a, Certain cancers are treated with antiangiogenic drugs (which affect blood-vessel formation), either alone or in combination with cytotoxic agents that inhibit the growth of cancer cells. Some of these cells (green) are particularly dangerous because they can be more resistant to cytotoxic therapy than the other tumour cells and can spread to other organs to seed new tumours. b,c, A decreased blood supply to the tumour, which is the main benefit from antiangiogenic therapy, is also the basis for the therapy's limitations. b, It can reduce the distribution of cytotoxic agents in the tumour, and hence their efficacy. Van der Veldt et al.¹ report one such undesirable outcome in patients with non-small-cell lung cancer. c, By reducing oxygen levels in the tumour, antiangiogenic drugs can induce the accumulation of more aggressive cells that have an increased capacity to spread to other organs. Conley et al.² document this phenomenon in mouse models of breast cancer in the absence of a cytotoxic agent.

To test the theory in a clinical setting, Van der Veldt and colleagues¹ studied the uptake and retention of a cytotoxic drug (docetaxel) in 10 patients with advanced-stage non-small-cell lung cancer (NSCLC). By using radiolabelled docetaxel together with a sensitive imaging method (positron emission tomography), the authors demonstrate that VEGF inhibition with a drug called bevacizumab induces a fast and sustained decrease — not increase — in the penetration of both water and docetaxel in the tumours (Fig. 1b). These results contrast with those of previous studies in patients with rectal cancer⁶ and in patients with glioblastoma (a brain tumour)⁷, which showed that bevacizumab treatment induces vascular normalization and increased glucose uptake in the tumours. However, glucose uptake by tumour cells does not, in my opinion, necessarily correlate with cytotoxic-drug delivery and penetration into tumours.

The discrepancies between the authors' observations¹ and previous results^{6, 7} could also be due to differences in blood-vessel networks and in the response to angiogenesis inhibitors between the three cancer types, as it is known that these agents can affect blood vessels in different ways in different tissues^{7, 8}. In any case, the finding¹ that, at least in patients with NSCLC, antiangiogenic therapy does not improve cytotoxic drug delivery to tumours — but rather has the opposite effect — exposes a perturbing drawback to such treatments. Indeed, this could be the cause of the modest benefits of these combination therapies in NSCLC and other tumour types⁹. Such a potential shortcoming could be circumvented by optimizing the scheduling of the therapeutic agents. For example, rather than administering both types of drug to a patient during the same period, treatment with blood-distributed cytotoxic agents could be followed by antiangiogenic therapy.

Other limitations of the use of angiogenesis inhibitors derive from the fact that tumours are highly adaptable. Although their ability to become resistant to cytotoxic drugs and radiation — another common anticancer therapy — has long been known, it was initially postulated¹⁰ that antiangiogenic drugs would not suffer from the same problem because they target blood vessels rather than tumour cells. Yet preclinical and clinical evidence^{11, 12} has revealed that tumours can indeed adapt and become resistant to antiangiogenic therapy.

As if that was not bad enough, angiogenesis inhibitors have been shown^{13, 14} to make some tumours more aggressive in animal models (Fig. 1c). To explore this issue, Conley and co-workers² implanted human cancer cells (derived from established breast- cancer cell lines) in mice. They then treated the animals with the antiangiogenic agents sunitinib — which inhibits VEGF's main cell-surface receptors — and bevacizumab. The treatment induced an accumulation of certain cancer cells that expressed the enzyme aldehyde dehydrogenase and that, like cancer progenitor cells, could initiate tumours when reimplanted in other mice. Similar cell populations have been described in tissue samples from patients with inflammatory breast cancer¹⁵ and in glioblastoma in mice given combination therapies¹⁶.

Conley et al.² go on to delineate a possible cellular and molecular mechanism for the increased aggressiveness and spread capacity of tumours treated with antiangiogenic drugs. They find that the drugs, by inducing oxygen deficiency (hypoxia) in the tumours, activate not only a hypoxia-response program but also the Akt/β-catenin signalling pathway, which regulates cell growth and adhesion between cells. This pathway has been previously implicated in the regulation of breast-cancer progenitor cells¹⁷. The authors suggest that the drug-induced hypoxia response activates the Akt/β-catenin pathway, which in turn stimulates the growth of specific, more aggressive, cancer-cell populations.

How could this drawback of antiangiogenic therapies be overcome? One possibility would be to combine angiogenesis inhibitors with drugs that suppress the cancer cells' response to hypoxia, or with inhibitors of the Akt/β-catenin pathway. Another alternative could be the use of molecules such as modified semaphorin proteins¹⁸, which can exert dual (or multiple) anticancer effects by simultaneously targeting angiogenesis and blocking tumour spread.

Overall, the papers by Van der Veldt et al.¹ and Conley et al.² emphasize the need for a carefully balanced evaluation of the benefits and limitations of antiangiogenic therapies. As mentioned above, such treatments could be improved by sequential scheduling of cytotoxic and antiangiogenic drugs, or by smarter combinations of these drugs with others targeting progenitor-cell pathways. In any case, despite many open questions, there is hope that an understanding of the therapies' weaknesses will translate into therapeutic gains.

http://www.nature.com/nature/journal/v484/n7392/full/484044a.html#/references