MRI of a patient with lymphoma of the brain before (left) and after (right) chemotherapy.

MRI of a patient with lymphoma of the brain before (left) and after (right) chemotherapy.

More than 20,000 people in the United States will be diagnosed with cancers arising in brain or spinal cord this year, according to the American Cancer Society. Cancer is the second leading cause of death in children in the United States, and brain tumors are second only to leukemia as a cause of cancer in children. Brain cancer poses a unique challenge to cancer research and neuroscience, and its study demands a unique research environment, one that recognizes the special nature of the central nervous system and the tumors that develop there.

Brain tumor research is a relatively modern science that parallels the explosion of molecular biology and modern neuroimaging. While the science is new, Mayo Clinic has a long history of brain cancer treatment and research stretching back over 100 years.

Mayo Clinic ranks as one of the largest brain tumor centers in the world. Patients are seen at all three Mayo Clinic campuses in Scottsdale/Phoenix, Arizona; Jacksonville, Florida; and Rochester, Minnesota. Many patients under treatment also participate in clinical trials and associated research, including studies of tumor biology, neuropathology, and epidemiology.

Central to the department's mission is the discovery of novel therapies for central nervous system (CNS) tumors and metastases through clinical and laboratory research. In the laboratory researchers test new therapeutic agents and new approaches to gene therapy to target and kill cancer cells, or to disrupt the specific signal transduction pathways that control tumor growth. Mayo Clinic neuro-oncologists are developing novel combination strategies to augment traditional treatment approaches for chemotherapy, surgery, and radiation therapy.

Another important goal of the program is translational research so that new agents and strategies developed in the laboratories are brought quickly to the clinical setting. Translating laboratory findings into clinical studies can offer new, more effective therapies for CNS tumors. Improvements in survival rates and quality of life for brain tumor patients will depend on extending the understanding of the genetic abnormalities and molecular events responsible for the development, progression, and spread of brain tumors, in order to develop new and more effective treatment options.

Ongoing research

Oligodendrogliomas are a unique form of brain tumor that affect young adults. The initial symptoms typically are seizures. Oligodendrogliomas are occasionally able to be surgically removed because of limited invasiveness, and they frequently respond to initial treatments. Robert Jenkins, M.D., Ph.D., co-leader of the Neuro-Oncology Program, Mayo Clinic Cancer Center Associate Director, and principal investigator of the "Glioma Markers" P01 grant, a research program funded by the National Cancer Institute (NCI), contributed to the seminal discovery that deletions of regions in specific chromosomes (1p and 19q) are associated with oligodendrogliomas. Specifically, these deletions seem to predict prolonged survival and better sensitivity to chemotherapy in certain patients with those deletions. Dr. Jenkins, in collaboration with colleagues at Massachusetts General Hospital Cancer Center, has completed a detailed map of important chromosomal regions (19q13.3) in gliomas. These data form much of the basis of the research supported by Dr. Jenkins' NCI grant. Based on work done at Mayo, there is sufficient reason to believe that most of the tumor's clinical characteristics are biologically driven by genetic characteristics of the tumor.

Dr. Jenkins' research team also identified other relatively small chromosomal deletions (19q13.3) in two glioma cell lines (U87 and A172). These deletions completely encompass the commonly deleted region that the researchers mapped in patient gliomas, and therefore represent a potential resource to study these observations in the laboratory. In addition, the research team is attempting to develop oligodendroglioma human tumor cell lines. Such cell lines will facilitate the continued genetic studies. The identification of the 19q tumor suppressor gene and the evaluation of its function should lead to new diagnostic and new therapeutic approaches for treatment of glioma patients.

Haplotypes containing the GLTSCR1-exon-1 T allele are significantly associated with increased oligodendroglioma risk. The haplotype containing the GLTSCR1-exon-1 C allele are significantly inversely associated with glioma risk. Ping Yang, M.D., Ph.D., and Dr. Jenkins are currently validating these results using independent sets of Mayo Clinic gliomas and oligodendrogliomas in Radiation Therapy Oncology Group trial 9402. This work is also a major component of the GLIOGENE consortium.

Research advances

Primary central nervous system lymphoma (PCNSL) is a unique lymphoma and is invariably lethal without treatment. Current treatments are extending survival but are not increasing the number of cured patients; the quality of life of survivors is poor. Variable tumor characteristics may reflect differing stimuli from within the tumor itself and/or differing responses by surrounding normal tissues, particularly specific cellular functions that control the normal B cell growth. A multidisciplinary team of individuals led by Brian O'Neill, M.D., has been working together for more than a decade on this tumor. Dr. O'Neill's group has systematically contributed to the knowledge base of PCNSL by sequentially studying large numbers of patients with regard to their disease, treatment, and outcomes with the intent of identifying risk factors and studying better approaches to management and treatment of this disease. From these studies has come emerging basic and translational PCNSL research focused on the identification of an infectious agent(s) that might stimulate the tumor-producing cascade and abnormal immune cell (B cell cellular functions). Preliminary data suggest that a specific virus is involved in at least some of the cases. This work has continued within the Iowa/Mayo Lymphoma SPORE grant. Other outgrowths of this ongoing research include the development, launch, conduct, completion, analysis, and publication of a number of investigator-initiated clinical trials sponsored by the North Central Cancer Treatment Group.

The research of Panos Anastasiadis, Ph.D., and Joseph Loftus, Ph.D., demonstrates the program's Mayo Clinic system-wide functionality. Although each laboratory is investigating mediators of invasion, their work focuses on different mechanisms and is complementary. The research teams exchange data and have emerging collaborations between themselves and other Mayo Clinic colleagues. A unique focus of the Anastasiadis laboratory is the differential expression of invasion mediators including key components of the cadherin-catenin complex in pediatric age group astrocytomas compared to their adult counterparts. (Loftus: Biochemical and Biophysical Research Communications. 2006; 349(3): 939-947; Anastasiadis: Journal of Cell Biology. 2006; 174: 1087-96).

Joon Uhm, M.D., and colleagues have accrued compelling evidence that galectin-1 is an important mediator of glioma invasion. This research extensively utilized the Brain SPORE Xenograft Core. RNA was extracted by way of laser capture microdissection from the periphery (invasive edge) and core of the tumor, and the RNA was analyzed for differential gene expression by Affymetrix microarray analysis. Galectin-1 was identified as one of the genes overexpressed uniquely at the invasive margin of human glioblastoma tumor. Overexpression of galectin-1 resulted in increased invasiveness in vitro. Moreover, these galectin-1 overexpressing cells, when implanted intracranially, lead to shortened survival compared to their wild-type counterparts.

Mayo Clinic neuro-oncology research has a broad geographic reach, with multidisciplinary teams of basic, clinical, and population science investigators at Mayo's three campuses in Scottsdale/Phoenix, Arizona; Jacksonville, Florida; and Rochester, Minnesota. Mayo physicians and researchers form powerful collaborations across programs, organ areas, and specialties such as neurology, surgery, and radiology to multiply the potential for breakthroughs in research and treatment. A brief overview of selected advances is listed below.

Astrocytomas are the most common glial tumor. These tumors affect patients of all ages, especially young children and older adults. They may be more likely to occur in patients with several genetic diseases that lead to high incidences of certain cancers. These tumors are only occasionally surgically removable because of their large size prior to diagnosis and because of their invasiveness. They appear to evolve from low-grade to high-grade forms by gradually acquiring molecular changes in the tumors, similar to the course of many colon cancers. Often, their pattern of failure is such that they acquire a number of these changes and "dedifferentiate" to become much more malignant, high-grade astrocytoma or glioblastoma multiforme, the most common and most lethal form of primary brain tumor of adults. Mayo researchers determined that an important molecule called epidermal growth factor receptor (EGFR) appears in higher concentrations in astrocytomas, another of the three principal kinds of glioma and the most common. Researchers are looking for ways to inhibit the cellular receptors and the pathways of EGFR to impede malignant behavior of tumors. Approaches include small molecule anticancer agents, gene therapy, and radiation sensitization.

Medulloblastomas are the most common brain tumor of children. Unlike adult brain tumors, these tumors most often occur in the hind part of the brain and arise from cells of neuronal lineage rather than glial lineage. Because of their compact growth, they are potentially curable by surgery. However, despite the tendency of these tumors to present as compact masses, they often spread via the cerebrospinal fluid. This increases the challenge of their treatment, increases the risk of toxicity to the developing nervous system of these pediatric patients, and considerably reduces the chance of cure or prolonged disease-free survival. The basic biological and genetic features responsible for this behavior are at best poorly understood and were the basis for Mayo researchers becoming the first to demonstrate mutations in important gene "patched" (PTCH) mutations in medulloblastomas. Based on current understanding of the functions of PTCH and the targets of this protein in tumor cell growth, the investigators hypothesized that other genes involved in those PTCH-driven cellular functions may also be mutated in medulloblastomas. Research based on this hypothesis led to discovery of mutations in a gene called b-catenin in a subset of these tumors.

Mayo's SPORE-based panel of serially-transplantable glioblastoma multiforme xenografts established from patient tumors is now being used to evaluate novel therapeutic strategies including combinations of novel signal transduction inhibitors with radiation and temozolomide. In addition, in a collaborative effort between the Brain SPORE, Mayo's Neuro-Oncology and Gene and Virus Therapy Programs, and the Department of Molecular Medicine, Eva Galanis, M.D., recently activated a trial of intratumoral and resection cavity administration of a measles virus derivative producing CEA in patients with recurrent glioblastoma multiforme. This trial represents the first human application of measles vibrotherapy in the treatment of GBM. The whole translational process to include generation and efficacy testing of the engineered virus strain, vector production, and toxicology testing were performed through the Cancer Center. This completely 'in house' bench-to-bedside effort is just one of many innovative ideas being moved from the laboratory to clinical application through the Neuro-Oncology Program's translational research efforts.

Caterina Giannini, M.D., and neuropathology colleagues, in conjunction with the TACMA and the Biospecimens Accessioning and Processing Shared Resources, have completed a study of O6-methylguanine DNA methyltransferase (MGMT) promoter methylation status and expression in human glioblastoma multiforme. MGMT DNA repair gene methylation silencing has been associated with longer survival in patients with glioblastoma multiforme, especially receiving chemotherapy with alkylating agents. This study provided a cautionary note regarding the usefulness of MGMT analysis. Methylation-specific polymerase chain reaction (PCR) was successful in only 78 percent of specimens. MGMT promoter methylation was present in 15 cases (38.5 percent) and absent in 24 (61.5 percent). Immunostain scoring was difficult due to MGMT expression in non-neoplastic cells (endothelial cells, glia, microglia, macrophages, and lymphocytes). MGMT expression was present in tumor cells in seven cases (18 percent). In 70 percent of specimens, promoter methylation and IHC-defined MGMT expression were discordant. There was no significant correlation between MGMT methylation and expression nor significant survival difference observed in the methylated versus unmethylated groups (mean survival 487 and 453 days, respectively), or negative versus positive immunohistochemistry.

A team of Mayo Clinic Neuro-Oncology Program investigators led by Robert Jenkins, M.D., Ph.D., discovered that combined deletion of chromosomes 1p and 19q is associated with improved prognosis and responsiveness to therapy in patients with anaplastic oligodendroglioma, one of three principal kinds of glioma. The deletions usually involve whole chromosome arms, suggesting a t(1;19)(q10;p10). Using stem cell media the researchers cultured a small number of oligodendrogliomas; one was found to have an unbalanced 45,XX,t(1;19)(q10;p10). Interphase fusion of CEP1 and 19p12 probes were developed to detect the t(1;19). Paraffin-embedded tissue was obtained from 21 Mayo Clinic patients and 98 patients enrolled in two North Central Cancer Treatment Group (NCCTG) low-grade glioma trials.

Among Mayo Clinic oligodendrogliomas, the prevalence of fusion was 81 percent. Among NCCTG tumors CEP1/19p12 fusion prevalence was 55 percent, 47 percent, and 0 percent among the oligodendrogliomas, mixed oligoastrocytomas, and astrocytomas, respectively. Ninety-one percent of NCCTG gliomas with and 12 percent without 1p/19q deletion had CEP1/19p12 fusion (p<0.001, chi-square test). The results demonstrate that a t(1;19)(q10;p10) mediates the combined 1p/19q deletion in human gliomas.

Dr. Jenkins and his colleagues further clarified the prognostic significance of combined 1p and 19q deletion and the t(1;19) in the low-grade NCCTG gliomas. The t(1;19) was associated with prolonged survival compared with the absence of the deletions. The median overall survival (OS) for all patients was 8.1 years without and 11.9 years with translocation (p=0.003). The median OS for patients with low-grade oligodendroglioma was 9.1 years without and 13.0 years with translocation (p=0.01). Similar significant median OS differences were observed for patients with combined 1p/19q deletions. The absence of alterations was associated with a significantly shorter OS for patients who received higher doses of radiotherapy. Like combined 1p/19q deletion, the 1;19 translocation is associated with superior OS and progression-free survival in low-grade glioma patients.

The identification of the 19q tumor suppressor gene and the evaluation of its function should lead to new diagnostic and therapeutic approaches for patients who have gliomas. In conjunction with Dr. Giannini, M.D., and other colleagues in Neuropathology, Dr. Jenkins developed a test utilizing fluorescent probes that can specifically detect the chromosomal deletion discussed above. This test may eventually improve the ability to target specific therapies to individual patients. A secondary aim has been to utilize Mayo resources and bring forward novel therapies in "orphan" tumor states and then launch them through clinical trial "portals." One such effort is related to neurofibromatosis, type I. Since NF1 is strongly associated with CNS and PNS tumors, discovery in this "orphan" may impact more common sporadic tumors that are histologically similar, such as pilocytic astrocytoma.

Dusica Babovic-Vuksanovic, M.D., led a Phase I study of pirfenidone, an anti-fibrotic cytokine developed "in house," on NF-associated plexiform neurofibromas. Her Department of Defense grant (DAMD17-2-1-064-01) was given a "no-cost extension" to continue a Phase II Children's Oncology Group study of pirfenidone. Dr. Babovic-Vuksanovic is also the primary investigator of a Mayo Phase 2 Consortium grant leading a pilot study of AZD2171, an anti-angiogenic compound, with novel translational components relevant to angiogenesis.

Ahmet Dogan, M.D., Ph.D., who specializes in B-cell malignancies, in conjunction with Brian O'Neill, M.D., and Dr. Giannini, used interphase FISH to demonstrate that the frequency of specific chromosomal abnormalities in PCNSL differed from systemic diffuse large B-cell lymphoma, suggesting a distinct pathogenesis (plenary presentation at the United States and Canadian Academy of Pathology 2007 meeting).

Dr. O'Neill, in conjunction with Ricardo Lloyd, M.D., Ph.D., extended preliminary observations regarding a role for SV40 in PCNSL tumorigenesis. Antigen was convincingly localized to CD20-positive large lymphocytes and informed a new R03 grant designed to localize antigen and determine its potential as a "homing" cue. Drs. O'Neill and Lloyd also identified prolonged survival in the cohort of PCNSL patients whose tumors demonstrated SV40, suggesting that identification of SV40 antigen may have biomarker potential.

Paul Brown, M.D., and colleagues in Mayo's Psychiatry, Behavioral Neurology, and Genetic Epidemiology and Risk Assessment Programs, as part of Mayo's Neurobehavioral Conference Group, have completed and published the results of a prospective study evaluating quality of life (QOL) in newly diagnosed high-grade glioma patients. As a component of this study they included analysis of the extent of surgical resection and its impact on QOL. Another completed study examined multidisciplinary interventions to improve QOL for patients with advanced cancers. Other prospective studies of specific interventions are underway, including erythropoietin for fatigue in central nervous system metastases, and citalopram for depression in glioma patients.