1. Anti-VEGF Inhibitors
1.1. Bevacizumab
In 2009, the FDA granted accelerated approval for bevacizumab, a humanized anti-vascular endothelial growth factor (VEGF)monoclonal antibody, for use in patients with recurrent glioblastoma
[1][27]. This was in response to initial phase II studies exploring the efficacy of bevacizumab in this patient population, which showed a reduction in tumor size, prolongation of progression-free survival, decreased cerebral edema, and improved neurological symptoms
[2][3][28,29]. Following the results of the phase III, multicenter, randomized study conducted by the European Organization for Research and Treatment of Cancer (EORTC 26101), the FDA granted full approval for treatment of recurrent glioblastoma in 2017.
As an anti-angiogenic agent, bevacizumab works by specifically binding to circulating VEGF-A ligand and consequently inhibiting its binding to cell-surface receptors
[4][5][12,30]. In response, there is a decrease in the growth of microvasculature and blood supply to the tumor, and down-regulation of angiogenesis
[4][5][12,30]. Given the highly vascularized nature of glioblastoma, bevacizumab is theoretically a suitable agent for these tumors, and it remains the most commonly used anti-angiogenic agent in the treatment of recurrent glioblastoma
[6][7][31,32], despite no significant improve in overall survival
[8][33] due to its role in reducing brain edema
[9][34]. The most common adverse events associated with bevacizumab are gastrointestinal perforations, hemorrhage, and arterial thromboembolism
[6][31].
Intriguingly, bevacizumab did not demonstrate an overall survival advantage as part of the first-line treatment for newly diagnosed glioblastoma patients
[8][10][33,35]. As it stands, the standard of care for this patient population continues to be maximal surgical resection followed by adjuvant radiation and concurrent treatment with temozolomide
[10][35].
1.2. Aflibercept
Aflibercept is a human recombinant fusion protein with anti-angiogenic properties that serves as a decoy receptor by binding to VEGF-A, VEGF-B, and placental growth factor (P1GF), thus acting as a “VEGF trap”
[11][36]. Structurally, this protein is composed of the second immunoglobulin (Ig) domain of VEGFR-1 and the third domain of VEGFR-2, fused to the constant region (Fc) of human IgG1
[11][36]. Mechanistically, it has a higher affinity than both VEGFR and bevacizumab for VEGF-A
[11][36]. Furthermore, PlGF has been shown to enhance VEGF signaling activity and mediate angiogenic escape
[12][37], as Batchelor et al. found, levels of P1GF were increased in recurrent glioblastoma patients following treatment with cediranib monotherapy, a VEGFR tyrosine kinase inhibitor
[13][38]. Conceptually, aflibercept should demonstrate increased efficacy in comparison to other anti-angiogenic drugs due to the dual inhibition of both VEGF and P1GF
[11][12][36,37]. However, phase II clinical trials did not find meaningful improvements in survival of patients with recurrent malignant glioma
[12][37]. The most common adverse events associated with this agent are proteinuria, fatigue, injection-site reactions, and hypertension
[11][36].
1.3. Ramucirumab
Whereas the aforementioned agents targeted VEGF, ramucirumab is a human monoclonal antibody that exerts its effect through its high affinity for the extracellular domain of VEGFR-2, blocking its binding to natural ligands
[14][39]. In 2014, ramucirumab received FDA approval as a single-agent treatment for advanced gastric cancer following prior chemotherapy
[14][39]. In a recent non-randomized phase II clinical trial of recurrent glioblastoma patients, ramucirumab was compared to an anti-platelet-derived growth factor receptor (PDGFR) monoclonal antibody, and offered slightly improved progression-free survival and overall survival, with a similar adverse-event profile
[15][40]. Side effects include hypertension, venous thrombosis, diarrhea, and epistaxis
[16][41].
1.4. Dovitinib
In addition to inhibiting VEGFR, dovitinib is also a potent inhibitor of basic fibroblast growth factor (bFGF), another pro-angiogenic growth factor that has been shown to be increased in glioblastoma
[17][18][42,43]. Binding of bFGF to its receptor activates the protein kinase Cα pathway and extracellular signal-regulated kinase (ERK) pathway
[18][43]. In addition, due to the role that bFGF plays in angiogenic escape and bevacizumab resistance in glioblastoma, dovitinib was theorized to offer promise as an anti-angiogenic agent for recurrent glioblastoma
[18][43]. However, the results from a recently published two-arm, phase II clinical trial, compared anti-angiogenic naïve patients with recurrent glioblastoma to patients with glioblastoma progression after prior anti-angiogenic treatment, and dovitinib failed to demonstrate improvements in survival outcomes
[18][43]. Common adverse effects that were reported included appendicitis, fatigue, and thrombocytopenia
[18][43].
2. Small Molecular Tyrosine Kinase Inhibitors (TKIs)
2.1. Sunitinib
Beyond targeting the VEGF pathway extracellularly, interference of downstream signaling molecules has also been examined
[19][20][44,45]. Small-molecule TKIs act by reversible, competitive inhibition of adenosine triphosphate (ATP) binding to the tyrosine domain of VEGFRs
[16][41]. Sunitinib is an oral kinase inhibitor of VEGFR, PDGFR, stem-cell-factor receptor (c-KIT),
RET oncogene tyrosine kinase, FMS-like tyrosine kinase, and colony-stimulating factor-1 receptor
[21][46]. Approved by the FDA as the first anti-VEGF therapy to treat a subset of pancreatic neuroendocrine tumors, this multi-targeted anti-angiogenic TKI blocks downstream signal transduction, thus affecting tumor angiogenesis and growth
[22][47]. The most common adverse reactions include fatigue, diarrhea, nausea, anorexia, vomiting, abdominal pain, hypertension, and thrombocytopenia
[22][47]. Due to its multi-targeted inhibition of angiogenic pathways, sunitinib held promise in glioblastoma therapy. Previously, a phase II study found that single-agent sunitinib therapy in continuous daily dose did not prolong progression-free survival in recurrent glioblastoma
[21][46]. Nevertheless, the STELLAR study, an ongoing multi-center randomized clinical trial, is currently evaluating the efficacy of high-dose, intermittent sunitinib in the treatment of recurrent glioblastoma, compared to lomustine, an alkylating agent of the nitrosourea family capable of permeating the blood–brain barrier
[23][48] (NCT03025893).
2.2. Sorafenib
Similar to sunitinib, sorafenib is another small-molecule TKI with multiple targets including VEGF, PDGFR, and the RAS/RAF/MEK signaling pathways. In a recently published meta-analysis comparing the efficacy and safety between sorafenib and sunitinib as first-line therapy for metastatic renal-cell carcinoma, Deng et al. found that sorafenib did not prolong overall survival as effectively as sunitinib; however, it conferred a lower toxicity
[24][49]. In earlier orthotopic glioblastoma models
[25][50], sorafenib was found to reduce angiogenesis. Following this, a phase II study was conducted to evaluate the efficacy of dual anti-angiogenic therapy with bevacizumab and sorafenib in the treatment of recurrent glioblastoma
[26][51]. While this particular combination did not improve patient outcomes compared to bevacizumab treatment alone, the potential synergistic effect of dual anti-angiogenic therapy simultaneously targeting multiple angiogenic pathways continues to warrant further investigation, as this approach may yield higher clinical efficacy
[26][51]. Common adverse effects of sorafenib therapy include diarrhea, nausea and vomiting, fatigue, rash, and hypertension
[24][49].
2.3. Cediranib
Another multi-kinase inhibitor capable of simultaneously targeting several angiogenic growth factor pathways is cedirnaib, an orally available VEGFR TKI that also targets c-KIT and to a lesser degree, PDGFR
[27][52]. In pre-clinical trials, this small-molecule receptor TKI has shown promising results by reducing microvessel density and metastasis
[28][53]. Additionally, it can be taken orally and is compatible with once-daily dosing due to is half-life of 22 h
[20][45]. However, in a phase III randomized controlled trial, cediranib did not yield any significant improvement in progression-free survival whether in the form of monotherapy or in combination with the chemotherapy agent lomustine, in recurrent glioblastoma patients
[20][45]. Lomustine has been increasingly used as a control arm in clinical trials, in part due to its reputation as the main standard of care for recurrent glioblastoma in Europe where bevacizumab has not been approved, and as such, it remains one of the most widely used drugs, second only to temozolomide, in the treatment of gliomas.
[23][48]. The most commonly reported adverse events include hypertension, dysphonia, fatigue, and diarrhea
[28][53].
2.4. Imatinib
A highly selective inhibitor of the tyrosine kinase family, PDGFR, and c-KIT, imatinib has previously been shown to exert anti-angiogenic effects through inhibition of PDGFR
[29][54]. An earlier phase II study evaluating the efficacy and safety of imatinib in combination with hydroxyurea in patients with recurrent meningioma was one of the first to examine combination therapy
[30][55]. This study found that this combination was well-tolerated among patients, and survival outcomes were significantly improved in recurrent meningioma patients with lower-grade tumors. However, recent results published from an open-label, non-randomized phase II trial evaluating imatinib with and without radiotherapy in newly diagnosed or recurrent glioblastoma failed to show an effect
[31][56]. The differences between these two trials may stem from the inherent differences between the tumors which may impact response to treatment. Notably, unlike slow-growing lower-grade meningiomas, glioblastomas represent the most aggressive adult primary brain tumor
[30][31][55,56]. Common adverse effects that were reported included constipation, fatigue, nausea, and thrombocytopenia
[30][55].
2.5. Pazopanib
Pazopanib is another multitargeted TKI of VEGFR, PDGFR, and c-KIT. Due to its ability to target multiple angiogenic pathways, pazopanib was reasoned to exert strong anti-tumor activity
[32][57]. When the efficacy of pazopanib as a single agent in the treatment of recurrent glioblastoma was evaluated in a phase II single-arm study, progression-free survival was not found to be prolonged at a clinically tolerated dose
[32][57]. Previous clinical trials have also evaluated pazopanib in combination with laptinib, a dual TKI of EGFR and HER-2 receptors, in patients with recurrent glioblastoma
[33][58].
3. Other Anti-Angiogenic Therapies
3.1. Cilengitide
In preclinical models, integrins αvβ3 and αvβ5, were identified to be implicated in angiogenic pathways and in glioblastoma blood vessels and tumor cells
[34][35][59,60]. This was the rationale behind using cilengitide, a pentapeptide integrin inhibitor. While early results from phase I and II trials evaluating cilengitide in recurrent glioblastoma were promising and suggested improvement in survival compared to historical controls
[36][37][61,62], findings from the phase III tCENTRIC trial failed to show improvements in outcome when cilengitide was added to temozolomide, in comparison to the control group
[37][62]. Despite the disappointing outcome, integrins remain a potential valuable target for further review in glioblastoma therapy due to their role in invasion and angiogenesis.
3.2. Marizomib
A recently developed small-molecule proteasome inhibitor, marizomib has been shown to perform a multitude of activities, including induction of apoptosis and down-regulation of cell growth and survival signaling pathways including angiogenesis by interfering with VEGF-dependent migration
[38][39][63,64]. Compared to other proteasome inhibitors, this irreversible inhibitor is unique in that it has been shown to cross the blood–brain barrier, making it a suitable and attractive agent for brain tumors
[40][65]. Most recently, the results from a phase I/II trial in patients evaluating marizomib alone or in combination with bevacizumab in patients with recurrent glioblastoma were published
[40][65]. Both marizomib monotherapy and dual anti-angiogenic treatment in combination with bevacizumab failed to demonstrate a meaningful benefit. Commonly reported adverse effects include hypertension, confusion, headache, and fatigue
[41][66].
4. Discovery of Novel Anti-Angiogenic Therapy Targets
Despite the theoretically suitable mechanism of anti-VEGF therapy and promising preclinical results
[4][42][43][7,12,67], these therapeutic agents have failed to produce definitively favorable outcomes in glioblastoma patients
[8][12][15][20][33,37,40,45]. The constantly evolving genetic composition of glioblastoma leads to high rates of intratumor heterogeneity, which subsequently facilitates anti-angiogenic therapy resistance
[44][45][68,69]. Knowing this, it is evident that targeting the VEGF pathway in glioblastoma via monotherapy is insufficient, and therefore there is an urgent need to discover novel targets that can be used in concert to improve patient survival
[7][32].
Anti-angiogenic therapy can be broken down to two approaches: (1) reducing pro-angiogenic gene expression, and (2) increasing anti-angiogenic gene expression
[46][70]. Apart from VEGF, reduction of other pro-angiogenic factors has also been evaluated
[47][48][71,72]. For instance, IL-8, a pro-inflammatory cytokine, is involved in increased VEGF expression and signaling
[46][49][10,70]. Yamanaka et al. reported that retroviral-mediated transfer of antisense IL-8 led to reduced tumor growth, suggesting its potential as a therapeutic target
[50][73]. Conversely, brain angiogenesis inhibitor 1 (BAI1) is an anti-angiogenic protein whose reduced expression has been noted in several malignancies such as colorectal cancer, renal-cell carcinoma, and glioblastoma
[46][70]. Increased expression of BAI1 via recombinant adenovirus in xenograft models led to reduced tumor vasculature
[51][74]. Other notable targets of this category include angiostatin
[52][53][75,76], endostatin
[54][55][77,78], and thrombospondin
[56][57][79,80]. Although the preliminary results have been encouraging, the clinical efficacy of these targets is yet to be determined.
A recent anti-angiogenic target of interest is epithelial membrane protein-2 (EMP2), a cell-surface protein encoded by growth-arrest-specific 3 (GAS3)/peripheral myelin protein 22 kDa (PMP22) gene family that localizes within the lipid raft domains
[57][58][59][60][80,81,82,83]. Under physiologic conditions, EMP2 appears to stabilize select integrins and modulates their adhesion onto various extracellular matrices
[61][84]. Its expression has been investigated in a number of neoplasms including endometrial carcinoma, breast cancer, and primary brain tumors
[7][62][63][32,85,86].
As EMP2 is involved in a variety of pathologies including non-cancerous diseases
[64][65][66][87,88,89], its signaling mechanism needs further elucidation. Using endometrial cancer xenografts, Gordon et al. demonstrated that modulation of EMP2 expression profoundly changed tumor microvasculature
[67][90]. Under hypoxic conditions, up-regulation of EMP2 promoted VEGF expression through an HIF1
α-dependent pathway whereas reduction of EMP2 directly correlated with reduced HIF1
α and VEGF expression, supporting its involvement in angiogenesis
[67][90]. More recently, Patel et al. investigated the potential impact of bevacizumab treatment on EMP2 levels in a cohort of 12 glioblastoma patients. In paired analysis, EMP2 histological scores were significantly higher following bevacizumab treatment, and this increase was proportional to the length of treatment
[7][32]. More importantly, patients with higher levels of EMP2 had significantly shorter time to repeat surgery, progression-free survival, and overall survival
[7][32]. Such findings underscore the potential to investigate the clinical implications of EMP2 in glioblastoma. Concurrent evaluation of these proteins along with hypoxia-inducible factors such as HIF1
α, or potentially HIF2a, may provide insightful information to better understanding EMP2′s involvement in angiogenesis and designing future therapeutic agents.