Recent Progress in Hormone Research 59:73-104 (2004)
© 2004 The Endocrine Society
Angiostatin and Anti-angiogenic Therapy in Human Disease
Miriam L. Wahl,
Tammy L. Moser and
Salvatore V. Pizzo
Department of Pathology, Duke University Medical Center, Durham, North Carolina 27710
 |
ABSTRACT
|
|---|
Many diseases have abnormal quality and/or quantity of vascularization as a characteristic feature. Cancer cells elicit the growth of new capillaries during neovascularization in a process termed angiogenesis. In diabetics, pathologic angiogenesis in various tissues is a clinical feature of many common complications. Therefore, the diabetic cancer patient warrants special consideration and extra care in the design of anti-angiogenic treatments without adverse side effects. Some treatment regimens that look promising in vitro, in animal models, or in early clinical trials may be contra-indicated for diabetics. This chapter will review the common complications of diabetes, with emphasis on the angiogenic pathology. Recent research related to the mechanism of action and basis for specificity of the anti-angiogenic peptide, angiostatin, will be the focus. The aim is to shed light on areas in which more research is needed with respect to angiostatin and other anti-angiogenic agents and the microenvironmental conditions that affect their activities, in order to develop improved therapeutic strategies for diabetic cancer patients.
|
I. Introduction
|
|---|
Normal tissue function relies on an adequate supply of nutrients and oxygen from pre-existing blood vessels. It is well established that tumor growth depends on development of a new vascular supply, a process known as angiogenesis (Folkman, 1971). The discovery of factors that mediate this process has significantly increased our understanding of many normal and pathological circumstances. For example, angiogenesis is desirable in wound healing and recovery from cardiac ischemia but undesirable in the pathogenesis of psoriasis (Oates, 2002) or when metastastic tumor tissue develops a blood supply, allowing it to grow and spread (Weidner et al., 1991,1993; Macchiarini et al., 1992). The quest has been to isolate naturally occurring agents or develop new ones that aid in the regulation of this process. Judah Folkman initially developed the hypothesis that naturally occurring agents existed that inhibited angiogenesis. The first two, discovered in his laboratory, were angiostatin (O’Reilly et al., 1994a) and endostatin (O’Reilly et al., 1997).
As tumors grow in size, they become hypoxic and acidotic and elaborate several growth factors to stimulate local blood vessels to sprout branches. Endothelial cells then begin proliferating towards the tumor and form tubular structures that become blood vessel branches. These new blood vessel branches differ from normal blood vessels. Some agents can target them while not harming intact, mature vasculature elsewhere in the body. For example, newly formed blood vessels are "leaky" or exceptionally permeable (Dvorak et al., 1988). They also are tortuous, with blind ends and incomplete drainage or backflow (Secomb et al., 1993; Kimura et al., 1996) and therefore contain a mixture of arterial and venous blood (Kallinowski et al., 1988; Vaupel, 1997; Vaupel et al., 1989,1998). This makes them poor at removing catabolites, including carbon dioxide (CO2) (Boyer and Tannock, 1992).
Angiostatin is endogenously produced in the tumor stroma. Therefore, circulating levels can control metastatic cell proliferation until a primary tumor is removed. Before the discovery of angiostatin, it was believed that when cancer recurred after surgical removal of a primary tumor, it was because the surgeon had left some tumor cells behind. It is now generally accepted that micrometastases already have seeded elsewhere in the body but were inhibited from proliferating by endogenous production of angiostatin or other inhibitors from the primary tumor. These micrometastases then came under permissive growth conditions when the primary tumor was removed. The discovery of angiostatin elicited great excitement because it seemed likely that exogenous administration after primary tumor removal could prevent metastatic growth. However, one problem is the short half-life of this peptide, leading to a need for continuous administration. Identifying receptors for these molecules was deemed a high priority, so that alternate molecules could be developed that have the same targets and are more stable, more effective, and easier to manufacture and administer.
Angiostatin is found naturally in significant amounts in the circulation of patients with primary tumors (Canfield et al., 1986; Cao et al., 2000; OSI Pharmaceuticals et al., 2001). When a primary tumor is removed, metastases may experience lower circulating levels of endogenous inhibitors. Local blood vessels respond to the malignant cells’ elaboration of vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF). Local blood vessels then respond by sprouting branches to feed the metastases. At this time, small micrometastases may grow beyond the 2-mm size, which is functionally dormant, and become a threat to the patient when rapid growth causes local damage. Angiostatin can maintain metastases in a dormant state in laboratory animals when administered exogenously (O’Reilly et al., 1994a,b,1997). Research on the mechanism of action of angiostatin delineates some parameters affecting its activity that will require evaluation, in order to design treatments that yield the most-effective combinations of agents. More-complicated issues are whether these treatments are applicable in patients having a variety of vascular pathologies. Diabetic patients have other body compartments that feature abnormal vascularization. Understanding how these anti-angiogenic agents work and the conditions in the various compartments will enable predictions about possible side effects, permitting further drug development. Many agents are in clinical trials (Table I), with alarmingly little known about their mechanisms of action and potential side effects.
|
II. Stages of Tumor Angiogenesis
|
|---|
Angiogenesis begins when a fibrin clot forms on the adventitial surface of an existing blood vessel (Muthukkaruppan and Auerback, 1970; van Hinsbergh et al., 2001), followed by sprouting of new capillaries. The initial phase begins with increased vascular permeability and local degradation of the vessel wall (i.e., extracellular matrix (ECM) or basement membrane). The endothelial cells enter the tumor stroma, migrate toward a stimulus such as VEGF or FGF, and proliferate behind the leading edge. At this time, the cells may be most vulnerable to agents that interfere with their proliferation, since they lack protection from other cell types (Egginton et al., 2000). The next step in vessel formation is recruitment of pericytes, followed by smooth muscle cells.
|
III. On/Off Switches for Angiogenesis: a Delicate Balance
|
|---|
Angiostatin is a fragment of a larger protein, plasminogen, with an activity distinct from the parent protein. It may be cleaved from the original precursor protein by enzymes under various conditions; normally, in cases where a pro-angiogenic stimulus first has promoted blood vessel development (such as during wound healing), which must then be inhibited when the desired level of vascularity has been achieved. Angiostatin has a short half-life of about 15 minutes (Gonzalez-Gronow et al., 1990; DeMoraes et al., 2001; Fortier et al., 2001). Consistent with these results, exogenous administration of angiostatin at doses ranging from 15–300 mg/m2/day showed rapid clearance of this protein and no side effects in a phase I clinical trial performed at Thomas Jefferson University (DeMoraes et al., 2001; Fortier et al., 2001). Based on animal studies, however, if administered continuously in addition to endostatin, metastasis may be discouraged from growth for prolonged lengths of time (Yokoyama et al., 2000). Combination therapies with two new agents are necessarily slow to be tested in humans. Each compound has to be tested individually first in patients and, if they are only effective together, the potential efficacy may not be discernable in early trials. Recent combinations that have been studied in animal models include angiostatin with chemotherapy (te Velde et al., 2002) and angiostatin with radiation (Mauceri et al., 1998). Basic research is needed to understand mechanisms and predict which combinations will be most effective, so that trials can be done in the correct context, the right tumor types, and thus in patients that will benefit the most. Furthermore, deleterious side effects should be predictable and avoidable if comparison between compartmental microenvironment and tumor microenvironment are made and the effects of these determinants on the drug activity are understood.
|
IV. Mechanisms of Action of Angiostatin
|
|---|
One major theme that has emerged from research done thus far on anti-angiogenic compounds is that the extent of cell attachment to a matrix and the nature of the matrix are critical determinants of the compound’s activity. This is a complex issue because tumor stroma is comprised of a variety of proteins, often in abnormal relative concentrations. Extensive literature indicates that, in addition to tumor cell modulation of stroma composition, stromal components impact tumor cells. This two-directional signaling is part of the malignant phenotype and often is regulated at the post-transcriptional level. This pioneering work, recently reviewed by Roskelley and Bissell (2002), has delineated the importance of tumor stroma in tumor cell behavior. Recent work in the field of angiogenesis has shown that many of the same microenvironmental factors are critical to the endothelial cell response to the tumor microenvironment and can determine the phenotype of these cells and the outcome of exposure to various stimuli (e.g., administration of exogenous angiostatin) (Wahl and Grant, 2002).
Angiostatin is an internal fragment of plasminogen and may contain either the first three (K1–3) or four (K1–4) kringle domains. A similar activity has been reported for kringle 5 (K-5) of plasminogen (Liu et al., 2000; Gonzalez-Gronow et al., 2003). Three receptors on endothelial cells have been reported for angiostatin to date. Adenosine triphosphate (ATP) synthase was the first receptor identified by Moser and colleagues (1999). The presence of this typically mitochondrial enzyme on the endothelial cell surface was somewhat surprising. Binding of angiostatin to surface-associated ATP synthase since has been confirmed by other research groups (Wahl and Grant, 2002; Arakaki et al., 2003). Additional reports have confirmed the unexpected finding that many of the enzymes and components of the mitochondrial electron transport chain and ATP synthesis generating mechanisms are located on the plasma membrane of endothelial cells (Yegutikin et al., 2002; Arakaki et al., 2003). Two other potential target receptors of angiostatin that have been reported are angiomotin (Troyanovsky et al., 2001) and integrin 
vß3 (Tarui et al., 2001). The distribution of these receptors on the surface of endothelial cells may be located at sites so proximal to one another that angiostatin may be able to interact with more than one simultaneously. Alternatively, multiple targets could be implicated in binding to the different kringles of angiostatin as a function of receptor and peptide presentation.
Parameters affecting angiostatin’s interaction with the ATP synthase receptor have not been completely elucidated but recent work indicates that the synthase may be situated in a caveolar compartment (Moser et al., 2001) and generates ATP on the cell surface (T.L. Moser, D.J. Cheek, J.A. Roy, T.A. Ashley, M.D. Goodman, A.E. Paradis, B. Li, D.J. Kenan, and S.V. Pizzo, unpublished results). It has been reported that one of angiostatin’s receptors is proximal to focal adhesion kinase (FAK) and that binding catalyzes FAK phosphorylation in the absence of integrin clustering that usually is triggered by cell attachment to substrate (Claesson-Welsh et al., 1998). This implies that FAK phosphorylation occurs under aberrant circumstances (i.e., when endothelial cells are not tightly attached to their normal substratum). In the usual scenario, attachment to a substrate such as fibronectin causes integrins to form tetramers. FAK is phosphorylated and organizes, allowing cytoskeletal elements to form structurally organized arrays to carry out cell spreading (Figures 1 and 2). In the tumor microenvironment, several parameters differ. Matrix composition includes collagen, laminin, and many other proteins (Canfield et al., 1986; Baatout and Cheta, 1996; Grant and Kleinman, 1997; Aoudjit and Vuori, 2001). The pH and the oxygen levels are low and the degree of attachment is compromised because cells are rapidly dividing and moving into areas requiring a blood supply. This means that different integrin isoforms may be activated to differing extents (Schwartz et al., 1991b; Agius et al., 1996; Coopman et al., 1996; Erdreich-Epstein et al., 2000). Thus, the distribution of other molecules on the cell surface and their activities will be affected. When extracellular pH is low, typically the sodium/proton antiporter, isoform 1 (NHE-1), is activated to maintain intracellular pH in the viable range. This occurs as a function of attachment, phosphorylation, ATP binding, and the presence of growth factors in fibroblasts, where it has been studied in the most detail (Schwartz et al., 1991a; McSwine et al., 1994). It has been reported that migrating cells have NHE-1 clustering at the leading edge of lamellopodia in work done using fibroblasts (Grinstein et al., 1993) and melanoma cells (Akasaka et al., 1995) and that they co-localize in partially attached and moving cells with FAK (Schwartz et al., 1991a).
View larger version (57K):
[in this window]
[in a new window]
|
FIG. 2. Endothelial cell attached to fibronectin at normal pH ( 7.3). Integrins are clustered into tetramers in response to attachment to fibronectin, FAK is phosphorylated, NHE is more active, cytoskeletal elements are assembled to promote cell attachment and spreading, and ATP synthase remains in mitochondria.
|
|
We recently demonstrated that when coupled with extracellular acidification, angiostatin caused a precipitous decline in cytosolic intracellular pH (Wahl et al., 2001,2002a; Wahl and Grant, 2002). Low extracellular pH could affect receptor levels, receptor distribution, angiostatin binding, and angiostatin conformation. ATP synthase distribution on the endothelial cell surface was reported to be more focal at low pH (Wahl and Grant, 2002). Receptor binding interactions with angiostatin as a function of extracellular pH currently are being studied in our laboratories. At normal extracellular pH, both ATP synthase and NHE-1 would be less active, whether or not cells are attached. However, in the tumor microenvironment (Figures 3 and 4), there is less attachment to a variety of substrate proteins, a lack of integrin clustering, and aberrant FAK phosphorylation. Because of the low extracellular tumor pH, NHE-1 is activated and ATP synthase is activated and organized focally (Wahl and Grant, 2002). When angiostatin enters this scenario, it could bind to ATP synthase, which could, in turn, disrupt FAK, ATP synthesis, and the function of NHE-1. A model for how angiostatin binding affects endothelial cells in a tumor microenvironment — based on recent research, the literature, and conjecture — is shown in Figure 4.
View larger version (56K):
[in this window]
[in a new window]
|
FIG. 3. Endothelial cell in a tumor microenvironment with a low and variable pH. Cells are partially attached to various matrix proteins, including collagen and laminin, different integrin isoforms cluster in tetramers, and ATP synthase appears in caveolae situated near FAK. This ATP synthase now generates ATP on the plasma membrane. FAK is phosphorylated. Extracellular pH is lower (6.7 on average) and migration of endothelial cells is required; therefore, NHE is very active, highly phosphorylated, and has ATP bound on the cytoplasmic tail. NHE molecules congregate at the leading edge of lamellopodia to generate a sodium gradient to bring about cell movement. Signals are transduced to cytoskeleton, which catalyzes a rearrangement of cytoskeletal elements. Cells then proliferate and can differentiate into tubular structures.
|
|
View larger version (51K):
[in this window]
[in a new window]
|
FIG. 4. Proliferating endothelial cells in the tumor microenvironment in the presence of angiostatin. Angiostatin binds to the catalytic subunit of ATP synthase. ATP synthesis ceases and inhibits the synthesis of extracellular ATP. Rapid proliferation loosens attachment of the cells, integrin tetramers dissociate into dimers, and NHE molecules disaggregate. FAK is phosphorylated (as in Figure 2) but under conditions where attachment is not being maintained. NHE is inhibited by lack of attachment, dephosphorylations, and loss of ATP. Intracellular pH decreases below 6.5 and metabolism ceases. The cell has received conflicting signals, which trigger a signal transduction cascade ending in apoptosis.
|
|
|
V. Other Anti-angiogenic Agents
|
|---|
Other anti-angiogenic agents have been discovered or developed. In some cases, the receptors have been identified (Table I). However, little is known about how their signals are transduced, the basis for their selectivity, or their degree of selectivity. Features of the tumor microenvironment that have been shown to be critical for demonstrating angiostatin activity are delineated below, emphasizing parameters that should be evaluated for other anti-angiogenic agents that may be useful to determine in which clinical scenario(s) each is likely to be most effective.
Some characteristic features of the tumor microenvironment may be shared by other improperly vascularized areas. This means that anti-angiogenic therapies could cause complications for cancer patients with diabetes mellitus. Drugs for treating tumors may have undesirable side effects in other tissues in diabetics to a greater extent than in nondiabetics. Some of these features have been described but have not been characterized fully in terms of application of anti-angiogenic agents. Differences between vasculature arising in different locations, in endothelial cells from the various tumor types, and in different ECM microenvironments are also areas where further basic research is warranted. Some major parameters that need to be examined for each anti-angiogenic agent under consideration for clinical use are described in Section VI.
|
VI. Features of the Tumor Microenvironment: Potential Mediators of the Response of Tumor Stromal Endothelial Cells to Anti-angiogenic Drugs
|
|---|
A. VEGF & FGF
VEGF and FGF are two growth factors secreted by metastatic or primary tumor cells to encourage new vessel branch development. These growth factors are used as markers for vessel growth. In the first phase I clinical trial of angiostatin at Thomas Jefferson University, the protocol was designed to administer daily doses of intravenous (IV) angiostatin for 2-week intervals, separated by 1-week interruptions. Both VEGF and FGF rebounded rapidly in the circulation just 1 day after daily dosing stopped after each dosing interval (DeMoraes et al., 2001). This made it evident that daily dosing without interruption would be necessary to maintain the effects of angiostatin. It indicated that agents with a slower clearance time would probably be more practical clinically. Table I lists some anti-angiogenic agents currently in clinical trial. If these are expressed in excess and/or at inappropriate times, hypervascularization could occur.
B. ENDOGENOUS INHIBITORS
Other endogenous inhibitors of angiogenesis have been reported, including angiogenin (Fett et al., 1985), endostatin (O’Reilly et al., 1997), thrombospondin-1 (Nicosia and Tuszynski, 1994), interferons (Dinney et al., 1998), platelet factor-4 (Nicosia et al., 1994), and 16 kDa prolactin (Hanahan and Folkman, 1996). Numerous other inhibitors, both naturally occurring and synthetic, are listed in Table I. Their relative abundance can play a role in the overall picture in various body compartments. Angiostatin has proven to be one of the most potent. In vitro, angiostatin is relatively specific for tumor endothelial cells and inhibits their proliferation (Moser et al., 1999), migration (Moser et al., 1999), tube formation (Moser et al., 1999; Wahl et al., 2002a), vessel network formation in the embryonic body model (Eriksson et al., 2003), and formation of sprouts from mouse aortic rings (Hajitou et al., 2002). Exceptions to this specificity include a published study on smooth muscle cells (Walter and Sane, 1999) and one on neutrophils (Benelli et al., 2002). In vivo, however, angiostatin inhibits primary tumor metastasis in mice (O’Reilly et al., 1994b), vascular proliferation in the chick chorioallantoic membrane (CAM) assay (Gately et al., 1996), and neovascularization in the corneal pocket assay (White et al., 2003).
Angiostatin is found in normal human plasma at a concentration of 6–12 nM (Soff et al., 1999). The concentration of angiostatin in the urine of cancer patients is significantly higher than in other patients (Cao et al., 2000). Elevated levels of angiostatin have been observed in the ascites fluid from ovarian cancer patients as well (Yokoyama et al., 2000). More recently, angiostatin was found to be elevated during impaired production of nitric oxide in coronary angiogenesis (Matsunaga et al., 2002) and in the bronchoalveolar lavage fluids from patients with acute respiratory distress syndrome (ARDS) (Lucas et al., 1998). These findings may indicate that angiostatin functions not only as a tumor-specific inhibitor but also as a regulator of angiogenesis in other scenarios.
C. PERICYTES & SMOOTH MUSCLE CELLS
Pericytes and smooth muscle cells are recruited to surround the endothelial cell tubes in developing blood vessels, so reduced numbers of them could affect the levels of various stimulators and inhibitors to which endothelial cells are exposed. Platelet-derived growth factor (PDGF) plays a major role in pericyte recruitment. When vessels initially form, their vulnerability may be maximal in part because these other cell types are not there to provide a mechanical barrier. A recent study indicates that retinal capillary coverage by pericytes was crucial for survival of endothelial cells in retina subjected to hypoxic conditions, particularly under the stress conditions of diabetes (Hammes et al., 2002). Pericytes also secrete similar growth factors and thus modulate the ECM (Allt and Lawrenson, 2001). Smooth muscle cell proliferation also is inhibited by angiostatin (Walter and Sane, 1999).
D. INTRACELLULAR & EXTRACELLULAR pH
Both intracellular and extracellular pH affect angiostatin activity. The average extracellular pH (5.6–7.6) in tumors is lower and more variable than in normal tissue (7.2–7.6), yet tumors have a normal average intracellular pH (Yamagata and Tannock, 1996). It has been reported that angiostatin profoundly affects intracellular pH in endothelial cells (Wahl and Grant, 2002; Wahl et al., 2002a). This effect is manifested only at low extracellular pH. This implicates pH-regulating transporters that are activated at low extracellular pH as a potential class of targets for anti-angiogenic agents in areas of tumors where extracellular pH is low. These potentially include NHE and the H+-linked monocarboxylate transporter (MCT).
E. ATP SYNTHASE RECEPTOR LEVELS
Modulation of angiostatin’s receptor levels may be a determinant of angiostatin activity. Research has shown that ATP synthase distribution can be altered as a function of ECM composition and extracellular pH. When cells are allowed to attach to fibronectin, histochemical assays using a primary antibody directed against the ß subunit of ATP synthase showed no detectable enzyme on the cell surface. However, when cells were plated on Matrigel to simulate the tumor stroma, the enzyme was revealed on the cell surface (Wahl and Grant, 2002). When cells at normal and low extracellular pH were compared, the distribution at low pH was more punctate. This was likely related to the organization of focal adhesion plaques and how they assemble for migration (Wahl and Grant, 2002). Other microenvironmental parameters have not been evaluated in terms of this receptor and these factors have not been evaluated for other anti-angiogenic agents.
F. HYPOXIA
Hypoxia often — but not always — goes hand in hand with acidity, although there is some disparity in spatial distribution (Vaupel et al., 1989,1998; Brizel et al., 1996; Gullege and Dewhirst, 1996; Helmlinger et al., 1997). These two features have not been teased apart experimentally to determine the relative importance of each and the degree of overlap. Most studies thus far are either conducted at low pH or done under hypoxic conditions but not both. Future studies with new anti-angiogenic agents, such as 2-methoxyestradiol (2ME2) or Panzem ® (Entremed, Inc.), a hypoxia-inducible factor (HIF)-1 inhibitor (Mabjeesh et al., 2003), will help determine the extent to which targeting this receptor will be useful in a variety of tumor types. There is evidence that hypoxia is a feature of diabetic complications such as renal disease (Ries et al., 2003), poor circulation in lower extremities (Fife et al., 2002), and retinopathy (Drasdo et al., 2002).
G. BICARBONATE STATUS
Bicarbonate (HCO3-) is a parameter that has not been considered in most in vitro work but should receive more attention. In the tumor microenvironment, catabolite removal by inefficient and abnormal blood vessels is poor, resulting in CO2 buildup (Newell et al., 1993; Helminger et al., 2002). Both these aspects contribute to the low pH environment. In some work at low pH, the pH is lowered by altering sodium bicarbonate concentration (Chu and Dewey, 1988; Chu et al., 1990; Wahl et al., 1996; Owen et al., 1997). In this type of experiment, the buffering capacity of the medium will be low, causing more potential extracellular pH fluctuation. Furthermore, this approach has two experimental variables, pH and bicarbonate, without distinction as to which is causing the effects on cells in the tumor microenvironment. Research studies have been performed where bicarbonate is maintained at 26 mM and the incubator CO2 is raised to 17% from 5%, to create acidic and tumor-like conditions (Wahl and Grant, 2002; Wahl et al., 2002a,b). This is an appealing model, since physiologic bicarbonate is 26 mM and the tumor is high in CO2, which is a likely cause of elevated carbonic anhydrase levels in tumors (Beasley et al., 2001; Giatromanolaki et al., 2001; Koukourakis et al., 2001; Olive et al., 2001; Bui et al., 2003; Swinson et al., 2003). It has been proposed that carbonic anhydrase inhibitors could relax pericytes, cause vasodilation, and enhance blood supply to the retina by increasing intracellular pH and decreasing extracellular pH (Reber et al., 2003). It would be even more physiologic to study tumor cell behavior in a hypoxic chamber inside a 17% CO2 incubator, to further mimic the part of the tumor microenvironment most likely to promote new blood vessel branch growth. Evaluation of low extracellular pH, low oxygen, and the combination in terms of therapeutic responses to agents and phenotype of cells in such environments should help characterize responses in the tumor microenvironment more completely.
H. LACTIC ACIDOSIS
Another reason, other than high CO2 and poor perfusion, for the acidic microenvironment in tumors is excessive production of lactic acid (Vaupel et al., 1989; Walenta et al., 1997; Brizel et al., 2001; Wahl et al., 2002b). One indicator of the contribution of lactic acidosis is measuring levels of the transporter used to remove it from the cell. This transporter, MCT, is elevated in human melanoma (Wahl et al., 2002b) and may be elevated in other tumors of neural crest origin, including brain tumors. MCT levels are elevated in the retina of diabetics, indicative of elevated lactic acid in that compartment (Knott et al., 1999). Intracellular and extracellular pH levels, functionality of other transporters used to regulate intracellular pH, and degree to which MCT is elevated may be critical to understanding the degree to which a pH regulation inhibitor can affect a certain cell type and to which angiostatin can affect intracellular pH in endothelial cells in the low pH tumor microenvironment. Intracellular and extracellular pH measurements in normal and pathologic body compartments aside from the tumor microenvironment will help address the relative importance of these parameters. In addition, a relationship between areas of hypoxia and acidity is likely, although the relative importance of each has not been addressed in any one study.
I. GLUCOSE, HYPERGLYCEMIA & ACIDIFICATION
Oral glucose is used clinically to lower tumor pH before thermoradiotherapy (Thistlethwaite et al., 1987; Leeper et al., 1994,1998; Engin et al., 1995). The mechanism is via increased glycolysis in the tumor microenvironment, producing lactic acid (Burd et al., 2001). In diabetes, glucose overload may occur, so the patient’s tissues already may be more acidic. This could lead to potentiation of angiostatin’s activity in the body, since its activity is greatest when pH is low (Wahl and Grant, 2002; Wahl et al., 2002a). Other enzymes relating to glycolysis may impact this axis, including lactate dehydrogenase (Koslowski et al., 2002), glut-1 receptor levels (Thews et al., 2003), and hexokinase levels (Wachsberger et al., 2002). In the diabetic, it is believed that complications arise to a greater extent in patients with the most poorly managed glucose levels (DCCT Research Group, 1993), although this is still a matter of considerable debate (Craighead, 1994). Individual variations in glycation (non-enzymatic glycosylation) of proteins are not only affected by degree of management but also by other unknown parameters. Facchiano and coworkers (2002) have reported that FGF (isoform 2) is more glycosylated in diabetic mice and is less chemotactic and pro-angiogenic as a result.
J. MATRIX
Matrix composition could be used to distinguish physiologic compartments, if qualitative and/or quantitative differences can be characterized in nondiabetics and diabetics with and without cancer. With respect to angiostatin, research indicates that the vulnerability of endothelial cells in the tumor microenvironment stems in part from the fact that when they are rapidly proliferating, their attachments are compromised. Furthermore, substrates they come in contact with in the tumor microenvironment differ, leading to different integrin isoform expression, with subsequent differences in signal transduction. This would contribute to the explanation of why mature vessels appear to be unaffected by angiostatin. It also would explain the targeting of endostatin to integrins (Sudhakar et al., 2003), which then would be less apt to be engaged in attachment. Migration of tumor cells is often along the ECM of the basal lamina. When mice are injected with a melanoma cell line selected for metastatic seeding in the lung, the cells migrate specifically and secondary lung tumors are formed. When they are inhibited from binding fibronectin or laminin, over 90% of the cells fail to reach the lungs (Humphries et al., 1986). In order to enter a blood vessel, tumor cells lyse the collagenous matrix by secreting various plasminogen activators. Antibodies directed against plasminogen activators have been shown to inhibit metastasis (Ossowski and Reich, 1983). The urinary-type plasminogen activation complex (u-PA) is one of the major regulators of ECM remodeling. u-PA converts plasminogen to plasmin, which degrades matrix and indirectly activates other metalloproteinases (Vassalli et al., 1991). Enriched levels of u-PA and its receptor (u-PAR) are found on the leading edge of migrating cells. Plasminogen activator inhibitor type 1 (PAI-1), a u-PA antagonist, mediates cell adhesion and spreading by forming a bridge between the cell surface and the matrix directly regulating adhesion (Planus et al., 1997). u-PA expression has been correlated with angiogenesis and poor prognosis (Kaneko et al., 2003).
Matrix metalloproteinases (MMP) are enzymes that digest/degrade matrix proteins, enabling metastasizing cells to migrate. Matrix metalloproteinase inhibitors (MMPI) exist in a balance and can be offset in pathologic conditions (Spranger et al., 2000). Another factor warranting further study is the composition of matrix in various locations, which will affect the degree of enzyme activity needed to impact upon the matrix. A third consideration is that enzymatic alteration of matrix composition may affect where tumor cells disseminate and may cause digestion of other pathologic matrices in nearby or distant areas. This issue may require consideration in glomerulosclerosis and other complications of diabetes in which basement membrane thickening is a feature. Finally, MMP/MMPI balance may affect wound healing, an important consideration for patients that may need surgery during their course of treatment (Lockhart et al., 2003).
K. TUMOR VESSEL ARCHITECTURE
Tumor blood vessel branches are inefficient and tortuous. In tumor vessels, backflow often occurs, mixing arterial and venous blood, and vessels leak (Secomb et al., 1993; McDonald and Baluk, 2002). Some of these characteristics are potentially exploitable by anti-angiogenic agents and antivascular agents, again, hopefully, with some specificity for the uniquely abnormal features of these vessels. More studies are needed to characterize vasculature in different tumor types, as a function of different body locations of metastasis, and the degree of resemblance to other vasculature in normal and pathologic instances that can coexist in patients.
|
VII. Models Used to Study Angiogenesis in Various Tissues
|
|---|
A. IN VITRO MODELS
Many cell types are used to study angiogenesis from macrovascular vessels, microvascular vessels, various locations in the body, and from various animal species. Although little information is available about the differences between these various cell types, these differences are extremely important if research findings are to be translated into practical, predictive information. Models include the tube differentiation assay (Kubota et al., 1988; Grant et al., 1990,1995; Grant and Kleinman, 1997), the migration/scratch assay, and the rat aortic ring assay (Nicosia and Ottinetti, 1990; Grant and Kleinman, 1997). For angiostatin, data have been confirmed in all three of these models (Wahl et al., 2002a). However, for other agents, one system may demonstrate activity more than another, which will lead to elucidation of what models may be most useful for which agents or diseases. Physiologic models are critical to getting meaningful results.
B. ANIMAL MODELS
Several animal models are available to study angiogenesis, including the corneal pocket (Muthukkaruppan and Auerback, 1970; O’Reilly et al., 1994b; Sood et al., 1999), the chick embryo (Risau and Lemmon, 1988; O’Reilly et al., 1994b; Eriksson et al., 2003), implantation of human cells into a mouse or a rat (xenografts) in various locations (Danielsen and Rofstad, 2000; Shan et al., 2003), and the angioreactor model (Guedez et al., 2003). Which are best to model which tumor types and clinical scenarios? Considerations include the fact that blood vessels in a human xenograft usually are a combination of human and animal vessels. Tumors often are placed in animals in locations other than the ones where they naturally occur. Model evaluation is critical to determining which one may be most physiologically meaningful. A knowledge of the pathogenesis of the particular type of cancer, the degree of vascularity it possesses, and the microenvironment of the location into which it will be grafted are all critical considerations.
C. DESIGN OF CLINICAL TRIALS: TIMING IS EVERYTHING
In the angiostatin phase I clinical trial at Thomas Jefferson University, wound healing was normal in a patient needing acute unexpected surgery while on angiostatin (DeMoraes et al., 2001). Other anti-angiogenic agents may differ, however, so this must be considered, preferably prior to introduction of agents in the clinic. Another consideration in trials is that the typical measure of efficacy — tumor growth delay assessments — may not be perfectly applicable to anti-angiogenic drugs. Some are initially slow to show responses (D’Amato et al., 1994). Others may cause necrosis followed by swelling, which is mostly inflammatory cells and fluid. Thus, even though a large part of the tumor mass is killed, the initial effect is an apparent increase in tumor size (Kaban et al., 1999). Most traditional therapies are directed at achieving maximum results during a brief window of opportunity, while most anti-angiogenic therapies will need to be administered chronically and indefinitely to prevent blood vessel branches from sprouting (Kaban et al., 1999). Therefore, animal model experiments will have to be designed to take these differences into account.
The parameters described above are emerging as important for cancer pathogenesis and treatment outcome, yet most are not considered during experimental work on anti-angiogenic drugs. Conversely, when pro-angiogenic drugs are administered for other conditions, patients with neoplastic disease may have counter-indications. Some of these drugs are listed in Table II.
|
VIII. Diabetes
|
|---|
Some sequelae of diabetes mellitus include impaired wound healing, characterized by both spontaneous and wound-induced ulcerations, particularly of the lower limbs, and retinopathy. Both these conditions are consequences of insufficient perfusion. If limbs become acidotic and hypoxic, why don’t they invoke the growth of new and normal blood vessels? Do they produce compounds that antagonize angiogenesis (e.g., angiostatin, endostatin) in higher amounts than those that promote it (e.g., VEGF, FGF) initially, then not turn off those signals at the right time? If so, exogenous administration of either type of compound could make things worse. With retinopathy, hypervascularization occurs at later stages, where there could be a positive impact from anti-angiogenic therapy. All potential side effects should be evaluated in diabetics, whether or not they have complications, since the eventual complication rate is high. Those that are known to have a vascular component are discussed in the following sections.
A. RETINOPATHY
Retinopathy, followed by macular degeneration, develops in 50–98% of diabetics within 15 years of diagnosis and is the most-devastating complication. It is characterized by basement membrane thickening, pericyte degeneration, microaneurysm formation, and focal capillary closure and acellularity, with subsequent abnormal proliferation of endothelial cells. Although laser treatment can prevent blindness in most cases, vitreous cavity bleeding and retinal detachment are problematic and preventative therapies would be more desirable (Danis et al., 2001).
Among those issues being investigated are the effects of chronic intracellular hyperglycemia and abnormal glycation (Kaban et al., 1999; Noma et al., 2002; Verstappen et al., 2003), disturbances in the polyol pathway (Funatsu et al., 2003), and protein glycation (Funatsu et al., 2002). All sugars, including D-glucose and D-galactose, can initiate protein glycation. They produce stable adducts that accumulate on long-lived biological molecules such as hemoglobin and proteins of the extracellular matrix (e.g., collagen). The rate of this reaction increases with glucose concentration. When basement membranes were glycated heavily in vitro, increased proliferation of endothelial cells and decreased proliferation of pericytes was reported (Kalfa et al., 1995; Beltramo et al., 2003). Synergistic effects of diabetes and hypertension have been reported in an animal model and increased number of caveolae in endothelial cells and pericytes, basement membrane thickening, and decreased cell-cell contact (Hillman et al., 2001). As the most metabolically active tissue in the body, the retina has a high demand for oxygen. In diabetes, capillaries supplying the retina may become clogged, resulting in oxygen deprivation (Barinaga, 1995). The retina then produces VEGF but the vessels don’t form in a proper or orderly manner. A similar process happens in premature infants. If this abnormal development could be stopped or re-directed, the blindness that often ensues could be preventable. One successful approach to therapy for retinopathy is photocoagulation. After this therapy was administered to diabetic retinas, endogenous angiostatin release was detected, which likely contributes to the success of the treatment (Spranger et al., 2000). Some studies have linked high production of VEGF and low production of endostatin with severity of diabetic retinopathy (Noma et al., 2002), diminished benefits from surgical intervention to treat cataracts (Funatsu et al., 2003), and increased risk of macular edema (Funatsu et al., 2002). Inflammatory mechanisms may influence retinal neovascularisation. Induction of cyclooxygenase-2 inhibitors has been shown to inhibit it by antagonizing prostaglandin E2 (Sennlaub et al., 2003).
B. ULCERATIONS AND DRY GANGRENE
Ulceration and dry gangrene occur in diabetes due to insufficiency of circulation to the lower limbs. What is the mechanism by which this hypovascularity develops? Is there a lack of pro-angiogenic cues or responses to them or counterbalancing anti-angiogenic factors? Recently, Regranex (Ortho MacNeil Pharmaceuticals, Raritan, NJ) was approved for use in combating circulation impairment in the lower extremities. It is important to keep in mind that if a person had an undiagnosed cancer at the time of treatment, his/her malignant condition could be made worse as a result of treatment.
C. OTHER COMPLICATIONS
Other complications in which the role of angiogenesis is unknown include glomerulosclerosis, necrotizing papillitis, focal demyelination, and peripheral neuropathy. Peripheral neuropathy is also a side effect of some chemotherapeutic drugs (Verstappen et al., 2003); thus, alternatives to these should be prescribed for diabetic cancer patients. Neuropathy may be related to the polyol pathway, in which aldose reductase and sorbitol dehydrogenase are the two key enzymes that handle excess glucose by converting it to sorbitol (Oates, 2002). Sorbitol toxicity has been implicated in neuron, kidney, and retinal damage. It may act by creating an osmotic gradient leading to swelling or have direct toxicities (Craighead, 1994). Diabetics have impaired wound healing, increased rejection of transplanted organs, and impaired formation of coronary collaterals, all of which have the lack of angiogenesis in common.
|
IX. Understanding the Basis for Selectivity Is the Key to Prudent Trial Design
|
|---|
Anti-angiogenic therapies for cancer patients are most desirable if they don’t inhibit wound healing. Conversely, development of pro-angiogenic drugs for diabetics that work well in the retina and wounds but don’t stimulate tumor growth is needed. Combination therapies that aren’t antagonistic can be developed only if there is detailed analysis of the microenvironment and cell types in the different physiologic and pathologic compartments of the patient, so the desirable selectivity can be achieved and undesirable side effects can be avoided. The key is selectivity as a function of cell type, microenvironment, and for the various pathologies in which abnormal vascularity is a feature.
Much remains to be learned about the mechanisms for selectivity of anti-angiogenic agents in the tumor microenvironment and other locations. There is evidence that the tumor microenvironment is acidotic, hypoxic, and pro-angiogenic, so it could follow that pH or oxygen elevation could enhance effects in desired locations for pro-angiogenesis. Could impairment of intracellular pH regulation be causative in the etiology of the reduced perfusion in the limbs of diabetics? If it is compromised, how? There may be abnormalities in pH regulation in the diabetic retina or the ulcerative wound. If so, could counteractive drugs be developed that stimulate ion transport and encourage angiogenesis? Another aspect of drug potency is pH-dependent activity. Many features must be taken into account to understand the interactions that will lead to a positive, negative, or no effect for each agent.
|
X. Conclusions
|
|---|
In summary, concepts are emerging from research reported on the mechanism of action and basis for specificity of angiostatin. The data indicate that the many new anti-angiogenic agents that have been developed and tested need to be characterized in terms of microenvironmental influences that affect their activity and other undesirable actions in other types of tissue, in patients with and without secondary pathologies. This chapter delineates some of activity parameters for this class of agents that are needed to enable effective translation to the clinic, with predictive information regarding harmful side effects, so that patients at risk can be handled accordingly.
|
ACKNOWLEDGEMENTS
|
|---|
This work was supported by U.S. Public Health Service Grant CA-59960 (M.L.W.) and National Cancer Institute CA86344 (S.V.P.). The authors would like to thank collaborators who contributed to the research described herein. These include Roderick A. Capaldi, Robert L. Capizzi, Adam P. Dicker, Judah Folkman, David W. Golde, Derrick S. Grant, Andrew P. Halestrap, Dennis B. Leeper, Charles S. Owen, and Judith A. Owen. The authors thank Steve Conlon, Robert W. Veneziale, and Soman N. Abraham for contributions towards artwork in Figures 1–4.
|
REFERENCES
|
|---|
Agarwala SS, Kirkwood JM 2003
Temozolomide in combination with interferon -2b in patients with metastatic melanoma: a phase I dose-escalation study. Cancer 97
: 121
–127[CrossRef][Medline]
Agius E, Sagot Y, Duprat AM, Cochard P 1996
Antibodies directed against the ß1-integrin subunit and peptides containing the IKVAV sequence of laminin perturb neurite outgrowth of peripheral neurons on immature spinal cord substrata. Neuroscience 71
: 773
–786[CrossRef][Medline]
Akasaka T, vanLeeuwen RL, Yoshinaga IG, Mihm MC, Byers HR 1995
Focal adhesion kinase (p125FAK) expression correlates with motility of human melanoma cell lines. J Invest Dermatol 105
: 104
–108[CrossRef][Medline]
Akhter S, Nath SK, Tse CM, Williams J, Zasloff M, Donowitz M 1999
Squalamine, a novel cationic steroid, specifically inhibits the brush-border Na+/H+ exchanger isoform NHE3. Am J Physiol 276
: C136
C144
Albanell J, Rojo F, Averbuch S, Feyereislova A, Mascaro JM, Herbst R, LoRusso P, Rischin D, Sauleda S, Gee J, Nicholson RI, Baselga J 2002
Pharmacodynamic studies of the epidermal growth factor receptor inhibitor ZD1839 in skin from cancer patients: histopathologic and molecular consequences of receptor inhibition. J Clin Oncol 20
: 110
–124[Abstract/Free Full Text]
Allt G, Lawrenson JG 2001
Pericytes: cell biology and pathology. Cells Tiss Organs 169
: 1
–11[CrossRef][Medline]
Aoudjit F, Vuori K 2001
Matrix attachment regulates Fas-induced apoptosis in endothelial cells: a role for c-flip and implications for anoikis. J Cell Biol 152
: 633
–643[Abstract/Free Full Text]
Arakaki N, Nagao T, Niki R, Toyofuku A, Tanaka H, Kuramoto Y, Emoto Y, Shibata H, Magota K, Higuti T 2003
Possible role of cell surface H+-ATP synthase in the extracellular ATP synthesis and proliferation of human umbilical vein endothelial cells (HUVECs). Mol Cancer Res 1
: 931
–939[Abstract/Free Full Text]
Baatout S, Cheta N 1996
Matrigel: a useful tool to study endothelial differentiation. Rom J Intern Med 34
: 263
–269[Medline]
Baidas SM, Winer EP, Fleming GF, Harris L, Pluda JM, Crawford JG, Yamauchi H, Isaacs C, Hanfelt J, Tefft M, Flockhart D, Johnson MD, Hawkins MJ, Lippman ME, Hayes DF 2000
Phase II evaluation of thalidomide in patients with metastatic breast cancer. J Clin Oncol 18
: 2710
–2717[Abstract/Free Full Text]
Barinaga M 1995
Shedding light on blindness. Science 267
: 452
–453[Free Full Text]
Baselga J, Rischin D, Ranson M, Calvert H, Raymond E, Kieback DB, Kaye SB, Gianni L, Harris A, Bjork T, Averbuch SD, Feyereislova A, Swaisland H, Rojo F, Albanell J 2002
Phase I safety, pharmacokinetic, and pharmacodynamic trial of ZD1839, a selective oral epidermal growth factor receptor tyrosine kinase inhibitor, in patients with five selected solid tumor types. J Clin Oncol 20
: 4292
–4302[Abstract/Free Full Text]
Batist G, Patenaude F, Champagne P, Croteau D, Levinton C, Hariton C, Escudier B, Dupont E 2002
Neovastat (AE-941) in refractory renal cell carcinoma patients: report of a phase II trial with two dose levels. Ann Oncol 13
: 1259
–1263[Abstract/Free Full Text]
Beasley NJ, Wykoff CC, Watson PH, Leek R, Turley H, Gatter K, Pastorek J, Cox GJ, Ratcliffe P, Harris AL 2001
Carbonic anhydrase IX, an endogenous hypoxia marker, expression in head and neck squamous cell carcinoma and its relationship to hypoxia, necrosis, and microvessel density. Cancer Res 61
: 5262
–5257[Abstract/Free Full Text]
Beltramo E, Buttiglieri S, Pomero F, Allione A, D’Alu F, Ponte E, Porta M 2003
A study of capillary pericyte viability on extracellular matrix produced by endothelial cells in high glucose. Diabetologia 46
: 409
–415[Medline]
Benelli R, Morini M, Carrozzino F, Ferrari N, Minghelli S, Santi L, Cassatella M, Noonan DM, Albini A 2002
Neutrophils as a key cellular target for angiostatin: implications for regulation of angiogenesis and inflammation. FASEB J 16
: 267
–269[Free Full Text]
Bergers G, Song S, Meyer-Morse N, Bergsland E, Hanahan D 2003
Benefits of targeting both pericytes and endothelial cells in the tumor vasculature with kinase inhibitors. J Clin Invest 111
: 1277
–1280[CrossRef][Medline]
Bhargava P, Marshall JL, Bahut W, Rizvi N, Tocky N, Williams JI, Hait H, Song S, Holroyd KJ, Hawkins MJ 2001
A phase I and pharmacokinetic study of squalamine, a novel antiangiogenic agent, in patients with advanced cancers. Clin. Cancer Res 7
: 3912
–3919
Bier H, Reiffen KA, Haas I, Stasiecki P 1995
Dose-dependent access of omurine anti-epidermal growth factor receptor monoclonal antibody to tumor cells in patients with advanced laryngeal and hypopharyngeal carcinoma. Eur Arch Otorhinolaryngol 252
: 433
–439[Medline]
Boyer MJ, Tannock IF 1992
Regulation of intracellular pH in tumor cell lines: influence of microenvironmental conditions. Cancer Res 52
: 4441
–4447[Abstract/Free Full Text]
Braybrooke JP, O’Byrne KJ, Propper DJ, Blann A, Saunders M, Dobbs N, Han C, Woodhull J, Mitchell K, Crew J, Smith K, Stephens R, Ganesan TS, Talbot DC, Harris AL 2000
A phase II study of razoxane, an antiangiogenic topoisomerase II inhibitor, in renal cell cancer with assessment of potential surrogate markers of angiogenesis. Clin Cancer Res 6
: 4697
–4704[Abstract/Free Full Text]
Brewer GJ, Dick RD, Grover DK, LeClaire V, Tseng M, Wicha M, Pienta K, Redman BG, Jahan T, Sondak VK, Strawderman M, LeCarpentier G, Merajver SD 2000
Treatment of metastatic cancer with tetrathiomolybdate, an anticopper, antiangiogenic agent: phase I study. Clin Cancer Res 6
: 1
–10[Abstract/Free Full Text]
Brivio F, Lissoni P, Rovelli F, Nespoli A, Uggeri F, Fumagalli L, Gardani G 2002
Effects of IL-2 preoperative immunotherapy on surgery-induced changes in angiogenic regulation and its prevention of VEGF increase and IL-12 decline. Hepatogastroenterology 49
: 385
–387[Medline]
Brizel DM, Scully SP, Harrelson JM, Layfield LJ, Bean JM, Prosnitz LR, Dewhirst MW 1996
Tumor oxygenation predicts for the likelihood of distant metastases in human soft tissue sarcoma. Cancer Res 56
: 941
–943[Abstract/Free Full Text]
Brizel DM, Schroeder T, Scher RL, Walenta S, Clough RW, Dewhirst MW, Mueller-Klieser W 2001
Elevated tumor lactate concentrations predict for an increased risk of metastases in head-and-neck cancer. Intl J Radiat Oncol Biol Phys 51
: 349
–353[Medline]
Bui MH, Seligson D, Han KR, Pantuck AJ, Dorey FJ, Huang Y, Horvath S, Leibovich BC, Chopra S, Liao SY, Stanbridge E, Lerman MI, Palotie A, Figlin RA, Belldegrun AS 2003
Carbonic anhydrase IX is an independent predictor of survival in advanced renal clear cell carcinoma: implications for prognosis and therapy. Clin Cancer Res 9
: 802
–811[Abstract/Free Full Text]
Burd R, Wachsberger PR, Biaglow JE, Wahl ML, Leeper DB 2001
Absence of Crabtree effect in human melanoma cells adapted to growth at low pH: reversal by respiratory inhibitors. Cancer Res 61
: 5630
–5635[Abstract/Free Full Text]
Canfield AE, Schor AM, Schor SL, Grant ME 1986
The biosynthesis of extracellular-matrix components by bovine retinal endothelial cells displaying distinctive morphological phenotypes. Biochem J 235
: 375
–383[Medline]
Cao Y, Veitonmaki N, Keough K, Cheng H, Lee LS, Zurakowski D 2000
Elevated levels of urine angiostatin and plasminogen/plasmin in cancer patients. Intl J Mol Med 5
: 547
–551[Medline]
Cascone I, Audero E, Giraudo E, Napione L, Maniero F, Philips MR, Collard JG, Serini G, Bussolino F 2003
Tie-2-dependent activation of RhoA and Rac1 participates in endothelial cell motility triggered by angiopoietin-1. Blood 102
: 2482
–2490[Abstract/Free Full Text]
Chen HX, Gore-Langton RE, Cheson BD 2001
Clinical trials referral resource: current clinical trials of the anti-VEGF monoclonal antibody bevacizumab. Oncology 15
: 1017
–1026[Medline]
Chu G, Dewey WC 1988
The role of low intracellular or extracellular pH in sensitization to hyperthermia. Radiat Res 114
: 154
–167[CrossRef][Medline]
Chu GL, Wang Z, Hyun W, Pershadingh H, Fulwyler M, Dewey WC 1990
The role of intracellular pH and its variance in low pH sensitization of killing by hyperthermia. Radiat Res 122
: 288
–293[CrossRef][Medline]
Claesson-Welsh L, Welsh M, Ito N, Ananda-Apte B, Soker S, Zetter B, O’Reilly M, Folkman J 1998
Angiostatin induces endothelial cell apoptosis and activation of focal adhesion kinase independently of the integrin-binding motif RGD. Proc Natl Acad Sci USA 95
: 5579
–5583[Abstract/Free Full Text]
Colorado PC, Torre A, Kamphaus G, Maeshima Y, Hopfer H, Takahashi K, Volk R, Zamborsky ED, Herman S, Sarkar PK, Ericksen MB, Dhanabal M, Simons M, Post M, Kufe DW, Weichselbaum RR, Sukhatme VP, Kalluri R 2000
Anti-angiogenic cues from vascular basement membrane collagen. Cancer Res 1
: 2520
–2526
Coopman PJ, Thomas DM, Gehlsen KR, Mueller SC 1996
Integrin 3ß1 participates in the phagocytosis of extracellular matrix molecules by human breast cancer cells. Mol Biol Cell 7
: 1789
–1804[Abstract]
Craighead JE 1994
Diabetes. In: Rubin E, Farber J, eds. Pathology, chap 22. Philadelphia: J.B. Lippincott Co.; 1148
–1160
Daliani DD, Papandreous CN, Thall PF, Wang X, Perez C, Oliva R, Pagliaro L, Amato R 2002
A pilot study of thalidomide therapy in refractory solid tumour patients. Cancer 95
: 758
–765[CrossRef][Medline]
D’Amato RJ, Loughnan MS, Flynn E, Folkman J 1994
Thalidomide is an inhibitor of angiogenesis. Proc Natl Acad Sci USA 91
: 4082
–4085[Abstract/Free Full Text]
Danielsen T, Rofstad EK 2000
The constitutive level of vascular endothelial growth factor (VEGF) is more important than hypoxia-induced VEGF up-regulation in the angiogenesis of human melanoma xenografts. Br J Cancer 82
: 1528
–1534[CrossRef][Medline]
Danis RP, Ciulla TA, Criswell M, Pratt L 2001
Anti-angiogenic therapy of proliferative diabetic retinopathy. Expert Opin Pharmacother 2
: 395
–407[CrossRef][Medline]
DCCT Research Group 1993
The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 329
: 977
–986[Abstract/Free Full Text]
DeMoraes ED, Fogler WE, Grant DS, Wahl ML, Leeper DB, Zrada S, Malin A, Connors S, Fortier AH, Dabrow M, Sidor C, Capizzi RL 2001
Recombinant human angiostatin (rhA): a phase I clinical trial assessing safety, pharmacokinetics (PK) and pharmacodynamics (PD). In: Abstracts of the 37th Annual Meeting of the American Society for Clinical Oncology. May 12–15, San Francisco, abstract 10
Dinney CP, Bielenberg DR, Perrotte P, Reich R, Eve BY, Bucana CD, Fidler IJ 1998
Inhibition of basic fibroblast growth factor expression, angiogenesis, and growth of human bladder carcinoma in mice by systemic interferon- adminstration. Cancer Res 58
: 808
–814[Abstract/Free Full Text]
Drasdo N, Chiti Z, Owens DR, North RV 2002
Effect of darkness on inner retinal hypoxia in diabetes. Lancet 359
: 2251
–2253[CrossRef][Medline]
Dvorak HF, Nagy JA, Dvorak JT, Dvorak AM 1988
Identification and characterization of the blood vessels of solid tumors that are leaky to circulating macromolecules. Am J Pathol 133
: 95
–109[Abstract]
Eckhardt SG, Burris HA, Eckardt JR, Weiss G, Rodriguez G, Rothenberg M, Rinaldi D, Barrington R, Kuhn JG, Masuo K, Sudo K, Atsumi R, Oguma T, Higashi L, Fields S, Smetzer L, Von Hoff DD 1996
A phase I clinical and pharmacokinetic study of the angiogenesis inhibitor, tecogalan sodium. Ann Oncol 7
: 491
–496[Abstract/Free Full Text]
Eckhardt SG, Rizzo J, Sweeney KR, Cropp G, Baker SD, Kraynak MA, Kuhn JG, Villalona-Calero MA, Hammond L, Weiss G, Thurman A, Smith L, Drengler R, Eckardt JR, Moczygemba J, Hannah AL, Von Hoff DD, Rowinsky EK 1999
Phase I and pharmacologic study of the tyrosine kinase inhibitor SU101 in patients with advanced solid tumors. J Clin Oncol 17
: 1095
–1104[Abstract/Free Full Text]
Eder JP, Supko JG, Clark JW, Puchalski TA, Garcia-Carbonero R, Ryan DP, Shulman LN, Proper J 2002
Phase I clinical trial of recombinant human ednostatin administered as a short intravenous infusion repeated daily. J Clin Oncol 20
: 3772
–3784[Abstract/Free Full Text]
Egginton S, Zhou AL, Brown MD, Hudlicka O 2000
The role of pericytes in controlling angiogenesis in vivo. Adv Exp Med Biol 476
: 81
–99[Medline]
Ekstrand AJ, Cao
TOP
ABSTRACT
I. Introduction
