Advertisement
Research Articles

A Human Monoclonal Anti-ANG2 Antibody Leads to Broad Antitumor Activity in Combination with VEGF Inhibitors and Chemotherapy Agents in Preclinical Models

Jeffrey L. Brown, Z. Alexander Cao, Maria Pinzon-Ortiz, Jane Kendrew, Corinne Reimer, Shenghua Wen, Joe Q. Zhou, Mohammad Tabrizi, Steve Emery, Brenda McDermott, Lourdes Pablo, Patricia McCoon, Vahe Bedian and David C. Blakey
DOI: 10.1158/1535-7163.MCT-09-0554 Published January 2010

Abstract

Localized angiopoietin-2 (Ang2) expression has been shown to function as a key regulator of blood vessel remodeling and tumor angiogenesis, making it an attractive candidate for antiangiogenic therapy. A fully human monoclonal antibody (3.19.3) was developed, which may have significant pharmaceutical advantages over synthetic peptide-based approaches in terms of reduced immunogenicity and increased half-life to block Ang2 function. The 3.19.3 antibody potently binds Ang2 with an equilibrium dissociation constant of 86 pmol/L, leading to inhibition of Tie2 receptor phosphorylation in cell-based assays. In preclinical models, 3.19.3 treatment blocked blood vessel formation in Matrigel plug assays and in human tumor xenografts. In vivo studies with 3.19.3 consistently showed broad antitumor activity as a single agent across a panel of diverse subcutaneous and orthotopic xenograft models. Combination studies of 3.19.3 with cytotoxic drugs or anti–vascular endothelial growth factor agents showed significant improvements in antitumor activity over single-agent treatments alone with no apparent evidence of increased toxicity. Initial pharmacokinetic profiling studies in mice and nonhuman primates suggested that 3.19.3 has a predicted human half-life of 10 to 14 days. These studies provide preclinical data for 3.19.3 as a potential new antiangiogenic therapy as a single agent or in combination with chemotherapy or vascular endothelial growth factor inhibitors for the treatment of cancer. Mol Cancer Ther; 9(1); 145–56

Keywords
  • Ang2
  • angiogenesis
  • antibody

Introduction

The context-dependent role of angiopoietin-2 (Ang2) as a key regulator of vascular development and tumor-associated blood vessel formation consistently points to its potential as a next-generation antiangiogenesis cancer therapy (1–3). New insights into Ang2-Tie2 signaling interactions in developing vasculature and the dynamic interplay between tumor cells and proangiogenic cytokines in the tumor microenvironment have reshaped our views of tumor biology and blood vessel formation (4, 5). The highly coordinated recruitment of bone marrow–derived myeloid precursor cells and the temporal roles of vascular endothelial growth factor (VEGF), Ang1/2-Tie2, and placental growth factor (PlGF) in regulating endothelial cell proliferation and vessel formation remain active areas of research. Therapeutic approaches to targeting interactions in the tumor microenvironment, including VEGF antibodies (bevacizumab, VEGF-TRAP) and VEGF receptor small molecule antagonists (sorafinib, sunitinib, ZD6474, AZD2171), have clearly shown value to clinical oncology by improving patient survival and rapidly moving to front-line therapy across a number of tumor indications (6–8). The emerging biology involving the Ang2-Tie2 axis in regulating localized tumor blood vessel plasticity, recruitment of Tie2 expressing monocytes, and the regulation of extracellular matrix interactions in the vascular endothelium reinforces the role of Ang2 as an important angiogenesis target with potential for clinical development (2).

The angiopoeitin family comprises Ang1, Ang2, and Ang3/4, with their cognate receptor Tie2 and homologous receptor Tie1 (9). Whereas Ang1 and Ang2 share ∼55% identity in their primary amino acid sequence and seem to bind Tie2 with similar affinity, distinct functional roles have been described for each ligand (10). The downstream effects of Ang1 or Ang2 interactions with Tie2 are complex and context dependent, with both agonist and antagonist effects described (10). Recent data suggest that Ang1 regulates Tie2 activation at cell-cell and cell-matrix contacts (11, 12), whereas Ang2 acts as either an agonist or antagonist to Tie2 signaling. Ang2-Tie2 interactions have been shown to increase the permeability of the vascular endothelial layer of mature blood vessels, exposing endothelial cells to proliferation signals from angiogenesis inducers, including VEGF (3, 10, 13, 14). Ang2 upregulation effectively promotes a localized region of vessel “plasticity,” a vascular state that facilitates endothelial cell migration, and is responsive to vessel sprouting and proliferation drives, such as VEGF and fibroblast growth factor. As blood vessels mature, Ang2 levels decrease, reestablishing Ang1 and Tie2 interactions. Ang1, together with platelet-derived growth factor and transforming growth factor β, recruits periendothelial cells, and subsequently a more stable, rigid blood vasculature is formed. The modulation of Tie2 receptor activity by Ang1 and Ang2 seems to play a key role in new blood vessel maturation and the overall integrity of the developing tumor vascular endothelium.

There is growing evidence supporting the hypothesis that blocking Ang2-Tie2 receptor interactions would be an effective antiangiogenic therapy for the treatment of solid tumors (15). Deletion of Ang2 in mice results in postnatal lethality and terminal vascular defects associated with poor blood vessel and lymphatic integrity and associated edema (16, 17). Recently, anti-Ang2 antibodies, peptide-Fc fragments, and Ang2-specific RNA aptamers have shown in vivo efficacy and antiangiogenic activity in tumor models (18–20). One of these agents, AMG386, has recently progressed into early clinical trials (21). Evidence for elevated Ang2 levels has been reported in human tumor histopathology studies, and increased Ang2 expression has been associated with diverse tumor types, including breast, colon, lung, prostate, HCC, gastric, glioma, renal, ovarian, and melanoma (9, 10, 22–27). Ang2 expression seems dynamic and heterogeneous in tumors, correlating spatially with regions of high angiogenic activity. Conversely, Ang1 seems to be expressed constitutively in adults, consistent with a role for Ang1 in blood vessel maintenance and stability (10, 22, 28). Ang2 may also have additional roles in early stages of tumor growth, promoting the recruitment of Tie2-expressing monocyte populations (29) and through interaction with the α5β1 integrin leading to a stimulation of tumor cell motility and invasion (30).

In this report, we describe the first fully human Ang2-specific monoclonal antibody, 3.19.3, directed against Ang2. We show potent inhibition of angiogenesis and tumor growth inhibition across a broad panel of xenograft models and increased antitumor activity in combination with anti-VEGF agents and cytotoxic chemotherapy. In preclinical pharmacokinetic studies in nonhuman primates, 3.19.3 showed a sustained serum half-life and clearance profile consistent with a 10- to 14-day half-life in human. Taken together, these data provide preclinical rationale and support for the further clinical investigation of 3.19.3 as a novel antiangiogenic therapy for the treatment of cancer.

Materials and Methods

Antibody Generation

Abgenix XenoMouse strains XMG2 and XMG4 were immunized using recombinant human Ang2 purified from R&D Systems, Inc., to produce human anti-Ang2 (IgG2, κ) antibodies (31). Immunizations were carried out twice weekly for 4 wk, and anti-Ang2 specific immune responses were confirmed by ELISA. High titer mice were sacrificed, and B cells were isolated using CD90+ magnetic beads. Cell fusions with P3 myeloma cells in a 1:1 ratio were done using an Electro-Cell-Fusion generator (Genetronic, Inc.) according to manufacturer's instructions and seeded in 96-well plates at 5 × 106 cells per well. Hybridomas were grown for 10 to 30 d, and cell culture supernatants were subsequently screened for Ang2 binding by ELISA. Approximately 170 Ang2-positive hybridoma lines were identified, and supernatants were further tested for lack of cross reactivity to Ang1 and inhibition of Ang2 binding to Tie2 in an ELISA format. The top 27 neutralizing antibodies were progressed to subcloning, and 10 monoclonal lineages were characterized in detail from which 3.19.3 was selected as a top lead for further testing.

Antibody Binding Affinity and Selectivity

The affinity of 3.19.3 binding to human Ang2 was determined using a KinExA 3000 instrument in standard KD-controlled titrations against purified carrier-free recombinant Ang2 ligand (R&D Systems) coupled to polymethyl methacrylate beads (Sapidyne). For KD-controlled titration experiments, samples were allowed to equilibrate at room temperature for at least 5 h, whereas 1.5 h were sufficient for antibody-controlled titration experiments to fully reach equilibrium. All samples were prepared in degassed HBS-P buffer (0.01 mol/L HEPES, 0.15 mol/L NaCl, 0.005% Tween 20 (pH 7.4) with 100 μg/mL bovine serum albumin. Cy5-labeled goat α-human polyclonal antibody (Jackson ImmunoResearch Laboratories) was diluted (1,000-fold for KD-controlled or 10,000-fold for antibody-controlled titrations) and primed into the KinExA instrument injection line to serve as the secondary labeling antibody. Samples were run in triplicate, and affinity experiments were independently repeated at least thrice.

The binding of 3.19.3 to human Ang1 could not be accurately determined using KinExA, so Biacore technology was used to determine the affinity of the Ang1 interaction. Antibody 3.19.3 was immobilized on Biacore chips using routine amine coupling, and Ang1 titration samples were run against constant Ang1 ligand concentrations (ranging from 5.8 to 17.4 nmol/L) allowed to reach equilibrium in HBS-P running buffer containing 100 μg/mL bovine serum albumin. Binding curves were plotted as a function of antibody binding site concentration following a 1:1 binding model. Binding responses were converted to percentage of free Ang1, and a KD value was calculated to be 58 ± 20.4 nmol/L, with 95% upper and lower confidence intervals. Estimates of cross reactivity to murine Ang2 were determined using similar procedures.

Competitive Ang1-Ang2-Tie2 Binding ELISA

Antibody 3.19.3 was tested for its ability to bind Ang2 and Ang1 and block interactions with the human Tie2 receptor in a competitive ELISA assay. Ninety-six–well plates (Nunc Immunoplates) were coated with 100 μL/well of 4 μg/mL Tie-2 Fc (R&D Systems) diluted in 0.1 mol/L carbonate buffer (pH 9.6) at 4°C overnight. The plates were then blocked for at least 1 h with 200 L blocking buffer containing 0.5% bovine serum albumin (Sigma A-3059) and 0.1% Tween 20 in 1× PBS and further washed thrice with 0.05% PBS-T. Antibody 3.19.3 was added to the plates at various concentrations followed by 50 μL of a 2× stock of Biotin Ang2 (R&D Systems) at 100 ng/mL final concentration and incubated at room temperature for 2 h in 1× PBS. As a control, recombinant human Ang2 was mixed with the antibody and was also included. Plates were washed before the addition of 100 μl of streptavidin horseradish peroxidase (Pierce) diluted 1/5,000 in PBS + 0.05% Tween 20, and absorbance at 450 nm was determined.

Inhibition of Ang2-Tie2 Cell-Based Fluorescence-Activated Cell Sorting Assay

The ability of 3.19.3 to bind Ang2 and block interaction with human Tie2 receptor was evaluated in a competitive cell-based HEK293F/Tie2 fluorescence-activated cell sorting assay. Stably transfected HEK293F cells expressing human Tie2 on the cell surface were incubated in the presence of recombinant human Ang1 or Ang2 at a constant of 3 nmol/L ligand concentration. The ability of 3.19.3 to block binding to the receptor was then determined after a 2-h incubation at room temperature. Ang1 or Ang2 binding to human Tie2 was detected using anti-6x histidine antibodies or fluorescein-conjugated streptavidin and analyzed with a Calibur cytometer (Becton Dickinson). Experiments were done in duplicate and repeated independently at least thrice.

Inhibition of Downstream Tie2 Phosphorylation

The ability of 3.19.3 antibody to bind Ang2 and block Ang2-mediated Tie2 receptor phosphorylation was determined. HEK293-Tie2 cells expressing the full-length human Tie2 receptor were grown in 60-mm culture dishes and then serum starved for 18 h. Cells were stimulated for 15 to 20 min (37°C, 5% CO2) in the presence of purified human recombinant Ang2 (1 μg/mL, 36 nmol/L), preincubated with either 3.19.3 antibody or IgG isotype controls for 15 to 20 min at room temperature at 2:1 or 1:2 molar ratios corresponding to antibody concentrations of 72 nmol/L (5.4 μg/mL) or 17 nmol/L (1.35 μg/mL), respectively. Cells were harvested and lysed, and 0.8 mg total protein was immunoprecipitated using an anti-Tie2 antibody (MAB313, R&D Systems). Immune complexes were captured using Protein G beads. Tie2 protein was resolved on denaturing acrylamide gels under reduced conditions and transferred to polyvinylidene difluoride membranes. Blots were probed with either total Tie2 (Ab33, Cell Signaling Technology) or pTie2 (Y992) antibodies (R&D Systems) and detected using Amersham's horseradish peroxidase–conjugated chemiluminescent detection.

Matrigel Plug Angiogenesis Assays

The ability of 3.19.3 to inhibit in vivo angiogenesis was evaluated using Matrigel plug angiogenesis assays. Matrigel (0.5 mL) was premixed with 2 × 106 MCF7 cells and antibody 3.19.3 (100 μg/mL) before subcutaneous implantation into the right flank of BALB/c nude mice in cohorts of n = 5 animals per group. After 7 d, Matrigel plugs were removed and frozen for analysis. Cryostat sections (10μmol/L) were stained with phycoerythin-conjugated antimouse CD31 monoclonal antibodies, and immunoflurescence images analyzed to quantity vessel length, vessel ends, and nodes, as previously described (32). In subsequent assays, Matrigel (0.5 mL) was premixed with 2 × 106 MCF7 cells and implanted on both flanks of BALB/c nude mice. A single 20 mg/kg treatment of 3.19.3 antibody, IgG controls, or vehicle alone was given by i.p. injection to groups of five to six mice 24 h after implantation. The Matrigel plugs were grown for 10 d, and 0.1 mL of FITC-Dextran solution at 25 mg/mL was injected via the tail vein to each mouse, before being euthanized 20 min postinjection. The Matrigel plugs were weighed before homogenization, and the FITC-Dextran concentration in each supernatant was determined using a linear standard curve.

Xenograft Antitumor Efficacy Studies

The in vivo efficacy of 3.19.3 was evaluated in human tumor xenograft models in Swiss, NCr, or severe combined immunodeficient nude mice maintained in specific accordance with institutional and Institutional Animal Care and Use Committee guidelines. Briefly, human tumor cell lines, including SW620, LoVo, HT29, Colo205 and A431 cells, were cultured under standard conditions, and ∼1 to 5 × 106 cells implanted s.c. or orthotopically in 8- to 10-wk-old mice with cohort sizes averaging n = 10 per group. Tumor-bearing mice were randomized into control and treatment groups when tumors reached ∼0.2 cm3 in size. Antibody 3.19.3 treatment was given by i.p. injection after a twice weekly schedule at doses of 2 or 10 mg/kg for three to four cycles. Tumor volumes were calculated as follows: volume = 0.5*X^2*Y, or volume = (π/6)*X*Y*(X*Y)^(1/2), where X = tumor width and Y = tumor length. Growth inhibition was calculated from the start of treatment by comparison of the Geomean change in tumor volume for control and treated groups. Statistical analysis was done using the Student's t test. Xenograft tumor samples were treated for histology following standard zinc fixation and in paraffin imbedding procedures. Tissue sections were stained with anti-CD31 antibody (Becton Dickinson) using a Ventana automated slide processing system. Images were scanned using an Aperio imaging system and IHC-positive signal analyzed using color deconvolution imaging software.

Pharmacokinetics

The pharmacokinetics properties of the 3.19.3 antibody were evaluated in male CD1 mice and in cynomolgus monkeys to determine the antibody's predicted human half-life and clearance. Cohorts of 6-wk-old male mice were randomized to three subgroups (three mice per group), and serum samples were collected from individual animals after a single i.p. dose of 20 mg/kg (500 μg) per animal at prespecified time points throughout the study. Serum concentrations of the 3.19.3 antibody were determined using a goat anti-human IgG Fc ELISA, and the composite serum concentration-time profile plotted after a two-compartment pharmacokinetics model. Similarly, the pharmacokinetics profile of 3.19.3 was evaluated in cynomolgus monkeys after single bolus i.v. doses of 6.0 or 60 mg/kg given to cohorts of four animals each. Serum samples were collected and analyzed for 3.19.3 levels before dosing, at 4- and 8-h time points on day 1, and daily intervals over the 22 days. Serum was also collected on day 22 to test for monkey anti-human antibody (MAHA) immune responses at the end of study.

Results

Generation of Ang2-Specific Monoclonal Antibodies

Ang2-positive hybridomas were identified producing human IgG2 antibodies after XenoMouse immunization and from a total of 176 lines were prioritized on the basis of their ability to block Ang2-Tie2 binding in a Tie2/Fc competitive ELISA assay. Epitope binning was used to identify antibodies binding to distinct conformational epitopes in the fibrinogen-like domain of Ang2, and 27 hybridomas were selected for cloning and further characterization. Ten clones were found to inhibit Ang2-induced phosphorylation of Tie2. One of the 10 antibodies (clone 3.19.3) was also found to cross-react with mouse Ang2 and was chosen for further investigation.

Potent and Specific Binding of 3.19.3 to Ang2

The binding affinity of 3.19.3 to purified recombinant human Ang1 and Ang2 ligands was determined using Biacore and KinExA technology. Initial Biacore results for 3.19.3 suggested potent (subpicomolar) binding affinity for Ang2 and an estimated equilibrium dissociation constant (KD) of 58 nm for Ang1 (Supplementary Fig. S1). However, subsequent studies showed the Ang2 binding displayed complex nonlinear kinetics and slow off rates, suggesting the initial Ang2 results may be unreliable. To more accurately determine the 3.19.3 binding equilibrium and off-rates for Ang2, solution phase KinExA studies were done and a KD of 86.4 pmol/L was determined with a 95% confidence interval ranging from 64.3 to 98.7 pmol/L (Supplementary Fig. S1). These results suggest 3.19.3 potently binds Ang2 with at least 500-fold or greater affinity compared with Ang1.

Inhibition of Ang2-Tie2 Binding and Tie2 Phosphorylation

A competitive Tie2-Fc ELISA assay was used to independently confirm the relative affinity of 3.19.3 binding to Ang2 and Ang1. In these experiments, 3.19.3 was found to bind Ang2 with an IC50 value of 506 pmol/L, whereas Ang1 binding was essentially undetectable (Fig. 1A). The 3.19.3 antibody was further tested for the ability to block functional interactions with Ang2 and full-length human Tie2 receptor expressed on the cell surface in a cell-based fluorescence-activated cell sorting assay. Stably transfected HEK293F cells expressing human Tie2 were incubated in the presence of 3.19.3 antibody and biotinylated human Ang2 or Ang1 (3 nmol/L), and the Ang2 that remained bound to the Tie2 receptor following incubation was determined by quantitative fluorescence-activated cell sorting analysis. As shown in Fig. 1B, antibody 3.19.3 inhibited Ang2 binding to Tie2 with an EC50 of 157 pmol/L. In contrast, 3.19.3's ability to inhibit Ang1 binding to Tie2 resulted in an estimated EC50 value of >170 nmol/L (Fig. 1B). This experiment was independently repeated thrice, with an average EC50 value of 478 pmol/L for Ang2 whereas the average EC50 value of 3.19.3 binding to Ang1 was > 194 nmol/L. Thus, in terms of functional inhibition of Ang2 binding to its receptor, 3.19.3 has a high degree of selectivity for Ang2 over Ang1.

Figure 1.
Figure 1.

Antibody 3.19.3 binds human Ang2 and blocks Tie2-mediated phosphorylation and downstream signaling. A, the ability of antibody 3.19.3 to selectively bind Ang2 in a competitive Tie2-Fc ELISA assay was determined with an IC50 value of 506 pmol/L, whereas Ang1 binding was essentially undetectable. B, inhibition of Ang2 binding to human Tie2 in a cell-based HEK293 assay. Antibody 3.19.3 was tested for the ability to block Ang2 interactions with Tie2 expressed on the cell surface using biotinylated human Ang1 or Ang2. Percentage of inhibition of binding was calculated as a fraction of the maximum signal obtained without antibody and suggested EC50 values of 150 pmol/L affinity to Ang2, whereas the Ang1 binding did not reach saturation up to concentrations of 1 μmol/L antibody. C, inhibition of Ang2-induced Tie2 phosphorylation in HEK293/Tie2 cells. IgG control antibody (lanes 1, 2) was compared alongside 3.19.3 (lanes 3, 4) for the ability to block recombinant Ang2 (1 μg/mL)–mediated Tie2 phosphorlylation at 2:1 molar ratios (lanes 1, 3) or 1:2 molar ratios (lanes 2, 4) corresponding to 1.35 μg/mL or 5.4 μg/mL antibody, respectively. Controls for maximal Ang2 induction are shown (lane 5) or no treatment (lane 6), indicating baseline pTie2 activity in this assay.

Antibody 3.19.3 was tested for the ability to directly block Ang2-induced Tie2 receptor phosphorylation using stably transfected HEK293F-Tie2 cells treated with recombinant human Ang2 to stimulate Tie2 phosphorylation (Fig. 1C). Tie2-expressing cells were preincubated with Ang2 (1 μg/mL), and antibody 3.19.3 at 2:1 or 1:2 molar ratios, and 3.19.3 was found to significantly block pTie2 with an estimated IC50 value of 580 pmol/L. Using Ang1 to stimulate the receptor, 3.19.3 antibody inhibited Ang1-mediated Tie2 phosphorylation with an IC50 value of 92 nmol/L. These results are consistent with the previous binding results and confirm that 3.19.3 antibody is a potent and selective inhibitor of Ang2-mediated functional interactions with the Tie2 receptor.

Pharmacokinetics of 3.19.3 Antibody in CD1 Mice and Nonhuman Primates

The pharmacokinetic profile of 3.19.3 was determined in mice at various time intervals after a single dose of 20 mg/kg. Plasma levels of 3.19.3 were found to reach a Cmax of 350 μg/mL after 8 to 10 hours and were then slowly cleared from the blood stream after a two-compartment pharmacokinetics model (Supplementary Fig. S2). The single-dose terminal elimination half-life of the antibody in mice was estimated at 11 to 12 days. The pharmacokinetics profile of 3.19.3 was also evaluated in cynomolgus monkeys (Supplementary Table S1). Two dose levels of 3.19.3 (6.0 and 60 mg/kg) were evaluated after i.v. administration to cohorts of four animals each. The initial estimated half-life of 3.19.3 was determined to be between 3 and 4 days. The volume of distribution of 3.19.3 was comparable with total plasma volume, and the intrinsic clearance seemed to be dose independent across the range tested. Analysis of the MAHA immunogenicity assay indicated no positive responses or blocking antibodies against 3.19.3 could be detected at the end of study. These results suggest 3.19.3 is estimated to have a predicted human half-life of 10 to 14 days, which is similar to that observed for other human therapeutic antibodies (33).

Activity of 3.19.3 in Matrigel Plug Angiogenesis Assays

The ability of 3.19.3 to inhibit angiogenesis in vivo was evaluated using two independent Matrigel plug assays. MCF7 tumor cells were premixed with Matrigel and 3.19.3 antibody or IgG controls at final concentrations of 100 μg/mL and implanted in nude mice for 7 days in cohorts of n = 5 per group. The ability of 3.19.3 to inhibit new blood vessel formation was subsequently evaluated with anti-CD31 antibodies to quantify the vessel length, number, and nodes (Fig. 2A–C). Matrigel plugs from the 3.19.3-treated groups were noticeably pale white in appearance compared with controls, consistent with a reduction in new blood vessel formation (Supplementary Fig. S3). The 3.19.3-treated group showed 86% reduction (P < 0.01) in mean vessel length compared with the MCF7 control group (Fig. 2A). Similarly, an 86% reduction in mean vessel nodes (Fig. 2B) and an 82% reduction in mean vessel ends (Fig. 2C) were observed, whereas control treatments had no significant effects on any of these parameters (Fig. 2A–C). The ability of 3.19.3 to inhibit functional blood vessel formation was also determined after a single dose of 3.19.3 (20 mg/kg) given 24 hours after implantation (Fig. 2D). In this assay, FITC-Dextran was used to quantify blood vessel perfusion into MCF7 containing Matrigel plugs after 10 days of growth. A single dose of 3.19.3 antibody resulted in 89% reduction (P < 0.05) in blood vessel formation compared with vehicle controls (Fig. 2D). Taken together, these results suggest that 3.19.3 inhibits angiogenesis and functional blood vessel formation in vivo in a MCF7-driven Matrigel plug angiogenesis assay.

Figure 2.
Figure 2.

Inhibition of blood vessel formation in Matrigel plug assays. The antiangiogenic effects of 3.19.3 was determined using MCF7-driven Matrigel plug implants and quantified according to Wild et al. (32) by measuring CD31-stained frozen sections at the end of the study (day 7). Determination of the effects on (A) vessel length, (B) vessel nodes, (C) vessel ends, relative to IgG controls. D, inhibition of functional tumor blood vessel formation after 3.19.3 treatment (20 mg/kg) given 24 h after Matrigel implantation. Mean FITC-Dextran uptake was determined by measuring fluorescence at the end of the study (day 10). Matrigel plugs alone with no MCF7 cells implanted (Control IgG) are shown as an indication of background, compared with Matrigel plugs containing MCF7 cells with no treatment (MCF7 cells) to indicate maximal angiogenesis, alongside Matrigel plugs with MCF7 cells + Control IgG treatment, and mice harboring Matrigel plugs with MCF7 cells + 3.19.3 treatment.

Activity of 3.19.3 in Xenograft Tumor Efficacy Models

The antitumor efficacy of 3.19.3 was evaluated in a series of subcutaneous and orthotopic human tumor xenograft models. Antibody 3.19.3 was given at 2 or 10 mg/kg by i.p. injection after a twice-weekly dosing schedule. Significant antitumor activity was seen with single agent of 3.19.3 treatment in four models tested at 2 mg/kg dose and across seven tumor xenograft models at the 10 mg/kg dose (Table 1A; Fig. 3). The tumor growth inhibition observed was found to range from 33% to 71% (P < 0.05) across these models, and no significant toxicity was observed in any of these studies as judged by changes in body weights (Supplementary Fig. S4). Antibody 3.19.3 also showed significant antitumor activity at doses of 10 mg/kg in the MDA-MB-231 or MCF7 orthotopic breast cancer models implanted in the mammary fat pad of mice (P < 0.05; Table 1). Antibody 3.19.3 was found to inhibit the growth of orthotopic MCF7 tumors by 73% (P < 0.05) and MDA-MB-231 tumors by 57% (P < 0.05) compared with vehicle-treated controls. Based on these studies, we chose a twice weekly 10 mg/kg dosing regimen for subsequent in vivo experiments to ensure we maintained high circulating plasma levels of 3.19.3 (in molar excess of circulating Ang2) to block Ang2 function and mitigate against any potential MAHA (mouse anti-human antibody) responses, which could affect pharmacokinetics and clearance after repeated dosing.

View this table:
Table 1.

In vivo single-agent and combination efficacy of 3.19.3 in subcutaneous and orthotopic xenograft models

Figure 3.
Figure 3.

Single-agent tumor xenograft efficacy observed after 3.19.3 treatment. Antibody 3.19.3 was given by i.p. injection after a twice weekly schedule for 3 to 4 wk. A, activity in xenograft A431 (vehicle alone ▪; 2 mg/kg •; 10 mg/kg ▴). B, activity in xenograft Colo 205 (vehicle alone ▪; 2 mg/kg •; 10 mg/kg ▴). C, activity in xenograft HT29 (vehicle alone ▪; 10 mg/kg •). D, activity in orthotopic xenograft MDA-MB231 (vehicle alone▪; 10 mg/kg •). Efficacy results and statistical analysis are summarized in Table 1.

To determine if the antitumor activity observed with 3.19.3 correlated with an antiangiogenic mechanism of action, tumor models representing subcutaneous and orthotopic sites of implantation (Colo205 and MDA-MB-231, respectively) were evaluated for CD31 staining by immunohistochemistry. Reductions in tumor-associated blood vessels could be readily observed in micrographs from sections of Colo205 and MDA-MB-231–treated tumors (Fig. 4). Subsequent quantitative analysis of CD31 staining confirmed these findings and indicated Colo205 IgG-treated samples (Fig. 4A) contained 4.20% CD31-positive vascular staining, whereas 3.19.3 samples showed 1.53% of total CD31 staining, representing a 64% reduction in blood vessel formation after treatment. Interestingly, other histopathologic changes corresponding to increased areas of tumor necrosis and restricted tumor growth to regions immediately surrounding the vasculature were also observed (Fig. 4A). Similarly, the orthotopic MDA-MB-231–treated tumors also displayed marked changes in overall tumor blood vessel microdensity (Fig. 4B). Quantitative analysis of CD31 staining after antibody 3.19.3 treatment from these studies showed a similar 40% reduction in CD31-positive blood vessels (P < 0.05) compared with IgG controls (Supplementary Fig. S5). No obvious evidence of increased tumor invasion after 3.19.3 treatment could be determined from this analysis. These results suggest 3.19.3 treatment significantly inhibits in vivo blood vessel formation and blocks tumor angiogenesis in both subcutaneous and orthotopic models.

Figure 4.
Figure 4.

Inhibition of tumor angiogenesis after 3.19.3 treatment. A, immunohistochemistry of Colo205 tumors fixed and stained with anti-CD31 antibody after treatment to visualize endothelial cells and tumor blood vessel formation. Representative (20×) images of IgG controls or 3.19.3 treatment (after 4 doses of 10 mg/kg) are shown, and quantitative image analysis indicated a 64% overall reduction in CD31 staining relative to controls. B, immunohistochemistry analysis of orthotopic MDA MB231-treated tumors stained with anti-CD31 antibody. Representative (10×) images of IgG controls or 3.19.3 treatment are shown, and quantitative image analysis (Supplementary Fig. S5) indicated a 40% reduction in CD31 staining from IgG-treated controls.

Activity of 3.19.3 in Combination with VEGF Inhibitors and Cytotoxic Drugs in Xenograft Models

The antitumor activity of 3.19.3 was evaluated in combination with small molecule VEGF receptor tyrosine kinase inhibitors vandetanib (ZD6474, ZACTIMA) and cediranib (AZD2171, Recentin) in subcutaneous human tumor xenograft models (Fig. 5; Table 1B). Antibody 3.19.3 was also evaluated in combination with AZ10167514, a structurally similar analogue of AZD2171 (34), and DC101 antibody to block binding of VEGF to murine VEGF receptor-2 (Supplementary Fig. S6). Results of the combined effects of antibody 3.19.3 and VEGF inhibitor treatments and statistical analysis are also found summarized in Table 1B. Interestingly, the combination of 3.19.3 treatment together with anti-VEGF agents consistently led to significant improvements (P < 0.05) in tumor growth inhibition and efficacy ranging from 81% to 96% (Fig. 5A, B) over the single-agent treatments alone, with an average body weight loss of 2.3%, with a range of 0.4% to a maximum of 6.4% (Supplementary Fig. S4C).

Figure 5.
Figure 5.

Combination studies with 3.19.3 and VEGF inhibitors or chemotherapy agents. A, combination efficacy with vandetanib (Zactima, ZD 6474) in xenograft LoVo (vehicle alone ▪; 3.19.3 at 10 mg/kg i.p., twice weekly •; Vandetanib at 50 mg/kg p.o. daily ▴; combination ♦). B, combination with cediranib (Recentin, AZD 2171) in xenograft HT29 (vehicle alone ▪; 3.19.3 at 10 mg/kg i.p., twice weekly •; cediranib at 1.5 mg/kg p.o. daily ▴; combination ♦). C, combination with 5FU in xenograft LoVo (vehicle alone ▪; 3.19.3 at 10 mg/kg i.p., twice weekly •; 5FU at 100 mg/kg i.p. weekly ▴; combination ♦). D, Combination with irinotecan in xenograft HT29 (vehicle alone ▪; 3.19.3 at 10 mg/kg i.p., twice weekly •; Irinotecan at 35 mg/kg i.v. weekly ▴; combination ♦). Efficacy results and statistical analysis are summarized in Table 1.

The antitumor activity of 3.19.3 in combination with chemotherapy agents, including docetaxel, 5-fluorouracil, irinotecan, oxaliplatin, and gemcitabine, was also determined to evaluate the effects of these combinations on in vivo efficacy and tolerability. Results from these studies are summarized in Table 1B. In each study, antibody 3.19.3, given in combination with these widely used cancer therapies, resulted in additive efficacy compared with either antibody or chemotherapy agent alone (Fig. 5C, D; Table 1B) with no evidence of body weight loss (Supplementary Fig. S4). Taken together, these results suggest 3.19.3 acts in vivo to inhibit tumor angiogenesis by blocking Ang2 function, resulting in broad antitumor effects as a single agent and enhancing antitumor effects in combination with VEGF inhibitors and chemotherapy agents in xenograft tumor models.

Discussion

The Ang2-Tie2 system serves an essential role in blood vessel formation during embryogenesis and in pathologic blood vessel remodeling associated with inflammation and tumorigenesis (10, 22–27). Here we report data for a novel fully human anti-Ang2 monoclonal antibody, 3.19.3, generated by immunization of the XenoMouse strains with human recombinant Ang2 protein. We showed that 3.19.3 selectively binds Ang2 with 86 pmol/L affinity and blocks Ang2-mediated phosphorylation of the Tie2 receptor in cell-based assays. Epitope mapping studies suggested the 3.19.3 antibody binds in a manner consistent with interactions in the COOH terminal fibrinogen-like domain of Ang2, which has previously been shown to be important for Ang2-Tie2 receptor binding (10). In cell-based functional assays, 3.19.3 also showed >200-fold selectivity for Ang2 over the related angiopoietin Ang1. These results are consistent with in vitro affinity measurements and suggest the primary mechanism of action of 3.19.3 in vivo is mediated through inhibition of Ang2.

Consistent with a direct effect on tumor angiogenesis, the 3.19.3 antibody was found to potently inhibit in vivo blood vessel development. Treatment with 3.19.3 significantly reduced blood vessel formation in Matrigel plug assays containing MCF7 cells and reduced CD31 staining in all tumor models examined. These results are further supported by the observation of necrosis and the restriction of tumor cell growth to diffusion-limited region immediately surrounding the surviving vasculature. These data are consistent with Ang2 functioning as a proangiogenic cytokine and blocking Ang2 function leads to an antiangiogenic phenotype, rather than direct effects on tumor proliferation. The specific role of Ang2 in mediating recruitment of Tie2-expressing monocytes (29) and the potential effects of Ang2 signaling through extracellular matrix interactions including α5β1 integrins (30) remain to be determined and are the subject of ongoing investigations.

The in vivo efficacy of 3.19.3 was determined against a panel of xenograft tumor models to establish whether the effects of Ang2 inhibition on tumor growth were model specific or broader, as have been reported in preclinical studies with bevacizumab (35). Data from these studies consistently showed 3.19.3 single-agent activity ranging from 30% to 70% inhibition across both subcutaneous and orthotopic xenograft tumor models. Previous reports of Ang2 inhibition of xenograft growth using soluble Tie2 receptors, Ang2 peptibody or synthetic aptamers, showed similar effects; however, the results were somewhat more limited (18–20). Whether the enhanced in vivo activity observed with 3.19.3 is due to more effective Ang2 binding, selectively binding Ang2 and not Ang1, or blocking some additional functions associated with murine-derived Ang2 biology remains to be determined. Interestingly, preliminary pharmacokinetics modeling indicates the circulating plasma levels of 3.19.3 after 10 mg/kg twice weekly dosing remain in molar excess of free Ang2 ligand, suggesting complete suppression of Ang2 occurs in these studies. Interestingly, 3.19.3 binding to mouse Ang2 in cross reactivity studies indicated significantly less affinity to murine Ang2 with KD = 3 nm, which is less than that observed for human Ang2 (KD = 86 pmol/L). This raises the possibility that the in vivo efficacy observed with 3.19.3 may be achievable with lower predicted human doses or with less frequent administration in clinical trials. Treatments with 3.19.3 resulted in no significant differences in body weight loss observed in any of these studies across multiple genetic backgrounds. The pharmacokinetic profile of antibody 3.19.3 in both mouse and primates was similar those observed for other human monoclonal antibodies in these species (33), suggesting 3.19.3 should have a predicted 10- to 14-day half-life in humans. Taken together, these results are consistent with Ang2 functioning as a key target in tumor angiogenesis, and selective inhibition of Ang2 may translate to an effective antiangiogenesis therapy with broad potential similar to VEGF.

We evaluated the in vivo activity of 3.19.3 together with VEGF inhibitors or chemotherapy agents to test the potential for administration as a combination therapy. Rational clinical combinations involving 3.19.3 and inhibition of angiogenesis through the Ang2-Tie2 pathway may offer significant clinical opportunities to improve response rates and extend progression-free survival. The antitumor efficacy of both VEGF mediated angiogenesis inhibition, and the effects of cytotoxic drug treatments were both significantly improved in combination with 3.19.3 at doses and schedules that resulted in no adverse effects. Previous studies involving Ang2 inhibitors have only reported single-agent activity in similar models (18). VEGF inhibitors have shown encouraging clinical activity, improving patient time to progression and survival across a number of clinical indications (6–8). The potential for 3.19.3 to be used in addition to these treatments without increasing toxicity holds promise for future trials, as the majority of patients receiving VEGF therapy will eventually stop treatment due to toxicity or disease progression (36–40). The mechanisms behind the enhanced activity observed in these preclinical models may be due to a more complete angiogenic blockade or may suggest that Ang2 and VEGF function independently to promote tumor angiogenesis. Subsequent studies with 3.19.3 could provide insights regarding intrinsic or acquired VEGF resistance and may eventually lead to a more comprehensive clinical treatment for vascular modulation.

When combined with chemotherapy, 3.19.3 also displayed significant improvements in antitumor efficacy with no evidence of increased toxicity in preclinical models. Similarly, VEGF-based therapies have shown greater clinical activity when combined with cytotoxic drugs, presumably due in part to normalization of the existing tumor vasculature in the clinic (41, 42). Whether the combined effects of Ang2-Tie2 inhibition with cytotoxic therapy are due to normalization of the vascular endothelium in a manner similar to VEGF inhibitors has not been fully determined (43, 44). Inhibition of Ang2 does seem to have some direct effect on tumor blood vessel size and diameter, but whether this represents blood vessel normalization, inhibition of bone marrow–derived hematopoeitic progenitor cell recruitment, or some alternative mechanism will require further investigation.

In summary, this work describes the first novel human monoclonal antibody directed against Ang2 and provides additional preclinical data in support of Ang2 inhibition as an effective antiangiogenic therapy. The predicted human half-life and reduced immunogenicity profile associated with a fully human anti-Ang2 antibody may offer distinct clinical advantages and provide options as a single agent and in combinations with other antiangiogenic therapies for the treatment of solid tumors.

Disclosure of Potential Conflicts of Interest

J.L. Brown, A.Z. Cao, M. Pinzon-Ortiz, J. Kendrew, C. Reimer, S. Wen, S. Emery, B. McDermott, L. Pablo, P. McCoon, V. Bedian, and D.C. Blakey are employees of AstraZeneca. J.Q. Zhou, and M. Tabrizi are employees of Intradigm and AnaptysBio, respectively.

Acknowledgments

We thank Juliane Jurgensmeier and Andy Ryan (AstraZeneca Pharmaceuticals) for Recentin and Zactima, Andrew Drake and Scott Klakamp (Abgenix, Inc.) for the affinity determinations with KinExA and Biacore instruments, Theresa LaVallee (MedImmune, LLC) for reviewing the manuscript, and members of the MedImmune and AstraZeneca EPT3 project teams for collaborative and insightful comments.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Footnotes

  • Note: Supplementary material for this article is available at Molecular Cancer Therapeutics Online (http://mct.aacrjournals.org/).

    • Received June 22, 2009.
    • Revision received October 16, 2009.
    • Accepted November 9, 2009.

References

  1. ↵
    1. Ferrara N,
    2. Mass RD,
    3. Campa C,
    4. Kim R
    . Targeting VEGF-A to treat cancer and age-related macular degeneration. Annu Rev Med 2007;58:491–504.
    OpenUrlCrossRefPubMed
  2. ↵
    1. Augustin HG,
    2. Koh GY,
    3. Thurston G,
    4. Alitalo K
    . Control of vascular morphogenesis and homeostasis through the angiopoietin-Tie system. Nat Rev Mol Cell Biol 2009;10:165–77.
    OpenUrlCrossRefPubMed
  3. ↵
    1. Hanahan D
    . Signaling vascular morphogenesis and maintenance. Science 1997;277:48–50.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Lewis CE,
    2. Pollard JW
    . Distinct role of macrophages in different tumor microenvironments. Cancer Res 2006;66:605–12.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. De Palma M,
    2. Murdoch C,
    3. Venneri MA,
    4. Naldini L,
    5. Lewis CE
    . Tie2-expressing monocytes:regulation of tumor angiogenesis and therapeutic implications. Trends Immunol 2007;28:519–24.
    OpenUrlCrossRefPubMed
  6. ↵
    1. Hurwitz H
    . Integrating the anti-VEGF-A humanized monoclonal antibody bevacizumab with chemotherapy in advanced colorectal cancer. Clin Colorectal Cancer 2004;4:S62–8.
    OpenUrlCrossRefPubMed
    1. Marx J
    . Cancer: encouraging results for second-generation antiangiogenesis drugs. Science 2005;308:1248–9.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Wilhelm S,
    2. Carter C,
    3. Lynch M,
    4. et al
    . Discovery and development of sorafenib: a multikinase inhibitor for treating cancer. Nat Rev Drug Discov 2006;5:835–44.
    OpenUrlCrossRefPubMed
  9. ↵
    1. Shim WS,
    2. Ho IA,
    3. Wong PE
    . Angiopoietin: a TIE(d) balance in tumor angiogenesis. Mol Cancer Res 2007;5:655–65.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    1. Maisonpierre PC,
    2. Suri C,
    3. Jones PF,
    4. et al
    . Angiopoietin-2, a natural antagonist for Tie2 that disrupts in vivo angiogenesis. Science 1997;277:55–60.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Saharinen P,
    2. Eklund L,
    3. Miettinen J,
    4. et al
    . Angiopoietins assemble distinct Tie2 signalling complexes in endothelial cell-cell and cell-matrix contacts. Nat Cell Biol 2008;10:527–37.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Fukuhara S,
    2. Sako K,
    3. Minami T,
    4. et al
    . Differential function of Tie2 at cell-cell contacts and cell-substratum contacts regulated by angiopoietin-1. Nat Cell Biol 2008;10:513–26.
    OpenUrlCrossRefPubMed
  13. ↵
    1. Asahara T,
    2. Chen D,
    3. Takahashi T,
    4. et al
    . Tie2 receptor ligands, angiopoietin-1 and angiopoietin-2, modulate VEGF-induced postnatal neovascularization. Circ Res 1998;83:233–40.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Inoue M,
    2. Hager JH,
    3. Ferrara N,
    4. Gerber HP,
    5. Hanahan D
    . VEGF-A has a critical, nonredundant role in angiogenic switching and pancreatic β cell carcinogenesis. Cancer Cell 2002;1:193–202.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Nasarre P,
    2. Thomas M,
    3. Kruse K,
    4. et al
    . Host-derived angiopoietin-2 affects early stages of tumor development and vessel maturation but is dispensable for later stages of tumor growth. Cancer Res 2009;69:1324–33.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Gale NW,
    2. Thurston G,
    3. Hackett SF,
    4. et al
    . Angiopoietin-2 is required for postnatal angiogenesis and lymphatic patterning, and only the latter role is rescued by angiopoietin-1. Dev Cell 2002;3:411–23.
    OpenUrlCrossRefPubMed
  17. ↵
    1. Dellinger M,
    2. Hunter R,
    3. Bernas M,
    4. et al
    . Defective remodeling and maturation of the lymphatic vasculature in angiopoietin-2 deficient mice. Dev Biol 2008;319:309–20.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Oliner J,
    2. Min H,
    3. Leal J,
    4. et al
    . Suppression of angiogenesis and tumor growth by selective inhibition of angiopoietin-2. Cancer Cell 2004;6:507–16.
    OpenUrlCrossRefPubMed
    1. White RR,
    2. Shan S,
    3. Rusconi CP,
    4. et al
    . Inhibition of rat corneal angiogenesis by a nuclease-resistant RNA aptamer specific for angiopoietin-2. Proc Natl Acad Sci U S A 2003;100:5028–33.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    1. Sarraf-Yazdi S,
    2. Mi J,
    3. Moeller BJ,
    4. et al
    . Inhibition of in vivo tumor angiogenesis and growth via systemic delivery of an angiopoietin 2-specific RNA aptamer. J Surg Res 2008;146:16–23.
    OpenUrlCrossRefPubMed
  21. ↵
    43rd Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, USA; June 1–5, 2007.
  22. ↵
    1. Tait CR,
    2. Jones PF
    . Angiopoietins in tumours: the angiogenic switch. J Pathol 2004;204:1–10.
    OpenUrlCrossRefPubMed
    1. Ahmad SA,
    2. Liu W,
    3. Jung YD,
    4. et al
    . Differential expression of angiopoietin-1 and angiopoietin-2 in colon carcinoma. A possible mechanism for the initiation of angiogenesis. Cancer 2001;92:1138–43.
    OpenUrlCrossRefPubMed
    1. Tanaka S,
    2. Mori M,
    3. Sakamoto Y,
    4. Makuuchi M,
    5. Sugimachi K,
    6. Wands JR
    . Biologic significance of angiopoietin-2 expression in human hepatocellular carcinoma. J Clin Invest 1999;103:341–5.
    OpenUrlPubMed
    1. Eggert A,
    2. Ikegaki N,
    3. Kwiatkowski J,
    4. Zhao H,
    5. Brodeur GM,
    6. Himelstein BP
    . High-level expression of angiogenic factors is associated with advanced tumor stage in human neuroblastomas. Clin Cancer Res 2000;6:1900–8.
    OpenUrlAbstract/FREE Full Text
    1. Etoh T,
    2. Inoue H,
    3. Tanaka S,
    4. Barnard GF,
    5. Kitano S,
    6. Mori M
    . Angiopoietin-2 is related to tumor angiogenesis in gastric carcinoma: possible in vivo regulation via induction of proteases. Cancer Res 2001;61:2145–53.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    1. Holash J,
    2. Maisonpierre PC,
    3. Compton D,
    4. et al
    . Vessel cooption, regression, and growth in tumors mediated by angiopoietins and VEGF. Science 1999;284:1994–8.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Zagzag D,
    2. Hooper A,
    3. Friedlander DR,
    4. et al
    . In situ expression of angiopoietins in astrocytomas identifies angiopoietin-2 as an early marker of tumor angiogenesis. Exp Neurol 1999;159:391–400.
    OpenUrlCrossRefPubMed
  29. ↵
    1. De Palma M,
    2. Venneri MA,
    3. Galli R,
    4. et al
    . Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 2005;8:211–26.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Imanishi Y,
    2. Hu B,
    3. Jarzynka MJ,
    4. et al
    . Angiopoietin-2 stimulates breast cancer metastasis through the α5β1 integrin-mediated pathway. Cancer Res 2007;67:4254–63.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    1. Mendez MJ,
    2. Green LL,
    3. Corvalan JR,
    4. et al
    . Functional transplant of megabase human immunoglobulin loci recapitulates human antibody response in mice. Nat Genet 1997;15:146–56.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Wild R,
    2. Ramakrishnan S,
    3. Sedgewick J,
    4. Griffioen AW
    . Quantitative assessment of angiogenesis and tumour vessels architecture by computer-assisted digital image analysis: effects of VEGF-toxin conjugate on tumour microvessel density. Microvasc Res 2000;59:368–76.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Tabrizi MA,
    2. Tseng CM,
    3. Roskos LK
    . Elimination mechanisms of therapeutic monoclonal antibodies. Drug Discov Today 2006;11:81–8.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Wedge SR,
    2. Kendrew J,
    3. Hennequin LF,
    4. et al
    . AZD2171: a highly potent, orally bioavailable, vascular endothelial growth factor receptor-2 tyrosine kinase inhibitor for the treatment of cancer. Cancer Res 2005;65:4389–400.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. Kim KJ,
    2. Li B,
    3. Winer J,
    4. et al
    . Inhibition of vascular endothelial growth factor-induced angiogenesis suppresses tumour growth in vivo. Nature 1993;362:841–4.
    OpenUrlCrossRefPubMed
  36. ↵
    1. Herbst RS
    . Therapeutic options to target angiogenesis in human malignancies. Expert Opin Emerg Drugs 2006;11:635–50.
    OpenUrlCrossRefPubMed
    1. Wysocki PJ,
    2. Zolnierek J,
    3. Szczylik C,
    4. Mackiewicz A
    . Targeted therapy of renal cell cancer. Curr Opin Investig Drugs 2008;9:570–5.
    OpenUrlPubMed
    1. Gettinger S
    . Targeted therapy in advanced non-small-cell lung cancer. Semin Respir Crit Care Med 2008;29:291–301.
    OpenUrlCrossRefPubMed
    1. Seiwert TY,
    2. Cohen EE
    . Targeting angiogenesis in head and neck cancer. Semin Oncol 2008;35:274–85.
    OpenUrlCrossRefPubMed
  40. ↵
    1. Rosa DD,
    2. Clamp AR,
    3. Collinson F,
    4. Jayson GC
    . Antiangiogenic therapy for ovarian cancer. Curr Opin Oncol 2007;19:497–505.
    OpenUrlCrossRefPubMed
  41. ↵
    1. Hurwitz H,
    2. Fehrenbacher L,
    3. Novotny W,
    4. et al
    . Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335–42.
    OpenUrlCrossRefPubMed
  42. ↵
    1. Jain RK,
    2. Duda DG,
    3. Clark JW,
    4. Loeffler JS
    . Lessons from phase III clinical trials on anti-VEGF therapy for cancer. Nat Clin Pract Oncol 2006;3:24–40.
    OpenUrlCrossRefPubMed
  43. ↵
    1. Jain RK
    . Normalization of tumor vasculature: an emerging concept in antiangiogenic therapy. Science 2005;307:58–62.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Fukumura D,
    2. Jain RK
    . Tumor microvasculature and microenvironment: targets for anti-angiogenesis and normalization. Microvasc Res 2007;74:72–84.
    OpenUrlCrossRefPubMed
Back to top
Molecular Cancer Therapeutics: 9 (1)
January 2010
Volume 9, Issue 1

Sign up for alerts

View this article with LENS

Open full page PDF
Article Alerts
Email Article
Citation Tools
A Human Monoclonal Anti-ANG2 Antibody Leads to Broad Antitumor Activity in Combination with VEGF Inhibitors and Chemotherapy Agents in Preclinical Models
Jeffrey L. Brown, Z. Alexander Cao, Maria Pinzon-Ortiz, Jane Kendrew, Corinne Reimer, Shenghua Wen, Joe Q. Zhou, Mohammad Tabrizi, Steve Emery, Brenda McDermott, Lourdes Pablo, Patricia McCoon, Vahe Bedian and David C. Blakey
Mol Cancer Ther January 1 2010 (9) (1) 145-156; DOI: 10.1158/1535-7163.MCT-09-0554
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Advertisement