Targeting VEGF-mediated Tumor Angiogenesis in Cancer Therapy

Abstracts, Day 1: Monday, April 28, 2014

Keynote Lecture

Heterogeneity of the Tumor Vasculature: Why Doesn't Anti-VEGF / VEGF Receptor Therapy Work Better?
Harold F. Dvorak, MD, The Center for Vascular Biology Research and the Department of Pathology, Beth Israel Deaconess Medical Center, Boston; and Harvard Medical School, Boston

The microvasculature that supplies human cancers and mouse tumors is similar, and similarly heterogeneous, and includes at least six distinct blood vessel types. Some of these vessel types are present in all tumors thus far examined, whereas others are present in only some cancers and not in others. We have used an adenovirus expressing VEGF-A¹⁶⁴ (Ad-VEGF-A¹⁶⁴) to replicate each of these vessel types in immunodeficient nude mice and have found that anti-VEGF/VEGF receptor (VEGFR) therapy targets some vessel types, particularly early-formed mother vessels, but not other vessels that form later. Even so, the mechanism(s) by which anti-VEGF/VEGFR therapy damages Ad-VEGF-A¹⁶⁴-induced surrogates and tumor vessels is not entirely clear. Our findings in the Ad-VEGF-A¹⁶⁴ model are consistent with those of others demonstrating that similar types of therapy are effective in regressing rapidly growing mouse tumors but are less effective in treating many human cancers, as well as more slowly growing mouse tumors that have already acquired an established vasculature. If anti-vascular cancer therapy is to become more effective, new targets in addition to VEGF/VEGFR may be needed.
 
 

Session I: Mechanisms of Tumor Angiogenesis

From Isolation and Cloning of VEGF-A to FDA Approval of Multiple VEGF Inhibitors for Cancer and Intraocular Diseases
Napoleone Ferrara, MD, University of California, San Diego

Angiogenesis, the development of new blood vessels, is a fundamental pathophysiological process. Vascular endothelial growth factor (VEGF)-A was isolated and cloned in 1989 and during the following decade was shown to be a key regulator of blood vessel growth in embryonic development and in a variety of other physiological processes. Several VEGF inhibitors were reported to block tumor growth in numerous preclinical models, consistent with an important role of VEGF-A in tumor angiogenesis. Bevacizumab, a humanized anti- VEGF-A monoclonal antibody, and several other VEGF pathway inhibitors have been approved for the treatment of advanced tumors. Furthermore, blocking VEGF-A prevented vision loss and had a dramatic impact on the progression of neovascular age-related macular degeneration as well as other intraocular neovascular disorders.
 
We have recently been studying the mechanisms of resistance to anti-VEGF therapies in various tumor models. These studies indicate that multiple pro- angiogenic mechanisms may be implicated. We identified factors produced by myeloid cells and by fibroblasts. In recent studies, we identified Interleukin 17, a key product of Th17 Helper T cells as a factor mediating angiogenic escape and resistance to VEGF inhibitors. Efforts are ongoing to determine the translational and clinical significance of such findings.
 

Effects of Extracellular Matrix Binding on the Biology of VEGF and VEGF-Therapy
M. Luisa Iruela-Arispe, PhD, University of California, Los Angeles

Vascular Endothelial Growth Factor (VEGF) is essential for the formation, growth, and survival of blood vessels during both developmental and pathological conditions. The Vegf gene is alternatively spliced into multiple isoforms with varying affinities to the extracellular matrix (ECM), dictated by the presence of exon 7, commonly referred to as the heparin binding domain. Compared to soluble VEGF, ECM-associated VEGF provides distinct signaling outcomes, resulting in differences in vascular patterning. While the importance of matrix interactions to VEGF signaling has been well documented, much less is known about the specific components that bind to VEGF in vivo. Surprisingly, endothelial specific knockout of Ext1, an enzyme necessary for heparin sulfate biosynthesis, had no effect on blood vessel development or function. Here we present work on the identification of VEGF binding partners both in vitro and in vivo. Using surface plasmon resonance, we observed tight association of VEGF with various ECM molecules in vitro. We extended this work with mass spectrometry experiments to determine proteins that interact with VEGF in vivo. Using this unbiased approach, we discovered ECM components that have not yet been described as VEGF interactors. Finally, we explored the functional consequences of VEGF binding to the ECM and showed that this association significantly increases the half-life of VEGF by roughly 100-fold. Overall, our findings demonstrate that VEGF is capable of binding to a vast array of ECM molecules and that these associations can alter the bioavailability of VEGF.
 
Co-authors: Christiana Ruhrberg, PhD, Carmen M. Warren, PhD, Tom Chen, PhD, University of California, Los Angeles
 

The miR-126–VEGFR2 Axis Controls the Innate Response to Pathogen-associated Nucleic Acids
Judith Agudo, PhD, Icahn School of Medicine at Mount Sinai, New York

MicroRNA-126 (miR-126) is a microRNA predominately expressed by endothelial cells and controls angiogenesis. Unexpectedly, we found miR-126 is also required for the innate response to pathogen-associated nucleic acids, and that miR-126- deficient mice had increased susceptibility to pseudotyped-HIV infection. miRNA profiling and deep-sequencing indicated that miR-126 was highly and specifically expressed by plasmacytoid dendritic cells (pDCs), the major type I interferon producing cells. In the absence of miR-126, pDCs were unable to become properly activated or secrete high levels of interferons in response to stimulation, and underwent premature apoptosis, indicating that miR-126 controlled the survival and function of pDCs. Molecular analysis revealed that miR-126 regulated the expression of a number of innate response genes. Interestingly, in pDCs miR-126 also controlled the expression of VEGF-receptor 2 (VEGFR2), a receptor described to be predominately restricted to endothelial cells. However, gene expression analysis of more than 200 different immune cell populations indicated that VEGFR2 is also highly and exclusively expressed on pDCs amongst hematopoietic cells.
 
Deletion of the gene encoding VEGFR2 specifically in DCs resulted in impaired pDC development and reduced type I interferon production; indicating that VEGFR2 signaling is also important for pDC function, and suggesting that miR-126 regulates pDCs, at least in part, through its control of VEGFR2 expression. These studies indicate that the miR-126–VEGFR2 axis is an important regulator of the innate response to pathogen infection, which operates through multiscale control of pDCs, and assigns an important new function to the clinically targeted receptor VEGFR2 and its ligand VEGF-A.
 
Co-authors: Albert Ruzo, Navpreet Tung, Hélène Salmon, Marylène Leboeuf, Daigo Hashimoto, Christian Becker, Alessia Baccarini, Miriam Merad, and Brian D. Brown,Icahn School of Medicine at Mount Sinai, New York; Lee-Ann Garrett-Sinha, State University of New York at Buffalo.
 

Overcoming Resistance to VEGF-Blockade by Combinatorial Targeting of the Angiopoietin/Tie2 Axis
Alexander Scholz, PhD, University Cancer Center & Institute of Neurology, Goethe University Medical School, Frankfurt; and Laboratory of Immunology and Vascular Biology, Stanford University

Glioblastoma multiforme (GBM) is the most common and the most malignant form of brain tumor with a median overall survival of just about 14 months. Until now, standard treatment consists of combined radio- and chemotherapy. More recently, anti-angiogenic therapy targeting VEGF has been approved for the treatment of recurrent GBM. Even though anti-VEGF therapy has prolonged symptom-free survival, its role in first-line treatment is less clear. Furthermore, several pre-clinical studies suggest that resistance to anti-angiogenic therapy could be mediated by infiltration of pro-angiogenic myeloid cells.
 
To define new therapeutic options for anti-angiogenic therapy in GBM, we addressed the role of the Angiopoietin(Ang)/Tie signaling pathway. We found Ang-2 to be upregulated in vessels of malignant human brain tumors while being absent in normal brain. Furthermore, Ang-2 expression correlated with WHO grade and the amount of infiltrating myeloid cells. In a mouse model with endothelial cell-specific overexpression of Ang-2, we observed significantly increased numbers of myeloid cells in an orthotopic, syngeneic glioblastoma model, thus corroborating our findings in the human system. Interfering with Ang-2/Tie2 signaling by either blockade of Ang-2 or overactivation of the receptor lead to increased overall survival, vessel normalization, and reduced infiltration of myeloid cells in the Gl261 glioma model. Combining anti-Ang-2 and anti-VEGF treatment showed synergistic effects with regard to survival and myeloid cell infiltration.
 
Thus, targeting Ang-2 and VEGF in glioblastoma represents a promising approach for future therapy by targeting two major components of solid tumors, angiogenesis, and inflammation.
 
Co-authors: Sebastian Cremer¹, Patrick N. Harter¹, Michel Mittelbronn¹, Karl H. Plate¹, Yvonne Reiss¹, Paul van Slyke², and Daniel J. Dumont²

1. University Cancer Center & Institute of Neurology, Goethe University Medical School, Frankfurt
2. Sunnybrook Health Center, Toronto
 

Overcoming Resistance to VEGF-Blockade by Combinatorial Targeting of the Angiopoietin/Tie2 Axis
Alexander Scholz, PhD, University Cancer Center & Institute of Neurology, Goethe University Medical School, Frankfurt; and Laboratory of Immunology and Vascular Biology, Stanford University

Interacting Signaling Pathways in Tumor Angiogenesis: VEGF, Delta/Notch, and Angiopoetin/Tie-2
Gavin Thurston, PhD, Regeneron Pharmaceuticals

Blocking angiogenesis in cancer is recognized as a useful approach for the treat- ment of many solid tumors. Although angiogenesis is a complex process that requires the coordinated interplay of a variety of signaling pathways, much of the focus to date has been on targeting the VEGF pathway, which is an impor- tant upstream activator of tumor angiogenesis. However, not all preclinical and clinical tumors are equally responsive to VEGF inhibition. Therefore, more recent regimens are exploring VEGF blockers in combination with other anti-angiogenic agents or with cytotoxic chemotherapy. Additional angiogenic targets include Delta-like ligand 4 (Dll4) and angiopoietin-2 (Ang2), which are both expressed by endothelial cells in tumors and act as ligands for Notch or Tie2 receptors in the vasculature, respectively. In preclinical tumor models, combined blockade of VEGF plus these other angiogenic pathways can produce enhanced anti-vascular effects, such as greater decreases in tumor perfusion, compared to single agent treatments. These enhanced anti-vascular effects are associated with more potent tumor growth inhibition. Full benefit of combination anti-angiogenic therapy in the clinic will require better mechanistic understanding of which tumors vessels are resistant to individual treatments, as well as the development of biomarkers to identify resistant and susceptible tumor vessels.
 

Instructive Angiocrine Contribution of Vascular Niche to Tumor Progression
Shahin Rafii, MD, Weill Cornell Medical College

Abstract to come.
 
 

Session II: Immunemodulation in Angiogenesis

Mechanisms of Macrophage Promotion of Tumor Angiogenesis
Evelyn Aranda, PhD, Albert Einstein College of Medicine

Tumor-associated macrophages (TAM) play a major role in promoting malignancy in a wide variety of cancers. Our previous results in a mouse model of breast cancer caused by the expression of the Polyoma Middle T oncoprotein (PyMT) in the mammary epithelium, have demonstrated that the ablation of the macrophage colony stimulating factor-1  (CSF-1) resulted in a reduced rate of progression of tumors to malignancy and an inhibition of metastases. This was associated with a dramatic reduction in TAMs and inhibition of the angiogenic switch. Vascular endothelial growth factor (VEGF) and Wnts have been shown to be regulators of normal and pathological angiogenesis but little is known about the mechanisms that macrophages use to increase angiogenesis. Here we aimed to find the role of macrophages in tumoral blood vessel formation. Our results show that sorted TAMs from PyMT mice increase in vitro endothelial cell tube formation and hemoglobin content in Matrigel plugs. Furthermore, a significant reduction of hemoglobin content and infiltration of CD34-positive cells was found in plugs implanted in macrophage-deficient Csf1op null mice when compared to wild-type mice. Targeted gene ablation of VEGF in macrophages shows less angiogenesis in plugs while ablation of Porcupine (Porcn), a protein involved in the secretion of Wnts, increases the levels of VEGF in macrophages, tube formation, and infiltration of cells in Matrigel. These data show that macrophages are essential players in the formation of blood vessels and suggest a crosstalk of VEGF and Wnts in the regulation of tumoral angiogenesis.
 
Co-authors: Bin-Zhi Qian, Jeffrey W. Pollard, PhD, Albert Einstein College of Medicine
 

Novel Immunotherapeutic Approaches for Cancer Therapy
David F. McDermott, MD, Beth Israel Deaconess Medical Center, Boston

An improved understanding of the molecular mechanisms governing the host response to tumors has led to the identification of signaling pathways involved in limiting the anti-cancer immune response. One of the most critical pathways responsible for mediating tumor-induced immune suppression, or "immune checkpoint," is the programmed death-1 (PD-1) pathway. Many human solid tumors express programmed death-1 ligand 1 (PD-L1), and this is often, though not always, associated with a worse prognosis. Tumor-infiltrating lymphocytes (TILs) from patients with cancer typically express PD-1 and have impaired anti-tumor functionality. Proof-of-concept has come from several preclinical studies in which blockade of PD-1 or PD-L1 has enhanced T-cell function and tumor-cell lysis. Three monoclonal antibodies against PD-1, and one against PD-L1, have reported phase 1 data. All four agents have shown encouraging preliminary activity with good safety profiles, and additional data are awaited. Other factors in the tumor microenvironment have been shown to suppress tumor immune responses in preclinical models. For example, there is evidence that VEGF decreases dendritic cell (DC) function, and thus by extension impacts antigen presentation and T-cell activation. Concurrently restoring DC and T-cell functionality may improve the anti-tumor T-cell response and lead to improved clinical activity. Trials evaluating the combination of "immune checkpoint inhibitors" and angiogenesis inhibitors are ongoing. This talk will summarize the emerging data in this field and discuss the potential for collaboration between tumor immunotherapists and cancer "micro-environmentalists."
 

Tackling Resistance to Anti-angiogenic Therapy: a Tumor-Immune Cell Dialogue
Gabriele Bergers, PhD, Helen Diller Family Comprehensive Cancer Center and Brain Tumor Research Center, University of California, San Francisco

Bevacizumab (Avastin; Genentech/Roche), a humanized monoclonal antibody directed against VEGF-A, and several receptor tyrosine kinase inhibitors blocking the VEGF/VEGFR pathway (e.g., Sunitinib, sorafenib) have been FDA-approved for use in various tumor types in the US. Despite encouraging beneficial effects, patients inevitably develop resistance and frequently fail to demonstrate significantly better overall survival. Emerging evidence suggests that tumors bypass anti-angiogenic therapy while VEGF signaling remains inhibited through a variety of mechanisms by either re-establishing neovascularization, or by altering their behavior to propagate and progress without the need to satisfactorily reinitiate angiogenesis. Leveraging from the results obtained during the course of anti-angiogenic therapy in pancreatic and brain tumor mouse models and patient tumor samples, we found that relapsing tumors exposed cellular signals associated with resistance that were rather transiently regulated in response to contextual signals that tumor cells received from the immune cell compartment and responded to during the course of treatment. These data also revealed that the nature of tumor cells determines which resistance pathways will be facilitated by immune cells. Further, we observed that various innate immune populations exhibit an oscillatory behavior by which they substitute for each other to promote resistance by using the same central signaling pathway. These studies support the concept of reprogramming innate immune cell populations, rather than targeting them, as a sufficient and feasible approach to abrogate angiogenesis and restore T cell-mediated antitumor immunity.
 
Co-authors: Lee B. Rivera¹, Kan Lu¹, David Meyronet², and Judy Varner³.
 
1. Helen Diller Family Comprehensive Cancer Center, UCSF
2. Hospices Civils de Lyon
3. Moores Cancer Center, UCSD
 

Overcoming Tumor Resistance to VEGFA Signaling Blockade by Co-targeting ANG2-TIE2 Signaling and Macrophages
Michele De Palma, PhD, The Swiss Institute for Experimental Cancer Research (ISREC), School of Life Sciences, Swiss Federal Institute of Technology Lausanne (EPFL), Lausanne, Switzerland

Angiopoietin-2 (ANG2/ANGPT2) is a context-dependent TIE2 receptor agonist that enables angiogenesis in concert with vascular-endothelial growth factor (VEGF)-A. Indeed, ANG2 neutralization improves tumor angiogenesis and growth inhibition by VEGFA or VEGF receptor-2 (VEGFR2) blockade in several tumor models. However, it is unclear whether ANG2 may also sustain VEGFA-independent tumor angiogenesis. To address this question, we employed two tumor models—late-stage RIP1-Tag2 pancreatic neuroendocrine tumors (PNETs) and orthotopic MMTV-PyMT mammary carcinomas—which are transiently responsive but rapidly become resistant to VEGFR2 blockade. We found that ANG2 expression was higher in the PNETs than mammary tumors and was further increased by VEGFR2 inhibition. Combined ANG2/VEGFR2 blockade suppressed revascularization and progression in the vast majority of the PNETs, whereas it only had minor additive effects in the mammary tumors. Notably, a minority of the PNETs revascularized after ANG2/VEGFR2 blockade, thus showing adaptive resistance to the double anti-angiogenic treatment. We found increased tumor-associated macrophage (TAM) numbers in association with the resistant tumors, and therefore explored the utility of pharmacologically depleting TAMs in combination with the anti-angiogenic treatment. The results of these studies will be presented at the meeting.
 

Imaging Immune Cells in the Context of Tumor Angiogenesis
John S. Condeelis, PhD, Albert Einstein College of Medicine

Imaging at single cell resolution in vivo has been used to discover that carcinoma cells form migratory streams and intravasate when associated with macrophages and linear collagen-I fibers. The tumor cell-macrophage tropism results from EGF/ CSF1-paracrine chemotaxis causing the formation of cell pairs and streams that move unidirectionally to blood vessels [Roussos et al, Nat. Rev. Cancer, 2011]. This emergent behavior is cell autonomous and can be reconstituted in vitro using the three purified cell types and collagen-I fibers [Sharma et al, IntraVital, 2012]. Once the three cell types touch each other they spontaneously assemble a cell complex involving the direct contact between a tumor cell, endothelial cell, and macrophage [Roh-Johnson et al, Oncogene, 2013]. This triad is called the "Tumor MicroEnvironment of Metastasis (TMEM)" and its presence in human breast tumors has been used as a prognostic to predict distant metastatic recurrence [Robinson et al, Clin. Cancer Res., 2009; Rohan et al, JNCI, 2014 in Press].
 
TMEM functions as a dedicated site of tumor cell transendothelial migration for intravasation and extravasation [Roh-Johnson et al, Oncogene, 2013]. The gene expression profile of the tumor cells involved in TMEM function has been determined and has supplied prognostic markers for predicting tumor cell dissemination in breast tumors [Robinson et al, Clin. Cancer Res., 2009; Rohan et al, JNCI, 2014 in Press; Agarwal et al, Breast Cancer Res., 2012]. The micro-pharmacology of the signaling between cells within TMEM suggest that TMEM is similar in function to an angiogenic sprout suggesting that re-tasking of drugs directed toward inhibition of angiogenesis, but at much lower concentrations, may block tumor cell dissemination.
 
 

Session III: Vascular Normalization in Angiogenesis

Point – Counter Point Debate: Does Vascular Normalization Improve Delivery of Cancer Therapeutics?
Pro-Debater: Dan G. Duda, DMD, PhD, Massachusetts General Hospital and Harvard Medical School, Boston

Anti-angiogenic therapy has shown clear activity and improved survival benefit for certain tumor types. However, an incomplete understanding of the mechanisms of action of anti-angiogenic agents has hindered optimization and broader application of this new therapeutic modality. In particular, the impact of anti-angiogenic therapy on tumor blood flow and drug and oxygen delivery (i.e., the role of vessel pruning versus normalization) remains controversial. This controversy has become critical as multiple phase III trials of anti-VEGF agents combined with cytotoxics failed to show overall survival benefit in newly diagnosed glioblastoma (nGBM) patients and several other cancers. Using MRI techniques and blood biomarkers in prospective phase II clinical trials of cediranib with chemoradiation vs. chemoradiation alone in nGBM patients, we shed light on mechanisms of nGBM response to cediranib, a pan-VEGF receptor tyrosine kinase inhibitor. We demonstrate that improved perfu- sion occurs only in a subset of patients in cediranib-containing regimens, and is associated with improved overall survival in these nGBM patients. Moreover, an increase in perfusion is associated with improved tumor oxygenation status as well as with pharmacodynamic biomarkers, such as changes in plasma placenta growth factor and sVEGFR2. In conclusion, tumor perfusion changes after anti-angiogenic therapy may distinguish patients who benefit from those who do not early in the course of this expensive and potentially toxic form of therapy, and these results may provide new insights into the selection of glioblastoma patients most likely to benefit from anti-VEGF treatments.
 
Co-authors: Kyrre E. Emblem, Marek Ancukiewicz, Christine Lu-Emerson, Elizabeth R. Gerstner, Tracy T. Batchelor, A. Greg Sorensen, and Rakesh K. Jain, Massachusetts General Hospital and Harvard Medical School, Boston.
 
Contra-Debater: Elisabeth G.E. de Vries, MD, PhD, The University Medical Center Groningen, the Netherlands.
 

---

Abstracts, Day 2: Friday, June 20, 2014

Editor's Guide to Writing and Publishing Your Paper
Brooke Grindlinger, PhD, The New York Academy of Sciences

Publishing is critical to the scientific profession, yet training and guidance on the topic is very limited. Brooke Grindlinger, PhD, former Science Editor at The Journal of Clinical Investigation, will discuss what makes a good paper, strategies for selecting the appropriate journal, the review process, and how to navigate resubmissions.
 
 

Session IV: Challenges to Anti-angiogenic Therapy — Learning from Successes and Failures

Synergy of Chemotherapy and Anti-VEGF Treatment (Mechanisms of Action of Targeting VEGF)
Lee M. Ellis, MD, The University of Texas MD Anderson Cancer Center, Houston

Targeting VEGF has been a successful anti-neoplastic approach in numerous cancers. However, the magnitude of benefit has been less than expected. A better understanding of the mechanism(s) of action of this class of drugs would help provide insight on how to improve efficacy, and biomarker identification, but despite valiant attempts, we still do not fully understand how these drugs work.
 
The mechanisms of action of VEGF targeted therapies are diverse and probably cannot be attributed primarily to any single effect. This is more complicated by the fact that multi-kinase inhibitors affect numerous pathways in addition to VEGF signaling. With rare exception, the addition of VEGF-Receptor (VEGFR) targeted tyrosine kinase inhibitors (TKIs) to chemotherapy has not led to meaningful improvements in survival in patients with solid malignancies. In fact, in recent meta- analysis of VEGFR TKIs and chemotherapy (Funakoshi et al, Cancer Treatment Reviews, 2014), the relative risk of a fatal adverse event was increased by nearly 50%. Antibodies to VEGF (Bevacizumab) and more recently VEGFR2 (Ramucirumab) have led to improvements in survival in some, but not all clinical trials where these combinations have been studied. The same holds true for VEGF-Trap (Aflibercept). This presentation will discuss various mechanisms of action of VEGF targeted therapies.
 

Optimal Strategies for Resistance Avoidance and Long-Term Treatment with Anti-angiogenic Therapy
Axel Grothey, MD, Mayo Clinic College of Medicine

Anti-angiogenic therapy, in particular, anti-VEGF (vascular endothelial growth factor) therapy has become a component of standard treatment algorithms in various malignancies over the last decade. In contrast to cytotoxic chemotherapy or agents targeting tumor cells directly, anti-VEGF agents presumably exert their anti-tumor effect by modulating the tumor-host interaction or tumor microenvironment and affecting the host-provided tumor vasculature. The cellular targets of anti-angiogenic therapy are genetically stable since they are routinely not derived from tumor cells. Thus, mechanisms of resistance to anti-angiogenic agents could differ from the ones associated with directly tumor-targeting agents. In fact, it has been questioned if resistance to anti-angiogenic agents exists at all. Recent data of a phase III trial in metastatic colorectal cancer confirmed that continuing the anti-VEGF-A antibody bevacizumab beyond tumor progression from first- to second-line therapy is associated with improved survival, thereby challenging the established concept of tumor resistance to anti-angiogenic therapy. A large body of evidence has been generated over the last years demonstrating that a prolonged duration of anti-VEGF therapy is important for optimizing outcomes. Research on mechanisms of secondary resistance has suggested that upregulation of alternate pro-angiogenic factors beyond VEGF could play an important role. A longitudinal analysis of these factors could provide valuable insights regarding treatment approaches to counteract these mechanisms of resistance. It is conceivable that additional anti-angiogenic agents will have to be added to protracted and continued anti-VEGF therapy over time to maximize the duration of tumor control.
 

Vessel Architectural Imaging Reveals Early Vascular Response to Anti-angiogenic Therapy
Kyrre E. Emblem, PhD, Oslo University Hospital; and Massachusetts General Hospital and Harvard Medical School, Boston

Randomized phase III trials of bevacizumab with standard chemoradiation in glioblastoma (RTOG 0825 and AVAglio) indicate that first-line anti-VEGF treatment does not extend survival in unselected patient populations. Because anti-angiogenic drugs reduce vessel permeability, conventional MRI alone may not be adequate for monitoring treatment responses. Here, we address the lack of appropriate imaging biomarkers in clinical trials in brain tumors by going beyond conventional MRI. Our initial experience in 70 glioblastoma patients enrolled in two phase II trials (NCT00305656, NCT00662506) of cediranib, an oral pan-VEGFR inhibitor, with- and without chemoradiation, shows the promise for Vessel Architectural Imaging (VAI) [Emblem et al, Nature Medicine, 2013].
 
VAI combines perfusion MRI from small (<10µm) and large vessels in one exam for in vivo assessment of vessel caliber, vessel type (arterioles/capillaries/venules), and relative oxygen saturation levels (ΔSO2). As early as day one of therapy, 21/40 newly diagnosed glioblastoma patients treated with cediranib+chemoradiation [Batchelor et al, PNAS, 2013] and 10/30 patients with recurrent glioblastomas treated with cediranib only [Emblem et al, Nature Medicine, 2013] responded to therapy by increased perfusion, normalization of arterio-venous ratios and ΔSO2 levels, and reduction of abnormal vessel calibers. Compared to non-responding patients, newly diagnosed glioblastoma responders had a survival of 696 versus 381 days (+10 months) and recurrent glioblastoma responders had 341 versus 146 days (+6 months). These effects were not observed in newly diagnosed glioblastoma patients receiving chemoradiation only (NCT00756106).
 
Collectively, our results indicate that vascular normalization may be a mechanism of survival benefit in selected glioblastoma patients, and that VAI enables early identification of patients most likely to benefit from anti-angiogenic therapy.
 
Co-authors: Elizabeth R. Gerstner², Tracy T. Batchelor², Dan G. Duda², Bruce Rosen², Rakesh K. Jain², Marco C. Pinho^2,3, A. Greg Sorensen^2,4, Patrick Y. Wen⁵.
 
1. Oslo University Hospital
2. Massachusetts General Hospital and Harvard Medical School, Boston
3. University of Texas Southwestern Medical Center, Dallas
4. Siemens Healthcare Health Services, Malvern, Pennsylvania
5. Dana-Farber Cancer Center and Harvard Medical School, Boston
 

Population Pharmacodynamics: Mechanism-based Modeling of the Angiogenesis Receptor Kinome in Cancer
Feilim Mac Gabhann, PhD, Johns Hopkins University, Baltimore

We have previously built computational pharmacodynamic models of anti-angiogenic therapies in human cancer. These models incorporate detailed protein-protein interaction networks to simulate the complex dynamics of the VEGF family and its receptors. By integrating this molecular detail into whole-body simulations of tumor and endothelial cells in multiple tissues, we can evaluate many different therapeutic approaches—different drugs, doses, schedules, and routes of administration. Our models make predictions of the dynamics of receptor tyrosine kinase activity and of key blood-borne biomarkers following therapeutic intervention. These predictions can and have been validated against clinical experimental data.
 
We have previously used these models to simulate treatment of 'average' tumors of particular types, and compared the results to ensemble experimental data. However, tumors are highly diverse, exhibiting significant variability of gene expression between individuals and heterogeneity within an individual. In our current work, we incorporate patient-specific gene expression data sets into multiple instances of our mechanistic pharmacodynamic models. We simulate the impact of multiple drug and dosing regimens on a population of in silico individuals ('virtual patients') based on real patient data, allowing us to run thousands of 'virtual clinical trials.' We use these patient-specific computational models to predict the efficacy of multiple anti- angiogenic drugs—anti-ligand antibodies, receptor-blocking peptides, and tyrosine kinase inhibitors—and compare the predictions to the outcomes of clinical trials. We also compare the predicted responsiveness of primary tumors to those at metastatic sites to design optimal therapeutic approaches.
 
Co-author: R. Joseph Bender, MSc, Johns Hopkins University, Baltimore.
 

Understanding VEGFR2 Targeting for Cancer Therapy
Laura E. Benjamin, PhD, ImClone Systems, a wholly owned subsidiary of Eli Lilly

Angiogenesis is one of the hallmarks of cancer with associated therapeutic approaches proven to extend life for patients with cancer. Of the anti-angiogenic approaches approved by the U.S. Food and Drug Administration (FDA), only Ramucirumab selectively blocks Vascular Endothelial Growth Factor Receptor-2 (VEGFR2). Ramucirumab is a monocolonal antibody to the human VEGFR2 recently approved as a single-agent treatment for patients with advanced gastric cancer after prior chemotherapy. This was the first regulatory approval for CYRAMZA (Ramucirumab), which has also shown positive data in lung cancer and as a combination therapy with paclitaxel in gastric cancer. It is also being studied in several other types of cancer. This presentation will review some of the clinical data associated with Ramucirumab as well as preclinical data into unique aspects of selective VEGFR2 blockade. Finally, preclinical research will be presented that furthers our understanding of tumor vessels heterogeneity. This line of research may help inform future clinical opportunities for anti-angiogenesis strategies.
 

Anti-angiogenic Therapy: Impact on Invasion and Metastatic Disease Outcomes
Robert S. Kerbel, PhD, Sunnybrook Research Institute, University of Toronto

In 2009, two preclinical papers published in Cancer Cell reported results showing that treatment of mouse or human tumors in vivo with single agent anti-angiogenic drugs that target the VEGF pathway could promote local invasion or metastasis. A large number of subsequent preclinical studies—though not all—have reported similar results. Nevertheless, such preclinical findings remain highly controversial primarily because there is very little clinical evidence to substantiate them. In this regard, there has been much confusion regarding how the results of the original two studies have been interpreted and evaluated: 1) one of original Cancer Cell studies did not show a decrease in survival despite the increase in metastasis—indeed there was a modest increase in survival times (Paez-Ribes et al, Cancer Cell, 2009), a finding which may help explain why the positive clinical benefits of anti-angiogenic therapies are usually modest, and often do not lead to an increase in overall survival despite a benefit in progression free survival; 2) clinical study results, especially with drugs such as bevacizumab, involve concurrent upfront treatment with chemotherapy, thus raising the possibility that the chemotherapy treatment blocks any potential pro-invasive effect of the anti-angiogenic drug. If so, this suggests that it is not only that anti-angiogenic drugs improve the efficacy of chemotherapy, but also the reverse—it is a 'two-way street'; 3) one of the Cancer Cell papers (Ebos et al, Cancer Cell, 2009) did not deal with the impact of the therapy (using sunitinib) on established metastatic disease (of breast cancer), but rather on early stage microscopic disease, i.e., it was an adjuvant therapy study. Subsequent studies did not show a decrease in survival when mice with established metastatic disease were treated with sunitinib (E Guerin et al., Cancer Res., 2013). As such, our results (Ebos et al, Cancer Cell, 2009) represented a prospective cautionary note that anti-angiogenic drugs might not be effective in the adjuvant setting. Four subsequent phase III trials have shown this to be the case. Overall, these considerations may help explain some of the clinical outcomes of anti-angiogenic drug-based therapies, and the benefits of improved preclinical models to evaluate the impact of anti-angiogenic drugs on micro- and macroscopic disease as well as local invasion.
 

Point – Counter Point Debate: Does Anti-angiogenic Therapy Induce Invasion and Metastasis?
Pro-Debater: Oriol Casanovas, PhD, Translational Research Laboratory, Catalan Institute of Oncology (IDIBELL), Barcelona

Drugs targeting VEGF-mediated angiogenesis are well established for the treatment of several tumor types. Unfortunately, several clinical trials report a failure in long- lasting effects of anti-angiogenic agents with modest modifications of patients' long-term survival as consequence of tumor adaptation to the therapy. Indeed, several mechanisms of tumor adaptation and resistance to anti-angiogenic therapy have been described to limit their long-lasting effects in blocking vascularization, tumor growth, and dissemination. Furthermore, contrary to the expectations, a rapidly growing list of preclinical studies, which still need to be fully validated in the clinic, report promotion rather than inhibition of the tumor invasive/metastatic behavior after anti-angiogenic treatments. In this debate, the intricate relationship between anti-angiogenic agents and metastasis will be discussed, including recent advances in understanding the mechanisms of tumor adaptation to these therapies. Finally, we will discuss the studies and observations of the occurrence of this insidious event in the clinical setting.
 
Contra-Debater: Napoleone Ferrara, MD, University of California, San Diego.
 
 

Session V: Biomarkers

Genetic and Circulating Markers Associated with Response to Anti-VEGF Therapy
John V. Heymach, MD, PhD, MD Anderson Cancer Center

Abstract to come.
 

Novel Biomarkers that are of Prognostic Value and can Predict Response or Resistance to Anti-angiogenic Therapy
Thomas Powles, MBBS, MRCP, MD, Barts Cancer Institute, London

VEGF and mammalian target of rapamycin (mTOR) targeted therapy are both established treatments in metastatic clear cell renal cancer. These agents have had a profound effect on median survival for the unselected population, but benefits for individual patients remains variable. Biomarker analysis of the targets of these therapies (VEGFR2 and mTOR/PI3K) within cancer tissue has not resulted in predictive biomarkers. The reasons for this remain unclear but factors such as method of tissue collection and tumor heterogeneity may be responsible. Further research into circulating chemokines and cytokines, DNA/RNA/protein analysis form frozen tissue, functional imaging, and single nucleotide polymorphisms from germ line material have given some insight into mechanisms of response and resistance to VEGF targeted therapy, but predictive biomarkers remain elusive. Sequential tissue investigation of dynamic changes to biomarkers may unlock the key to resistance but studies remain sparse. There is still a major lack of research into mechanisms resistance to mTOR inhibitors, although mutations to key genes may be important. Circulating tumor free DNA is an area of intense research for the future.
 

Direct Evidence of Targeted Inhibition by Anti-angiogenic Agents from a Wound-healing Model in Patients with Advance Solid Tumors
Jingquan Jia, MD, PhD, Duke University Medical Center, Durham, North Carolina

In early clinical testing, most novel targeted anti-cancer therapies have limited toxicities and limited efficacy, which complicates dose and schedule selection for these agents. Confirmation of target inhibition is critical for rational drug development; however, confirmation by repeated tumor biopsies is often impractical and peripheral blood mononuclear cells and normal skin are often inadequate surrogates for tumor tissue. Based upon the similarities of tumor and wound stroma, we have developed a clinical dermal granulation tissue model for the pharmacodyanamic (PD) evaluation of novel targeted therapies. Granulation tissue biopsies are less expensive and more readily accessible than tumor biopsies. Granulation tissue was assessed at baseline and on-treatment for four targeted agents—bevacizumab, everolimus, erlotinib, and panitumumab—across three different clinical trials. Total and phosphorylated levels of VEGFR2, S6RP, and EGFR, the respective targets of the above-mentioned drugs, were evaluated using ELISA- based methodologies. Our results demonstrated that total and phosphorylated levels of VEGFR2, S6RP, and EGFR were reduced by bevacizumab, everolimus, and panitumumab or erlotinib, respectively. Interestingly, stronger inhibition of EGFR was seen with panitumumab compared to erlotinib. Molecular analyses of dermal granulation tissue can be used as a convenient and quantitative PD biomarker platform for multiple classes of targeted anti-cancer therapies. These features make granulation tissue biopsies a practical and reliable tissue source for further exploration of the correlation between individual PD responses and clinical responses to targeted therapies.
 
Co-authors: Andrew Dellinger¹, Eric S. Weiss², Anuradha Bulusu¹, Haiyan Li³, Leigh Howard¹, Neal Kaplan¹, Herbert Pang⁴, Herbert I. Hurwitz¹, and Andrew B. Nixon¹.
 
1. Duke University Medical Center, Durham, North Carolina
2. Temple University School of Medicine, Philadelphia
3. Department of Medicine, Center for Musculoskeletal Research, University of Rochester Medical Center
4. School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong
 

Metabolic PET Imaging Approaches as Early Biomarkers of Cardiac Toxicity Following Treatment with the Tyrosine Kinase Inhibitor Sunitinib
Ian S. Miller, PhD, Royal College of Surgeons in Ireland, Dublin

Sunitinib is a small molecule Receptor Tyrosine Kinase Inhibitor employed in the treatment of several cancer indications. Associated adverse effects include hypertension and congestive heart failure. We have sought to 'reverse translate' clinically relevant sunitinib treatment protocols in rodents in order to investigate novel imaging safety biomarkers.
 
Mice or rats were treated with sunitinib [40 or 20 mg/kg respectively, p.o.]. Blood pressure was measured weekly. Cardiovascular function was assessed using echocardiography. Positron emission tomography (PET) tracers ¹⁸F-FDG (Fluoro- deoxyglucose), ¹⁸F-FTHA (Fluoro-6-thia-heptadecanoic acid), and ¹¹C-acetate (myocardial perfusion) were used to interrogate changes in myocardial metabolism and blood flow. Immunohistochemistry, histology, and electron microscopy were carried out on rodent cardiac tissue to interrogate sub-cellular off-target effects.
 
Treated animals displayed a significant increase in blood pressure after 5 days of treatment, a decline in left ventricular perfusion, indicated by lower ¹¹C-acetate PET signal, and perturbation in cardiac function over the course of the study.
 
Importantly, changes in FTHA /FDG-PET signal and accumulation of lipid vesicles in cardiomyocytes indicated metabolic adjustments. Proteomic analyses of cardiac tissue from treated animals indicated major metabolic aberrations (mitochondrial dysfunction, increase in glycolytic enzymes, and increase in lipid storage). Metabolic PET imaging approaches may represent early cardiotoxicity biomarkers in patients treated with sunitinib.
 
Co-authors: Alice C. O'Farrell¹, Rhys Evans¹, Johanna M.U. Silvola², Emer Conroy³, Monika Jarzabek^1,5, David W. Murray^1,4, Liam Shiels¹, Marina Alamanou4, Sue Hector¹, Antti Saraste², Heidi Liljenbäck², Sami Mahrus⁵, Axel Ducret⁶, P. Cutler⁶, William Gallagher^3,4, Maurice Cary⁷, Juhani Knuuti², Anne Roivainen², and Annette T. Byrne¹.
 
1. Royal College of Surgeons in Ireland, Dublin
2. Turku PET Centre, Turku, Finland
3. University College Dublin, Dublin
4. Oncomark Ltd., Dublin
5. Genentech, San Francisco
6. Roche, Basel, Switzerland
7. Pathology Experts GmbH, Therwil, Switzerland
 
Acknowledgments: AngioTox is funded by a European Commission FP7 Industry Academia Pathways and Partnerships Marie Curie Award. The authors would like to thank M2i Limited Blackrock clinic for supplying 18F-FDG and the EuroBioimaging initiative and BACR for work carried out at the Turku PET Centre, Finland.
 

Identification and Analysis of In Vivo VEGF Downstream Markers Link VEGF Pathway Activity with Efficacy of Anti-VEGF Therapies
Carlos Bais, PhD, Genentech, Inc.

Purpose: The aim of this study was to identify conserved pharmacodynamic and potential predictive biomarkers of response to anti-VEGF therapy using gene expression profiling in preclinical tumor models and in patients.
 
Experimental Design: Surrogate markers of VEGF inhibition [VEGF-dependent genes or VEGF-dependent vasculature (VDV)] were identified by profiling gene expression changes induced in response to VEGF blockade in preclinical tumor models and in human biopsies from patients treated with anti-VEGF monoclonal antibodies. The potential value of VDV genes as candidate predictive biomarkers was tested by correlating high or low VDV gene expression levels in pretreatment clinical samples with the subsequent clinical efficacy of bevacizumab (anti-VEGF)-containing therapy.
 
Results: We show that VDV genes, including direct and more distal VEGF downstream endothelial targets, enable detection of VEGF signaling inhibition in mouse tumor models and human tumor biopsies. Retrospective analyses of clinical trial data indicate that patients with higher VDV expression in pretreatment tumor samples exhibited improved clinical outcome when treated with bevacizumab- containing therapies.
 
Conclusions: In this work, we identified surrogate markers (VDV genes) for in vivo VEGF signaling in tumors and showed clinical data supporting a correlation between pretreatment VEGF bioactivity and the subsequent efficacy of anti-VEGF therapy. We propose that VDV genes are candidate biomarkers with the potential to aid the selection of novel indications as well as patients likely to respond to anti- VEGF therapy. The data presented here define a diagnostic biomarker hypothesis based on translational research that warrants further evaluation in additional retrospective and prospective trials.
 
Co-authors: Matthew J. Brauer¹, Guanglei Zhuang¹, Maike Schmidt¹, Jenny Yao¹, Xiumin Wu1¹, Joshua S. Kaminker¹, Stefanie S. Jurinka¹, Ganesh Kolumam¹, Alicia S. Chung¹, Adrian Jubb¹, Zora Modrusan¹, Tomoko Ozawa², C. David James², Heidi Phillips¹, Benjamin Haley¹, Rachel N.W. Tam¹, Anne C. Clermont¹, Jason H. Cheng¹, Sherry X. Yang⁴, Sandra M. Swain⁵, Daniel Chen¹, Stefan J. Scherer¹, Hartmut Koeppen¹, Ru-Fang Yeh¹, Peng Yue¹, Jean-Philippe Stephan¹, Priti Hegde¹, Napoleone Ferrara¹, and Mallika Singh³.
 
1. Genentech, Inc., South San Francisco
2. University of California at San Francisco
3. Novartis Institutes for Biomedical Research, Emeryville, California
4. National Cancer Institute, NIH, Bethesda, Maryland
5. Washington Cancer Institute, Medstar Washington Hospital, Washington, DC
 

Point – Counter Point Debate: Hypertension is a Valid Biomarker for Anti-angiogenic Therapy
Pro-Debater: Brian I. Rini, MD, FACP, Cleveland Clinic Taussig Cancer Institute; and Glickman Urological Institute, Cleveland

Elevation of blood pressure is a commonly-encountered sequela of therapy targeting VEGF. Several clinical trials have examined the association of treatment- induced elevation in blood pressure and clinical outcome. Notably, these analyses—conducted across a diverse array of solid tumors including breast cancer, lung cancer, and renal cell carcinoma, and with various VEGF-targeting therapies including bevacizumab and VEGF receptor inhibitors—have consistently shown an association. Specifically, patients who experience elevations of blood pressure, defined as either above a given systolic (e.g. 140 mmHg) or diastolic (e.g. 90 mmHg) threshold, or defined as a change in blood pressure versus baseline, have improved clinical outcome measured by objective response rate, progression-free survival, and/or overall survival. These analyses continue to show significance when subjected to landmark analysis (thus correcting for time on therapy) and in multivariate analyses accounting for prognostic disease characteristics. Hypertension analyzed in the context of other toxicities continues to be significant, and thus this phenomenon does not merely reflect achievement of adequate drug levels. The underlying biology accounting for this association and how to incorporate this into clinical practice, however, remain elusive.
 
Contra-Debater: Herbert I. Hurwitz, MD, Duke University School of Medicine, Durham, North Carolina.