Search

4532 Background: In CALGB 90206, a phase III trial of IFN +/- Bev, the addition of Bev was associated with an improved progression-free survival (PFS) in RCC pts. Previously, we assessed a panel of...

448 Background: Previously, we identified potential predictive biomarkers of C sensitivity related to EGFR signaling from archived tumor tissue from CALGB 80203. Due to the fact that blood-based markers are more convenient and can be monitored over the course of treatment, baseline plasma samples were also collected and five (EGF, HB-EGF, sEGFR, sHER2, sHER3) of the 14 markers previously analyzed in archived tumor were evaluated in plasma. Methods: CALGB 80203 was a randomized (1:1) phase II trial of 238 pts with locally advanced or metastatic CRC comparing FOLFOX or FOLFIRI (chemo) vs. chemo combined with C. Baseline EDTA plasma samples from 154 pts were analyzed for the 5 candidate markers. The levels of each analyte were correlated with the primary endpoint of OS using univariate Cox proportional hazards models. Potential predictive markers were identified using a treatment by marker interaction term in the Cox model and the markers with significant p-values are reported. Hazard ratios between treatment groups are reported for low or high marker levels dichotomized at the median. Results: Univariate analyses indicated that plasma levels of EGF and sHER3 were negative prognostic markers (p

Acetyl-CoA:1-O-alkyl-2-lyso-sn-glycero-3-phosphocholine acetyltransferase, along with phospholipase A2, is a key regulator of platelet-activating factor biosynthesis via the remodeling pathway. We have now obtained evidence in human neutrophils indicating that this enzyme is regulated by a specific member of the mitogen-activated protein kinases, namely the p38 kinase. We earlier demonstrated that tumor necrosis factor-alpha (TNF-alpha) as well as N-formyl-methionyl-leucyl-phenylalanine treatment leads to increased phosphorylation and activation of p38 kinase in human neutrophils. Strikingly, in the present study these stimuli increased the catalytic activity of acetyltransferase up to 3-fold, whereas 4-phorbol 12-myristate 13-acetate, which activates the extracellular-regulated kinases (ERKs) but not p38 kinase, had no effect. Furthermore, a selective inhibitor of p38 kinase, SB 203580, was able to abolish the TNF-alpha- and N-formyl-methionyl-leucyl-phenylalanine-induced activation of acetyltransferase. The same effect was not observed in the presence of an inhibitor that blocked ERK activation (PD 98059). Complementing the findings in intact cells, we have shown that recombinant, activated p38 kinase added to microsomes in the presence of Mg2+ and ATP increased acetyltransferase activity to the same degree as in microsomes obtained from TNF-alpha-stimulated cells. No activation of acetyltransferase occurred upon treatment of microsomes with either recombinant, activated ERK-1 or ERK-2. Finally, the increases in acetyltransferase activity induced by TNF-alpha could be ablated by treating the microsomes with alkaline phosphatase. Thus acetyltransferase appears to be a downstream target for p38 kinase but not ERKs. These data from whole cells as well as cell-free systems fit a model wherein stimulus-induced acetyltransferase activation is mediated by a phosphorylation event catalyzed directly by p38 kinase.

We have shown previously that both 1,2-diacylglycerol (AAG) and 1-O-alkyl-2-acylglycerol (EAG) prime neutrophil release of arachidonic acid via uncharacterized phospholipases A2. Therefore, we investigated the actions of EAG and AAG specifically on neutrophil cytosolic (cPLA2) and secretory (sPLA2) phospholipase A2s. We hypothesized that AAG as a protein kinase activator would activate cPLA2 via phosphorylation events. EAG is antagonistic to the AAG activation of PKC, thus it was not expected to act via phosphorylation of cPLA2. Neutrophils were primed with either AAG or EAG and then stimulated with fMLP. When neutrophils were primed with 5-20 microM 1,2-diacylglycerol, a shift was observed in cPLA2 migration on SDS-PAGE gels, consistent with phosphorylation of the protein. This gel shift was not seen after exposure to EAG. AAG also caused a parallel increase in enzymatic activity of cPLA2 that was not seen with EAG. We also investigated whether either diglyceride would cause similar priming or direct secretion of sPLA2. Both AAG and EAG directly caused significant secretion of neutrophil sPLA2. EAG also increased the release of sPLA2 in cells subsequently stimulated with fMLP. Thus, AAG activated cPLA2 and stimulated secretion of sPLA2. In contrast, EAG did not activate cPLA2, but directly activated secretion of sPLA2. We also demonstrated that human synovial fluid sPLA2 increased AA release from resting and fMLP-stimulated neutrophils. Given that diglycerides prime for release of AA, PAF, and LTB4, these current data support the hypothesis that such priming may be mediated by phosphorylation dependent (cPLA2) or phosphorylation independent (e.g. secretion of sPLA2) events.

In human neutrophils, the choline-containing phosphoglycerides contain almost equal amounts of alkylacyl- and diacyl-linked subclasses. In contrast to phosphatidylinositol hydrolysis which yields diacylglycerol, hydrolysis of choline-containing phosphoglycerides by phospholipase D coupled with phosphohydrolase yields both alkylacyl- and diacylglycerol. While diacylglycerol activates protein kinase C, alkylacylglycerol does not, and its role is unclear. Yet previous studies have shown that exogenous alkylacyl- and diacylglycerols can prime for the release of radiolabeled arachidonic acid (AA) in intact neutrophils stimulated by formyl-methionyl-leucyl-phenylalanine. We have now examined the effects of both diacylglycerol (1-oleoyl-2-acetylglycerol; OAG) and alkylacylglycerol (1-O-hexadecyl-2-acetylglycerol; EAG) on the activation of mitogen-activated protein (MAP) kinase and the 85-kDa cytosolic phospholipase A2 (cPLA2) in human neutrophils. We observed that while OAG could effectively activate p42 and p44 MAP kinases along with cPLA2 in a time- and concentration-dependent manner, EAG could not. A novel p40 MAP kinase isoform is also present and activated in response to OAG treatment; the behavior of this MAP kinase isoform is discussed. The activation of cPLA2 and MAP kinase by 20 microM OAG could be inhibited by pretreatment with 1 microM GF-109203X, a selective inhibitor of protein kinase C. Although only OAG activated cPLA2, both OAG and EAG primed for the release of AA mass as determined by gas chromatography/mass spectrometry. The priming of AA release by OAG may be explained by the phosphorylation of cPLA2 through the activation of protein kinase C linked to MAP kinase. However, priming by EAG appears to involve a separate mechanism that is dependent on a different PLA2. Our results support a role for phospholipase D-derived products modulating the activation of cPLA2, further supporting the idea of cross-talk among various phospholipases.

11036 Background: Src inhibition may augment sensitivity to chemotherapy, but in unselected patients (pts) with advanced solid tumors, src inhibitors have shown limited clinical activity. Biomarkers to predict benefit from src inhibitors in advanced solid tumors are not yet known. Methods: 22 pts (dose escalation cohort= 12 pts; colorectal cancer [CRC] expansion cohort= 10 pts) were enrolled in a phase I study to determine the safety and tolerability of the src inhibitor dasatinib with capecitabine, oxaliplatin, and bevacizumab (J Clin Oncol 29: 2011 [suppl; abstr 3586]). Src activation (src-a) was assessed in tumors from 16 evaluable pts. Src-a was measured by immunohistochemistry (IHC) in formalin-fixed, paraffin-embedded tumor samples using an antibody that selectively recognizes the active conformation of src (clone 28). A GI pathologist who was blinded to pt outcomes graded membranous src-a using a standard semi-quantitative method. The endpoint of this exploratory analysis was objective response rate ([ORR]= PR+CR). 2-sided Fisher’s Exact test was used to evaluate the association between ORR and src-a. Results: Across all tumor types, 8 tumors had no/faint src-a (IHC=0/1); 8 tumors had moderate/strong src-a (IHC≥2). Benign colonic epithelium had no src-a (IHC=0). The ORR was 75% (6/8) for pts with moderate/strong src-a versus (vs) 0% (0/8) for pts with no/faint src-a (p =0.007). In the CRC expansion cohort, the ORR was 83% (5/6) for patients with moderate/strong src-a vs 0% (0/2) for pts with no/faint src-a (p=0.107); progression free survival range was 7.9-24.4 months for pts with moderate/strong src-a. Conclusions: In this small phase I study, src-a is associated with benefit from the combination of dasatinib and oxaliplatin-based chemotherapy. Further evaluation of dasatinib in patients whose tumors demonstrate high levels of src-a may be warranted. Clinical trial information: NCT00920868.

11011 Background: Beyond KRAS status, there are no validated markers for anti-EGFR therapy in mCRC. While expression of genes within the EGF signaling axis has been reported to correlate with benefit, most reports used data from non-randomized trials which cannot distinguish between general prognostic markers and markers that predict for benefit from C. CALGB 80203 was a 238-patient (pt), 4-arm, randomized phase II study of FOLFOX or FOLFIRI +/- C in mCRC pts. Formalin-fixed, paraffin-embedded (FFPE) tumor samples from CALGB 80203 were analyzed for gene expression of HER family ligands, receptors and regulators. Methods: FFPE tumor samples from consenting pts were analyzed for AREG, BTC, CD73, DUSP4, EGF, EGFR, EPGN, EREG, HB-EGF, HER2, HER3, HER4, PHLDA1 and TGFa by RT-qPCR. Interaction between baseline gene expression levels and treatment (C) with respect to progression-free survival (PFS) and overall survival (OS) was modeled using a multiplicative Cox proportional hazards model. Results: Baseline tumor tissue was available from 103 pts; 55% of tumors were KRAS wild type (WT). Two prognostic markers were identified; higher tumor mRNA levels of HER2 (HR=0.64, p=0.002) and EREG (HR=0.89, p=0.016) were associated with longer PFS across all pts. Potential predictive markers of benefit for C were identified. In KRAS WT tumors, lower HER3 expression was associated with longer OS from C treatment while higher HER3 expression was associated with shorter OS from C treatment (chemo + C: HR=1.15; chemo only: HR=0.48, interaction p=0.029). No association was observed for HER3 and outcome in KRAS mutant (MT) tumors. Interestingly, higher CD73 expression was associated with longer PFS in C-treated pts in both KRASWT (chemo + C: HR=0.91; chemo only: HR=1.57, interaction p=0.026) and MT tumors (chemo + C: HR=0.80; chemo only: HR=1.29, p=0.025). Conclusions: In one of the first reports using data from a randomized study, gene expression of HER3 and CD73 were identified as potential predictive markers for C. These data implicate not only HER axis signaling but also immune modulation as potential mechanisms of C action and sensitivity, and warrant confirmation in other large randomized trials.

4520 Background: CALGB 90206 was a phase III trial of 732 pts with RCC comparing B+I versus I alone demonstrating no difference in OS. To date, there are no validated predictive biomarkers for B in RCC. For this reason, baseline plasma samples from CALGB 90206 pts were analyzed to identify and test predictive markers for B+I in RCC pts. Methods: Baseline EDTA plasma samples from 424 consenting pts were analyzed using an optimized multiplex ELISA platform for 32 candidate factors related to tumor growth, angiogenesis, and inflammation. The data were randomly split into training (n=286) and validation (n=138) sets. The proportional hazards model was used to test for treatment-marker interactions of OS. The estimated coefficients from the training set were used to compute a risk score (RS) for each pt in the validation set. The RS classified pts by risk in the validation set. The model was assessed for its predictive accuracy using area under the curve (AUC). Results: A statistically significant 3-way interaction between interleukin-6 (IL-6), hepatocyte growth factor (HGF) and treatment was observed in the training set (p

Growth factors and their receptors coordinate neuronal differentiation during development, yet their roles in the pediatric tumor neuroblastoma remain unclear. Here we report that expression of type III TGF-beta receptor (TβRIII) mRNA and protein decreases with advancing stage of neuroblastoma and positively correlates with prognosis. TβRIII expression is epigenetically suppressed by MYCN oncogene amplification and TβRIII expression can be used as a prognostic marker in neuroblastoma patients with MYCN amplification. TβRIII expression in neuroblastoma cells promotes neuronal differentiation and enhances the differentiating effects of FGF2 treatment. Mechanistically, glycosaminoglycan modifications on TβRIII bind FGF2 and FGFR1 to promote neuronal differentiation via Erk MAPK and the transcription factor ID1. TβRIII-mediated differentiation suppresses tumor cell proliferation in vitro and in vivo. These studies characterize a novel co-receptor function for TβRIII in FGF2-mediated neuronal differentiation of neuroblastoma cells, while identifying potential therapeutic targets and clinical biomarkers for advanced-stage disease. More generally, our results suggest that the targeting of growth factor receptors and downstream signaling pathways may prove useful in promoting neuronal differentiation to suppress neuroblastoma tumor growth. Citation Format: Erik H. Knelson, Angela L. Gaviglio, Alok K. Tewari, Michael B. Armstrong, Andrew B. Nixon, Mark D. Starr, Karthikeyan Mythreye, Gerard C. Blobe. The type III TGF-beta receptor promotes FGF2-mediated neuronal differentiation in neuroblastoma. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 5041. doi:10.1158/1538-7445.AM2013-5041

The newly defined eicosatetraenoates (ETEs), 5-oxoETE and 5-oxo-15(OH)-ETE, share structural motifs, synthetic origins, and bioactions with leukotriene B4 (LTB4). All three eicosanoids stimulate Ca2+ transients and chemotaxis in human neutrophils (PMN). However, unlike LTB4, 5-oxoETE and 5-oxo-15(OH)-ETE alone cause little degranulation and no superoxide anion production. However, we show herein that, in PMN pretreated with granulocyte-macrophage or granulocyte colony-stimulating factor (GM-CSF or G-CSF), the oxoETEs become potent activators of the last responses. The oxoETEs also induce translocation of secretory vesicles from the cytosol to the plasmalemma, an effect not requiring cytokine priming. To study the mechanism of PMN activation in response to the eicosanoids, we examined the activation of mitogen-activated protein kinase (MAPK) and cytosolic phospholipase A2 (cPLA2). PMN expressed three proteins (40, 42, and 44 kDa) that reacted with anti-MAPK antibodies. The oxoETEs, LTB4, GM-CSF, and G-CSF all stimulated PMN to activate the MAPKs and cPLA2, as defined by shifts in these proteins' electrophoretic mobility and tyrosine phosphorylation of the MAPKs. However, the speed and duration of the MAPK response varied markedly depending on the stimulus. 5-OxoETE caused a very rapid and transient activation of MAPK. In contrast, the response to the cytokines was rather slow and persistent. PMN pretreated with GM-CSF demonstrated a dramatic increase in the extent of MAPK tyrosine phosphorylation and electrophoretic mobility shift in response to 5-oxoETE. Similarly, 5-oxoETE induced PMN to release some preincorporated [14C]arachidonic acid, while GM-CSF greatly enhanced the extent of this release. Thus, the synergism exhibited by these agents is prominent at the level of MAPK stimulation and phospholipid deacylation. Pertussis toxin, but not Ca2+ depletion, inhibited MAPK responses to 5-oxoETE and LTB4, indicating that responses to both agents are coupled through G proteins but not dependent upon Ca2+ transients. 15-OxoETE and 15(OH)-ETE were inactive while 5-oxo-15(OH)-ETE and 5(OH)-ETE had 3- and 10-fold less potency than 5-oxoETE, indicating a rather strict structural specificity for the 5-keto group. LY 255283, a LTB4 antagonist, blocked the responses to LTB4 but not to 5-oxoETE. Therefore, the oxoETEs do not appear to operate through the LTB4 receptor. In summary, the oxoETEs are potent activators of PMN that share some but not all activities with LTB4. The response to the oxoETEs is greatly enhanced by pretreatment with cytokines, indicating that combinations of these mediators may be very important in the pathogenesis of inflammation.

The initial step in the remodeling pathway of PAF synthesis is catalyzed by phospholipase A2 (PLA2).1 It has generally been accepted that PLA2 acts directly on 1-0-alkyl-2-arachidonoyl-sn-glycero-3-phosphocholine (alkyl-2-AA-GPC) to form lyso-PAF (1-alkyl-2-lyso-GPC) which is then acetylated to yield PAF. More recent studies suggest that PAF synthesis may also be initiated indirectly through the action of CoA-independent transacylase following the hydrolysis of l-0-alk-1′-enyl-2-AA-sn-glycero-3-phosphoethanolamine (alkenyl-2-AA-GPE) by PLA2.2,3 In the putative indirect pathway, an accumulation of alkenyl-2-lyso-GPE triggers the transfer of AA from alkyl-2-AA-GPC to the alkenyl-2-lyso-GPE thus forming lyso-PAF. It seems logical that both the direct and indirect pathways may contribute to PAF synthesis. However, the relative physiological importance of the two pathways has not been established.

The identities and roles of arachidonic acid (AA) metabolites in human diseases are often elusive. For example, allergens cause tissues to secrete agents that attract eosinophils (Eo) and induce these cells to release granule enzymes, Superoxide anion, and bioactive molecules that lead to the organ dysfunctions seen in allergy1. Zileuten, an anti-lipoxygenase drug, relaxes the airways of asthmatics, perhaps by blocking the synthesis of an Eo-targeting eicosanoid2-4. The drug clearly inhibits allergen-induced Eo fluxes to lung as well as bronchospasm in primates and lower mammals5,6. Thus, an eicosatetraenoate (ETE) may link Eo to allergy. Among the eicosanoids occupying human allergen-reactive sites, 15(S)-hydroxy-ETE (15-HETE), prostaglandins, and thromboxanes are not chemotactic for Eo; leukotriene (LT)B4 attracts Eo but is more active on polymorphonuclear neutrophils (PMN); and peptido-LTs, lipoxins, and 5(S)-hydroxy-ETE (5-HETE) act weakly on both cell types1,7-10. These agents are unlikely to mediate Eo-based lesions in vivo. However, a newly defined 5-HETE analog, 5-oxoETE, has in vitro actions suggesting that it could be such a mediator.

e21038 Background: TRC105, an anti-CD105 monoclonal antibody, has completed phase 1 testing and is being studied in multiple phase 2 trials. We previously reported that low dose TRC105 (0.01-1.0 mg/kg) modulated the expression of soluble angiogenic biomarkers in patients [J Clin Oncol 29: 2011 (suppl; abstr 10565)]. In this report, we evaluated angiogenic biomarkers in patients receiving higher doses of TRC105 including the recommended phase 2 dose. Methods: Patients with advanced refractory solid tumors were treated with escalating doses of TRC105 until disease progression. Serial plasma samples were analyzed via an optimized multiplex ELISA platform. 36 biomarkers related to tumor growth, angiogenesis, and inflammation were assayed at baseline (BL), after 1 month (C2D1), concurrent with radiological restaging near the end of cycle 2 (C2D22), and at end of study (EOS). Results: 32 patients treated with TRC105 at doses of 0.3 to 15 mg/kg were evaluated for biomarker expression. Wilcoxon signed rank tests indicated that the following analytes were significantly different at C2D1 when compared with baseline (p

e13589 Background: Tumor growth and angiogenesis are regulated by multiple cellular pathways, including the VEGF, mTOR, and EGFR pathways. We have previously reported our clinical experience combining Bev, Ev and Erl therapies. Taking advantage of the similarities between tumor growth and wound healing, we conducted a PD analysis of these agents using novel dermal wound angiogenesis assays. Methods: Patients with advanced solid tumors received either 1) Bev-Ev doublet therapy; 2) Bev-Ev-Erl triplet therapy; 3) Ev monotherapy for 14 days followed by addition of Bev-Erl; 4) Erl monotherapy for 14 days followed by addition of Bev-Ev. Full-thickness dermal granulation biopsies were harvested; the phospho (p) and total (t) levels of VEGFR2, EGFR, and S6RP were measured at baseline, end of the monotherapy lead-in period (day 14), and 35 days post doublet or triplet therapy. A dermal wound angiogenesis assay, which quantifies peripheral wound vascularization was also incorporated in the study. Results: 38 pts were evaluable for biomarker analysis. Wilcoxon signed rank tests indicated that both VEGFR2 and S6RP were significantly down-regulated after Bev-Ev-Erl triplet therapy: tVEGFR2 (q=5.1E-05), pVEGFR2 (q=0.01), tS6RP (q=0.007), pS6RP (q=9.7E-04). Levels of tVEGFR2 (q=5.5E-04) and pS6RP (q=6.9E-04) were also found to be significantly down-regulated after Bev-Ev doublet therapy. Ev monotherapy for 14 days led to significant tVEGFR2 down-regulation (q=0.035). No significant changes in EGFR levels were observed in any cohort. Additionally, Bev-Ev-Erl triplet therapy led to a slight decrease in dermal wound vascularization (p=0.098), which positively correlated with changes in tVEGFR2 levels (q=0.008). Conclusions: This is the first investigation evaluating the PD effects of VEGF, mTOR, and EGFR inhibition in a clinical wound-healing model. Our results indicate that targeting of all three pathways is associated with overall down-regulation of both VEGF and mTOR pathways. These data suggest that this dermal wound model can be used to quantify PD markers in evaluating multiple classes of targeted therapies.

10508 Background: CALGB 80303 was a phase III trial of 602 patients (pts) with locally advanced or metastatic pancreatic cancer (PC) comparing gemcitabine + bevacizumab (GB) vs. gemcitabine + place...

10565 Background: CD105 (endoglin) is an important mediator of tumor angiogenesis that is upregulated by hypoxia and VEGF inhibitors. TRC105 is an anti-CD105 monoclonal antibody currently being eva...

e21009 Background: Biomarkers that predict benefit and define mechanisms of resistance to VEGF inhibitors are urgently needed. Methods: Plasma samples from patients (pts) treated on a phase II tria...

The synthesis of 1-O-alkyl-2-(R)-hydroxypropane-3-phosphonocholine is described. An efficient alkylation procedure using (NaH/DMSO) catalysis is also described and applied to the synthetic scheme. The key intermediate 1-O-alkyl-2-(R)-O-benzyl-3-bromopropane was phosphonylated using tris(trimethylsilyl)phosphite; the resulting phophonic acid was coupled to choline using trichloroacetonitrile/pyridine or triisopropylbenzenesulfonyl chloride/pyridine followed by catalytic hydrogenation to yield 1-O-alkyl-2(R)-hydroxypropane-3-phosphonocholine.

e14609 Background: Integrins are commonly upregulated in tumor cells and are important regulators of invasion and metastasis. Integrin signaling is initiated upon engagement of ECM and requires Src as well as various other proteins including ILK, FAK, and paxillin. Methods: Presence of αv integrins on CRC cell lines was established by flow cytometry. CNTO 95 (10μg/ml, a monoclonal anti-αv integrin antibody; Ortho Biotech), and dasatinib ( [Table: see text]

3551 Background: In preclinical models, VEGF, mTOR, and EGFR inhibitors have anti-tumor and anti-angiogenesis effects as monotherapies and in combination. B inhibits VEGF; E inhibits mTOR; P inhibits EGFR. There is also potential for interaction between the pathways. Previously BE and BE + erlotinib were evaluated and showed signs of clinical activity. Methods: Patients (pts) with refractory advanced solid tumors were accrued in a phase I dose escalation of B + E + P on a 28d cycle. Dose levels are shown in the table below. DLT was defined as any treatment-related grade 4 heme, grade 3/4 non-heme adverse event (AE), or receiving [Table: see text] [Table: see text]

e15506 Background: Dasatinib (D) is a small molecule tyrosine kinase inhibitor with activity against both bcr-abl and src. Cetuximab (C) is a monoclonal antibody that blocks EGFR. Preclinical models suggest D reverses resistance to G. In addition, src and EGFR pathways interact; synergism of dual blockade by D + C is possible. We evaluated two combination regimens, GD and GDC, in a Phase I dose escalation study. Methods: Patients (pts) with advanced solid tumors were enrolled in cohorts of 3–6 to either GD or GDC. G was dosed in mg/m2 weekly for 3 of 4 weeks, D was dosed in mg PO BID, and C was dosed at 250 mg/m2 weekly after loading dose of C=400; cycle length was 28 days. Dose levels were as follows: 1) G 1000 + D 50 ± C; 2) G 1,000 + D 70 ± C; 3) G 1,000 + D 100 ± C. Standard cycle 1 DLT definitions were used. Eligible pts had advanced solid tumors, adequate organ and marrow function, and no co-morbidities that would increase risk of toxicity. Serum, plasma, and skin biopsy biomarkers were obtained pre- and on treatment. Results: 25 pts have been enrolled, including 21 with pancreatic adenocarcinoma, 3 of whom had received prior G. 21 pts were evaluable for toxicity and 18 for efficacy. Four DLT were observed: Gr 3 ANC with infection (GDC1, n=1), Gr 3 ALT (GD2, n=2), and Gr 5 pneumonitis (GDC2, n=1). Possible treatment-related adverse events in later cycles included: Gr3–4 ANC (n=4), Gr4 colitis (n=1), Gr3 bilirubin (n=2), Gr3 Hgb (n=2), Gr3 Plt (n=2), Gr3 edema/fluid retention syndrome (n=1), and Gr3 vomiting (n=2). One previously untreated pt had a partial response. Eight of 18 pts, 3 of whom had received prior G, had stable disease as best response, median duration = 5 months (range 1–7). Biomarker results are pending. Conclusions: The MTD of the GD arm is G1000/D50BID. Stable disease in previous G-refractory pts was noted. Hematologic toxicities were dose-limiting; later toxicities including hematologic, LFT changes, pneumonitis, and fluid retention were seen. To address these toxicities, once daily dosing of D will be explored, followed by an expanded cohort of G + daily D vs G + bid D in pts with treatment-naïve pancreatic cancer. [Table: see text]

4080 Background: For patients (pts) with mCRC, no standard therapy exists after progression on 5-FU, oxaliplatin, irinotecan, bevacizumab, and/or cetuximab/panitumumab. Preclinical data demonstrate combined VEGF and mTOR inhibition has greater anti-angiogenic and anti-tumor activity than either monotherapy. B inhibits VEGF; E inhibits mTOR. Phase I data in patients demonstrated B + E was safe and activity was seen in several pts with refractory mCRC. Methods: 25 pts with refractory mCRC were enrolled in an expanded cohort of B + E. Doses: B 10 mg/kg q2 wks; E 10 mg PO QD. Blood, skin, and tumor biopsies pre- and on-treatment were collected for markers of response and resistance. Results: At this time, 19 pts (10M: 9F) are evaluable for toxicity; 17 for efficacy. Median age 57 years (range 35–78). Median number prior regimens 3. All pts had prior B exposure; 17 pts had progressive disease on prior B-based therapy. There was one Grade (Gr) 4 adverse event (AE) of hypokalemia. Grade 3 AE related to treatment were bowel perforation/fistula, (n=2), hyperglycemia (3), hypokalemia (3), hypertension (2), fatigue (1), alk phos elevation (1; lab only), hypoalbuminemia (1), and volume depletion (1). Other events of interest were: Gr1–2 mucositis (n=10), Gr1 hyperlipidemia (11). Of 17 pts evaluable for response, 4 pts had SD as best response (median 24 wks, range 17–31+ wks); there were 3 minor responses in pts who had progressed on B (median 16 wks, range 16–27 wks). No CR or PR were seen. Biomarker data is pending. Conclusions: B+E has activity in refractory mCRC in pts who had progressed on a B-based regimen, suggesting B+E may overcome resistance to B. Patient accrual is continuing and updated data will be presented. [Table: see text]

4063 Background: FOLFOX/XELOX + BV are standard 1st line regimens for the treatment of mCRC. Based upon the activity of CET in mCRC, promising data with the XELOX-A regimen, and preclinical and early clinical data suggesting benefit from simultaneous targeting of the VEGF and EGFR pathways, we evaluated the safety and efficacy adding CET to the XELOX-A regimen. Methods: Pts with untreated mCRC received XELOX-AE: OX 130 mg/m2 d1, CAP 850 mg/m2 BID d1–14, BV 7.5 mg/kg d1, and CET 250 mg/m2 d1, 8, and 15 (loading dose CET = 400 mg/m2). Cycles were repeated every 3 weeks. Blood was collected for markers of response and resistance. Results: 29 pts have mature data for toxicity; 24 for efficacy. Median age 56 (range 33–76). Grade 3/4 adverse events which occurred in > 10 % of patients were: diarrhea (7/29 G3) and skin rash (9/29 G3). Other events of interest were: neuropathy (21/29, 2 G3), hypomagnesemia (15/29, 3 G 3) proteinuria (3 G2), HTN (1 G1), hand-foot syndrome (1 G1, 4 G2). 17/29 pts required dose modi...

3548 Background: BV inhibits vascular endothelial growth factor (VEGF). EV is an mTOR (mammalian target of rapamycin) inhibitor. E inhibits epidermal growth factor receptor (EGFR) tyrosine kinase. VEGF, mTOR, and EGFR inhibitors have anti-tumor and anti-angiogenesis effects alone and in combination in preclinical models. As a combination targeted therapy, we evaluated BV + EV + E in a phase I, pharmacokinetic (PK), biomarker study. Methods: Cycle length was 28 days. Doses: BV 10mg/kg IV q14d. EV 5mg PO QD, escalating to 10mg QD. Once the recommended phase II dose (RPTD) of BV + EV was reached, E was added, starting at 75 mg PO QD. DLT was defined as any treatment-related grade 4 heme or grade 3/4 non-heme event in Cycle 1. Results: 34 pts have been enrolled (18 F, 16 M), 28 evaluable for DLT, 24 for efficacy. Median age is 58y (range 29–73). Dose level 1 (BV 10mg/EV 5mg) had no DLT’s. Dose level 2 (BV 10mg/EV 10mg) had no DLT’s and the cohort was expanded to 13 evaluable pts. E (75mg) was added to BV 10mg/ EV 10mg. 2/6 patients had DLT (grade 3 mucositis and grade 3 rash). The doses were adjusted to BV 5mg/EV 5mg/E 75mg. 3 patients had no DLT and this dose is the MTD and RPTD for the 3-drug combination. 20 more patients are being enrolled at the RPTD for biomarker studies. Other grade ¾ toxicity included: nephrotic syndrome, cardiac ischemia, ventricular thrombus, portacath thrombosis, and bowel perforation. 2 patients had PR: 1 renal and 1 osteosarcoma. 16/24 pts had SD (10–112+ weeks). 5/6 patients with colorectal cancer (CRC) previously progressing on BV had SD (16+ - 112 weeks). One CRC patient had 19% radiologic decrease. Conclusions: BV + EV + E preliminary clinical activity (notably in refractory CRC) and class-related side effects were seen. The MTD is BV 5mg/kg IV q14d + EV 5mg PO QD + E 75mg PO QD. Updated data will be presented. No significant financial relationships to disclose.

3541 Background: BV is a potent inhibitor of vascular endothelial growth factor (VEGF) with broad clinical activity both alone and in combination. ABT-510 is a 9 amino acid synthetic analog of the N-terminal region of thrombospondin (TSP-1), an endogenous inhibitor of angiogenesis. ABT-510 exhibits anti-angiogenic effects in preclinical models and has been well-tolerated as a single agent in previous Phase I & 2 studies. As a combination anti-angiogenesis therapy, we evaluated BV + ABT-510 in a phase I biomarker study. Methods: In cohorts of 3–6 patients, BV and ABT-510 were evaluated. BV was dosed in mg/kg IV q2weeks and ABT-510 was given daily in mg SC BID; cycle length was 28 days. Dose levels were as follows: (1) BV 5/ ABT-510 50; (2) BV 10/ ABT-510 50; (3) BV 10/ ABT-510 100. At the recommended phase II dose, 20 patients are being enrolled for detailed biomarker studies. DLT was defined as any hematological toxicity = grade 4 or grade = 3 non- hematological event in Cycle 1 related to treatment, with the exception of grade 3 hypertension adequately controlled by medication. Eligible patients had advanced solid tumors, adequate organ and marrow function, and no co-morbidities suggesting increased risk for class-related toxicities. Dermal wound angiogenesis assays were analyzed for visualization as well as phospho-VEGFR2, TGFβ, phospho-AKT, and thrombospondin-1 and -2, both pre- and on-treatment. Plasma was assayed for multiple angiogenic factors. Results: 22 patients have been enrolled; 17 are currently evaluable for toxicity and 14 for efficacy. No DLT were seen in any cohort. Possible treatment related adverse events occurring in later cycles included: gr3 GI bleed (n=1) and gr3 headache (n=1). 7 patients had stable disease as best response (range 8- 64+ weeks); 2 remain on therapy, one at 64+ weeks and one at 24+ weeks. Conclusions: BV + ABT-510 is well-tolerated and has hints of activity in refractory tumors. The recommended phase II dose is BV 10mg/kg IV q14d and ABT-510 100mg SC BID. Updated clinical and biomarker data will be presented. No significant financial relationships to disclose.

3093 Background: Preclinical data have suggested at least additive antitumor effects when imatinib (I) and 5-FU are combined. The goals of this study were to determine the maximal tolerated dose for the combination of I and capecitabine (C) in solid tumor patients (pts), as well as to describe non-dose limiting toxicities (non-DLT), pharmacokinetics (PK), and pharmacodynamics (PDGFR inhibition in granulation tissue). Methods: 24 pts (12 M, 12 F) with solid tumor malignancies received C plus I using a 21 day (d) cycle. C was dosed BID for 14/21 d. I was dosed QD x 21d. Results: 24 of 24 pts were fully evaluable for toxicity and efficacy. 6 pts were treated at dose level 1 (C 1000 mg/m2 BID, I 300 mg QD). The only dose limiting toxicity (DLT) at dose level 1 was grade (gr) 3 diarrhea in 1 patient; however 5 of 6 pts required dose reductions in a subsequent cycle. Therefore, the next 18 pts were treated at dose level -1 (C 750 mg/m2 BID and I 300 mg QD). One DLT (gr 3 fatigue) was seen at dose level -1. Non-DLT gr ¾ toxicities at any point at dose level -1 were gr 4 anemia (n=1) and gr 3 lymphopenia (n=6). Toxicities requiring dose reductions at any point on dose level -1 occurred in 7/18 patients. Gr ½ toxicities found in > 10% of all patients at any point in the trial were anemia (46%), leukopenia (17%), neutropenia (21%), lymphopenia (29%), thrombocytopenia (29%), nausea (71%), anorexia (41%), edema (42%), diarrhea (38%), hand foot syndrome (33%), vomiting (33%), headache (25%), insomnia (25%), skin rash (25%), constipation (21%), stomatitis (17%), lightheadedness (13%), fatigue (13%), numbness/tingling (13%), taste change (13%), and elevated transaminases (13%). 1 pt with carcinoid tumor had a partial response. 1 pt with non-small cell lung cancer had a minor response, and 5/21 pts had stable disease as their best response. 1 pt had stable disease for 8 months and another pt was stable for 6 months. Conculsions: On a 21 d cycle, the recommended phase 2 dose of C plus I is C 750 mg/m2 BID for 14 d, and I 300 mg QD. [Table: see text]

3097 Background: BV is a potent inhibitor of vascular endothelial growth factor (VEGF) with broad clinical activity. E is an mTOR (mammalian target of rapamycin) inhibitor in development for cancer and solid organ transplant therapy. VEGF and mTOR inhibitors have anti-tumor and anti-angiogenesis effects alone and in combination in preclinical models. As a combination anti-angiogenesis therapy, we evaluated BV + E in a phase I, pharmacokinetic (PK), biomarker study. Methods: BV was dosed at 10mg/kg IV q14d. E was dosed at 5mg PO QD, escalating to 10mg QD. Cycle length was 28 days. DLT was defined as any grade 4 heme or grade 3/4 non-heme event in Cycle 1 related to treatment. Pts had advanced solid tumors, adequate organ function, and no increased risks for class-related toxicities. Serial blood samples were collected for PK studies of E. Dermal wound angiogenesis assays were performed pre and on treatment for phospho VEGFR2, AKT, mTOR, and S6K. Results: 14 pts have been enrolled (8 F, 6 M), 12 evaluable for toxicity, 14 for efficacy. Median age is 58y (range 29–73). At dose level 1 (BV 10mg/E 5mg) there were no DLT’s in 5 pts. At dose level 2 (BV 10mg/E 10mg), no DLT’s were noted in the initial 3 pts and the cohort was expanded to 9 pts. Side effects were primarily grade 1–2: pain (10/14), mucositis (9/14), anorexia (8/14), rash (7/14), bleeding (7/14), hyperlipidemia (6/14), fatigue (6/14), and HTN (4/14). 1 pt had a myocardial infarction at day 72 and one pt developed nephrotic syndrome at day 70. 7/14 pts had stable disease as best response (70–278d). Conclusions: BV + E is generally well-tolerated. Preliminary clinical activity and class-related side effects were noted. The recommended phase II dose is BV 10mg/kg IV q14d and E 10mg PO QD. [Table: see text]

Bone morphogenetic proteins (BMPs) are members of the TGF-β superfamily that are over-expressed in breast cancer, with context dependent effects on breast cancer pathogenesis. The type III TGF-β receptor (TβRIII) mediates BMP signaling. While TβRIII expression is lost during breast cancer progression, the role of TβRIII in regulating BMP signaling in normal mammary epithelium and breast cancer cells has not been examined. Restoring TβRIII expression in a 4T1 murine syngeneic model of breast cancer suppressed Smad1/5/8 phosphorylation and inhibited the expression of the BMP transcriptional targets, Id1 and Smad6, in vivo. Similarly, restoring TβRIII expression in human breast cancer cell lines or treatment with sTβRIII inhibited BMP-induced Smad1/5/8 phosphorylation and BMP-stimulated migration and invasion. In normal mammary epithelial cells, shRNA-mediated silencing of TβRIII, TβRIII over-expression, or treatment with sTβRIII inhibited BMP-mediated phosphorylation of Smad1/5/8 and BMP induced migration. Inhibition of TβRIII shedding through treatment with TAPI-2 or expression of a non-shedding TβRIII mutant rescued TβRIII mediated inhibition of BMP induced Smad1/5/8 phosphorylation and BMP induced migration and/or invasion in both in normal mammary epithelial cells and breast cancer cells. Conversely, expression of a TβRIII mutant, which exhibited increased shedding, significantly reduced BMP-mediated Smad1/5/8 phosphorylation, migration, and invasion. These data demonstrate that TβRIII regulates BMP-mediated signaling and biological effects, primarily through the ligand sequestration effects of sTβRIII in normal and cancerous mammary epithelial cells and suggest that the ratio of membrane bound versus sTβRIII plays an important role in mediating these effects.

SummaryEndoglin, or CD105, is a cell membrane glycoprotein that is overexpressed on proliferating endothelial cells (EC), including those found in malignancies and choroidal neovascularization. Endoglin mediates the transition from quiescent endothelium, characterized by the relatively dominant state of Smad 2/3 phosphorylation, to active angiogenesis by preferentially phosphorylating Smad 1/5/8. The monoclonal antibody TRC105 binds endoglin with high avidity and is currently being tested in phase 1b and phase 2 clinical trials. In this report, we evaluated the effects of TRC105 on primary human umbilical vascular endothelial cells (HUVEC) as a single agent and in combination with bevacizumab. As single agents, both TRC105 and bevacizumab efficiently blocked HUVEC tube formation, and the combination of both agents achieved even greater levels of inhibition. We further assessed the effects of each drug on various aspects of HUVEC function. While bevacizumab was observed to inhibit HUVEC viability in nutrient-limited medium, TRC105 had little effect on HUVEC viability, either alone or in combination with bevacizumab. Additionally, both drugs inhibited HUVEC migration and induced apoptosis. At the molecular level, TRC105 treatment of HUVEC lead to decreased Smad 1/5/8 phosphorylation in response to BMP-9, a primary ligand for endoglin. Together, these results indicate that TRC105 acts as an effective anti-angiogenic agent alone and in combination with bevacizumab.

4042 Background: T is a reversible and highly selective allosteric inhibitor of MEK1/MEK2. The addition of T to G did not improve overall survival (OS) as first-line treatment for pts with metastatic adenocarcinoma of the pancreas (ASCO GI 2013 #291). CAF profiles have shown potential for identification of prognostic and predictive markers in cancer pts (Tran, Lancet, 2012). Methods: Plasma samples (n = 144 baseline [BL]; n = 112 day 15) from pts who consented to participate in MEK113487 study were analyzed for 30 CAFs (ANG2, IGFBP3, IL2R, IP10, MMP2, MMP9, OPN, PDGFBB, SCF, TIMP1, IL6, MIP1B, SDF1, TRAIL, VEGF, MIP3A, MCP2, FGFB, IL8, VEGFR1, IL10, IL1A, IL12P40, PIGF, EGF, IL1B, TNFA, IL4, MIP1A, MCP3) using SearchLight multiplex assays in a CLIA-certified laboratory. Change from BL was assessed using either paired t tests or Wilcoxon tests. BL and change from BL CAF levels were tested for association with clinical outcome using proportional hazards regression within arm (P < .01 for significance) and between arms (treatment arm by CAF level interaction, P < .05 for significance). Results: Lower levels (< median) of BL ANG2, IL6, TIMP1, and IL8 were associated with an average of 5 mo longer OS in both the T+G and G arms. Lower BL levels (less than first quartile) of IGFBP3 and PDGFBB were associated with 6 mo longer OS in the T+G arm, while higher levels (greater than third quartile) of IGFBP3 and PDGFBB were associated with 4.8 mo longer OS in the G arm. A combined model of low IGFBP3 and/or PDGFBB showed improved survival for T+G vs G (median OS, 10.3 vs 5.6 mo). Decreases of ANG2 and IL2R levels at day 15 from BL were observed in the T+G arm only; additional results will be presented. Conclusions: This study suggests that plasma CAF profiling may aid in the prognostic evaluation of pts, assessing therapeutic response, and identifying pathways modulated by treatment in pancreatic cancer pts. Low IGFBP3 and PDGFBB may identify patients who receive greater benefit from T+G compared with G in this population and merit further investigation as predictive markers. Clinical trial information: NCT01231581.