Targeted therapy for malignant glioma: neural stem cells

For better or for worse, stem cells are here to stay. In fact, it is hard to open a newspaper, turn on the television, or go to any scientific meeting without being bombarded about the potential promise which these cells hold for the treatment of cancer. Science aside, the ethical and legal issues which surround the use of stem cells are enough to fuel the controversy for years to come. Nevertheless, human stem cells are gaining increasing importance in neuro-oncology, and given the poor prognosis of most patients with malignant gliomas, their application to the treatment of brain tumors warrants further investigation. This review highlights some of the key milestones in the development of stem cell therapy for brain tumors.

Initial treatment options utilizing neural stem cells (NSCs) focused mostly on their ability to reconstitute abnormal or destroyed tissue, as in the case of neurodegenerative disorders or stroke. The appeal of this strategy was obvious, NSCs are self-renewing, totipotent cells that exhibit multilineage differentiation. With the aid of appropriate growth and differentiation signals, NSCs can yield cells with neuronal and glial phenotypes. However, three properties of NSCs, namely proliferation, self-renewal and migration, are reminiscent of neoplastic cells. While it is not absolutely clear whether cancer stem cells originate from normal stem cells due to genetic changes and/or redifferentiation from somatic tumor cells to the stem-like cells, the similarities inherent to both were quickly explored in the context of glioma genesis.

Recently, several groups have described the existence of a cancer stem cell population in human brain tumors of different phenotypes from both children and adults [1–3]. These brain tumor stem cells (BTSC) express the CD133 marker, as well as nestin, but show no evidence of differentiated neural lineage markers. Although BTSCs represent a minority fraction of the entire brain tumor cell population, they have been shown to possess a marked capacity for proliferation, self-renewal and differentiation. Moreover, BTSCs can be induced to differentiate in vitro into tumor cells that phenotypically resemble the tumor isolated from the patient.

However, the true measure of a BTSC is the capacity to recapitulate the original tumor in vivo. In a seminal article published in Nature, Singh and collegues have shown that only CD133+ brain tumor fractions were capable of tumor initiation in nonobese diabetic, severe combined immunodeficient (NOD-SCID) mice [4]. In fact, injection of as few as 100 CD133+ cells produced a tumor that could be serially transplanted and was a phenocopy of the patient’s original tumor, whereas injection of up to 10⁵ CD133- cells engrafted but did not cause tumor growth. The identification of tumor stem cells within adult gliomas represents a major step in understanding the origin of malignant brain tumors and clearly establishes a previously unidentified target for more effective cancer therapies.

Do brain tumor stem cells originate from NSCs? Evidence in support of this conclusion is strong and current literature supports the origin of malignant gliomas from a neural precursor stem cell which has undergone a process of self-renewal dysregulation [5–10]. Investigators have demonstrated that NSCs possess robust tropism for infiltrating tumor cells and, in fact, NSCs have been shown to track to intracerebral glioma microsatellites in vivo. As such, a lot of research has focused on whether or not NSCs can be used to deliver therapeutic agents directly to tumors, with the potential of significant therapeutic benefits.

The first clear evidence that NSCs could migrate towards intracranial glioma was shown by Aboody and colleagues [11]. Using an immortalized murine NSC line expressing β-galactosidase, the authors were able to localize the NSCs within the intracranial tumor following either intracerebral or intravenous injections. NSCs engineered to carry cytosine deaminase, when injected into mice with intracranial tumors, led to significant reduction in tumor burden as compared with controls. Such use of NSCs as delivery vehicles for tumor-toxic molecules represented the first experimental strategy aimed specifically at targeted disseminated tumor pockets.

Several other groups have built upon these initial efforts to expand the spectrum of NSCs. For instance, Ehtesham and colleagues explored the use of nonimmortalized, syngeneic NSCs [12]. After harvesting the cells from fetal mice and growing them in vitro, the authors infected NSCs with a replication–defective vector expressing β-galactosidase. The cells were then injected into an established syngeneic model of intracranial glioma. NSCs were readily identifiable within inoculated tumors and were clearly visible tracking glioma cells as they migrated away from the main tumor mass. In fact, injection of NSCs contralateral to the site of the tumor caused visible tracking across the brain into the immediate vicinity of the tumor. There was no evidence of random dispersion into adjacent normal tissue, nor could the NSCs be visible tracking to any distant nontumorous region of the brain. Local delivery of interleukin (IL)-12 or tumor necrosis factor (TNF)-α related apoptosis inducing ligand (TRAIL) led to significant tumor growth inhibition in vivo [12,13]. Taken together, these results provided strong evidence that NSCs can act as tumor-tracking delivery vehicles for therapeutic gene product delivery.

To date, one of the major limitations of stem cell research involves the use of fetal-derived NSCs. Significant ethical and legal concerns have been raised in this regard and in order to translate this therapy into a meaningful treatment, alternative sources of NSCs will have to be identified. When one further considers the immunologic factors and the potential for tissue rejection of allogeneic cells, the need to use autologous cells becomes clear. The identification of NSCs in the adult brain represents one way to overcome the present limitations. However, the need for an invasive procedure along with a low yield make this approach impractical in most clinical applications. Current research efforts have therefore focused on isolation of NSCs from adult bone marrow.

Several groups have successfully harvested cells with neuronal and glial phenotypes from adult bone marrow [14–18]. These cells are capable of self-renewal and differentiation into neurons, astrocytes, and oligodendroglia. They are also morphologically and phenotypically indistinguishable from fetal stem cells. Most importantly, just as in the case of fetal NSCs, adult bone-marrow derived NSCs track glioma tumor invasion in vivo. Nakamizo and colleagues have further shown that human bone-marrow derived mesenchymal stem cells (hMSCs) localize exclusively within brain tumors regardless of whether the cells are injected into ipsilateral or contralateral carotid artery [19]. When injected into the contralateral hemisphere, hMSCs were also shown to migrate and track human glioma xenografts. Most importantly, hMSCs engineered to release interferon (IFN)-β significantly increased animal survival compared with controls. These results indicate that hMSCs can integrate into human gliomas after intravascular or local delivery, and that this engraftment can be exploited to therapeutic advantage.

In spite of encouraging results in preclinical models, there are significant impediments that must be overcome prior to clinical implementation of stem cell therapy. Clearly, an understanding of the specific mechanisms which mediate NSC migration will allow us to precisely target these cells to the underlying pathology. In the case of intracranial tumors, early evidence suggests that brain tumor tropism of transplanted human NSCs may be induced by vascular endothelial growth factor (VEGF) [20]. This factor has been shown to induce long-range attraction of transplanted human NSCs from distant sites in the adult brain. Tumor–upregulated VEGF and angiogenic-activated microvasculature are therefore important signals for NSC tropism toward malignant brain tumors.

Another important consideration is the phenotypic characteristics of NSCs that exhibit tumor-tracking activity. Current data suggest that only a small subpopulation of NSCs migrate to the site of the tumor. This implies that not all stem cells will be available as vectors for delivery of therapeutic agents. Whether this subpopulation of cells expresses receptors for chemokines such as VEGF or as yet unidentified factors secreted by intracranial tumors remains to be shown. Furthermore, the inherent heterogeneity found in primary gliomas might result in only a proportion of tumor cells secreting the appropriate chemokine signal necessary for stem cell chemotaxis. A thorough understanding of the mechanisms which govern the tropism and migration of these cells will allow further refinement in the use of NSCs.

Although recent advances in the use of bone marrow-derived stem cells have put aside some of the concerns related to the use of fetal/embryonic cells, the yield of these cells remains very low when compared to fetal or embryonic tissue. As such, a critical impediment to successful use of adult bone marrow-derived stem cells will be harvesting enough of the cells to make a clinically significant impact in humans. This is especially important since stem cells will have to traverse significantly larger distances in humans in order to be of therapeutic value. When one further considers the need for stable and long-term expression of therapeutic gene products which are going to be delivered by NSCs, the importance of developing a sufficient supply of bone marrow-derived NSCs becomes clear. Such a goal will require not only further optimization of harvest techniques but also improvements in cell culture techniques so as to achieve an adequate number of stem cells.

Finally, the safety of stem cell therapy will have to be fully addressed prior to implementing this treatment in human clinical trials. As previously stated, three properties of stem cells – proliferation, the capacity for self-renewal and migration – are highly reminiscent of neoplastic cells. The potential for tumorigencity of transplanted primary NSCs has been established [21,22] and remains an important criteria by which to judge the success of any potential short- or long-term therapy. Elucidating the mechanisms involved in turning stem cells ‘on’ and ‘off’ may help to assuage the safety concerns and allow a way to regulate the delivery of the NSC vector and the gene to the cancer cell.

It is amazing to look at the research accomplished in the last 5 years and consider the great strides which have been undertaken to further our understanding of stem cell therapy for brain tumors. The appeals of this therapeutic approach are obvious; the challenges are equally daunting. Yet, given the fact that malignant brain tumors remain incurable and the currently available modes of therapy prolong survival by only several months, new and improved approaches are clearly needed if we are going to make an impact on this devastating disease. In many ways, the field of neuro-oncology has seen more advances in the last decade than during the last 100 years. While it may be optimistic to think that we are at a point of a significant breakthrough, the only impediments to our success are those which are self-imposed. The challenge is to create novel and targeted therapies which will spare normal and healthy brain tissue while effectively killing and removing cancer cells. Whether by themselves or in combination with other therapies, stem cells are likely to play a significant role in this evolving treatment paradigm for malignant glioma.