New findings on brain tumors called ependymomas

Brain tumors called ependymomas that occur in different parts of the central nervous system appear to arise from subpopulations of stem cells called radial glia cells (RGCs), according to investigators at St. Jude Children's Research Hospital. The discovery explains why some identical-looking ependymomas are actually distinctly different diseases, the researchers said.

This new information, in combination with the techniques used to conduct the study, holds promise for designing more effective treatments for ependymomas as well as for other solid tumors. A report on this work appears in the October issue of Cancer Cell. RGCs are unspecialized cells that line the surface of the ventricles (fluid-filled spaces in the brain) and the spinal cord, and give rise to normal mature cells in the nervous system. The St. Jude study found strong evidence that when rare populations of RGCs acquire mutations that disrupt the cell signaling pathways controlling growth and differentiation, these cells reproduce continually and give rise to an ependymoma.

The St. Jude finding that RGCs can give rise to these tumors is consistent with evidence from a variety of researchers that cancers arise from, and are maintained by, a rare number of mutated stem cells called cancer stem cells, according to Richard Gilbertson, M.D., Ph.D., associate member in the Developmental Neurobiology and the Hematology-Oncology departments. Gilbertson is senior author of the Cancer Cell paper.

The current discovery at St. Jude explains why ependymomas arising in various parts of the central nervous system are clinically different, even though they look the same histologically (as seen under a microscope), the St. Jude researcher said. For example, although all ependymomas look alike, supratentorial ependymomas arise in the top part of the brain in both adults and children; often cause weakness in the arms and legs, visual problems and seizures; and have a survival rate of 50-60 percent. Posterior fossa ependymomas arise in the back of the brain and cause patients to have an unsteady walk and neck pain; and they occur mainly in children and have a slightly worse prognosis than do surpratentorial tumors. A spinal ependymoma occurs mainly in adults, and more than 70 percent of patients who undergo surgery to remove this tumor survive.

"Historically, physicians based their diagnosis and treatment of cancer primarily on the histology of tumors," Gilbertson said. "So our demonstration that identical-looking ependymomas that arise in different regions of the central nervous system are distinct diseases at the cellular and molecular level is an important insight. This suggests that treatments should be designed to kill the cancer stem cells. If you kill only the cells making up the bulk of the tumor, the disease will likely return because you haven't eliminated the stem cells that are the source of the tumor."

The St. Jude study is also important because ependymomas are the third most common central nervous system tumor in children and no effective chemotherapy exists for them. "If surgery and radiation doesn't treat the entire tumor, then resistant stem cells left behind might re-grow the cancer," Gilbertson said. "And since children don't tolerate radiation treatment well, we need new treatments that completely eliminate cells that produce the tumors."

The researchers made their discovery by first determining the patterns of gene expression in more than 100 tumor samples from patients with different types of ependymoma. Gene expression patterns, called signatures, reflect the specific genes that have been activated in the tumors. The signatures that distinguished supratentorial, posterior fossa and spinal ependymomas included genes that regulate the proliferation and differentiation of normal primitive cells in the corresponding region of the embryonic nervous system. For example, the St. Jude team showed that more than 80 percent of genes expressed at high levels in the supratentorial and spinal ependymomas are also expressed in the corresponding regions of the nervous system during development.

The team also demonstrated that each subtype of ependymoma contains rare populations of cells that resemble RGCs. Moreover, when these RGC-like ependymoma cancer stem cells were inserted into laboratory models that lacked protective immune systems, the stem cells formed tumors. This was additional strong evidence that mutated RGCs can give rise to ependymomas.

The different genetic signatures found in each subtype of ependymoma represent potential targets for new drugs designed to kill RGCs that give rise to each subtype of this tumor, said Gilbertson. Such individualized treatment might allow physicians to prevent the recurrence of ependymoma following treatment to remove the primary (original) tumor by eliminating the cancer stem cells that give rise to the tumor.

The technique the St. Jude team used to identify populations of RGCs as cells of origin of ependymoma could also be used to identify cancer stem cells for other solid tumors, according to Helen Poppleton, PhD, an associate scientist. "That new knowledge could lead to the development of new drugs that significantly improve the outcomes of a variety of cancers," Poppleton said. One of the authors of the paper, Poppleton did much of the work on this project.