In the present work, A-07-GFP and R-18-GFP melanoma xenografts grown in dorsal window chambers were used as preclinical models. It has previously been shown that intradermal A-07 and R-18 xenografts retain several characteristic features of the original patient tumors, including histological appearance, angiogenic potential, and vessel density [31
]. Moreover, intradermal A-07 and R-18 xenografts have been shown to differ substantially in angiogenic potential, vessel density, growth rate, and oxygenation status [31
]. These differences were maintained when A-07-GFP and R-18-GFP tumors were grown in dorsal window chambers despite the transfection with GFP, the confinement of tumor growth by the chamber preparations, the small size of the tumors, and the elevated temperature during tumor growth. Consequently, A-07-GFP and R-18-GFP tumors grown in dorsal window chambers should be appropriate models for investigating the effect of sunitinib treatment on tumor vasculature and oxygenation.
Tumors were treated with two different doses of sunitinib in the current study. Both sunitinib doses have been shown to result in sufficient plasma concentrations in athymic mice to inhibit VEGFR and PDGFR phosphorylation in xenografts of human melanoma, human glioma, and human colon carcinoma [38
]. Lower sunitinib doses have been shown to result in insufficient plasma concentrations and no inhibition of VEGFR and PDGFR phosphorylation in the same xenograft models [38
]. Moreover, the higher sunitinib dose (40
mg/kg/day) has been shown to reduce vessel density and improve vascular function in human glioma xenografts [14
]. The two sunitinib doses should therefore be well suited to evaluate the effect of sunitinib treatment on tumor vasculature and tumor oxygenation.
The morphology of tumor vasculature was assessed by mapping tumor vascular networks with high-resolution transillumination images and filters for green light. In these images only vessels with erythrocytes can be seen and, consequently, the morphological analysis was based on vessels with erythrocytes as opposed to dysfunctional vessels with plasma only. The function of tumor vasculature was assessed by using a novel first-pass imaging method which involves recording movies of the dynamic distribution of a fluorescent vascular tracer after an intravenous bolus injection [33
]. From the recorded first-pass imaging movies, BST images and corresponding BST frequency distributions were produced. We have previously shown that the BST-assay is highly reproducible, sufficiently sensitive to detect gradients in BST along vessel segments, and sufficiently sensitive to indentify the majority of tumor vessels [33
It has previously been shown that A-07 and R-18 cells express and secrete VEGF-A and interleukin-8 (IL-8), and that the angiogenic activity can be significantly reduced by inhibiting VEGF-A in both xenograft lines [40
]. The secretion rate of VEGF-A and IL-8 has been shown to be higher for A-07 cells than for R-18 cells and, in addition, A-07 cells have been shown to express and secrete basic fibroblast growth factor (bFGF) whereas R-18 cells do not secrete this factor [40
]. In the present work, we show that sunitinib treatment significantly reduced vessel densities in both A-07-GFP and R-18-GFP tumors. The sunitinib-induced antiangiogenic effects were more pronounced for A-07-GFP tumors than for R-18-GFP tumors, and these differences probably reflected differences in the angiogenic profiles. Treatment-induced reductions in vessel densities are expected to reduce tumor growth. In the present study, prolonged sunitinib treatment significantly reduced tumor growth, whereas short-term sunitinib treatment did not affect tumor size. This observation illustrates that effects on tumor size generally are delayed compared to the effects on tumor vasculature after antiangiogenic treatment [41
Sunitinib treatment did not affect BST and, consequently, improved vascular function was not observed in the current study. Narrow, elongated, and tortuous tumor vessels are expected to elevate the geometric resistance to blood flow, and a potential increase in vessel diameter, decrease in vessel segment length, or reduction in vessel tortuosity is expected to enhance tumor blood flow [6
]. In the present study, we observed increased vessel segment lengths and unchanged vessel tortuosities after sunitinib treatment, and the remaining vessels in sunitinib-treated tumors showed similar or smaller increases in vessel diameter than vessels in untreated tumors. The effects on vascular morphology are thus in accordance with the observation that sunitinib treatment did not improve vascular function. Reduction in vessel density combined with no improvement of vascular function is expected to impair oxygen supply. In accordance with this, sunitinib treatment induced hypoxia in A-07-GFP and R-18-GFP tumors. Interestingly, A-07-GFP tumors differed substantially from R-18-GFP tumors in vessel maturation. Our study thus illustrates that sunitinib treatment fails to improve vascular function in melanoma xenografts with both high and low degrees of vessel maturation.
Sunitinib-induced improvement of vascular function has been reported in a preclinical study of human glioma xenografts [14
]. In that study, increased red blood cell velocities were observed 2, 4, and 6
days after the start of sunitinib treatment. The time points where improved vascular function was observed in that study thus correspond well to the time points where BST was assessed in our study (2, 4, and 8
days after the start of sunitinib treatment). Consequently, the lack of improved vascular function observed in our study was unlikely to be due to inadequate observation time points.
Treatment with anti-VEGF-A or anti-VEGFR-2 antibody has improved vascular function and tumor oxygenation in some preclinical models [13
]. The same antibodies have failed to improve vascular function and increased hypoxic fractions or have not affected tumor oxygenation in other preclinical models [19
]. Similarly, sunitinib-induced blockade of VEGFR and PDGFR has been reported to improve vascular function in human glioma xenografts [14
], whereas the current study shows that sunitinib treatment does not improve vascular function in human melanoma xenografts. Consequently, whether antiangiogenic treatment improves vascular function, does not reflect whether the antiangiogenic agent blocks PDGFR in addition to VEGFR, but more likely reflects differences in tumor models.
In addition to improve tumor blood supply and oxygenation, antiangiogenic agents have also been reported to reduce vessel permeability and lower tumor IFP in preclinical models. These effects have collectively been referred to as vascular normalization [17
]. Moreover, increased hypoxic fractions, reduced vessel permeability, and lowered tumor IFP have been observed simultaneously after antiangiogenic treatment [27
]. This observation suggests that improved tumor oxygenation, normalized vessel permeability, and normalized tumor IFP may not necessarily occur in parallel temporal windows [27
]. Consequently, although sunitinib treatment did not improve blood supply and oxygenation in A-07-GFP and R-18-GFP tumors, we cannot rule out the possibility that sunitinib treatment may normalize vessel permeability and tumor IFP in these tumor models.
Sunitinib has been shown to prolong progression-free and overall survival in patients with imatinib-refractory GIST and metastatic RCC in clinical phase III trials, and has been approved by the US Food and Drug Administration for these indications [29
]. However, the tumors eventually become unresponsive to sunitinib, and the benefits in progression-free and overall survival are measured in months. Treatment regimes that combine sunitinib with ionizing radiation, different chemotherapeutic agents, or other antiangiogenic agents may enhance and prolong the effects of sunitinib, and clinical studies that evaluate such combinations are ongoing [42
The current study and previously reported preclinical studies suggest that antiangiogenic treatment improves vascular function and tumor oxygenation in some clinical tumors, and induces hypoxia in others. Neoadjuvant antiangiogenic therapy may enhance the effect of ionizing radiation or chemotherapy in clinical tumors where antiangiogenic treatment improves vascular function. The feasibility of this strategy has been demonstrated in preclinical studies where maximal antitumor effect of ionizing radiation was achieved when tumors were irradiated within the time period when tumor oxygenation was improved by antiangiogenic treatment [13
]. On the other hand, in clinical tumors where antiangiogenic treatment induces hypoxia, neoadjuvant antiangiogenic therapy is expected to reduce the effect of ionizing radiation or chemotherapy [4
]. Consequently, antiangiogenic agents should not be considered as neoadjuvant therapy in combination with radiation or chemotherapy in such tumors. Whether it is possible to predict if antiangiogenic treatment can improve vascular function in a specific tumor is currently unknown. The effect of antiangiogenic treatment on tumor vasculature and on tumor oxygenation should thus be monitored closely if antiangiogenic treatment is considered as neoadjuvant therapy.