Posted: February 1, 2011

TCGA Researchers Identify a Distinct Subtype of Glioma Linked to Less Severe Outcome

Catherine Evans

Glioblastoma multiforme (GBM) is an advanced form of glioma, a cancer that develops in the brain cells that support neurons. Once diagnosed, less than half of GBM patients survive one year. Even if initial treatments counter cancer growth, tumors inevitably recur. Researchers from The Cancer Genome Atlas (TCGA) are using a variety of technologies to identify genetic changes in these tumors that they hope will lead to better diagnostic methods and treatments. In a paper published by TCGA scientists in the May 18, 2010 issue of Cancer Cell, Houtan Noushmehr, Ph.D. and colleagues found that a subset of GBM samples displayed a different pattern of chemical reactions, called DNA methylation, than the rest of the samples. The patients whose tumors had this distinct pattern of methylation were younger and lived longer than other GBM patients, forming a potential patient subgroup for tailored treatment. The researchers called the subtype of tumors in these patients “G-CIMP,” for glioma CpG island methylator phenotype.

Methylation Patterns Define Patient Subgroups

DNA methylation is a normal chemical reaction that occurs when a chemical ‘mark’ is placed on specific pieces of DNA throughout the genome. These chemical marks influence whether genes are switched on and turned into proteins that direct growth and maintenance of cells. Inappropriate methylation can lead to an imbalance of the proteins that control cell growth and is thought to be one way cancer develops. Different patterns of improper methylation can occur in tumors of the same cancer type, pointing to the existence of different molecular pathways that cause tumors to develop within one cancer. Scientists want to define each type of methylation pattern so that they can develop treatments specific to both tumors and patients. 

In examining the distinct methylation pattern of this new subtype of GBM, the researchers observed that the G-CIMP tumors had several features in common with less advanced gliomas. The G-CIMP tumors came from patients who were younger at diagnosis, with a median age of 36, compared to a median age of 56 for patients without G-CIMP tumors. These patients also survived longer, with a median survival time of 150 weeks, compared to a median of 42 weeks in patients without G-CIMP tumors. The G-CIMP tumors tended to belong to tumor groups that are slower progressing, less invasive and lacking in genetic changes normally seen in GBM. Overall, a picture emerged of a GBM subtype that behaved more like a low-grade glioma than like the uniformly aggressive GBM family to which it belonged. TCGA researchers then sought to determine what happened at the molecular level to set these tumors apart.

Implications of Methylation Profiles

The first hints came from a more detailed look at the specific genes affected. The research team compared the genes that were methylated in each tumor subtype. A subset of genes already known to be linked to tumor progression was methylated only in the G-CIMP tumors, an event that may account for their less aggressive nature. The researchers believe that the distinct methylation profiles and the resulting gene expression differences between G-CIMP and non-G-CIMP glioblastomas point to the likely mechanisms that cause them to behave so differently in a clinical setting.

More work remains to help us understand the full range of gene and protein changes that result from the methylation profiles specific to each glioblastoma subtype. However, discovering the existence of these subtypes is a big step toward creating personalized treatments for glioblastoma patients, as well as in understanding what makes glioblastoma so aggressive.


Noushmehr, H., Weisenberger, D.J., Diefes, K., Phillips, H.S., Pujara, K., Berman, B.P., Pan F., Pelloski, K.E., Suhmar, E.P., Bhat, K.P., et al. (2010) Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma. Cancer Cell. 17(5):510-522. View PubMed abstract.