Researchers Identify Distinct Subtypes of Glioblastoma

The Bottom Line

A detailed genetic analysis of brain tumors from people with glioblastoma has allowed researchers to identify four distinct subtypes of the disease. This work lays the groundwork for improved understanding of what goes wrong in this deadly cancer, for better prediction of outcomes, and, ultimately, for the development of more targeted and more effective treatments.

The Whole Story

Glioblastoma multiforme, the most common type of malignant brain tumor in adults, is a particularly fast-growing cancer. The median survival for patients diagnosed with this disease is approximately 15 months, even with aggressive treatment. One of the challenges doctors face in treating patients with glioblastoma is that they don’t have a good way to predict the likely outcome or course of the disease for a given patient. Therefore, researchers have been searching for genetic “signatures” that could help in determining patient prognosis or predicting response to treatment.

This vital genetic information is now emerging from studies of glioblastoma by The Cancer Genome Atlas (TCGA) Research Network. TCGA, a collaborative effort of the National Cancer Institute and the National Human Genome Research Institute, both components of the National Institutes of Health, uses the latest techniques in genomic analysis to investigate the molecular basis of cancer. TCGA involves more than 150 researchers at dozens of institutions across the United States. Data from TCGA studies are made rapidly available to the research community through an online database.

In one recent study, TCGA investigators used gene expression profiling, which measures the activity (expression) of thousands of genes in a single sample, to analyze 200 glioblastoma specimens and 2 normal brain specimens. Based on the expression patterns of 840 genes that were expressed at either higher or lower levels in glioblastomas than in normal brain tissue, the researchers were able to classify tumors into four distinct molecular subtypes: proneural, neural, classical, and mesenchymal. Subsequent studies of 260 independent glioblastoma samples confirmed the gene expression patterns of these four subtypes.

In addition, the researchers found that mutations in specific genes and certain chromosomal changes were associated with particular subtypes. Moreover, they found connections between several clinical features of glioblastoma, such as the age of the patient and the patient’s responsiveness to aggressive therapy, and molecular subtype. For example, the proneural subtype was more common in younger patients. And, the use of aggressive therapy reduced the rates of death for patients with the classical and mesenchymal subtypes but not the proneural subtype.

In a second study, TCGA researchers analyzed glioblastoma specimens to look for a type of DNA modification known as methylation. The methylation of genes can reduce their expression. The researchers analyzed 272 tumor specimens and found that a subset of glioblastomas, which they called G-CIMP tumors, showed increased methylation of a large group of genes. When the DNA methylation data and the gene expression data were analyzed jointly, the investigators found that G-CIMP tumors make up a subset of proneural tumors. Patients with G-CIMP tumors are generally young and have a relatively good prognosis. 

Better understanding of glioblastoma at the molecular level from these and other ongoing studies is setting the stage for the development of more effective and personalized therapies for glioblastoma patients. These discoveries also show the potential of data being generated by TCGA. Results from this program are expected to help researchers understand the molecular underpinnings not only of glioblastoma but also of more than 20 additional cancer types.

References:

Verhaak RGW, Hoadley KA, Purdom E, et al. Integrated genomic analysis identifies clinically relevant subtypes of Glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell 2010; 17: 98-110.

Noushmehr H, Weisenberger DJ, Diefes K, et al. Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma. Cancer Cell 2010; 17:510-522.