Forty years ago, President Richard Nixon signed the National Cancer Act, declaring a war on cancer, with the goal of eliminating the disease as a leading cause of death in the United States. Since then, many new advances in the field of cancer biology and therapeutics have been developed. Judah Folkman initiated a major advancement in the war on cancer in the 1970s by hypothesizing that blocking tumor angiogenesis, the process by which tumors grow new blood vessels, malignancies would essentially be starved of oxygen and nutrients. Based on Folkman's proposed mechanism, several therapeutic agents that target neovascularization have been developed. These agents include antibodies such as bevacizumab (Roche), targeting vascular endothelial growth factor (VEGF), and receptor tyrosine kinase inhibitors (RTKI), such as sunitinib (Pfizer), that block VEGF receptor signaling.1 However, preclinical and clinical studies have recently brought the effectiveness of such therapies, especially for breast cancer, into question. Clinical trials have revealed limited benefit, with rare lasting responses, only a moderate increase in progression-free survival, and little benefit in overall survival of patients.2,3 In light of the limited clinical benefits demonstrated, a USFDA panel has recently revoked its approval of bevacizumab for treatment of advanced breast cancer.
Meanwhile, paradigm-shifting developments in cancer research have recently widened our understanding of tumor biology. Emerging evidence supports the cancer stem cell (CSC) hypothesis, which posits that many cancers, including breast cancer, are driven by a subpopulation of cells displaying stem cell properties. These CSCs may also be responsible for mediating tumor metastasis and therapeutic resistance. Several lines of evidence have demonstrated that, similar to both embyonic stem cells and normal adult stem cells, CSCs are regulated by the oxygen levels within their microenvironment. Since the administration of antiangiogenic agents has been shown to generate intratumoral hypoxia, these observations led us to hypothesize that hypoxia generated by antiangiogenic therapies may stimulate the population of CSCs and contribute to the resistance of these drugs. In a recent study, we found that antiangiogenic drugs, including sunitinib and bevacizumab, do, in fact, increase breast CSCs.4 Although the antiangiogenic drugs worked well in terms of reducing tumor volume of human breast xenografts grown in immunocompromised mice, measuring the levels of the stem cell marker aldehyde dehydrogenase and the growth of secondary tumors identified an increase in the CSCs. The greatest increase in CSC populations was seen when therapy was initiated to treat large, established tumors, likely owing to greater overall volume for hypoxia to develop. Most interestingly, immunohistochemical studies revealed numerous clusters of CSCs residing directly within and around the tumors' hypoxic zones. Our results are supported by preclinical data indicating that antiangiogenic agents in fact increase metastatic and invasive properties of breast cancer xenografts in mice.5,6 Moreover, a newly published clinical study examining sunitinib in combination with chemotherapy reported outcomes consistent with our findings, where treatment of advanced breast cancer patients with combined sunitinib and chemotherapy resulted in an increased response rate but had no effect on progression free- or overall survival compared with chemotherapy alone.7
The findings in our study provide a potential explanation for the limited clinical effectiveness of antiangiogenic therapies in breast cancer as well as a number of other solid malignancies. It has long been hypothesized that the hypoxia generated within tumors by antiangiogenics may stimulate invasion and metastasis of cancer cells via gene regulation by hypoxiainducible factor (HIF), leading to drug resistance.8 Our findings extend this mechanism to include a hypoxia/HIF-1α-mediated increase in CSCs. We may now add antiangiogenics to the list of existing cancer therapies to which CSCs display intrinsic resistance, including chemo- and radiation therapy, as well as some molecularly targeted therapeutics. Although these traditional treatments are highly effective at debulking the majority of the tumor mass, the CSCs end up repopulating the cells lost within the tumor following treatment. This suggests that improving the clinical efficacy of current cancer therapies, including antiangiogenic treatments, will require combined use with CSC-targeting agents. The benefit of drugs specifically targeting CSCs is that they should truly produce a lasting clinical response. Several CSC-targeted agents are under development, and we have initiated studies to examine the strategy of combining CSC-targeting drugs with antiangiogenic agents. This suggests that improving the clinical efficacy of current cancer therapies, including antiangiogenic treatments, will require combined use with CSC-targeting agents as illustrated in Figure 1. We anticipate that successful outcomes of these studies will lead to the development of new clinical trials utilizing these combinations. By simultaneously targeting CSCs and bulk tumor populations, these approaches hold the promise of improving patient outcome.