Bev Treatment Research Articles

NIMG-07RADIOGRAPHIC PATTERNS OF PROGRESSION WITH ASSOCIATED OUTCOMES AFTER BEVACIZUMAB THERAPY IN GLIOBLASTOMA PATIENTS

INTRODUCTION: Patterns of progression following bevacizumab (bev) treatment and associated outcomes remain poorly characterized. In patients (pts) with glioblastoma (GB) treated with bev, we describe radiographic patterns of progression and their association with outcome. METHODS: 64 pts treated at MD Anderson matched the predetermined inclusion criteria. Tumor progression after bev treatment was assessed according to the RANO criteria and pts categorized into groups based on previously published data: Group1:exclusively T2-diffuse hyperintense tumor (T2-diffuse), Group2:initial decrease and subsequent flare-up of contrast enhancement (CE) at progression (cT1 Flare-up), Group3:no decrease in CE or development of new lesions at first follow-up imaging (non- responders), Group4:exclusively T2-circumscribed hyperintense tumor progression (T2-circumscribed). In addition, we screened for new diffusion-restricted lesions or pre-contrast T1-hyperintense lesions or both (double-positive). RESULTS: Pts were categorized into Group1:11%, Group2: 33%, Group3: 45%, Group4: 11%. 16 pts had T1-hyperintense lesions and 37 had restricted diffusion;10 pts had double-positive lesions. There was no significant difference in time-to-initiation of bev treatment in the 4 groups. After starting bev, median OS and PFS (months) was Group1:8.6, 4.2 Group2:12.3,3.9 Group3:5.6,1.4 and Group4:7.0,3.2 respectively. Comparing non-responders vs the rest of the groups (responders), OS from initiation of bev was 5.6 vs 10 months (p = <0.001). OS from diagnosis of GB was not significantly different between the 4 groups. Post-progression on bev, OS was: Group1:3.9,Group2:6.4, Group3: 3.8 and Group4:3.0. Pts with restricted diffusion had worse OS (23.4 vs17.4 months). There was no difference in survival based on presence/absence of T1-hyperintense or double-positive lesions. CONCLUSION: Compared to non-responders, cT1 Flare-up have better PFS/OS after bev initiation. Post-progression on bev, cT1 Flare pts lived significantly longer than T2-diffuse and non-responders. Though T2-diffuse and T2-circumscribed have a better PFS after bev initiation compared to non-responders, no OS benefit was seen. Restricted diffusion was a radiographic marker of worse outcomes.

Amino acid PET tracers are reliable markers of treatment responses to single-agent or combination therapies including temozolomide, interferon-β, and/or bevacizumab for glioblastoma

IntroductionWe examined whether the amino acid PET tracers, trans-1-amino-3-18F-fluorocyclobutanecarboxylic acid (anti-18F-FACBC) and 11C-methyl-l-methionine (11C-Met), are suitable for detecting early responses to combination therapies including temozolomide (TMZ), interferon-β (IFN), and bevacizumab (Bev) in glioblastoma. MethodsHuman glioblastoma U87MG (U87) cells were incubated with low dose TMZ to induce chemoresistance. Both trans-1-amino-3-fluoro-1-14C-cyclobutanecarboxylic acid (anti-14C-FACBC) and 3H-methyl-l-methionine (3H-Met) uptake were quantified using triple-label accumulation assays to examine the relationship between tracer uptake and proliferation (3H-thymidine (TdR) accumulation) in vitro. U87 and U87R (TMZ-resistant subculture) cells were inoculated into the right and left basal ganglia, respectively, of F344/N-rnu rats. The efficacy of single-agent (TMZ, Bev) and combination therapy (TMZ/IFN, TMZ/Bev, TMZ/IFN/Bev) was examined in orthotopic gliomas using MRI, Evans blue extravasation, anti-14C-FACBC, and 3H-Met autoradiography, and MIB-1 immunostaining. ResultsTMZ treatment decreased 3H-TdR accumulation and the volume distribution of anti-14C-FACBC and 3H-Met in U87 but not U87R cells. TMZ/IFN combination therapy significantly decreased these parameters in U87R cells; however, Bev had no additional effect in vitro. In vivo, U87R-derived gliomas were observed as equivocal tumors on MRI and T2-high intensity lesions. Bev treatment, either alone or in combination, markedly decreased U87 enhancing lesions. By contrast, autoradiographic images using anti-14C-FACBC and 3H-Met clearly delineated tumor extent, which spread widely beyond T2-high intensity lesions and enhancing lesions. TMZ therapy significantly decreased tracer accumulation and proliferation of U87- but not U87R-derived tumors. TMZ/IFN combination treatment significantly decreased these parameters in U87R tumors, which were further reduced (in both tumor types) by Bev addition. Tracer uptake correlated with the MIB-1 proliferation index. However, MRI was unsuitable for tumor delineation and assessment of Bev treatment response. ConclusionsTriple-agent therapy (TMZ/IFN/Bev) was effective against even TMZ-resistant glioblastomas. PET with amino acid tracers provides useful information on the early response of glioblastomas to single-agent and combination therapy.

AT-20 * SINGLE INSTITUTION RETROSPECTIVE REVIEW OF PERFORMANCE STATUS AND CORTICOSTEROID USE IN NEWLY DIAGNOSED GLIOBLASTOMA (GBM) PATIENTS (PTS) TREATED ON BEVACIZUMAB (BEV)

BACKGROUND: Two multicenter phase III trials, RTOG0825 and AVAglio, evaluated the addition of BEV to radiation and temozolomide in newly diagnosed GBM and demonstrated a lack of statistically significant increased survival and discordant health-related quality of life results. We performed a retrospective chart review in newly diagnosed GBM to understand clinical benefit in BEV-treated patients in the community. METHODS: Newly diagnosed adult GBM pts treated at Duke were identified through our Primary and Recurrent Glioblastoma Registry. We conducted a retrospective chart review of pts who initiated BEV between 1/1/2009 and 5/14/2012. Pts who participated in a clinical trial prior to receiving BEV were excluded. Pts alive on 5/15/2012 consented to the prospective collection of their data. Demographic and clinical data, including corticosteroid use and Karnofsky performance status (KPS), were analyzed. OS, PFS and time to KPS reduction were measured from BEV initiation. RESULTS: A total of 188 pts have been analyzed. At BEV initiation, median age was 57 years, 64% were male, 86% had KPS ≥80, and 46% were on corticosteroids. Median OS was 23.3 months (95% CI, 20.0-26.2 months). Median PFS was 14.8 months (95%CI, 12.2-17.9 months). Median time to KPS reduction ≥10-points and ≥20-points were 10.8 and 34.0 months, respectively. Of 86 pts on steroids pre-BEV, 78% discontinued steroids at a median of 64 days. Of 102 pts not on steroids pre-BEV, 57% did not initiate steroids, 31% initiated steroids while on BEV, and 12% initiated steroids while on another regimen. CONCLUSION: In newly diagnosed GBM pts treated on BEV in clinical practice, this retrospective review suggests a median OS and median PFS of 23.3 and 14.8 months, respectively. A majority of patients discontinued steroids, or remained off steroids during BEV treatment. Initial KPS levels were maintained by 86% of pts through the first year of follow-up.

AI-25 * THE NOVEL c-MET INHIBITOR ALTIRATINIB (DCC-2701) INHIBITS TUMOR GROWTH AND INVASION IN A BEVACIZUMAB RESISTANT GLIOBLASTOMA MOUSE MODEL

Glioblastoma with acquired resistance to bevacizumab therapy overexpresses c-MET which mediates tumor growth, EMT, invasion, and metastasis. We evaluated the expression level of total and phosphorylated c-MET (p-c-MET) and the inhibitory activity of altiratinib (DCC-2701), a novel c-MET/TIE2/VEGFR inhibitor (Deciphera Pharmaceuticals) that has demonstrated significant BBB penetration into the CNS. Four of 12 (30%) mesenchymal glioma stem cell lines (GSCs) highly expressed c-Met and were sensitive to altiratinib treatment in vitro for blocking cell viability. After HGF stimulation, the expression level of p-cMET was significantly increased in these 4 cell lines. Altiratinib markedly suppressed HGF stimulated p-c-MET expression in both paracrine (MET+) and autocrine (MET + /HGF+) GSC cell lines. In an intracranial glioma mouse model, animals were treated with bevacizumab (Bev) alone, altiratinib alone or the combination. Mice were euthanized at 3, 4 and 5 weeks. H&E staining revealed a significant reduction in tumor size in the combination group compared with the tumors treated with either agent alone. Altiratinib alone or combination with Bev treatment dramatically suppressed EMT and tumor invasion in vivo. This anti-invasive effect correlated with decreased N-cadherin and vimentin levels. In vivo, microvascular density was significantly inhibited in the groups treated with altiratinib and altiratinib plus Bev compared with control or the group treated with Bev alone. Similarly, F4/80+ BMDC cell infiltration was significantly suppressed in the groups treated with altiratinib and altiratinib plus Bev combination compared to control Bev-treated groups. In summary, our data indicated that altiratinib inhibited tumor growth, EMT and invasion both in vitro and in vivo. The combination of altiratinib and anti-VEGF therapy may provide a new strategy to overcome Bev resistance and prolong overall survival in patients with glioblastoma.

P3-9-4 - Importance of Blood Pressure Monitoring in Colorectal Cancer Patients Treated with Bevacizumab: the Current Situation

Background: Hypertension associated with bevacizumab (Bev) is a well-known adverse drug reaction. However, we tend to overlook the symptom as other adverse drug events are induced by chemotherapy combined with Bev. In order to highlight the current issues in our hospital on blood pressure monitoring during Bev treatment we conducted this study.Methods: Thirty one colorectal cancer patients treated with Bev, from August 2007 to September 2011, at Fujita health university hospital were enrolled in this study. We retrospectively researched antihypertensive medication use and blood pressure change in the patients. The common terminology criteria for adverse events ver.3.0 (CTCAE) was used for adverse event classification.Results: Blood pressure elevation was observed in 14 (grade1 (G1); 45.1%), 6 (grade2 (G2); 19.3%) and 1 patient (grade 3 (G3); 3.2%). The number of Bev cycles administered was 2-22 (G1; median 3), 4-24 (G2; median 7.5) and 20 (G3). 5/6 patients (G2) showed G1 hypertension at early Bev treatment cycles then reached G2 with following treatment. 1 patient (G3) also showed G3 hypertension after G1/G2 at early cycles. Ca-blockers were prescribed to two (G2) and one patients (G3). In G3 case, Ca-blocker was added to ARB. 21 patients (67.7%) showed systolic blood pressure >140mmHg, which hypertension treatment guidelines recommend medication.Conclusions: The current issue in our hospital is that Bev-induced hypertension is not well controlled. It is necessary that the adverse drug reaction is controlled with medication as chemotherapy with Bev have resulted in good prognosis. Moreover, it is important to start hypertension control at G1 because G2/G3 is caused from uncontrolled hypertension during Bev treatment. Chemotherapy with Bev is given outpatients. It is considered that building a blood pressure monitoring system for outpatients and promoting patients education on home blood pressure measurement are urgent tasks.

Abstract CT222: Differences in pharmacokinetics of TRC105 (anti-endoglin antibody) when administered as a single agent versus in combination with bevacizumab (Bev)

Abstract Background: TRC105 is an anti-endoglin chimeric monoclonal antibody that inhibits angiogenesis and tumor growth and is being studied in randomized Phase 2 trials with Bev. TRC105 is cleared through binding to endoglin expressed on proliferating endothelium when given as a single agent to cancer patients (Spencer et al, ASCO 2012). Preclinical data indicate that endoglin expression is increased in response to VEGF targeted treatment, and increased endoglin expression in response to Bev may increase the clearance of TRC105 when administered in combination to cancer patients. Methods: Patients (pts) with solid tumors (ST) or ovarian cancer (OC) received 10 mg/kg/wk TRC105 as a single agent, and additional patients with ST (primarily ovarian and colorectal) received 10 mg/kg/wk TRC105 with Bev. Peak and trough levels were assessed by ELISA in 39 patients who received TRC105 alone and compared to that of 35 patients (largely Bev refractory) who received TRC105 and Bev. Pts administered TRC105 and Bev were considered a population sub-group and treated as a covariate. A population pharmacokinetic model of TRC105 disposition was built using rich sampling from the ST trial, with sparse data from OC and TRC105+Bev included in the base model. A two-compartment model with nonlinear elimination best fit the data, utilizing Michaelis-Menten parameters for saturable clearance. Results: TRC105 peak and trough concentrations exceeded target serum concentrations of TRC105 known to saturate endoglin receptors in all pts dosed with 10 mg/kg/wk of TRC105 with and without Bev. The PK of TRC105 given with Bev had mean predicted (following 10,000 simulations) parameters of volume of distribution in the central compartment (VC), VMAX, and KM that were increased compared to population estimated parameters of TRC105 given as a single agent [VC= 44.5±2 (SE) (mL/kg) Pop mean estimate vs. 68.8±4 (mL/kg) Bev predicted; VMAX = 92.6±16 (μg/hr) vs. 297.5±40 (μg/hr) predicted, and KM= 5.91±2 (μg/mL) vs. 61.6±8.8 (μg/mL)]. All differences were significant (p&amp;lt;0.001). Observable data from patients administered the combination yielded PK parameters that were consistent with increased target-mediated clearance of TRC105 when given with Bev. Conclusions: Peak and trough TRC105 serum levels exceed target serum concentrations when given at 10 mg/kg/wk as a single agent or with Bev. Central compartment distribution of TRC105 increased when given with Bev, which is consistent with increased endoglin expression on proliferating endothelium following Bev treatment. The maximum rate of elimination (Vmax) also increased, consistent with increased turnover; however the intrinsic clearance ratio of Vmax/Km remained the same, suggesting no change in endoglin turnover efficiency. Future studies will assess whether PK parameters correlate with responses to the combination of TRC105 and Bev in Bev refractory patients. Citation Format: Shawn D. Spencer, Lee S. Rosen, Michael S. Gordon, Francisco Robert, Daniela Matei, Cody J. Peer, Bonne Adams, Delia Alvarez, Ben K. Seon, Charles P. Theuer, W. Douglas Figg. Differences in pharmacokinetics of TRC105 (anti-endoglin antibody) when administered as a single agent versus in combination with bevacizumab (Bev). [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr CT222. doi:10.1158/1538-7445.AM2014-CT222

434P - Rechallenge with Bevacizumab (Bev) in Patients (P) with Glioblastoma (Gbm) and Previous Objective Response to Bev

ABSTRACT Aim: GBM is the most common primary brain tumor and is characterized by high vascularitzation and overexpression of vascular endothelial growth factor A (VEGF-A). BEV has been approved in several countries for the treatment of recurrent GBM. While the optimal duration of treatment with BEV is unknown, treatment until progression or for a maximum of 1 year has been established in phase II studies. No therapy has demonstrated efficacy after BEV. Methods: We have retrospectively analyzed GBM p who stopped BEV after an objective response and received rechallenge with BEV at the time of recurrence. Results: 17 p were included. 7 p (41.2%) had received neoadjuvant BEV plus temozolomide (TMZ) followed by BEV/TMZ concomitant with radiotherapy as part of the GENOM 009 trial. 5 p had attained a partial response (PR) and 2 had stable disease (SD). 6 of these p received rechallenge with BEV at the time of the first recurrence and 1 at the time of the second recurrence. Response to BEV rechallenge in these 7 p: 1 complete response (CR) (14.3%); 4 PR (57.1%); 2 SD (28.6%). The remaining 10 p had received the standard EORTC regimen as first-line therapy; 6 then received second-line treatment with BEV at the first recurrence and 4 at the second recurrence. Median duration of BEV treatment was 15.8 months (m). Outcome to second-line BEV: 5 CR (50%); 5 PR (50%); progression-free survival (PFS), 24.1 m. Of these 10 p, 8 were rechallenged with BEV at the time of recurrence. Outcome to BEV rechallenge in 6 evaluable p: 4 PR (66.6%); 1 SD (16.7%); 1 progressive disease (16.7%); PFS from start of BEV rechallenge was 6.7 m. Median overall survival from the time of diagnosis is 38.66 m for all 17 p. Conclusions: The optimal duration of treatment with BEV at recurrence is unknown. In patients with an objective response, BEV can be stopped and new responses to rechallenge with BEV can be attained. A clinical trial is warranted to examine the comparative benefit of stopping or continuing treatment with BEV in responders. Disclosure: All authors have declared no conflicts of interest.

Abstract B2: Switching to aflibercept treatment resulted in greater tumor responses than continuous bevacizumab treatment in patient-derived xenograft models of colorectal cancer.

Abstract Background: The recombinant fusion protein aflibercept (afl) binds VEGF-A, -B, and placental growth factor (PlGF). Based on an OS benefit observed in the phase 3 VELOUR trial, aflibercept in combination with FOLFIRI was approved for patients with metastatic colorectal cancer (mCRC) following a prior oxaliplatin-containing regimen. In vitro, afl inhibited VEGFR signaling more potently and blocked cell migration more effectively vs. bevacizumab (bev) which only binds VEGF-A. Recent in vivo studies comparing the anti-tumor activity of afl vs. bev were conducted in patient-derived xenograft (PDX) CRC models. The PDX models are clinically relevant as they represent common genetic backgrounds in CRC and were derived from primary tumors or from metastases. A sequential treatment regimen of aflibercept post-bev vs. continuation of bev treatment was investigated. Methods: Forty eight different CRC adenocarcinoma tumors were engrafted subcutaneously (SC) into NMRI nude mice to generate 48 distinct PDX CRC models. ELISA and Luminex measured the expression of mouse and human VEGF-A and mouse and human PlGF in untreated tumors. Tumor-bearing mice (8 mice/group) were dosed with afl, bev, or placebo 2x/wk for 3 wks and SC tumor measurements were recorded 2x or 3x/wk. In 3 PDX models, mice that were treated with bev were randomized and either switched to afl treatment or continued to receive bev. Activity was expressed as relative change in tumor volume between treatment (ΔT) and control (ΔC) at 3 wks. A ΔT/ΔC ≤0% indicated complete tumor stasis. Student's t tests evaluated statistical differences in final tumor volumes. Results: Aflbercept induced complete tumor stasis in 31/48 PDX CRC models and bev in 2/48 PDX CRC models. The PDX models were characterized into 2 phenotypes based on statistical analysis: phenotype A, where afl activity was greater than bev activity (39 /48 PDX), and phenotype B, where there was no significant difference between the 2 drugs (9/48 PDX). Three models in which bev treatment resulted in eventual tumor escape vs. continuous inhibition by afl were selected for sequential treatment of bev followed by afl. Afl treatment began after bev treatment resulted in a greater than 2-fold increase in tumor volume. Greater anti-tumor activity occurred when afl was administered after bev vs. continuous bev treatment. At the end of the study, the tumor volumes were statistically significantly different between the afl and bev groups (T test p values &amp;lt; 0.0001). Analysis of angiogenic factors revealed that the tumor cells were the major source of VEGF whereas PlGF was primarily produced by the tumor stroma. Conclusions: Aflibercept induced complete tumor stasis more frequently in CRC PDX models than bevacizumab. As tumor levels of hVEGF-A were far greater than mVEGF-A, the inability of bevacizumab to bind mVEGF-A is unlikely to explain the higher and more consistent activity of aflibercept. In all 3 PDX models tested, switching to aflibercept treatment after bev treatment resulted in a greater tumor response than continuous bev treatment. The greater response to aflibercept than bev after an initial treatment with bev suggests that individual patients may derive more benefit from aflibercept than bev in second line treatment after an initial bev-based regimen. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):B2. Citation Format: Marielle Chiron, Rebecca G. Bagley, Jack Pollard, Christophe Henry, Parminder Mankoo, Loic Vincent, Catherine Geslin, Tatjana Kloss, Donald A. Bergstrom. Switching to aflibercept treatment resulted in greater tumor responses than continuous bevacizumab treatment in patient-derived xenograft models of colorectal cancer. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr B2.

Adjuvant bevacizumab as treatment for melanoma patients at high risk of recurrence: Preplanned interim results for the AVAST-M trial.

LBA9000 Background: Bevacizumab (Bev) is a recombinant humanized monoclonal antibody to vascular endothelial growth factor (VEGF) shown to improve survival in several advanced solid tumors. Multiple adjuvant trials are underway, but trials that have reported in colon and triple-negative breast cancer did not meet their primary end points. Since VEGF is a relevant target in melanoma, AVAST-M aimed to evaluate the role of Bev in patients (pts) with resected melanoma at high risk of recurrence. Methods: AVAST-M is a randomized phase III trial evaluating single agent Bev (7.5mg/kg IV 3 weekly for 1 year) as adjuvant therapy following resection of AJCC stage IIB, IIC, and III cutaneous melanoma compared to standard observation (Obs). 1,320 pts were required to detect 8% differences in 5-year overall survival (OS) rate from 40% to 48%; 85% power, 5% alpha level. Primary endpoint is OS; secondary endpoints are disease free interval (DFI), distant-metastasis free interval (DMFI), safety, and quality of life (QoL). An associated translational study is ongoing. Results of the first pre-planned interim analysis (agreed by the IDSMC) are reported here. Results: Between July 2007 and March 2012, 1,343 pts were recruited. 56% were male; median age 56 years (range 18-88 years), 16% were stage IIB, 11% IIC, 15% IIIA, 36% IIIB, 20% IIIC, and 2% unknown stage. Ulceration status of the primary melanoma was: 38% present, 45% absent, 17% unknown. At the time of the interim analysis, 286 (21%) patients had died. Median follow-up for survival was 25 months. Median duration of Bev treatment in 671 treated pts was 51 weeks (dose intensity 86%). Main outcomes are shown in the table. Grade 3/4 adverse events were experienced in 101 (15%) Bev pts and 36 (5%) Obs pts. Conclusions: Interim analysis of this large, multicenter trial of melanoma patients at high risk of recurrence has shown that adjuvant Bev monotherapy is well tolerated and improved DFI. Longer follow-up is required to determine an impact on the primary endpoint of 5-year OS. Clinical trial information: 81261306. [Table: see text]

RTOG 0825: Phase III double-blind placebo-controlled trial evaluating bevacizumab (Bev) in patients (Pts) with newly diagnosed glioblastoma (GBM).

1 Background: Chemoradiation (CRT) with temozolomide (TMZ/RT→TMZ) is the standard of care for newly diagnosed GBM. This trial determined if the addition of Bev to standard CRT improves survival (OS) or progression-free survival (PFS) in newly diagnosed GBM. Methods: This phase III trial was conducted by the RTOG, NCCTG, and ECOG. Neurologically stable pts &gt; 18 yrs with KPS ≥ 60, and &gt; 1cm3 tumor tissue block, were randomized to Arm 1: standard CRT + placebo or Arm 2: standard CRT plus Bev (10 mg/kg iv q 2wks). Experimental treatment began at wk 4 of radiation then thru 6-12 cycles of maintenance chemotherapy. Protocol specified co-primary endpoints were OS and PFS, with significance levels of .023 and .002, respectively. At progression, treatment was unblinded and pts allowed to crossover or continue Bev. Symptom, QOL and neurocognitive (NCF) testing was performed in the majority of pts. Secondary analyses evaluated impact of MGMT methylation (meth) and prognostic 9 gene signature status. Results: From 978 registered pts, 637 were randomized. Inadequate tissue (n=105) and blood on imaging (n=40) were key reasons for non-randomization. No difference was found between arms for OS (median 16.1 vs. 15.7 mo, p = 0.11). PFS was extended for Arm 2 (7.3 vs. 10.7 mo, p = 0.004). Pts with MGMT meth had superior OS (23.2 vs. 14.3 mo, p &lt; 0.001) and PFS (14.1 vs. 8.2 mo, p &lt; 0.001). Neither the 9 gene signature nor MGMT predicted selective benefit for Bev treatment, but best prognosis pts (MGMT meth, favorable 9-gene), had a worse survival trend with Bev (15.7 vs 25 mo p = 0.08). To date, 128 pts were unblinded on Arm 1 (salvage Bev in 86) and 87 pts on Arm 2 (continued Bev in 39). Increased grade ≥ 3 toxicity was seen with Bev, mostly neutropenia, hypertension, and DVT/PE. Conclusions: The addition of Bev for newly diagnosed GBM did not improve OS, did improve PFS but did not reach the significance criterion. MGMT and 9 gene profile did not identify selective benefit, but risk subset results suggested strongly against the upfront use of Bev in the best prognosis pts. Full interpretation of the PFS results incorporating symptom burden, QOL, and NCF is ongoing. Support: NCI U10 CA 21661, U10 CA37422, and Genentech. Clinical trial information: NCT00884741.

Post-bevacizumab treatment and clinical outcomes in recurrent malignant glioma.

2098 Background: Bevacizumab (Bev) and irinotecan combination therapy is effective against recurrent malignant glioma. However, post-Bev treatment and its clinical outcomes are not well investigated. Methods: We identified 103 consecutive recurrent malignant glioma patients who received Bev plus irinotecan at our institutions.Clinical records and magnetic resonance images were reviewed. Response and progression were assessed by RANO criteria. Results: Bev and irinotecan treatment produced response rate of 37.9% (95% CI, 29.1-47.5%). At a median follow-up time of 41 weeks, the median progression-free survival (PFS) was 17.1 weeks (95% CI, 14.3-20.0), and 6-month PFS (6M-PFS) was 27.8% (95% CI, 18.4-37.2). The median overall survival (OS) was 33.4 weeks (95% CI, 27.8-39.1). Response predicted for superior PFS (25.0 weeks vs. 11.0 weeks, p &lt; .001) and OS (45.9 weeks vs. 26.7 weeks, p &lt; .001). A total of 93 patients discontinued Bev treatment and the reasons for discontinuation were: disease progression in 59 (63.4%), toxicities in 4 (4.3%), physician’s decision in 5 (5.4%), patient’s refusal to further treatment in 25 (26.9%). The median OS was 26.7 weeks in 59 patients who discontinued Bev due to disease progression, and 45.7 weeks in 34 patients who discontinued Bev for reasons other than disease progression (p &lt; .001). Among 85 patients who progressed after Bev, 42 (49.4%) received further therapy: chemotherapy in 32 (37.6%), radiotherapy in 9 (10.6%), and surgery in 1 (1.2%). Further chemotherapy regimens included temozolomide (31.2%), ACNU/CDDP (25.0%), Bev reintroduction (18.8%), erlotinib (12.5%), PCV (9.4%), and intrathecal methotrexate (3.1%). The median survival time after Bev failure was 15.6 weeks (95% CI, 13.3-17.8). Patients who received further therapy showed longer median OS (18.6 weeks vs. 12.9 weeks, p &lt; .001). In patients who received chemotherapy, the median PFS and OS was 6.6 weeks and 20.6 weeks, respectively. Conclusions: Prognosis after Bev failure was poor. Proper selection of patients who may benefit from further treatment is warranted.

Association of the development of bone metastases with the development of brain metastases in patients with non-small cell lung cancer.

e13030 Background: In patients (pts) with non-small cell lung cancer (NSCLC), it is unknown whether pts with bone metastases are predisposed to the development of brain metastases. The impact of bevacizumab (Bev) on the development of bone and brain metastases is not fully characterized. Methods: Retrospective review of pts with stage IV NSCLC without brain metastasis at diagnosis was undertaken. The primary endpoint was to determine whether the development of bone metastases is predictive of the development of brain metastases. Secondary endpoints involved the proportion of pts who developed brain and/or bone metastases while being treated with Bev. Data were analyzed using competing risks methods. Results: A total of 175 pts (52% males, median (range) age: 60 y (35-80)) were studied. Of whom 79% received Bev and 21% did not receive Bev as part of their treatment. Overall 34% of pts had bone metastases at the start of therapy. Pts with pre-existing bone metastases tended to develop new bone disease more frequently than pts who did not initially have bone metastases (39% vs. 19%, p =. 01). The incidence of the development of brain metastases among pts who had pre-existing bone metastases was 17% vs.16 % in pts without pre-existing bone metastases. However, the incidence of brain metastases among pts with pre-existing bone disease who developed new bone metastases was 33% vs. 6% in pts who did not develop new bone disease but had pre-existing bone metastases. Adjusting for initial bone disease, development of new bone metastases was associated with a shorter brain metastases-free interval HR 5.06 (2.03-12.65; p = 0.005). In a subgroup analysis based on Bev exposure, the likelihood of developing brain and bone metastases within 2 years of starting Bev was estimated to be 25% and 27%, respectively, while the likelihood of developing brain and bone metastases without Bev treatment was 33% and 43 %, respectively. Conclusions: In pts with NSCLC, the development of new bone metastases may be indicative of the subsequent development of brain metastases. Additionally, Bev therapy may have an effect on the development of both bone and brain metastases. A prospective investigation may be warranted.

Tumor perfusion during bevacizumab and irinotecan in recurrent glioblastoma: A multimodal approach.

2074 Background: Angiogenesis is a requirement for progression of glioblastoma (GBM) and vascular endothelial growth factor (VEGF) is a mediator of neo-angiogenesis in this tumor. Bevacizumab (Bev), an antibody directed to VEGF, was recently used to treat GBM. However in vivo modifications induced by treatment are still not clearly understood. Aim of this study is to analyze tumor changes induced by Irinotecan (Ir) and Bev, using two different methodologies: relative CBV variation (rCBV) and Difference Perfusion Maps (DPMs). Methods: 42 recurrent GBM patients underwent Bev (10 mg/kg) and Ir (125 or 340 mg/m2) treatment every 2 weeks and were followed up with a radiological protocol, including Dynamic Susceptibility Contrast MRI every 8 weeks. Radiological responses were assessed based on RANO criteria (Wen et al, 2012). Two methods were used to assess perfusion changes. In method A, relative CBV variation after 8 weeks of treatment was calculated through semi-automatic ROI placement in the same anatomic region as in baseline. In method B, relative CBV variations with respect to baseline values were calculated into the evolving tumour region by means of a voxel-by-voxel difference. DPMs were created showing where rCBV significantly increased, decreased or remained unchanged. Results: After a median follow-up of 33.5 months median overall survival (OS) was 35.0 weeks and median progression free survival (PFS) was 20.0 weeks. Method A showed a significant decrease of rCBV for patients with stable disease or partial response after 8 weeks of treatment (p = 0.01) while progressing patients maintained elevated levels of rCBV (p = 0.38). Method B, based on DPMs, showed that patients presenting rCBV increase higher than 25° percentile (corresponding to rCBV increase higher than 18% of tumor volume) had a significantly longer PFS (p = 0.045) and OS (p = 0.016). Conclusions: Using DPM we observed that early increase in global perfusion is related to better survival. This may suggest that Bev, when effective, reduces blood-brain barrier permeability with a higher contrast retention in vessels. If confirmed by further studies this measure could be useful in identifying responders to Bev.

Bevacizumab (Bev) for treatment of recurrent serous borderline (SB) or low-grade serous (LGS) ovarian cancer: A retrospective review of the Memorial Sloan-Kettering Cancer Center (MSKCC) experience.

5545 Background: LGS ovarian cancer is a rare subtype of ovarian cancer, accounting for 10% of ovarian cancer cases. Patients typically present at an early age, exhibit a protracted clinical course, and have response rates to chemotherapy of &lt; 4%. Limited clinical data suggests that bev may have activity in this disease. The objective of this study was to determine the response rate to treatment with bev in patients with SB or LGS ovarian cancer treated at MSKCC. Methods: Following IRB approval, all patients with a diagnosis of SB or LGS ovarian or primary peritoneal cancer treated with ≥ 1 dose of bev for persistent or recurrent disease were identified. 17 patients were treated at MSKCC between July 2005 and June 2012. Diagnosis was confirmed by a gynecologic pathologist. All imaging was independently reviewed by the study radiologist and response was determined by RECIST 1.1 criteria. Results: 17 patients, 10 with LGS ovarian cancer, 3 with LGS primary peritoneal cancer, and 4 with SB disease were included in the analysis. The mean number of prior therapies was 3.4 (range: 1-9, median: 2). Two patients were treated with bev alone, the remainder (15) received bev in combination with paclitaxel (Pac, 6), topotecan (1), pemetrexed (1), oral cyclophosphamide (3), gemcitabine (Gem, 2), Gem and carboplatin (Carbo) (1), or Pac and Carbo (1) .Two patients were not evaluable for response due to termination of treatment prior to first radiographic assessment. The median duration of bev treatment was 23 weeks (mean: 32.2; range 6-79.4). Conclusions: This data suggests that the addition of bev to cytotoxic chemotherapy may produce dramatically higher response rates than chemotherapy alone in patients with SB and LGS ovarian cancer. A prospective study of bev for treatment of this chemotherapy resistant disease is warranted. [Table: see text]

Anti-vascular endothelial growth factor antibody attenuates inflammation and decreases mortality in an experimental model of severe sepsis

IntroductionSevere sepsis is associated with an unacceptably high rate of mortality. Recent studies revealed elevated levels of vascular endothelial growth factor (VEGF), a potent angiogenic and vascular permeability factor, in patients with sepsis. There was also an association between VEGF levels and sepsis severity. Here we investigate the effects of an anti-VEGF antibody (Bevacizumab, Bev) in an experimental model of sepsis.MethodsHuman umbilical vein endothelial cells (HUVECs), murine cecal ligation and puncture (CLP), and endotoxemia models of sepsis were used. HUVECs were treated with lipopolysaccharide (LPS) and/or Bev, harvested and cytokine mRNA levels determined using a semi-quantitative reverse transcription-polymerase chain reaction assay. The levels of inflammatory cytokine were also determined in HUVECs supernatants. In addition, the effects of Bev on mortality in the CLP and endotoxemia models of sepsis were evaluated.ResultsTreatment with Bev and LPS significantly decreased the expression and the level of inflammatory cytokines in HUVECs relative to LPS alone. In CLP and endotoxemia models, survival benefits were evident in mice given 0.1 mg/kg of Bev relative to the CLP or LPS alone (P <0.001 and P = 0.028, respectively), and in 6 h post-treated mice relative to the CLP alone for the effect of different time of Bev (P = 0.033). In addition, Bev treatment inhibited LPS-induced vascular leak in the lung, spleen and kidney in the murine endotoxemia model (P <0.05).ConclusionsAnti-VEGF antibody may be a promising therapeutic agent due to its beneficial effects on the survival of sepsis by decreasing inflammatory responses and endothelial permeability.

The Addition of Bevacizumab to Standard Therapy with R-CHOP in Patients with Previously Untreated Diffuse Large B-Cell Lymphoma Is Associated with an Increased Rate of Cardiac Adverse Events: Final Analysis of Safety and Efficacy Outcomes From the Placebo-Controlled Phase 3 MAIN Study

Abstract 58 Vascular endothelial growth factor (VEGF) is involved in lymphoma growth, suggesting a potential therapeutic role for anti-VEGF therapies. This randomized, placebo (PBO)-controlled, double-blind, multicenter, phase 3 trial (MAIN study) evaluated if adding the anti-VEGF antibody bevacizumab (Bev) to standard R-CHOP therapy improves outcomes for patients (pts) with previously untreated CD20+ diffuse large B-cell lymphoma (DLBCL). The study was designed to detect a 27% reduction in the risk of a progression-free survival (PFS) event (Hazard ratio [HR] = 0.73, R-CHOP + Bev vs R-CHOP). Pts with DLBCL (IPI score ≥2 or bulky disease, ≥7.5 cm irrespective of IPI) and normal cardiac function were randomized (1:1) to receive Bev or PBO + site pre-specified immunochemotherapy (8 cycles R-CHOP-21 or 6 cycles R-CHOP-14 followed by 2 cycles R). R-CHOP-21 & R-CHOP-14 cohorts received Bev at 15 mg/kg q21d & 10 mg/kg q14d, respectively. Response was assessed at the end of induction; pts achieving complete remission continued Bev monotherapy (15 mg/kg q21d, up to 1y from study start, Bev arm) or observation (PBO arm). To observe the required 326 events the planned sample size was 1060 pts followed for 60 mo. The protocol specified strict cardiac monitoring; left ventricular ejection fraction (LVEF) was measured at baseline (bl), after 4 & 8 cycles of therapy, and at 1y. LVEF adverse events (AEs) were defined as a ≥20% absolute decline from bl and/or a decline of ≥10% from bl to As part of ongoing serial safety monitoring, in May 2010 the independent Data and Safety Monitoring Board (DSMB) observed an increased number of LVEF AEs in the Bev arm. Given a low probability of the study meeting the primary efficacy endpoint, the DSMB concluded an unfavorable benefit–risk profile for pts treated with R-CHOP + Bev. Therefore, Bev treatment was immediately discontinued and further enrollment stopped. The study was revised and completed as a safety-oriented follow-up of the enrolled pts. The final analysis included 787 pts (enrolled 7/07 to 6/10) who were followed until 12 mo after the last pt had received the last dose of chemotherapy. Study arms were balanced for age, gender, planned chest irradiation, history of hypertension, cardiac risk factors, and bl LVEF. Efficacy of R-CHOP was not significantly affected by the addition of Bev; median PFS was 42.9 (PBO) vs 40.2 mo (Bev) (HR = 1.09, P =.49). While survival data are immature, no difference was evident (deaths 83/397 [20.9%] PBO vs 82/390 [21%] Bev). Pts in the Bev arm vs the PBO arm had higher rates of LVEF (18.0 vs 8.0%; OR 2.51, 95% CI [1.60;3.93], respectively) and CHF AEs (16.2 vs 6.5%; OR 2.79; 95% CI [1.72;4.54], respectively). Further analysis of CHF AEs by treatment and age group indicated that pts ≥65y had a higher rate than younger pts; both in the Bev arm (20.0 vs 14.0%) and PBO (9.3% vs 4.7%). Incidence of CHF AEs varied according to backbone regimen and in the Bev arm was 16.8% (R-CHOP-21) and 13.8% (R-CHOP-14), and in the PBO arm was 4.6% (R-CHOP-21) and 13.8% (R-CHOP-14). As this was an unplanned analysis on a non-randomized pt population an imbalance of relevant bl features was observed (e.g. age and race composition). CHF events began to increase (determined by Kaplan–Meier analysis) at a cumulative doxorubicin dose of 200 mg/m2 and above in both treatment arms, and increased further at a cumulative doxorubicin dose ≥300 mg/m2 in the Bev arm vs. the PBO arm. Pts in the PBO arm were more likely to resolve their CHF events without sequelae (by investigators reporting, 77.8% in PBO arm vs 63.2% in Bev arm), and also had fewer unresolved CHF events (14.8% vs 20.6%). Fatal CHF AEs were comparable in both arms (1/386 vs 2/395). In conclusion, this study demonstrates that the addition of Bev to R-CHOP is associated with an increased risk of LVEF/CHF events in pts with DLBCL. Following the DSMB recommendation, recruitment was stopped and Bev was discontinued. Data suggest that, regardless of randomization CHF risk is increased for pts ≥65 y of age, and the addition of Bev increases the cardiac toxicity rate of doxorubicin. The current data do not support further evaluation of Bev with R-CHOP in DLBCL. Disclosures: Seymour:Roche/Genetech: Honoraria. Off Label Use: The addition of bevacizumab to standard therapy with R-CHOP in patients with previously untreated diffuse large B-cell lymphoma. Pfreundshuh:Roche: Consultancy, Research Funding; Celgene: Consultancy; Pfizer: Consultancy; Mundi Pharma: Consultancy; Onyx: Consultancy; Amgen: Research Funding. Coiffier:Roche: Consultancy. Trneny:Roche: Consultancy, Honoraria, Research Funding. Sehn:Roche: Consultancy, Honoraria, Membership on an entity9s Board of Directors or advisory committees, Research Funding. Csinady:Roche: Employment.

DNA mismatch repair deficiency and benefit from adjuvant bevacizumab in NSABP C-08: Molecular profiling results.

55 Background: The purpose of this study was to identify biomarkers that define a subset of patients who received benefit from bevacizumab (bev) in NSABP trial C-08, even though bev did not improve outcomes over standard adjuvant chemotherapy (CT) in the treatment of stage II and III colon cancer. Methods: A randomly selected cohort of C-08 cases (N=500) were profiled for whole genome expression (N=445) and for mutations (N=463) in KRAS, NRAS, PIK3CA, and MET. BRAF mutations and mismatch repair (MMR) status were profiled on the available cases (N=1,764 and 1,993, respectively). Cox proportional hazard models were used to assess prognosis and prediction for the value of bev using overall survival (0S) and time to recurrence (TTR) as end points. Results: The effect of bev was different for MMR deficient (MMR-d) and proficient tumors for OS (interaction p=.035) but not TTR (interaction p=.08). Patients with MMR-d (N=252) tumors showed a significant benefit from the addition of bev to CT for OS (hazard ratio =0.52 (95% CI: 0.29-0.94, p=0.028). KRAS, NRAS, PIK3CA, and MET were not significant for interaction with bev in the discovery cohort. BRAF mutations were associated with MMR status (p&lt;.0001) and the prognostic value of MMR depended on BRAF for TTR (interaction p=.027) but not OS (interaction p=.31). The effect of bev was independent of BRAF (interaction p=.28 TTR and .37 OS). Three-factor interaction tests for bev, MMR, and BRAF were not significant for either endpoint. Gene expression analysis with BRB array tools identified 5 BioCarta pathways (p&lt;0.05) which differentially expressed in 4 statistical tests; 4 of these pathways were directly or indirectly involved in T cell activation and one was involved in the activation of VEGF. Conclusions: Patients in C-08 with MMR-d tumors received benefit from bev treatment but these results need to be validated in a separate study. Gene expression data suggest that T-cells may be differentially expressed based on MMR status. Activation of VEGF has been shown to suppress T-cell development (Ohm et al. Blood. 2003:10;4878). A speculative possibility for the benefit of bev in MMR-d tumors may be due to blocking of VEGF, releasing T cells from VEGF suppression.

PLS-5 - Results of ML18147: A Phase III Study of Bevacizumab in Second-Line MCRC for Patients Pretreated with Bevacizumab in First Line

ABSTRACT Background BEV in combination with fluoropyrimidine-based CT is a standard treatment of mCRC in the first-line (1L) and BEV-naive second-line (2L) settings. This is the first randomized study evaluating the benefit of continuing BEV in combination with standard CT as 2L treatment of patients with mCRC who progressed after receiving a standard BEV-containing regimen in the 1L setting. Methods Patients with unresectable, histologically confirmed mCRC who progressed within 3 months after discontinuation of 1L BEV + CT were randomized to 2L fluoropyrimidine-based CT ± BEV (2.5 mg/kg/week equivalent). Choice of oxaliplatin- or irinotecan-based 2L CT was dependent on the regimen used in 1L (crossover) and included as a stratification variable. The primary end point was overall survival (OS); secondary end points included progression-free survival (PFS), response rate and safety. Results Eight hundred and twenty patients were randomized from February 2006 to June 2010. Baseline patient and disease characteristics were well balanced between the arms. The study met its primary end point of OS. Patients who received BEV + CT as 1L for their mCRC and then continued on BEV with a different CT after their cancer progressed lived significantly longer than patients who received only CT in the 2L mCRC. The adverse event profile was consistent with previously reported data for BEV + CT. Compared with the historical data from BEV treatment in 1L or 2L mCRC, BEV-related adverse events were not increased when continuing BEV beyond progression. Conclusions This is the first randomized study to prospectively investigate the impact of continuing BEV treatment in 2L mCRC for patients who progressed after receiving a BEV-containing regimen in 1L. Our findings demonstrate that BEV + CT (crossed over from 1L regimen) continued beyond progression significantly prolongs OS in 2L mCRC. Additional analysis (including biomarker evaluation) is ongoing. Roche (SIX: RO, ROG; OTCQX: RHHBY) announced today that a phase III study (ML 18147) in metastatic colorectal cancer met its primary end point of overall survival. People who received Avastin® (bevacizumab) plus standard chemotherapy as initial treatment (so-called ‘first-line’ treatment) for their metastatic colorectal cancer and then continued on Avastin with a different chemotherapy after their cancer progressed (so-called ‘second-line’ treatment) lived significantly longer than people who received only chemotherapy in the second-line setting. No new safety findings were observed and adverse events were consistent with those seen in previous trials of Avastin. Full data from the ML 18147 study will be submitted for presentation at an upcoming medical meeting.

Impact of duration of bevacizumab (Bev) treatment in the prognosis of adults with recurrent malignant gliomas.

2064 Background: Bev is the standard treatment for patients with recurrent glioblastoma (GB) but is also used in treating recurrent anaplastic gliomas (AG). Differences in outcome between these groups and optimal duration of treatment with Bev in pts with recurrent malignant gliomas are not well defined. We examined the relationship between the duration of Bev treatment and the outcome in pts with GB and AG. Methods: In this retrospective chart and data review derived from our longitudinal database, we identified pts with recurrent AG and GB who were treated with Bev alone or Bev-containing regimens between 2005 and 2009; the data was analyzed to determine the overall survival (OS) and the progression free survival (PFS). Results: A total of 261 patients with recurrent malignant gliomas (196 with GB and 65 with AG) were identified. There was no significant difference between the median length of treatment between AG and GB (5.81±0.66 months vs. 6.77±0.52 months, p=0.32). PFS6 was 34.2% (95% CI, 27.8-41.3) for patients with GB and 44.2% (95% CI, 32.5-56.7) for patients with AG. Patients with GB who were treated ≥6 months had a significantly higher OS (29.13 months vs. 20.16 months, p= 0.001) compared to those treated &lt;6 months, and a significantly higher PFS compared to those treated &lt;6months (11.33 months vs. 3.7 months, p=0.0001). For patients with AG, although treatment ≥6 months had a significantly higher PFS (13.93 months vs. 3.53 months, p&lt;0.0001), OS was not significantly different (months 38.6 vs. 52.5 months, p=0.6) compared with those treated &lt;6 months. Conclusions: Length of treatment ≥6 mo with Bev or Bev-containing regimen was associated with improved PFS in both AG and GB but only the GB subgroup showed improved OS. These results suggest equivocal survival benefit in patients with AG with longer duration of bevacizumab treatment, which requires further study in prospective trials.

Abstract LB-308: Combination of CTx-0294945 a highly selective inhibitor of focal adhesion kinase with bevacizumab in pre-clinical models of breast cancer

Abstract Background: Focal adhesion kinase (FAK) is a non-receptor tyrosine kinase that provides a critical hub for signalling from at least four different classes of cellular sensing mechanisms including growth factor receptors, GPCRs, integrins and mechanical stress forces. By temporal and spatial integration of signals from these sources, FAK plays a central role in cell migration, invasion and proliferation; processes vital for tumorigenesis. The significance of FAK to the function of signal transduction pathways provides a strong rationale for the combination of FAK inhibitors with other targeted agents to achieve improved efficacy against a range of cancers. Others have demonstrated the importance of FAK in angiogenesis and therefore combining a FAKi with anti-VEGF agents is attractive as it employs two complementary mechanisms of suppressing the formation of tumor vasculature. Here we present results from the co-administration of CTx-0294945, a highly selective FAKi, and bevacizumab (bev) in an orthotopic model of human breast cancer. Methods and Results: CTx-0294945 is an orally bioavailable small molecule ATP competitive inhibitor of focal adhesion kinase (FAK KD=0.21 nM). It exhibits high selectivity against a diverse panel of 125 kinases including the closely related Pyk2. CTx-0294945 inhibits autophosphorylation of 397Y-FAK in MDA-MB-231 cells with an IC50 = 7 nM and exhibits low general cellular toxicity (IC50 = 2.7 µM, MDA-MB-231 cells). CTx-0294945 is suitable for oral administration (%F=58 and t1/2=5.1 h at 20 mg/Kg in rat) and does not inhibit (IC50 &amp;gt;20 µM) any of the cytochrome p450 isoforms tested to date. To assess the co-administration of CTx-0294945 with bev, mice were injected orthotopically with MDA-MB-231 cells (106). After 14 days, when tumors were palpable, mice were randomized into 4 groups and dosing commenced. The groups were treated with CTx-0294945 (80 mg/kg QD, PO), bev (12.5 mg/Kg IP, x2/week), CTx-0294945 (80 mg/Kg QD, PO) and bev (12.5 mg/Kg IP, x2/week) or vehicle. Tumor growth was monitored and on day 28 animals in the vehicle and CTx-0294945 arms were culled when the size of the tumors reached ethical end point (1000 mm3). Tumor growth in the bev only and the CTx-0294945 + bev arms was significantly inhibited (75% and 88% TGI respectively). At this time the treatment regimes for both cohorts were stopped and tumor growth allowed to progress. After an additional 14 days the experiment was terminated when the bev treatment group reached ethical end point; however the average size of the tumors in the CTx-0294945 + bev cohort was still was only 562 mm3. Conclusions: Our data suggest the potential utility of combining a selective FAK inhibitor with bevacizumab to prevent tumour progression and enhance the durability of response. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-308. doi:1538-7445.AM2012-LB-308