Medulloblastoma originates in normal brain "stem" cells that turn malignant

An aggressive childhood brain tumor known as medulloblastoma originates in normal brain "stem" cells that turn malignant when acted on by a known mutant, cancer-causing oncogene, say researchers from Dana-Farber Cancer Institute and the University of California, San Francisco (UCSF).

Reporting in the Aug. 12 issue of Cancer Cell, the scientists say they have uncovered new origins for these tumors from early stem cells as well as more mature cells. Previously, scientists had assumed the tumors might only come from a single source: more mature cells which become neurons and do not have "stem" cell properties. The findings hint at potential new treatment approaches for medulloblastoma by targeting the origins of the tumors, and further suggest that not all patients' tumors may be born from the same cells.

"We now have a better idea of where these brain tumors come from and their relationship to normal stem cells in the brain," said Keith Ligon, MD, PhD, co-senior author of the report and an investigator at the Center for Molecular Oncologic Pathology at Dana-Farber and the Brigham and Women's Hospital in Boston.

Co-senior author, David Rowitch, MD, PhD, currently a professor of pediatrics and neurosurgery at UCSF and a Howard Hughes Medical Institute investigator, commented that mouse experiments shed light on how normal stem cells -- cells with the power to create all types of cells in the brain -- can be transformed into tumors. The transformation occurs when a cell-signaling pathway known as Sonic hedgehog (named for a cartoon character) is reactivated by a chance mutation.

Sonic hedgehog plays an important role during the embryonic development of the brain, but normally shuts down when it's no longer needed. When turned on again by a mutation, the signals can trigger cell processes leading to tumors -- not just in the brain, but in other organs as well.

Medulloblastomas, usually diagnosed in children between 2 and 5 years of age, affect the brain's cerebellum region, which is involved in controlling body movements. They make up about 30 percent of childhood brain tumors, and account for 250 to 300 new cases per year.

With current treatments, approximately 60 to 70 percent of patients live at least five years, but often they are left with cognitive disabilities from surgery, chemotherapy and radiation, urgently suggesting a need for new, more-selective therapies.

"Medulloblastoma was one of the first tumors that was believed to fit the hypothesis that tumors are caused by 'cancer stem cells' that initiate malignancies and sustain them," said Ligon, who is also on faculty and an assistant professor at Harvard Medical School. "But the prevailing hypothesis -- that medulloblastomas originate from non-stem cells just did not make perfect sense with this."

The discoveries emerged from a series of experiments begun in the Rowitch Laboratory at Dana-Farber. The initial goal was to determine whether activating the Sonic hedgehog cancer pathway in multiple types of brain cells, including neural stem cells, could help pinpoint which cells brain cancers might come from. Surprisingly, the scientists generated just one tumor type, medulloblastoma, regardless of whether they activated the pathway in stem cells for other cell types called neurons and glia. This was a surprise: it had been thought that medulloblastoma arose purely from neuronal ("thinking") cells and not "glial" or supporting cells.

An intriguing question for the investigators is why these cells, known as granule neuron precursors, seem to be uniquely vulnerable to the tumor-triggering effects of the Sonic hedgehog pathway, while other brain stem and progenitor cells are not.

Explained Rowitch: "There must be susceptibility factors in the granule neuron precursor cell that predispose it to forming cancer, so we now must try to understand what it is about this cell type that makes it susceptible to forming cancer in response to Sonic hedgehog signaling. This relationship between stem cells and oncogenes suggests a new point of potential therapeutic intervention."