Пять VEGF-таргетируемых терапий — гуманизированное анти-VEGF моноклональное антитело бевацизумаб и четыре низкомолекулярных ингибитора VEGF рецепторной 2 тирозинкиназы (VEGFR2) (сорафениб, сунитиниб, пазопаниб и акситиниб) — одобрены FDA для лечения пациентов с метастатической RCС [1-3] и несколько других препаратов находятся в процессе исследования. Эти VEGF-таргетируемые тирозинкиназные ингибиторы (TKI) оказали самое большое воздействие на лечение RCС к настоящему времени. Выживаемость без прогрессирования пациентов с RCС, которые принимали сунитиниб или пазопаниб в виде терапии первой линии, например, была более 11 месяцев. Но несмотря на обнадеживающие результаты, по меньшей мере небольшая часть RCС, видимо, является изначально устойчивой к VEGF- таргетируемой терапии, и подавляющее большинство RCС, первоначально отвечающих на эти препараты, позже прогрессируют несмотря на длительное лечение [4, 5]. Неэффективность этих препаратов индуцировать длительные или полные ответы и ограниченное число терапевтических опций, доступных пациентам с RCС, ведет к развитию резистентности к TKI.
Адаптация к стрессу, индуцированному гипоксией и нутриентной депривацией: HIF, AMPK, p53 и UPR
Лечение с VEGFR-таргетируемыми ингибиторами ангиогенеза ведет к временной деэндотелиализации и уменьшению в кровоснабжении опухоли . Падение перфузии опухоли вызывает гипоксию и нутриентную депривацию, которые в свою очередь инициируют адаптивные ответы в клетках, усиливающие выживаемость и способность переносить гипоксию и другие формы клеточного стресса (Фиг. 16.1). Подобные адаптивные ответы продвигаются частично гипоксия-сенситивными транскрипционными факторами HIF-1 и HIF-2. Это увеличение в HIF транскрипторной активности результирует в повышенную экспрессию HIF-зависимых генов, часть из которых промотирирует ангиогенез и поддерживает анаэробный гликолиз как доминантный механизм продукции энергии. Оба из действия усиливают выживаемость в процессе гипоксии.
Активация АМФ киназного пути, ответ на дефолдинг протеинов (UPR) и другие стрессорные пути (Фиг. 16.2) могут также способствовать выживанию опухолевых клеток при ингибировании ангиогенеза. Каждая из этих клеточных адаптаций ассоциирована с повышенной продукцией не только VEGF, но также других факторов ангиогенеза, действия которых не блокируются VEGFR2-таргетируемыми TKI.
Chapter 16. Development of resistance to targeted therapy: preclinical findings and clinical relevance
R.M. Bukowski et al. (eds.), Renal Cell Carcinoma (2015)
Five VEGF-targeted therapies—the humanized anti-VEGF monoclonal antibody bevacizumab and four small molecule VEGF receptor 2 (VEGFR2) tyrosine kinase inhibitors (sorafenib, sunitinib, pazopanib, and axitinib)—are now approved by the FDA for the treatment of patients with metastatic RCC [1–3], and several others are in the developmental pipeline. As a group, these VEGF-targeted tyrosine kinase inhibitors (TKIs) have had the greatest impact on the treatment of RCC to date. The progression-free survival of RCC patients treated with either sunitinib or pazopanib as first-line therapy, for example, is in excess of 11 months. However, despite these encouraging results, at least a minority of RCC appear to be innately resistant to VEGF-targeted therapies, and the overwhelming majority of RCC initially responsive to these drugs later progress despite continued treatment [4, 5]. The failure of these drugs to induce durable or complete responses and the limited number of therapeutic options available to RCC patients once TKI resistance develops have led investigators to redouble their efforts to acquire a more thorough understanding of the molecular mechanisms by which TKI resistance develops. This chapter will review the various experimental models that have informed our current view of this problem, some of which have provided insights into possible therapeutic solutions.
Although there is an abundant literature on the various evasive mechanisms utilized by tumors to escape the effects of angiogenesis inhibitors (summarized in ), most of the experimental data concerning resistance to VEGF-targeted therapies has been generated from xenograft models. Very little of the information that has molded our current understanding of this subject has been derived from analyses of tumor tissue obtained from RCC patients. Furthermore, most of the xenograft studies designed to investigate the problem of TKI resistance have been carried out in tumor types other than RCC (e.g., lung and islet cell carcinomas), and it is possible that the conclusions from these studies may not pertain to a VEGF-driven malignancy such as VHL-deficient RCC. Many of these xenograft studies have made use of agents highly specific for the VEGF signaling pathway (e.g., neutralizing anti-VEGF or anti-VEGFR2 antibodies) and do not take into account the fact that the various FDA-approved TKIs now in common use (e.g., sunitinib, pazopanib) target other kinases (e.g., the PDGF receptors, c-kit) in addition to VEGFR2. Although these previous studies have been informative, many of the candidate resistance mechanisms they have identified involve cytokines (e.g., PlGF, PDGF-C) and signaling pathways known to be blocked by the less specific TKIs. These pathways may therefore not factor into the development of resistance against more broadly targeted TKIs. Many of the mechanisms of TKI resistance discussed in this chapter should therefore be viewed as provisional pending validation in studies based on serial tumor biopsies from patients with RCC.
Adaptation to stress induced by hypoxia and nutrient deprivation: HIF, AMPK, p53, and the UPR
Treatment with a VEGFR-targeted angiogenesis inhibitor results in the transient deendothelialization of the tumor and a reduction in tumor blood flow . This decrease in tumor perfusion causes worsening hypoxia and nutrient deprivation, which in turn trigger adaptive responses in the surviving cells that enhance survival and the ability to tolerate hypoxia and other forms of cellular stress (see Fig. 16.1). These adaptive responses are driven in part by the hypoxia-sensing transcription factors HIF-1 and HIF-2. HIF-2 and in some instances HIF-1 levels are constitutively elevated in clear cell RCC as a result of the loss of VHL function and both are further increased by hypoxia. This increase in HIF transcriptional activity results in the increased expression of HIF-dependent genes, several of which promote angiogenesis and maintain anaerobic glycolysis as the dominant mechanism of energy production. Both of these effects would be predicted to enhance survival in the setting of hypoxia.
The activation of the AMP kinase pathway, the unfolded protein response (UPR), and other stress pathways (Fig. 16.2) may contribute as well to the ability of tumor cells to survive the effects of angiogenesis inhibition. Each of these cellular adaptations is associated with the increased production of not only VEGF but also other angiogenesis factors, the effects of which would not be blocked by VEGFR2-targeted TKIs. These non-VEGF factors could facilitate the restoration of the tumor microcirculation essential for the resumption of tumor growth in the setting of drug-induced VEGFR2 blockade. Treatment-induced hypoxia, for example, is known to increase the production of placental growth factor (PlGF) and stromal-derived factor-1 (SDF-1, CXCL12) by tumor cells and associated stromal elements. The biological activities of neither PlGF nor SDF-1 would be affected by an anti-VEGF antibody, and since their proangiogenic effects are largely due to their engagement of VEGFR1 (flt-1) and CXCR4, respectively, they would not likely to be affected by an anti-VEGFR2 antibody. Thus, either of these two cytokines could theoretically drive angiogenesis and the development of resistance in those situations in which treatment involves either an anti-VEGF or anti-VEGFR2 antibody.
Fig. 16.1. Signaling pathways that contribute to stress tolerance and tumor angiogenesis in the setting of VEGFR blockade. Treatment with a VEGF-targeted therapy induces a transient involution of the tumor microvasculature, resulting in worsening hypoxia and diminished access to glucose. This metabolic stress increases HIF activity and augments the production of VEGF, PlGF, SDF-1, and other HIF-dependent proangiogenic cytokines. The reduced production of energy results in increased AMP levels and the activation of AMPK, which further enhances the production of VEGF independently of HIF. The accumulation of misfolded protein in the endoplasmic reticulum (ER) during hypoxia activates the unfolded protein response (UPR), which selectively increases the translation of mRNAs encoding several angiogenesis factors including VEGF, IL-8, and IL-6. The signaling pathway involved in the increased FGF production in this setting is unclear. The activation of p53 in response to hypoxia increases the ability of cells to tolerate ROS, low glucose, and the absence of certain amino acids (e.g., serine). p53 activation in this setting is apt to be transient and must ultimately be disabled before tumor cell proliferation can resume. The emergence of tumor cells resistant to VEGF-targeted therapy is thought to arise as a result of a selection process favoring tumor cells in which these adaptive pathways are particularly robust
Hypoxia and the lack of glucose limit ATP production, which results in the accumulation of AMP. Increased AMP levels in turn activate AMPK, which phosphorylates numerous substrates that promote energy production and inhibit anabolism. The activation of AMPK retards cell proliferation through its effects on p53 and mTORC1 but preserves cell viability by increasing glucose import and promoting fatty acid oxidation . Although it reduces global protein synthesis through the suppression of the mTORC1 pathway, AMPK activation increases the translation of a select group of mRNAs, many of which encode proangiogenic factors . It is therefore possible that AMPK activation in response to metabolic stress assists in the recovery of the microvasculature from the initial effects of VEGF/VEGFR blockade.
Fig. 16.2. Collaborative interactions between tumor cells, tumor-associated fibroblasts (TAF/CAF), endothelial cells (EC), and various bone marrow-derived myeloid cells present within the tumor infiltrate, including Tie-2-expressing macrophages (TEM) and myeloid-derived suppressor cells (MDSC). Myeloid cells are stimulated by numerous cytokines derived from virtually all of the tumor cellular compartments. They, in turn, produce the angiogenic factors FGF and Bv8 as well as large amounts of MMP9, which drives angiogenesis by liberating VEGF immobilized in the tumor extracellular matrix
Proper protein folding in the ER involves the function of numerous ER-based chaperones, the introduction of disulfide bridges, and the N’-linked attachment of complex mannose-containing branched oligosaccharide chains in order to ensure proper catalytic function of the proteins and to prevent self-association and precipitation of newly synthesized protein within the ER. Each of these processes requires energy, oxygen, and glucose. The metabolic stresses resulting from angiogenesis inhibition can lead to protein misfolding in the ER (i.e., ER stress) and the activation of the unfolded protein response (UPR), an integrated adaptation that enhances protein folding capacity and facilitates the degradation of misfolded protein in the ER [10, 11]. Several of the proteins (e.g., PERK, IRE-1, ATF4, XBP-1) that mediate various aspects of the UPR are absolutely required for cell survival in the setting of hypoxia [12–14]. These same proteins would be predicted to play a similar prosurvival role in tumor cells subjected to the effects of an angiogenesis inhibitor. One of the downstream effects triggered by the UPR is the increased production of angiogenesis factors such as VEGF, IL-8, and IL-6 [15, 16], the latter two of which would be expected to drive angiogenesis independent of the VEGF signaling and to substitute for that cytokine in the restoration of the tumor vasculature during TKI treatment. The AMPK signaling pathway and the UPR are HIF-independent mechanisms activated by hypoxia that could contribute to the reconstitution of a vasculature depleted by VEGF-targeted therapy.
Hypoxia is known to activate the tumor suppressor p53. At least two of the kinases that phosphorylate p53 in response to DNA damage and enhance its stability are redox sensitive and activated by hypoxia . Panka et al. recently showed that p53 is activated in RCC xenografts in response to treatment with sunitinib , presumably as a result of the hypoxia induced by the diminution in tumor perfusion. Although generally regarded as antiproliferative and antiangiogenic, p53 activation can in some circumstances promote cell survival. For example, p53 activation protects cells from the effects of low glucose and enhances cellular tolerance of ER stress . p53 is required for cells to survive in the absence of the amino acid serine  and has an antioxidant effect that increases the ability of cells to tolerate ROS . It is therefore possible that the p53 activation induced early in response to an angiogenesis inhibitor has a transient protective effect against some of the metabolic stresses encountered during treatment. These salutary effects of p53, however, cannot be durable or substantial since the absence of p53 does not sensitize either RCC or CRC xenografts to VEGFR2-targeted drugs—in fact, it has the opposite effect of rendering these agents ineffective [18, 22]. p53 can mediate programmed cell death in response to hypoxia, and it appears that this effect trumps the cytoprotective effects of p53 in the setting of TKI treatment.
Of the various adaptations to metabolic stress, HIF activation may be the only one that promotes angiogenesis without suppressing anabolic pathways in the tumor cell. AMPK activation and the UPR, for example, are both associated with increased production of proangiogenic factors. However, they both limit tumor cell proliferation through their effects on protein translation via TORC1 and eIF-2a phosphorylation, respectively. Although p53 may protect against some forms of metabolic stress, p53 activation is both antiproliferative and antiangiogenic. Thus, although these stressactivated signaling pathways may confer a survival advantage on metabolically stressed tumor cells, their persistent activity is not compatible with tumor growth. The increase in p53 transcriptional activity induced in RCC xenografts during sunitinib treatment, for example, is quite transient . Although p53 persists in tumor cells during treatment, the expression of p53-dependent genes is brief and lost prior to the onset of TKI resistance. In fact, it is possible that the development of resistance to VEGF-targeted therapies requires the subversion of p53 function.
Developmental pathways involved in pathologic angiogenesis: potential value of Notch and BMP blockade as a means of preventing TKI resistance
Several signaling pathways essential for embryonic vascular development play a prominent role in pathologic angiogenesis as well. Although they are not necessarily hyperactivated in response to metabolic stress as is the case with the UPR, AMPK, and other prosurvival pathways discussed in the preceding section, they are essential for tumor angiogenesis and their disruption augments the effects of VEGF-targeted agents. The Notch pathway, for example, is required for tumor microvessel development, and its inhibition enhances the sensitivity of tumor microvessels to the effects of VEGF-targeted therapies [23, 24]. The engineered overproduction of the Notch ligand DLL4 by glioma cells results in large, abnormal tumor vessels that are highly resistant to the effects of VEGF neutralization by bevacizumab or VEGF receptor inhibition . As expected, blockade of the Notch pathway with a ?-secretase inhibitor dibenzazepine blocked drug resistance in these tumors. In another study, neutralization of DLL4 with a specific antibody was shown to give rise to prematurely branched, malformed capillaries that limit tumor perfusion and growth . These microvessels were hypersensitive to VEGFR inhibitors. In fact, DLL4 neutralization was able to render TKI-resistant tumors sensitive to these agents. Despite these encouraging results, however, protracted DLL4 blockade was associated with the involution of the thymus, abnormalities in the hepatic sinusoids, and the development of subcutaneous tumors of vascular origin . These effects suggest that inhibition of the Notch pathway might be fraught with too many side effects to be considered a safe strategy for averting TKI resistance in RCC.
BMP (bone morphogenic proteins)-9 and BMP-10 are ligands for activin-like kinase-1 (ALK-1) and endoglin (CD105, ENG). ALK-1, a type II TGF-Я receptor, and its coreceptor ENG are expressed on endothelial cells and are known to regulate angiogenesis. ALK-1and ENG-deficient mice die of vascular defects early during embryonic development [26, 27]. Patients with mutations in ENG or ALK-1 have the autosomal dominant disorder hereditary hemorrhagic telangiectasia (HHT) [28, 29]. HHT-1 has been attributed to mutations in ENG and HHT-2 to mutations in ALK-1. These disorders are characterized by abnormal vessel development characterized by the formation of telangectasias on the skin and arteriovenous malformations which are predisposed to bleed. It has been shown that BMP-9 and BMP-10 via binding to ALK-1 and ENG mediate angiogenesis in vitro and in vivo [30, 31]. Preclinical studies support the idea that both ALK-1 and ENG may be attractive targets for angiogenesis inhibition. Inhibition of either of these receptors has additive antiangiogenic effects with VEGFR inhibitors [32, 33], suggesting that agents that block BMP-9/BMP-10 signaling may be useful as a means of forestalling TKI resistance.
Reversibility of resistance to VEGF-targeted therapy
The development of resistance to drugs that target receptor tyrosine kinases is not unique to agents that block VEGF signaling. Resistance to drugs that block the mutated EGF receptor in NSCLC, for example, or the Bcr-Abl fusion protein in CML is common—even inevitable—and is often attributable to secondary mutations in the genes encoding the targeted kinases. In some cases, it is due to additional mutations involving genes that encode tyrosine kinases other than the one originally targeted (e.g., c-met) [34–36]. Resistance to VEGFR antagonists, on the other hand, does not appear to have a genetic basis and at least in some circumstances is readily reversible. Zhang et al., for example, have shown that human RCC xenografts that become resistant to sorafenib reacquire their initial sensitivity to the drug when cells from resistant tumors are disaggregated and reimplanted into mice . These sorafenib-resistant xenografts can, in fact, be serially reimplanted and each new implant retraces the growth curve of the first implant, responding initially to the drug and then becoming resistant. Hammers et al. described a similar reversible phenotype in an aggressive RCC that had developed resistance to sunitinib after an initial response to the drug . When implanted into nude mice, this tumor lost many of its aggressive features and acquired a more epithelial phenotype as well as its original sensitivity to VEGF-targeted treatment.
Reversible resistance to VEGFR-targeted therapies is quite familiar to most clinicians who treat RCC patients. It is well known that patients who fail sorafenib or bevacizumab can respond to other VEGFR antagonists such as sunitinib, although the PFS of these patients (5.8 months) is generally less than that reported for patients receiving sunitinib as first-line therapy (11 months) . Patients who develop resistance to sunitinib can even be retreated later with the same agent with some degree of success. In fact, approximately one in four such patients respond to sunitinib “rechallenge” . These observations all attest to the potential reversibility of resistance to VEGF-targeted therapies. They suggest that the underlying mechanism(s) may involve an adaptation to hypoxia and other metabolic stresses and the progressive selection of tumor cells in which the adaptive responses may be particularly robust. It is possible that these adaptations may place the tumor cells at a proliferative disadvantage once the stress is removed—hence the reversion to the initial “sensitive” phenotype when drug treatment is terminated. Why resistance to EGFR inhibitors is permanent and genetically based whereas that which develops to VEGFR antagonists is reversible is unclear. One plausible explanation for this difference is the genetic stability of the cell targeted by drug. The targets of VEGFR antagonists are endothelial cells, which are not particularly prone to mutation, whereas the cells targeted by EGFR inhibitors are genetically unstable tumor cells.
The various mechanisms by which tumors develop resistance to VEGFor VEGFR-targeted therapies are presented in two sections of this chapter, the first of which reviews the contribution of specific cytokines such as HGF and FGF and the second of which discusses the stromal and myeloid cell types that infiltrate tumors and induce resistance through the production of several mediators. Although much of the material presented in this chapter is derived from xenograft studies (mostly non-RCC), the resistance models reviewed involve stereotypical mechanisms by which tumor (and even normal) cells respond to hypoxic stress. The data presented are therefore likely to apply to RCC during treatment with VEGF-targeted therapies.
Enhanced production of alternative proangiogenic growth factors
As mentioned previously, the adaptation to treatment-induced hypoxia involves the activation of HIF, the UPR, and the AMPK pathway and the increased production of factors capable of promoting the restoration of blood flow. Several investigators have, for example, demonstrated increased levels of VEGF and PlGF in the blood of patients undergoing treatment with VEGFR-targeted drugs . These proangiogenic factors were first thought to be produced by ischemic tumor tissue. More recent studies, however, have refuted this notion since the same increase in proangiogenic cytokine levels is observed in tumor-free mice treated with these agents . Others have shown an increase in FGF or IL-8 production by tumor cells and/or their associated stromal elements during treatment with VEGFR antagonists [43, 44]. These two cytokines are of particular interest since, unlike PlGF and VEGF, they activate endothelial signaling pathways not likely to be affected by the VEGFR2 inhibitors currently used to treat RCC. They could therefore promote angiogenesis in the presence of drugs such as sunitinib or pazopanib. Furthermore, their increased production during treatment has been shown to correlate with the development of resistance to VEGF-targeted therapy. The following is a brief review of the various angiogenesis factors that have been shown to contribute to the development of resistance to VEGF-targeted therapies (see Fig. 16.2).
Chemokines are 8–12 kDa proteins produced primarily by inflammatory cells but other cell types, including most tumor cells, are capable of producing them in some circumstances . Chemokines regulate several aspects of leukocyte biology including chemotactic responses, respiration, and metabolism. Some have the ability to promote or suppress angiogenesis in addition to their effects on inflammatory cells. The chemokines can be categorized into four subgroups based on the specific arrangement of certain cysteines within the proteins (i.e., CC, CXC, C, and CX3C), and those with the CXC motif can be further divided into two classes based on whether they have a specific glutamine-leucine-arginine (ELR) motif. Interleukin-8 (IL-8) is one of several ELR(+) CXC chemokines capable of binding to the G proteincoupled receptor CXCR2 present on the endothelium and promoting angiogenesis . Its expression is readily induced by hypoxia, proinflammatory cytokines, and other stimuli , and it is thought to contribute to the development of the tumor microcirculation through the recruitment of inflammatory cells and endothelial progenitors into tumor tissue.
One of the first observations implicating IL-8 in the development of resistance to primary antiangiogenic therapy was that of Mizukami et al., who demonstrated that IL-8 production could compensate for the loss of HIF-1 in DLD-1 colon cancer xenografts . In this study, tumors in which HIF-1 had been knocked down were shown to be well vascularized despite the loss of a transcription factor thought by most investigators to be essential to the generation and maintenance of the tumor microcirculation. These HIF-1-deficient tumors produced large amounts of IL-8, the neutralization of which reduced the tumor microvessel density and retarded tumor growth. This study clearly showed that IL-8 production could maintain the tumor microvasculature in the absence of HIF-1-dependent angiogenesis factors (e.g., VEGF).
The question of whether IL-8 mediates the resistance to sunitinib that inevitably develops in RCC was recently addressed by Huang et al. . These investigators measured the levels of some 89 proangiogenic factors in plasma samples from mice harboring sunitinib-responsive and sunitinib-resistant RCC xenografts and found elevated levels of IL-8 in the mice with resistant tumors. To determine if the increased IL-8 levels were functionally significant, they treated the mice with resistant tumors with either a murine antihuman IL-8 monoclonal antibody, sunitinib alone, or both sunitinib and the anti-IL-8 antibody. Although the antibody alone had no significant antitumor effect, it was able to restore the responsiveness to sunitinib. The sunitinib/anti-IL-8 antibody combination not only inhibited tumor growth but reduced tumor microvessel density, suggesting that the primary effect of IL-8 neutralization was the suppression of tumor angiogenesis. Finally, to determine if these observations were relevant to human RCC, they analyzed IL-8 expression in primary RCC specimens and demonstrated that IL-8 expression and the response to sunitinib treatment were inversely correlated. These data provide the most convincing evidence to date that IL-8 production is an important escape mechanism for RCC subjected to the stress of VEGFR blockade.
Not all CXC chemokines have the ability to promote angiogenesis—in fact, the non-ELR-containing CXC chemokines CXCL9 (Mig), CXCL10 (IP-10), and CXCL11 (ITAC) actually suppress tumor neovascularization . These three chemokines CXCL9–11 are induced by interferon and their production accounts for a substantial portion of the antiangiogenic effects of that cytokine. They all bind the G protein-coupled receptor CXCR3 on endothelial cells and inhibit endothelial proliferation and motility. To determine how sunitinib administration might affect the expression of these angiostatic chemokines, Bhatt et al. analyzed lysates from RCC xenografts by western blot and found that sunitinib treatment down-modulated the expression of these interferon-inducible chemokines as well as that of the interferon-? receptor . To determine the functional significance of these data, they injected recombinant CXCL9 directly into RCC xenografts and noted that, although the injections had little effect on tumor growth by themselves, they delayed the onset of sunitinib resistance. Intratumoral CXCL9 augmented the ability of sunitinib to reduce tumor microvessel density and perfusion, suggesting that the enhanced therapeutic effect of the combination was due to the inhibitory effects of CXCL9 on tumor angiogenesis. It is unknown whether the down-modulation of these non-ELR chemokines that occurs during sunitinib treatment contributes to the development of drug resistance. However, one might predict that the disappearance of these angiostatic chemokines from the tumor might lower the threshold of response to the stimulatory effects of IL-8 and other proangiogenic factors and thereby promote angiogenesis (and the development of resistance) indirectly through this mechanism.
Fibroblast growth factor
One of the first studies to address the mechanism of acquired resistance to VEGFtargeted therapy was performed by Casanovas et al. using the RIP-Tag2 spontaneous islet cell carcinoma model . This tumor is known to respond to agents that block signaling through VEGFR2, but not VEGFR1. Treatment of mice bearing these islet cell tumors with the rat anti-murine VEGFR2 antibody DC-101, for example, induced partial tumor regression accompanied by a marked reduction in tumor vascularity. These antitumor effects, however, were associated with increased tumor aggressiveness manifested as increased invasion by tumor cells into the tumor capsule and infiltration into the surrounding normal pancreatic tissue. This disturbing observation was one of the first to suggest that the hypoxia induced by antiangiogenic therapy might result in a more malignant tumor phenotype.
In the aforementioned Casanovas study, the islet cell carcinomas developed resistance to the DC-101 anti-VEGFR2 antibody within a few weeks despite continued treatment. The resumption of tumor growth was associated with the restoration of the vasculature and the increased expression of several proangiogenic factors including members of the ephrin, angiopoietin, and FGF families. To determine the cellular origin of these factors, cells derived from the tumors were fractionated and the tumor cells and stromal elements were analyzed separately by RT-PCR for mRNAs encoding these proteins. The epithelial tumor cells were shown to produce increased amounts of FGF1, FGF2, FGF7, FGF9, Ephrin A1, and angiopoietin-2 (Ang-2) with the onset of resistance. Tumor-associated endothelial cells expressed an overlapping array of transcripts, including those encoding FGF-1 and FGF-2 as well as angiopoietin-1 and angiopoietin-2. The increased expression of FGF-2 and Ang-1 was also observed at the protein level. Many of these gene products could be induced in islet cell carcinoma cells in vitro by hypoxia, suggesting that the enhanced expression of these genes observed during the course of DC-101 treatment might have been due to the hypoxia that results from the attenuation of the vasculature. To determine if the development of resistance to the DC-101 antibody might have been due to increased production of FGF family members, mice were treated with a soluble form of the FGF receptor FGFR-2 (FGF trap), a protein that binds FGF1, FGF3, and FGF7. The neutralization of these FGFs by the FGF trap significantly delayed the development of resistance to the DC-101 antibody. These studies were the first to implicate FGF production as a strategy by which tumor cells might evade the biological effects of VEGFR2-targeted therapy.
In a related in vitro study with cultured endothelial cells, Welti et al. showed that FGF2 was able to induce endothelial cell proliferation and tubule formation in the presence of the VEGFR antagonist sunitinib . They also showed that human renal cell carcinoma specimens strongly and consistently express FGF2. Together, these observations suggest that tumor cell FGF2 production might be able to override the inhibitory effects of VEGFR2 antagonists on tumor angiogenesis and that FGF2 expression might therefore play a role in the development of resistance to VEGFR2 antagonists. This hypothesis has since been corroborated by xenograft studies examining the antitumor activity of small molecule inhibitors of both VEGFR2 and the FGFR. For example, E-3810, a potent inhibitor of VEGFR-1, VEGFR-2, and VEGFR-3 and FGFR-1 and FGFR-2 tyrosine kinase activities, induces tumor regression in numerous xenograft models, including A498 human RCC xenografts that had become resistant to sunitinib . Collectively, these studies demonstrate that the production of FGF2 enables tumors to maintain their vasculature and to thrive despite treatment with VEGFR antagonists.
Hepatocyte growth factor (HGF)
HGF is produced primarily by the non-endothelial stromal elements within tumors rather than by the tumor cells or associated microvasculature and is readily detectable in most tumors [53, 54]. Its receptor, c-met, is present on some tumor cells but is particularly well expressed by tumor vascular cells including endothelial cells and pericytes. When bound by its ligand HGF, c-met autophosphorylates on certain tyrosine residues, after which numerous adaptor proteins and downstream signaling molecules are recruited to its cytosolic domain. This activates several canonical pathways (e.g., MAPK, PI3-K) shared by other receptor tyrosine kinases that serve to promote proliferation, motility, and survival .
Several previous observations have suggested the involvement of the HGF/c-met signaling pathway in the development of resistance to VEGFR antagonists. For example, Shojaei et al. showed that HGF is more abundant in sunitinib-resistant than in sunitinib-responsive tumors . They also demonstrated that the administration of recombinant HGF reduces the antitumor and antiangiogenic effects of sunitinib in otherwise sensitive tumors. Finally, they showed that the concurrent administration of the selective c-met inhibitor PF-04217903 amplified the antitumor activity of sunitinib. These two drugs individually and in combination retarded the growth of cultured endothelial cells but had no effect on the proliferation of tumor cells in vitro, suggesting that the antitumor activity of the drug combination might be due to the additive effects of the drugs on tumor angiogenesis. It should be kept in mind that these studies were carried out in melanoma and lymphoma cells and their negative conclusions regarding the potential direct effects of c-met inhibition on tumor cells may not apply to RCC. Several studies have, in fact suggested that RCC cells may rely on c-met signaling to maintain their oncogenicity. For example, a synthetic lethal shRNA screen designed to identify kinases whose absence was selectively toxic for VHL(-/-) RCC (but not RCC in which VHL function had been restored) specifically identified c-met as one of the kinases essential for the viability of the VHL-deficient tumor cells . It therefore appears almost a foregone conclusion that a c-met inhibitor would have at least some intrinsic antitumor activity in RCC independent of its antiangiogenic effects. Regardless of whether HGF has direct prosurvival effects on RCC cells or functions solely as an angiogenesis factor, the findings discussed above suggest that the HGF/c-met signaling pathway can be exploited by tumor cells to escape from the effects of VEGF-targeted therapies and that inhibitors of this pathway might prove to be useful adjuncts to VEGFR antagonists in the treatment of RCC.
Epidermal growth factor receptor (EGFR) ligands
The EGFR ligands TGF-a and amphiregulin as well as the EGF receptor (EGFR, erbB1) are abundantly expressed by RCC . This observation suggests that EGF family-EGFR interactions might contribute to the proliferation, invasiveness, or metastatic behavior of RCC cells. This hypothesis was in fact validated in studies by Weber et al., who demonstrated that the development and growth of RCC bone metastases in an orthotopic (i.e., tibial implant) xenograft model were partly dependent on EGFR activation . Given its central role in the biology of RCC, the question has arisen whether EGFR signaling might contribute to the development of resistance to VEGF-targeted therapy. The idea that EGFR ligands might be able to substitute for VEGF in the promotion of angiogenesis was first proposed by Cascone et al. . Using a lung cancer xenograft model, these investigators observed that both de novo and acquired resistance to bevacizumab was associated with reduced endothelial cell apoptosis despite persistent inhibition of VEGFR2 signaling (i.e., absent endothelial cell VEGFR2 phosphorylation). Using species-specific gene expression profiling, they showed that the cells whose gene expression was most affected with the development of resistance were stromal and that many of the changes observed were consistent with enhanced EGFR signaling. They subsequently demonstrated increased EGFR phosphorylation on the endothelial cells of tumors with de novo resistance to bevacizumab and on the pericytes of tumors that developed resistance during treatment. Finally, they showed that concurrent EGFR and VEGFR blockade with either a combination of erlotinib and bevacizumab or with the dual EGFR/VEGFR inhibitor vandetanib yielded superior antitumor activity compared with that induced by VEGFR blockade alone. These data establish a role for EGFR ligands in both primary and acquired resistance to bevacizumab in lung cancer.
Whether EGFR ligands play a similar role in the resistance to VEGFR-targeted therapy that develops in RCC is, however, unclear. There are virtually no studies that support the view that the EGFR is critically involved in the biology of RCC in humans. The PFS of RCC patients receiving the EGFR monoclonal antibody ABXEGF, for example, was only 3 months and only 3 of 88 patients had major responses to the antibody . The gene encoding the EGFR ligand TGF-a is HIF dependent and constitutively overexpressed in VHL-deficient RCC. If TGF-a/EGFR signaling were able to override the effects of VEGFR blockade, the biological consequences of producing large amounts of TGF-a would be apparent at the outset of treatment as most RCC would be resistant de novo to sunitinib and related TKIs. In a recent randomized, double-blinded, phase II clinical trial comparing bevacizumab alone with a bevacizumab/erlotinib combination in RCC patients, the combination arm was not found to be superior . The PFS of the combination vs. bevacizumab alone arms was 9.9 and 8.5 months, respectively, and the response rates (complete plus partial) were 14 and 13 %. Neither of these differences was statistically significant. The failure to demonstrate that erlotinib could augment the clinical efficacy of bevacizumab in this clinical trial casts doubt on the importance of EGFR signaling as a mechanism of resistance to VEGFR antagonists in RCC.
Angiopoietin-2 is produced primarily by hypoxic tumor-associated endothelial cells. This cytokine engages a tyrosine kinase receptor (Tie-2) through which it augments several of the effects of VEGF on the endothelium. Several lines of evidence suggest that its production may limit the effectiveness of VEGFR-targeted therapies and predispose to drug resistance. For example, Hashizume et al. showed in Colo-205 colon carcinoma xenografts that the inhibition of Ang-2 enhanced the antitumor and antiangiogenic effects of an anti-VEGF antibody . In a subsequent study, Falcon et al. showed that Ang-2 inhibition could “normalize” the vasculature of these Colo-205 xenografts . Specifically, the Ang-2 inhibitor was shown to increase the extent of endothelial pericyte coverage and the expression of VE cadherin at endothelial junctions, both of which are indicative of vascular maturation. These effects were antagonized by the concurrent administration of a selective Ang-1 inhibitor.
Independent of its direct effects on endothelial cells, at least some of the effects of Ang-2 appear to be mediated through the activation of a population of Tie-2expressing monocytes/macrophages (TEM) that infiltrate tumor tissue. For example, Mazzieri et al. showed that the effects of Ang-2 inhibition on the growth of murine mammary and islet cell carcinomas could be attributed to the loss of TEM function as these cells failed to associate with the tumor endothelium in the absence of Ang-2 signaling . The role played by Ang-2 and TEM in the development of resistance to VEGF-targeted therapies is further discussed below in the section on “Tumor-Associated Macrophages”.
Sphingosine-1-phosphate (S1P) is a bioactive sphingolipid produced by sphingokinases 1 and 2. The biological effects of S1P are mediated through the engagement of membrane-associated G protein-coupled S1P receptors (S1PR) as well as through the binding to various intracellular targets. The S1PR mediate many cellular functions including angiogenesis, cell proliferation, autophagy, and apoptosis. SPHK is known to be regulated by hypoxia, likely via HIF-1 and HIF-2. SPHK1 expression has been demonstrated in several tumor types and is associated with poor prognosis . It is possible that SPHK/S1P signaling is an adaptation to hypoxia. For example, Wang et al. have shown that SPHK is increased in a murine model of resistance to antiangiogenic therapy—likely secondary to treatmentinduced hypoxia . Three classes of S1P pathway inhibitors are currently being explored. SPHK1 inhibitors decrease S1P production by hypoxic tumor cells, blocking S1P-mediated tumor cell survival and angiogenesis [66, 67]. S1P-neutralizing molecules can sequester S1P and prevent its binding to S1PRs. This approach has been shown to have antiangiogenic activity in colon cancer models . S1PR inhibitors are also potentially beneficial therapeutic agents, preventing tumor growth in preclinical models .
Cyclooxygenase-2 (COX-2) has been shown to play a role in inflammation, tumor growth, invasiveness, metastasis, angiogenesis, and survival . COX-2 catalyzes the production of prostaglandin E2 (PGE2) from arachidonic acid. Inhibition of COX-2 has been shown to be a promising antitumor and antiangiogenic strategy in several tumor types including RCC [71–74]. In preclinical models, COX-2 inhibition has an activity as a single agent as well as in combination with immunotherapy and chemotherapy [75–77]. Clinical testing of COX-2 inhibitors has been performed in many tumor types. Although initial reports suggested improved response rates for patients bearing tumors expressing COX-2, a subsequent study of the combination of a COX-2 inhibitor and interferon alpha confined to these patients did not demonstrate a significant benefit for the combination relative to interferon alone. Wang et al. found that COX-2 expression in RCC models was elevated in areas of hypoxia induced by VEGFR inhibition. In RCC xenografts generated both from established cell lines and from fresh patient-derived tumors, the concurrent administration of the selective COX-2 inhibitor celecoxib with sunitinib delayed the emergence of treatment resistance longer than that achieved with sunitinib alone , suggesting that COX-2 inhibition might be a useful strategy to prevent TKI resistance.
Contribution of bone marrow-derived myeloid cells and other stromal elements to the development of resistance to VEGFR antagonists
Tumor-associated fibroblasts (TAF)
Several studies suggest that tumor-associated fibroblasts (TAF) stimulate tumor growth and angiogenesis and contribute to the development of resistance to VEGF-targeted therapy. Olumi et al., for example, showed that fibroblasts obtained from prostate carcinomas promote the growth of “primed” (i.e., SV40 large T antigen-immortalized) prostate cells both in vitro and in vivo . This growth promotion could not be induced with fibroblasts isolated from normal prostatic tissue and appeared to depend on the “priming” effect of the SV40 large T antigen since the TAF had no demonstrable effect on the proliferation of normal prostate epithelial cells. Crawford et al. observed a similar tumor cell-fibroblast symbiotic interaction in their studies with EL4 and TIB6 tumors . In their studies, however, the tumor-promoting effect of TAF was observed only with fibroblasts isolated from tumors resistant to VEGF-directed therapy (e.g., EL4) and not with either normal skin-derived fibroblasts or TAF derived from tumors that were responsive to anti-VEGF treatment (e.g., TIB6). TAF from the resistant EL4 tumors promoted the growth of TIB6 tumors even in the setting of VEGF blockade and roughly doubled the tumor microvessel density. Furthermore, they showed by confocal microscopy that the admixture of EL4-derived TAF with TIB6 tumor cells increased the number of vascular branch points in the resulting tumors up to sixfold and increased vessel area and volume by approximately 2.5-fold. The EL4-derived TAF, but not normal skin fibroblasts or TAF from TIB6, were capable of supporting angiogenesis in implanted Matrigel plugs, indicating that TAF did not require the continued presence of tumor cells to retain their ability to promote angiogenesis. The genetic or epigenetic alterations in the TAF that maintain their proangiogenic phenotype in the absence of an ongoing inductive effect from tumor cells were not characterized in this study.
The tumor-promoting effects of TAF have been ascribed to their ability to produce cytokines such as TGF-Я , HGF , and SDF-1 . Gene expression profiling of EL4-derived TAF as well as direct cytokine measurements suggested that the production of the PDGF isoform PDGF-C might also contribute to their proangiogenic effects . This hypothesis was validated in subsequent studies in which the ability of EL4-derived TAF to generate a microvasculature in Matrigel plugs was shown to be suppressible with a neutralizing anti-PDGF-C antibody. This antibody also suppressed the growth of EL4 tumors and augmented the growth inhibitory effects of an anti-VEGF antibody. The anti-PDGF-C antibody, however, did not suppress the growth of TIB6 tumors, which were responsive to the anti-VEGF antibody. Collectively, these findings indicate that the production of the novel PDGF isoform, PDGF-C, by TAF promotes tumor angiogenesis and is responsible for the primary resistance to anti-VEGF therapy observed in some tumors. The collaborative interactions between TAF-generated PDGF-C and the other cytokines produced by stromal cells that infiltrate tumor tissue are depicted in Fig. 16.2.
Although the Crawford data discussed above clearly demonstrate that TAF—in particular, those obtained from tumors that are resistant to VEGF-targeted therapies—are able to promote tumor growth and angiogenesis through mechanisms that are largely independent of VEGF, it should be kept in mind that these studies were not done with RCC and the extent to which these data are applicable to RCC is not known. The individual PDGF polypeptides are encoded by four genes, and five different homoand heterodimers can be assembled from these gene products . All of these PDGF dimers, including the PDGF-CC isoform implicated as a resistance factor in the Crawford study, signal through PDGFR-a and PDGFR-Я receptor tyrosine kinases. Many of the VEGFR2-targeted small molecule inhibitors (e.g., sunitinib, pazopanib) also efficiently block the PDGF receptors so it appears unlikely that the production of PDGF-C would provide a means of escape from the TKIs currently used to treat RCC.
The ability of TAF to promote tumor growth and angiogenesis is not peculiar to prostate and lung carcinoma. Breast carcinoma-associated fibroblasts (CAF) have similar biologic properties . Orimo et al., for example, has shown that CAF, but not normal fibroblasts, cocultivated with human breast cancer (MCF7) cells prior to implantation accelerate the growth and enhance the vascularity of the resulting xenografts . Fibroblasts isolated from normal breast tissue had no effect on tumor growth or vascularity. Tumors derived from the CAF-MCF-7 cell mixture contained far more Sca1+CD31+ endothelial progenitor cells (EPC) than those generated from MCF-7 cells alone or cells mixed with normal fibroblasts. These EPC were also much more abundant in the circulation of mice harboring tumors derived from the cocultivation of MCF-7 cells with CAF than in control mice. This observation suggests that the relative hypervascularity of the CAF-MCF-7 tumors was due in part to the enhanced EPC mobilization from the bone marrow. This suspicion was confirmed in a subsequent MCF-7 xenograft study involving H2k-d mice injected with H2k-b bone marrow-derived Sca1+CD31+ cells. In this study, the tumor endothelial cells stained positively for H2k-b by immunofluorescence, corroborating their bone marrow origin. RT-PCR analyses of the breast carcinoma CAF revealed abundant mRNA encoding the chemokine SDF-
1, and ELISA of the CAF-conditioned media demonstrated abundant SDF-1 secretion. A similar high degree of SDF-1 expression was observed in the a-smooth muscle actin (SMA)-staining fibroblasts present within invasive human breast carcinomas. To determine the functional significance of this SDF-1 production, mice bearing MCF-7 xenografts were treated with a neutralizing anti-SDF-1 antibody. The intraperitoneal instillation of this antibody suppressed the growth and vascularity of the tumors and reduced the number of Sca1+CD31+ cells present within the tumor infiltrate .
The gene encoding the chemokine SDF-1 and that encoding one of the SDF-1 G protein-coupled receptors CXCR4 are both HIF dependent and abundantly expressed by VHL-deficient RCC. It is likely that tumor-associated fibroblasts and other stromal elements also contribute to the SDF-1 produced by these tumors, especially in the setting of hypoxia induced by treatment with VEGF-targeted agents. In a SCID mouse model of human RCC, CXCR4 expression was shown to correlate with increased metastatic behavior and the neutralization of SDF-1 shown to reduce metastases . It is therefore likely that the SDF-1/CXCR4 axis plays a role in the growth and hypervascularity of human RCC, regardless of the identity of the cells from which the SDF-1 is derived. To the extent that this is the case, one would expect that treatment strategies that incorporate the neutralization of SDF-1 or blockade of CXCR4 function (e.g., with AMD3100) might serve as useful adjuncts to VEGFR-targeted therapies in the treatment of RCC .
Tumor-associated macrophages (TAM)
Among the most abundant leukocytes infiltrating tumors, macrophages accumulate in hypoxic/necrotic areas where they scavenge dead cells and promote tissue remodeling through the secretion of VEGF, MMP-9, and other factors . These cells recognize and respond to a wide variety of chemoattractants including endothelin (ET), the chemokine CCL5, VEGF, and PlGF. Approximately 20 % of macrophages (as well as circulating monocytes) express the tyrosine kinase receptor Tie-2 and respond to the Tie-2 ligands Ang-1 and Ang-2 as well . These Tie-2+ macrophages also express the generic leukocyte marker CD45 and the myeloid marker CD11b, but not the neutrophil marker Gr-1 or any of the well-characterized markers found on endothelial progenitor cells such as CD34 or CD146.
Of all of the myeloid cell subpopulations, Tie-2+ monocyte/macrophages are particular adept at promoting tumor growth and angiogenesis. For example, cocultivation of glioma cells with Tie-2+CD14+ monocytes (TEM), but not their Tie-2counterparts, prior to implantation into nude mice has been shown to enhance the growth and vascularity of the resulting tumors . The proangiogenic effect of TEM was at least in part due to their production of FGF . Exposure to angiopoietin-2 enhanced the ability of TEM to promote angiogenesis further by suppressing their production of the angiostatic cytokine Interleukin-12 (IL-12) . The importance of these cells was further illustrated in studies involving transgenic mice carrying the gene encoding thymidine kinase (TK) placed under the control of the Tie-2 promoter/enhancer. These mice express the enzyme in their Tie-2+ cells, rendering the TEM of these mice vulnerable to the antiviral prodrug ganciclovir . In these studies, bone marrow from the transgenic mice was adoptively transferred to recipient mice that were later implanted with mammary carcinoma or glioma cells and then treated with ganciclovir to ablate the TEM population. Ganciclovir treatment reduced the size and vascularity of the tumors growing in these mice, indicating that TEM contribute to the development of the tumor microcirculation.
The VEGF family member PlGF is another cytokine that activates macrophages. It stimulates the revascularization of ischemic tissue through its interaction with VEGFR1 and neuropilins 1 and 2 . PlGF is produced by both tumor cells and stromal elements in response to hypoxia and high levels can be detected in the plasma of patients undergoing treatment with VEGFR antagonists such as sunitinib . In a series of studies examining the role of PlGF in tumor growth and angiogenesis, Fischer et al. demonstrated that the neutralization of the cytokine with an anti-PlGF antibody resulted in the regression of several tumors, including some (e.g., the colon carcinoma CT26) resistant to treatment with an anti-VEGFR2 antibody . This antitumor effect was additive to that of an anti-VEGFR2 antibody. One of the most obvious histologic effects of PlGF neutralization was a marked reduction in tumor-infiltrating macrophages. Macrophage depletion by treatment with clodronate liposomes reduced tumor growth, but this effect was not additive to that of the anti-PlGF antibody [92, 93]. These data show that tumor-infiltrating macrophages enhance tumor growth and angiogenesis and that PlGF is one of the cytokines responsible for their recruitment into tumor tissue. The data, in fact, suggest that the promotion of macrophage recruitment may be the dominant mechanism by which PlGF stimulates tumor growth.
It is unknown whether PlGF-induced macrophage recruitment is a contributing factor in the development of TKI resistance by RCC. The fact that PlGF neutralization inhibits the growth of tumors resistant to VEGF-targeted therapy and enhances the efficacy of an anti-VEGFR2 antibody in other tumor models supports this hypothesis, as does the detection of high circulating PlGF levels in RCC patients undergoing sunitinib treatment . However, many of the small molecule VEGFR2 inhibitors FDA-approved for use in RCC (e.g., sunitinib, pazopanib) also potently block VEGFR1, which should incapacitate PlGF-mediated signaling in endothelial and other cell types. It is therefore possible that the increased production of PlGF during treatment and the PlGF-dependent recruitment of macrophages into tumors contributes to the development of resistance only to VEGF-specific therapies (e.g., bevacizumab), but not to agents that block both VEGFRs 1 and 2 (e.g., sunitinib, pazopanib).
Gr1+CD11b+ marrow-derived myeloid suppressor cells (MDSC)
Several studies have implicated a heterogeneous population of bone marrow-derived myeloid cells that express both the granulocyte phenotypic marker Gr-1 and the macrophage marker CD11b in the development of resistance to VEGF-targeted therapies [94–96]. MDSC are responsive to several cytokines and chemokines, some of which (e.g., SDF-1) are produced in hypoxic areas within the tumor through a HIF-dependent mechanism. Kioi et al., for example, showed that the recruitment of MDSC into irradiated glioblastoma xenografts was driven predominantly by the chemokine SDF-1. Tumor infiltration by these cells could be prevented by the administration of AMD3100, a drug that blocks the function of the SDF-1 receptor CXCR4 . These cells are also responsive to the chemokines CXCL5 , CXCL1, and CXCL2  and to GM-CSF and TNF [99, 100]. Finally, in a study by Chan et al., the recruitment of these cells into tumors was found to be mediated by IL-8 and angiogenin through an NF-?B-dependent mechanism . Thus, it appears that there a number of factors that regulate the influx of these MDSC into tumor tissue, some, but not all, of which are hypoxia (HIF) dependent.
Regardless of the specific chemotactic factors responsible for MDSC recruitment, the influx of these cells into tumor tissue is governed by the p53 status of the stromal elements within the tumor. Guo et al., for example, recently demonstrated that melanoma cells elicit a much more intense MDSC infiltrate when implanted into p53(-/-) mice than in p53-WT mice . This observation suggests that the production of chemotactic factors by the tumor stroma is suppressed by even baseline p53 activity in these cells. In a related study, Panka et al. demonstrated that the infiltration of CD11b+Gr-1+ MDSC into RCC xenografts is augmented by antiangiogenic therapy (i.e., sunitinib) and that this enhanced influx can be prevented by the concurrent administration of an HDM2 antagonist (MI-319, Sanofi-Aventis) . MI-319 increases p53 levels and p53-dependent gene expression by blocking the interaction between p53 and its dominant E3 ligase HDM2. In this study, MI-319 was also shown to suppress the increase in SDF-1 production within the tumor induced by sunitinib treatment, suggesting that this chemokine might be responsible for the MDSC recruitment. The interactions between p53, SDF-1, and MDSC are depicted in Fig. 16.3. The addition of MI-319 to sunitinib markedly extended the interval during which the growth of the RCC xenografts was constrained by sunitinib. These data suggest that the ability of HDM2 antagonists to suppress MDSC influx might be exploitable as a means of preventing TKI resistance.
Fig. 16.3. VHL-deficient RCC are known to produce SDF-1 constitutively. The gene encoding this chemokine is regulated by HIF and its expression is induced by hypoxia—as occurs, for example, in tumor stromal cells during treatment with VEGF-targeted drugs. SDF-1 is one of several chemokines that regulate MDSC trafficking into tumors. In this model, the induction of SDF-1 by hypoxia results in the recruitment of MDSC, whereas the suppression of SDF-1 production by p53 limits the influx of these cells
Through a mechanism that is not entirely understood, MDSC are able to confer on adjacent tumor cells the ability to tolerate cellular stress and to render the tumor cells resistant to many forms of treatment, including even chemotherapy. MDSC are relatively abundant, especially in tumors refractory to anti-VEGF therapies, and their depletion by treatment with an anti-Gr1 antibody has been reported to increase tumor responsiveness to treatment with an anti-VEGF antibody . Yang et al. showed that the implantation of an admixture of tumor cells with Gr1+CD11b+ MDSC resulted in tumors that grew faster than control tumors and had increased microvessel density and reduced areas of necrosis . These proangiogenic effects of MDSC were attributed to their ability to produce the matrix metalloproteinase MMP9 as the deletion of the gene encoding this matrix metalloproteinase from the MDSC rendered them unable to promote tumor vascularity. MMP9 is thought to stimulate angiogenesis through the liberation of high molecular weight isoforms of VEGF immobilized in the extracellular matrix. In other studies, Bv8 (prokineticin), a secreted protein generated in response to tumor-derived cytokines such as G-CSF, was found to be the dominant proangiogenic factor produced by Gr1+CD11b+ cells [95, 103]. Shojaei et al., for example, showed that forced Bv8 expression by tumor cells increased tumor angiogenesis . In addition to the production of MMP9 and Bv8, at least a subset of Gr1+CD11b+ cells has the capacity to insinuate into the developing tumor endothelium and to contribute structurally to the developing tumor microcirculation. This extent to which the incorporation of these cells into the tumor microvasculature contributes to their proangiogenic agenda is unclear.
As implied in the acronym, Gr1+CD11b+ MDSC are immunosuppressive as well as proangiogenic. These cells express arginase and deplete the microenvironment of arginine, which results in the accumulation of uncharged arginine-tRNA and the activation of the eIF2a kinase GCN2 . The translational arrest induced by the activation of eIF2a results in the selective down-modulation of the TCR-? chain, the src-related kinase p56lck, as well as the components of the NF-?B family in T lymphocytes and the loss of T cell function . MDSC produce immunosuppressive cytokines such as TGF-Я. They also express high levels of iNOS, which enables them to generate large amounts of peroxynitrite, a radical that directly nitrosylates the TCR, rendering it incapable of recognizing antigens . Thus, there are several means by which MDSC undermine the immune response to tumor-associated antigens. Whether these immunosuppressive effects of MDSC contribute to the development of resistance to VEGF-targeted therapy is less clear.
Clinical trials of regimens designed to delay/prevent resistance to VEGF-targeted therapy
Several clinical trials have been conducted or are now underway in which a VEGFR2-targeted drug is administered in combination with second agent chosen on the basis of its ability to block one or more of the signaling pathways suspected of playing a role in the development of TKI resistance. Similar studies involving single agents that target both VEGFR2 and a non-VEGFR signaling pathway implicated in TKI resistance are also underway. The preclinical and clinical data available for some of the agents involved in these trials are reviewed in detail in other chapters of this book and will therefore be discussed here only as they relate to the problem of TKI resistance.
As discussed above, the HGF/c-met pathway is one of several suspected of providing a means of escape from the effects of VEGFR2 blockade . Cabozantinib (XL184), an agent active against both c-met and VEGFR2, has been tested in a small phase II trial in patients with metastatic RCC. Twenty-five patients were enrolled, 17 of whom had received more than two prior agents including 13 whose disease had progressed following receipt of a VEGF pathway and an mTOR pathway inhibitor. Tumor responses were seen in 7 patients (28 %) and an additional 13 patients exhibited disease stabilization. Responses were seen in multiple disease sites including in four patients with bone metastases. Median PFS was 14.7 months, which was quite impressive given the heavy prior treatment of the patient population. This encouraging data has formed the basis of an Alliance Cooperative Group randomized phase II trial comparing cabozantinib to sunitinib in VEGFR TKI-naпve patients as well as an industry-sponsored phase III trial comparing cabozantinib to everolimus in patients whose disease has progressed on one of more VEGF pathway inhibitors .
Several reversible ATP-competitive TKIs that block FGFR-1 as well as VEGFR-2 and PDGFR-Я with IC50 values of <100 nM are active in various murine tumor models, and phase I studies with two such agents—dovitinib (Novartis) and E7080 (Eisai)—have been completed and found to have antitumor activity. The ability of dovitinib to inhibit FGFR-1 appears to be a crucial component of the drug’s activity since in two preclinical studies, the antitumor effects of the drug were shown to correlate with FGFR expression and/or the presence of an activating FGFR mutation in the tumor cells [107–109]. The other agent, E7080, shares with sunitinib and pazopanib the ability to inhibit c-kit, and its antiangiogenic effects in at least some tumors (e.g., human H146 SCLC xenografts) may depend on this activity . The extent to which its ability to block FGFR signaling contributes to its antitumor effect is unknown. E7080 is currently being tested in a randomized phase II trial comparing its efficacy to everolimus in the VEGFR TKI refractory population of patients with metastatic RCC. In addition, dovitinib is being compared to sorafenib in a randomized phase III trial involving patients whose disease has progressed following both VEGFR TKI and mTOR inhibitor therapies. This latter trial has completed accrual with results anticipated shortly. These clinical trials should hopefully clarify the role played by FGFR signaling in the development of resistance to VEGFR2 antagonists.
Amgen has developed a soluble Tie2-Fc “peptibody” AMG386 that blocks the interaction of angiopoietin-1 and angiopoietin-2 with their tyrosine kinase receptor Tie-2. This drug has marked antitumor activity in tumor xenograft models  and potent antiangiogenic activity in some tumors as determined by dynamic contrastenhanced magnetic resonance imaging . Although the dual nature of this drug precludes an analysis of the individual effects of Ang-1 and Ang-2 neutralization, at least one study suggests that both may contribute to the antitumor activity of AMG386 . A phase I clinical trial examined AMG386 in combination with either sorafenib or sunitinib in patients with metastatic RCC . The combination was fairly well tolerated with toxicity primarily attributed to the VEGFR TKI. Further, significant antitumor activity was noted with tumor responses seen in 5 of 17 patients treated with AMG386 and sorafenib and 8 of 15 patients treated with the agent and sunitinib. Subsequently a randomized, phase II, placebo-controlled clinical trial was performed examining sorafenib with or without AMG386 administered at either 3 or 10 mg/kg intravenously every week. Although the response rate was higher in both of the AMG 386 arms (37 and 38 %) than that seen with sorafenib alone (24 %), there was no difference in median progression-free survival. In addition, a multi-institutional phase II clinical trial of standard dose and schedule sunitinib in combination with AMG386 at either 10 or 15 mg/kg has also been performed. Results showed response rates for the two cohorts of 58 % and 63 %, respectively, and median PFS of 13.6 and 16.3 months and very little toxicity that was not attributable to sunitinib . These results appear to be potentially superior to sunitinib alone but are difficult to reconcile with the failure of AMG386 to prolong median PFS in combination with sorafenib in the randomized phase II trial mentioned previously. A randomized trial of sunitinib ± AMG386 at the 15 mg/kg dose would seem indicated, but at the present time is not being considered. The Pfizer Ang-2 inhibitor PF-04856884 (CVX-060) is also undergoing clinical evaluation. Unlike AMG-386, however, this agent selectively blocks Ang-2 and has no effect on the other angiopoietins. PF-04856884 has been shown to enhance the antitumor activity of sunitinib and bevacizumab in tumor xenografts, and a phase I trial of the drug in combination with axitinib in patients with previously treated RCC has been planned. Based on trials to date, the extent to which Ang-2 production contributes to the development of resistance to VEGFR-targeted therapies remains to be firmly established.
Although there is no evidence that the BMP-9 and BMP-10 activin receptor-like kinase-1 (ALK-1) signaling pathway is upregulated in response to treatment with a VEGF-targeted therapy, this pathway is essential for vascular remodeling and pathologic angiogenesis , and agents that block this pathway may be exploitable in delaying or preventing resistance to VEGFR antagonists. Two drugs that block this pathway—an ALK-1-Fc fusion protein (Acceleron) and an ALK-1-specific antibody (Pfizer)—have been shown to have antitumor and antiangiogenic effects in xenograft models [32, 115]. The anti-ALK-1 antibody has, in fact, been shown to cooperate with the VEGFR2 inhibitor axitinib in a melanoma xenograft model , suggesting that these agents might be useful as adjuncts to VEGFR2 antagonists as a means of delaying the emergence of drug resistance. A phase I trial to determine the antitumor activity of the ALK-1-Fc fusion protein administered in combination with axitinib in previously treated RCC patients was recently launched.
HIF-1 and HIF-2 are regulated by mTORC1 and mTORC2, respectively, and agents that block these signaling complexes (or the upstream kinase PI-3K) would be expected to suppress the production of numerous HIF-dependent proangiogenic factors such as VEGF. It is therefore possible that the use of an mTOR or PI3-K inhibitor in conjunction with a VEGF-targeted therapy would delay or prevent the development of resistance. Unfortunately, such combinations have been poorly tolerated and have necessitated significant reductions in the doses of the VEGF pathway inhibitor. Of note, despite some encouraging early data , randomized studies of either bevacizumab or sorafenib with temsirolimus have produced more toxicity and less activity than single agent VEGF pathway inhibitors alone [118, 119].
Finally, it may be possible to delay the emergence of resistance and enhance the PFS of drugs such as sunitinib by the concurrent administration of conventional chemotherapy. The combination of sunitinib with gemcitabine, for example, is active even in RCC patients whose tumors have become refractory to single agent sunitinib . Although the mechanism has not been verified experimentally, interference with the recruitment of various proangiogenic myeloid cells into the tumor may account for this additive effect.
The cytokines, cell types, and signaling pathways that have been proposed to mediate the development of resistance to VEGF-targeted therapies are numerous and diverse. Our inability to identify a particular single cytokine or factor that is consistently responsible for TKI resistance attests to the complexity of the cellular response to the hypoxia and nutrient deprivation induced by drugs whose primary mode of action is the disruption of the tumor microvasculature. The sheer number of signaling pathways that enhance the ability of tumor cells to tolerate hypoxia and other forms of metabolic stress and facilitate the restoration of the microcirculation in the setting of VEGF/VEGFR blockade tends to undermine the notion that the blockade of any one additional cytokine (e.g., HGF) or signaling pathway would solve the problem of TKI resistance for all tumors. The success or failure of the upcoming clinical trials with drug combinations that target VEGFR2 and either c-met or the FGFR, for example, will determine whether such therapeutic nihilism is justified. A careful delineation of the roles played by the various stress-induced signaling pathways (e.g., HIF, APMK, the UPR) activated in the setting of hypoxia and the development of agents that block these pathways may yield a solution to the problem of sunitinib resistance that cannot be solved through the continued focus on individual tyrosine kinases. Drugs that block the expression or activity of HIF-2, for example, might prove extremely useful as adjuncts to VEGF-targeted therapies. Several investigators are in fact in the process of developing agents with this capability . An effort to better define the contribution of tumor-infiltrating myeloid cells to the problem of TKI resistance might also prove useful as would an analysis of the factors that regulate their trafficking. HDM2 antagonists and drugs that block the interaction between the chemokine SDF-1 and its receptor CXCR4 may be effective when used in combination with TKIs because of their ability to regulate MDSC influx.
- Bayne LJ, Beatty GL, Jhala N, Clark CE, Rhim AD, Stanger BZ, Vonderheide RH. Tumorderived granulocyte-macrophage colony-stimulating factor regulates myeloid inflammation and T cell immunity in pancreatic cancer. Cancer Cell. 2012;21:822–35.
- Acharyy S, Oskarsson T, Vanharanta S, Malladi S, Kim J, Morris PG, et al. A CXCL1 paracrine network links cancer chemoresistance and metastasis. Cell. 2012;150:165–78.
- Yang L, Huang J, Ren X, Gorska AE, Chytil A, Aakre M, et al. Abrogation of TGFЯ signaling in mammary carcinomas recruits Gr-1 + CD11b + myeloid cells that promote metastasis. Cancer Cell. 2008;13:23–35.
- Finke J, Ko J, Rini B, Rayman P, Ireland J, Cohen P. MDSC as a mechanism of tumor escape from sunitinib mediated anti-angiogenic therapy. Int Immunopharmacol. 2011;11:856–61.
- Yang L, DeBusk LM, Fukuda K, Fingleton B, Green-Jarvis B, Shyr Y, et al. Expansion of myeloid immune suppressor Gr + CD11b + cells in tumor-bearing host directly promotes tumor angiogenesis. Cancer Cell. 2004;6:409–21.
- Shojaei F, Wu X, Malik AK, Zhong C, Baldwin ME, Schanz S, Fuh G, Gerber HP, Ferrera N. Tumor refractoriness to anti-VEGF treatment is mediated by CD11b+Gr1+ myeloid cells. Nat Biotechnol. 2007;25:911–20.
- Zeisberger SM, Odermatt B, Marty C, Zehnder-Fjallman AH, Ballmer-Hofer K, Schwendener RA. Clodronate-liposome-mediated depletion of tumour-associated macrophages: a new and highly effective antiangiogenic therapy approach. Br J Cancer. 2006;95:272–81.
- Fischer C, Jonckx B, Mazzone M, Zacchigna S, Loges S, Pattarini L, et al. Anti-PlGF inhibits growth of VEGFR-inhibitor-resistant tumors without affecting healthy vessels. Cell. 2007;131:463–75.
- Luttun A, Tjwa M, Moons L, Wu Y, Angelillo-Scherrer A, Liao F, et al. Revascularization of ischemic tissues by PlGF treatment and inhibition of tumor angiogenesis, arthritis, and atherosclerosis by anti-Flt1. Nat Med. 2002;8:831–40.
- Murdoch C, Tazzyman S, Webster S, Lewis CE. Expression of Tie-2 by human monocytes and their responses to angiopoietin-2. J Immunol. 2007;178:7405–11.
- DePalma M, Venneri MA, Galli R, Sergi Sergi L, Politi LS, Sampaclesi M, Naldini L. Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell. 2005;8: 211–26.
- Venneri MA, DePalma M, Ponzoni M, Pucci F, Scielzo C, Zonari F, et al. Identification of proangiogenic TIE2-expressing monocytes (TEMs) in human peripheral blood and cancer. Blood. 2007;109:5276–85.
- Lewis CE, DePalma M, Naldini L. Tie2-expressing monocytes and tumor angiogenesis: regulation by hypoxia and angiopoietin-2. Cancer Res. 2007;67:8429–32.
- Murdoch C, Giannoudis A, Lewis CE. Mechanisms regulating the recruitment of macrophages into hypoxic areas of tumors and other ischemic tissues. Blood. 2004;104:2224–34.
- Kioi M, Vogel H, Schultz G, Hoffman RM, Harsh GR, Brown JM. Inhibition of vasculogenesis, but not angiogenesis, prevents recurrence of glioblastoma after irradiation in mice. J Clin Invest. 2010;120:694–705.
- Pan J, Mestas J, Burdick MD, Phillips RJ, Thomas GV, Reckamp K, et al. Stromal-derived factor-1 (SDF-1/CXCL12) and CXCR4 in renal cell carcinoma metastasis. Mol Cancer. 2006;5:56.
- Li X, Eriksson U. Novel PDGF family members: PDGF-C and PDGF-D. Cytokine Growth Factor Rev. 2003;14:91–8.
- Orimo A, Gupta PB, Sgroi DC, Arenzana-Seisdedcs F, Delaunay T, Naeem R, et al. Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-1/CXCL12 secretion. Cell. 2005;121:335–48.
- Bhowmick NA, Chytil A, Plieth D, Gorska AE, Dumont N, Shappell S, et al. TGF-beta signaling in fibroblasts modulates the oncogenic potential of adjacent epithelia. Science. 2004;303:848–51.
- Crawford Y, Kasman I, Yu L, Zhong C, Wu X, Modrusan Z, Kaminker J, Ferrera N. PDGF-C mediates the angiogenic and tumorigenic properties of fibroblasts associated with tumors refractory to anti-VEGF treatment. Cancer Cell. 2009;15:21–34.
- Olumi AF, Grossfeld GD, Hayward SW, Carroll PR, Tlsty TD, Cunha GR. Carcinomaassociated fibroblasts direct tumor progression of initiated human prostatic epithelium. Cancer Res. 1999;59:5002–11.
- Wang X, Zhang L, O’Neill A, Bahamon B, Alsop DC, Mier JW, et al. Cox-2 inhibition enhances the activity of sunitinib in human renal cell carcinoma xenografts. Br J Cancer. 2013;108:319–26.
- Legge F, Paglia A, D’Asta M, Fuoco G, Scambia G, Ferrandina G. Phase II study of the combination carboplatin plus celecoxib in heavily pre-treated recurrent ovarian cancer patients. BMC Cancer. 2011;11:214.
- Vivaldi A, Ciampi R, Tacito A, Molinaro E, Agate L, Bottici V, et al. Celecoxib a cyclooxygenase-2 inhibitor, potentiates the chemotherapeutic effect of vinorelbine in the medullary thyroid cancer TT cell line. Mol Cell Endocrinol. 2012;355:41–8.
- Veltman JD, Lambers ME, van Nimwegen M, Hendriks RW, Hoogsteden HC, Aerts JG, Hegmans JP. COX-2 inhibition improves immunotherapy and is associated with decreased numbers of myeloid-derived suppressor cells in mesothelioma. Celecoxib influences MDSC function. BMC Cancer. 2010;10:464.
- Chen Q, Shinohara N, Abe T, Harabayashi T, Nonomura K. Impact of cyclooxygenase-2 gene expression on tumor invasiveness in a human renal cell carcinoma cell line. J Urol. 2004;172:2153–7.
- Song X, Lin HP, Johnson AJ, Tseng PH, Yang YT, Kulp SK, Chen CS. Cyclooxygenase-2, player of spectator in cyclooxygenase-2 inhibitor-induced apoptosis in prostate cancer cells. J Natl Cancer Inst. 2002;94:585–91.
- Hida T, Kozaki K, Muramatsu H, Masuda A, Shimizu S, Mitsudomi T, et al. Cyclooxygenase-2 inhibitor induces apoptosis and enhances cytotoxicity of various anticancer agents in nonsmall cell lung cancer cell lines. Clin Cancer Res. 2010;6:2006–11.
- Arber N, Eagle CJ, Spicak J, Racz J, Dite P, Hajer J, et al. Celecoxib for the prevention of colorectal adenomatous polyps. N Engl J Med. 2006;355:885–95.
- Ghosh N, Chaki R, Mandal V, Mandal SC. COX-2 as a target for cancer chemotherapy. Pharmacol Rep. 2010;62:233–44.
- Liu Y, Deng J, Wang L, Lee H, Armstrong B, Scuto A, et al. S1PR1 is an effective target to block STAT3 signaling in activated B cell-like diffuse large B-cell lymphoma. Blood. 2012;120:1458–65.
- Visentin B, Vekich JA, Sibbald BJ, Cavalli AL, Moreno KM, Matteo RG, et al. Validation of an anti-sphingosine-1-phosphate antibody as a potential therapeutic in reducing growth, invasion, and angiogenesis in multiple tumor lineages. Cancer Cell. 2006;9:225–38.
- Anelli V, Gault CR, Cheng AB, Cheng AB, Obeid LM. Sphingosine kinase 1 is up-regulated during hypoxia in U87MG glioma cells: role of hypoxia-inducible factors 1 and 2. J Biol Chem. 2008;283:3365–75.
- Schnitzer SE, Weigert A, Zhou J, Brune B. Hypoxia enhances sphingosine kinase 2 activity and provokes-1-phosphate-mediated chemoresistance in A549 lung cancer cells. Mol Cancer Res. 2009;7:393–401.
- Cuvillier O, Ader I, Bouquerel P, Brizuela L, Malavaud B, Mazerolles C, Rischmann P. Activation of sphingosine kinase-1 in cancer: implications for therapeutic targeting. Curr Mol Pharmacol. 2010;3:53–65.
- Mazzieri R, Pucci F, Moi D, Zonari E, Ranghetti A, Berti A, et al. Targeting the ANG2/TIE2 axis inhibits tumor growth and metastasis by impairing angiogenesis and disabling rebounds of proangiogenic myeloid cells. Cancer Cell. 2011;19:512–26.
- Falcon BL, Hashizume H, Koumoutsakos P, Chou J, Bready JV, Coxon A, Oliner JD, McDonald DM. Contrasting actions of selective inhibitors of angiopoietin-1 and angiopoietin-2 on the normalization of tumor blood vessels. Am J Pathol. 2009;175:2159–70.
- Hashizume H, Falcon BL, Kuroda T, Baluk P, Coxon A, Yu D, et al. Complementary actions of inhibitors of angiopoietin-2 and VEGF on tumor angiogenesis and growth. Cancer Res. 2010;70:2213–23.
- Bukowski RM, Kabbinavar FF, Figlin RA, Flaherty K, Srinivas S, Vaishampayan U, et al. Randomized phase II study of erlotinib combined with bevacizumab compared with bevacizumab alone in metastatic renal cancer. J Clin Oncol. 2007;25:4536–41.
- Rowinsky EK, Schwartz GH, Gollob JA, Thompson JA, Vogelzang NJ, Figlin R, et al. Safety, pharmacokinetics, and activity of ABX-EGF, a fully human anti-epidermal growth factor receptor monoclonal antibody in patients with metastatic renal cell cancer. J Clin Oncol. 2004;22:3003–15.
- Cascone T, Herynk MH, Xu L, Du Z, Kadara H, Nilsson MB, et al. Upregulated stromal EGFR and vascular remodeling in mouse xenograft models of angiogenesis inhibitor-resistant human lung adenocarcinoma. J Clin Invest. 2011;121:1313–28.
- Weber KL, Doucet M, Price JE, Baker C, Kin SJ, Fidler IJ. Blockade of epidermal growth factor receptor signaling leads to inhibition of renal cell carcinoma growth in the bone of nude mice. Cancer Res. 2003;63:2940–7.
- Yotsumoto F, Yagi H, Suzuki SO, Oki E, Tsujioka H, Hachisuga T, et al. Validation of HB-EGF and amphiregulin as targets for human cancer therapy. Biochem Biophys Res Commun. 2008;365:55–61.
- Bommi-Reddy A, Almeciga I, Sawyer J, Geisen C, Li W, Harlow E, Kaelin WG, Gruenberg DA. Kinase requirements in human cells: III. Altered kinase requirements in VHL-/cancer cells detected in a pilot synthetic lethal screen. Proc Natl Acad Sci U S A. 2008;105:16484–9.
- Maulik G, Shrikhande A, Kijima T, Ma PC, Morrison PT, Salgia R. Role of the hepatocyte growth factor receptor c-Met in oncogenesis and potential for therapeutic inhibition. Cytokine Growth Factor Rev. 2002;13:41–59.
- Shojaei F, Lee JH, Simmons BH, Wong A, Esparza CO, Plumlee PA, et al. HGF/c-met acts as an alternative angiogenic pathway in sunitinib-resistant tumors. Cancer Res. 2010;70:10090–100.
- Peruzzi B, Bottaro DP. Targeting the c-met signaling pathway in cancer. Clin Cancer Res. 2006;12:3657–60.
- Bello E, Colella G, Scarlato V, Oliva P, Berndt A, Valbusa G, et al. E-3810 is a potent dual inhibitor of VEGFR and FGFR that exerts antitumor activity in multiple preclinical models. Cancer Res. 2011;71:1396–405.
- Welti JC, Gourlaouen M, Powles T, Kudahetti SC, Wilson P, Berney DM, Reynolds AR. Fibroblast growth factor 2 regulates endothelial cell sensitivity to sunitinib. Oncogene. 2011;30:1183–93.
- Bhatt RS, Wang X, Zhang L, Collins MP, Signoretti S, Alsop DC, et al. Renal cancer resistance to antiangiogenic therapy is delayed by restoration of angiostatic signaling. Mol Cancer Ther. 2010;9:2793–802.
- Strieter RM, Kunkel SL, Arenberg DA, Burdick MD, Polverini PJ. Interferon gammainducible protein (IP-10), a member of the CXC chemokine family, is an inhibitor of angiogenesis. Biochem Biophys Res Commun. 2005;210:51–7.
- Mizukami Y, Jo WS, Duerr EM, Gala M, Li J, Zhang X, et al. Induction of interleukin-8 preserves the angiogenic response in HIF-1-alpha-deficient colon cancer cells. Nat Med. 2005;11:992–7.
- Karakurum M, Shreeniwas R, Chen J, Pinsky D, Yan SD, Anderson M, et al. Hypoxic induction of interleukin-8 gene expression in human endothelial cells. J Clin Invest. 1994;93:1564–70.
- Strieter RM, Polverini PJ, Arenberg DA, Kunkel SL. The role of CXC chemokines as regulators of angiogenesis. Shock. 1995;4:155–60.
- Gale LM, McColl SR. Chemokines: extracellular messengers for all occasions? Bioessays. 1999;21:17–28.
- Huang D, Ding Y, Zhou M, Rini BI, Petillo D, Qian CN, et al. Interleukin-8 mediates resistance to antiangiogenic agent sunitinib in renal cell carcinoma. Cancer Res. 2010;70: 1063–71.
- Casanovas O, Hicklin DJ, Bergers G, Hanahan D. Drug resistance by evasion of antiangiogenic targeting of VEGF signaling in late-stage pancreatic islet tumors. Cancer Cell. 2005;8:299–309.
- Ebos JM, Lee CR, Christensen JG, Mutsaers AJ, Kerbel RS. Multiple circulating proangiogenic factors induced by sunitinib malate are tumor-independent and correlate with antitumor efficacy. Proc Natl Acad Sci U S A. 2007;104:17069–74.
- Deprimo SE, Bello CL, Smeraglia J, Baum CM, Spinell D, Rini BI, et al. Circulating protein biomarkers of pharmacodynamic activity of sunitinib in patients with metastatic renal cell carcinoma: modulation of VEGF and VEGF-related proteins. J Transl Med. 2007;5:32.
- Zama IN, Hutson TE, Elson P, Cleary JM, Choueiri TK, Heng DY, et al. Sunitinib rechallenge in metastatic renal cell carcinoma patients. Cancer. 2010;116:5400–6.
- Elfiky AA, Cho DC, McDermott DF, Rosenberg JE, Fortner B, Antras L, et al. Predictors of response to sequential sunitinib and the impact of prior VEGF-targeted drug washout in patients with metastatic clear cell renal cell carcinoma. Urol Oncol. 2011;29:756–63.
- Hammers HJ, Verheul HM, Salumbides B, Sharma R, Rudek M, Jaspers J, et al. Reversible epithelial to mesenchymal transition and acquired resistance to sunitinib in patients with renal cell carcinoma: evidence from a xenograft study. Mol Cancer Ther. 2010;9:11525–35.
- Zhang L, Bhasin M, Schor-Bardach R, Wang X, Collin MP, Panka D, et al. Resistance of renal cell carcinoma to sorafenib is mediated by potentially reversible gene expression. PLoS One. 2011;6:e19144.
- Diamond JM, Melo JV. Mechanisms of resistance to BCR-ABL kinase inhibitors. Leuk 2011;52 Suppl 1:12–22.
- Bonanno L, Jirillo A, Favaretto A. Mechanisms of acquired resistance to epidermal growth factor receptor tyrosine kinase inhibitors and new therapeutic perspectives in non-small cell lung cancer. Curr Drug Targets. 2011;12:922–33.
- Benedettini E, Sholl LM, Peyton M, Reilly J, Ware C, Davis L, et al. Met activation in nonsmall cell lung cancer is associated with de novo resistance to EGFR inhibitors and the development of brain metastasis. Am J Pathol. 2010;177:415–23.
- Wang X, Solban N, Bhasin MK. ALK1-Fc inhibits tumor growth in a VEGF pathway resistance model of renal cell carcinoma. Proc AACR. 2012;LB-313.
- Hu-Lowe DD, Chen E, Zhang L, Watson KD, Mancuso P, Lappin P, et al. Targeting activin receptor-like kinase 1 inhibits angiogenesis and tumorigenesis through a mechanism of action complementary to anti-VEGF therapies. Cancer Res. 2011;71:1362–73.
- Cunha SI, Pardali E, Thorikay M, Anderberg C, Hawinkels L, Goumans MJ, et al. Genetic and pharmacological targeting of activin receptor-like kinase 1 impairs tumor growth and angiogenesis. J Exp Med. 2010;207:85–100.
- Suzuki Y, Ohga N, Moroshita Y, Hida K, Miyazono K, Watabe T. Bmp-9 induces proliferation of multiple types of endothelial cells in vitro and in vivo. J Cell Sci. 2010;123: 1684–92.
- Johnson DW, Berg JN, Baldwin MA, Gallione CJ, Marondel I, Yoon SJ, et al. Mutations in the activin receptor-like kinase 1 gene in hereditary haemorrhagic telangiectasia type 2. Nat Genet. 1996;13:189–95.
- McAllister KA, Grogg KM, Johnson DW, Gallione CJ, Baldwin MA, Jackson CE, et al. Endoglin, a TGF-Я binding protein of endothelial cells, is the gene for hereditary haemorrhagic telangiectasia type 1. Nat Genet. 1994;8:345–51.
- Arthur HM, Ure J, Smith AJ, Renforth G, Wilson DI, Torsney E, et al. Endogin, an ancillary TGFЯ receptor, is required for extraembryonic angiogenesis and plays a key role in heart development. Hum Mol Genet. 2000;107:1653–7.
- Oh SP, Seki T, Goss KA, Imamura T, Yi Y, Donahoe PK, et al. Activin receptor-like kinase 1 modulates transforming growth factor-Я 1 signaling in the regulation of angiogenesis. Proc Natl Acad Sci U S A. 2000;97:2626–31.
- Yan M, Callahan CA, Beyer JC, Allamneni KP, Zhang G, Ridgway JB, et al. Chronic DLL4 blockade induces vascular neoplasms. Nature. 2010;463:e6–7.
- Ridgway J, Zhang G, Wu Y, Stawicki S, Liang WC, Chanthery Y, et al. Inhibition of DLL4 signalling inhibits tumour growth by deregulating angiogenesis. Nature. 2006;444:1083–7.
- Li JL, Sainson RC, Oon CE, Turley H, Leek R, Sheldon H, et al. DLL4-Notch signaling mediates tumor resistance to anti-VEGF therapy in vivo. Cancer Res. 2011;71:6073–383.
- Yu JL, Rak JW, Coomber BL, Hicklin DJ, Kerbel RS. Effect of p53 status on tumor response to antiangiogenic therapy. Science. 2002;295:1526–8.
- Sablina AA, Budanov AV, Ilyinskaya GV, Agapova LS, Kravchenko JE, Chumakov PM. The antioxidant function of the p53 tumor suppressor. Nat Med. 2005;11:1306–13.
- Maddocks OD, Berkers CR, Mason SM, Zheng L, Blyth K, Gottleib E, Vousden KH. Serine starvation induces stress and p53-dependent metabolic remodeling in cancer cells. Nature. 2013;493:542–6.
- Dioufa N, Chatzistamou I, Farmaki E, Papavassiliou AG, Klaris H. p53 antagonizes the unfolded protein response and inhibits ground glass hepatocyte development during endoplasmic reticulum stress. Exp Biol Med. 2012;237:1173–80.
- Panka DJ, Liu Q, Geissler AK, Mier JW. HDM2 antagonism delays the development of sunitinib resistance in RCC xenografts: effects of MI-319 on sunitinib-induced p53 activation, SDF-1 induction, and tumor infiltration by CD11b+/Gr-1+ myeloid suppressor cells. Mol Cancer. 2013;12:17.
- Ditch S, Paull TT. The ATM protein kinase and cellular redox signaling: beyond the DNA damage response. Trends Biochem Sci. 2012;37:15–22.
- Pereira ER, Liao N, Neale GA, Hendershot LM. Transcriptional and post-transcriptional regulation of proangiogenic factors by the unfolded protein response. PLoS One. 2010; 5:e12521.
- Auf G, Jabouille A, Guerit S, Pineau R, Delugin M, Bouchecareilh M, et al. Inositol-requiring enzyme-1a is a key regulator of angiogenesis and invasion in malignant glioma. Proc Natl Acad Sci U S A. 2010;107:15553–9.
- Romero-Ramirez L, Cao H, Nelson D, Hammond E, Lee AH, Yoshida H, et al. XBP1 is essential for survival under hypoxic conditions and is required for tumor growth. Cancer Res. 2004;64:5943–51.
- Fels DR, Koumenis C. The PERK/eIF2a/ATF4 module of the UPR in hypoxia resistance and tumor growth. Cancer Biol Ther. 2006;5:723–32.
- Blais JD, Addison CL, Edge R, Falls T, Zhao H, Wary K, et al. PERK-dependent translational regulation promotes tumor cell adaptation and angiogenesis in response to hypoxic stress. Mol Cell Biol. 2006;26:9517–27.
- Sitia R, Molteni SN. Stress, protein misfolding and signaling: the redox connection. Sci 2004;239:27–33.
- Bi M, Naczki C, Koritzinsky M, Fels D, Blais J, Hu N, et al. ER stress-regulated translation increases tolerance to extreme hypoxia and promotes tumor growth. EMBO J. 2005; 24:3470–82.
- Ouchi N, Shibata R, Walsh K. AMP-activated protein kinase signaling stimulates VEGF expression and angiogenesis in skeletal muscle. Circ Res. 2005;96:838–46.
- Laderoute KR, Amin K, Calaoagan JM, Knapp M, Le T, Orduna J, et al. 5′ AMP-activated protein kinase (AMPK) is induced by low oxygen and glucose deprivation conditions found in solid tumor microenvironments. Mol Cell Biol. 2006;26:5336–47.
- Schor-Bardach R, Alsop DC, Pedrosa I, Solazzo SA, Wang X, Marquis RP, et al. Does arterial spin-labeling MR imaging-measured tumor perfusion correlate with renal cell cancer response to antiangiogenic therapy in a mouse model? Radiology. 2009;251:731–42.
- Sennino B, McDonald DM. Controlling escape from angiogenesis inhibitors. Nat Rev Cancer. 2012;12:699–709.
- Ebos JM, Lee CR, Kerbel RS. Tumor and host-mediated pathways of resistance and disease progression in response to antiangiogenic therapy. Clin Cancer Res. 2009;15:5020–5.
- Bergers G, Hanahan D. Modes of resistance to anti-angiogenic therapy. Nat Rev Cancer. 2008;8:592–603.
- Rini BI, Escudier B, Tomczak P, Kaprin A, Szczylik C, Hutson TE, et al. Comparative effectiveness of axitinib versus sorafenib in advanced renal cell carcinoma (AXIS): a randomized phase 3 trial. Lancet. 2011;378:1931–9.
- Ward JE, Stadler WM. Pazopanib in renal cell carcinoma. Clin Cancer Res. 2010;16:5923–7.
- Motzer RJ, Michaelson MD, Redman BG, Hudes GR, Wilding G, Figlin RA, et al. Activity of SU11248, a multitargeted inhibitor of vascular endothelial growth factor receptor and platelet-derived growth factor receptor, in patients with metastatic renal cell carcinoma. J Clin Oncol. 2006;24:16–24.
- Zhao X, Rong L, Zhao X, Li X, Liu X, Deng J, et al. TNF signaling drives myeloid-derived suppressor cell accumulation. J Clin Invest. 2012;122:4094–104.
- Chan DA, Kawahara TLA, Sutphin PD, Chang HY, Chi JT, Giaccia AJ. Tumor vasculature is regulated by PHD2-mediated angiogenesis and bone marrow-derived cell recruitment. Cancer Cell. 2009;15:527–38.
- Guo G, Marrero L, Rodriguez P, Dei Valle L, Ochoa A, Cui Y. Trp53 inactivation in the tumor microcirculation promotes tumor progression by expanding the immunosuppressive lymphoid-like stromal network. Cancer Res. 2013;73:1668–75.
- Shojaei F, Wu X, Zhong C, Yu L, Liang XH, Yao J, et al. Bv8 regulates myeloid-celldependent tumour angiogenesis. Nature. 2007;450:825–31.
- Zea AH, Rodriguez PC, Atkins MB, Hernandez C, Signoretti S, Zabaleta J, et al. Arginaseproducing myeloid suppressor cells in renal cell carcinoma patients: a mechanism of tumor evasion. Cancer Res. 2005;65:3044–8.
- Nagaraj S, Gupta K, Pisarev V, Kinarsky L, Sherman S, Kang L, et al. Altered recognition of antigen is a mechanism of CD8+ T cell tolerance in cancer. Nat Med. 2007;13:828–35.
- Choueiri TK, Pal SK, McDermott DF, Ramies DA, Morrisey S, Lee Y, et al. Efficacy of cabozantinib (XL184) in patients with metastatic, refractory renal cell carcinoma (RCC). J Clin Oncol. 2012;30(Suppl):abstr 4504.
- Dancey JE. Kinase Inhibitor 4 Minisymposium summary. Expert Rev Anticancer Ther. 2009;9:891–4.
- Lamont FR, Tomlinson DC, Cooper PA, Shnyder SD, Chester JD, Knowles MA. Small molecule FGF receptor inhibitors block FGFR-dependent urothelial carcinoma growth in vitro and in vivo. Br J Cancer. 2011;104:75–82.
- Renhowe PA, Pecchi S, Shafer CM, Machajewski TD, Jazan EM, Taylor C, et al. Design, structure-activity relationships and in vivo characterization of 4-amino-3-benzimidazol-2ylhydroquinolin-2-ones: a novel class of receptor tyrosine kinase inhibitors. J Med Chem. 2009;52:278–92.
- Matsui J, Yamamoto Y, Funahashi Y, Tsuruoka A, Watanabe T, Wakabayashi T, et al. E7080, a novel inhibitor that targets multiple kinases, has potent antitumor activities against stem cell factor producing human small cell lung cancer H146, based on angiogenesis inhibition. Int J Cancer. 2008;122:664–71.
- Neal J, Wakelee H. AMG-386, a selective angiopoietin-1/2-neutralizing peptibody for the potential treatment of cancer. Curr Opin Mol Ther. 2010;12:487–95.
- Herbst RS, Hong D, Chap L, Kurzrock R, Jackson E, Silverman JM, et al. Safety, pharmacokinetics, and antitumor activity of AMG386, a selective angiopoietin inhibitor, in adult patients with advanced solid tumors. J Clin Oncol. 2009;27:3557–65.
- Coxon A, Bready J, Min H, Kaufman S, Leal J, Yu D, et al. Context-dependent role of angiopoietin-1 inhibition in the suppression of angiogenesis and tumor growth: implications for AMG386, an angiopoietin-1/2-neutralizing peptibody. Mol Cancer Ther. 2010;9:2641–51.
- Atkins MB, Ravaud A, Gravis G, Drosik K, Demkow T, Tomczak P, et al. Safety and efficacy of AMG386 in combination with sunitinib in patients with metastatic renal cell carcinoma (mRCC) in an open-label multicenter phase II study. J Clin Oncol. 2012;30(Suppl):abstr
- Mitchell D, Pobre EG, Mulivor AW, Grinberg AV, Castonquay R, Monell TE, et al. ALK1-Fc inhibits multiple mediators of angiogenesis and suppresses tumor growth. Mol Cancer Ther. 2010;9:379–88.
- Zhang L, Lira M. Combination strategy to circumvent the acquired resistance to VEGF receptor tyrosine kinase inhibitor in M24met melanoma in mice. Proc AACR. 2011;Abstr. 3270.
- Merchan JR, Pitot HC, Qin R, Liu G, Fitch R, Maples WJ, et al. Final phase II safety and efficacy results of study MC0452: phase I/II trial of CCI779 and bevacizumab in advanced renal cell carcinoma. J Clin Oncol. 2011;29(Suppl):abstr 4548.
- Negrier S, Gravis G, Perol D, Chevreau C, Delva R, Bat JO, et al. Temsirolimus and bevacizumab, or sunitinib, or interferon alfa and bevacizumab for patients with advanced renal cell carcinoma (TORAVA): a randomized phase 2 trial. Lancet Oncol. 2011;12:673–80.
- McDermott DF, Manola J, Pins M, Flaherty KT, Atkins MB, Dutcher JP, et al. The BEST trial (E2804): a randomized phase II study of VEGF, RAF kinase, and mTOR combination targeted therapy (CTT) with bevacizumab (bev), sorafenib (sor), and temsirolimus (tem) in advanced renal cell carcinoma (RCC). J Clin Oncol. 2013;31 Suppl 6:abstr 345.
- Pandya SS, Mier JW, McDermott DF, Cho DC. Addition of gemcitabine at the time of sunitinib resistance in metastatic renal cell cancer. BJU Int. 2011;108:E245–9.
- Zimmer M, Ebert BL, Neil C, Brenner K, Papaloannou I, Melas A, et al. Small molecule inhibitors of HIF-2a translation link its 5’UTR iron-responsive element to oxygen sensing. Mol Cell. 2008;32:838–48.