Mol. Hum. Reprod. Advance Access originally published online on April 22, 2008
Molecular Human Reproduction 2008 14(5):259-268; doi:10.1093/molehr/gan019
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Antiangiogenic and vascular-disrupting agents in endometriosis: pitfalls and promises
1Department of Gynaecology, Université Catholique de Louvain, Avenue Hippocrate 10, 1200 Brussels, Belgium 2 Research Institute GROW, Department of Obstetrics and Gynaecology, University Hospital Maastricht, Maastricht, The Netherlands 3Organon Biosciences, Schering-Plough, P.O. Box 20 5340, BH, Oss, The Netherlands
4 Correspondence address. Tel: +32-2-764-95-01; Fax: +32-2-764-95-07; E-mail: donnez{at}gyne.ucl.be
 |
Abstract |
|---|
Top
Abstract
IntroductionIt is widely known that angiogenesis plays a key role in endometriotic lesion formation and development. Antiangiogenic treatments aimed at inhibiting new vessel formation have proven efficient in experimental models. However, as antiangiogenic strategies do not target pre-existing pericyte-protected vessels, they require chronic administration and are likely to be beneficial for early-stage disease only or to prevent recurrence after surgery. Moreover, they may have detrimental effects on reproductive function. Vascular-disrupting agents (VDAs) have emerged as a promising new tool for the treatment of tumors. VDAs target established blood vessels, resulting in tumor ischemia and necrosis. These agents may therefore be more efficient against advanced disease. Two major types of VDAs are being developed for cancer: ligand-directed VDAs using antibodies, peptides and growth factors to deliver toxic effectors to tumor endothelium; and small-molecule VDAs exploiting physiological differences between tumor and normal endothelium to induce acute vascular shutdown. The ongoing evolution in genomics and proteomics is revolutionizing the discovery of novel endothelial markers. Several studies suggest that the vasculature of endometriotic lesions may have particular pathophysiological properties, which could be exploited for the development of selective VDAs. The aim of this review is to explore the merits and limitations of vascular therapy for the treatment of endometriosis.
Key words: endometriosis/angiogenesis/vascular-disrupting agent/antiangiogenic therapy/chronic inflammation
|
Introduction |
|---|
Endometriosis is a common benign gynecological disorder, which may develop into aggressive disease in some women. It is defined as the presence of endometrial tissue, including both glandular epithelium and stroma, outside the uterine cavity. Endometriosis is associated with various distressing symptoms such as dysmenorrhea, dyspareunia, pelvic pain and subfertility, and often has serious long-term detrimental effects on the professional, social and marital life of sufferers.
Based on its clinical presentation, the disease can be classified into three different types: ovarian, peritoneal and rectovaginal (or deep-infiltrating) endometriosis (Donnez et al., 1996; Nisolle and Donnez, 1997). Ovarian endometriosis presents as large hemorrhagic cysts of the ovary, whereas peritoneal endometriosis usually appears as superficial implants that are red, black or white in color (as seen during laparoscopy), depending on the level of angiogenic activity, hemorrhage and fibrosis. The third type, rectovaginal or deep invasive endometriosis, is usually found in the pouch of Douglas behind the uterus, in the rectovaginal septum or on the bladder. Deep-infiltrating endometriosis is often an aggressive disease associated with massive fibromuscular reactions, sometimes leading to ureteral obstruction and loss of kidney function (Donnez et al., 2002a).
The broad spectrum of clinical presentations of endometriosis has led to a wide variety of surgical and medical treatment options (Kennedy et al., 2005; Practice Committee of the American Society for Reproductive Medicine, 2006a, b). The efficacy of medical and surgical treatment of endometriosis-associated infertility and pelvic pain is a source of ongoing controversy. Complete resolution of endometriosis is not yet possible and current therapy has three main objectives: (i) to reduce pain; (ii) to increase the possibility of pregnancy and (iii) to delay recurrence for as long as possible. The choice of therapy should be tailored to the individual, taking into account factors such as age, symptoms, reproductive status, future childbearing expectations and initial or recurrent diagnosis.
The main trophic hormone for endometriotic lesions is estrogen (Giudice and Kao, 2004), and so the majority of endocrine and pharmacological therapies currently in use are based on generating a hypoestrogenic milieu. Unfortunately, the disease often recurs after cessation of treatment, underlining the importance of developing new treatment strategies.
Surgical treatment, either conservative or radical, is becoming the treatment of choice. In case of moderate and severe endometriosis-associated infertility, the combined approach (operative laparoscopy and post-operative medical treatment) may be recommended (Donnez et al., 2002b). Cumulative recurrence rates after conservative surgery can be as high as 50–70% (Fedele et al., 1994). It is evident that any treatment option which reduces the need for these invasive procedures would be a welcome alternative.
Thanks to recent advances in the understanding of the mechanisms underlying the development of this multifactorial disease, new therapeutic strategies have been proposed and are currently being tested in experimental models (Ferrero et al., 2005; Mihalyi et al., 2006). Antiangiogenic approaches, designed to prevent new vessel formation, have been the subject of growing interest during the last 10 years (Healy et al., 1998; Groothuis et al., 2005; Becker and D'Amato, 2007). The aim of this review is to explore the potential, limitations and challenges of antiangiogenic therapy for the treatment of endometriosis in terms of efficacy and risk of side effects. Vascular-disrupting agents (VDAs) have recently emerged as promising drugs for targeting vessels in tumors, and their potential use in the context of endometriosis is discussed.
|
Angiogenesis |
|---|
Angiogenesis is the process by which new vessels develop from pre-existing ones. Physiological angiogenesis is essential for embryogenesis, but rarely occurs in adults, except in reproductive tissues and wound healing. In women, profound non-pathological angiogenesis is seen in the ovary during corpus luteum formation and in the endometrium during the menstrual cycle (Smith, 1998). This is unusual since the vasculature is normally in a state of quiescence maintained by a balance between inhibitors and activators of angiogenesis. Disruption of this balance in favor of excessive angiogenesis results in conditions such as rheumatoid arthritis, diabetic retinopathy, psoriasis and cancer. At least four different mechanisms of angiogenesis have been identified: sprouting, intussusception (internal division of vessels resulting in vessel splitting), elongation or incorporation of circulating or progenitor endothelial cells into vessels (Risau, 1997; Girling and Rogers, 2005). It is a complex and precisely regulated process involving a number of coordinated sequences of cellular interactions. The maturation of newly formed angiogenic blood vessels implies the recruitment of pericytes and smooth muscle cells, which are essential to stabilize vessels. In sprouting angiogenesis, angiogenic stimuli induce vascular endothelial cell activation, breakdown of the basement membrane, migration and proliferation of endothelial cells and tube formation (Folkman and Shing, 1992).
|
Angiogenic and mature vessels in endometriotic lesions |
|---|
It is generally recognized that the establishment of an effective blood supply is essential for the survival of endometrial implants and the development of endometriosis (Donnez et al., 1998; Taylor et al., 2002; Groothuis et al., 2005). Neovascularization of newly formed endometriotic lesions is likely to play a key role in their development and persistence, by providing them with nutrients and growth factors and promoting the recruitment of inflammatory cells. However, surprisingly little information is available on the maturity and functionality of the vascular network of endometriotic lesions collected from patients. Jondet et al. (2006) concluded, after a precise evaluation of the microvasculature of endometrium, pelvic endometriotic lesions and deep endometriotic lesions, that different lesions are heterogeneous with respect to their vasculature.
Only a few studies have examined the vasculature of peritoneal endometriotic lesions (Nisolle et al., 1993; Matsuzaki et al., 2001; Khan et al., 2003). Both angiogenic and mature vessels reside in peritoneal endometriosis depending on lesion stage. Early endometriotic lesions, which are considered to be the most active, are characterized by a high vascular density that gives them the typical pink-red appearance (Nisolle et al., 1993; McLaren, 2000). These lesions contain angiogenic vessels, as indicated by the high mitotic index and absence of 
-smooth muscle actin (SMA)-positive pericytes, as well as mature pericyte-covered vessels. There is a greater percentage of mature vessels in black endometriotic lesions, which are thought to be a later stage of lesion development (Matsuzaki et al., 2001).
Ovarian endometriotic lesions are also strongly vascularized with a high proportion (84%) of immature pericyte-free vessels (Hull et al., 2003). The ovarian vascular surface density index is higher in endometriotic than in follicular and luteal ovarian samples (Inan et al., 2003).
Available data on the vascularization of rectovaginal or deep invasive endometriosis are also limited. Deep-infiltrating endometriosis in the sigmoid and rectovaginal septum is well vascularized (Groothuis et al., 2005), and vascularization is increased in deep invasive endometriosis affecting the rectum (Machado et al., 2008). However, most blood vessels appear to be non-angiogenic because they are covered with SMA-positive smooth muscle cells (Conway et al., 2001). These findings clearly show that angiogenesis is frequently associated with the disease and its immediate environment.
|
Early endometriosis development depends on angiogenesis |
|---|
In order to study the initial steps of endometriotic lesion formation, various animal, ex vivo and in vitro experimental models have been developed (Laschke and Menger, 2007). Analysis of endometriotic lesion revascularization in these models shows that an adequate angiogenic response is required for the successful survival and growth of ectopic endometrium (Groothuis et al., 2005; Becker and D'Amato, 2007). Studies on murine models have demonstrated that revascularization of human endometriotic lesions using transplanted human tissue occurs as early as 4–5 days after transplantation (Grümmer et al., 2001; Eggermont et al., 2005). In the dorsal skinfold chamber angiogenesis model, immature vessels in ectopic endometrial implants were already perfused and exhibited typical signs of angiogenesis 4 days after transplantation (Laschke et al., 2005). Blocking the angiogenic cascade impairs endometriosis-like lesion development, as shown in various animal models (Dabrosin et al., 2002; Hull et al., 2003; Nap et al., 2004; Becker et al., 2005; Laschke et al., 2006) and in the chicken chorioallantoic membrane (Nap et al., 2005).
Crucial information on factors triggering angiogenesis and ectopic endometrial tissue revascularization was obtained from these models. As recently reviewed, a chronic inflammatory environment and the innate properties of the human endometrium ensure continuous angiogenic stimuli and impulses for vascular remodeling to meet the needs of growing endometriotic tissue (Groothuis et al., 2005; Becker and D'Amato, 2007). There is an imbalance between pro- and antiangiogenic growth factors in peritoneal fluid from endometriosis patients (Koninckx et al., 1998; Laschke and Menger, 2007), and inflammatory cells such as activated macrophages produce proangiogenic factors, contributing to this proangiogenic milieu (Gazvani and Templeton, 2002; Lin et al., 2006). Interestingly, dendritic cells were recently implicated in supporting angiogenesis in endometriosis. Folkman's team showed that dendritic cells infiltrate sites of angiogenesis in endometriotic lesions and promote their growth in a murine model (Fainaru et al., 2008).
There are numerous indications that the human endometrium itself is highly angiogenic, and it has been hypothesized that pathological changes in angiogenesis in eutopic endometrium can contribute to the initiation of endometriosis (Laschke and Menger, 2007). Endothelial cell proliferation is increased in endometrium from patients with endometriosis (Wingfield et al., 1995) and microvessel density is higher (Bourlev et al., 2006). Several proangiogenic factors are overexpressed or disregulated in endometrium from these patients (Groothuis et al., 2005).
Although many angiogenic factors associated with endometriosis have been identified, the mechanisms underlying revascularization of endometriotic lesions remain largely unknown. It appears that angiogenic processes in endometriosis share common markers with tumor angiogenesis, since many factors overexpressed in endothelial cells from eutopic and ectopic endometrium of endometriosis patients, such as vascular endothelial growth factor (VEGF) receptor-2, endoglin, vβ3 integrin, urokinase-type plasminogen activator, interleukin-8, matrix metalloproteinase-2 and -9 (MMP-2 and -9) and fibronectin (Table I), are also found in activated endothelial cells in tumors (McCarty et al., 2003; Brack et al., 2004). However, it is not clear whether endometriotic lesion revascularization involves sprouting angiogenesis, as observed in tumorogenesis, or elongation and intussusception processes that were recently suggested to underly physiological endometrial capillary growth (Girling and Rogers, 2005). Moreover, one cannot exclude the possibility that different mechanisms are implicated in the vascularization of pelvic, ovarian and rectovaginal endometriosis, whose etiology (Nisolle and Donnez, 1997) and local environment (Machado et al., 2008) appear to be different.
|
|
Antiangiogenic therapy in cancer and endometriosis |
|---|
In the early 1970s, Folkman (1971) demonstrated that angiogenesis is important for the growth and survival of tumor cells, and suggested that targeting the tumor vasculature could be a strategy for the treatment of malignancies. Targeting the endothelium offers numerous advantages: (i) endothelial cells are easily accessible; (ii) drugs do not have to penetrate tissues in order to reach their target cells; (iii) drug resistance is avoided because endothelial cells are genetically very stable due to their low proliferation rate, and most drugs are cytotoxic, directly killing the endothelial cells and (iv) angiogenesis in adults occurs only in a limited number of processes, including menstrual cycles in the endometrium, corpus luteum development, early pregnancy and wound healing. Blood vessels in tumors are disorganized and highly angiogenic, and therefore theoretically sensitive to angiostatic drugs. Angiostatic therapy can be directed at various elements of the angiogenic response, i.e. activation, proliferation, adhesion, migration and maturation of endothelial cells. This has already resulted in a variety of antiangiogenic drugs that are currently being tested in human clinical anti-cancer trials (McCarty et al., 2003). Some drugs inhibit endothelial cell proliferation (TNP-470), some block endothelium-specific integrin survival signaling (i.e. the humanized anti-
vβ3-integrin antibody Vitaxin), some block extracellular matrix breakdown (the MMP inhibitor batimastat), whereas others neutralize activators of angiogenesis (bevacizumab, a humanized VEGF-neutralizing antibody), as reviewed by McCarty et al. (2003). Recently, the FDA approved the use of bevacizumab, an anti-VEGF agent, for the treatment of certain defined cancer indications (Lien and Lowman, 2008). The efficacy of antiangiogenic agents on the establishment, maintenance and progression of endometriosis has been demonstrated in various laboratory and animal models (Dabrosin et al., 2002; Hull et al., 2003; Nap et al., 2004, Becker et al., 2005; Laschke et al., 2006; Becker et al., 2008). Promising antiangiogenic drugs for the treatment of endometriosis include the antiangiogenic agent TNP-470 (Nap et al., 2004), caplostatin, a non-toxic form of TNP-470 (Becker et al., 2006), anginex (Nap et al., 2004), VEGF-A- blocking antibodies and flt-1 decoy receptor (Hull et al., 2003), endostatin (Nap et al., 2004; Becker et al., 2005, 2006), 2-methoxyestradiol (Becker et al., 2008) and selective cyclo-oxygenase inhibitors (Matsuzaki et al., 2004), as recently reviewed by Becker and D'Amato (2007). Transient overexpression of the gene encoding angiostatin was also shown to induce disease regression (Dabrosin et al., 2002). The species of animal used, the model design and the route/dose of antiangiogenic agent administration are summarized in Table II. The encouraging responses observed, however, have yet to be confirmed in a clinical setting (Ferrero et al., 2006).
|
As pointed out by Efstathiou et al. (2005), unlike in murine models, endometriotic disease in humans is dynamic and, at any given time, characterized by reseeding of new endometriotic foci and progression of established lesions.
|
Potential disadvantages of antiangiogenic therapies in endometriosis |
|---|
Inhibitors of angiogenesis interfere with new vessel formation and thus confer a preventive effect. They require chronic administration and are likely to be of particular benefit in early-stage disease or to prevent recurrence after surgery. When diagnosed early enough, endometriotic lesions have not yet progressed beyond the superficial lesion stage, which is usually highly angiogenic (Nisolle et al., 1993; Matsuzaki et al., 2001). Inhibition of angiogenesis should therefore be effective. However, antiangiogenic approaches cannot target mature vessels surrounded by pericytes (Benjamin and Keshet, 1997). This is supported by the observation that angiostatic agents induced a regression of newly formed vessels but not smooth muscle cell-protected vessels in endometriotic lesions established in nude mice (Nap et al., 2004). This limitation may prove particularly critical for the treatment of rectovaginal endometriotic nodules, which contain a high proportion of mature vessels (Groothuis et al., 2005). In this respect, combining treatment inhibiting both angiogenesis and vessel maturation may prove more efficient in the treatment of endometriosis. Laschke et al. (2006) showed, using the skinfold chamber model, that combined inhibition of VEGF, fibroblast growth factor and platelet-derived growth factor was much more effective to prevent vascularization of endometrial grafts than VEGF treatment alone. Nevertheless, the efficacy of antiangiogenic therapies may be limited by the fact that spatiotemporal differences in neovascularization are likely to exist in active endometriotic lesions (Molema, 2002). Parts of the diseased tissue may be dormant, while other parts may be actively proliferating with concomitant intense and, most of the time, heterogeneous neovascularization. Due to this heterogeneity, it is not likely that any single angiostatic compound will be sufficient in itself to treat all diseased tissue (Nap et al., 2004). A significant role for conventional antiangiogenic therapy in the treatment of endometriosis can therefore only be envisaged if combined with other conventional hormone therapies to suppress progression, as the drugs work independently on different cellular targets.
One of the main challenges of antiangiogenic therapies is to selectively target pathological angiogenesis, without altering physiological angiogenesis essential for reproductive function or wound healing. This is particularly crucial in the context of endometriosis pathology, which usually affects young women of reproductive age. Several anticancer drugs with antiangiogenic potential have been found to have a detrimental effect on reproductive function in both animal models and patients (Klauber et al., 1997; Pauli et al., 2005).
There is little available information on the possible side effects of antiangiogenic treatment in the context of endometriosis (Laschke and Menger, 2007). No obvious adverse effects of endostatin treatment were observed on murine reproductive function (Nap et al., 2004; Becker et al., 2005), yielding normal offspring, but angiostatin treatment was shown to impair ovarian function and decrease uterine weight (Dabrosin et al., 2002). However, the most important concern remains the risk of teratogenic effects associated with antiangiogenic therapies in case of pregnancy. VEGFR-2-mediated endothelial cell signals are critical to maintain the functionality of luteal blood vessels during pregnancy (Pauli et al., 2005). Treatment with TNP-470 was found to completely inhibit embryonic growth (Klauber et al., 1997). The safety, efficacy and mechanisms underlying antiangiogenic therapies therefore need to be thoroughly investigated before clinical trials can be envisaged in patients with endometriosis (Ferrero et al., 2006; Becker and D'Amato, 2007).
|
VDAs: definition and potential benefits for endometriosis treatment |
|---|
A new type of vascular-targeted therapy has recently emerged for cancer treatment: VDAs designed to act upon the established vasculature. They differ from antiangiogenic approaches in three key respects: their physiological target, the type or extent of disease likely to be susceptible and the treatment modalities (Table III). Unlike antiangiogenic drugs effective for preventing neovascularization, VDAs target established blood vessels (Siemann et al., 2005). These agents are therefore given acutely and may be particularly effective against advanced disease (Lippert, 2007). Two major types of VDAs are being developed for cancer: small-molecule VDAs that exploit physiological differences between tumor endothelium and normal tissue endothelium to induce acute vascular shutdown in tumors, and ligand-directed VDAs that use antibodies, peptides and growth factors to deliver toxins, procoagulants and proapoptotic effectors to tumor endothelium (Siemann et al., 2005). Unlike angiostatic drugs that inhibit the formation of new vessels, vascular-targeting agents occlude pre-existing blood vessels of lesions (or tumors) to cause cell death from ischemia and hemorrhagic necrosis (Thorpe, 2004). Vascular embolization works in the same way, but ischemia is induced by mechanical obstruction of local arteries (Pelage et al., 2005).
|
When advanced endometriosis is diagnosed, it has usually been present for years. As a result, the vasculature has had time to develop fully and consists mainly of pericyte-protected blood vessels (Groothuis et al., 2005). These vessels are resistant to conventional antiangiogenic therapy (Benjamin and Keshet, 1997; Nap et al., 2004), but may well respond to a vascular-disrupting approach. The search for safe, selective and efficient targeting agents therefore remains one of the major challenges in the pursuit of more effective treatment of cancer, as well as cardiovascular, neurodegenerative and chronic inflammatory diseases (Hajitou et al., 2006). VDAs specifically designed to target endometriotic lesions would thus provide a safer alternative to current antiangiogenic therapy.
|
VDAs as potential endometriosis treatment |
|---|
Small-molecule VDAs
VDAs exploit pathophysiological differences between endothelium in normal and diseased tissue to achieve selective occlusion of blood vessels. Disease-related characteristics that can be taken advantage of include increased proliferation, permeability and reliance on cytoskeletal proteins to maintain cell shape (Thorpe, 2004).
VDAs induce a much greater reduction in blood flow in tumors than in normal tissues, which forms the basis for their acceptance into clinical trials (Tozer et al., 2005). Most VDAs currently under development [i.e. combretastatin A4 disodium phosphate (CA-4-P) and 5, 6-dimethylxanthenone-4-acetic acid (DMXAA)] are microtubule-destabilizing agents. They disrupt the cytoskeleton in rapidly proliferating endothelial cells, which is required to maintain cell shape during this process. In addition, an increase in vascular permeability is observed, resulting in the leakage of proteins, greater interstitial pressure and vessel collapse. This approach causes mitotic arrest and cytotoxicity in endothelial cells and reduced tumor blood flow (Nihei et al., 1999). Phase I and II clinical trials have confirmed the tumor selectivity of CA-4-P and DMXAA, and other small-molecule VDAs are presently being tested (Lippert, 2007). Improved antitumor effects were observed when VDAs were combined with other therapies (Thorpe, 2004). VDAs were shown to destroy the central hypoxic region of the tumor, while antiproliferative agents, such as cisplatin or radiation, killed the rapidly proliferating cells repopulating the tumor. As the actions of these agents are mostly directed toward angiogenic endothelial cells, this approach may be less appropriate to treat long-standing endometriotic lesions, which are mainly vascularized by pericyte-protected blood vessels.
One exception, however, may be DMXAA. In contrast to most other VDAs, the actions of DMXAA are not mediated through tubulin disruption and strongly depend on local production of the inflammatory cytokine tumor necrosis factor (TNF)- ( Baguley and Ching, 2002). The antivascular action of DMXAA involves a cascade of vasoactive processes, including a direct effect on vascular endothelial cells, and indirect effects involving induction of TNF-, serotonin and nitric oxide synthesis. DMXAA is itself an antivascular agent, inducing endothelial apoptosis (Woon et al., 2007). Such damage would expose the basement membrane and in turn result in platelet activation. DMXAA treatment has been shown to cause a measurable reduction in blood flow within 30 min (Zwi et al., 1990). There are also indications that the effect of DMXAA leads to activation of transcription factor NF
B-mediated gene expression (Baguley, 2003). These findings suggest that this subclass of VDAs may also be suitable for use in inflammatory conditions, which are usually associated with increased levels of inflammatory cytokines and NFB activity (Kuldo et al., 2005; González-Ramos et al., 2007). This immunodependency is further supported by the fact that DMXAA was shown to enhance the antitumor effects of immunotherapy and overcome immune resistance (Pedley et al., 1996; Kanwar et al., 2001).
Unfortunately, systemic VDAs are still associated with significant side effects, as they do not selectively target endothelium in diseased tissues. For this reason, they do not yet offer an acceptable alternative for the treatment of benign disease.
Ligand-based targeting
An alternative way of delivering drugs to diseased tissues is by means of selective binding to molecules specific to the endothelium of the tissue. Ligand-directed VDAs contain molecules that specifically bind to antigens on or in endothelial cells to achieve selectivity, and which are coupled to a therapeutic agent that kills the endothelial cells, causing vascular thrombosis. Similarly to small-molecule VDAs, treatment delivery via the blood stream ensures immediate access to luminal endothelial cells.
Ligand-based tumor-targeting approaches have been found to yield ligand concentrations in the tumor microenvironment showing tumor-to-control ratios of >10:1 only a few hours after intravenous administration (Hajitou et al., 2006). In contrast, tumor-to-control ratios for non-targeted chemotherapy are estimated to be <1:10 (Bosslet et al., 1998).
The ligand used for the selective delivery of a therapeutic agent can either be a known endothelial cell luminal surface antigen associated with endothelial cell activation or inflammation, or a disease-specific endothelial cell marker. Indeed, several lines of evidence show that the vasculature of each organ expresses a unique set of functional molecules that dictate endothelial heterogeneity specific to that particular organ or disease state (Rajotte et al., 1998).
Ligand-based inflammatory adhesion molecule targeting
Endometriosis is a chronic inflammatory condition, as evidenced by the local accumulation and activation of various white blood cell populations such as T and B lymphocytes, granulocytes, macrophages and mast cells (Witz et al., 1994; Jones et al., 1998; Matarese et al., 2003; Antsiferova et al., 2005). The inflammatory character of endometriotic lesions is also illustrated by the ‘activated’ state of the endothelium in deep-infiltrating endometriosis, as confirmed by the high expression of inflammation-associated genes like E-selectin and vascular adhesion protein-1 (VCAM-1) (Springer, 1994; Takahashi et al., 2001; Van Langendonckt et al., 2007). On the basis of these characteristics, Eniola and Hammer (2003) developed an interesting vascular-targeting approach to treat chronic inflammatory conditions. The investigators designed a leukocyte mimetic to target the endothelium. Leukocytes usually employ two types of adhesion molecules concurrently to home to diseased tissues: selectins, involved in the initial capture and ‘rolling’ of leukocytes, and cellular adhesion molecules, i.e. intercellular adhesion molecule (ICAM-1) and VCAM-1, responsible for the firm adhesion of leukocytes prior to extravasation. Individually, these proteins are expressed at basal levels in normal tissues too, but they are upregulated in inflamed tissues (Springer, 1994; Takahashi et al., 2001; Barthel et al., 2007). A biodegradable polymer was co-functionalized with the selectin ligand, sialyl LewisX (sLeX), and an antibody against ICAM-1. These particles firmly adhered to substrate surfaces in flow. Particle affinity was determined by the sLeX/ICAM-1 ratio and the density of these molecules on the microsphere. Microspheres with only ICAM-1 or sLeX did not bind under these conditions. This system looks promising for the selective delivery of drugs to inflamed sites, as additional ligands can be added to the polymer microsphere, conferring better lesion selectivity.
Despite the presence of activated leukocytes, ectopic endometrium can persist for years without any signs of necrosis or being cleared. Circumstantial evidence suggests that the (local) immune system may be impaired (Matarese et al., 2003). It may therefore be beneficial to boost the immune system of patients to attack blood vessels in lesions. One way of achieving this is by means of vaccination. The proof of principle of this approach was provided by Chen et al. (2006), who showed that immunity against tumor endothelium can be evoked by immunization with viable human umbilical vein endothelial cells (HUVECs), which resulted in a highly effective antitumor response in Lewis lung carcinoma and myeloma tumor models. Similar observations were made by Wu et al. (2004) in a rat model of endometriosis. The investigators injected Lewis rats bearing experimentally induced endometriosis with HUVECs. The endometriotic implants became smaller, which was associated with a significant decrease in microvessel density. Triggering the immune system to attack blood vessels in endometriotic lesions by means of vaccination therefore shows promise. However, safeguarding the selectivity of the response poses a major challenge.
Ligand-based disease-specific molecule targeting
Potential targets for ligand-based vascular disruption in cancer include VEGFR, vβ3 integrin, fibronectin and annexin A1 (Brack et al., 2004). Most of these markers, typically associated with endothelial cell activation, were also reported to be overexpressed in endothelial cells of eutopic endometrium or endometriotic lesions (Table I). For example, vβ3 integrin and extra-domain B of fibronectin, which were shown to be overexpressed in endothelial cells from endometrium in case of endometriosis (Hii and Rogers, 1998; Sha et al., 2007) are absent or minimally expressed in normal endothelial cells, but induced in the angiogenic vasculature of tumors. Both have been extensively studied as possible targets for the delivery of toxic drugs, and have been successfully used for the selective destruction of blood vessels in tumors (Nilsson et al., 2001; Hood et al., 2002).
During this last decade, the discovery of novel disease- and tissue-specific endothelial cell markers that may be used as vascular addresses has opened up new perspectives for the selective delivery of VDAs (Siemann et al., 2005).
Conjugating these targeting peptides to an apoptosis-inducing peptide allowed selective ablation of prostate and white fat tissue in mice (Rajotte and Ruoslahti, 1999; Arap et al., 2002; Kolonin et al., 2004). Radioimmunotherapy to annexin A1 was shown to effectively destroy tumors and increase animal survival (Oh et al., 2004). Similarly, use of cancer-homing peptide linked to liposome-carrying chemotherapeutic drugs proved efficient for disrupting tumor vessels in lung cancer-bearing xenografted mice (Lee et al., 2007). These studies in experimental animal models provide the proof of concept for ligand-based vascular targeting. While the development of ligand-directed vascular disrupting therapeutics is still in its infancy, the findings of these studies are nevertheless very important, indicating that vascular therapy may well be applicable to treat endometriosis in the future. Endothelial cells are genetically stable and therefore alterations in endothelial cell surface expression must result from changes in the local (tumor) environment. Fernandez-Shaw et al. (1993) clearly showed that women with endometriosis have antiendothelial antibodies in their circulation, which bind more tenaciously to vessels in endometrium and endometriotic lesions. This suggests that novel antigens are exposed to the endothelial cell surface in eutopic and ectopic endometrium of women with endometriosis. It also appears to indicate that in non-malignant diseases too, such as rheumatoid arthritis, cardiovascular disease and endometriosis, antigen expression is dictated by the local inflammatory milieu. This is supported by a recent study in a baboon model of endometriosis, showing that the development of pelvic endometriotic lesions induced a marked increase in expression of the proangiogenic factor Cyr61 in endothelial cells from eutopic endometrium (Gashaw et al., 2006). Endothelial cells recovered from endometrium from ovarian endometriosis patients also exhibited altered gene expression, as evidenced by serial analysis of gene expression and microarray, and upregulation of a number of proteins having no previous association with endothelial cells, such as GREM1 (Sha et al., 2007). Disease-specific molecular signatures may thus be present in the context of such pro-inflammatory diseases.
We recently showed selective upregulation of various molecules in vessels of deep-infiltrating endometriosis, whose biological relevance and applicability for diagnosis and targeted therapy is currently under investigation (Van Langendonckt et al., 2007). In this study, vascular cells were collected from endometriotic nodules by laser capture microdissection and gene expression was analyzed by microarray. A broad range of endothelial cell-specific genes were shown to be upregulated in vessels of deep invasive endometriotic lesions, indicative of the ‘activated’ state of these blood vessels, as well as several other markers with no previous association with endothelial cells, such as matrix Gla protein. Further studies are clearly warranted to map molecular expression in endothelial cells from pelvic, ovarian and rectovaginal endometriotic lesions. Such studies might identify precise targets for the design of more specific VDAs, but are also essential to better understand mechanisms associated with pathological angiogenesis in endometriosis.
Screening approaches to identify potential endothelial markers
Various high-throughput screening approaches have been applied to discover novel endothelial cell surface markers in disease models or human tissue biopsies, as summarized in Table IV.
|
In animal models, screening of tissue-specific endothelial cell surface markers can be achieved by in vivo labeling of vascular structure, followed by recovery of endothelial cells and comparative proteomic or genomic screening approaches (Brack et al., 2004). Several strategies can be used to selectively recover endothelial cells, such as endothelial cell labeling by perfusion of silica beads and recovery from homogenized tissue by density centrifugation (Brack et al., 2004), biotinylation and separation on a streptavidin column (Scheurer et al., 2005), labeling with fluorescent cationic liposomes and selection by fluorescence-activated cell sorting (Favre et al., 2003), and in vivo rhodamine-RCA angiogenic vessel labeling followed by laser capture microdissection (Hunter et al., 2006).
Another approach which is widely applied to identify endothelium-specific markers in animal models is in vivo phage display. Ruoslahti's team developed a procedure to select phage-displayed peptides that home to receptors differentially accessible in the vasculature of specific organs or diseased tissue (Pasqualini and Ruoslahti, 1996). Phage display allowed identification of peptide motifs selectively homing to tissues, including brain, kidney, lung, skin, pancreas, intestine, uterus, adrenal gland, retina, lymph nodes, muscle, prostate, breast, placenta and white adipose tissue, as reviewed by Sergeeva et al. (2006). The selectivity of this approach may be enhanced by converting the phage-displayed antibodies into scFv-Fc fusion proteins, which are then able to rapidly target the selected organ(s) in vivo (Valadon et al., 2006).
However, because of species-specific differences in expression of many vascular targets, some endothelial markers evidenced in murine experimental models could not be translated into real clinical applications (Sergeeva et al., 2006). To overcome this problem of species specificity, endothelial cells may be recovered from patient biopsies for gene or protein expression mapping by classic endothelial cell isolation procedures, such as immunomagnetic cell sorting (Lacorre et al., 2004) or laser capture microdissection (Kinnecom and Pachter, 2005; Van Langendonckt et al., 2007). The purity of endothelial cell preparations and the maintenance of protein and RNA integrity are key factors in the success of these procedures. Alternatively, endothelial cells may also be enriched and analyzed after in vitro culture (Brack et al., 2004). However, isolating endothelial cells from their in vivo environment extensively and rapidly alters their protein expression (Lacorre et al., 2004). For example, 40% of endothelial cell surface proteins expressed in rat lungs in vivo are not detected in isolated rat lung endothelial cells grown in cell culture (Durr et al., 2004).
These screening approaches have revealed a wide heterogeneity in endothelial cell expression and allowed identification of several promising tissue-specific targets.
Vascular embolization
Another emerging vascular-disrupting technique is vascular embolization, which has already made its way into clinical practice for the treatment of uterine adenomyosis and fibroids (Goodwin et al., 1999; Pelage et al., 2005). The uterine artery is occluded by injecting 355–500 µm polyvinyl alcohol particles after selective catheterization (Siskin et al., 2001). Uterine fibroid embolization was shown to contribute to the improvement of menorrhagia and pelvic pain symptoms in 
80–90% of women and resulted in substantial fibroid and uterine shrinkage. Artery embolization in adenomyosis led to significant relief of symptoms, i.e. abnormal uterine bleeding and dysmenorrhea associated with decreased uterine size and junctional zone thickness (Goodwin et al., 1999; Pelage et al., 2005; Lohle et al., 2007; Siskin et al., 2001). Endometriotic lesions and fibromuscular tissue are also strongly innervated (Anaf et al., 2000; Tokushige et al., 2006) and the extent of innervation correlates with preoperative pain scores, dysmenorrhea and deep dyspareunia. Surgical removal of lesions was shown to significantly improve pain scores (Anaf et al., 2000). Mechanical obstruction of local blood flow, resulting in smaller lesion size, should therefore lead to reduced pressure pain.
|
Conclusions |
|---|
Endometriosis is known to be a chronic inflammatory and angiogenesis-dependent multifactorial disease. Encouraging results obtained in experimental models and the growing need for new therapeutic modalities make vascular therapy an attractive option for the treatment of endometriosis. The vasculature is a promising target because of its genetic stability, easy access via the circulation and amplifying action during treatment.
Antiangiogenic agents tested so far have proven effective for preventing neovascularization of endometriotic lesions and are likely to be efficient for early-stage disease. However, antiangiogenic treatments may alter reproductive function by impairing physiological angiogenesis, the greatest concern being the potential risk of teratogenicity. Vascular therapy is thus particularly challenging in the context of endometriosis, which is a benign disease affecting young women of reproductive age. Another major challenge of vascular therapy in endometriosis is developing more efficient drugs to target pericyte-coated vessels found in more advanced pelvic endometriotic lesions, as well as ovarian and rectovaginal endometriosis.
Identification of tissue-specific and disease-specific endothelial cell markers has opened up new perspectives for the development of novel drugs designed to selectively disrupt tumor vessels. Since inflammatory angiogenesis in endometriosis has many features in common with tumor angiogenesis, it is likely that these findings could be extrapolated to this disease. However, the specifics of each individual chronic inflammatory situation should be taken into account.
The molecular signature on the endothelial cell surface of any organ or diseased tissue is distinct from that found in other tissues, and can be probed for specific cell surface antigens. Alternatively, the vasculature of lesions may have different pathophysiological properties from normal vessels, which can be exploited for the development of selective drugs. Unfortunately, the initial promise of vascular therapy has not yet been fulfilled, mainly due to the limited selectivity of the vascular therapies available. The ongoing evolution in genomics and proteomics is revolutionizing the discovery of novel, disease-specific endothelial markers, leading to improved ligand-based treatments. A better understanding of the mechanisms underlying physiological angiogenesis in reproductive tissues, and those responsible for pathological angiogenic processes in endometriosis, is essential to design effective, selective and safe strategies for vascular therapy.
Combining angiostatic agents and VDAs with medical and surgical therapies in an adjuvant setting, e.g. with a view to preventing relapse or improving drug efficacy, may serve to accelerate the introduction of vascular drugs in the context of endometriosis treatment. This would probably be the quickest and most efficient way of enhancing current treatment modalities.
|
Funding |
|---|
The present study was supported by grants from ‘the Région Wallonne’ and the ‘Fonds Special de la Recherche’ of the Université catholique de Louvain, and by grant no 1.5.010.06 from the ‘Fonds National de la Recherche Scientifique de Belgique’.
|
Acknowledgements |
|---|
The authors thank Mira Hryniuk, B.A., for reviewing the manuscript.
|
References |
|---|
Anaf V, Simon P, El N I, Fayt I, Buxant F, Simonart T, Peny MO, Noël JC. Relationship between endometriotic foci and nerves in rectovaginal endometriotic nodules. Hum Reprod (2000) 15:1744–1750.
Antsiferova YS, Sotnikova NY, Posiseeva LV, Shor AL. Changes in the T-helper cytokine profile and in lymphocyte activation at the systemic and local levels in women with endometriosis. Fertil Steril (2005) 84:1705–1711.[CrossRef][Web of Science][Medline]
Arap W, Haedicke W, Bernasconi M, Kain R, Rajotte D, Krajewski S, Ellerby HM, Bredesen DE, Pasqualini R, Ruoslahti E. Targeting the prostate for destruction through a vascular address. Proc Natl Acad Sci USA (2002) 99:1527–1531.
Baguley BC. Antivascular therapy of cancer: DMXAA. Lancet Oncol (2003) 4:141–148.[CrossRef][Web of Science][Medline]
Baguley BC, Ching LM. DMXAA: an antivascular agent with multiple host responses. Int J Radiat Oncol Biol Phys (2002) 54:1503–1511.[CrossRef][Web of Science][Medline]
Barthel SR, Gavino JD, Descheny L, Dimitroff CJ. Targeting selectins and selectin ligands in inflammation and cancer. Expert Opin Ther Targets (2007) 11:1473–1491.[CrossRef][Web of Science][Medline]
Becker CM, D'Amato RJ. Angiogenesis and antiangiogenic therapy in endometriosis. Microvasc Res (2007) 74:121–130.[CrossRef][Web of Science][Medline]
Becker CM, Sampson DA, Rupnick MA, Rohan RM, Efstathiou JA, Short SM, Taylor GA, Folkman J, D'Amato RJ. Endostatin inhibits the growth of endometriotic lesions but does not affect fertility. Fertil Steril (2005) 84(Suppl 2):1144–1155.[CrossRef][Web of Science][Medline]
Becker CM, Wright RD, Satchi-Fainaro R, Funakoshi T, Folkman J, Kung AL, D'Amato RJ. A novel noninvasive model of endometriosis for monitoring the efficacy of antiangiogenic therapy. Am J Pathol (2006) 168:2074–2084.
Becker CM, Rohwer N, Funakoshi T, Cramer T, Bernhardt W, Birsner A, Folkman J, D'Amato RJ. 2-Methoxyestradiol inhibits Hypoxia-Inducible Factor-1{alpha} and suppresses growth of lesions in a mouse model of endometriosis. Am J Pathol (2008) 172:534–544.
Benjamin LE, Keshet E. Conditional switching of vascular endothelial growth factor (VEGF) expression in tumors: induction of endothelial cell shedding and regression of hemangioblastoma-like vessels by VEGF withdrawal. Proc Natl Acad Sci USA (1997) 94:8761–8766.
Bosslet K, Straub R, Blumrich M, Czech J, Gerken M, Sperker B, Kroemer HK, Gesson JP, Koch M, Monneret C. Elucidation of the mechanism enabling tumor selective prodrug monotherapy. Cancer Res (1998) 58:1195–1201.
Bourlev V, Volkov N, Pavlovitch S, Lets N, Larsson A, Olovsson M. The relationship between microvessel density, proliferative activity and expression of vascular endothelial growth factor-A and its receptors in eutopic endometrium and endometriotic lesions. Reproduction (2006) 132:501–509.
Brack SS, Dinkelborg LM, Neri D. Molecular targeting of angiogenesis for imaging and therapy. Eur J Nucl Med Mol Imaging (2004) 31:1327–1341.[Web of Science][Medline]
Chen XY, Zhang W, Zhang W, Wu S, Bi F, Su YJ, Tan XY, Liu JN, Zhang J. Vaccination with viable human umbilical vein endothelial cells prevents metastatic tumors by attack on tumor vasculature with both cellular and humoral immunity. Clin Cancer Res (2006) 12:5834–5840.
Conway EM, Collen D, Carmeliet P. Molecular mechanisms of blood vessel growth. Cardiovasc Res (2001) 49:507–521.
Dabrosin C, Gyorffy S, Margetts P, Ross C, Gauldie J. Therapeutic effect of angiostatin gene transfer in a murine model of endometriosis. Am J Pathol (2002) 161:909–918.
Donnez J, Nisolle M, Smoes P, Gillet N, Beguin S, Casanas-Roux F. Peritoneal endometriosis and "endometriotic" nodules of the rectovaginal septum are two different entities. Fertil Steril (1996) 66:362–368.[Web of Science][Medline]
Donnez J, Smoes P, Gillerot S, Casanas-Roux F, Nisolle M. Vascular endothelial growth factor (VEGF) in endometriosis. Hum Reprod (1998) 13:1686–1690.
Donnez J, Nisolle M, Squifflet J. Ureteral endometriosis: a complication of rectovaginal endometriotic(adenomyotic) nodules. Fertil Steril (2002) a 77:32–37.[Web of Science][Medline]
Donnez J, Chantraine F, Nisolle M. The efficacy of medical and surgical treatment of endometriosis-associated infertility: arguments in favour of a medico-surgical aproach. Hum Reprod Update (2002) b 8:89–94.
Durr E, Yu J, Krasinska KM, Carver LA, Yates JR, Testa JE, Oh P, Schnitzer JE. Direct proteomic mapping of the lung microvascular endothelial cell surface in vivo and in cell culture. Nat Biotechnol (2004) 22:985–992.[CrossRef][Web of Science][Medline]
Efstathiou JA, Sampson DA, Levine Z, Rohan RM, Zurakowski D, Folkman J, D'Amato RJ, Rupnick MA. Nonsteroidal antiinflammatory drugs differentially suppress endometriosis in a murine model. Fertil Steril (2005) 83:171–181.[CrossRef][Web of Science][Medline]
Eggermont J, Donnez J, Casanas-Roux F, Scholtes H, Van Langendonckt A. Time course of pelvic endometriotic lesion revascularization in a nude mouse model. Fertil Steril (2005) 84:492–499.[CrossRef][Web of Science][Medline]
Eniola AO, Hammer DA. Artificial polymeric cells for targeted drug delivery. J Control Release (2003) 87:15–22.[CrossRef][Web of Science][Medline]
Fainaru O, Adini A, Benny O, Adini I, Short S, Bazinet L, Nakai K, Pravda E, Hornstein MD, D'Amato RJ, et al. Dendritic cells support angiogenesis and promote lesion growth in a murine model of endometriosis. FASEB J (2008) 22:522–529.
Favre CJ, Mancuso M, Maas K, McLean JW, Baluk P, McDonald DM. Expression of genes involved in vascular development and angiogenesis in endothelial cells of adult lung. Am J Physiol Heart Circ Physiol (2003) 285:1917–1938.
Fedele L, Bianchi S, Di Nola G, Candiani M, Busacca M, Vignali M. The recurrence of endometriosis. Ann N Y Acad Sci (1994) 734:358–364.[CrossRef][Web of Science][Medline]
Fernández-Shaw S, Hicks BR, Yudkin PL, Kennedy S, Barlow DH, Starkey PM. Anti-endometrial and anti-endothelial auto-antibodies in women with endometriosis. Hum Reprod (1993) 8:310–315.
Ferrero S, Abbamonte LH, Anserini P, Remorgida V, Ragni N. Future perspectives in the medical treatment of endometriosis. Obstet Gynecol Surv (2005) 60:817–826.[CrossRef][Web of Science][Medline]
Ferrero S, Ragni N, Remorgida V. Antiangiogenic therapies in endometriosis. Br J Pharmacol (2006) 149:133–135.[CrossRef][Web of Science][Medline]
Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med (1971) 285:1182–1186.[Web of Science][Medline]
Folkman J, Shing Y. Angiogenesis. J Biol Chem (1992) 267:10931–10934.