Funded Grants

Glioblastoma and the Brain's Immune System

Glioblastoma is the most malignant type of brain cancer known. It can affect people at any age but is most commonly found in adults with a peak incidence between 45 and 70 years of age. Glioblastomas tend to be more frequent in males than in females, and a significant portion of the patients are children (as much as 9%). The prognosis for patients diagnosed with glioblastoma is usually poor with the 5-year survival being less than 5%. What makes glioblastoma a particularly insidious disease is the fact that it spreads rapidly within the central nervous system. From the main tumor mass individual tumor cells can migrate away and invade aggressively neighboring normal brain structures. Surgical resection of the tumor almost never cures the disease because small microscopic cohorts of tumor cells which have already migrated away from the main tumor mass are not removed by the surgery. It is these remaining tumor cells that will give rise to recurrent tumors in a relatively short period of time. In most cases, the initial surgical treatment of glioblastoma is followed by radiotherapy or chemotherapy, and these additional treatments help in prolonging patient survival beyond the increased survival achieved by surgery alone. However, a discouraging bottom line remains, despite aggressive treatment most patients die of the disease, with a median survival of about one year.

The focus of Dr. Streit's research is to develop a new type of therapy for the treatment of glioblastoma which takes advantage of recent advances made in the field of neuroimmunology. Neuroimmunology combines neuroscience with immunology and most of the research in this area is focused on how immunological processes contribute towards the development of brain disease. The classic example of a `neuroimmunological' disease is multiple sclerosis where the immune system becomes autoaggressive and destroys portions of the central nervous system. Normally, the brain and spinal cord are protected from being attacked by cells of the immune system because the leukocytes (white cells), which are abundant in the blood, are prevented from entering the central nervous system. However, is the exclusion of white cells not dangerous since it leaves the brain unprotected from invading microorganisms? As it turns out, the brain is not at all unprotected because it has its own immune system [Streit and Colton, The brain's immune system, Scientific American, November 1995]. The key component of the brain's immune system are so-called microglial cells. For conceptual purposes, one can think of microglia as crossbreeds between blood-borne white cells and glial cells of the central nervous system (glial cells are cells that support neurons). In other words, microglia share some of the same characteristics as white cells, but they have evolved to live with these potentially dangerous properties inside the brain without causing harm. They are specially adapted immune-competent cells of the brain. One of the properties that microglia share with white cells is their ability to become cytotoxic effector cells. This means that microglia can acquire the ability to kill other cells, including normal brain cells but also brain tumor cells. The ultimate goal of Dr. Streit's research is to learn how to manipulate the cytotoxic ability of microglia and to use it for selectively destroying tumors cells. Perhaps the greatest potential advantage that comes with this strategy is that fact that microglia, which are quite numerous within brain tumors, could be used as microscopical weapons to eliminate those small remaining contingents of brain tumor cells that remain after surgical resection and are responsible for tumor recurrences. Following is a brief summary of microglia in brain tumors and on the mechanisms thought to play a role in controlling interactions between microglial and tumor cells.

It has been observed repeatedly that malignant gliomas, both spontaneously occurring ones in humans and experimentally induced ones in rodents, contain large numbers of microglial cells. Many of the microglia observed appear to be activated, suggesting that they are in a functional cytotoxic state that enables them to kill tumor cells. However, tumor cytotoxicity does not seem to occur to any significant extent since glioblastomas continue to thrive and expand eventually killing their victims. Thus, questions have arisen as to how exactly microglia kill tumor cells and why they appear to be unable to do so in the tumor environment. Part of the answers to these questions comes from research involving analysis of tumor and microglia-associated cytokines, small peptides important for mediating cell-cell interactions. Cell culture studies have shown that microglia can kill tumor cells by producing a particular cytokine called tumor necrosis factor-alpha (TNF-a). However, when TNF-a content was analyzed in tissue samples of rat brain tumors, it was found that TNF-a was not being produced in these samples clearly supporting the idea that microglia inside brain tumors are unable to produce the tumorkilling TNF-a. In addition, it turns out that glioblastomas produce large amounts of another cytokine, called transforming growth factor beta (TGF-Beta), which has potent immunosuppressive effects. It has been shown that TGF-Beta, when added to cultured microglial cells, can inhibit their ability to produce TNF-a and to kill glioma cells. The picture that emerges from all of this is that glioma cells can survive immune attack by producing immunosuppressive cytokines, such as TGF-Beta. This relationship between microglia and glioma cells is illustrated schematically in the figure below.


The next question in this line of work is, can the immunosuppressive action of tumorderived (TGF-Beta), be overcome by direct immune stimulation of microglial cells inside of brain tumors. In other words, can direct intra-tumoral injection of immune stimulatory agents induce microglial cytotoxicity. This question is precisely the focus of the proposed research. Dr. Streit's laboratory has obtained promising, albeit preliminary, results using injections of bacterial endotoxin and gamma interferon into rat brain tumors. The initial results have shown that these injections result in greatly increased areas of tumor cell death. Since it is known that endotoxin and interferon can stimulate microglia to become cytotoxic for tumor cells in cell culture, these preliminary findings strongly suggest immune stimulation applied directly into a brain tumor has the same effect. The research plan will continue to test and refine the immunostimulatory protocol in rats with glioblastoma, and to follow both treated and untreated animals long term using high resolution magnetic resonance imaging. The MRI will allow us to determine whether shrinkage of gliomas, and perhaps even eradication, can be achieved using repeat injections of endotoxin and interferon. The proposed work will employ a rat glioma model which has been used in the Streit laboratory since 1992. This model has been characterized extensively and has been found to mimic many of the same features of human glioblastomas, including a high density of microglia and high levels of TGF-P production. One can therefore be reasonably confident that if a successful treatment protocol can be developed for the rat brain tumors, it might be directly applicable for treatment of human patients with glioblastoma and could lead to clinical trials. The proposed immunotherapy could become a follow?up treatment to the initial surgical resection and it could result in the successful removal of microscopic foci of leftover tumor cells which are impossible to remove with current regimens. Elimination of all tumor cells may prevent recurrent tumor formation.