Modulating Angiogenesis: Title and subTitle BreakMore vs Less

It is now more than 30 years since Folkman first proposed that formation of new blood vessels (angiogenesis) was critical to tumor growth and development.¹ What was then a novel concept has during the intervening years gained widespread acceptance and is the basis not only of groundbreaking research in many fields, ranging from cancer through rheumatoid arthritis (RA) and cardiovascular disease, but also of exciting therapeutic approaches. In the 1970s, as an indicator of the emerging importance of this area, there were approximately 100 references in the National Library of Medicine's database with the key words angiogenesis or angiogenic. This increased to more than 7000 in the 1990s, although since the advent of the new millennium, there have been (at the last count) more than 13 000. Indeed, it is significant that during the preparation of this article, the first antiangiogenic agent, bevacizumab, an antibody that binds to vascular endothelial growth factor (VEGF), was approved by the US Food and Drug Administration (FDA) for the treatment of metastatic colorectal cancer.²

The requirement for blood vessels arises from the need to maintain oxygen homeostasis, as well as to deliver nutrients and to remove waste products in vivo. The earliest vessels provide nutrients and oxygen to the growing embryo, as a result of vasculogenesis, during which precursor cells commit to form endothelial cells rather than hematopoietic cells and fuse to form a primitive meshwork. Subsequently, angiogenesis extends and remodels these structures to form the primordial aorta and vein, as well as yolk sac arteries and veins. In contrast, endothelial cell turnover in adults can often be measured in years, because the body can normally cope with changes in oxygen delivery without compromising respiration.

Only under certain situations, physiological and pathological, does active angiogenesis occur. In general, the common denominators are an increase in tissue mass, a reduction in oxygen levels, or both. These may occur simultaneously (eg, during tumor growth or atherosclerotic plaque development) and frequently are associated with expression of cytokines, such as tumor necrosis factor α (TNF-α) or transforming growth factor β (TGF-β), as well as angiogenic factors, such as VEGF. In certain conditions, hypoxia and inflammatory molecules seem to be the predominant driving force, as is the case when the vasculature is disrupted during bone fractures.³

Members of the VEGF family, including the original member, VEGF (also known as VEGF-A), VEGF-B, and placental growth factor, play important roles in angiogenesis. The effects of VEGF are mediated via receptors whose cytoplasmic regions contain sequences with tyrosine kinase enzyme activities, termed Flt-1 or VEGF-R1 and KDR/Flk-1VEGFR-2.⁴ Of relevance is the fact that VEGF levels are increased by hypoxia, making VEGF-driven angiogenesis a central response to low oxygen tension. Hypoxic induction of VEGF results from stabilization of hypoxia-inducible transcription factor 1 (HIF-1) and factor 2 (HIF-2). Other proangiogenic factors include fibroblast growth factors, epidermal growth factor, and platelet-derived growth factor (PDGF).³

Angiogenesis plays a key role in many pathological conditions. VEGF, in particular, appears to be a master regulator of this process. For example, excessive angiogenesis is thought to promote and maintain RA, tumor growth and metastasis, loss of vision as part of diabetic retinopathy or age-related macular degeneration, and atherosclerosis. Conversely, insufficient angiogenesis may be involved in complications of fracture healing, stroke, and myocardial ischemia.

Currently, angiogenesis can be targeted at several stages, including inhibition of factors such as VEGF, interruption of downstream signaling, in particular the receptor tyrosine kinases, blockade of matrix degrading enzymes, or using endogenous inhibitors such as endostatin (Figure 1). Many of these approaches have been used with varying degrees of success for human cancers. Some inhibitors are entering trials for other angiogenesis-dependent diseases, including RA. Furthermore, the VEGF aptamer pegaptanib sodium, a small nucleic acid molecule that binds to and inhibits VEGF, has shown promise as a treatment for age-related macular degeneration, the most common cause of irreversible central visual loss in elderly populations of the industrialized world, which is also characterized by excessive angiogenesis.⁵

Malignancies. Angiogenesis has been a putative target for anticancer therapy since it was first linked to tumor growth and metastases in the 1970s.¹ VEGF is overexpressed by numerous solid angiogenic tumors and hematological malignancies. Therefore, interrupting the VEGF pathway has become a major focus of oncological research⁶ and many clinical studies are currently in progress. Some of these, as well as other trials of inhibitors designed to block other steps in the angiogenic cascade, are briefly reviewed herein.

The most successful antiangiogenic approach is bevacizumab, a humanized monoclonal anti-VEGF antibody. A recent phase 3 study in metastatic colorectal cancer demonstrated increased response rates, prolonged time to cancer progression, and significantly increased survival times when bevacizumab was used with first-line chemotherapy (irinotecan, 5-fluorouracil, and leucovorin).⁷ Following the success of this pivotal trial, the FDA very recently approved bevacizumab to be used in combination with intravenous 5-fluorouracil–based chemotherapy as a treatment for patients with first-line or previously untreated metastatic cancer of the colon or rectum.² This makes bevacizumab the first approved antiangiogenic therapy for cancer treatment and more broadly represents a major advance in antiangiogenic therapy.

Other approaches to block angiogenesis are based on VEGF receptors. High-affinity decoy receptors known as "VEGF traps" have been developed, combining domains from both VEGF-R1 and VEGF-R2, and presumably binding and thereby inactivating circulating VEGF. Initial animal experiments have shown VEGF trap capable of halting tumor growth and even producing a degree of regression.⁸ Phase 1 trials of VEGF trap in patients with advanced solid tumors and non-Hodgkin lymphoma are ongoing. Use of anti-VEGF receptor antibodies is less well described, although a phase 1 trial of anti–VEGF-R2 (IMC-1C11) in patients with liver metastases with colorectal cancer demonstrated evidence of safety and low drug-related toxicity. However, human antichimeric antibodies were detected in 50% of patients.⁹ This observation suggests that the antibody is itself immunogenic, a feature that has previously been observed with other antibodies, such as infliximab. However, in the case of infliximab, this effect was reduced by coadministration of methotrexate, and therefore future use of IMC-1C11 is feasible.

Many pharmaceutical companies have attempted to target the tyrosine kinase activity of VEGF receptors and a number of such inhibitors are currently in clinical trials (Table 1). The most advanced of these is PTK787/ZK222584, an orally available VEGF-R1 and VEGF-R2 inhibitor. Phase 1 and 2 trials have produced promising results with a minimum of adverse effects.¹⁰ Phase 3 trials investigating the effectiveness of PTK787/ZK222584 in combination with 5-fluorouracil, irinotecan with and without oxaliplatin chemotherapy in patients with metastatic colorectal cancer, are ongoing. SU11248, an inhibitor with specificity against not only VEGF receptors but also PDGF and Flt-3 receptor tyrosine kinases, is now entering a phase 3 trial for gastrointestinal stromal tumors.

Other approaches have attempted to mimic naturally occurring angiogenesis inhibitors, such as thrombospondin 1. Structurally modified thrombospondin 1 mimetics have been made that include the active antiangiogenic N-terminal region. Other naturally occurring angiogenesis inhibitors that are currently under evaluation include endostatin and angiostatin. Furthermore, inhibitors of αv integrins, which prevent the interaction between proliferating endothelial cells and matrix components, have been developed and Medi-552, a humanized anti-αvβ3 monoclonal antibody, is in phase 1 trials.

Antiangiogenic therapies offer a number of advantages over conventional chemotherapy, including reduced toxicity, a broader spectrum of activity, and reduced development of drug resistance. However, despite these attributes, the main achievement so far appears to be a stabilization of disease progression, and so their full potential in cancer remains to be determined.

Rheumatoid Arthritis. Rheumatoid arthritis is a chronic systemic inflammatory disease of unknown etiology. It occurs worldwide with a prevalence of approximately 1% and is one of the most common causes of disability in the western world.¹¹ The characteristic pathological feature of RA is the formation of an inflammatory erosive synovitis.¹² Early changes in the synovium are characterized by angiogenesis, inflammatory cell infiltration, and synoviocyte hyperplasia. Destruction of cartilage, bone, and soft tissues results in long-term deformity leading to loss of function.

Recent work suggests that synovial angiogenesis precedes all other pathological features of RA. Indeed, exuberant angiogenesis has been shown in studies of joint synovium from patients who have RA.¹³ Synovial blood vessel number has been found to correlate with hyperplasia, mononuclear cell infiltration, and indices of joint tenderness.¹⁴ Endothelial cells lining blood vessels within RA synovium have been shown to express cell-cycle antigen and integrin αvβ3.¹⁵ As a result of these studies, angiogenesis is now recognized as a fundamental component of disease progression in RA.¹⁶ A variety of angiogenic mediators, including cytokines and growth factors, have been identified in rheumatoid joints.¹⁶ Serum VEGF levels have been measured at higher levels in patients with inflammatory arthritis compared with healthy controls. Levels of VEGF are higher in patients with RA compared with patients with osteoarthritis.¹⁷ A significant correlation was found between serum VEGF levels at presentation and the magnitude of deterioration observed in hand and feet radiographs during the first year, suggesting a prognostic role for serum VEGF levels in RA.¹⁷ Despite active angiogenesis, the RA joint has been shown to be hypoxic. Using a murine arthritis model, onset of disease has been shown to be associated with a reduction in synovial oxygen tension.¹⁸ Joint inflammation has been shown to be associated with an increase in cellular hypoxia and HIF-1 expression in an adjuvant-induced arthritis model.¹⁹ These observations suggest that hypoxia may provide a pivotal stimulus in RA progression. Importantly, hypoxia has been shown to synergize with inflammatory cytokines expressed in RA (TGF-β and IL-1) to up-regulate VEGF, suggesting that these stimuli may act in concert to promote RA synovial angiogenesis.²⁰

Considerable insights into the importance of angiogenesis in RA were also gained from trials of the anti–TNF-α antibody infliximab. Patients with RA who are treated with infliximab have reductions in serum VEGF levels. Furthermore, combining infliximab with methotrexate in multiple infusions prolonged the reduction of VEGF observed with infliximab alone.²¹ These observations imply reducing angiogenesis may play a role in the beneficial actions of anti–TNF-α therapy.

More recently, the role of angiogenesis has been studied in RA tenosynovitis. Approximately 50% of patients with RA have tendon involvement and dorsal tenosynovitis is often the first presentation of the disease.²² Proliferation of the synovial lining of tendons causes scarring and adhesion formation and 50% of patients with tendon disease will also show tenosynovial invasion into the tendon substance itself.²³ This invasion is associated with multiple tendon ruptures and a poorer prognosis for long-term hand function.²⁴ VEGF has been detected in RA tenosynovium at levels equivalent to those found in RA joint synovial samples in culture systems using cells obtained from whole tissue,²⁵ suggesting a role for angiogenesis in tenosynovitis.

The potential for inhibiting the enhanced angiogenic drive observed in RA has been explored through numerous animal studies using broad-spectrum angiogenesis inhibitors. Human soluble VEGF receptor 1 (sFlt-1) has been shown to significantly reduce joint inflammation, bone and cartilage destruction, and disease severity in a murine collagen-induced arthritis (CIA) model.²⁶ The effectiveness of VEGF blockade in CIA models has been highlighted in other studies using anti-VEGF polyclonal antibodies.^{27 - 28} In a detailed collaborative study, the effects of antibodies against VEGF-R1 and VEGF-R2 were compared in arthritis. In contrast to murine tumor models, in which anti–VEGF-R2 was nearly as effective as anti–VEGF-R1, Luttun et al²⁹ observed reduced joint inflammation and destruction in a model of CIA anti–VEGF-R1 but not anti–VEGF-R2. However, only 1 clinical trial has been initiated to date in RA, involving anti-αvβ3 antibody Medi-552, a phase 2 randomized, double-blind, placebo-controlled study, the results of which are still awaiting publication.

The reviewed studies support the role of angiogenesis in RA development and implicate VEGF as a key player in this process. It is likely that the blockade of angiogenesis holds the potential for considerable therapeutic benefit in the future for RA. This will most likely produce greatest benefit through combination with existing treatments such as anti–TNF-α therapy.

Therapeutic angiogenesis has been heralded as a significant therapeutic avenue for treatment of ischemic disease, in which arterial thickening or occlusion can lead to hypoxia, as well as in union of bone fractures, which leads to disruption of the normal blood supply, resulting in necrosis and hypoxia.

Cardiovascular Disease. The awareness that angiogenesis is pertinent in the context of cardiovascular disease has arisen from the fact that occlusion or narrowing of arteries is likely to result in hypoxia, in response to which the ischemic myocardium develops collateral vessels. However, this compensatory angiogenesis seems to be insufficient. The concept of therapeutic angiogenesis exploits and supplements the physiological response to hypoxia or ischemia. A number of approaches have been studied, with varying degrees of success.^{30 - 31}

In their most fundamental form, trials have involved administration of proangiogenic factors. For example, the VIVA trial (VEGF in Ischemia for Vascular Angiogenesis) was a trial of 2 doses of VEGF protein in stable angina.³² At 2 months, there was no significant effect of VEGF beyond that of placebo on the primary end point, namely exercise treadmill test times, and no improvements were observed in myocardial perfusion. However, this study did demonstrate a significant improvement at 4 months in angina class in patients receiving the higher dose of VEGF and a trend toward improved exercise times. In another study³³ in patients with myocardial ischemia, VEGF plasmid reduced angina and single photon emission computed tomography ischemia scores. These investigators have also recently reported improvements in a placebo-controlled trial of VEGF-2 (VEGF-C) plasmid. In contrast, administration of adenovirus expressing VEGF did not improve exercise performance or quality of life in patients with peripheral arterial disease.³⁴

A more sophisticated conduit for enhanced angiogenesis has involved cell-based therapy. This is based on the recent discovery that postnatal vasculogenesis may contribute to blood vessel formation in adults. A scarce population of bone marrow–derived endothelial progenitor cells are mobilized by VEGF and angiopoietin 1. These cells have the characteristics of, and therefore share surface markers with, both mature endothelial cells and subsets of hematopoietic stem cells. In a pig coronary artery ligation model, transfer of bone marrow–derived mononuclear cells increased collateral vessels, with cells becoming incorporated into new capillaries.³⁵ Early phase clinical trials in humans appear promising. For example, autologous mononuclear bone marrow cells implanted into ischemic myocardium resulted in improvement in symptoms and myocardial perfusion.³⁶

However, it is also worth noting that atherosclerosis may involve plaque angiogenesis. Normally, the arterial intima is devoid of blood vessels, with oxygen and nutrients being supplied from the lumen and by the adventitial vasa vasorum. However, during plaque development, the intima, and to a lesser extent the media, become thickened and this is accompanied by the appearance of numerous blood vessels.³⁷ VEGF has been shown to actually promote plaque development in various in vivo models.³⁸ Further support for a role of angiogenesis in plaque progression came from a cholesterol-fed rabbit arterial injury model, in which adventitial delivery of angiogenesis inhibitors (paclitaxel or angiostatin) inhibited neointima progression.³⁹

Angioplasty and coronary artery bypass graft surgery are the current interventions of choice for myocardial ischemia. Whether promoting angiogenesis will supplement or overtake these is still open to debate. A recent study⁴⁰ reported improvements in myocardial perfusion, systolic function, and exercise capacity following intracoronary infusion of peripheral blood stem cells in patients with myocardial infarction who had undergone coronary stenting. However, those patients also had a high rate of in-stent restenosis, suggesting that stem cell therapy accelerated neointimal growth in the bare metal stents.

Fracture Healing.Angiogenesis has also been highlighted as an integral component of fracture repair.⁴¹ Many cytokines, such as TGF-β and PDGF, play a key role in promoting fracture healing⁴² and may regulate angiogenesis in fractures through the induction of molecules such as VEGF.²⁰ A fracture disrupts the normally plentiful bone-blood supply. Revascularization of the fracture site is an essential step before resorption of necrotic tissue and introduction of osteogenic stem cells.^{43 - 44} Later, during the phase of endochondral ossification, vascularization of the cartilage is necessary for the subsequent apoptosis of hypertrophic chondrocytes and mineralization of the matrix to form woven bone prior to remodeling.

Much of the evidence demonstrating a role for VEGF in angiogenesis during fracture healing has been drawn from studies of the growth plate of long bones, where it has been shown that the latter stages of endochondral ossification recapitulate those occurring at the fracture site. For example, blood vessel invasion into the cartilage analogue in the growth plate has been shown to be coupled with osteogenesis, with VEGF playing a pivotal role.⁴⁵ VEGF expression precedes blood vessel formation in developing mouse bones and its expression is tightly associated with cells involved in bone formation.⁴⁶ Other studies have shown that blockade of angiogenesis in a closed rat femoral fracture model by the inhibitor O-(chloroacetylcarbamoyl)fumagillol suppressed both intramembranous and endochondral ossification.⁴⁷ Inhibition of VEGF using soluble VEGF-R1 in animal fracture models dramatically reduced healing, as well as angiogenesis, bone formation, and callus mineralization.⁴⁸ Human fracture hematomas express angiogenic activity, most likely VEGF.^{49 - 50} The initial hematoma formed at the site of injury is inherently angiogenic, primarily due to expression of VEGF.⁵¹

VEGF has therefore been shown to be vital for angiogenesis in fracture repair and mineralization of the cartilage to form bone in response to injury, suggesting that treatment with exogenous VEGF might be expected to promote angiogenesis and bone formation after injury. There are no clinical trials of VEGF therapy for this purpose at present, although several studies support this hypothesis. Street et al⁴⁸ demonstrated that addition of VEGF enhanced bone formation in mouse femur fractures and rabbit radial defects. In a model of distraction osteogenesis in a rabbit tibial fracture, locally applied exogenous VEGF increased bone-blood flow.⁵² Finally, local adenoviral delivery of VEGF into rabbit femurs increases bone formation and decreases bone resorption.⁵³

Although these in vitro and in vivo studies are encouraging, the timing and dose of administration together with the method of application still need to be investigated. Furthermore, the osteogenic properties of VEGF need to be further elucidated, so that appropriate enhancement of the fracture healing process can be achieved.

Angiogenesis is generally a response to increased tissue mass, alterations in oxygen tension, or both. In adults, angiogenesis generally does not occur, except under specific physiological situations, such as during fracture healing. However, angiogenesis, excessive or insufficient, underlies many pathological situations, such as RA, malignancies, or cardiovascular disease. Targeting angiogenesis to make more or less vessels should yield new therapeutic options in the future.