Ligand-dependent c-MET activation through paracrine HGF induces proliferation, survival, and invasiveness of solid tumor through a series of signaling pathways including PIK3/Akt, STAT, and Ras/Mek. It also regulates tumor angiogenesis by inducing endothelial cell migration and tubular formation. The significant role of the HGF/c-met pathway in the mechanism of resistance to anti-VEGFR therapy has been indicated by several authors [30, 31]. Shojaei et al. [32] demonstrated that HGF levels in the sunitinib-resistant xenograft tumor such as mouse lymphoma (EL4) and Lewis lung carcinoma (LLC) are much higher when compared to that of sunitinib-sensitive tumor such as B16F1 mouse melanoma and TIB6 mouse myeloma cells, especially after treated by sunitinib. Combination of sunitinib and selective c-MET inhibitor had an additive effect of inhibiting the growth of sunitinib-resistant xenograft. Moreover, treatment efficacy of sunitinib for B16F1 or TIB6 cells was significantly decreased after the exposure of tumor cells to HGF. This approach clearly confirms the significance of c-MET activation in the mechanism of developing anti-VEGFR drug resistance in xenograft of several malignancies. Recently, cabozantinib, which is a novel tyrosine kinases inhibitor of c-MET and VEGFR2, has been highlighted for targeting both axes, expected to offer significant benefit comparing to targeting each individual pathway [33]. In phase I trial for RCC, the patients, most of whom had VEGF pathway inhibiting therapy, have shown a favorable safety profile and antitumor activity [34].

Casanova et al. provided the first report of fibroblast growth factor (FGF) involvement in the resistance to anti-VEGFR therapy by using RIP-Tag2 model, which is a genetically engineered mouse model producing pancreatic islet carcinoma [35]. In their study, although VEGFR2 blockade with an antibody initially achieved an impaired tumor growth and angiogenesis, significant regrowth of the treated tumor occurred during 4 weeks of treatment. Those tumors which grew during anti-VEGFR2 therapy showed a more invasive and malignant phenotype. Quantitative mRNA expression analysis in relapsing tumors revealed a significant upregulation of a set of proangiogenic factors, such as FGFs, ephrins, and angiopoietin, when compared with untreated tumors. In the condition of hypoxia, upregulation of some of these proangiogenic families was also observed in the cultured cell line, derived from RIP-Tag2 tumors. Interestingly, blocking of FGF by using a soluble form of the FGF receptor achieved a remarkable decrease in tumor volume and angiogenesis during the regrowth phase, especially when they are combined with anti-VEGFR2 treatment. Those indicated that increased production of FGFs could be involved in alternative pro-angiogenic pathway and partly explained the mechanism of development of resistance to VEGFR blockade. Subsequent related studies also have provided the effect of FGF blockade in the face of VEGF inhibition. Welti et al. reported that although sunitinib alone could provide a significant inhibition of endothelial tube formation in the presence of VEGF [36], FGF2 could suppress this antiangiogenic activity of sunitinib. Addition of FGF2 to cultures incubated with VEGF and sunitinib increased the endothelial proliferation and de novo tubule formation. Furthermore, Bello et al. demonstrated by using a human RCC xenograft model that renal tumors that had developed resistance to sunitinib still had sensitivity to a dual inhibitor of VEGFR and FGFR tyrosine kinase, E-3810 [37], suggesting the clinical utility of FGF antagonist in advanced RCC. However, the recent clinical trial using dovitinib, a multi-tyrosine kinase inhibitor of FGFR, VEGFR, and PDGFR, could not prove the superiority of FGFR-targeting therapy in the treatment of anti-VEGFR-refractory RCC patients [38].

Many of the proangiogenic factors upregulated under the hypoxic microenvironment of tumor contain hypoxia-response element (HRE) on their gene promoter [39]. This suggests that the HIF pathway partly regulates the compensatory gene expression in the tumors which has responded to antiangiogenic treatment, leading to the development of intratumoral hypoxia. However, Mizukami et al. revealed that even in the absence of functional HIF-1 protein, IL-8 could be induced and maintain vascularity of colon cancer xenograft. They also confirmed that neutralizing antibody to IL-8 achieves the regression of xenograft tumor lacking HIF-1 [40]. Wyscozynski et al. showed that blockade of both MAPK and PI3K-AKT pathways was necessary to inhibit the hypoxia-induced IL-8 expression in rhabdomyosarcoma cells, suggesting that nuclear factor (NF)-KB and activating protein (AP)-1 have an essential role in regulating its gene expression [41].

The potential role of IL-8 on mediating resistance to sunitinib treatment in RCC was demonstrated by Huang et al. [42]. They established a sunitinib-resistant human RCC xenograft model by administrating minimal dosage of drug with an intermittent dosing schedule. The levels of various secreted human cytokines derived from the xenograft tumor were screened. They found that IL-8 level of plasma sample from mice bearing sunitinib-resistant tumor was higher than that from mice with sunitinib-sensitive tumor. To demonstrate the functional significance of IL-8 in development of sunitinib resistance, neutralizing antibody to IL-8 was used in xenograft models. The combination treatment with IL-8 neutralizing antibody and sunitinib had a significant effect in decreasing the size and microvascular density of sunitinib-refractory tumor, even though the neutralizing antibody alone did not reduce the growth of the tumor. They conclude that inhibition of IL-8 function could resensitize the resistance tumors to sunitinib treatment, which result in decreasing RCC growth.

Although placental growth factor (PlGF) is a member of the VEGF family, it has a different characteristic from that of VEGF-A isoforms, which bind to VEGFR-1 and VEGFR-2. There is much evidence that VEGFR-2 is the major mediator of VEGF signaling pathway in endothelial cells [43], but PlGF exclusively binds to VEGFR-1, which is expressed by tumor cells, endothelial cells, bone marrow-derived proangiogenic and proinflammatory cells, and stromal cells [44]. The functional properties of VEGFR-1 are still under debate, because they seem to be different depending on the cell type. Although VEGFR-1 has ten times higher affinity to VEGF than VEGFR-2, it undergoes weak tyrosine phosphorylation in response to VEGF [43]. Therefore, VEGFR-1 is considered rather as a decoy receptor, which is able to negatively regulate the activation of VEGFR-2 by VEGF. Indeed, PlGF-deficient mice are healthy and fertile under physiological condition. However, in pathological conditions such as wound healing and inflammation, defect in pathological angiogenesis becomes obvious when a PlGF-deficient mouse was analyzed [45, 46]. The therapeutic potential of PlGF blocking in tumor growth was initially demonstrated by Luttun et al. [47]. Later on, Fisher et al. [48] demonstrated in a xenograft model of melanoma and colonic and pancreatic cancer that blockade of PlGF with a monoclonal antibody enhanced the antitumor activity of anti-VEGFR-2 therapy by inhibiting tumor angiogenesis and intratumoral macrophage recruitment. As for RCC, Rini et al. have revealed that serum levels of PlGF are increased in patients treated with anti-VEGFR therapy [49]; however, the treatment efficacy of PlGF blocking strategy in anti-VEGF treatment-resistance RCC is still under research [50].

In addition to the VEGF-VEGFR system, Tie receptors, and their ligands, angiopoietin has been regarded as second tyrosine kinase receptor signaling system which is specific to vascular endothelial cells [51]. This signaling is supposed to have an essential role in maturation and stabilization of the blood vessels. In RIP-Tag2 model, upregulation of angiopoietin mRNA was demonstrated in the tumor relapsing after initial anti-VEGFR2 treatment [35]. The association between evasive tumor resistance and anti-VEGF therapy and adaptive upregulation of Ang2 was revealed in the same model recently by Rigamonti et al. [52]. Therefore, it is not surprising that several recent evidences have shown that angiopoietin production decreases the effect of VEGFR-targeted therapies in other types of cancer. For example, Hashizume et al. demonstrated that the combination of Ang2 inhibitor with anti-VEGF antibody could limit the tumor growth and vasculature expansion more effectively than either of those agents alone in a colon cancer xenograft model [53]. The efficacy of human anti-Ang2 monoclonal antibody was confirmed in several other types of xenograft tumor, including RCC [54]. However, as for RCC, there is only limited evidence which suggests the therapeutic benefit of the combination therapy targeting Ang2 along with VEGF inhibition [55].