Molecular Mechanisms and Future Implications of VEGF/VEGFR in Cancer Therapy

Abstract

Angiogenesis, the sprouting of new blood vessels from existing vessels, is one of six known mechanisms employed by solid tumors to recruit blood vessels necessary for their initiation, growth, and metastatic spread. The vascular network within the tumor facilitates the transport of nutrients, oxygen, and immune cells and is regulated by pro- and anti-angiogenic factors. Nearly four decades ago, vascular endothelial growth factor (VEGF) was identified as a critical factor promoting vascular permeability and angiogenesis, followed by identification of VEGF family ligands and their receptors (VEGFRs). Since then, over a dozen drugs targeting the VEGF/VEGFR pathway have been approved for ~20 solid tumor types, usually in combination with other therapies. Initially designed to starve tumors, these agents transiently “normalize” tumor vessels in preclinical and clinical studies, and in the clinic, increased tumor blood perfusion or oxygenation in response to these agents is associated with improved outcomes. Nevertheless, the survival benefit has been modest in most tumor types, and there are currently no biomarkers in routine clinical use for identifying which patients are most likely to benefit from treatment. However, the ability of these agents to reprogram the immunosuppressive tumor microenvironment into an immunostimulatory milieu has rekindled interest and has led to the FDA-approval of 7 different combinations of VEGF/VEGFR pathway inhibitors with immune checkpoint blockers for many solid tumors in the past 3 years. In this review, we discuss our understanding of the mechanisms of response and resistance to blocking VEGF/VEGFR, and potential strategies to develop more effective therapeutic approaches.

INTRODUCTION/ BACKGROUND

The development of new blood vessels from existing vasculature, termed angiogenesis, is a crucial step in the expansive growth and metastasis of tumors. As tumors increase in diameter beyond several millimeters, they become more hypoxic which limits further growth, but the initiation of angiogenesis can enable small, dormant, avascular tumors to rapidly expand. Angiogenesis was proposed as a therapeutic target for cancer in the 1970s, by Dr. Judah Folkman [1], but it was not until years of preclinical and clinical validation later that this hypothesis was fully accepted as a viable theory. However, the original paradigm of targeting angiogenesis to deprive tumors of nutrients for their growth could not explain why anti-angiogenesis improved the efficacy of co-administered cytotoxic drugs that required the blood supply to reach tumors. A potential explanation for this paradox was provided when one of us (R.K.J.) introduced a counterintuitive hypothesis in 2001 that anti-angiogenic drugs – originally developed to starve tumors – can transiently “normalize” tumor vessels, improve perfusion, decrease hypoxia and improve drug delivery and the outcome of concurrent cytotoxic therapy [2–4].

Vascular endothelial growth factor (VEGF or VEGF-A), initially identified as vascular permeability factor (VPF), is one of the major molecules associated with angiogenesis [5–7]. Since these early discoveries, angiogenesis has been found to be regulated by the balance of numerous pro-angiogenic factors such as VEGF-A, basic fibroblast growth factors (bFGF), platelet-derived growth factor (PDGF), and angiopoietin molecules, as well as endogenous angiogenesis inhibitors such as thrombospondin-1. Moreover, in addition to angiogenesis, tumors can establish a vascular network through the multiple non-angiogenic mechanisms, including the recruitment of bone-marrow derived progenitor cells which then differentiate into endothelial cells; co-opting existing vessels; vessel “mimicry” in which channels comprised of tumor cells act as vasculature; transdifferentiation of cancer cells into endothelial cells; and the splitting of blood vessels, a process known as intussusception – that can confer resistance to antiangiogenic therapy [8].

VEGF/VEGFR family members

VEGF is a 40–45 kDa homodimeric protein secreted by a wide variety of cells in both physiologic and pathologic settings. The VEGF family ligands also include VEGF-B, VEGF-C, VEGF-D, and placenta growth factor (PlGF). VEGF-A is known to be a critical factor modulating endothelial cell sprouting, mitogenesis, cell migration, vasodilation, and vascular permeability. VEGF-B activates embryonic angiogenesis and can increase perivascular cell coverage and mediate metastasis independent of effects via VEGF-A [9]. VEGF-C and VEGF-D regulate lymphangiogenesis, and PlGF is critical for a variety of functions including vasculogenesis, inflammation, wound healing, and survival of certain cancer cells (e.g. medulloblastoma) [10]. VEGF family members exert their biological functions through binding to VEGF receptors (VEGFRs). VEGF-A, VEGF-B, and PIGF binds VEGFR1 primarily localized on blood vascular endothelial cells. VEGFR2 is expressed on both blood vascular and growing lymphatic vessels, and VEGF-A and processed VEGF-C/VEGF-D are ligands for this receptor. VEGFR3 serves as a receptor for VEGF-C and VEGF-D and is expressed on blood vascular and lymphatic endothelial cells. Neuropilins-1 and −2 (NRP-1 and NRP-2) also serve as co-receptors for VEGF ligands [10]. VEGF is comprised of 8 exons and alternative splicing of its mRNA yields six distinct transcripts of 111, 121, 145, 165, 189, and 206 amino acids. VEGF₁₆₅ is the predominant molecular species produced by normal and cancer cells. There are also multiple anti-angiogenic splice isoforms of VEGF, VEGFxxxb, which can counteract activity of their respective pro-angiogenic spliced isoforms. Tumors may overexpress proangiogenic isoforms but downregulate expression of anti-angiogenic isoforms. For example, VEGF_165b is a dominant isoform which can act as a competitive inhibitor to VEGF₁₆₅ [11]. Moreover, studies have identified specific splicing factors which regulate expression of these isoforms and could serve as therapeutic targets to inhibit tumor angiogenesis [12, 13].

Regulation of VEGF expression

VEGF-A is upregulated in tumors by both hypoxia-dependent and -independent mechanisms. In response to hypoxic stress, hypoxia-inducible factor (HIF) is stabilized and directly promotes VEGF-A transcription [14–16]. The hypoxia-dependent accumulation of HIF-1α protein is mediated by a mechanism through which in low oxygen conditions HIF-1α becomes de-hydroxylated and in turn is no longer targeted for degradation by the von Hippel-Lindau (VHL) protein. The 2019 Nobel Prize in Physiology and Medicine was awarded to Drs. Kaelin, Ratcliffe, and Semenza for key discoveries in this area. The HIF pathway, and thus VEGF-A secretion, can be deregulated by genomic alterations in the VHL gene, most commonly observed in renal cell carcinoma (RCC) and von-Hippel Lindau disease. Moreover, abnormal activation of receptor tyrosine kinases in tumor cells can also trigger hypoxia-independent increases in HIF-1α and VEGF-A. For example, in non-small cell lung cancer (NSCLC), EGFR activating mutations increase HIF-1α levels and in turn drive VEGF-A production even in normoxic conditions [17–20]. It has been proposed that this enhanced VEGF-A production via hypoxia-independent activation of HIF-1α promotes a more “VEGF-dependent” phenotype, and greater responsiveness to VEGF-A inhibition, in RCC and EGFR-mutant NSCLC [15, 17, 21, 22]. Studies in neuroblastoma have identified that signaling through multiple receptor tyrosine kinases including RET, PDGFR-α and-β, c-Kit, EGFR, and VEGFR-1, −2 and −3 can drive HIF1α and 2α expression in both normoxic and hypoxic conditions [23]. Finally, there are several transcription factors which can directly regulate production of VEGF-A. AP-1, Sp-1, C/EBPB, STAT3, and 17β-estradiol (E2) [24, 25] are known effectors which can modulate transcription by binding to the VEGF-A promoter [26].

Downstream VEGF signaling

VEGF family members exert their biological effects by binding to VEGFR on target cells and activating downstream signaling pathways (Fig 1). VEGF-A binds VEGFR-1 with a higher affinity than VEGFR-2 though most functional activity results from VEGF-A binding to VEGFR-2 on the tumor endothelium. Ligand binding and receptor dimerization triggers VEGFR-2 auto-phosphorylation [27], and activation of downstream signal transduction including the PI3K, PLC-γ, Akt, Ras, and MAPK pathways promoting proliferation, cell survival, migration, permeability, differentiation, modulation of cell-adhesion molecules [28] and other effects [29]. AXL is a known mediator of VEGF-A dependent activation of PI3K/Akt signaling and the associated changes in vascular permeability [30]. VEGFR signaling is also regulated by the presence of co-receptors such as NRP-1 [31] and co-activating proteins as well as crosstalk with other receptors including fibroblast growth factor receptor (FGFR), AXL, and the Hippo pathway.

Figure 1. The effect of VEGF on the function and growth of endothelial, tumor, and immune cells in the tumor microenvironment.

VEGF drives the formation of new blood vessels within the tumor in part by stimulating endothelial cell sprouting, mitogenesis, and endothelial cell migration. VEGF is not only produced by cancer cells, but also stromal cells, including endothelial cells and fibroblasts. VEGFR-1 is mostly expressed on myeloid cells; VEGFR-2 is primarily expressed on endothelial and sporadically on tumor cells; VEGFR-3 is primarily expressed on lymphatic endothelial cells. Original figure created with Biorender.

MECHANISMS

Impact of VEGF signaling in tumor biology

The initial therapeutic targeting of the VEGF/VEGFR pathway was accomplished by Ferrara and colleagues using a monoclonal antibody against VEGF-A [32]. VEGF blockade resulted in a decrease in tumor vessels, supporting the hypothesis that inhibition of tumor angiogenesis- and the ensuing reduction in tumor perfusion- was the underlying mechanism of tumor growth inhibition. Subsequent studies, however, demonstrated that this hypothesis fails to capture the complex and varied effects of VEGF-A inhibition, especially how anti-VEGF-A therapy can improve the outcome of concurrently administered cytotoxic therapy that requires blood vessels for drug delivery. As noted previously, based on the “vascular normalization” model, VEGF-A inhibitors can transiently normalize the abnormal structure and function of inefficient, tortuous tumor vessels, making them at least temporarily more efficient for oxygen and drug delivery [2, 3]. This model was not accepted initially – as it was against the prevailing hypothesis at that time about using anti-VEGF-A therapy to starve tumors. However, as discussed in more detail below, extensive preclinical and clinical studies provided compelling support for this concept and demonstrated that patients whose tumor perfusion initially increases with anti-VEGF/R agents have better PFS and OS [33, 34]. VEGF-A is also known to directly induce tumor cell invasiveness and motility [23, 35] and to promote immunosuppression in the tumor microenvironment [36]. The major effects VEGF-A signaling on endothelial, tumor, and immune cells are illustrated in Figure 1 and detailed below.

VEGF modulates proliferation of endothelial cells, vascular permeability and recruitment of circulating endothelial progenitors

VEGF-A was initially found to regulate endothelial sprouting, migration and differentiation, predominantly through the binding of VEGF-A and VEGFR-2 [27]. Subsequent studies showed that tumors commonly produced high levels of VEGF-A, often driven by tumor-associated hypoxia, and that such vessels take on an aberrant and tortuous pattern [2]. In addition, VEGF-A increases the microvascular permeability and endothelial fenestrations [37], in part through modulation of adhesion factors, P-selectin, E-selectin, ICAM-1, and inducing synthesis of platelet-activating factor (PAF) [28, 38, 39]. Moreover, autocrine VEGF-A signaling by endothelial cells promotes the development of endothelial cell networks, required to support tumor growth [40]. Other sources of VEGF-A include activated fibroblasts [26], endothelial cells [40], and immune cells [41].

VEGF-A can also initiate the recruitment of circulating endothelial progenitor cells (CEPs), which originate in the bone marrow, and can be incorporated into existing vasculature to promote tumor angiogenesis [42–45]. In addition to CEPs, mature circulating endothelial cells (CECs) can be shed from tumor vasculature, a process enhanced by drugs targeting tumor vasculature [46–50]. For this reason, there have been investigations into the utility of CECs and CEPs as biomarkers for the efficacy of agents targeting the VEGF/VEGFR pathway although this approach has not been validated clinically [51, 52].

VEGF acts directly on tumor cells and promotes an aggressive phenotype

It is well known that tumor hypoxia – potentially exacerbated by anti-VEGF/R therapies – can induce EMT, invasiveness and metastatic spread [33]. More recently, there is a growing appreciation that VEGF-A may also exert direct effects on tumor cells via VEGFRs, including the induction of self-renewing cancer stem cells [53]. Preclinical studies in breast cancer have also shown that VEGF/NRP1-mediated activation of the Wnt/β-catenin pathway [54] and VEGF/NRP2-mediated activation of the Hippo pathway [55] induce cancer cell stemness. Other reports have associated epithelial to mesenchymal transition (EMT)-driven secretion of VEGF-A with the propagation of cancer stem cells and increased tumor angiogenesis [56]. Moreover, VEGF-A signaling can directly promote tumor invasiveness [57], motility [58], and migration [59–62]. Studies in colorectal cancer have identified that intracrine VEGF-A signaling mediates cancer cell migration and invasion [58], and in preclinical NSCLC models, tumor cell growth was found to be mediated by autocrine VEGF-A signaling through co-receptor NRP-1 [62]. In addition, VEGF/VEGFR signaling in tumor cells is known to activate survival pathways and promote chemotherapeutic resistance [63, 64]. Taken together these studies suggest that, in addition to its effects on the tumor microenvironment, VEGF may directly impact tumor cell invasiveness, stemness, and survival.

Immunosuppressive effects of VEGF

VEGF-A can act as an immunosuppressive cytokine in the tumor microenvironment, impairing the function of multiple immune cell types. Specifically, VEGF-A impairs maturation and antigen presentation of dendritic cells which in turn can inhibit T cell mediated cytotoxic killing [65, 66]. Similarly, VEGF-A can inhibit activation of NF-κB in hematopoietic progenitor cells (HPCs) which are supporting players in the activation of dendritic cell differentiation [67].

VEGF-A exerts effects on macrophage function as well. Macrophages that typically exhibit immunosuppressive activity are labelled as M2 macrophages while M1 macrophages display a more pro-inflammatory phenotype. VEGF-A is a primary cytokine secreted by tumor-associated M2 macrophages, which in turn promote tumor vascularization and remodeling [68]. Tumor associated macrophages are a major source of degradative matrix metalloproteinase proteins which can assist with metastatic spread and angiogenesis [69].

VEGF-A also serves to mobilize and attract myeloid derived suppressor cells (MDSCs) from bone marrow to the periphery. MDSCs express VEGF-A as well as VEGFR1 and VEGFR2 on their surface which supports a positive autocrine feedback loop [70, 71]. Granulocytic-MDSCs are a subpopulation of immune cells which are significantly reduced by combination of VEGF-A blockade and chemotherapy in peripheral blood of NSCLC patients [72].

VEGF-A also inhibits antitumor immunity via multiple mechanisms, including effects on T cells and immunosuppressive myeloid populations [73]. Specifically, in the lung, VEGF-A has been shown to increase VEGFR2 receptor expression on T cells [74]. CD8+ T cell activation can also be directly inhibited by signal transduction pathways activated by VEGF [75, 76]. Immunosuppressive Foxp3+ T-regulatory cells also express VEGFR-2 and can be recruited to the tumor microenvironment via VEGF secreted by tumor cells [77]. Moreover, high secretion of VEGF by tumor cells results in an abnormal vascular network that can decrease CD8+ T cell infiltration to the tumor [78]. This VEGF-induced abnormal vascular network can compromise the trafficking of immune cells and delivery of therapeutics including immunotherapeutic agents to the tumor [79]. Moreover, T cell exhaustion marker expression is induced by exposure to VEGF [80]. Taken together, these findings indicate that within the tumor microenvironment, VEGF promotes an immunosuppressive microenvironment through multiple mechanisms.

Mechanisms of resistance to VEGF/VEGFR targeting

Therapeutic strategies targeting the VEGF/VEGFR pathway were initially expected to reduce tumor perfusion, hindering the ability for cells to utilize oxygen and nutrients for survival. It was initially hoped that drugs targeting this pathway would be less prone to the development of therapeutic resistance because the primary targets of these drugs, tumor endothelial cells, are genetically stable compared to tumor cells [81, 82]. It became clear from early trials of VEGF-A inhibition that despite the clear improvements in clinical outcomes particularly in renal cell [83] and in colorectal [84] and NSCLC [85] when used in combination with chemotherapy, both de novo (primary) and acquired resistance were common. The mechanisms underlying this resistance are varied, and appear to involve both tumor cell intrinsic and tumor microenvironmental factors [86].

Specific tumor genomic alterations may promote resistance to VEGF/VEGFR inhibition, potentially by rendering tumor cells less sensitive to oxygen or nutrient deprivation, or, alternatively, by enhancing VEGF-independent tumor angiogenesis as well as other modes of recruiting blood vessels [2]. For example, p53 inactivation is known to render tumor cells less sensitive to hypoxia-induced apoptosis [87]; alter HIF-1 stability and VEGF-A expression [88, 89]; and suppress endogenous angiogenesis inhibitors such as TSP-1 [90]. Similarly, loss of LKB1 protein, typically via mutations in the STK11 gene, is associated with worse outcomes in patients treated with bevacizumab combinations in NSCLC [89].

In patients that initially respond to VEGF/VEGFR inhibition, acquired resistance may emerge through multiple mechanisms. Treatment-induced hypoxia can upregulate expression of multiple other pro-angiogenic signaling molecules in tumor cells [91–93] or cancer-associated fibroblasts [94] changes that may be associated with a shift to a mesenchymal phenotype [95]. Secretion of IL-6 and FGF-2, HGF [96] as well as induction of DLL4/Notch signaling, have all been associated with resistance to VEGF-A therapy [97, 98].

Host factors may also impact response. Estrogen has been implicated as a driver of resistance to bevacizumab therapy in NSCLC models [99]. Vasculogenic mimicry, the development of networks of endothelial-like cells which have vascular characteristics, may in part shape a tumor microenvironment that acquires resistance to VEGF-A therapy and furthers immunosuppression. Altered blood vessel development including vessel cooption [100], post-natal vasculogenesis, and intussusception also have been demonstrated or implicated in resistance to anti-angiogenic therapy [8]. Resistance to VEGF therapy can also be mediated by upregulation of alternative pro-angiogenic molecules including Fgf1/2, Efna1, Angpt1, and others [101]. Moreover, a study using multiple tumor model systems identified the correlation between VEGF-A resistance and production of IL-17 which assists the mobilization of immunosuppressive MDSC cells to the tumor microenvironment [102]. The recruitment of bone marrow-derived cells (BMDCs) [103] and pro-angiogenic monocytes [104] have also been implicated in the development of resistance to VEGF therapies. Resistance to VEGF/VEGFR targeting agents can also be mediated by increased pericyte coverage of tumor blood vessels [105], increases in other factors which promote tumor invasiveness and metastasis [106], and increased hypoxia-induced autophagy [107]. Taken as a whole, these studies demonstrate that a wide range of tumor cell, microenvironmental, and host factors can impact response and resistance to VEGF/VEGFR inhibitors, highlighting the challenges in identifying biomarkers for predicting response and developing combination regimens that broadly target potential resistance pathways.

CLINICAL APPLICATIONS

Approved VEGF/VEGFR targeting agents

The first VEGF-targeting agent approved for the treatment of cancer was bevacizumab, largely used in combination with chemotherapy (Table 1). Bevacizumab is a recombinant humanized monoclonal antibody approved for the treatment of colorectal cancer, NSCLC, glioblastoma, renal cell carcinoma, cervical cancer, ovarian cancer, fallopian tube cancer, peritoneal cancer, and hepatocellular carcinoma. Additionally, there are 3 bio-similars of bevacizumab which have been recently FDA approved with many others in trial [108]. Ramucirumab is a monoclonal antibody against VEGFR2 and was first approved for the treatment of gastric cancer or gastro-esophageal junction adenocarcinoma [109] and later in NSCLC in combination with docetaxel [110] or, more recently, with erlotinib for patients with tumors bearing EGFR mutations [21]. Other multi-kinase inhibitors approved include sunitinib, sorafenib, and pazopanib (VEGFR-1, VEGFR-2, and VEGFR-3, PDGFR-α/β, and c-kit). These agents have generally demonstrated higher anticancer activity compared to single-target agents. The “off-target” receptor specificity for each inhibitor, as well as their pharmacokinetics and potency, are likely to be key determinants of their clinical activity. Aflibercept is a recombinant fusion protein of VEGFR1 and VEGFR2 which binds to VEGF-A, VEGF-B, and PIGF1 and 2 which has displayed efficacy in metastatic colorectal cancer [111]. Outside of RCC, VEGF inhibitors typically have limited activity as monotherapy and are approved for use in combination with chemotherapy, immunotherapy, and EGFR inhibitors [15, 17, 21, 22] [112, 113], although combinations with inhibitors of FGFR [113], PDGFR [114], MET [115], and Ang1/2 [116, 117] pathways have been explored.

Table 1.

Current VEGF/VEGFR Therapy Approvals including those approved for use in combination with anti-PD-1/PD-L1 Therapy.

Approved VEGF/VEGFR Drugs
Agent	Type	Target	Approved Indications*
Bevacizumab	mAb	VEGF-A	CRC♦, NSCLC♦†, platinum-resistant ovarian cancer†, cervical cancer♦, HCC†, breast cancer (Japan and Europe), and glioblastoma
Ramucirumab	mAb	VEGFR-2	Gastric♦, gastro-esophageal junction adenocarcinoma♦, metastatic NSCLC♦
Aflibercept	decoy receptor	VEGF-A, VEGF-B, PIGF	Metastatic CRC♦
Apatinib	RTKI	VEGFR2	Adenoid cystic carcinoma, gastric (Approved by CDFA)
Axitinib	RTKI	VEGFR-1/2/3, PDGFR	RCC†
Cabozantinib	RTKI	VEGFR2, MET, RET, AXL, Flt-3, c-Kit	HCC, RCC†
Cediranib	RTKI	VEGFR-1/2/3, PDGFR-a, CSF-1R, Flt3	Ovarian cancer, BRCA mt HER2- metastatic breast cancer
Levantinib	RTKI	VEGFR-2/3	Thyroid cancer, RCC†, HCC, endometrial carcinoma†
Nintedanib	RTKI	VEGFR-1/2/3, PDGFR, FGFR-1/3	Radiopathic pulmonary fibrosis; NSCLC (EU approval)
Pazopanib	RTKI	VEGFR2	RCC, soft tissue sarcoma
Regorafenib	RTKI	VEGFR-2/3, PDGFR-b, FGFR-1/2, c-Kit, RET, B-Raf	CRC, HCC, GIST
Sorafenib	RTKI	VEGFR-2/3, PDGFR-b, Flt-3, c-Kit, B-Raf	RCC, hepatocellular cancers†, metastatic differentiated thyroid carcinoma
Sunitinib	RTKI	VEGFR-1/2/3, PDGFR, Flt-3, RET	RCC, GIST, NE tumors of the pancreas
Vandetanib	RTKI	VEGFR-2, EGFR, RET	MTC

Agents approved in Combination with anti-PD1/PD-L1 therapy
Agents	Disease	FDA approval
Bevacizumab/ Atezolizumab/ Chemotherapy	NSCLC	2018 [148]
Atixinib/Pembrolizumab	RCC	2019 [149]
Axitinib/Avelumab	RCC	2019 [150]
Bevacizumab/Atezolizumab	HCC	2020 [151]
Cabozantinib/Nivolumab	RCC	2021 [152]
Lenvatinib/Pembrolizumab	Endometrial Cancer (not including MSI-H or dMMR)	2021 [153]
Lenvatinib/Pembrolizumab	RCC	2021 [154]

mAb, monoclonal antibody; mt, mutant; RTKI, receptor tyrosine kinase inhibitor; CRC, colorectal cancer; RCC, renal cell carcinoma, GIST, gastrointestinal stromal tumors; NE, neuroendocrine; HCC, hepatocellular carcinoma; MTC, medullary thyroid cancer; EU, European Union. CDFA, China General Administration of Food Drug Administration

^*FDA-approved unless otherwise indicated.

^♦Indicates agent is approved in combination with cytotoxic therapy.

^†Indicates agent is approved in combination with immunotherapy.

Combinations of VEGF/VEGFR therapy with chemotherapy

Preclinical and clinical studies have demonstrated that anti-angiogenic therapy improves the efficacy of cytotoxic therapies [118, 119]. This might seem counterintuitive given that inhibition of angiogenesis may be expected to reduce drug delivery to tumors. However, the tumor vasculature is known to develop as a disorganized network of abnormal, tortuous vessels during tumorigenesis. These abnormal vascular structures are hyperpermeable and leaky leading to high interstitial fluid pressure [33], which ultimately impairs drug delivery to tumor cells. In preclinical models and in rectal cancer in patients, VEGF/VEGFR inhibition increased mature, pericyte-covered vessels [3, 33, 120], resulting in a vascular normalization or remodeling and decreased tumor interstitial fluid pressure [3, 51, 121–123]. VEGF-A inhibition has also been found to enhance chemotherapy-induced endothelial apoptosis [124]. Combinations of VEGF/VEFR inhibitors with chemotherapy have demonstrated clear benefit in a variety of solid tumor types (Table 1). Moreover, a number of clinical studies have highlighted that patients treated with anti-VEGF/VEGFR therapy alone or in combination with chemotherapy who have increased tumor blood perfusion or oxygenation have better clinical outcomes (Table 2).

Table 2.

Increased blood flow is correlated with better clinical outcome for patients treated with anti-VEGF therapy.

Agents	Disease	Imaging parameter	Clinical Outcome
Cediranib	Recurrent GBM	Increased blood flow (MRI)	Increased PFS and OS [155]
Cediranib + chemoradiotherapy	Newly diagnosed GBM	Increased blood flow (MRI)	Increased PFS and OS [156]
Bevacizumab alone and then with chemotherapy	Advanced NSCLC	Increased blood flow (dCT after bevacizumab single agent)	Increased ORR [157]
Neoadjuvant bevacizumab alone and then with chemotherapy	Chemo-naïve breast cancer	Increased oxygenation (FMISO-PET after bevacizumab single agent)	Increased ORR [158]
Neoadjuvant bevacizumab alone and then with chemotherapy	TNBC	Vessel density and pericyte coverage (IHC in serial biopsies after bevacizumab single agent)	Path response (Miller-Payne score) [159]

GBM, glioblastoma; NSCLC, non-small cell lung cancer; RCC, renal cell carcinoma; TNBC, triple negative breast cancer. PFS, progression-free survival; OS, overall survival; ORR, objective response rate.

VEGF/VEGFR therapy in combination with immune checkpoint inhibitors

Immune checkpoint inhibitors prevent the binding of checkpoint proteins (PD-1, CTLA-4, and LAG-3) on T cells, which are negative regulators of immune response, to their binding partners expressed on tumor cells, thus preventing tumor immune evasion. Since its first approval over a decade ago, immune checkpoint blockade (ICB) therapy has provided durable clinical benefit for many cancer patients but the majority do not initially respond and the determinants of response within the tumor microenvironment remain incompletely understood. The VEGF/VEGFR pathway has emerged as an important regulator of the tumor microenvironment because of the interactions between immune cells and vasculature as well as direct effects of the VEGF/VEGFR pathway on immune cells detailed above. VEGF/VEGFR inhibition would therefore be predicted to enhance the effects of various immunotherapies, including ICB, through multiple mechanisms, a hypothesis supported by multiple preclinical studies [79, 125, 126]. These preclinical results have informed the design of clinical trials on combination of VEGF/VEGFR agents with ICB which ultimately led to 7 FDA approvals in the past 3 years (Table 1). The majority of ongoing clinical trials using VEGFR/VEGFR therapeutics are evaluating these agents in combination with ICB.

Emerging therapeutics to target the VEGF/VEGFR pathway

Given the importance of the VEGF/VEGFR pathway for multiple aspects of tumor growth and progression, there continues to be novel VEGF/VEGFR-targeting therapies in development. Sulfatinib, also known as surufatinib, is a novel multi-target inhibitor of VEGFR-1/2/3, FGFR1, and CSF1R. There have been multiple phase II (NCT02267967, NCT02614495) and phase III (NCT02588170) [127] which have demonstrated its clinical efficacy in patients with solid tumors with more trials ongoing (NCT04579757, NCT02549937). Additionally, concurrent inhibition of VEGF-A signaling in combination with other targets such as c-MET is being explored and has shown promising activity in preclinical studies, including models resistant to VEGF/VEGFR inhibition [128]. Other therapies under initial stages of development include bispecific antibodies like CTX-009 which simultaneously targets delta-like ligand 4 and VEGF-A which is currently being evaluated in additional trials (NCT03292783, NCT04492033). Finally fruquintinib, an inhibitor of VEGFR1/2/3 is being evaluated in phase II (NCT04866108) and phase III (NCT04866108, NCT04322539) trials [129, 130]. Other anti-angiogenic treatments have targeted alternate drivers of angiogenesis beyond VEGF including fibroblast growth factors (FGFs) [131], angiopoietins [116], platelet-derived growth factor (PDGF) [93], hepatocyte growth factor (HGF)/c-MET [132], and placental growth factor (PIGF) [10, 133]. Agents targeting these angiogenic factors have displayed anti-tumor activity in preclinical models. More recently, co-targeting agents have been utilized such as the novel humanized VEGF/ANG2 bi-specific monoclonal antibody which demonstrated an acceptable safety and tolerability profile in a recent first-in-human phase I study in patients with advanced solid tumors [134].

Additional innovative approaches for targeting VEGF/VEGFR signaling are in development. This includes decoy receptor fusion proteins conbercept, which binds VEGFB, PIGF, and VEGFA, and VEGF-grab, which bind to VEGF-A and PIGF. These have both shown promising results in inhibiting tumor growth in models of colon cancer [135, 136]. Monoclonal antibodies such as brolucizumab (DLX1008) have demonstrated pre-clinical activity in Kaposi sarcomas [137], and bispecific antibodies are being evaluated including those which target VEGF-A in combination with Ang-2, PD(L)1, or DLL4 [138–140]. Novel tyrosine kinase inhibitors against VEGFRs like anlotinib, tivozanib, and YLT192 are also being currently investigated [141–143].

FUTURE DIRECTIONS

It is a testament to the progress in the field that there are now more than a dozen VEGF/VEGFR inhibitors approved for routine clinical use (Table 1), covering the majority of common solid tumor types including NSCLC, CRC, RCC, and hepatocellular cancer. At the same time, the magnitude of benefit in most of these tumor types has been relatively modest, and there remain gaping deficiencies in our knowledge about how to optimally use these drugs. For example, to date, there remain no biomarkers in routine clinical use for selecting which patients are likely to benefit from treatment with these drugs, and in many cases it is unclear whether the benefits of treatment occur through effects on tumor angiogenesis, the immune microenvironment, the cancer cells themselves, or some combination of the three (Fig. 1). This remains an active areas of investigation, as the potential use of VEGF-A, VEGFRs, P1GF, IL-6, IL-8, FGF-2, and other angiogenic markers continue to be evaluated as biomarkers to predict clinical outcome and response to VEGF/VEGFR targeted therapies [144–147]. Moving forward, the emergence of next-generation genomic profiling, which is done routinely for many tumor types, as well as growing use of transcriptomic profiling provide the opportunity to examine the association of specific oncogenic drivers (e.g. KRAS, HER2, EGFR), tumor suppressors (e.g. p53, LKB1), or transcriptionally-defined phenotypes (e.g. “inflamed” or “mesenchymal” tumors) and their association with clinical response, and use this information to develop predictive markers and more rational combination regimens. Targeted therapies and immunotherapies are typically being given in a tailored, biomarker-driven manner, and there is an opportunity for VEGF/VEGFR inhibitors to join in this progress.