INSIGHT REVIEW
`
`NATURE|Vol 438|15 December 2005|doi:10.1038/nature04483
`
`Angiogenesis as a therapeutic target
`
`Napoleone Ferrara1 & Robert S. Kerbel2
`
`Inhibiting angiogenesis is a promising strategy for treatment of cancer and several other disorders,
`including age-related macular degeneration. Major progress towards a treatment has been achieved
`over the past few years, and the first antiangiogenic agents have been recently approved for use in several
`countries. Therapeutic angiogenesis (promoting new vessel growth to treat ischaemic disorders)
`is an exciting frontier of cardiovascular medicine, but further understanding of the mechanisms of
`vascular morphogenesis is needed first.
`
`Early pioneers of angiogenic research observed over a century ago that
`the growth of human tumours is often accompanied by increased vas-
`cularity. They suggested that a key aspect of the cancer process is a dis-
`ease of the vasculature in the whole area affected (reviewed in ref. 1).
`The existence of tumour-derived factors responsible for promoting
`new vessel growth was postulated over 65 years ago2, and a few years
`later it was proposed that tumour growth is crucially dependent on the
`development of a neovascular supply3. In 1971, it was hypothesized
`that inhibition of angiogenesis (antiangiogenesis) would be an effec-
`tive strategy to treat human cancer, and an active search for angiogen-
`esis inducers and inhibitors began4. Extensive research has led to the
`identification and isolation of several regulators of angiogenesis, some
`of which represent therapeutic targets.
`Despite some initial setbacks and negative clinical trial results,
`major progress has been made over the past few years in targeting
`angiogenesis for human therapy. In February 2004, the US Food and
`Drug Administration (FDA) approved bevacizumab, a humanized
`anti-VEGF (vascular endothelial growth factor)-A monoclonal anti-
`body, for the treatment of metastatic colorectal cancer in combination
`with 5-fluorouracil (FU)-based chemotherapy regimens. This fol-
`
`BMC
`
`PDGF–B
`
`c
`
`Pericyte
`
`Capillary bud
`
`Endothelial cells
`
`Other angiogenic
`factors
`such as bFGF
`
`VEGF-A
`
`BMC
`
`BMC
`
`VEGF-A
`
`Other angiogenic
`factors
`
`a
`
`Tumour cells
`
`SDF-1
`
`HGF
`TGF-α
`EGF
`
`PDGF-A
`PDGF-C
`TGF-β
`
`b
`Stromal cells
`
`lowed from a phase III study showing a survival benefit5. In December
`2004, the FDA approved pegaptinib, an aptamer that blocks the 165
`amino-acid isoform of VEGF-A, for the treatment of the wet (neovas-
`cular) form of age-related macular degeneration (AMD)6.
`These achievements have validated the notion that angiogenesis
`is an important target for cancer and other diseases. These advances
`notwithstanding, much progress is needed on a variety of important
`issues; for example, how do we achieve the most effective combina-
`tions of antiangiogenic agents with chemotherapy or other biologi-
`cal agents and how do we select patients that are most likely to
`respond to the treatment? Another issue is that resistance to antian-
`giogenic therapy is emerging7 and thus a better understanding of
`pathways that may mediate tumour angiogenesis in various cir-
`cumstances is necessary. Furthermore, the hope that ‘therapeutic
`angiogenesis’ will provide a treatment for ischaemic disorders still
`remains unfulfilled, in spite of considerable preclinical and clinical
`efforts.
`The main purpose of this review is to summarize recent progress and
`emphasize the issues that need to be resolved before the field of angio-
`genic therapy can make further significant advances.
`
`Figure 1 | A few of the molecular and cellular players in the
`tumour/microvascular microenvironment. a, Tumour cells produce VEGF-
`A and other angiogenic factors such as bFGF, angiopoietins, interleukin-8,
`PlGF and VEGF-C. These stimulate resident endothelial cells to proliferate
`and migrate. b, An additional source of angiogenic factors is the stroma.
`This is a heterogeneous compartment, comprising fibroblastic,
`inflammatory and immune cells. Recent studies indicate that tumour-
`associated fibroblasts produce chemokines such as SDF-1, which may
`recruit bone-marrow-derived angiogenic cells (BMC). The various
`hypotheses on the nature and role of such cells in angiogenesis and tumour
`progression are discussed in the text. VEGF-A or PlGF may also recruit
`BMC. Tumour cells may also release stromal cell-recruitment factors, such
`as PDGF-A, PDGF-C or transforming growth factor (TGF)-ȋ. A well-
`established function of tumour-associated fibroblasts is the production of
`growth/survival factor for tumour cells such as EGFR ligands, hepatocyte
`growth factor and heregulin. c, Endothelial cells produce PDGF-B, which
`promotes recruitment of pericytes in the microvasculature after activation
`of PDGFR-ȋ. HGF, hepatocyte growth factor.
`
`1Genentech, 1 DNA Way, South San Francisco, California 94080, USA; 2Sunnybrook and Women’s College Health Sciences Centre and the University of Toronto, Ontario M5G 2M9, Canada.
`
`© 2005
`
`Nature Publishing Group
`
`© 2005 Nature Publishing Group
`
`967
`
`CELLTRION - EXHIBIT 1046
`
`

`

`INSIGHT REVIEW
`
`NATURE|Vol 438|15 December 2005
`
`Chemotherapy drug targets
`
`Tumour cells
`
`Bone-marrow
`progenitors
`
`Hair follicle
`cells
`
`Epithelial
`mucosal
`cells
`
`Tumour shrinkage/
`response
`
`Myelo-suppression,
`for example,
`neutro-penia
`
`?
`
`Drop in pro-
`angiogenic
`monocytes
`and pericyte
`precursors
`
`a
`
`Antiangiogenic
`effect
`
`Alopecia
`
`Mucositis
`
`Dividing
`endothelial
`cells in
`sprouting
`vessels
`
`Endothelial
`progenitor
`cells (EPCs)
`in bone marrow
`or peripheral
`blood
`
`b
`
`c
`
`Antiangiogenic
`effect
`
`Antiangiogenic
`effect
`
`Figure 2 | Chemotherapy targets. In addition to tumour cells, the intended target for chemotherapy in cancer patients, conventional chemotherapy drugs
`can inhibit the proliferation of, or kill, a number of normal host cell types, including several that, in principle, can contribute to an antiangiogenic effect.
`Targeting of various normal cell populations is generally associated with harmful or undesirable side effects such as myelosuppression, alopecia or
`mucositis. A desirable effect could be antiangiogenesis as a result of targeting. a, Bone-marrow-derived proangiogenic cells that adhere to the walls of new
`blood vessels and further stimulate their growth by paracrine mechanisms. Whether these latter cell types, which probably include monocytes and pericyte
`precursors, are affected directly by chemotherapy or are reduced in numbers by elimination of more primitive bone marrow progenitors which give rise to
`such cells is not yet clearly established. b, Cycling endothelial cells present in sprouting blood vessel capillaries; and c, authentic bone-marrow-derived
`circulating endothelial progenitor cells (EPC) that can incorporate into the lumen of growing vessels and differentiate into endothelial cells. Inhibiting the
`levels or function of VEGF can augment these various antiangiogenic mechanisms of chemotherapy. For example, VEGF is a potent mobilizer of EPC, a pro-
`survival (anti-apoptotic) factor for differentiated, activated endothelial cells, and also may be one of the more important paracrine growth factors secreted
`by proangiogenic vessel adherent bone-marrow-derived monocytes.
`
`The major signalling pathways in tumour angiogenesis
`VEGF/VEGF receptors
`Angiogenesis is a fundamental developmental and adult physiological
`process, requiring the coordinated action of a variety of growth factors
`and cell-adhesion molecules in endothelial and mural cells (reviewed in
`this issue by Coultas, Chawengsaksophak and Rossant, p. 937). So far,
`VEGF-A and its receptors are the best-characterized signalling pathway
`in developmental angiogenesis1,8,9. Loss of a single VEGF-A allele results
`in embryonic lethality1,8,9. This pathway also has an essential role in
`reproductive and bone angiogenesis8. Much research has also estab-
`lished the role of VEGF-A in tumour angiogenesis8,10. VEGF-A binds to
`two receptor tyrosine kinases (RTK), VEGFR-1 (Flt-1) and VEGFR-2
`(KDR, Flk-1) (reviewed in ref. 10). Of the two, it is now generally agreed
`that VEGFR-2 is the major mediator of the mitogenic, angiogenic and
`permeability-enhancing effects of VEGF-A. The significance of
`VEGFR-1 in the regulation of angiogenesis is more complex. Under
`some circumstances, VEGFR-1 may function as a ‘decoy’ receptor that
`sequesters VEGF and prevents its interaction with VEGFR-2 (ref. 10).
`However, there is growing evidence that VEGFR-1 has significant roles
`in haematopoiesis and in the recruitment of monocytes and other bone-
`marrow-derived cells that may home in on the tumour vasculature and
`promote angiogenesis11–13. In addition, VEGFR-1 is involved in the
`induction of matrix metalloproteinases (MMPs)14 and in the paracrine
`release of growth factors from endothelial cells15. Thus the VEGFR-1-
`selective ligands VEGF-B and placental-like growth factor (PlGF) may
`also have a role in these processes. Furthermore, in some cases VEGFR-
`1 is expressed by tumour cells and may mediate a chemotactic signal,
`thus potentially extending the role of this receptor in cancer growth16.
`VEGF-A gene expression is upregulated by hypoxia17. The tran-
`scription factor hypoxia inducible factor (HIF), which operates in con-
`cert with the product of the von Hippel–Lindau (VHL) tumour
`suppressor gene, has a major role in such regulation. Under normoxic
`conditions, the VHL protein targets HIF for ubiquitination and degra-
`dation17.
`In situ hybridization studies demonstrate that VEGF-A messenger
`
`RNA is expressed in many human tumours18. Renal cell carcinomas
`have a particularly high level of VEGF-A expression, consistent with
`the notion that inactivating VHL mutations occur in about 50% of
`such tumours19, thus providing a further explanation for the respon-
`siveness of this tumour type to a VEGF-A blockade20. However,
`VEGF-A upregulation in tumours is not only linked to hypoxia or
`VHL mutations. Indeed, a very broad and diverse spectrum of onco-
`genes is associated with VEGF-A upregulation, including mutant ras,
`erbB-2/Her2, activated EGFR and bcr-abl7,21. Besides VHL, inactiva-
`tion/mutation of various other suppressor genes can also result in
`VEGF upregulation. These genes include those associated with famil-
`ial syndromes characterized by well-vascularized hamartomas22.
`In 1993, it was reported that a murine anti-human VEGF-A mono-
`clonal antibody inhibited the growth of several tumour cell lines in
`nude mice, whereas the antibody had no effect on tumour-cell prolif-
`eration in vitro23. Subsequent studies have shown that many additional
`tumour cell lines, regardless of the tumour’s origin, are inhibited in
`vivo by the same anti-VEGF monoclonal antibody (reviewed in ref.
`24). Tumour-growth inhibition has also been demonstrated using
`independent anti-VEGF approaches including a dominant-negative
`VEGFR-2 mutant25, anti-VEGFR-2 antibodies26, small molecule
`inhibitors of VEGF RTKs27 and soluble VEGF receptors28,29. VEGF-A
`gene inactivation also suppresses angiogenesis in a transgenic model
`of multi-stage tumorigenesis30.
`
`Platelet-derived growth factor (PDGF) and angiopoietins
`Other signalling molecules that have an established role in the devel-
`opment and differentiation of the vessel wall such as PDGF-
`B/PDGFR-ȋ31 and the angiopoietins (Ang), the ligands of the Tie2
`receptor9, may also be therapeutic targets. PDGF-B is required for
`recruitment of pericytes and maturation of the microvasculature31.
`Inhibition of PDGFR-ȋ signalling has been reported to result in a
`tumour microvascular tree that is particularly dependent on VEGF-
`mediated survival signals. Withdrawal of VEGF-A leads to endothe-
`lial apoptosis and vascular regression32. In this context, newly formed
`
`968
`
`© 2005
`
`Nature Publishing Group
`
`© 2005 Nature Publishing Group
`
`

`

`NATURE|Vol 438|15 December 2005
`
`INSIGHT REVIEW
`
`vessels, whether they are tumour-associated or not, are particularly
`vulnerable to VEGF-A blockade, whereas mature vessels, covered by
`extracellular matrix and pericytes, may be resistant to VEGF inhibitors
`and other antiangiogenic agents. Furthermore, recent studies have
`emphasized the significance of tumour-derived PDGF-A (and poten-
`tially PDGF-C) and PDGFR-Ȋ signalling in the recruitment of an
`angiogenic stroma that produces VEGF-A and other angiogenic fac-
`tors33 (Fig. 1). Therefore, combining PDGF and VEGF inhibitors is an
`attractive anti-vascular and anti-tumour strategy.
`Ang-1 is required for further remodelling and maturation of the ini-
`tially immature vasculature. Unlike mouse embryos lacking VEGF-A
`or VEGFR-2, embryos lacking Ang-1 or its receptor Tie2 develop a
`rather normal primary vasculature, but this vasculature fails to undergo
`effective remodelling (reviewed in ref. 9). The generally accepted view
`is that Ang-1 is the major agonist for Tie2, whereas Ang-2 may act as an
`antagonist or a partial agonist34. However, more recent evidence indi-
`cates that, unexpectedly, Ang-2 has a positive role, at least in tumour
`angiogenesis35. Administration of Ang-2 inhibitors to tumour-bearing
`mice has been reported to result in delayed tumour growth, accompa-
`nied by reduced endothelial cell proliferation, consistent with an
`antiangiogenic mechanism. Therefore, inhibitors of Ang-2 may be can-
`didates for clinical development35.
`
`Axon-guidance molecules
`Recently, the role of axon-guidance receptors and ligands in develop-
`mental angiogenesis has received much attention. There are four main
`families: the neuropilins (NRP)/semaphorins, the ephrins, Robo/Slit
`and netrin/Unc5. For recent reviews, see refs 36, 37. Although the sig-
`nificance of these pathways in tumour angiogenesis is far from clear,
`there is emerging evidence that they have a role in some cancer models
`and therefore may be potential therapeutic targets.
`NRP1 and NRP2, previously shown to bind the collapsin/sema-
`phorin family and implicated in axon guidance, are also receptors for
`the heparin-binding isoforms of VEGF-A and seem to potentiate the
`activation of VEGFR-2 by VEGF165 (ref. 38). Therefore, NRPs may
`participate in tumour angiogenesis as positive modulators of VEGF
`signalling in endothelial cells. Furthermore, NRP1 and NRP2 are
`expressed on the cell surface of several tumour cell lines that bind
`VEGF165 and display a chemotactic response to this ligand, suggest-
`ing a pro-tumour activity of NRPs, with or without the involvement of
`VEGF RTK signalling37.
`The ephrins and their tyrosine kinase Eph receptors are a large fam-
`
`ily, initially implicated in neuronal guidance during development and
`subsequently found to have activities in other cell types, including vas-
`cular cells (for a review see ref. 39). The earliest evidence for a role of
`this family in angiogenesis was the report by Pandey et al. that ephrin
`A1 mediates TNF-Ȋ-induced angiogenesis in vivo40. Ephrin B2 and its
`receptor EphB4 are important for distinguishing between developing
`arterial and venous vessels (see p. 937). Recent studies suggest a role for
`Eph/ephrin interactions in malignant tumour progression and angio-
`genesis. Soluble EphB4-expressing human melanoma A375 cells grown
`subcutaneously in nude mice showed reduced tumour growth com-
`pared with control tumours41. Interfering with EphA signalling has
`been also reported to result in some inhibition of angiogenesis in
`tumour models42.
`Slits are secreted proteins that function as chemorepellents in axon
`guidance and neuronal migration through the Roundabout (Robo)
`receptor (reviewed in refs 36, 37). Wang et al. reported the expression
`of Slit2 in several tumour cell types and that Robo1 expression was
`localized to vascular endothelial cells43. Recombinant Slit2 protein
`attracted endothelial cells and promoted tube formation. Neutraliza-
`tion of Robo1 reduced microvessel density and growth of A375 cells
`transplanted in nude mice43.
`
`Negative regulators of angiogenesis
`Angiogenesis is a tightly regulated process and seems to be under the
`control of both positive and negative regulatory factors. Although sev-
`eral potential negative regulators of angiogenesis have been identified,
`relatively little is known about their role in the physiological regulation
`of angiogenesis. Thrombospondin, a large multifunctional glycopro-
`tein secreted by most epithelial cells in the extracellular matrix, inhibits
`angiogenesis associated with tumour growth and metastasis44. Several
`fragments of larger proteins have been described as endogenous
`inhibitors of angiogenesis including endostatin45, tumstatin46 and vaso-
`statin47. The most recently described endogenous inhibitor of angio-
`genesis is vasohibin, which seems to be derived from the endothelium
`and to operate as a feedback regulator48. The precise mechanism of
`action of these proteins remains to be more clearly defined, although
`several hypotheses have been proposed, including that they bind to spe-
`cific integrins in the case of endostatin and tumstatin49.
`
`Role of bone-marrow-derived cells in angiogenesis
`An intensively debated issue in the field is the contribution (as well as
`the precise nature) of bone-marrow-derived endothelial progenitor
`
`Figure 3 | Various strategies to inhibit VEGF signalling.
`These include monoclonal antibodies targeting
`VEGF-A (a) or the VEGF receptors (b, c). d, Chimaeric
`soluble receptors such as the ‘VEGF-trap’ (domain 2 of
`VEGFR-1 and domain 3 of VEGFR-2 fused to a Fc
`fragment of an antibody) are also undergoing clinical
`development. e, Additional extracellular inhibitors are
`aptamers that bind the heparin-binding domain of
`VEGF165 (pegaptanib). A variety of small-molecule
`VEGF RTK inhibitors that inhibit ligand-dependent
`receptor autophosphorylation of VEGFR-1 and
`VEGFR-2 are being tested. Additional strategies to
`inhibit VEGF signalling include antisense and siRNA
`targeting VEGF-A or its receptors.
`
`a
`
`Anti-VEGF
`antibodies
`
`Anti-VEGF-1
`antibodies
`
`b
`
`VEGF
`
`P
`
`P
`
`P
`
`P
`
`VEGFR-1
`
`VEGFR-2
`
`P
`
`P
`
`P
`
`P
`
`Endothelial cell
`
`e
`
`Small-molecule
`VEGFR TK
`inhibitors
`
`d
`
`Soluble
`VEGF
`receptors
`
`e
`
`Aptamers
`
`c
`Anti-VEGFR-2
`antibodies
`
`© 2005
`
`Nature Publishing Group
`
`© 2005 Nature Publishing Group
`
`969
`
`

`

`INSIGHT REVIEW
`
`NATURE|Vol 438|15 December 2005
`
`cells (EPC) to angiogenesis. However, there is little doubt that bone-
`marrow-derived cells participate in angiogenic processes, at least as a
`source of angiogenic factors. In 1997, Asahara et al. reported the iso-
`lation of putative EPC from human peripheral blood, on the basis of
`cell-surface expression of CD34 and other endothelial markers50.
`These cells were reported to differentiate in vitro into endothelial cells
`and seemed to be incorporated at sites of active angiogenesis in vari-
`ous animal models of ischaemia. These findings suggested that incor-
`poration into the lumen of bone-marrow-derived endothelial
`precursor cells contributes to the growing vessels, complementing res-
`ident endothelial cells in sprouting new vessels. Also, ischaemia and
`various cytokines, including VEGF and granulocyte-macrophage
`colony-stimulating factor (GM-CSF), were reported to mobilize EPC
`into sites of neovascularization51. However, the precise contribution of
`these cells in various pathophysiological circumstances was not clearly
`defined.
`Subsequent studies have suggested that the contribution of such
`cells to angiogenesis is dependent on the experimental system
`employed. In the angiogenic-defective, tumour-resistant Id-mutant
`mice, EPC accounted for a large proportion of endothelial cells asso-
`ciated with xenografted tumours52. Rafii and collaborators proposed
`that mobilization of EPC from bone marrow requires angiogenic-fac-
`tor-mediated activation of MMP-9, which leads to the release of the
`soluble KIT ligand. This ligand would in turn promote proliferation
`and motility of EPC within the bone-marrow microenvironment, thus
`creating permissive conditions for their mobilization into the periph-
`eral circulation53. However, in spontaneous tumours occurring in Id-
`deficient mice in the tumour-prone PTEN+/ǁ genetic background, the
`contribution of EPC was less significant54. Also, De Palma et al. sug-
`gested that the percentage of EPC that are truly incorporated into a
`growing vessel wall is very low and that the majority of bone-marrow-
`derived cells homing in on the tumour vasculature are adherent
`perivascular mononuclear cells, which may contain angiogenic fac-
`tors55. Peters et al. recently analysed the tumour endothelial cells in six
`individuals who developed cancers after bone-marrow transplantation
`with donor cells derived from individuals of the opposite sex and
`found that an average of 4.9% of cells of the total endothelial cell pop-
`ulation were derived from the bone marrow56.
`In summary, bone-marrow-derived cells seem to contribute to
`tumour angiogenesis, of which a small and variable proportion are
`probably true EPCs. Bone-marrow-derived circulating pro-angiogenic
`cells, regardless of their precise nature, may be a common target for
`antiangiogenic therapies and may be exploitable as surrogate bio-
`markers for the angiogenic process as well as antiangiogenic thera-
`pies57.
`
`Combination therapies
`It is increasingly likely that cancer therapy, with a few exceptions, will
`need to be combinatorial. It seems logical to target multiple pathways
`simultaneously. Much preclinical evidence indicates that combining
`antiangiogenic agents with conventional cytotoxic agents or radiation
`therapy results in additive or even synergistic anti-tumour effects58. So
`far, it is unclear whether such positive interaction takes place prefer-
`entially with specific types of antiangiogenic or cytotoxic agents. An
`issue that is being debated is the mechanism of such potentiation, as it
`would seem counterintuitive that ‘tumour-starving’ antiangiogenic
`drugs that suppress blood flow in tumours actually increase the effi-
`cacy of chemotherapy. Browder et al.59 and Klement et al.60 proposed
`that chemotherapy, especially when delivered at close regular intervals
`using relatively low doses with no prolonged drug-free break periods
`(‘metronomic therapy’), preferentially damages endothelial cells in
`tumour blood vessels. These cells are presumably dividing, and the
`simultaneous blockade of VEGF-A is thought to blunt a key survival
`signal for endothelial cells, thus selectively amplifying the endothelial
`cell targeting effects of chemotherapy, leading to improved subsequent
`killing of cancer cells.
`A similar process, in principle, may take place when combining
`
`a
`
`b
`
`Figure 4 | Computed tomography chest scans. These scans were taken of a
`NSCLC patient before (a) and after (b) three cycles (nine weeks) of
`treatment with bevacizumab plus carboplatin and taxol (reproduced from
`ref. 71 with permission from American Society of Clinical Oncology). Note
`that the tumour mass in a (arrow) underwent extensive necrosis and
`cavitation in b (arrow). This pattern was seen more frequently in patients
`treated with bevacizumab plus chemotherapy relative to chemotherapy
`alone. Cavitation may be associated with serious bleeding, especially when it
`occurs in proximity to large vessels71.
`
`more conventional maximum-tolerated dose chemotherapy regimens
`with a drug such as bevacizumab61. In addition, bone-marrow-derived
`pro-angiogenic circulating cells, probably including authentic EPC,
`seem to be very sensitive to both conventional cytotoxic and low-dose
`metronomic chemotherapy62. However, levels of such cells can rapidly
`rebound, returning to normal or even increased levels during the
`drug-free break periods after maximum-tolerated dose cytotoxic
`chemotherapy62. Because VEGF-A acts as a mobilizing and probably a
`survival agent for such cells, co-administration of a VEGF-targeting
`agent, especially one with a long half-life in the circulation (for exam-
`ple, anti-VEGF antibodies), would be expected to amplify and sustain
`the suppressive effects of standard (as well as metronomic) chemother-
`apy on bone-marrow-derived circulating pro-angiogenic cells63. Fur-
`thermore, it has been proposed that progressively accelerated
`proliferation and repopulation of cancer cells during intervals of radio-
`therapy or chemotherapy is an important cause of treatment failure64.
`It is tempting to speculate that antiangiogenic treatment during these
`intervals inhibits such repopulation process. Figure 2 illustrates the
`various cellular targets of chemotherapy.
`An alternative hypothesis has been proposed by Jain65. Submaximal
`doses of an antiangiogenic agent such as an anti-VEGFR-2 antibody
`would ‘normalize’ the vasculature that is characteristic of many vessels
`in tumours. This would result in pruning of excessive endothelial and
`perivascular cells, in a decrease in the normally high interstitial pres-
`sures detected in solid tumours and in temporarily improved oxy-
`genation and delivery of chemotherapy to tumour cells65. However,
`according to recent studies, the tumour vasculature can be ‘normal-
`ized’ transiently, and eliciting synergistic effects through this mecha-
`nism requires administration of chemotherapy or radiation therapy
`over a defined time window after the angiogenesis inhibitor66. Con-
`sidering also that in most clinical protocols no such intentional
`sequential administration is performed, it remains to be established
`whether such a mechanism accounts for the long-term beneficial
`effects of combination treatments observed in some trials. By contrast,
`acute administration of angiogenic inhibitors induces vascular
`changes consistent with ‘normalization’ in humans. In this regard, Wil-
`lett et al. reported that a single infusion of bevacizumab to patients
`with rectal carcinoma rapidly decreased tumour perfusion, vascular
`volume, microvascular density and interstitial fluid pressure as well as
`the number of viable, circulating endothelial cells in six colorectal can-
`cer patients67.
`Combinatorial therapies with antiangiogenic agents are not limited
`to those including cytotoxic chemotherapy. Several preclinical and
`clinical trials are exploring the combination of various angiogenesis
`
`970
`
`© 2005
`
`Nature Publishing Group
`
`© 2005 Nature Publishing Group
`
`

`

`NATURE|Vol 438|15 December 2005
`
`INSIGHT REVIEW
`
`inhibitors with other targeted therapies, such as EGFR or Her2
`inhibitors (cetuximab, erlotinib and trastuzumab), PDGFR/ bcr-abl
`inhibitors (imatinib), proteasome inhibitors (bortezomib) and other
`antiangiogenic agents such as inhibitors of integrins (for example
`Ȋvȋ3 and Ȋ5ȋ1).
`
`Clinical trials for antiangiogenesis
`Many angiogenesis inhibitors are currently in clinical trials. It is note-
`worthy that, in parallel to angiogenesis inhibitors, another class of vas-
`cular-targeting or vascular-modulating drugs is being tested, namely
`‘vascular-disrupting agents’. These drugs primarily target existing,
`recently formed vasculature and cause acute vascular occlusion and
`disruption of tumour blood flow68. For an overview of these trials, see
`http://www.cancer.gov/clinicaltrials/developments/anti-angio-table.
`The inhibitors tested include a variety of agents with diverse mecha-
`nisms of action (several of which are not known). At present,
`inhibitors of the VEGF pathway are the most clinically advanced, and
`bevacizumab, a humanized variant of a murine anti-VEGF-A mono-
`clonal antibody that was used in early proof-of-concept studies23, is the
`only FDA-approved antiangiogenic treatment for cancer therapy69.
`Figure 3 illustrates several methods for inhibiting the VEGF pathway.
`Several important clinical studies testing angiogenesis inhibitors
`have been presented at recent oncology meetings, such as the Ameri-
`can Society of Clinical Oncology meeting. Typically, clinical studies
`are presented and discussed at such meetings in advance of peer-
`reviewed publication. Therefore, in the interest of an up-to-date
`overview of the field, a discussion of some of these studies will be
`included here, with the caveat that the data are preliminary and require
`further analysis.
`The clinical trial that resulted in FDA approval of bevacizumab was
`a large, randomized, double-blind, phase III study in which beva-
`cizumab was administered in combination with bolus IFL (irinotecan,
`5FU and leucovorin) chemotherapy as first-line therapy for metasta-
`tic colorectal cancer5. Median survival was increased from 15.6
`months in the bolus-IFL plus placebo arm of the trial to 20.3 months
`in the bolus IFL plus bevacizumab arm. Similar increases were seen in
`progression-free survival, response rate and duration of response. The
`clinical benefit of bevacizumab was seen in all subject subgroups,
`including those defined by performance status, location of primary
`tumour, number of organs involved and duration of metastatic dis-
`ease5. Although bevacizumab was generally well tolerated, some seri-
`ous and unusual toxicities have been noted, albeit at low frequencies.
`Bevacizumab was associated with gastrointestinal perforations and
`wound healing complications in about 2% of patients. In addition, the
`incidence of arterial thromboembolic complications were increased
`about twofold relative to chemotherapy alone, with patients 65 years
`or older with a history of arterial thromboembolic events being at
`higher risk. Although the precise mechanism of this effect is unknown,
`it is conceivable that vascular damage induced by cytotoxic agents can
`be exacerbated by the blockade of VEGF-A.
`Preliminary data of a phase III study indicate that bevacizumab
`confers a survival advantage on patients with previously treated,
`relapsed, metastatic colorectal cancer in combination with FOLFOX4
`chemotherapy (5-fluorouracil, leucovorin and oxaliplatin), relative to
`chemotherapy alone (B. Giantonio, P. J. Catalono, N. J. Meropol, E. P.
`Mitchell, M. A. Schwartz et al., unpublished data).
`The role of bevacizumab in other tumour types and settings is
`currently under investigation, and phase III clinical trials of this drug
`in non-small-cell-lung cancer (NSCLC), renal cell cancer and metasta-
`tic breast cancer are ongoing. An early phase III trial of advanced,
`heavily pretreated, metastatic breast cancer showed that adding beva-
`cizumab to capecitabine chemotherapy did not improve progression-
`free survival, despite a doubling of the response rate (that is, tumour
`shrinkage of 50% or more) in the bevacizumab-treated arm of the
`trial70. Thus, the responses seemed to be very short in duration. How-
`ever, an interim analysis of a phase III study of women with previously
`untreated metastatic breast cancer treated with bevacizumab in com-
`
`bination with weekly paclitaxel chemotherapy showed that the study
`met its primary efficacy endpoint of improving progression-free sur-
`vival, compared with paclitaxel alone (K. Miller, unpublished data).
`Furthermore, administration of bevacizumab in combination with
`paclitaxel and carboplatin to patients with NSCLC resulted in
`increased response rate and time to progression relative to chemother-
`apy alone in a randomized phase II trial71. The most significant adverse
`event was serious haemoptysis. This was primarily associated with
`centrally located tumours with squamous histology, cavitation and
`central necrosis and proximity of disease to large vessels71. Figure 4
`illustrates the extensive tumour necrosis and cavitation that may result
`from the combination treatment71. More recently, preliminary results
`from a large, randomized phase III clinical trial for patients with pre-
`viously untreated advanced non-squamous NSCLC show that patients
`who received bevacizumab in combination with paclitaxel and carbo-
`platin lived longer than patients who received chemotherapy alone (A.
`B. Sandler, R. Gray, J. Brahmer, A. Dowlati, J. H. Schiller et al., unpub-
`lished data). Serious bleeding was infrequent but occurred more com-
`monly in the bevacizumab arm of the trial.
`Besides bevacizumab, several other VEGF inhibitors are being clin-
`ically pursued. A variety of small-molecule RTK inhibitors targeting
`the VEGF receptors have been developed. The most advanced are
`SU11248 and Bay 43-9006. SU11248 inhibits VEGFRs, PDGFR, c-kit
`and Flt-3 (ref. 72) and has been reported to have considerable efficacy
`in imatinib-resistant gastrointestinal stromal tumour (R. G. Maki, J. A.
`Fletcher, M. C. Heinrich, J. A. Morgan, S. George et al., unpublished
`data). Bay 43-9006 was initially identified as a raf kinase inhibitor and
`subsequently shown to inhibit several RTKs including VEGFRs. An
`interim analysis of phase III data indicates that Bay 43-9006
`monotherapy results in a significant increase in progression-free sur-
`vival in patients with advanced renal cell carcinoma (B. Escudier, C.
`Szczylik, T. Eisen, W. M. Stadler, B. Schwartz et al., unpublished data).
`Follow-up of such phase III data is ongoing to determine whether an
`overall survival benefit occurs. An earlier randomize