Summary

Tumours need to develop a new blood supply to grow and metastasise. This process is called angiogenesis. Drugs that inhibit angiogenesis are therefore being evaluated in combination with chemotherapy for the treatment of various cancers. Vascular endothelial growth factor and its receptors are intimately involved in angiogenesis so they are targets for new drugs. Bevacizumab is a monoclonal antibody against vascular endothelial growth factor and is approved for use in metastatic colon cancer. Thalidomide also inhibits angiogenesis and may be used in the treatment of multiple myeloma.

 

Introduction

Small tumours are able to grow because they can obtain nutrients and oxygen by diffusion. For tumours to enlarge further they need to develop new collateral blood vessels to provide the essential nutrients for invasion, growth and subsequent metastasis. The formation of new vessels is termed neovascularisation. This is driven by the process of angiogenesis, the sprouting of new vessels from existing vasculature.

Angiogenesis has been an appealing target for anticancer drugs for 30 years, but it is only recently that this promise has borne some fruit. There are now over 30 angiogenesis inhibitors currently in clinical trials for the treatment of malignancy (Table 1). These drugs appear to have a cytostatic rather than cytotoxic effect, leading to tumour dormancy. The available data suggest that anti-angiogenic drugs work best in conjunction with chemotherapy. Their development also involves the identification and management of a new range of toxicities.

 

Angiogenesis

The development of blood vessels is a complex equilibrium regulated by anti-and pro-angiogenic factors. The balance may be tilted in favour of angiogenesis by hypoxia or inflammation. This has a physiological advantage, for example in wound healing, but may be part of the pathological process in chronic inflammatory disease or cancer.

With tumour-associated angiogenesis, the cancer releases various pro-angiogenic factors (including angiogenin, vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), and transforming growth factor-β (TGF-β)). These stimulate endothelial cell proliferation, migration and invasion resulting in new vascular structures sprouting from the patient's blood vessels. Cell adhesion molecules, such as integrins, are critical to the attachment and migration of endothelial cells to the extracellular matrix. The receptor for platelet derived growth factor is also important in angiogenesis as it is central to the recruitment of pericytes, the cells that surround and support capillaries.

When a tumour stimulates the growth of new vessels, it is said to have undergone an 'angiogenic switch'. The principal stimulus for this angiogenic switch appears to be oxygen deprivation, although other stimuli such as inflammation, oncogenic mutations and mechanical stress may also play a role.

The angiogenic switch leads to tumour expression of pro-angiogenic factors and increased tumour vascularisation. It is associated with more advanced tumour stages and worse prognosis in several human malignancies, including malignant melanoma and gastrointestinal, breast, prostate and lung cancers.

Vascular endothelial growth factor

Of the angiogenic factors secreted, VEGF is perhaps the most specific for endothelial cells. When VEGF binds to its receptor it triggers signalling pathways that result in endothelial cell migration, differentiation and proliferation, increased vascular permeability and release of endothelial cell precursors from the bone marrow. VEGF also prevents endothelial cell apoptosis. Higher concentrations have been associated with malignant effusions.

The VEGF-related family of genes involved in angiogenesis and lymphangiogenesis produce a number of glycoproteins, called VEGFs A-E and placental growth factors (PlGF) 1 and 2. These glycoproteins have several biologically active isoforms.

Table 1 Angiogenesis inhibitors and their targets
Drug Target Clinical development
Monoclonal antibody
Bevacizumab VEGF-A Approved in Australia
IMC-1121B VEGFR-2 Phase I
2C3 VEGF-A Preclinical
Receptor tyrosine kinase inhibitors
PTK-787 VEGFR-1, -2 Phase III
AEE788 VEGFR-2, EGFR Preclinical
ZD6474 VEGFR-1, -2, -3, EGFR Phase II
ZD2171 VEGFR-1, -2 Phase I
SU11248 (sunitinib) VEGFR-1, -2, PDGFR Phase II/III
G013736 VEGFR-1, -2 Phase II
EP-7055 VEGFR-1, -2, -3 Phase I
P-547,632 VEGFR-1, -2 Phase I/II
GW786034 VEGFR-1, -2, -3 Phase I
BAY 43-9006 VEGFR-1, -2, PDGFR Phase III
AMG706 VEGFR-1, -2, -3 Phase I
Soluble receptor chimeric protein
VEGF-Trap VEGF-A, PlGF Phase I
Inhibitors of endothelial cell proliferation
ABT-510 Endothelial CD36 Phase I/II
Angiostatin Various Phase I
Thalidomide Reduction of TNF-α Approved in Australia
Inhibitors of integrin's pro-angiogenic activity
Medi-522 Integrin αV Phase I/II
EMD12194 (Cilengitide) Integrin αV Phase I/II
Matrix metalloproteinase inhibitors
Marimastat MMP-1, -2, -3, -7, -9 Phase III
Prinomastat MMP-2, -9 Phase III
BMS 275291 MMP-1, -2, -8, -9, -13, -14 Phase III
Neovastat MMP-2, -9, -12, VEGF Phase III
Other
CDP-791 VEGFR-2 Phase I
Vascular targeting drugs
Combretastatin Endothelin tubulin Phase I/II
AVE8062A Endothelin tubulin Phase I
ZD6126 Endothelin tubulin Phase I
AS1404 Induction of TNF-α Phase I
Key VEGF Vascular endothelial growth factor
VEGFR Vascular endothelial growth factor receptor
EGFR Epidermal growth factor receptor
PDGFR Platelet derived growth factor receptor
PlGF Placental growth factor
TNF Tumour necrosis factor
MMP Matrix metalloproteinases

The VEGFs are produced either by direct secretion from the tumour, or by cleavage of isoforms sequestered in the extracellular matrix by enzymes such as plasmin or the matrix metalloproteinases.

VEGFs bind to at least three receptors (VEGFR-1, VEGFR-2, VEGFR-3). The isoforms can bind with other receptors (neuropilin receptors) which may also have a role in angiogenesis. The structure of each VEGF receptor includes a kinase. For example, ms-like tyrosine kinase (Flt-1) is part of VEGFR-1. These enzymes are involved in intracellular signalling when VEGFs bind to their receptors (Fig. 1).

VEGF promotes the release of other angiogenic factors and proteolytic enzymes. The release of proteolytic enzymes results in degradation of the vascular basement membrane. New vessels are formed as endothelial cells are organised into functional tubular structures. Individual vessels then connect to form networks that allow blood to circulate. The new blood vessels formed are derived from the host and not the tumour, however they are more tortuous and leaky than normal vessels.

 

Angiogenesis inhibition in the treatment of cancer

To stop angiogenesis requires treatment with anti-angiogenic factors, or drugs which reduce the production of pro-angiogenic factors, prevent them binding to their receptors or block their actions. The drugs being studied can be broadly defined as those that are exclusively anti-angiogenic, such as bevacizumab, and those that have additional functions, such as thalidomide and the cyclo-oxygenase (COX)-2 inhibitors.

Endogenous anti-angiogenic factors

Endostatin is the carboxy-terminal fragment of collagen XVII. It is thought to induce apoptosis in endothelial cells and inhibition of their migration to sites of neovascularisation, probably by interfering with endothelial cell adhesion. In preclinical models, endostatin has inhibited the growth of a wide variety of human primary and metastatic tumours. Clinical trials suggest that endostatin is well tolerated, but only minor evidence of antitumour activity has been observed.

Another endogenous inhibitor of angiogenesis is angiostatin. Like endostatin, it directly induces apoptosis of endothelial cells by disrupting the normal adhesion contacts between the endothelial cells. Angiostatin also acts by inhibiting VEGF and basic fibroblast growth factor (bFGF).

Interferon-alfa has an anti-angiogenic effect by inhibiting endothelial cell migration. It has been successfully used to treat haemagiomas, refractory giant cell tumours and angioblastomas.

Thalidomide

There has been renewed interest in this potent teratogen since it has been shown to be both an immunomodulatory and anti-angiogenic drug. Thalidomide is thought to inhibit angiogenesis by reducing levels of bFGF,VEGF, COX-2 and tumour necrosis factor (TNF-α). It may also reduce tumour-induced overproduction of circulating precursors of endothelial cells.

In patients with multiple myeloma there is an increased rate of angiogenesis within the bone marrow. Thalidomide has been used in the treatment of resistant multiple myeloma as it has anti-angiogenic effects and can directly inhibit the growth and survival of myeloma cells.


Fig. 1
Vascular endothelial growth factor receptors2

The vascular endothelial growth factor receptors (VEGFR) consist of a binding domain and a tyrosine kinase domain. Each receptor is associated with a different form of tyrosine kinase (Flt-1, Flk-1/KDR, Flt-4). The neuropilin receptors (NRP) act as co-receptors for vascular endothelial growth factor (VEGF). Some of the VEGF molecules bind with more than one receptor. Placental growth factor (PlGF) binds with VEGFR-1.


VEGF inhibitors

Inhibition of the VEGF pathway has become the focus of angiogenesis research as approximately 60% of malignant tumours express high concentrations of VEGF. Strategies to inhibit the VEGF pathway include antibodies directed against VEGF or VEGFR, soluble VEGFR/VEGFR hybrids, soluble analogues of the VEGFR (VEGF-Trap) and tyrosine kinase inhibitors. One of the earliest strategies to inhibit VEGF activity involved the use of antibodies directed against VEGFRs. For example, preclinical data with anti-VEGFR-2 antibodies demonstrated decreased VEGF-induced signalling, decreased angiogenesis and decreased primary and metastatic growth in a variety of tumour systems.

VEGF-Trap is a decoy receptor. It consists of parts of VEGFR-1, VEGFR-2 and immunoglobulin G (IgG). The molecule is soluble and binds to VEGF-A before it can reach its normal receptors. VEGF-Trap binds VEGF-A 100-to 1000-fold more tightly than monoclonal antibodies. It inactivates all circulating and tissue VEGF-A isoforms and PlGF.

Several small molecule inhibitors of tyrosine kinase activity have been developed. These have activity not only against VEGFR-2, but also on other VEGFRs, fibroblast growth factor receptor, the epidermal growth factor receptors (EGFR*) and platelet derived growth factor receptors (PDGFR-α, PDGFR-β). For example, sunitinib (SU11248) has activity against VEGFR-2 and PDGFR (see Table 1).

Bevacizumab

Bevacizumab is derived from a monoclonal antibody to murine VEGF. Genetic engineering produces a 93% human and 7% murine protein sequence. The molecule has the same biochemical and pharmacologic properties as the natural antibody, but with reduced immunogenicity and a longer biological half-life. By binding to VEGF-A bevacizumab prevents it from binding with its receptors.

Preclinical studies reported impressive responses and prevention of tumour growth in almost all tumour xenografts. Bevacizumab has been studied in a number of clinical trials and is approved for use in metastatic colon cancer.

Other strategies

There are a variety of other drugs that have at least some anti-angiogenic properties. It has been known for some time that low-dose chemotherapy with cytotoxic drugs, such as the taxoids, produces some anti-angiogenic effects. In addition, inhibition of other molecular targets also has the potential to interfere with angiogenesis. These targets include EGFR, COX-2, TGF-α and the proteosome. Other drugs that have been shown to have anti-angiogenic effects in preclinical models are zoledronic acid and rosiglitazone.

COX-2 is an important mediator of angiogenesis and tumour growth. COX-2 expression occurs in a wide range of preneoplastic and malignant conditions. The enzyme has been localised to neoplastic cells, endothelial cells, immune cells, and stromal fibroblasts within tumours. It mediates its pro-angiogenic effects primarily by three products of arachidonic acid metabolism - thromboxane A2, prostaglandin E2 and prostaglandin I2. These products promote angiogenesis by a number of mechanisms including stimulation of VEGF, promotion of vascular sprouting and tube formation, increased survival of endothelial cells and activation of EGFR-mediated angiogenesis. Studies have shown that selective inhibition of COX-2 activity will suppress angiogenesis in vitro and in vivo and therefore COX-2 inhibitors could be a useful adjunct to therapy.

Expression of human epidermal growth factor receptor 2 (HER-2) within tumour cells is closely associated with angiogenesis and VEGF expression. This is thought to be mediated by transregulation of HER-2 by proteins called heregulins. These heregulins regulate the expression and secretion of VEGF in breast cancer cells. Trastuzumab is a monoclonal antibody that blocks HER-2.1 This reduces tumour cell growth and VEGF expression by the inhibition of heregulin-mediated angiogenesis both in vitro and in vivo. Trastuzumab is currently available for patients with metastatic breast cancer if the tumour overexpresses HER-2.

It is not known how much of the anticancer effects of drugs aimed at molecular structures are due to their angiogenic effects. Angiogenesis is a complex process and successful inhibition of angiogenesis may involve the combination of multiple drugs with differing modes of action.

Another strategy related to angiogenesis is the destruction of new vessels. This has led to the development of vascular targeting drugs (Table 1).

 

Adverse effects

The full spectrum and aetiology of toxicities produced by the angiogenesis inhibitors has yet to be defined. The induction of venous and arterial thromboses, bleeding, hypertension and proteinuria by drugs, such as bevacizumab, is probably directly related to their effects on endothelial cells. The gut perforation occasionally associated with bevacizumab may be related to induction of ischaemia. The teratogenic effects of thalidomide may have been due to its action on peripheral blood vessel development in the fetus. However, it is unlikely that the common adverse effects of thalidomide such as somnolence, rash and neuropathy are related to its effect on angiogenesis.

 

Conclusion

The use of angiogenesis inhibitors is an exciting new area of cancer research. Their optimal use has yet to be defined.


* EGFR is also known as human epidermal growth factor receptor (HER) and these abbreviations are interchangeable.

Stephen Clarke is the Principal investigator for Asia Pacific for two Roche-sponsored studies involving bevacizumab. He has been a member of a Roche colorectal advisory board and an invited speaker at Roche-sponsored clinical meetings.
 

Further reading

Folkman J. Tumor angiogenesis: therapeutic implications [review]. N Engl J Med 1971;285:1182-6.

Folkman J. Endogenous angiogenesis inhibitors [review]. APMIS 2004;112:496-507.

Ferrara N. Vascular endothelial growth factor: basic science and clinical progress [review]. Endocr Rev 2004;25:581-611.

Sparano JA, Gray R, Giantonio B, O'Dwyer P, Comis RL. Evaluating antiangiogenesis agents in the clinic: the Eastern Cooperative Oncology Group portfolio of clinical trials. Clin Cancer Res 2004;10:1206-11.

Hurwitz H, Fehrenbacher L, Novotny W, Cartwright T, Hainsworth J, Heim W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335-42.

Bilenker JH, Haller DG. Future directions with angiogenesis inhibitors in colorectal cancer [review]. Clin Colorectal Cancer 2004;4 Suppl 2:S86-93.

 

Self-test questions

The following statements are either true or false.

1. Increased expression of angiogenic factors is associated with an improved prognosis for patients with cancer.

2. Bevacizumab may cause thrombosis and haemorrhage.

Answers to self-test questions

1. False

2. True

 

Stephen J. Clarke

Sydney Cancer Centre and Department of Medicine, University of Sydney, Concord Hospital Clinical School, Sydney

Rohini Sharma

Sydney Cancer Centre and Department of Medicine, University of Sydney, Concord Hospital Clinical School, Sydney