NEUROSURGERY ARTICLES

UC IRVINE BRAIN TUMOR RESEARCH LABORATORY

Mark Linskey, M.D. and Yi-Hong Zhou, Ph.D.

Molecular biology is the study of the genetic building blocks of cell identity and behavior. Genes are made of DNA and reside in chromosomes in the cell nucleus, while RNA molecules communicate the gene instructions to the intracellular factories that make the proteins that actually carry out the genetic instructions. In the Brain Tumor Research Laboratory of the Department of Neurological Surgery at the University of California, Irvine, we study brain tumor molecular biology in order to try and overcome the limits of traditional tumor diagnosis and classification.

The most common malignant tumor in the brain arises from the glial cells that support the electrical cells of the brain called neurons. As tumors arising from glial cells, they are called gliomas. Malignant gliomas are one of the most aggressive forms of cancer to strike the human body. Since they arise in and damage the brain, they can have a significant effect on both functional independence and the quality of a patient's life.

Doctors divide malignant gliomas into their various types based on how the individual cells look under the pathologist's microscope and how they form structures in relation to one another as well as normal tissue cells (histology). Tumor classifications based on histology allow doctors to predict patient survival (prognosis), as well as choose among different therapies, that might be effective for some classes of gliomas and not for others. The most aggressive and dangerous classification of malignant glioma is the grade 4 astrocytoma also known as "glioblastoma multiforme" (GBM).

There are two problems with classifying tumors based on histology. The first is that the distinctions based on appearance under the microscope are very subjective and open to differing interpretation among even experienced pathologists. Indeed, in two separate studies performed in 1996 and 1997 at major university hospitals, four different fellowship-trained neuropathologists in each study only agreed with one another on what to call a given glioma specimen 64% and 52% of the time, respectively. This means that there is a significant probability that the patient could be given an improper diagnosis. Therefore, the patient might not receive correct information regarding prognosis and might even be referred for therapy that was not targeted at their tumor 36-48% of the time.

Secondly, even if pathologists agree on what to call a glioma based on histology, different tumors that look exactly the same actually behave very differently biologically. Some are much more aggressive than others. Some respond to one therapy quite well, while others do not respond to that therapy at all. Doctors call this unpredictable mix of behavior and response in similar looking tumors -"heterogeneity". If we could define tumor heterogeneity, we could potentially improve prognosis for patients, choose the most effective treatment method for a given patient, and avoid toxicities of treatment with a low chance of success.

Several experimental studies have shown that cluster analysis of unidentified gene expression can better distinguish different glioma types than pathologists reviewing the histology of these specimens under the microscope. In 2005, our own lab demonstrated that a model looking at the quantitative expression of only 8 genes coupled with three readily-available clinical variables could account for 67% of all heterogeneity in patient outcome within a histology tumor class. In the future, it is very likely that brain tumors will be diagnosed by gene expression and not by appearance under the microscope.

The first hint that molecular biology also effected glioma response to treatment came from the observation that certain gliomas that lacked normal genes on the short arm of chromosome #1 (1p - each human chromosome has a paired short [p] and long [q] arm) responded better to chemotherapy. The exact genes involved have yet to be worked out. The next indication that this might be true for other therapies as well, came from the observation that the majority of GBM patients who responded well to the newer treatment of temazolamide chemotherapy at the same time as radiation therapy, had inactivation of a gene that interferes with chemotherapy function (methylation of the MGMT gene promoter). Our lab and others are actively searching for gene expression signatures that predict responsiveness to different current and experimental therapies. Just as importantly, we are searching for gene signatures that will identify patients who will not respond to therapies, such as radiation therapy, so that we can spare these patients six weeks of time lost on a predictably ineffective therapy, as well as the potential side effects involved.

Our laboratory is also using molecular biology techniques to identify molecules that might be measurable in serum, blood, or spinal fluid. These findings may be able to tell us how effective a treatment is. They may also indicate when a therapy is no longer effective. These "biomarkers" could allow us to change to an effective therapy before a brain tumor has a chance to re-grow or enlarge and cause symptoms that might require further surgery.

Whether for accurate diagnosis, better prognosis, optimal choice of therapy, avoidance of ineffective therapy, or monitoring therapy response independent of tumor growth, molecular biology holds the key to advances in brain tumor research and future clinical trial design. Our brain tumor research laboratory is always seeking additional research grant funding support as well as philanthropic support for our efforts. A research endowment fund has recently been created to facilitate outside donations. It is only through aggressive molecular biology research efforts at the only university multidisciplinary neuro-oncology program, based in the only National Cancer Institute designated Comprehensive Cancer Center in Orange County, that we are likely to improve the outlook for our families and friends who suffer from malignant brain tumors.

University of California, Irvine • Irvine, CA 92697
(949) 824-5011
© 2006 The Regents of the University of California.
All Rights Reserved.

Comments & Questions: Privacy & Legal Notice
Copyright Inquiries