Skip to content
BACK to News

Smarter Classification of Gliomas: NBTS Funded Research

Published on June 22, 2015 in Brain Tumor Information, Research, Our Impact, Take Action

Guest Author: National Brain Tumor Society

NBTS Associate Director of Scientific Operations, Jennifer Helfer, PhD, contributed to this post. 

Following the 2014 Society for Neuro-Oncology Annual Meeting last November, we recapped how the scientific and medical field was moving toward a smarter classification for the many different types, and subtypes, of brain tumors.

Here’s what we wrote at the time:

Currently, gliomas by-and-large are classified into grades (which affects diagnosis, prognosis, and treatment decisions) based on what’s called histopathology. In this paradigm, a pathologist examines a tissue sample from a patient’s biopsy under a microscope and produces a report on the tumor’s grade and diagnosis from an observable set of standard characteristics.

However, more and more studies over recent years are showing that genomics and molecular analysis of tumors can provide a deeper level of understanding of each tumor’s unique biology (and thus, again, better inform diagnosis, prognosis, and treatment).

Just this month, two major studies were published on this topic in the same issue of the New England Journal of Medicine.

The studies focused on a wide-range of adult, diffuse gliomas. Diffuse gliomas are currently classified as astrocytomas, oligodendrogliomas, or oligoastrocytomas and span in grade from II to IV.


The first study – from The Cancer Genome Atlas (TCGA) consortium of more than 300 scientists from 44 different institutions – analyzed lower-grade gliomas (grades II and III). Using the latest in molecular and genomic sequencing technologies, the TCGA study produced three distinct subgroups of low-grade gliomas based on two molecular changes that were discovered to happen early in the process of tumor development: mutations to a gene known as IDH, and simultaneous alterations on arms of two chromosomes (the 1p/19q co-deletion).

  • IDH mutation & 1p/19q co-deletion: The group with both a mutation in the IDH gene and the 1p/19q co-deletion corresponded with the best prognosis (median 8 year survival). These tumors were otherwise classified mainly as grade II oligodendrogliomas. This group was associated with additional mutations to the TERT, CIC, and FUBP1 genes (which were rarely seen in the other two subgroups).
  • IDH mutation only: The group with an IDH mutation, but no 1p/19q co-deletion had the youngest age at diagnosis, and a slightly lower mean survival (6.3 years). This group was associated with additional mutations to the TP53 and ATRX. They were otherwise typically classified as grade II astrocytomas or oligoastrocytomas.
  • Neither marker: The group with no mutation to their IDH gene was associated with an older age at diagnosis. This group typically would have been classified as a mix of grade III gliomas, but actually revealed mutations more commonly associated with grade IV glioblastomas (GBM), including alterations to the EGFR, NF1, and PTEN genes, and thus had the worst prognosis of the subgroups.

We hope this will impact how physicians both diagnose and plan therapies for brain cancer.

Attribution: Dr. W.K. Alfred Yung, study leader and top NBTS advisor

In the second study, researchers from the Mayo Clinic and the University of California, San Francisco (UCSF) analyzed 1087 diffuse gliomas (both high-grade and low-grade: grades II, II and IV) for the presence of IDH mutations, 1p/19q co-deletions, and mutations to the “promoter” region (the area in a gene’s DNA sequences that can turn the gene “on” or “off”) of a gene known as TERT.

Virtually all of the 1087 tumors analyzed (97%), could be stratified into five, similar, but even more specific subgroups than the TCGA study:

  • Triple-Positive: This subgroup had all three of the IDHTERT  and 1p/19q alterations. This subgroup associates mostly with grade II and III and oligodendroglioma histological classification. Mean age at diagnosis is 44 and median survival is greater than 10 years.
  • IDH & TERT mutations: This subgroup associates mostly with grade II and III astrocytoma histological classification. Mean age at diagnosis is 46 and median survival is greater than 10 years.
  • IDH mutation only: This subgroup is associated with TP53 and ATRX mutations and associates mostly with a mix of grade II and III astrocytoma and oligodendroglioma histological classification. Mean age at diagnosis is 37 and median survival is 8.9 years.
  • TERT mutation only: This subgroup only harbors the TERT mutation and is associated with EGFREGFRvIII, NF1, PTEN, RB1, and PIK3 genetic alterations. This subgroup associates mostly with grade IV and astrocytoma histological classification. Mean age at diagnosis is 59 and median survival is 1.8 years.
  • Triple-Negative: This subgroup harbors none of the defining genetic alterations, but is associated with EFGR, NF1, and PTEN alterations. This subgroup associates mostly with grade IV and astrocytoma histological classification. Mean age at diagnosis is 50 and median survival is 3.5 years.

Unfortunately, classifying a tumor only by appearance and grade has not provided sufficient information about the way the tumor is likely to behave, how it will respond to treatment, or the patients likely survival time. These markers will potentially allow us to predict the course of gliomas more accurately, treat them more effectively, and identify more clearly what causes them in the first place.

Attribution: Dr. Margaret Wrensch, UCSF

Funding for this research was provided to Dr. Wrensch from the National Brain Tumor Foundation (one of two organizations that merged to become NBTS). NBTS is currently funding the work of Dr. Jenkins and Dr. Lachance through Oligo Community Research Fund Phase II, and Dr. Jenkins is on our Scientific Advisory Council.


Taken together, these studies provide the latest evidence that gliomas are more accurately defined and characterized by molecular and genetic markers than current methods.

This molecular data helps us better classify gliomas patients, so we can begin to understand who needs to be treated more aggressively and who might be able to avoid unnecessary therapies.

Attribution: Dr. Daniel Lachance, Mayo Clinic

Under the current system, Dr. Wrensch expanded, “someone with a grade III glioma, for example, may not have been treated as aggressively as someone with a grade IV. But now, if you determine that it’s a TERT-mutated-only tumor, there is more confidence that it will behave more like a grade IV tumor and could be treated more aggressively. A grade III tumor that only has an IDH mutation might be treated les aggressively. Glioma treatments can be very toxic, so it’s important to know how aggressive treatment needs to be.”

Leaders from both study groups were careful to say that their results could benefit from more data and study, especially in terms of the association between genetic make-up and potential survival rates. However, these studies are likely to significantly add to the information and considerations that the World Health Organization is currently reviewing as its looks to issue a revised classification for brain tumors later this year.

This genome-wide analysis will be much more objective and likely will be practice-changing. It can easily be implemented and will markedly improve diagnosis, patient care, and treatment planning.”

Attribution: Dr. Andrew Sloan, Director of the Brain Tumor and Neuro-Oncology Center at Case Western Reserve University School of Medicine

Stay Informed & Connected