Program Co-Leaders Mitchel Berger, MD Michael Prados, MD

The Neurologic Oncology Program comprises 30 faculty members who work to support and stimulate basic, clinical, and population research in brain cancer, and to facilitate translation of these findings into improved cancer management and control. Most investigators are also members of the UCSF Department of Neurological Surgery's Brain Tumor Research Center, which has been in existence for more than 27 years. Programmatic research is focused on five strategic themes, as described below.

Program research is supported additionally by a Brain Tumor Specialized Program Of Research Excellence (SPORE) grant, awarded in 2002 by the National Cancer Institute.

Creation of models to better assess the cause and treatment of brain tumors
A fundamental problem in the treatment of brain tumors is our lack of understanding of the underlying causes of the disease. Considerable insight has been gained by the careful examination of genetic alterations that accompany the development of tumors. This type of analysis, however, is correlative in nature, and cannot definitively pinpoint pathways that might serve as therapeutic targets. To address this limitation, program investigators have developed systems by which genetic alterations can be introduced into normal cells, after which the ability of these lesions to form brain tumors can be assessed. In one model, normal human astrocytes can be retrovirally infected with various cDNAs, after which the ability of the cells to form intracranial gliomas in rats can be assessed. A complementary model accurately reflects the formation of spontaneous brain tumors that strongly resemble human oligodendroglioma. These models provide insight into the causes of astrocytomas and oligodendrogliomas, and provide a means by which the treatment of gliomas can be tested.

Development of methodologies to better predict the outcome of patients with brain tumors
Brain tumors are a heterogeneous group of malignancies, ranging from benign meningiomas to the nearly universally fatal glioblastoma multiforme (GBM). Even within what histologically appears to be similar tumor classifications, patient survival varies greatly. Thus a fundamental problem is in classifying tumors and predicting how patients will respond to therapeutic options. Program investigators are applying statistical, molecular, and epidemiological methods to address the question of how mutations and polymorphisms in various genes might affect brain tumor susceptibility, and how these alterations may also contribute to survival. Their work has shown that overexpression of Epidermal Growth Factor Receptor (EGFR) in the absence of p53 mutations predicted a poorer survival among young GBM patients, and that GBM in non-whites, females, and young patients were more likely to have p53 mutations than GBM in older people, whites, and males. Other program members are mapping genetic aberrations and changes in gene expression in human astrocytic tumors, toward identifying genes that modify response to therapy and clinical outcome, and providing tailored, tumor-specific therapeutic targets.

Focus on pre-clinical studies that will enable development of new and better therapies for patients
Promising program initiatives in this area include:

• immunoliposomes designed to bind and internalize in EGFR-overexpressing tumor cells, or to mutant EGFR-expressing tumor cells, and thereby deliver toxic molecules encapsulated within the liposome;

• studies that address the problem of hypoxia, which occurs when brain tumors grow beyond their blood supply. Hypoxic areas are much less sensitive to killing by radiation, yet upon therapy-induced tumor shrinkage can become re-oxygenated and contribute to tumor regrowth and progression;

• therapeutic strategies based on the PKB pathway, which promotes growth and survival of cells, and is in turn controlled by the tumor-suppressor protein PTEN;

• using DNA array technology to better define the molecular phenotype of childhood brain tumors such as brain stem gliomas and ependymomas; and

• investigating normal brain tissue toxicity -- because any new treatment has the potential to induce serious injury to normal tissues surrounding the tumor -- as well as utilizing stem cells as a source of potentially mobile, undifferentiated cells that can be mobilized after injury.

Finally, researchers seek improvements in existing therapies for brain tumors, which combine the use of surgical resection, radiation, and alkylating agent-based chemotherapy. For example, to help surgeons better distinguish between tumor and non-tumor tissue in real time, investigators employ spectroscopic monitoring of cellular metabolites, which provides more detailed imaging than conventional MRI scanning. Additionally, because drug resistance is a common problem in chemotherapy, one initiative is investigating the mechanisms of action and resistance to temozolomide.

Conduct clinical trials based on translational research to improve selective tumor targeting and enhance local/regional tumor control
The clinical program seeks to continue its research into more selective targeting of tumors within the brain and improve local/regional control of disease. Clinical trials that have started or are soon to open target the Epidermal Growth Factor Receptor (EGFR), Platelet Derived Growth Factor Receptor (PDGFR), IL-13, TGF-alpha, and mTOR signaling pathways. This clinical research supports ongoing molecular and cell biology research in receptor-mediated and cell cycle regulation of brain tumors. Specific agents include the use of STI-571, ZD1839, OSI-774, and CCI-771, and the immunotoxins IL-13/Pseudomonas Exotoxin (PE) and TGF-alpha/PE. Testing has also begun on OSI-774, a small molecule inhibitor of EGFR, in combination with a commonly used methlyating agent, Temodar. Other studies are looking at IL-13/PE and TGF-alpha/PE, given over a 96-hour infusion directly into tumor and brain surrounding tumor, to selectively target tumor cells using convection-enhanced delivery techniques.

Expand efforts on population-based research
Large ongoing studies are investigating the genetic epidemiology of cancer, the role of the immune system in brain tumor etiology, and risk factors for childhood brain tumors. The San Francisco Bay Area Adult Glioma Study, in existence since 1991, has accrued approximately 900 brain tumor cases and an equivalent number of controls. An upcoming series of the study will accrue another 700 cases and 500 controls. Results to date have provided clues to the causes of adult glioma, the most common form of primary malignant brain tumor. Immune-system work builds on the observations that brain tumor patients have far lower frequency of seropositivity to Varicella Zoster Virus, and that they also have a far lower reported frequency of common allergies than do controls. Researchers are using descriptive and functional biomarkers of immune-cell function to explore the mechanistic nature of these associations. Research on childhood brain tumors is looking at maternal exposure to N-nitroso compounds during pregnancy and the child's exposures prior to diagnosis.