Severance Hospital said it and two other institutions have jointly developed a biomarker and treatment for predicting the prognosis of glioblastoma, incurable cancer.

Glioblastoma is the most common brain tumor that occurs in the brain. Hospitals conventionally conduct surgery to remove the tumor and follow up with additional therapies such as chemotherapy and radiotherapy after surgery.

A group of researchers said in a paper that they have confirmed new prognostic biomarkers and therapies for glioblastoma. From left are Professors Kim Hyun-seok, Lee Cheol-ju, and Nam Do-hyun,

However, the treatment method still has a very poor prognosis. Wild-type glioblastoma, which accounts for 90 percent of all glioblastomas and has no gene mutation in the isocitrate dehydrogenase (IDH) enzyme, has a poor prognosis and no approved treatments.

To overcome such obstacles, the team led by Professor Kim Hyun-seok, analyzed the samples of 50 glioblastomas by mass spectrometry-based proteomics and obtained the first large-scale glioblastoma protein data. Professor Lee Cheol-ju at Korea Institute of Science and Technology and Professor Nam Do-hyun at Samsung Medical Center participated in the study.

Afterward, the team performed an analysis of drug response on the obtained protein data and genetic transcripts collected from the same sample.

As a result of the analysis, the team was able to classify the IDH wild-type malignant glioblastoma into two GBM Proteomic Cluster (GPC) groups –GPC1 and GPC2-- with regards to different metabolism, immunomodulation, and tumor origin.

In GPC1, the expressions of FKBP Prolyl Isomerase 9 (FKBP9), a malignant prognostic biomarker, and PD-L1, a target of an immune checkpoint inhibitor, was high. The team also confirmed that the characteristics of tumor-origin cells in the group were strong,

Regarding GPC2, the team confirmed that the expression of an excellent prognostic biomarker D-3-phosphoglycerate dehydrogenase (PHGDH) and metabolic oxidative protein expression were high. Consequently, the researchers could suggest targeted therapeutics for each of the two groups.

"For GPC1 tumor-derived cancer cells, treatments such as vistusertib, tandutinib, and crizotinib showed high treatment response, while hedgehog inhibitor erismodegib and the pan-ERBB inhibitor canertinib showed high treatment response for GPC2," the teams said. "Notably, in the PHGDH-negative patient group, predicted to have the worst prognosis, the mTORC1/2 dual inhibitor vistusertib showed a good therapeutic."

Professor Kim said, "This study presented the world's first classification system based on protein information for IDH wild type glioblastoma, one of the representative incurable cancers."

While there needs to be additional clinical verification before applying the theory to real patients, the research is significant in that it simultaneously presented new prognostic biomarkers and therapies for glioblastoma by integrating the analysis of glioblastoma genome-protein data, Kim added.

The journal Nature Communications published the research in its latest issue.

Copyright © KBR Unauthorized reproduction, redistribution prohibited