The past decade has seen considerable advances in the understanding of angiogenesis. Blood vessel development and growth in the central nervous system (CNS) are tightly controlled processes that are regulated by pro-angiogenic and anti-angiogenic factors. When this balance becomes deranged, angiogenesis can progress unabated and support the development of neoplastic and non-neoplastic disorders.1,2 In fact, angiogenic factors have been implicated in the pathogenesis of a wide variety of disorders, including primary and metastatic brain tumours. To date, more than 19 angiogenic factors and 300 angiogenesis inhibitors have been identified.1 The potential clinical implications of angiogenesis research include inhibition of angiogenesis to control brain tumours. However, anti-angiogenic agents targeting one single pathway or growth factor have not proven to be of clinical benefit as yet, and hence, further investigation continues. This article summarizes the processes of blood vessel formation in the brain, the angiogenic factors that are prominent in the CNS and the clinical use of angiogenesis inhibitors.
Angiogenesis is a fundamental process in reproduction and wound healing. It is a tightly regulated process causing neovascularization. However, if angiogenesis becomes unregulated, it may be responsible for several disease processes such as brain tumour growth and metastasis. An understanding of the factors implicated in angiogenesis and its inhibition is essential if they are to be exploited as possible clinical treatments for brain tumours. Unfortunately, there are multiple factors known to be involved in the regulation of angiogenesis, and hence, the clinical application of any single agent may not be effective. This article summarizes the processes of blood vessel formation in the brain, examines the angiogenic factors that are important in the nervous system and discusses their role in brain tumour development and possible treatment.
Blood vessel development in the normal brain
Vascularization is a crucial requirement for normal embryonal development and physiological function. Blood vessel formation throughout the body occurs in three ways: vasculogenesis, angiogenesis and arteriogenesis (Fig. 1).3,4,5 These processes depend on tight regulation of blood vessel growth, maintained by a delicate balance between pro-angiogenic and anti-angiogenic factors. Specific angiogenic factor-signalling systems correspond to each process of blood vessel formation. Recent evidence suggests that vasculogenesis, angiogenesis and arteriogenesis are not exclusive processes and are actually part of a continuum.2
Vasculogenesis involves two distinct steps: the in situ differentiation of mesoderm-derived angioblasts (endothelial cell precursors that have not formed a lumen) and the association of those cells to form blood vessels.1,3,5 Early in embryogenesis, angioblasts migrate into the head region to cover the developing brain.1 The differentiation of angioblasts to endothelial cells depends on a paracrine-signalling system consisting of vascular endothelial growth factor (VEGF) and its high-affinity receptor, tyrosine kinase VEGF receptor-1 (VEGFR-1).6 Further differentiation of endothelial cells, lumen formation, basal lamina production and vessel wall morphogenesis seems to involve a variety of molecules, including VEGF and other angiogenic factors, cell adhesion molecules and transcription factors. The primary vascular plexus is established by vasculogenesis, before the heart begins to beat. In the brain, embryonic vasculogenesis seems to be limited to the formation of the extracerebral vascular plexus, a fibronectin-rich matrix in the primitive meninges.1 Recent evidence suggests that vasculogenesis may also occur in the adult brain.7
Vascularization of the brain occurs primarily via angiogenesis.1 Angiogenesis involves the formation of new blood vessels via (i) sprouting or (ii) splitting from pre-existing vessels and is primarily responsible for the development of blood vessels during later embryogenesis and after birth.6,8,9
Sprouting angiogenesis takes place in two phases, namely growth and stabilization. The initial step during the growth phase is nitric oxide-mediated vasodilation. Nitric oxide also promotes the synthesis of the VEGF family of angiogenic factors. Vasodilation is followed by a VEGF-mediated increase in vascular permeability.1 This permits extravasation of plasma proteins to support endothelial cell proliferation. Simultaneously, the basement membrane and surrounding interstitial matrix of the parent vessel are dissolved by matrix metalloproteinases.1 Endothelial cells and pericytes migrate into the space created by matrix dissolution, a lumen is formed within the endothelial sprout, and loops are created via fusion of individual sprouts.1 During the stabilization phase, cell replication stops, the basement membrane is reformed and the new vessel is invested with pericytes, which inhibit endothelial cell proliferation.10
Splitting angiogenesis involves the placement of cellular columns into the lumina of pre-existing vessels. Growth and stabilization of these columns lead to partitioning and remodelling of the local vascular network.1
Angiogenesis is abundant during brain development and seems to be driven by the metabolic demands of the expanding neuroectoderm.11 Capillary sprouts migrate into the neuroectoderm from the extracerebral vascular plexus, towards the periventricular matrix zone, in a wave-like pattern.12 These vessels seem to penetrate the developing cerebrum in response to soluble angiogenic factors produced by the neuroectoderm. VEGF is likely to have a prominent role in this process.13
In adults, angiogenesis is tightly down-regulated in the normal adult brain by balancing stimulators and inhibitors of the process.1 Normally, inhibitors predominate, and vascular quiescence is maintained. The balance is shifted in favour of neovascularization in the presence of brain hypoxia, ischaemia and neoplastic disease.14
Arteriogenesis refers to the enlargement of pre-existing collateral arterioles to form larger arteries.1 It serves to redirect blood flow to areas of higher demand and to maintain flow after arterial occlusion. Vessel growth in this manner is an active process, requiring endothelial cell and smooth muscle proliferation.1 Unlike angiogenesis, which is primarily modulated by angiogenic factors such as VEGF, arteriogenesis is dependent on inflammatory mediators and monocytes.
Stimulators of angiogenesis
A number of endogenous angiogenic factors have been identified (Table 1).1,2 The VEGF family and the angiopoietins are endothelium-specific angiogenic factors. Other angiogenic factors are growth factors or cytokines that have angiogenic properties in addition to other functions.
|VEGF family members||Stimulate angio/vasculogenesis, permeability, |
|VEGFR-1, soluble VEGFR-1, soluble NRP-1||Sink for VEGF, VEGF-B, PlGF|
|VEGFR, NRP-1, NRP-2||Integrate angiogenic and survival signals||Ang 2||Antagonist of Ang 1|
|EG-VEGF||Stimulate growth of endothelial cells |
derived from endocrine glands
|TSP-1,2||Inhibit endothelial migration, growth, |
adhesion and survival
|Ang 1 and Tie 2||Stabilize vessels||Angiostatin and related plasminogen kringles||Inhibit endothelial migration and survival|
|PDGF-BB and receptors||Recruit smooth muscle cells||Endostatin (collagen XVIII fragment)||Inhibit endothelial survival and migration|
|TGF-β1, endoglin||Stimulate extracellular matrix||Tumstatin (collagen IV fragment)||Inhibit endothelial protein synthesis|
|TGF-β receptors FGF, HGF, MCP-1||Production stimulate angio/arteriogenesis||Vasostatin, calreticulin||Inhibit endothelial growth|
|Integrins αvβ3, αvβ5, α5β1||Receptors for matrix macromolecules and proteinases||Patelet factor-4||Inhibit binding of bFGF and VEGF|
|VE-cadherin, PECAM (CD31)||Endothelial junctional molecules||Tissue-inhibitors of MMP, MMP-inhibitors; PEX||Suppress pathologic angiogenesis|
|Ephrins||Regulate arterial/venous specification||Meth-1, Meth-2||Inhibitors containing MMP-, TSP-, |
|Plasminogen activators, MMPs||Remodel matrix, release growth factor||IFN-α, -β, -γ IP-10, IL-4, IL-12, IL-18||Inhibit endothelial migration, down-regulate bFGF|
|PAI-1||Stabilize nascent vessels||Prothrombin kringle-2, anti-thrombin III fragment||Suppress endothelial growth|
|NOS, COX-2||Stimulate angiogenesis and vasodilation||16 kDa prolactin||Inhibit bFGF/VEGF|
|AC133||Regulate angioblast differentiation||VEGI||Modulate cell growth|
|Chemokines||Pleiotropic role in angiogenesis||Fragment of SPARC||Inhibit endothelial binding and activity of VEGF|
|Id1/Id3||Inhibit differentiation||Osteopontin fragment maspin canstatin, |
proliferin-related protein, restin
|Interfere with integrin-signalling protease inhibitor |
The VEGF family
VEGF is one of the principal regulators of angiogenesis in the brain.15,16 Five VEGF homologues have been identified [VEGF-B, VEGF-C, VEGF-D, VEGF-E and placenta growth factor (PlGF)]; along with VEGF, they constitute the VEGF family of angiogenic factors (Fig. 2).17 There are four isoforms of VEGF (VEGF-121, VEGF-165, VEGF-189 and VEGF-206), which arise from alternative splicing of mRNA.16 The functions of the various isoforms have not been fully elucidated; VEGF-165 is the most abundant isoform in the brain.18
VEGF is a relatively specific mitogen and chemotactic factor for endothelial cells and a survival factor necessary for sustaining new blood vessels.19,20 The effects of VEGF on endothelial cells are dose dependent, and this has been demonstrated by the fact that over-expression or exogenous administration of VEGF increases blood vessel density.21 VEGF stimulates endothelial nitric oxide synthase, resulting in the generation of nitric oxide and activation of the angiogenic cascade.22,23 In addition, VEGF induces endothelial cell production of proteases which are necessary for the degradation of basement membrane during angiogenesis and serves as a survival factor for endothelial cells. VEGF promotes microvascular permeability to plasma proteins at the level of the small capillaries and venules and is 50 000 times more potent than histamine in this regard.24 Increased vessel permeability results from the direct effects of VEGF on endothelial cells, through mobilization of endothelial cytosolic calcium, and increases in fenestrae and pinocytic vesicles.25 The permeability effects of VEGF can be modulated by other angiogenic factors in vivo.
The homologues of VEGF have diverse roles. VEGF-B exhibits approximately 44% sequence identity to VEGF and is expressed primarily in skeletal and cardiac muscle.1 It has recently been found to be constitutively expressed in normal brain vessel endothelium, which suggests that it may have a role in the maintenance of the blood–brain barrier. VEGF-D is a mitogen for endothelial cells and has been found to be up-regulated in high-grade gliomas as is PlGF, suggesting that they both may play a role in tumour angiogenesis.1
The effects of VEGF are mediated by specific tyrosine kinase receptors that are expressed selectively, but not exclusively, on the endothelial cell surface.16 There are three known VEGFRs: VEGFR-1, VEGFR-2 and neuropilin-1 (VEGFR-3) (Fig. 2).16 Although VEGFR-1 is necessary for normal vascular development, it seems to be responsible for endothelial cell morphogenesis and to regulate rather than to promote endothelial cell proliferation.16 VEGFR-2 is structurally similar to VEGFR-1, is considered to be the primary VEGFR form in endothelial cells, is necessary for normal vascular development and mediates endothelial cell proliferation via activation by VEGF.18,26 VEGFR-3 is involved in vascular remodelling. Normal brain parenchyma expresses very low levels of VEGFR-1 and -2, but their expression is increased after ischaemia and, in humans, with gliomas.27,28
The expression of VEGF is regulated by numerous factors.1,16 Potent stimulators of VEGF expression include insulin, growth factors and cytokines including platelet-derived growth factor (PDGF), epidermal growth factor, tumour necrosis factor-α (TNF-α), interleukin-1β (IL-1β), IL-6 and nitric oxide.16 Induction of VEGF expression by hypoxia is probably the most significant of these.16 Hypoxia promotes VEGF production at the transcriptional and translational levels. The transcriptional component of this mechanism is mediated by hypoxia-inducible factor-1α (HIF-1α), which accumulates under hypoxic conditions and binds to a VEGF gene hypoxia-response element.29 VEGF is secreted by tumour cells and tumour stromal cells in response to hypoxia.30 Over-expression of VEGF can enhance tumour invasive potential and is associated with poor prognosis in several solid tumours.31 Brain tumours are often associated with surrounding oedema secondary to increased vascular permeability, which itself is associated with VEGF up-regulation causing enhanced vessel permeability without causing cell injury.32 Dexamethasone, used to treat peritumoral oedema, can down-regulate VEGF expression in in vitro models.33
The angiopoietins and Tie-receptors
The angiopoietins are ligands for the endothelium-specific receptor tyrosine kinase Tie-2.1,2 Tie-1 and Tie-2 are tyrosine kinase receptors expressed exclusively on endothelial cells. Tie-1 appears to have a role in the maintenance of vascular integrity.20 Although four members of the angiopoietin family have been cloned, angiopoietin-1 (Ang-1) and Ang-2 are the best characterized, being involved primarily in vessel remodelling, maturation and stabilization and working synergistically with VEGF to increase vessel size and number.1,14 Ang-1 is a ligand for Tie-214 and is involved in pericyte and smooth muscle cell recruitment and in stabilization of new vasculature.34,35 Ang-2 shares approximately 60% sequence identity with Ang-1 and competes with Ang-1 for Tie-2 and antagonizes its action.34 By antagonizing Ang-1, Ang-2 may function to restrain Ang-1-mediated angiogenesis, preventing excessive branching and sprouting of vessels.36 When co-expressed with VEGF, Ang-2 facilitates endothelial cell proliferation and migration, suggesting that Ang-1 may also contribute to angiogenesis by destabilizing vessels in preparation for vessel growth.37 In adults, expression of the angiopoietin system occurs physiologically during ovarian angiogenesis and occurs in the brain after ischaemia.
Tie-1 expression is increased in the endothelium of gliomas and meningiomas in humans, but neither Ang-1 nor Ang-2 is detected in normal brain.38 In human gliomas, Ang-1, Ang-2 and Tie-2 are up-regulated; Tie-2 is expressed in glioma endothelium, while Ang-1 mRNA is found in glioma cells.38 Ang-2 is expressed both in endothelial and tumour cells and is prominent in the perinecrotic areas and at the periphery of the tumour. Ang-2 expression has also been found to increase with increasing glioma grade.39
The fibroblast growth factor (FGF) family
The FGFs are a family of broad-spectrum growth factors that influence a wide variety of cellular activities. The FGF family now includes 23 polypeptides and have roles in the development of the skeletal, nervous and vascular systems, in wound healing and in many pathological conditions.1,16 Two FGFs, FGF-1 (acidic FGF) and FGF-2 (bFGF), are potent stimulators of endothelial cell migration, proliferation, sprouting and tube formation.16 Acidic FGF and bFGF and their receptors are widely distributed within the CNS.40
In contrast to VEGF, which is mitogenic specifically for endothelial cells, the FGFs stimulate proliferation of most cells derived from embryonic mesoderm and neuroectoderm, including fibroblasts, smooth muscle cells, pericytes and osteoblasts.5 They up-regulate expression of HIF-1α, VEGF and VEGFR-2 and stimulate VEGF secretion in glioma cell lines.41 bFGF also up-regulates expression of the serine protease urokinase-type plasminogen activator (uPA) and its receptor.42
However, despite wide-ranging effects on vessels, the FGFs do not seem to have an important role in the development of the vascular system, but they do seem to be important for maintenance of the adult vascular system and recovery from injury.2 The FGFs seem to have prominent roles in cerebral ischaemia and some of the families are associated with glioma-associated angiogenesis. Tumour angiogenesis has been associated with increased bFGF levels in the serum and urine.43
Platelet-derived growth factor
PDGF, an endothelial cell mitogen, was originally purified from platelets and seems to be the major mediator of angiogenic activity for those cells.1 The PDGF system participates in the development and maturation of the CNS and is expressed by neurones and astrocytes during embryogenesis and in the adult brain.44,45 PDGF promotes angio- genesis, mediates the proliferation and migration of vascular smooth muscle cells and pericytes and enhances the expression of VEGF.46,47 Expression of the PDGF system is induced by ischaemia and by other angiogenic factors such as TNF-α.1 PDGF also seems to have a role in glioma-associated and meningioma-associated angiogenesis.2,48
Inhibitors of angiogenesis
In the normal brain, angiogenesis is tightly down-regulated.1 The precise mechanism by which angiogenesis is suppressed in the CNS is not yet clear. Over 30 endogenous molecules with anti-angiogenic properties have been identified. Some molecules seem to function primarily as anti-angiogenic factors such as thrombospondin-1 (TSP-1) and TSP-2, released by activated platelets, endothelial cells and fibroblasts (Table 1).1 Several endothelial cell inhibitors are proteolytic fragments of factors with functions unrelated to angiogenesis. For instance, angiostatin is an amino terminal fragment of plasminogen, and endostatin is a carboxy-terminal fragment of collagen XVIII.49
First identified in 1994, angiostatin is a 38-kDa fragment of plasminogen generated primarily by uPA-mediated proteolysis.50 It inhibits endothelial cell migration, proliferation and lumen formation.51 Angiostatin exerts its anti-angiogenic activity by blocking angiogenic factors and by direct effects on endothelial cells.52 It induces endothelial cell apoptosis and inhibits plasminogen activation.51,53 Angiostatin serum levels have been found to decrease following primary tumour removal in humans.16
Endostatin is a fragment of collagen XVIII present in the basement membrane of blood vessels. It inhibits endothelial cell proliferation and migration and induces endothelial cell apoptosis.54,55 As with angiostatin, serum levels of endostatin decrease after primary tumour removal.
There are four isoforms of TSP released by activated platelets, endothelial cells and fibroblasts.16 TSP-1 and TSP-2 inhibit endothelial cell proliferation in vitro and angiogenesis and tumour growth in vivo.16 TSP-1 also impacts endothelial cell migration and morphogenesis and its expression is regulated by p53, the loss of which results in its down-regulation.56 Although TSP-1 is expressed in normal brain and in low-grade gliomas, it is absent in glioblastoma multiforme.57
Angiogenic factors in cerebral neoplasia
Angiogenic factors have been implicated in a number of neurological disorders. Increased expression of VEGF has been linked to such diverse diseases such as cerebral vasospasm, Alzheimer's disease, atherosclerosis, bacterial meningitis, traumatic brain injury and motor neurone disease.1,2 Most work on angiogenic factors in neurological disorders has focused on ischaemia and brain tumours.
Solid tumours are dependent on angiogenesis for growth beyond 2 mm in diameter.1,58 As long as the tumour is supported by nutrients through diffusion from the host vasculature, there is a balance between cell proliferation and death, and tumour size is smaller than a few cubic millimetres. This is called the avascular phase of a tumour. Once the tumour cells are able to induce neovascular formation, neoangiogenesis starts and the tumour switches towards a vascular phase with rapid growth, intensified invasion and increased metastatic potential.59 Neovascularization is a feature of many neoplasms in the CNS, and it has been established that brain tumours are able to induce angiogenesis in surrounding normal brain tissue.1,16 Several mechanisms are responsible for starting the switch to tumour angiogenesis. The acquisition of the angiogenic phenotype is mediated by a balance between positive and negative regulators of microvessel growth.59 In contrast to normal vessels, tumour vessels are tortuous, with increased vessel diameter, length, density and permeability. Some tumour vessels are proposed to consist of extracellular matrix channels surrounded by tumour cells (Fig. 3) and without any participation by endothelial cells.60 This concept, known as vasculogenic mimicry, has gained credibility from recent evidence but is still quite controversial.61 Although vasculogenic mimicry has been described in uveal and cutaneous melanomas,62 in which it indicates a poor prognosis, it has not been described in cerebral neoplasia.
Peritumoural brain oedema seems to be directly related to the expression of angiogenic factors by brain tumours.1 Angiogenic factors are thought to have a role in the progression of metastatic brain tumours, pituitary adenomas and craniopharyngiomas in humans.63,64 Most investigations of angiogenic factors in brain tumours have focused on gliomas and meningiomas as well as the oedema common to those tumours.
Glioblastomas are among the most highly vascularized solid tumours16 with the vasculature not only florid but also abnormal.65 Endothelial cell proliferation in high-grade gliomas is 40 times greater than that in normal brain tissue.12 In addition, glioblastomas are able to induce angiogenesis in previously quiescent vessels in the surrounding brain tissue.1 The angiogenic factors that have been implicated in glioma angiogenesis are VEGF and FGF, although most data indicate that VEGF is the primary mediator.16 Evidence also indicates that VEGF is necessary for tumour vessel maintenance.16
VEGF expression correlates with the extent of tumour vascularity and the glioma grade.46 Additionally, it has recently been shown that cerebrospinal fluid (CSF) VEGF levels are associated with the degree of glial tumour vascularity and are adversely associated with patient survival.66 Although VEGFR-1 expression is present in the vascular cells of both low- and high-grade gliomas, VEGFR-2 expression is limited to high-grade gliomas, which may explain the absence of neovascularization in the low-grade tumours.27 Moreover, VEGFR-1 levels are increased in high-grade gliomas compared with that in low-grade gliomas.27 Expression of VEGFRs may be induced by VEGF itself via a paracrine mechanism, in which VEGF secreted by tumour cells promotes the expression of VEGFRs on endothelial cells.2 Therefore, progression from low-grade astrocytomas to high-grade highly vascularized glioblastomas may involve an ‘angiogenic switch’ in which the VEGF system is fully activated.23 Up-regulation of the VEGF system in glioblastomas may result from several factors. Fast-growing tumour cells create a local hypoxic environment that seems to activate VEGF expression via HIF-1α.67 Of note, both VEGF and HIF-1α expression are restricted to the pseudo-palisading cells around the areas of necrosis in glioblastoma multiforme.68 In addition, HIF-1α expression correlates with glioma grade and vascularity. However, expression of VEGF has been identified in experimental tumours much smaller than 1 mm3, indicating that hypoxia is not the only stimulus for VEGF expression in gliomas. Other factors found to be involved with such up-regulation include a mutant p53 gene, local metabolic derangements such as hypoglycaemia within the tumour, and non-angiogenic growth factors such as PDGF and EGF.69
Expression of acidic FGF and bFGF is increased in glioma cell lines and in histological sections of primary glioma tumours.70 bFGF levels appear to correlate with microvessel density and glioma grade, and similarly, FGF receptor expression is up-regulated in gliomas and increases with tumour grade.16
Meningiomas are characteristically highly vascular tumours, although regional heterogeneity in the degree of vascularity exists even within individual meningiomas.71 Much of the morbidity associated with intracranial meningiomas is related to the degree of tumour vascularity and the extent of peritumoural vasogenic oedema.59 Up-regulation of VEGF and increased expression of PlGF and VEGF-B have been observed in meningiomas.59 VEGF mRNA and protein levels are correlated with meningioma vascularity, and high levels of VEGF expression have been observed to predict meningioma recurrence after resection in humans.2,72 However, VEGF expression does not seem to correspond to meningioma size or degree of malignancy in humans.59,71 This is contradicted in other studies,73 but it is on this basis that it is suggested that an angiogenic switch involving VEGF does not exist in meningiomas in humans as it does in gliomas. Among different subtypes of benign meningiomas, VEGF levels have been found to be 10-fold higher in meningothelial than in fibrous meningiomas.73 Canine meningiomas also express high levels of VEGF and preliminary work confirms an association with outcome following surgical resection and radiotherapy.74 The level of VEGF expression may indicate the degree of malignancy in canine meningiomas, but it does not seem to be associated with cyclooxygenase-2 expression, which is an independent indicator of malignancy in human meningiomas.
The growth of meningiomas seems less dependent on the up-regulation of VEGF by hypoxia and tumour necrosis and may be corticosteroid dependent, with some evidence indicating that VEGF expression may be related to stimulation of the oestrogen/progesterone system. Alternatively, over-expression of EGF and PDGF, which is also observed in meningiomas, may promote VEGF expression.49
A common feature of malignant brain tumours is their ability to disturb the blood–brain barrier and to increase capillary permeability, which subsequently leads to vasogenic brain oedema. Peritumoural oedema (PTO) occurs variably (50–92%) in meningiomas.71,75 Some meningiomas have a disproportionately large amount of PTO, the mechanisms of which are not fully explained. It seems that the size and histological subtype of the meningioma have little effect on the PTO.71 However, one study demonstrated a significant association between neuroanatomical location of the meningioma and the severity of the PTO.72 Meningiomas of the frontal lobe or the frontotemporal base in humans tend to have more extensive PTO, whereas meningiomas of the occipital lobes or posterior fossa have little or no PTO. Such associations have not been investigated in veterinary medicine.
Several studies demonstrated that the increased vascular permeability in brain tumours is associated with morphological alterations in tumour capillaries.76 Although the morphological alterations are certainly part of the pathophysiological mechanism involved in the genesis of brain oedema, there is an increasing amount of biochemical research examining the role of individual compounds that are involved with tumour oedema. Peritumoural oedematogenesis is likely to be a multifactorial process. VEGF seems to play a critical role; VEGF expression is correlated with vascular permeability in human gliomas in situ, and a strong correlation has also been found between the amount of PTO and VEGF expression in meningiomas.15 Other related co-factors and biochemical vasoactive agents may work in concert with VEGF to affect PTO formation. Specifically, it has been postulated that Matrix metalloproteases (MMPs) and their inhibitors have some roles in PTO formation of meningiomas.71 The author is currently investigating whether an association exists between PTO and VEGF expression in canine meningiomas.
The concept of anti-angiogenic therapy for brain tumours is fundamentally different from other chemotherapeutic approaches, because it targets non-malignant vessel cells rather than neoplastic tumour cells.77 Solid tumours are dependent on vascular support for growth, and inhibition of angiogenesis effectively controls the growth of gliomas and metastatic brain tumours in experimental models.77 Because angiogenesis is virtually absent in normal adults, therapies aimed at specifically interrupting angiogenesis within tumours should be well tolerated.77 Unlike the situation with chemotherapeutic agents that target tumour cells, the blood–brain barrier is not an obstacle, because blood-borne anti-angiogenic factors can reach the endothelial cells directly.77
The identification of a number of angiogenesis inhibitors in recent years has provided the foundation for laboratory studies and clinical trials of anti-angiogenic therapy for intracranial tumours.78 The targets for anti-angiogenic therapy include angiogenic factors and their receptors, endothelial cells and vascular smooth muscle cells; therefore, multitargeted regimens may be possible. Because anti-angiogenic therapy involves a cytostatic strategy, it may be most effective when used in combination with cytotoxic chemotherapeutic regimens, surgical resection or radiotherapy.49,77 Additionally, once the anti-angiogenic effect is eliminated, tumour growth suppression might no longer be expected. Therefore, long-term therapy could be required for any anti-angiogenic drug.49,77 Long-term therapy, however, could interfere with normal physiologic angiogenic processes including wound healing.
Angiostatin and endostatin
These two proteins are endogenous inhibitors of angiogenesis. Endostatin is derived as a cleavage product of collagen XVIII, and angiostatin is cleaved from plasminogen. Their mechanisms of action are not well elucidated but seem to include inhibition of extracellular proliferation for endostatin and inhibition of extracellular migration and increased extracellular apoptosis for angiostatin.79 The two have been the focus of a great deal of attention and hope because of promising preclinical results.79 Endostatin achieved complete responses in several animal models, and re-treatment showed that resistance did not develop.80 Systemic administration of endostatin to rats with subcutaneous or intracranial gliomas results in decreased angiogenesis, decreased tumour growth and prolonged survival.81 Systemic administration of angiostatin blocks angiogenesis and growth of metastasis in mice with Lewis-lung cell carcinoma and is effective in a variety of tumour models.16 Clinical development of angiostatin is presently underway, as phase I trials have demonstrated safety, although only minimal responses have been achieved.82 No trial of endostatin specifically for human patients who have brain tumours is underway at this time, although preclinical studies of endostatin for brain tumours showed some promising results with novel methods of local delivery or viral delivery.83,84
As VEGF is one of the critical regulators of brain tumour angiogenesis, several agents are being developed that target its signalling pathway. Systemic administration of anti-VEGF antibodies results in decreased tumour growth and vessel density in nude mice with subcutaneous human glioma xenografts.85 Similarly, anti-VEGF antibody treatment reduces the vascular permeability of gliomas in human xenograft models.16 Inhibition of VEGF with anti-sense oligonucleotides results in decreased tumour growth and vessel density of subcutaneously or intracerebrally implanted C6 glioma cells.86 A phase I trial of a recombinant human monoclonal antibody to VEGF in patients with metastatic cancer found it to be tolerated well.87 Phase III studies are ongoing.
Inhibitors of VEGFRs are promising agents in malignant gliomas, with the potential to inhibit angiogenesis and tumour growth as well as reduce PTO.88 PTK787/ZK222584, an inhibitor of VEGFR tyrosine kinases, decreases glioma growth and vascularization in vivo17and is undergoing phase I/II studies in malignant gliomas, alone or in combination with lomustine or temozolamide.88
Matrix metalloprotease inhibitors
MMPs play a central role in breaking down the extracellular matrix, a key step in angiogenesis, and have been a focus of therapeutic attempts for many years. The development of batimastat, one of the earliest candidates, was terminated in favour of newer compounds, with better oral bioavailability, such as marimastat and solimastat.79 To date, clinical trials with these MMP inhibitors have been disappointing, with dosage limited by significant joint and muscle pain. A phase III trial of marimastat for patients who had newly diagnosed glioblastoma multiforme (GBM) demonstrated no improvement in progression-free or overall survival.79
Thalidomide is an inhibitor of TNF-α, but its precise mechanism of anti-angiogenic activity has not been elucidated.79 Thalidomide has been tested in phase II trials in humans for several malignancies, including gliomas, for which its use has shown some partial responses.79 When used in combination with 1,3-bis-(chloroethyl)-1-nitrosourea (BCNU) for glioma treatment, a synergistic effect was demonstrated.79 Trials with non-teratogenic thalidomide analogues are currently underway in human brain tumour patients.
Local delivery of anti-angiogenic therapy
Local delivery of anti-angiogenic therapy represents an attractive possibility for the treatment of brain tumours. Although anti-angiogenic therapies are often stated to be non-toxic, most do have dose-limiting toxicities to various organs.79 Therefore, it may be possible to achieve greater effect and less systemic toxicity through local delivery to the brain. Localized anti-angiogenic therapies could consist of a single delivery of the agent itself, but doubtless will be more effective if delivery is prolonged. One strategy for this consists of viral vector delivery of an anti-angiogenic protein; this has been successful in animal models of brain tumours.79 Delivery of polymer-protected-producing cells also has been applied in brain tumour models.84 The author is currently investigating the use of cell delivery systems and anti-angiogenic factors in dogs with cerebral neoplasia.
Angiogenesis and the veterinary experience
Currently, there is no published literature demonstrating the effect of anti-angiogenic therapy in dogs or cats with cerebral neoplasia. Recent work has highlighted increased levels of VEGF expressed in canine meningiomas and has demonstrated an association with patient survival.74 However, ongoing work at the author's institution includes investigating the association of VEGF expression and PTO in canine meningiomas and treatment trials using locally delivered cell-encapsulated anti-angiogenic therapy in dogs with cerebral neoplasia.
Investigations into the role of angiogenesis in the pathogenesis of other tumours in dogs is at a more advanced stage. The relationship among angiogenesis, biological behaviour and histological parameters of malignancy in canine cutaneous mast cell tumours,89 canine cutaneous squamous cell carcinomas,90 canine seminomas,91 canine mammary tumours92,93 and feline invasive mammary carcinomas94 has been found to be consistent with previous human neoplasia findings. Dogs with systemic neoplasias have been documented with significantly higher pre-treatment plasma VEGF than normal dogs,95 and preliminarily unsuccessful attempts have been made to investigate the value of these plasma VEGF measurements during radiation therapy of dogs.95 Serum endostatin concentrations have been demonstrated to be elevated in dogs with lymphoma and dogs with haemagiosarcoma in comparison with normal dogs, serving as a useful marker in dogs with these diseases.96 Urine angiostatin presence has been documented in dogs with spontaneous bone and prostate cancers and was shown to disappear after complete surgical removal of the primary tumour.97 Serum concentrations of matrix metalloproteinases from tumour-bearing cats have been found to be higher than those seen in normal cats;98 however, poor correlation has been found between these serum levels and the tissue concentrations of these enzymes in increasing histologic grades of malignancy.98
Anti-angiogenic therapy has been investigated in a xenotransplantation model of canine osteosarcoma in athymic mice using thalidomide.99 Unfortunately, no effect was proven at the doses of thalidomide used.
Vascular development, maintenance and repair in the CNS depend on complex interactions among angiogenic factors, other growth factors and the biological environment. Although VEGF occupies a central role in these processes, it is becoming increasingly clear that other angiogenic factors modulate the effects of VEGF and are equally important. Many opportunities exist for the development of new therapeutic strategies based on angiogenic factors. Anti-angiogenic therapy continues to be a promising approach in neuro-oncology, as clinical evidence accumulates to support the inhibition of angiogenesis.
Support for this work in part was funded by EU 5th foundation grant QLG1-CT-2000-00815.