Targeting Angiogenesis: The 2011 Dr. Paul Janssen Award Symposium

Journal Articles

Napoleone Ferrara

Ferrara N, Carver-Moore K, Chen H, et al. Heterozygous embryonic lethality induced by targeted inactivation of the VEGF gene. Nature 1996;380(6573):439-442.

Ferrara N, Chen H, Davis-Smyth T, et al. Vascular endothelial growth factor is essential for corpus luteum angiogenesis. Nat. Med. 1998;4(3):336-340.

Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N. Engl. J. Med. 2004;350(23):2335-2342.

Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N. Vascular endothelial growth factor is a secreted angiogenic mitogen. Science 1989;246(4935):1306-1309.

Mesiano S, Ferrara N, Jaffe RB. Role of vascular endothelial growth factor in ovarian cancer: inhibition of ascites formation by immunoneutralization. Am. J. Pathol. 1998;153(4):1249-1256.

Shojaei F, Wu X, Zhong C, et al. Bv8 regulates myeloid-cell-dependent tumour angiogenesis. Nature 2007;450(7171):825-831.

Shojaei F, Wu X, Qu X, et al. G-CSF-initiated myeloid cell mobilization and angiogenesis mediate tumor refractoriness to anti-VEGF therapy in mouse models. Proc. Natl. Acad. Sci. USA 2009;106(16):6742-6747.

Anthony Adamis

Gragoudas ES, Adamis AP, Cunningham ET, Feinsod M, Guyer DR. Pegaptanib for neovascular age-related macular degeneration. N. Engl. J. Med. 2004;351(27):2805-2816.

Rosenfeld PJ, Brown DM, Heier JS, et al. Ranibizumab for neovascular age-related macular degeneration. N. Engl. J. Med. 2006;355(14):1419-1431.

Luisa Iruela-Arispe

Chen TT, Luque A, Lee S, et al. Anchorage of VEGF to the extracellular matrix conveys differential signaling responses to endothelial cells. J. Cell Biol. 2010;188(4):595-609.

Lee S, Jilani SM, Nikolova GV, Carpizo D, Iruela-Arispe ML. Processing of VEGF-A by matrix metalloproteinases regulates bioavailability and vascular patterning in tumors. J. Cell Biol. 2005;169(4):681-691.

Dan G. Duda

di Tomaso E, Snuderl M, Kamoun WS, et al. Glioblastoma recurrence after cediranib therapy in patients: lack of "rebound" revascularization as mode of escape. Cancer Res. 2011;71(1):19-28.

Duda DG, Ancukiewicz M, Jain RK. Biomarkers of antiangiogenic therapy: how do we move from candidate biomarkers to valid biomarkers? J. Clin. Oncol. 2010;28(2):183-185.

Duda DG, Kozin SV, Kirkpatrick ND, et al. CXCL12 (SDF1α)-CXCR4/CXCR7 pathway inhibition: an emerging sensitizer for anticancer therapies? Clin. Cancer Res. 2011;17(8):2074-2080.

Jain RK, Duda DG, Willett CG, et al. Biomarkers of response and resistance to antiangiogenic therapy. Nat. Rev. Clin. Oncol. 2009;6(6):327-338.

Willett CG, Duda DG, di Tomaso E, et al. Efficacy, safety, and biomarkers of neoadjuvant bevacizumab, radiation therapy, and fluorouracil in rectal cancer: a multidisciplinary phase II study. J. Clin. Oncol. 2009;27(18):3020-3026.

Zhu AX, Duda DG, Sahani DV, Jain RK. HCC and angiogenesis: possible targets and future directions. Nat. Rev. Clin. Oncol. 2011;8(5):292-301.

Sponsors

This symposium is made possible with support from

Speaker:
Napoleone Ferrara, Genentech, Inc.

Highlights

• Vascular endothelial growth factor (VEGF) regulates angiogenesis in normal development and in diseases such as cancer.
• Therapies that target VEGF have been used to treat cancer and age-related macular degeneration successfully.
• Research is currently underway to better understand anti-VEGF resistance and to find targets for new therapies.

Angiogenesis, VEGF, and cancer

Angiogenesis, or the formation of new blood vessels, is essential both to the development of organisms and to the mechanisms that underlie diseases such as cancer and age-related macular degeneration. A series of signals and proteins are involved in angiogenesis, however in the last 30 years, it has been shown that vascular endothelial growth factor (VEGF) is particularly central to the process. After several initial failures, researchers have developed a number of drugs that target VEGF and that are now approved by the FDA to treat various diseases.

Many of the recent studies of angiogenesis are directed at finding cancer treatments that target the formation of new blood vessels. Solid tumors rely on new blood vessels for nutrients as they grow and metastasize. The production of pathogenic forms of VEGF is stimulated by a variety of oncogenic pathways in a cell including those that respond to low oxygen. Many tumor types express VEGF mRNA. Effective therapies that shut down the tumors' supply chain by limiting angiogenesis could treat a wide range of cancers.

A brief history of VEGF

In his award address Napoleone Ferrara provided a historical perspective on the field of angiogenesis and described efforts to target this process to treat disease. He also examined research that aims to circumvent resistance to therapies that target angiogenesis.

A first look at tumor angiogenesis.

In 1939, researchers took some of the earliest images that depict angiogenesis in an animal model. Just a few days after a tumor was transplanted into the animal, a burst of blood vessels formed around the tumor site, suggesting the presence of a factor that promoted the vessels' formation and growth. Other researchers confirmed these observations and hypothesized that angiogenesis was a critical part of tumorigenesis. Judah Folkman first proposed angiogenesis as a target in cancer treatment in 1971. If you starve a tumor, he proposed, it can't grow or spread.

Researchers in other areas were also interested in angiogenesis. In 1948, based on his observations of patients, opthalmologist Isaac Michaelson proposed that a factor related to the one described in the transplanted tumor model might be present in the diseased retina, spurring the formation of new blood vessels.

On his personal trajectory, Ferrara started by studying reproductive endocrinology and began thinking about angiogenesis because of the complex web of blood vessels within the pituitary gland. At the time, in the 1980s, the process of fully characterizing new growth factors was arduous, taking years and sometimes a decade or more to purify, sequence, and clone a protein, and then to study its function.

VEGF promotes angiogenesis.

In 1983 Harvard researchers reported vascular permeability factor (VPF), a protein factor from a tumor cell line that made blood vessels more permeable, but the protein wasn't fully sequenced or cloned. Ferrara and his colleagues at Genentech isolated VEGF from bovine pituitary cells and subsequently cloned the protein. In that same year, a group from Monsanto purified and sequenced VPF and confirmed that its sequence matched that of VEGF. This evidence linked angiogenesis with the increased permeability of vessel walls.

VEGF form and function

The VEGF protein occurs in up to nine different isoforms, each of which splices together a particular combination of eight different exons of the VEGF gene. The primary changes in activity of these isoforms result from differential binding to heparin: longer isoforms, which have 206 and 189 amino acids, respectively, bind more strongly to heparin and thus remain tightly bound to the surfaces of cells. A shorter isoform, VEGF-121, which is formed when proteases cleave the C-terminal portion of the protein, can diffuse away from cells. Tumor proteases, such as plasmin and matrix metalloproteases, can cleave the heparin-binding regions from VEGF to produce the more soluble form of the protein. These soluble isoforms are more likely to have pathogenic effects. The difference in binding strengths appears to produce angiogenic gradients, a range of VEGF proteins with varying levels of solubility, Ferrara said.

Initially two different VEGF receptors were identified, now known as VEGF receptor 1 (VEGFR1) and VEGF receptor 2 (VEGFR2). It is now clear that most of the known physiological functions of VEGF result from binding to VEGFR2. VEGF is essential for angiogenesis during development. Even small modifications in VEGF expression are lethal to embryos, and VEGF is essential for angiogenesis in the corpus luteum as well.

Anti-VEGF therapies

Ferrara and his colleagues developed a mouse antibody to VEGF and showed that this antibody suppressed tumor growth in vivo. In experiments with an ovarian cancer model, they discovered that if they stopped anti-VEGF therapy, tumors would develop new vessels (revascularize) and would restart their growth.

In the meantime a number of other groups had been working on anti-VEGF therapies. In the late 1990s, several therapies looked promising in preclinical studies, but failed to show benefit in clinical trials. Ferrara and his colleagues developed a humanized version of their antibody, and despite relatively modest preclinical results compared with other drug candidates, it proved successful in the clinic. After a series of clinical trials, in 2004 bevacizumab (Avastin) became the first anti-angiogenesis drug approved by the FDA. A modified version of the antibody, ranibizumab (Lucentis) is approved for the treatment of intra-ocular diseases such as age-related macular degeneration.

Countering resistance to anti-VEGF therapies

Because anti-angiogenesis therapies target the area around the tumor (the tumor niche) rather than the cancer cells themselves, researchers initially thought that drugs that targeted VEGF would not lead to resistance. These cells within the tumor niche are much less likely to mutate rapidly and therefore less likely to develop resistance to therapies. However, some cancers do not respond to VEGF therapies, while others that initially respond to anti-VEGF therapies later become resistant to these therapies. The resistance could come from an almost infinite array of sources, Ferrara said. Tumors could produce other factors that prompt angiogenesis or could find mechanisms to circumvent their dependence on angiogenesis. In addition, the tumor cell niche could recruit bone marrow cells which carry a payload of molecules that promote angiogenesis.

Ferrara and his colleagues have identified two mouse tumor cell lines resistant to anti-VEGF therapies and have found that these cells recruit large numbers of bone marrow cells while cells sensitive to anti-VEGF therapies did not. The hematopoietic growth factor, G-CSF, boosts expression of a protein Bv8/Pk2 in these bone marrow cells. Bv8 promotes angiogenesis and shares 62% sequence identity with VEGF, including several functional cysteine residues, indicating that it may fulfill VEGF's angiogenesis functions when VEGF activity is blocked. This work highlights the variety of cell types and feedback loops that can promote angiogenesis in tumors.

Ferrara's laboratory is exploring other avenues of angiogenesis promotion, beyond the VEGF and G-CSF mechanisms. The group has profiled the proteins that are expressed in larger quantities in fibroblasts from tumors that do not respond to treatment compared to responsive tumors. One of these proteins was PDGF-C, a growth factor that is a key component in the signaling of the platelet-derived growth factor receptor-alpha (PDGFRα) pathway. The pathway is required to recruit the network of cells that promote angiogenesis. PDGF-C could serve as a critical signal that links cancer cells with niche fibroblasts, and it could become a target for new therapies.

Speaker:
Anthony Adamis, Genentech, Inc.

Highlights

• Anti-angiogenesis therapies that improve vision in patients with age-related macular degeneration and other related eye diseases are now available.
• Further research is underway to make those treatments both less frequent and more effective.

Angiogenesis in the eye

Although anti-angiogenesis therapy is most often discussed in reference to cancer, anti-VEGF treatments have shown the greatest benefit in the treatment of diseases within the eye. The retina is the most metabolically active tissue in the body, said Anthony Adamis of Genentech. Two layers of vasculature feed this critical tissue that translates photons of light into electrical signals. Diabetic retinopathy, a complication of diabetes mellitus, results from malformations in the inner layer of blood vessels. In age-related macular degeneration (AMD), also a type of retinopathy, blood leakage from the outer layer of blood vessels damages retinal tissue and leads to blindness.

In 1948, Isaac Michaelson first described the "Factor X" hypothesis to explain AMD. New, leaky blood vessels grow from the retinal vasculature, first on the surface of the retina. These vessels have a fibrous component, and a biochemical change in the vitreous fluid of the eye then shrinks these nets of blood vessels. This added strain on these weakened vessels causes them to break open and hemorrhage, and the eye fills with blood. Previously the only treatment for this type of process in the eye had been pan-retinal photocoagulation, a laser therapy that destroys 90% of the retinal tissue to prevent angiogenesis. Though this therapy stops the formation of new blood vessels, patients must accept a significant reduction in their vision in order to keep a small fraction of their eyesight.

VEGF in the eye

In work carried out in Judah Folkman's laboratory at Children's Hospital Boston, Adamis and his colleagues conducted experiments that began to link VEGF and angiogenesis in the eye. They found that VEGF is expressed in retinal tissue where the blood supply has become restricted because of abnormal clotting. They then followed VEGF and angiogenesis in an animal model of angiogenic eye disease. The pattern of VEGF expression levels paralleled the production of new blood vessels in the iris: as VEGF expression increased, more blood vessels formed in the eye.

Blocking VEGF stops the formation of new blood vessels in the iris.

Adamis and his colleagues then collaborated with Ferrara to establish the relationship between VEGF and vascular disorders in the eye. In a series of papers, they demonstrated that VEGF is the primary angiogenic factor in these diseases, and that anti-VEGF antibodies blocked neo-vascularization. They also described the altered vascular architecture in these tissues in response to VEGF. These trends held for all tissues of the eye, validating VEGF as a drug target, Adamis remarked.

Anti-VEGF therapies in the eye

Adamis founded a small company, Eyetech Pharmaceuticals, where he and his colleagues developed a small molecule that targeted one VEGF isoform, VEGF-165. The small molecule, pegaptanib, was FDA-approved and slowed vision loss by 50% in patients with AMD. However, Genentech's modified VEGF antibody ranibizumab (Lucentis) blocked all isoforms of VEGF-A and improved vision in patients who received injections in the eye.

VEGF antibody, ranibizumab, revolutionizes treatment of patients with age-related macular degeneration.

As a result of Lucentis, AMD is no longer the leading cause of blindness in older adults, Adamis said. The antibody is now being used to treat retinal vein occlusion, an eye disorder resulting from a blockage of one of the retinal veins. Genentech will soon seek approval for its use for diabetic macular edema and diabetic retinopathy as well.

Improving therapy

Despite the success of Lucentis, it is still possible to improve patient outcomes. For example, even with therapy more than half of patients have eyesight 20/50 or worse, Adamis noted, which leaves these individuals unable to qualify for a driver's license.

Genentech is currently carrying out a randomized study to examine whether increasing the dose four-fold from 0.5 mg to 2.0 mg will further improve vision in AMD patients. They're also trying to identify patients who are most likely to respond to the drug. Using genetic profiles of patients from previous clinical trials, Genentech researchers have identified 6 single nucleotide polymorphisms (SNPs) in patients that respond to Lucentis. Patients with all 6 SNPs show the greatest improvements in visual acuity after treatment, but those who lack all of these genetic markers show increased vision impairment after taking the drug. A clinical study is underway at Harvard Medical School to validate this connection.

Treatment with Lucentis currently involves monthly intraocular injections, and AMD is a chronic condition. Patients tolerate these treatments well, and compliance rates are generally 90%, Adamis said, but he and his colleagues would like to reduce the frequency of injections. In past studies that have spaced the treatment out to once every three months, patient vision deteriorated. However, researchers are testing a variety of new technologies that might allow for the slow release of drugs over periods of up to 6 months with microspheres, semi-permeable membranes, or even specially engineered cells. Not only could slow delivery ease the burden on patients over the long term, but the need to deliver less of the drug per dose could also minimize the potential for cardiovascular side effects from these treatments.

Anti-VEGF therapies halt the formation of new blood vessels in the retinal tissues, but it is unclear whether removal of the new vessels that have already formed might also improve the visual acuity in patients. Older blood vessels resist VEGF inhibition, but newer blood vessels are susceptible to these drugs. Targeting pericytes (connective tissue cells near small blood vessels) could eliminate these unwanted blood vessels and enhance the effects of anti-VEGF therapies. PDGF-BB (platelet derived growth factor BB) recruits these cells, and a DNA aptamer (an oligonucleotide that binds to a specific target) is capable of stripping these cells from new blood vessels, thereby making them more susceptible to VEGF therapy. Phase I studies of the DNA aptamer indicated some regression of these newly formed blood vessels, and a phase II study is currently underway.

Another way to improve these treatments may be to target inflammation, Adamis added. Unlike normal vasculature, pathological angiogenesis in the eye includes inflammatory cells, and targeting inflammation appears to halt angiogenesis under some circumstances. Finally, early treatment appears to be important for minimizing neural damage from these angiogenic events, so greater emphasis needs to be placed on treating and protecting the neurons of the retina before major damage occurs.

Speakers:
Luisa Iruela-Arispe, University of California, Los Angeles
Dan Duda, Massachusetts General Hospital, Harvard Medical School
Robert Kerbel, University of Toronto

Highlights

• VEGF isoforms have different activities.
• The C-terminal component of cleaved VEGF may contribute to angiogenesis signaling independent of the soluble part of the protein.
• Parallel clinical and laboratory studies are shedding light on the mechanisms of anti-VEGF therapies and point to possible biomarkers of treatment efficacy.

Teasing out the details of VEGF biology

Although targeting angiogenesis has extended the lives of some patients with cancer, many of the details of how VEGF works in the tumor microenvironment remain unclear. Many of the connections between angiogenesis and tumor biology have proved more complicated than initially thought, and many of the anti-angiogenic molecules do not work as well in humans as they do in preclinical models of angiogenesis.

The response of blood vessels to VEGF depends on isoform length.

Luisa Iruela-Arispe and her colleagues at the University of California, Los Angeles, have been trying to understand the roles of the different isoforms of VEGF. These isoforms result from the cleavage of portions of the carboxy-terminal end of the protein. The carboxy-terminal portion of the protein helps to anchor VEGF to the extracellular matrix, and longer VEGF isoforms are typically insoluble. Tumor proteases often cleave these longer isoforms to produce higher concentrations of the shorter, soluble VEGF isoforms.

All isoforms bind to the VEGF1 and 2 receptor proteins, and this binding leads to many of the downstream effects observed in angiogenesis. The carboxy-terminal portion of the protein is critical for VEGF binding to other transmembrane proteins, neuropilin 1 and 2.

To better understand the role of the C-terminal portion of VEGF, Iruela-Arispe and her colleagues designed an uncleavable recombinant VEGF protein and compared its function with natural isoforms. Previously researchers had thought that only soluble VEGF stimulated vascular growth, but these new experiments demonstrate that all forms of VEGF promote the growth of new blood vessels. The growth patterns prompted by these isoforms differ: soluble VEGF produces fatter, leakier blood vessels, while longer VEGF isoforms spur the growth of smaller blood vessels in denser patches. The VEGF signaling kinetics of these proteins—the measure of their activity over time—vary with different isoforms. Soluble VEGF has a half-life of one hour, compared to 7 days for the bound isoform. In Iruela-Arispe's studies matrix-bound VEGF produced different phosphorylation patterns on the VEGF receptor, which also altered downstream effects in the cell signaling cascade.

Iruela-Arispe and her colleagues also wondered what happens to the C-terminal domain after its separation from the soluble VEGF protein. Using labeled peptides the researchers observed the C-terminal domain bound to the surface of endothelial cells. They also showed that the C-terminal portion of VEGF is required to recruit neuropilin to the VEGF receptor 2 and to activate downstream signaling. These signals stimulated both cell migration and branching of capillaries. Overall the work strongly suggests that both the C-terminal portion of VEGF and neuropilin could present additional targets for anti-angiogenesis therapies.

Tailoring anti-VEGF treatments to patients

Many of the cancer therapies that target angiogenesis can have significant side effects and are expensive. As a result, clinicians would like to be able to match the therapy with the patients who are most likely to benefit from a drug. Dan Duda from Harvard Medical School discussed clinical studies at Massachusetts General Hospital that are examining the mechanism and activity of anti-angiogenesis therapies to match treatments to individuals more effectively.

Duda and his colleagues and collaborators are imaging patient tumors and are completing parallel studies in mice. They are using laser capture microdissection, a technique for homing in on tissues of interest, to examine tissues before and after treatment with VEGF-blocking drugs. In addition, they are comparing gene expression patterns before and after treatment and are evaluating the numbers and characteristics of cell types such as progenitor or inflammatory cells using flow cytometry.

In one of example of this type of study, Duda and his colleagues looked at the response of glioblastoma tumors to cediranib, an inhibitor that blocks receptor tyrosine kinase activity of all three VEGF receptors. Over the course of treatment with cediranib, tumor blood vessels shrank and became less permeable, but the tumors did not shrink. Survival time doubled on average because patients were less likely to experience edema, swelling events that are common in these brain tumors.

Looking for physiological changes that they could monitor, they also found that as the vascular basement membrane in the glioma becomes less porous, the amount of collagen IV in the bloodstream increases. These results suggest that it might be possible to model the effectiveness of this treatment using a simple blood draw.

In a clinical trial involving individuals with colorectal cancer, Duda and his colleagues have also found that the soluble form of the VEGF receptor protein can serve as an indicator of which patients are less likely to have side effects from therapies. By examining the serum of patients, the researchers determined that those individuals with higher baseline levels of the soluble VEGF receptor were more likely to respond to anti-VEGF therapy and less likely to experience serious adverse events.

In addition, several chemokines, inflammatory biomarkers, are elevated in patients whose cancers progress despite anti-angiogenic therapies. Duda and his colleagues are currently conducting trials to tease out the connections between some of these chemokines and their role in the response of the tumor and the body to various therapies.

Panel discussion and conclusion

Before the panel discussion, moderator Robert Kerbel from the University of Toronto provided some context for both the progress in the targeting of angiogenesis in cancer and the ups and downs of this area of research.

Approved anti-angiogenic therapies.

After much of the early scientific work on angiogenesis by Ferrara and others, researchers developed a number of drug candidates to target angiogenesis, particularly in the late 1990s. That initial enthusiasm plummeted with the failure of endostatin, with problems in the early trials of bevacizumab, and with the failure of inhibitors of matrix metalloproteases. But the FDA approval of bevacizumab for colorectal cancer, non-small cell lung carcinoma, and metastatic breast cancer marked a resurgence in enthusiasm for anti-angiogenic therapies. At the same time tyrosine kinase inhibitors that target VEGF receptors in renal cell carcinoma and hepatocellular carcinoma have provided treatments for cancers that had been difficult to treat with other strategies.

The four-decade roller coaster of anti-angiogenesis drugs.

Unfortunately, the FDA's 2010 decision to remove breast cancer as a treatment indication for bevacizumab has dampened some of this enthusiasm. So far a follow up trial of bevacizumab for the treatment of breast cancer has shown an extension of survival time before further progression of the disease but not an increase in overall survival, Kerbel said. Debate continues about what the standards for approval should be in such trials, particularly considering concerns about the side effects of these drugs and their expense. Additional clinical trials are underway to confirm these findings.

Clinicians would like to be able to give anti-angiogenesis drugs as adjuvant therapies, after initial treatment to prevent recurrence, but the data do not suggest a benefit. In some cases, patients were worse off after bevacizumab as adjuvant therapy. More recently, clinicians administered bevacizumab as a neoadjuvant therapy, before surgery was performed. Initial results indicated a small, but not statistically significant, benefit. But when the researchers narrowed the pool of patients to women with triple negative tumors, a particularly aggressive form of the disease, those patients showed a benefit from therapy.

Kerbel's group and Douglas Hanahan's group at UCSF published papers in 2009 that suggest that anti-angiogenic therapy could promote tumor metastasis. Those findings remain controversial, and further research is underway to better understand the complex biology of angiogenesis and the overall impact of anti-VEGF therapies.

Why does a drug such as bevacizumab have to be paired with chemotherapy, upfront, to show efficacy?

Why do anti-angiogenic tyrosine kinase inhibitors not improve the efficacy of chemotherapy?

Could anti-angiogenic drugs someday be effective for adjuvant treatment of early stage microscopic metastatic disease?

What are clinically relevant mechanisms of intrinsic or acquired resistance to anti-angiogenic drugs?

Will practical, reliable, predictive biomarkers be developed for anti-angiogenic drugs?

Other than the VEGF pathway, what are most promising pro-angiogenic pathways to target?

Will toxicities emerge as a significant negative factor in elderly patients receiving chronic anti-angiogenic therapy for blinding conditions such as AMD? Could the frequency of treatments be reduced to mitigate these toxicities?