The Deadliest Brain Cancer
Regina Kelder

The Deadliest Brain Cancer

Among tumors, glioblastomas are among the toughest to treat. But do they have to be a dead end? A discussion with Charles River Immunologist David Harris

Recent cases of brain cancer among US public figures illustrate how far we have come and how far we need to go in arresting this rare and deadly disease.

Some cancers that metastasize to the brain can be treated aggressively. US President Jimmy Carter, whose malignant melanoma spread to his brain, successfully treated his tumors with surgery, radiation and a recently-approved drug that harnessed his immune system to fight the cancer. Carter, 92, shared the news that he was in remission at one of his regular Sunday School classes in Georgia.

But US Senator John McCain, diagnosed two weeks ago with the rare brain cancer glioblastoma, is facing much stiffer odds. Glioblastomas originate in the brain, grow aggressively, and in ways that make them extremely difficult to remove surgically or resect. The median survival following treatment is about 14 ½ months. Glioblastoma, which also claimed the life of McCain’s Senate colleague Ted Kennedy in 2009, kills an astonishing 95% of patients within five years of diagnosis. And because of the complexity of the central nervous system—most notably the brain—finding game-changing treatments that might provide a different outcome are extremely challenging.

Nonetheless, there is work being done on the preclinical side to refine the models used to study the disease and assess the effectiveness of new drugs and some promising drugs have advanced to clinical testing. There are also new and increasingly more sophisticated ways of imaging tumors that make it easier to visualize what is going on inside the brain of an animal implanted with glioblastoma tumor cells.

To hear more about these new advances in research, Eureka spoke recently with David Harris, a Charles River immunologist and Research Director based in Morrisville, North Carolina, the hub for CRL’s drug discovery oncology services that also include several other sites in the US and Europe. With so many life-changing drugs now on the market or emerging for the treatment of different cancers, Dr. Harris talked about how high the hurdle is to duplicate some of those optimistic findings in glioblastoma patients, but why we shouldn’t give up hope.

Here is an edited transcript of our discussion.

Glioblastoma grows rapidly. Why else is the tumor deadlier than other brain cancers?

DH: Well, they are very complex tumors. A complete surgical removal of the entire tumor is extremely difficult because these tumors are highly invasive, having finger-like tentacles that extend into surrounding normal brain tissue. Glioblastomas are also highly vascularized and made up of a heterogeneous mix of different cell types that often differ from patient to patients. Aside from that, the tumors have the potential to change, mutating and evolving so when you perform surgery to remove the tumor, it is almost impossible to get rid of all of it. Additionally, when glioblastomas recur or grow back, which almost always happens, their molecular profiles change dramatically. As a result, responses to treatment will vary a lot.

What about immunotherapies? Are they effective against glioblastoma?

DH: The standard of care is still radiation therapy and chemotherapy. Having said that, there are some instances where there have been some good results using immunotherapy, though the results are highly experimental. Among the experimental therapeutic immunotherapies are several therapeutic vaccines composed of a peptide, representing a mutated form of a growth factor receptor that targets the tumor cell. Researchers are also trying other immunotherapy approaches, such as immune checkpoint inhibitors, modified T cells, cord blood–derived natural killer (NK) cells, and STAT3 (signal transducer and activator of transcription) inhibitors that impact tumor growth, to attack glioblastomas.

So why haven’t we seen any immunotherapies approved for glioblastoma?

DH: When it comes to the brain, immune-based treatments face some substantial obstacles before they can even reach a tumor. One of the most significant challenge is the blood–brain barrier, which protects the brain from threats that may be circulating in the bloodstream, like viruses or toxins. At the same time it can also interfere with the delivery of cancer drugs. Another reason is that the immune response to brain tumors is generally weak and tumors tend to have a low mutational load. So even if you manage to reach the tumor, the immunotherapy may not have much of an impact. In addition, a recent review article on immunotherapy approaches in glioblastoma noted that tumor microenvironment is rich with immunosuppressive factors secreted by the tumor, like transforming growth factor beta (TGF-β) and vascular endothelial growth factor (VEGF) that suppresses T cell proliferation and cytotoxic function.

A recent study in the Journal of Immunology Research also points out how hard it is identifying suitable, immunogenic tumor antigens, and appropriate pre- and post-therapeutic biomarkers. The study also cited the limited number of glioblastoma patients eligible to join particular clinical studies, and difficulties in fully understanding various regulatory and stimulatory factors in the immune system and the glioblastoma microenvironment.

Can you describe some of the current animal models that Charles River uses to study glioblastoma?

DH: We use a xenograft model called U87MG which is a human tumor cell line that was derived from a malignant glioma taken from a female patient by explant technique. It grows in immunodeficient animals when implanted subcutaneously or following intracranial tumor implant. The orthotopic injection of tumor cells mimics clinical responses to tumor growth and the growth of these tumors has been characterized in response to various chemotherapeutics agents. The model is also available for bioimaging studies using luciferase transduced tumor cells. Another model we use, GL261, is a mouse glioblastoma which is implanted in C57BL/6 mice. This syngeneic model is also being developed for bioimaging. Aside from these, we also have a number of highly characterized glioblastoma PDX (patient-derived xenograft) models.

How challenging is it working with these models?

DH: Although important insights into mechanisms of tumor development have resulted from preclinical studies, none of the current animal models fully reflects the phenotypic and genotypic changes seen in human glioma development and progression. A recent study in the journal Neuro-Oncology suggests it is due to the fact that tumor models do not reflect the biological properties of the patient tumors, that the animals used do not display the same pharmacokinetics as humans and that the tumors established do not reflect the cellular heterogeneity of human tumors. The unique anatomy and physiology of the brain is also a factor. The authors of the study consider the best models to be ones that are advanced orthotopic models developed in conjunction with sophisticated in vivo imaging techniques.

What imaging technologies are being used to study these tumors?

DH: Bioluminescence is one method that is being used more and more. By making the tumor cells bioluminescent you can see how the tumor grows in vivo without killing the animal. This is a non-invasive way of studying tumor growth in the animal longitudinally over time, and provides the opportunity for ex vivo imaging of organs and metastatic model development. The imaging studies are generally done with either an orthotopic implantation –implanting tumors into the organ—in this case the brain—where the tumor was derived or for disseminated hematological tumor models, intraveneous implantation of tumor cells. Other imaging strategies include optical and microCT, which allow for 3D optical tomography. Very cool.

Are there next-generation tools that might offer a bigger winder into how these tumors develop and progress?

DH: Whole-exome sequencing of the tumor cells and molecular characterization of the T cell responses for these models can give us a detailed phenotypic and genetic analysis of immune cells in tumor tissue. Bioinformatics can also help us understand how these tumors behave at the genetic level.