Genomic Profiling Provides Much-Needed Prognostic Data in Glioma


Low-grade gliomas may be classified into three distinct prognostic categories based on genomic data.

Low-grade gliomas may be classified into three distinct prognostic categories based on genomic data. This conclusion came from a genomic and molecular analysis presented at the 2015 annual meeting of the American Society of Clinical Oncology (ASCO), a gathering of nearly 30,000 oncology professionals in Chicago on June 2, 2015,1 and published in Acta Neuropathology.2

Clinicians currently rely on histologic classifications of gliomas to help predict a patient’s prognosis and to select treatment, but for gliomas that the World Health Organization (WHO) classifies as grade II or III — astrocytic, oligodendroglial, and oligoastrocytic glioma — histologic information may not be enough, according to Michael Weller, chair of the Department of Neurology at the University Hospital Zurich, Switzerland, and his colleagues.2 Classifications in these groups by histology alone are not always consistent among clinicians, especially for oligoastrocytomas. The clinical outcomes vary greatly with the grades, and there is little standardization of treatment options for each grade.

“There is a paucity of biologic and clinical trial data on low-grade gliomas,” said Howard Alan Fine, of the Weill Cornell Medical Center in New York City, who provided an independent analysis of Weller’s presentation. This has made it difficult for clinicians to counsel patients and select treatment options.

To overcome the limitations of the current classification system, clinicians and clinical investigators are increasingly turning to molecular markers such as the expression of certain mutations or genomic patterns. They hope that more accurate classifications would provide more accurate prognostic information, help guide treatment choices, and lead to improved clinical trials.

“There is increasing use of molecular markers for diagnostic purposes, although there is a lot of knowledge and we are not [all] using it the same way,” Weller says.

Weller’s study1,2 demonstrates how adding molecular markers, such as copy number variations and messenger RNA (mRNA) expression profiles, to histologic analyses might improve classification of this hard-to-classify group of tumors.

The study analyzed WHO grade II or III cerebral gliomas from 137 patients who agreed to participate in the German Glioma Network study. Weller et al used microarrays to create genomic profiles of the tumors that were then matched to histologic information and patient data. Weller explained that comparing genomic hybridization patterns among tumors allows researchers or clinicians to see patterns that they would not see by looking at individual genetic variations.

The analyses showed that there are five distinct groups based on genomic profiles: three express mutations in the IDH1/2 gene, and two have the wild-type IDH1/2 gene. By linking the genomic and clinical data, Weller defined three main prognostic groups. The best prognosis was associated with tumors that had both an IDH1/2 mutation and 1p/19q deletions. Weller said these tumors represent a distinct disease entity.

The worst prognosis was associated with IDH1/2 wild-type tumors that had genomic characteristics similar to glioblastomas, such as an extra chromosome arm 7q, a missing chromosome arm 10q, a mutation in the TERT promoter gene, or amplified oncogenes. Weller argued that these may actually be unrecognized glioblastomas, which Fine agreed was a possibility.

Patients had moderate prognoses whose tumors expressed an IDH1/2 mutation but 1p/19q intact, or who had IDH1/2 wild-type gliomas without chromosome arm 7q and 10q copy variations or genotype and TERT promoter mutations.

The researchers also looked at gene expression patterns in the tumors. Gene expression data suggested eight different groups, but these did not overlap well with the genome-defined groups. Weller said it was disappointing that the gene expression data were not very useful.

“It didn’t add a lot,” he says.

Weller predicts a new classification system is on the horizon.

“We are probably moving to a new classification system in the next few years where we place IDH1 mutations at the top of the world,” he says.

Fine agreed that the molecular low-grade glioma typing data from Weller and other trials presented at ASCO would likely lead to a new classification system for gliomas in the near future and will have a lasting impact on the treatment of gliomas.

“What we [have] today is a better understanding of the genetic and molecular biology of the disease,” he says. “Genotyping each individual patient is going to be critical to targeted therapy.”

According to Fine, the Weller’s data also provide valuable prognostic information for patients with low-grade gliomas, showing that prolonged survival is possible in two of the three genomic groups, but unlikely in the third.


1. Weller M. Molecular classification of diffuse cerebral gliomas using genome- and transcriptome-wide profiling. Presented at: 2015 American Society of Clinical Oncology Annual Meeting; May 29-June 2, 2015; Chicago, IL. Abstract 2007.

2. Weller M, Weber RG, Willsher E, et al. Molecular classification of diffuse cerebral WHO grade II/III gliomas using genome- and transcriptome-wide profiling improves stratification of prognostically distinct patient groups. Acta Neuropathol.

Related Videos
Image of a man with dark hair wearing a suit with a light blue tie.
Related Content