Robert+Kousnetsov

Glioblastoma: A Deadly, Relentless Brain Cancer John Doe is a 57-year-old man presenting with severe headaches and minor seizures. The patient reported that he started experiencing sharp headaches a few months ago that progressively increased in duration and intensity. In addition, the patient reported that he began having small tremors in his arms and legs a few weeks ago. Based on interviews with the patient and his family as well as a review of the patient’s medical records, it was concluded that the patient does not have any notable medical history of or significant genetic predisposition to neurological disorders. In order to check for any neoplastic growths or other abnormalities that may have caused his symptoms, the patient was sent for magnetic resonance imaging (MRI) of his head. The MRI revealed a significant lesion in the patient’s brain that was suspected to have been caused by a tumor. After consultation with the patient, the affected region was surgically removed and sent for further study. Histological examination of the biopsied sample indicated that the patient has glioblastoma.
 * Patient description**

Glial cells are a diverse class of cells that reside in the nervous system and support neurons. Some important glial cells in the central nervous system (CNS) include astrocytes, oligodendrocytes, and microglia. Astrocytes nourish neurons and regulate the chemical environment at neuronal synapses (1). Oligodendrocytes are responsible for myelinating axons in order to accelerate neuronal transmission (1). Microglia serve a role similar to macrophages, helping clear cellular debris and coordinate immune responses (1). Glioblastoma is a cancer of the glial cells (a glioma); more specifically, glioblastoma arises in the CNS from astrocytes and their precursors. For this reason, glioblastoma is also known as grade IV astrocytoma, where astrocytoma refers to a cancer of the astrocytes and grade IV refers to the highly aggressive, invasive nature of the cancer (2). Note that grade IV is reserved for the most severe, malignant cancers according to the 4-point grading system used by the World Health Organization (WHO). Glioblastoma is sometimes referred to as glioblastoma multiforme (GBM) because of its cellular and genetic heterogeneity. That is, glioblastoma contains a diverse collection of cells in and surrounding the tumor site, including a core of necrotic cells, a vast array of blood vessels enveloping the tumor, an interspersed population of endothelial cells, fibroblasts, and immune cells, as well as a large layer of rapidly-dividing, partially differentiated, pleomorphic cells (pleomorphic refers to the wide variety of shapes and sizes of tumor cells and their nuclei when stained and observed under a microscope) (3, 4). Similarly, GBM may harbor many different genetic and epigenetic aberrations, including various gain of function mutations associated with proliferative signal transduction cascades and several loss of function mutations associated with tumor suppression and DNA repair pathways (4). The precise cause of glioblastoma is unknown. Glioblastoma can develop spontaneously or from lower grade astrocytomas; respectively, these cancers are classified as primary (de novo) and secondary glioblastomas (2). Risk factors for glioblastoma include various rare genetic diseases and ionizing radiation (5). There is no cure or reliable treatment method for glioblastoma. Currently, treatment of glioblastoma typically involves surgical resection of the initial tumor followed by a combination of radiation therapy and chemotherapy to delay the resurgence of the cancer and manage symptoms (5). Unfortunately, owing to the invasive, fast-growing nature of glioblastoma, such therapeutic methods often fail to completely eliminate the cancer. Consequently, glioblastoma tends to recur, often spreading to various parts of the brain (see figure 1) and making further treatment futile (2). Overall, the prognosis for glioblastoma is grim (see figure 2): on average, untreated patients die 2-3 months after diagnosis, while treated patients die approximately 12-15 months after diagnosis (3). Due to the deadliness of this disease and the unsuccessfulness of traditional cancer treatments, numerous immunotherapies and targeted drug therapies are being tested for the treatment of glioblastoma (6, 7).
 * Background on glioblastoma**


 * Figure 1.** MRI scans of glioblastoma demonstrating its resilience and invasiveness (3). (A) Glioblastoma in the right temporal lobe is marked by the arrowhead. (B) Resection cavity following surgical removal of glioblastoma marked by the arrowhead. (C) Recurrence of glioblastoma 6 months after surgery at both the original site and a new site (the frontal lobe), both marked by arrowheads. (D) Surgical removal of both recurrent glioblastomas. (E) Recurrence of glioblastoma 3 months after second surgical resection. Glioblastoma has spread to the left hemisphere via the corpus callosum.


 * Figure 2.** Kaplan-Meier survival plot for glioblastoma (3). The y-axis refers to the percent survival of patients. Extensive resection refers to surgical removal of tumors, XRT refers to radiation therapy, and chemo refers to chemotherapy. Observe that even with combination treatments, the long-term survival rate is approximately 10% (which is shockingly low).

The development of cancer, glioblastoma included, involves an accumulation of mutations that give rise to a set of defining characteristics which are termed “the hallmarks of cancer” (8). The following section will outline some of the main mutations observed in glioblastoma, identifying the hallmark principle they belong to and describing the molecular mechanism by which they contribute to cancer progression. Before continuing, it is important to note that glioblastoma has been shown to exhibit all the major hallmarks of cancer. While this is not surprising (cancer requires many of these hallmarks to exist), I consider glioblastoma to be particularly dangerous because of its ability to quickly divide. That is, I believe that sustained proliferative signaling is the most critical of all cancer hallmarks with respect to glioblastoma. Not only does this hallmark enable glioblastoma to rapidly grow and spread but it also facilitates the microevolution of the cancer. As a result, glioblastoma can swiftly develop methods of resistance towards therapeutic interventions of any kind. The fact that glioblastoma is highly aggressive and invasive is alone quite frightening. Glioblastoma’s emergence in and invasion of the brain is also quite disturbing in itself, given that the brain is both difficult to treat and is arguably the most important and cherished organ in the human body (note that glioblastoma is not solely restricted to the brain, it can develop in or spread to the central nervous system). Yet, from a medical standpoint, the idea that glioblastoma is highly adaptive is incredibly distressing because this phenomenon likely presents the most significant obstacle towards developing a reliable form of treatment.
 * Molecular mechanisms overview**

Glioblastoma tends to be an aggressive cancer, signifying that it can proliferate rapidly. There are multiple avenues by which glioblastoma acquires its fearsome proliferative capacity. Many such mutations involve proteins associated with the EGFR/Ras/MAPK and PI3K/Akt/mTOR pathways, both of which are critical in regulating proliferation via control over the cell cycle. There exist several classes of receptor tyrosine kinases (RTKs) which are responsible for initiating signal transduction cascades in response to a variety of growth factors that culminate in cell growth and division. Mutations associated with these RTKs are capable of producing sustained growth signals that cause excessive proliferation. The most commonly mutated RTK in glioblastoma is the epidermal growth factor receptor (EGFR). The gene encoding EGFR is often amplified in glioblastoma, resulting in overexpression of EGFR that renders a cancerous cell more sensitive to growth factors (such as EGF). Consequently, cancerous glial cells may proliferate excessively even under normal concentrations of growth factors. In addition, some EGFR variants, such as EGFRvIII (which contains a deletion in its coding region), are able to initiate signals independently of their ligand. In this manner, EGFR can lead to permanent activation of proliferative pathways (7). Other RTKs undergo similar mutations, such as amplification or structural modification of platlet derived growth factor receptor (PDGFR) and human epithelial receptor 2 (HER2) (7). The RTKs above can activate the Ras G-protein or Phosphoinositide 3-kinase (PI3K). Ras can activate several kinases, such as Raf or PI3K, which, in turn, can activate other downstream kinases. Ultimately, the activation of extracellular signal-regulated kinase/mitogen activated protein kinase (ERK/MAPK) via the Ras-Raf-MEK-ERK pathway and the activation of protein kinase B (PKB/Akt) via the PI3K/Akt/mTOR pathway results in cell proliferation, cell survival, and upregulated protein synthesis (7). While there are many proteins involved in these complex pathways, it is simplest to focus on the main controllers of each pathway: Ras and PI3K. In addition to persistent activation by upstream elements such as RTKs, it is possible to achieve sustained proliferative signaling by eliminating the inhibitors of Ras and PI3K, which include neurofibromin 1 (NF1) and phosphatase and tensin homolog (PTEN), respectively. NF1 is a GTPase activating protein (GAP): it accelerates the hydrolysis of GTP to GDP on Ras, thereby inactivating Ras. PTEN is a phosphatase: it cleaves a phosphate from PI3K, thereby inactivating PI3K. Homozygous deletions or mutations in NF1 and PTEN can cause Ras and PI3K to remain constitutively active, resulting in uncontrolled cell growth and proliferation (7).
 * Sustained proliferative signaling**

The decision of a cell to commit itself to division is a highly regulated process. Positive and negative growth signals are integrated and cellular conditions are checked prior to advancing the cell cycle. For this reason, mutations in proliferative pathways, such as the ones above, are often not sufficient to lead to cancer. Two of the critical pathways involved in cell cycle regulation involve the retinoblastoma protein (Rb) and tumor protein 53 (TP53). Tumor suppression in the p16INK4a/CDK4/RB1 system entails preventing the passage of cells from G1 phase to S phase via transcriptional and post-translational repression of cyclins (Cyc) and cyclin dependent kinases (CDKs). For example, the p16INK4a protein prevents the binding of CycD to CDK4/6, while the Rb protein prevents expression of various genes needed for the execution of S phase (such as replication machinery and CycE) (7). Among other functions, the CycD/CDK4 and CycE/CDK2 complexes hypo- and hyper-phosphorylate Rb, respectively. When Rb is inactivated by phosphorylation, it can no longer bind the E2F family of transcription factors, thereby permitting transcription of S phase associated genes. Thus, the Rb system is responsible for integration of external signals and acts as a major gatekeeper of the cell cycle via the G1 restriction point. Glioblastoma can bypass Rb regulation via deletions that result in the complete loss of genes encoding p16 or Rb or via point mutations that culminate in the loss of function of these proteins. These effects can be achieved via similar mutations in upstream regulatory elements, such as the cyclin-dependent kinase inhibitor genes (CDKN2A, CDKN2B, CDKN2C), which control p16and Rb expression and activity (7). Ultimately, mutations in the Rb pathway render a cancer cell unresponsive to growth suppressors. Tumor suppression by TP53 involves sensing adverse cell conditions, which may include metabolic stress or DNA damage (both of which are common in cancer). Not only can TP53 temporarily suspend various phases of the cell cycle until a favorable cell state is restored, but TP53 can also initiate an apoptotic cascade when a cell is deemed unsalvageable (8). In this manner, TP53 helps safeguard a cell from genomic alterations and helps contain the spread of aberrant cells. Dysregulation of TP53 as well as its regulators or effectors are common mechanisms by which glioblastoma escapes cell death and maintains genomic instability. For example, loss of function mutations in the DNA binding domain of TP53 are fairly prevalent, causing TP53 to be unable to bind and upregulate the effector genes that are responsible for halting the cell cycle or initiating programmed cell death (7). Alternatively, effector genes of TP53 may be deleted in order to have an equivalent effect of losing regulatory control over the cell. For instance, homozygous deletions of p14ARF can occur, thereby eliminating this tumor suppressor protein which is critical in promoting cell cycle arrest and apoptosis upon activation by TP53 (7). Lastly, another method of silencing TP53 pathways involves TP53 destruction. Herein, glioblastoma can overexpress Mdm2, which is an E3 ubiquitin ligase that ubiquitinates TP53, thereby targeting it for degradation by a proteasome (7).
 * Evading growth suppressors and resisting cell death**




 * Figure 3.** Summary of mutations in major molecular pathways associated with malignant gliomas (MG, which includes glioblastoma) (7). Lightning bolts denote a mutation, large red bars denote gene amplification, and small red bars denote homozygous deletion.

The molecular mechanisms discussed above were responsible for “sustained cell proliferation, evading growth suppressors, and resisting cell death” (8). These pathways are summarized in figure 3. While these are the most well-studied characteristics of glioblastoma, there are other proteins that contribute to cancer development. For example, upregulation of telomerase is critical for helping cancerous glial cells “enable replicative immortality” (7, 8). Herein, maintenance of telomeres on the ends of the chromosomes prevents DNA loss associated with the end replication problem, thereby preventing senescence or apoptosis from excessive replication. In addition, glioblastoma is able to recruit additional blood vessels to its tumor site via heightened expression of vascular endothelial growth factor (VEGF), thereby “inducing angiogenesis” (7, 8). In this manner, glioblastoma is able to fuel rapidly growing tumors with nutrients (most importantly glucose and oxygen). Lastly, it has been shown that matrix metalloproteases (MMPs), which are responsible for the breakdown of extracellular matrix and necessary for invasion and metastasis, are indeed upregulated in glioblastoma (7, 8). This is consistent with the highly invasive nature of glioblastoma.
 * Other classical hallmarks**

Interestingly, some glioblastomas also demonstrate altered metabolism via isocitrate dehydrogenase (IDH1/2) mutations. As a result, glioblastoma can produce 2-hydroxyglutarate (2-HG) from α-ketoglutarate (α-KG) during the citric acid cycle. 2-HG is an oncometabolite that can inhibit histone and DNA modifying enzymes (4). As a consequence of this interference with epigenetic regulation, oncogenes may be upregulated while tumor suppressor genes may be downregulated, thereby contributing to cancer onset. Finally, glioblastoma is particularly frightening due to its effective suppression of the immune system. According to the theory of immunoediting, the immune system plays an active role in eliminating cancer (10). During a type 1 immune response, various elements of the innate immune system, including macrophages (Mφs), dendritic cells (DCs), and natural killer (NK) cells, become activated via pro-inflammatory cytokines (such as IL-1β, IL-12, TNF-α, and IFN-γ) and proceed to attack and destroy cancerous cells. Components of the adaptive immune system, including cytotoxic T lymphocytes (CTLs) and type 1 helper T (Th1) cells, may also be mobilized for cancer elimination (10). Glioblastoma harnesses several immunosuppressive cytokines (such as IL-10 and TGF-β) in order to create an environment characterized by a type 2 immune response, which is involved in the resolution of inflammation (10, 11). Herein, NK cells, DCs, and CTLs are inactivated and destroyed, CD4+ cells may adopt the anti-inflammatory Th2 subtype instead of the pro-inflammatory Th1 subtype, Mφs also adopt an anti-inflammatory M2 phenotype instead of the pro-inflammatory M1 phenotype, and regulatory T cells (Tregs) and myeloid derived suppressor cells (MDSCs) are recruited to further enforce immunosuppression (10, 11). Thus, glioblastoma can cleverly manipulate its microenvironment in order to evade destruction by immune cells. These immunological phenomena are summarized in figure 4.
 * Emerging hallmarks**




 * Figure 4.** Summary of cancer immunology (10). (A) A type 1 immune response is triggered during the elimination phase to destroy cancer. (B) Following an equilibrium phase (not shown), a type 2 immune response is elicited by the tumor during the escape phase in order to facilitate cancer proliferation and immune evasion (via immunosuppression).

Glioblastoma is an incredibly resilient cancer that is difficult to treat because it is invasive and quickly develops resistance. This is further complicated by the fact that glioblastoma is shielded by the blood-brain barrier and that there are notable risks of damaging the brain using conventional treatment methods. The current standard of care consists of surgical resection followed by a combination of radiotherapy (XRT) and chemotherapy. The most common chemotherapeutic used is temozolomide (TMZ), a DNA alkylating agent that helps promote cell death. Unfortunately, most glioblastomas can become resistant to TMZ via the overexpression of MGMT, an enzyme involved in repairing DNA damage associated with alkylation (13). As a result, the combined XRT and TMZ treatment raises the average survival of a patient by approximately 3 months (whereas untreated patients survive approximately 1 year) (12). Given the poor improvements in overall survival using traditional therapeutic means, new targeted therapies and immunotherapies aimed at the various molecular markers of glioblastoma are being investigated.
 * Treatment**

As discussed in part 2, glioblastoma involves the amplification or mutation of several genes that are responsible for the hallmarks of cancer. Several monoclonal antibodies and small molecule drugs have been developed that target proteins involved in the signaling pathways associated with glioblastoma with the intent of disrupting these pathways. Recall that VEGFR may be amplified in glioblastoma, resulting in heightened angiogenesis. Consequently, bevacizumab (a monoclonal antibody) and cediranib (a small molecule drug), which bind VEGF and VEGFR, respectively, were tested as treatments. By binding either component with a drug, it is possible to interfere with VEGF/VEGFR interactions, thereby slowing angiogenic signaling. So, both drugs ideally lower the rate at which new blood vessels are produced and thus decrease the nutrient supply to the tumor. While bevacizumab and cediranib seemed promising from results in phase II clinical trials, they were shown in phase III trials to be ineffective for improving overall survival in newly diagnosed and recurrent glioblastoma, respectively, when used as a monotherapy or in combination with other therapeutic methods (12). There are numerous other targeted therapies that are undergoing testing and will not be discussed. However, as of today, there does not exist a targeted therapeutic that significantly improves patient outcomes beyond the standard of care.
 * Targeted drug therapy**

While targeted therapies may not yet be effective, the principle of using molecular markers that are characteristic of glioblastoma as targets for therapy is still promising. Immunotherapy uses such molecular markers as targets for immune cells, which proceed to attack and destroy cancer cells. Note that this requires that the molecular target be on the surface of a cell, so that an immune cell may recognize it (this theoretically lowers the number of potential molecular targets compared to targeted drug therapies). Recall that EGFRvIII is a common EGFR variant encountered in glioblastoma that is responsible for ligand-independent proliferative signaling. It is possible to target EGFRvIII in several ways in the context of immunotherapy. Vaccine based methods utilize peptides or antigen-presenting cells (APCs) in order to stimulate the immune system to eliminate cancer cells. Herein, a peptide corresponding to a particular target of interest is incorporated into a vaccine formulation and injected into a patient. In an analogous manner to a standard vaccine, the immune system recognizes the foreign antigen and mounts an immune response against the cancer cells bearing the vaccine antigen (11). Alternatively, it is possible to culture macrophages or dendritic cells (both of which are APCs) in the presence of the tumor antigen as well as a classical foreign antigen and various cytokines in order to develop APCs targeted against cancer cells. These primed APCs can be injected into a patient in order to achieve an anti-cancer immune response much like the peptide based approach (14). Cell based methods utilize T cells or their variants in order to attack cancer cells. Herein, T cells harvested from a patient are expanded in culture and then targeted against a particular antigen either via antigen presentation or genetic engineering. In this process naïve T cells are converted into cytotoxic T lymphocytes (CTLs), which can bind cells bearing the antigen of interest and cause them to undergo apoptosis. These CTLs may then be injected back into the patient, along with a suite of activating cytokines, in order to destroy cancer cells (14). Alternatively, a synthetic T cell bearing a chimeric antigen receptor (CAR) instead of a T cell receptor (TCR) may be used. A CAR contains the antigen binding domain (scFv) of an antibody bound to the ζ chain of a TCR via linker and co-stimulation domains. CARs can be genetically engineered into T cells, wherein the scFv can be customized to target tumor antigens. CAR T cells may be more potent than activated CTLs for immunotherapeutic purposes (14). Both vaccine and cell based methods show great potential for the treatment of glioblastoma. For instance, a peptide based vaccine against EGFRvIII known as rindopepimut was shown to increase median survival time of glioblastoma patients to approximately 23 months in a phase I clinical trial (15). In addition, a CAR T cell therapy targeted against EGFRvIII was shown to be effective in a mouse model at eliminating glioblastoma and is now beginning a phase I clinical trial (16). There are many other glioblastoma antigens that can be used for immunotherapeutic intervention that are under investigation and will not be discussed.
 * Immunotherapy**

Both targeted therapy and immunotherapy present exciting opportunities for the treatment of glioblastoma and other cancers. Both types of therapies attempt to minimize side effects by targeting molecular markers that are unique to or predominantly expressed by the cancer. At this point, it is difficult to conclude which of these next generation techniques will be better given that they are still undergoing clinical trials and further refinement. Perhaps both therapies may be combined, given that targeted therapies can interfere with cytoplasmic cancer biomarkers while immunotherapies can destroy cancer cells based on their cell membrane antigens. This type of pincer attack, wherein a cancer is targeted from the outside and inside, would ideally overwhelm the cancer and prevent it from developing resistance.
 * Comparison of experimental treatments**

During a consultation with the patient, surgical removal of the tumor mass followed by treatment with XRT and TMZ was recommended for managing the symptoms of glioblastoma. The patient consented to and received the standard of care. After further discussion, the patient requested to receive any experimental treatments which were available. Consequently, the patient was enrolled into a clinical trial for CAR T cell therapy targeting EGFRvIII. Remarkably, this immunotherapy has been effective in its elimination of cancer cells. The patient will continue to be monitored, given that glioblastoma recurrence is still possible. After all, glioblastoma may eventually adapt to the immunotherapy, perhaps by withdrawing EGFRvIII from its cell surface or by suppressing the activity of the T cells.
 * Course of treatment for the patient**

Glioblastoma is an aggressive, invasive brain cancer with no effective means of treatment. It is possible to remove tumors with surgery, burn them with radiation, and poison them with a wide variety of chemotherapeutics (temozolomide, carboplatin, lomustine, etc) but they will often grow back with resistance (13). New types of targeted drug therapy and immunotherapy are currently under development. Currently, however, the prognosis for glioblastoma is grim, with average survival ranging from 12-15 months with the standard of care (13). Truly, glioblastoma is a deadly, relentless cancer.
 * Conclusion**

In the present, glioblastoma is tantamount to a death sentence. And so, to the poor beings afflicted with this cancer, I suggest that you get your affairs in order and perhaps consider an experimental treatment.
 * Aperçu**


 * References**
 * 1) Barres, B. (2008). The Mystery and Magic of Glia: A Perspective on Their Roles in Health and Disease. Neuron, 60(3), 430-440. http://www.cell.com/neuron/fulltext/S0896-6273%2808%2900886-6
 * 2) Agnihotri, S., Burrell, K., Wolf, A., Jalali, S., Hawkins, C., Rutka, J., and Zadeh, G. (2012). Glioblastoma, a Brief Review of History, Molecular Genetics, Animal Models and Novel Therapeutic Strategies. Arch. Immunol. Ther. Exp., 61(1), 25-41. http://link.springer.com/article/10.1007%2Fs00005-012-0203-0
 * 3) Holland, E. C. (2000). Glioblastoma multiforme: The terminator. PNAS, 97(12), 6242-6244. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC33993/
 * 4) Sturm, D., Bender, S., Jones, D., Lichter, P., Grill, J., and Becher, O. et al. (2014). Paediatric and adult glioblastoma: multiform (epi)genomic culprits emerge. Nature Reviews Cancer, 14(2), 92-107. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4003223/
 * 5) Omuro, A. (2013). Glioblastoma and Other Malignant Gliomas. JAMA, 310(17), 1842. http://jama.jamanetwork.com/article.aspx?articleid=1764056
 * 6) Thomas, A. A., Ernstoff, M. S., and Fadul, C. E. (2012). Immunotherapy for the Treatment of Glioblastoma. Cancer Journal (Sudbury, Mass.), 18(1), 59-68. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3269657/
 * 7) Wang, H., Xu, T., Jiang, Y., Xu, H., Yan, Y., Fu, D., & Chen, J. (2015). The Challenges and the Promise of Molecular Targeted Therapy in Malignant Gliomas. Neoplasia (New York, N.Y.), 17(3), 239-255. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4372648/
 * 8) Hanahan, D. and Weinberg, R. (2011). Hallmarks of Cancer: The Next Generation. Cell 144, 646-674. http://www.cell.com/cell/fulltext/S0092-8674%2811%2900127-9
 * 9) Ware M., Berger, M., and Binder, D. (2003). Molecular biology of glioma tumorigenesis. Histology and Histopathology 18(1), 207-16. http://www.hh.um.es/pdf/Vol_18/18_1/Ware-18-207-216-2003.pdf
 * 10) Mittal, D., Gubin, M., Schreiber, R., and Smyth, M. (2014). New insights into cancer immunoediting and its three component phases—elimination, equilibrium and escape. Current Opinion In Immunology 27, 16-25. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4388310/
 * 11) Ooi, Y., Tran, P., Ung, N., Thill, K., Trang, A., Fong, B., Nagasawa, D., Lim, M., and Yang, I. (2014). The role of regulatory T-cells in glioma immunology. Clinical Neurology And Neurosurgery 119, 125-132. http://www.clineu-journal.com/article/S0303-8467%2813%2900508-8/fulltext
 * 12) Weathers, S. and Gilbert, M. (2014). Advances in treating glioblastoma. F1000prime Rep 6.http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4047946/
 * 13) Veliz, I., Loo, Y., Castillo, O., Karachaliou, N., Nigro, O., and Rosell, R. (2014). Advances and challenges in the molecular biology and treatment of glioblastoma—is there any hope for the future? Annals Of Translational Medicine 3. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4293478/
 * 14) Bovenberg, M., Degeling, M., and Tannous, B. (2013). Cell-based Immunotherapy Against Gliomas: From Bench to Bedside. Mol Ther 21, 1297-1305. http://www.nature.com/mt/journal/v21/n7/full/mt201380a.html
 * 15) Swartz, A., Li, Q., and Sampson, J. (2014). Rindopepimut: a promising immunotherapeutic for the treatment of glioblastoma multiforme. Immunotherapy 6, 679-690. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4524671/
 * 16) Johnson, L., Scholler, J., Ohkuri, T., Kosaka, A., Patel, P., McGettigan, S., Nace, A., Dentchev, T., Thekkat, P., and Loew, A. et al. (2015). Rational development and characterization of humanized anti-EGFR variant III chimeric antigen receptor T cells for glioblastoma. Science Translational Medicine 7, 275ra22-275ra22. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4467166/