel combinations in the MAINSAIL and CALGB 90410 trials we attempted to contrast and compare our studies with the failed phase III trials. Methods: Among the first 52 pts on the ART-P, 3 received L 15 mg daily, 3 had 20 mg daily, and the rest had 25 mg daily for 14 days of every 21−day cycle (C). We later enrolled 11 more pts at L 15 mg. All pts received D 75 mg/m2 and B 15 mg/kg on day 1 with P 10 mg and enoxaparin daily throughout each C. Pegfilgrastim was given on day 2. Patients on CALGB 90410 received D 75 mg/m2 and B 15 mg/kg on day 1 with P 10 mg and on MAINSAIL received D 75 mg/m2, L 25 mg daily for 14 days of every 21−day cycle with daily P. Patients on CALGB 90410 and MAINSAIL did not receive enoxaparin or pegfilgrastim prophylactically. Results: Median number of Cs in ART-P was 16 (3−38). PFS was 22 months and median OS has not been reached. Pts with measurable disease had 1 CR and 25 PR (86.7% RR). Two patients (3%) had deep vein thromboses. Of 1,219 cycles given, 14 cycles were complicated by febrile neutropenia (FN) (1.1%). There were no treatment related deaths. In comparison, median number of Cs in MAINSAIL L+DP arm was 6, with a PFS of 45 weeks and an OS of 77 weeks. Thirty-four pts (6.5%) developed pulmonary emboli and there were 2 deaths due to toxicity in the experimental arm. Nearly 12% of Cs were complicated by FN. In the experimental arm of CALGB 90410 trial, median OS was 22.6 months with median PFS of 9.9 months. Median number of Cs was 8, and 19 pts developed thromboses/emboli (3.6%). In addition, 7% of patients developed FN and treatment related deaths were reported at 4%. Conclusions: The use of supportive care allows the ART-P combination to be given for more cycles than were given in MAINSAIL and CALGB 90401 potentiating a longer PFS, RR and possibly OS with an improved toxicity profile. This data demonstrates the potential importance of supportive measures and is hypothesis generating for future combination studies. Clinical trial information: NCT00942578.

Dual antiangiogenic therapy using lenalidomide and bevacizumab with docetaxel and prednisone in patients with metastatic castration-resistant prostate cancer (mCRPC)

4569

Background: Previously, we had shown the potent anti−tumor activity of dual anti-angiogenic therapy by combining bevacizumab (B) and thalidomide (T) with docetaxel (D) and prednisone (P) in mCRPC (Ning JCO 2010). We hypothesized that combining lenalidomide (L), an analogue of T, with B, D, and P would have a more favorable efficacy/toxicity profile. Methods: All patients (pts) had chemotherapy−naïve mCRPC. Among the first 52 pts, 3 received L 15 mg daily, 3 had 20 mg daily, and the rest had 25 mg daily for 14 days of every 21−day cycle (C). The protocol was recently amended to enroll 11 more pts at L 15 mg; 2 pts have now been enrolled in this expansion cohort. All pts received D 75 mg/m2 and B 15 mg/kg on day 1 with P 10 mg and enoxaparin daily throughout each C. Pegfilgrastim was given on day 2. PSA each C with imaging after C2 and after every 3C. Dental exams with mandible CT scan at baseline, after C5, and every 6C. Results: 54 of 62 pts have been enrolled. Median age 65.5 (51−82), Gleason score 8 (5−10), on−study PSA 85.2 ng/ml (0.15−3520), and pre−study PSA doubling time 1.49 months (0.52−6.73). Median number of Cs was 16 (3−38). PFS was 22 months and probability of survival at 12 months was 90%. Forty-six (85.2%) and 42 (77.8%) pts had PSA declines of ≥50% and ≥75%, respectively. Of 30 pts with measurable disease there were 1 CR and 25 PR (86.7% overall RR). 17/54 pts were off study for radiographic disease progression and 8/54 for other reasons. Grade ≥2 toxicities included neutropenia (34/54), anemia (23/54), thrombocytopenia (7/54), hypertension (12/54), perianal fistula (3/54), rectal fissure (1/54), myocardial infarction (1/54), and osteonecrosis of the jaw (ONJ) (12/54, 22.0%). At the time of diagnosis of ONJ, 7/12pts were on bisphosphonates (BP), 2/12 had used BP previously, and 3/12 never used BP. The incidence of ONJ was comparable to 18.3% reported by Ning et al. A recent study of carboplatin plus weekly docetaxel reported an incidence of 29.3%. Conclusions: Dual anti-angiogenic therapy with, B and L, plus D and P was associated with high PSA (85.2%) and tumor (86.7%) responses in mCRPC, with manageable toxicities. The incidence of ONJ is comparable to other studies.

A pilot study on the clinical value of 18F-sodium fluoride PET/CT in advanced prostate cancer

10589

Background: We evaluated the clinical utility of 18F-sodium fluoride PET/CT bone scan (18F-NaF) in the detection of bone metastases in patients (pts) with prostate cancer in comparison with Technetium-99m MDP bone scan (TcBS). Methods: In a prospective study, from October 2010-December 2011, 30 prostate cancer pts (ages 51-79), 21 with known bone metastases and 9 without known bone metastases, had18F-NaF and a TcBS performed. Abnormal foci of uptake on both TcBS and 18F-NaFwere classified as benign, malignant or indeterminate. Benign lesions included uptake in the joints and linear uptake at the endplates of the vertebral bodies consistent with degenerative changes. Malignant uptake on 18F-NaF scans was confirmed by characteristic osteoblastic features on CT. All TcBS and 18F-NaF were reviewed by an experienced nuclear medicine physician. For the patient-based analysis, scan results were categorized as positive (POS) = any malignant lesion; indeterminate (IND) = not distinctly malignant or benign; negative (NEG) = benign lesions only. Results: In the lesion-based analysis, 21 of 30 (70%) pts had more malignant lesions identified on 18F-NaF than on TcBS. The mean number of additional malignant lesions per patient on 18F-NaF vs TcBS was 4. Eight of the 30 pts had same number of malignant lesions identified in both studies. One of 30 pts had one less malignant lesion identified on 18F-NaF than on TcBS. CT correlation by 18F-NaF PET/CT of this particular lesion did not confirm osteoblastic feature. Malignant lesion distribution on 18F-NaF included: spine (28%), thorax (26%), pelvis (24%), long bones (13%) and skull (10%). In the patient-based analysis, 24 pts (80%) were POS by 18F-NaF, of whom 14 pts were POS, 8 were IND, and 2 were NEG by corresponding TcBS; in the 4 pts with NEG 18F-NaF, zero were POS, 2 were IND and 2 were NEG by corresponding TcBS. Conclusions: 18F-NaF identified more malignant lesions than TcBS. 18F-NaF may also add useful information in the management of advanced prostate cancer pts with and without known bone metastases.

A phase I study of TRC105 (Anti-CD105 [endoglin] antibody) in metastatic castration resistant prostate cancer (mCRPC)

3043

Background: Pre−clinical and clinical evidence demonstrates an important role for angiogenesis in mCRPC biology. CD105 (endoglin) is a transmembrane protein expressed on the surface of proliferating vascular endothelial cells. The expression of CD105 is required for the formation of new blood vessels. TRC105 is a human/murine chimeric IgG1 kappa monoclonal antibody that binds to human CD105 (endoglin). It inhibits angiogenesis and tumor growth through inhibition of endothelial cell proliferation, antibody-dependent cellular cytotoxicity, and induction of apoptosis. The primary objective is to evaluate safety and identify the maximum tolerable dose (MTD) of TRC105. Secondary objectives include the assessment of TRC105 pharmacokinetics, PSA response rate, evaluation of progression free survival (PFS), overall response rate (ORR) and overall survival (OS). Methods: Patients with an ECOG performance status (PS) ≤ 2, progressive mCRPC and either chemotherapy-naïve or post-docetaxel treatment were eligible. Six cohorts of patients, on escalating dose levels, receive TRC105 intravenously at doses of 1, 3, 10, 15, or 20 mg/kg IV every 2 weeks (cohorts 1, 2, 3, 5, and 6) or 10 mg/kg IV weekly (cohort 4) on a 4 week cycle. Response is assessed with imaging studies every 2 months for the first four months and then every 3 months thereafter. Results: Sixteen patients are enrolled in cohorts 1-5. Median age is 65 (range 48-87), median ECOG PS is 1 (range 0−2), median Gleason score is 8 (range 6−10), median on−study PSA is 147.5 (range 0.1-3373), and median number of prior (non-hormonal) therapies is 3 (range 0−6). Median time on study is 16 weeks (range 8-28 weeks). One patient experienced a dose limiting toxicity (grade 4 vasovagal episode) in cohort 5. PSA declines were seen in 6 patients ranging from 20% to 57% from baseline. Ten out of 12 patients with measurable soft tissue disease achieved stable disease for at least two cycles. Conclusions: TRC105 is tolerated up to 15 mg/kg every two weeks with early evidence of clinical activity in mCRPC. An additional cohort (6), with dosage of 20 mg/kg, is currently under investigation. Accrual is ongoing to evaluate ORR, PFS, and OS in the phase II portion of this study.

External validation of prognostic models for overall survival (OS) in patients (pts) with advanced cancer (UC) treated with cisplatin-based chemotherapy

4592

Background: The most commonly used model predicting OS for UC pts treated with cisplatin-based chemotherapy is based on 2-variables (visceral metastases and performance status), developed at MSKCC in 1999, and validated in a phase III study (DeSantis JCO 2011). A prognostic model of OS for advanced UC pts based on 4 variables (visceral metastases, albumin, performance status, and hemoglobin) was developed using 308 pts from MSKCC (ASCO 2007 abstr 5055). We report the discriminative ability of the 4- and 2- variable models for advanced UC pts using an independent dataset from CALGB 90102. Methods: The analysis was performed using an external multi-institutional dataset from CALGB 90102. The primary measurement of predictive discrimination was Harrell’s c-index which was computed with 95% confidence interval (CI). To assess whether there was a statistically significant difference in discrimination between the two models, the U statistic was used to test whether the predictions of the 4-variable model in all possible pairs were more concordant with actual observations than the 2-variable model in the same pairs. Results: CALGB 90102 included 74 UC pts (58 males, 16 females), median age 64 years, treated with cisplatin, gemcitabine and gefitinib, enrolled from 7/02 to 4/05 with a median follow-up of 72.5 months. Visceral metastases were present in 64% (bone, 18%, liver, 31%, lung, 43%), median KPS 90%. The MSKCC 2-variable risk group distribution was 30% =0, 65%= 1 and 5%=2. The median OS =12.7 months (95% CI=10.4-20.5) with 68 deaths observed. When applied to the CALGB cohort, the predictive accuracy for the 4- and 2-variable models were 0.63 (95 CI= 0.56- 0.69) and 0.58 (95% CI= 0.52-0.65), respectively. There was a statistically significant difference in discrimination between the two models (p =0.019), with superiority of the 4-variable model compared to the 2-variable model. Conclusions: A 4-variable prognostic nomogram for survival in pts with advanced UC was superior to a 2-variable risk-group model. The 4-variable prognostic model may replace the widely used 2-variable model and can be used in the design and conduct of future phase II and III trials in advanced UC.

Phase II trial of neoadjuvant gemcitabine (G) and cisplatin (C) with sunitinib in patients (pts) with muscle-invasive bladder cancer (MIBC)

4581^

Background: Response to neoadjuvant chemotherapy (NC) prior to radical cystectomy (RC) predicts improved overall survival (OS) in MIBC. Associations with enhanced survival include complete pathologic response (pT0; Grossman, NEJM 2003) and eradication of the muscle-invasive component (<pT2; Splinter, J Urol 1992). Sunitinib (S) is active in pretreated pts with advanced disease. We tested if S added to GC was safe, improved the rate of pT0, and improved the rate of <pT2. Methods: Cisplatin-eligible pts with cT2-4aN0 bladder cancer received G 1000 mg/m2 and C 35 mg/m2 on day (D) 1 and D8 with S 25 mg orally daily D1-14 of a 21D cycle for 4 cycles. RC plus pelvic lymph node dissection was required to assess response of pT0 or <pT2. A Simon’s Minimax 2-stage design was used to test a null (H0) pT0 rate ≤ 20% against alternative (H1) pT0 rate ≥ 40% with Type I and II error rates of 0.05 and 0.10 respectively. Enrollment to the 2nd stage of 45 patients was planned if ≥ 6 of the initial 24 evaluable pts achieved pT0. Primary endpoint was pT0N0 and secondary endpoints were: response defined as <pT2N0; safety; time to progression (TTP), and OS. Results: 18 pts (15M, 3F), median age 63 (54-76) were enrolled from 6/09 and 10/11 after which financial support was withdrawn. 3 pts were inevaluable for response endpoints due to: 1.) withdrawal of consent, 2.) declining any surgery, 3.) partial cystectomy instead of RC. All 18 were evaluable for safety, TTP and OS. 1 of 15 pts had pT0N0 (6.6%; 95% CI 0.34 – 29.8%) and 5 had <pT2N0 (33%; 95% CI 15-58%). 4 of 5 pts with status <pT2N0 were pTisN0. Median TTP was 10 months (95% CI 3.5-NR). 3 pts were deceased at time of analysis; median OS not reached. Neutropenia due to the 3 drug combination required routine GCSF support on day 8 of each cycle. Grade 3/4 toxicities were anemia (11 pts), neutropenia (6), thromboembolic events (2), febrile neutropenia (2) and infection (2). Conclusions: Despite incomplete accrual, the pT0 rate was low suggesting that S does not add to GC. Residual non-invasive disease (<pT2) was common, including a large proportion of pts with pTisN0. Given these findings, response criteria for future NC studies should consider either <pT2 or < pTis as the primary endpoint.

Tumor regression and growth rates determined in two intramural NCI prostate cancer trials: The growth rate constant as an indicator of therapeutic efficacy

133

Background: Like ATTP [bevacizumab + thalidomide + docetaxel + prednisone], ARTP [bevacizumab + lenalidomide + docetaxel + prednisone] is active in CRPC and in both the data suggest acquired resistance does not develop, supporting a strategy that continues therapy if tolerable. Discerning amongst new therapies in CRPC would be helped by novel assessment strategies that yield answers from smaller trials and allow comparisons across trials. We have validated a novel assessment method that quantifies tumor regression (d) and growth (g) rate constants using data obtained while pts are treated in a trial. We utilized this method to evaluate a phase II trial of ARTP in CRPC contrasting this with a previous study using thalidomide instead of lenalidomide (ATTP).

Methods: Using PSA values and a two-phase mathematical equation we determined d and g. A three-phase equation was used to determine resistant fractions.

Results: The median log g value with ARTP (-2.84) was statistically similar (p=0.204) to that observed with ATTP (-3.16). Both therapies had similar effects on log d (ARTP=-2.18; ATTP=-2.64; p=0.404). In a subset of pts with robust data both regimen are highly effective with only 3.01% and 5.46% of tumor resistant to ATTP and ARTP, respectively. In individual pt, statistically valid g and d values could be estimated after the fourth PSA value had been obtained, long before PSA increased - providing an early indicator of treatment failure. In most pts receiving prolonged treatment – as long as 700 d – the growth rate constant did not change, despite rising PSA values, indicating acquired resistance did not develop, and that if tolerable, therapy can be continued for longer periods of time.

Conclusions: The substitution of daily thalidomide with lenalidomide 14/21 days in ARTP resulted in a combination statistically as effective as ATTP. As with ATTP majority of pts experienced marked reductions in the tumor’s g, and surprisingly in most, there was no evidence of acquired resistance, g remaining unchanged over prolonged time periods. Given the tolerability of this combination, pts with marked reductions in g may benefit from prolonged ARTP therapy.

Phase II trial of bevacizumab and lenalidomide with docetaxel and prednisone in patients with metastatic castration-resistant prostate cancer (mCRPC)

207

Background: Angiogenesis may be vital to mCRPC. Previously, we had shown the potent anti−tumor activity of dual antiangiogenic therapy by combining thalidomide (T) and bevacizumab (B) with docetaxel (D) and prednisone (P) in mCRPC (Ning JCO 2010). We hypothesized that combining lenalidomide (L), an analogue of T, with B, D, and P would have a more favorable efficacy/toxicity profile.

Methods: All patients (pts) had chemotherapy−naïve mCRPC. 3 pts received R 15 mg daily, 3 pts had 20 mg daily, and the rest had 25 mg daily for 14 days of every 21−day cycle (C). All pts received D 75 mg/m2 and B 15 mg/kg on day 1 with P 10 mg and enoxaparin daily throughout each C. Pegfilgrastim was given on day 2. PSA was assayed each C with imaging after C2 and then after every 3C.

Results: 47 of the planned 51 pts have been enrolled. Median age was 66 (51−82), Gleason score 8 (5−10), on−study PSA 91.6 ng/ml (0.15−3520), pre−study PSA doubling time 1.43 months (0.52−6.73), number of Cs 14 (1−31), and PFS was 19.3 months as of this analysis. Among 45 pts who have completed ≥2 cycles, 39 (86.7%) and 30 (66.7%) had PSA declines of ≥50% and ≥75%, respectively. Of 29 pts with measurable disease there were 2 CR, 21 PR, and 6 SD (79.3% overall RR). 10/47 pts were taken off study for radiographic disease progression and 5/47 for other reasons. Grade ≥3 toxicities included neutropenia (24/47), anemia (9/47), thrombocytopenia (5/47), weight loss (1/47), hypertension (3/47), and febrile neutropenia (4/47). Other toxicities included perianal fistula (3/47), rectal fissure (1/47), myocardial infarction (1/47), and osteonecrosis of the jaw (ONJ) (16/47, 34.0%). At the time of diagnosis of ONJ, 9/16 pts were on bisphosphonates and 3/16 had used bisphosphonates previously. Although the incidence of ONJ was higher than the 18.3% reported by Ning, a recent study of carboplatin plus weekly docetaxel reported an incidence of 29.3%.

Conclusions: Dual antiangiogenic therapy with, B and L, plus D and P was associated with high PSA (86.7%) and tumor (79.3%) responses with manageable toxicities. Further studies are underway to explore the high incidence of ONJ.

A phase I study of TRC105 (anti-CD105 [endoglin] antibody) in metastatic castration-resistant prostate cancer (mCRPC)

117

Background: Preclinical and clinical evidence demonstrates an important role for angiogenesis in mCRPC biology. CD105 (endoglin) is a transmembrane protein expressed on the surface of proliferating vascular endothelial cells. The expression of CD105 is required for the formation of new blood vessels. TRC105 is a human/murine chimeric IgG1 kappa monoclonal antibody that binds to human CD105 (endoglin). It inhibits angiogenesis and tumor growth through inhibition of endothelial cell proliferation, antibody-dependent cellular cytotoxicity, and induction of apoptosis. The primary objective is to evaluate safety and identify the maximum tolerable dose (MTD) of TRC105. Secondary objectives include the assessment of TRC105 pharmacokinetics, PSA response rate, evaluation of progression free survival (PFS), overall response rate (ORR) and overall survival (OS).

Methods: Patients with an ECOG performance status (PS) ≤ 2, progressive mCRPC and either chemotherapy-naïve or post-docetaxel treatment were eligible. Five cohorts of patients, on escalating dose levels, receive TRC105 intravenously at doses of 1, 3, 10 or 15 mg/kg IV every 2 weeks (cohorts 1, 2, 3, and 5) or 10 mg/kg IV weekly (cohort 4) on a 4 week cycle. Response is assessed with imaging studies every 2 months for the first four months and then every 3 months thereafter.

Results: Seventeen patients are enrolled in cohorts 1-5. Median age is 65 (range 48-87), median ECOG PS is 1 (range 0−2), median Gleason score is 8 (range 6−10), median on−study PSA is 147.5 (range 0.1-3373), and median number of prior (non-hormonal) therapies is 3 (range 0−6). Median time on study is 16 weeks (range 8-28 weeks). One patient experienced a dose limiting toxicity (grade 4 vasovagal episode) in cohort 5. PSA declines were seen in 6 patients ranging from 20% to 57% from baseline. Ten out of 12 patients with measurable soft tissue disease achieved stable disease for at least two cycles. Two patients remain on study (in cohort 5).

Conclusions: TRC105 is tolerated up to 15 mg/kg every two weeks with early evidence of clinical activity in mCRPC. Accrual is ongoing to evaluate ORR, PFS, and OS in the phase II portion of this study.

Examining the management of muscle-invasive bladder cancer (MIBC) by medical oncologists (MO)

298

Background: Neoadjuvant chemotherapy (NC) for the treatment of MIBC remains underutilized in the United States despite evidence supporting its use. The aim of this project is to examine the current management of MIBC by MO to move towards standardization of practice.

Methods: An electronic 26 question survey was emailed to 92 MO belonging to the Bladder Cancer Advocacy Network or the American Society of Clinical Oncology, and posted on the US Oncology portal for 8 wks. The study was approved by the Office of Management and Budget (0925-0046). Percentages are based on the fraction that responded to a given question.

Results: Of the 83 respondents: 48% were non-academic. 51% were general oncologists and 45% focused on GU malignancies. The majority of MIBC referrals came from urologists (79%). Initial CT staging with abd/pelvis was required by 88%, CT chest by only 72% and PET by 21%. NC is offered by 79% of MO to all MIBC pts and by 45% to all pts with high grade upper tract urothelial carcinoma. Adjuvant chemotherapy (AC) was offered by 46% to all MIBC pts and by 41% to pts with upper tract disease. NC was not offered if ECOG performance status (PS) was >3 (49%), in T2 disease without lymphovascular invasion (29%), or with GFR of <50 ml/min (35%). Chemotherapy regimens used for NC included gemcitabine/cisplatin (90%); methotrexate/vinblastine/adriamycin/cisplatin (MVAC) (30%), dose dense MVAC (20%); gemcitabine/carboplatin (36%); single-agent gemcitabine (10%), other regimens described include carboplatin/paclitaxel and ifosfamide/doxorubicin/gemcitabine. Response to NC was assessed by CT abd/pelvis (82%), CT chest (39%), cystoscopy (30%) and PET (12%). Pts with pathologic residual disease (>pT2 or positive LN) after NC were generally observed (74%).

Conclusions: The majority of MO do offer perioperative chemotherapy for MIBC pts with a trend towards increased use of NC over AC, which follows best evidence. The main reason for not offering perioperative chemotherapy was poor PS. The majority offer cisplatin-based combination therapy by preference. Increasing the rate of referrals of MIBC pts to MO, will result in more pts receiving perioperative cisplatin-based chemotherapy and may lead to better outcomes in this disease.