Tumor angiogenesis, from foe to friend

See allHide authors and affiliations

Science  14 Aug 2015:
Vol. 349, Issue 6249, pp. 694-695
DOI: 10.1126/science.aad0862

Targeting the tumor vasculature to “starve a tumor to death” instead of targeting tumor cells with chemotherapeutic drugs was conceived over four decades ago and has led to the development of antiangiogenic drugs approved for use against various human malignancies (1). So far, however, antiangiogenic therapy has not fulfilled expectations because it aids only a subset of cancer patients and provides only transitory improvements. Vascular-disrupting agents were developed to more rigorously deplete tumor vessels (2). However, this approach leads to hypoxia, which promotes neovascularization and tumor regrowth. The sobering realization is that the more we try to exterminate tumor vessels, the more aggressively tumors respond to impede these efforts, sometimes becoming more belligerent tumors. Is manipulating the vasculature to control tumor growth a promising strategy after all?

Conventional angiogenic inhibitors primarily block the vascular endothelial growth factor (VEGF) signaling pathway, and prune rather than eradicate tumor vessels. By contrast, some flavonoids and microtubule-binding molecules rapidly and selectively kill the tumor vasculature (2). The latter, wide-scale vessel destruction results in a more massive necrosis and hypoxia than occur with antiangiogenic therapy, and therefore is effective at killing the bulk of the tumor. This reaction seems to be primarily driven by low oxygen tension. Conversely, hypoxia also triggers several resistance mechanisms that drive tumor regrowth. It promotes the epithelial-mesenchymal transition and stemlike properties of tumor cells, increases the expression of proangiogenic and invasive factors, and drives the infiltration and polarization of angiogenic and immune-suppressive myeloid cells (1, 3).

The approach of eradicating tumor vessels therefore seemed to be at a dead end until a new opportunity arose with the concept of vessel normalization. Detailed analysis of the changes resulting from inhibition of VEGF and the VEGF receptor (VEGFR) revealed that antiangiogenic therapies leave behind a more mature and functional vasculature by selectively pruning immature blood vessels (4). This results in enhanced oxygenation and perfusion throughout the tumor and subsequent activation of antitumor immunity, thus creating conditions for better drug delivery and efficacy. Vascular normalization with VEGF and VEGFR inhibitors has indeed been demonstrated in several preclinical cancer models. However, the inability to finely tune antiangiogenic therapy to create persistent normalization without further pruning leads to a recurrence of hypoxia and the emergence of acquired resistance.

Recently, use of an alternative strategy showed that the antimalarial agent chloroquine decreased tumor blood vessel tortuosity and vessel density while increasing endothelial cell organization, the coverage of endothelial cells by pericytes, and vessel perfusion (5). This enhanced tumor oxygenation and increased efficacy of chemotherapy. Chloroquine's alkalinizing nature hinders the endosomal cycling process, a mechanism that controls the trafficking of surface proteins between the cytoplasm and cell surface membrane. Indeed, the compound's capacity to normalize the vasculature was due to abnormal cycling and signaling of the surface protein Notch1, a negative-regulator of vascular sprouts. Vessel normalization to enhance vascular function has thus become an important concept, with the goal of improving delivery of chemotherapy and promoting a more oxygenized and immune-stimulating tumor environment.

The idea of manipulating the tumor vasculature recently advanced a step further with the demonstration that stimulating tumor angiogenesis can sensitize tumors to chemotherapy (6). The study took advantage of the angiogenic properties of cilengitide, a cyclic pentapeptide that binds to the integrins αVβ3 and αVβ5 and blocks cells adhesion. It also used verapimil, a Ca2+-channel blocker. Cilengitide, when administered at low doses, can alter VEGF receptor trafficking on endothelial cells to favor VEGF-induced angiogenesis (7), and verapimil induces vasodilation and subsequent blood flow. The combination of verapimil and cilengitide increased vessel density, dilation, permeability, and perfusion within tumors, which in turn increased tumor oxygenation and enabled more efficient delivery and enhanced efficacy of the chemotherapeutic drug gemcitabine. Importantly, the beneficial effects in rodent models of cancer were not just obtained by better drug delivery to the tumor site but also by a more efficient drug influx into the tumor cells and its subesquent activation. Gemcitabine uptake into cells is regulated by equilibrative nucleoside transporters 1 and 2 (ENT1 and ENT2) and by concentrative nucleoside transporter 3 (CNT3), which mediates the unidirectional flow of the drug into the cell. On entering the tumor cell, gemcitabine is converted into its active form by deoxycytidine kinase (DCK). Treatment with the triple combination of verapimil, cilengitide, and gemcitabine enhanced ENT1, ENT2, CNT3, and DCK expression, in part by lowering hypoxia and thereby increasing the intratumoral influx of gemcitabine (6).

Exploiting the vasculature to control tumors.

Current antiangiogenic approaches aim to sustain vessel normalization—the first phase of blood vessel pruning (before the onset of increased hypoxia)—for enhanced oxygenation, drug delivery, and efficacy. These goals are shared by vascular promotion therapy. So far, antiangiogenic and vessel-disrupting agents have led to tumor relapses caused by reneovascularization or altered tumor behavior. It remains to be seen whether vascular promotion follows a similar path.


The success of the vascular promotion approach both in tumor models that were either already angiogenic (Lewis lung cancer) or hypovascular (a genetic model of pancreatic adenocarcinoma that responds poorly to its standard-of-care gemcitabine) suggests a more general applicability. A better mechanistic understanding of this strategy is needed to move vascular promotion forward to the clinic. Clearly, gemcitabine appears to be the drug of choice at this stage due to its enhanced uptake and activation, whereas efficacy of another chemotherapeutic drug, cisplatin, was only improved as a result of better drug delivery to the tumor.

Can other chemotherapeutic drugs that enter the cells via ENTs and CNTs be used with this approach? Are other drug transporters regulated by hypoxia and if so, which chemotherapeutics would then most benefit from this strategy? Given that celingitide can have both pro- and antiangiogenic effects in a concentration-dependent manner that may not be easily assessed in the clinic, alternative approaches to promote angiogenesis in a controlled manner should be considered. Also unclear is whether enhanced angiogenesis is even needed in already highly angiogenic tumors or if verapimil would suffice to improve blood flow and oxygenation. Regardless, the idea of vascular promotion supports the important revelation that one can target the vasculature but leave it intact and thereby provide therapeutic benefits. This has been successfully exploited to reveal angiocrine signaling cues between endothelial cells and tumor cells that promote tumor growth and invasion (8). For example, lymphoma-derived fibroblast growth factor 4 induced expression of the Notch ligand Jagged 1 in endothelial cells, which in turn induced Notch-dependent lymphoma invasion and resistance to chemotherapy; deletion of the gene encoding focal adhesion kinase in endothelial cells led to a reduction in several secreted factors without affecting vessel density and enhanced tumor sensitivity to chemotherapies (9, 10). These results suggest that targeting angiocrine factors can provide therapeutic benefits without affecting vascular function and potentially be exploited for vascular promotion therapy.

Concerns about the nature of tumor relapse from vascular promotion should be investigated, specifically if influx of chemotherapeutic drugs into tumor cells stops due to other mechanisms. This could theoretically lead to a fast regrowth of the tumor with potentially enhanced metastasis as a likely by-product of vessel leakiness and tumor angiogenesis. Designing trials of vascular promotion for tumors for which gemcitabine has been standard care would be a first important step toward clinical validation. If successful, vascular promotion in such or other combinations—for example, with immune therapies—could prove to be an important new interventional approach in cancer therapy.


View Abstract

Stay Connected to Science

Navigate This Article