Biomarkers Define Distinct Types of Diffuse Glioma

Two reports in The New England Journal of Medicine lay the foundation for molecular classification of diffuse gliomas, a heterogeneous group of brain tumors currently diagnosed by their appearance under the microscope. In the studies, genetic markers performed far better than histologic criteria to define three major disease subgroups, each associated with a different clinical profile.

The findings are timely, coming just as clinicians are meeting to revise the World Health Organization (WHO) diagnostic classification of brain tumors to include—for the first time—molecular criteria.

The two studies took different approaches. One, from The Cancer Genome Atlas (TCGA) Research Network, used unsupervised clustering of molecular data on protein expression, DNA methylation, mRNA and miRNA expression, and DNA copy number and mutations in 293 lower-grade (WHO grades II and III) gliomas (N Engl J Med 2015;372:2481–98). The other, from researchers at the Mayo Clinic in Rochester, MN, and the University of California, San Francisco, assessed three previously identified biomarkers in 1,087 gliomas encompassing grades II through IV (N Engl J Med 2015;372:2499–508).

In both studies, the molecular data settled into three robust, cohesive tumor groups that could be defined based on isocitrate dehydrogenase (IDH) mutational status and the presence or absence of a 1p/19q chromosome codeletion. The Mayo study distinguished two additional tumor groups, one with mutations in the telomerase reverse transcriptase gene (TERT) promoter and the second with both TERT and IDH mutations.

The major groups—tumors with neither IDH mutations nor 1p/19q codeletion, with IDH mutations only, or with IDH mutations and 1p/19q codeletion—accounted for more than 95% of the grade II and III tumors in the two studies, showed little overlap, and correlated only modestly with histologic class or grade. Both studies found that the distinct types had significantly different age of onset and median survival, and developed characteristic secondary genetic alterations.

Notably, the majority of low-grade tumors that were wild-type for IDH actually possessed the genetic profile and aggressive course seen in patients with grade IV glioblastomas. Many in this group had TERT promoter mutations, which are common in glioblastomas. In the TCGA study, these tumors had a median survival time of 1.7 years, compared with 6 to 8 years for tumors with IDH mutations.

“We think we are capturing early glioblastomas,” says neuropathologist Daniel J. Brat, MD, PhD, of Emory University School of Medicine in Atlanta, GA, the lead author of the TCGA study. “There is no way to identify that behavior under the microscope without additional testing,” he adds.

IDH mutations and 1p/19q codeletion are already routinely assessed in gliomas to provide ancillary information to the histologic diagnosis that affects treatment decisions. The new work supports the idea that the two, along with TERT promoter mutations, could improve diagnosis.

The findings will also affect preclinical research and drug development efforts, says neuro-oncologist Ingo K. Mellinghoff, MD, of Memorial Sloan Kettering Cancer Center in New York, NY, who was not involved with the study. Having the detailed genetic profiles of each glioma subtype will enable researchers to develop more appropriate animal models. For clinical trials, the results will allow testing of drugs geared toward particular tumor types and molecular targets. “It will really change the way people think about the disease,” Mellinghoff says.

For more news on cancer research, visit Cancer Discovery online at http://CDnews.aacrjournals.org.