VEGFR3: A New Target for Antiangiogenesis Therapy?

Abstract

VEGFR-3 signaling plays an important role in developmental, physiological, and pathological angiogenesis and lymphangiogenesis. Tammela et al. in Nature show that VEGFR-3, via Notch regulation, is present on endothelial tip cells and is critical to sprouting angiogenesis.

The production of new blood vessels during development is an exquisitely regulated process that results in organ-specific vascular networks that fulfill required physiological and metabolic needs. A major mode of new vessel growth is via sprouting angiogenesis, in which a nonmitotic tip cell uses filopodia to guide a sprouting vessel toward an angiogenic stimulus. The tip cell is followed by mitotic stalk cells that extend the vessel and create a lumen capable of transporting blood. Vascular endothelial growth factor (VEGF) binds to VEGF receptor (VEGFR)-2 on tip cells and causes the cells to migrate up a concentration gradient, thus directing new vessel growth. On stalk cells, VEGF signaling via VEGFR-2 increases the rate of proliferation, allowing the sprouting vessel to continue to extend. Notch signaling, activated by its ligand Dll4, maintains the appropriate balance of tip to stalk cells to create proper sprouting and branching patterns in response to VEGF (Hellstrom et al., 2007). The regulation of sprouting angiogenesis by VEGF and Notch signaling, however, is also mediated by other signaling pathways, including netrins, semaphorins, and, as shown by Tammela et al. (2008), VEGFR-3.

Under physiologic conditions, VEGFR-3 is restricted to lymphatic and some fenestrated vascular endothelium in the adult. However, it is upregulated in angiogenic blood vessels in tumors and wounds, and blocking VEGFR-3 inhibits angiogenesis and growth in some tumors (Laakkonen et al., 2007). During development, deletion of VEGFR-3 results in cardiovascular failure, suggesting a critical role for VEGFR-3 in blood vessel formation (Dumont et al., 1998). Additionally, a blockade of Notch signaling in zebrafish results in increased angiogenic sprouting and VEGFR-3 induction (Siekmann and Lawson, 2007). However, the causal link between VEGFR-3/Notch signaling and sprouting angiogenesis was not known until now. Using a number of genetic and pharmacologic approaches, Tammela et al. (2008) reveal that the upregulation of VEGFR-3 in endothelial tip cells—mediated by the inhibition of Notch—plays a causal role in sprouting angiogenesis during early embryonic development, in postnatal development, and in tumors.

VEGFR-2 is considered the primary signaling receptor for VEGF during angiogenesis. Hence, many antiangiogenic agents are designed to block the VEGF/ VEGFR-2 pathway in tumors. After initial response, however, tumors can evade anti-VEGF/VEGFR-2 treatment by switching to different molecular pathways (Jain, 2005). Tammela et al. now show that blocking VEGFR-3 signaling can reduce the number of vessel branches and endothelial sprouts both during development and in tumors, resulting in a lower vascular density. In these models, the authors find strong expression of VEGFR-3 on the endothelial tip cells. These tip cells are also the cells in which the regulation of VEGFR-3 by Notch is critical for normal vascular formation, a regulation that may be lost in some tumors. Tammela et al. then combine VEGFR-2 with VEGFR-3 blockade and show additive antiangiogenic and antitumor effects. Moreover, they show that VEGFR-3 signals can sustain a low level of angiogenesis even in the presence of VEGFR-2 blockers. Thus, they propose that a combination of VEGF/VEGFR-2 and VEGFR-3 blockade may improve the outcome of anti-VEGF therapies.

This proposal raises many critical questions about translating this exciting finding to the clinic: Will such an approach work for different tumor types? Will this lead to a durable response in patients? Finally, how can this dual targeting approach, which is designed to destroy tumor vessels, be combined with chemo and/or radiation therapy, which require blood vessels for the delivery of drugs and oxygen (a known sensitizer of radiation and various anticancer drugs)? Addressing these translational issues is both necessary and urgent to effectively use the extensive pipeline of tyrosine kinase inhibitors that target both VEGFR-2 and -3 pathways in various phases of clinical development (Jain, 2005).

How does tumor-to-tumor variability affect the response to combined VEGFR-2 and -3 blockade? Laakkonen et al. (2007), showed that different tumors respond to VEGFR-3 blockade to different degrees, including some tumors that do not respond at all. Furthermore, emerging data suggest that VEGFR-3 is expressed on cancer cells (Su et al., 2008), which can lead to a direct antitumor effect of VEGFR-3 blockade, although this remains controversial. Thus, the responsiveness of individual tumors to VEGFR-3 blockade needs to be considered when implementing this therapy.

Can combined VEGFR-2 and -3 blockade proposed by Tammela et al. lead to a durable antiangiogenic response in patients? In a Phase II trial, glioblastoma patients responded to cediranib, a VEGFR-2 and VEGFR-3 tyrosine kinase inhibitor, for several months, but progressed eventually in essentially the same time frame as with bevacizumab (Avastin), which targets VEGF and has no direct anti-VEGFR-3 activity (Batchelor et al., 2007; Norden et al., 2008). Additionally, in animal models, combined VEGFR- 2 and VEGFR-3 blockade with cediranib did not inhibit established lymphatic metastasis (Padera et al., 2008). These data suggest that even combined blockade of VEGFR-2 and VEGFR-3 may not be adequate.

Tammela et al. show that dual targeting prunes more vessels than targeting VEGFR-2 or -3 alone. However, even with this increased antivascular effect, the tumors do not regress. Thus, tumor regression and total eradication will require the addition of drugs and/or radiation that destroy cancer cells directly. It is interesting to note that blocking VEGFR-2 normalizes the abnormal vessels in skin adenovirally transduced with VEGF and VEGF-C, similar to what is seen in some tumors treated with this same anti- VEGFR-2 antibody (Jain, 2005). Furthermore, blocking both VEGFR-2 and –3 with antibodies makes these angiogenic skin vessels look even more normal, like the control vasculature (Tammela et al., 2008). So it is probable that dual targeting will lead to a tumor vasculature more efficient for drug and oxygen delivery, making these tumors more vulnerable to chemotherapy and radiation (Jain, 2005). The challenge will be to identify the optimal dose and schedule of these agents for different tumors.

In conclusion, Tammela et al. have linked the Notch pathway to the production of an abnormal tumor vasculature through the control of VEGFR-3 expression on endothelial tip cells and its role in endothelial sprouting. It is important to test if targeting VEGFR-3 in tumors can improve the outcome of combined anti- VEGF/VEGFR-2 and cytotoxic therapies and lead to longer progression-free and overall survival.