“A Complete Molecular Atlas of Glioblastoma”

A detailed examination of the genes, proteins, metabolites, regulatory RNAs, post-translational modifications, and epigenetic markers active in glioblastoma has revealed new molecular drivers and immune subtypes of the deadly brain cancer. The findings, from the largest multi-omics investigation of glioblastoma conducted to date, should inform treatment strategies and clinical trial design (Cancer Cell 2021;39:509–28.E20).

“This is a huge resource—a complete molecular atlas of glioblastoma,” says Karin Rodland, PhD, of the Pacific Northwest National Laboratory (PNNL) in Richland, WA, and the Oregon Health & Science University in Portland, who co-led the research.

Neuro-oncologists currently rely on genomic mutation profiles and gene expression signatures to identify subtypes of glioblastoma, but such differential diagnoses have not dramatically altered therapeutic management of the disease—usually some combination of surgery, chemotherapy, and radiotherapy. Cancer biologists have thus remained on the hunt for molecular features to better stratify glioblastomas and select the likeliest effective treatment for each patient.

To that end, Rodland and her collaborators, including co-lead investigators Li Ding, PhD, of Washington University in St. Louis, MO, and PNNL's Tao Liu, PhD, threw the kitchen sink of analytic tools at 99 treatment-naïve glioblastomas and 10 unmatched healthy samples. Their study formed part of the NCI's Clinical Proteomic Tumor Analysis Consortium (CPTAC), an effort launched 10 years ago to add a new protein-based layer of biological understanding to genomically characterized tumors (Cancer Discov 2013; 3:1108–12).

Combining 10 different omics platforms, the researchers filled in molecular details about the three known subtypes of glioblastoma, each originally defined by gene expression patterns alone. Focusing on protein acetylation patterns, for example, they found that the “classical” subtype was enriched for post-translational modification on chromatin modifiers and DNA repair proteins. And using single-cell RNA sequencing, they showed that the “mesenchymal” subtype was defined by transcriptomic patterns expressed by the tumor cells themselves, rather than the surrounding stroma.

The team also discovered four new subtypes of glioblastoma characterized by distinct immune cell infiltrates. Plus, they identified phosphorylation events linked to two proteins—PTPN11 and PLCG1—proposed to act downstream of oncogenic mutations in frequently altered genes to fuel tumor growth.

Those two phosphorylated proteins, and the signaling hub they collectively form, now provide promising new drug targets, says Hyun-Seok Kim, PhD, of Yonsei University College of Medicine in Seoul, South Korea, who was not involved in the work. Also, Kim is intrigued by the suggestion that the four new immune subtypes of glioblastoma outlined in the paper might affect patient responses to checkpoint blockade. But, he says, “it would be important to test those hypotheses experimentally in relevant in vitro and in vivo models.”

According to Rodland, various CPTAC-affiliated researchers are engaged in such follow-up studies.

Investigators from the consortium also recently described a proteogenomic characterization of 218 pediatric brain tumors across seven histologic types (Cell 2020;183:1962–85.E31). They found druggable pathways that span histologic boundaries, as well as proteomic features that help explain why particular treatments work only for certain childhood brain cancers.

All the data from both brain cancer studies and the CPTAC's investigations of more than a dozen other tumor types are now publicly available on the consortium's Data Portal, notes Liu (see http://cptac-data-portal.georgetown.edu/cptacpublic). “These are resources that other people can test their hypotheses on,” he says. –Elie Dolgin

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