SBTF Brain Tumor Research Grant Outcome Report

The Foundation received this update from SBTF 2017 Ann Faulkenberry Memorial Award Research Grant Recipients Dr. Ihrie and Dr. Irish of Vanderbilt University. They continue to make advancements in the treatment of brain tumors and search for a cure. You can review all grants awarded by the SBTF and the clinician’s updates on our Impact page.

Project Title: “Dissecting the Contribution of the Niche and Immune Cells to patient outcome using single cell protein measurements”

Principal Investigators: Rebecca Ihrie, Ph.D. and Jonathan Irish, Ph.D. – Vanderbilt University

Funding from the SBTF to the Ihrie and Irish groups allowed us to leap forward in our studies of primary brain tumor tissue using single cell approaches. With the support of this award, we developed a suite of bench and computational tools, used these to reveal previously unrecognized cells within brain tumor tissue, and identified novel targets in these cells.

Our approach focused on measuring many proteins at the same time in millions of tumor cells. This approach revealed types of cells that have not been observed in healthy brains. Out of all the new cell types seen, there were two that stood out – one that was seen only in very aggressive tumors, and another that was a hallmark of tumors with much better than average outcomes. We believe these findings, which were reported in eLife in 2020, represent a fundamental advance in our understanding of glioblastoma. No prior work had resolved glioblastoma cell types that were so closely linked to patient outcomes. We named the two new cell types based on the clinical outcomes of tumors that contain them. Perhaps the most important cell type found is the Glioblastoma Negative Prognostic (GNP) cell subset, which is found in the most aggressive glioblastomas. These GNP cells are the ‘worst of the worst’ within this terrible disease.


Once we find a new cell type like this, the next questions are 1) what defined these cells and 2) how do we target them for destruction? Unlike cells in the brains of healthy adults, the glioblastoma cells we identified appear to hack into multiple signaling pathways and turn them all on at the same time. In normal adult brain cells, only one of the pathways studied would be active at any given time. We think these hacked signaling networks not only mark the aggressive cancer cells, we think they are necessary in those cells. Thus, these cells may have a critical weakness – they may depend on keeping all these signals active at once. One of the important next steps on this project is to find treatments that target both pathways to kill the GNP cells while sparing healthy cells. We think this is likely, since the pattern of signals seen in GNP cells is not seen in healthy human brain cells.


Although our main focus continues to be glioblastoma, the computational tool we devised to identify these cells can be used across many data types and diseases to find cells whose abundance predicts especially good or bad outcomes. So, while we are applying the machine learning algorithms to brain tumors, we are also collaborating with groups applying them to other cancer types and immune diseases. Notably, all the data, tools, and computer code generated is freely available to the glioblastoma research community and other researchers, and we have incorporated teaching on these tools into annual workshops taught worldwide.”

Dr. Ihrie, Ph.D., Vanderbilt University


One of the main articles supported by this award can be freely accessed here: https://elifesciences.org/articles/56879
Popular coverage of the work and descriptions of our broader collaboration can be found here: https://news.vanderbilt.edu/2020/06/24/discovery-of-aggressive-cancer-cell-types-by-vanderbilt-researchers-made-possible-with-machine-learning-techniques/
https://www.healio.com/news/hematology-oncology/20200807/algorithm-identifies-riskstratifying-glioblastoma-tumor-cells
https://medschool.vanderbilt.edu/vanderbilt-medicine/profiles-in-discovery/