The finding adds a new layer of complexity to the quest to understand the causes of childhood brain cancers, according to senior author David H. Gutmann, M.D., Ph.D., the Donald O. Schnuck Family Professor of Neurology at Washington University School of Medicine in St. Louis and co-director of the neuro-oncology program at the Siteman Cancer Center.
"Our findings suggest that brain tumors arising in different regions may be genetically distinct as a consequence of their unique cellular origins," Gutmann says. "This is yet another factor we need to consider when trying to understand how pediatric brain tumors form."
Researchers use information about tumor origins to develop new tests and treatments for the tumors. Brain tumors are the leading cause of cancer-related death in children, and the most common childhood brain tumor is the pilocytic astrocytoma (PA). Approximately 15 percent of all PAs are linked to neurofibromatosis 1 (NF1), a genetic condition that causes childhood brain tumors and is a primary focus of Gutmann's research. However, the genetic basis for the majority of PAs is unexplained.
In the new study, Gutmann led six laboratories in the most detailed genetic analysis of PAs to date.
"We were hoping to identify genes that contribute to the formation of these tumors and find indicators that might help us predict which tumors will be relatively well-behaved and which will be more aggressive," Gutmann says. Previous studies have failed to produce any solid leads on the genetic alterations that predispose children to PAs.
"It should be recognized that the genetic alterations in this tumor may be very subtle," Gutmann notes. "When we looked at gene activity levels in the tumors as a f unction of brain location, though, a very interesting pattern began to emerge."
Cells in different parts of the brain carry the same genes, but they also contain factors that modify the use of those genes, suppressing some genes and activating others to allow the cells to take on specialized characteristics as the brain matures. These changes in gene activity levels are called changes in gene expression.
The researchers found that tumors arising in different regions of the brain retain distinct patterns of gene expression. These patterns provided genetic fingerprints or bar codes for the location of PAs, as well as for another glial cell tumor called an ependymoma. In addition, scientists also detected these distinct patterns of expression in normal glia and stem cells from these brain locations, suggesting that genetic fingerprints can be used to identify the potential origins of brain tumors.
"There's been a movement in recent years to link normal brain development to pediatric neuro-oncology, and these findings affirm that as a necessary approach," Gutmann says. "We won't fully understand the causes of pediatric brain tumors until we consider them in the context of factors that shape the development and specialization of different brain regions."