Summary: Researchers at the University of Toronto have identified actionable vulnerabilities in glioblastoma cancer stem cells, offering a promising route for more effective treatments against this aggressive brain tumor. By screening patient-derived stem cell lines with CRISPR technology, the team mapped two primary cell states that drive tumor growth—each with distinct genetic weaknesses that could be targeted with drugs.
This work points to strategies that target both major stem cell subtypes simultaneously or tailor treatments to the subtype dominant in a patient’s tumor, with the goal of preventing recurrence and improving outcomes. The study represents the largest CRISPR fitness screen performed on patient-derived glioblastoma stem cells to date and highlights candidate targets such as OLIG2, MEK, FAK and β1-Integrin.
Key Facts:
- Glioblastoma tumor growth is driven by two principal cancer stem cell states: a developmental-like state and an injury-response (inflammatory) state.
- Distinct genetic dependencies for each subtype suggest druggable targets, including OLIG2 and MEK for the developmental state and FAK and β1-Integrin for the injury-response state.
- The findings derive from the largest CRISPR/Cas9 fitness screen performed across patient-derived glioblastoma stem cell cultures (30 patients).
Source: University of Toronto
Overview
A team led by researchers at the University of Toronto used targeted CRISPR screens to uncover state-specific vulnerabilities in glioblastoma stem cells (GSCs). Glioblastoma is the most common and deadliest primary brain tumor in adults. Its tendency to recur after standard therapy is driven in large part by GSCs that survive treatment and later regenerate treatment-resistant tumors.
Glioblastoma tumors are highly heterogeneous—both between patients and within individual tumors—making it difficult for one-size-fits-all targeted therapies to succeed. The University of Toronto team characterized this heterogeneity at the level of GSCs and found that these cells span a gradient between two dominant transcriptional states: a developmental-like state that resembles aberrant neurodevelopment and an injury-response state marked by inflammatory programs.
Recognizing these two states allowed researchers to search for conserved and subtype-specific genetic dependencies that are present across a diverse set of patient-derived GSC cultures. Identifying such dependencies creates opportunities to pair inhibitors tailored to each state or to combine drugs that simultaneously attack both kinds of stem cells within a tumor.
To achieve this, the team designed a focused gRNA library (GBM5K) and conducted CRISPR/Cas9 fitness screens in 30 patient-derived glioblastoma stem cell cultures—the largest study of its kind. These cultures were generated in the lab of Peter Dirks, professor of surgery and molecular genetics and Chief of the Division of Neurosurgery at SickKids, and better represent true patient tumor biology than many conventional immortalized cell lines.
From these screens, the researchers identified genes required for growth in each GSC subtype. Developmental-state GSCs showed dependency on transcriptional regulators involved in neurodevelopment, with OLIG2 and MEK emerging as prominent targets. Injury-response GSCs relied on integrin and focal adhesion signaling, highlighting vulnerabilities to inhibitors of β1-Integrin and FAK. The screens also revealed subtype-specific differences in cyclin D reliance (CCND1 versus CCND2), offering additional mechanistic insight.
These subtype-specific dependencies translated into differential drug sensitivities in experimental tests, supporting the idea that targeting the dominant stem cell state, or both states together, may produce more durable clinical responses and reduce the chance of tumor relapse.
The study’s co-first authors, Graham MacLeod and Fatemeh Molaei, emphasize that using directly patient-derived GSC lines provides a clearer picture of clinically relevant vulnerabilities than traditional lab-adapted cell lines. Stéphane Angers, principal investigator and director of the Donnelly Centre, notes that the clinic has not yet addressed differences between GSC subtypes and that these findings could inform subtype-tailored therapies or combination approaches in the future.
Funding: This research was supported by the Canadian Institutes of Health Research.
About this brain cancer research news
Author: Anika Hazra
Source: University of Toronto
Contact: Anika Hazra – University of Toronto
Image: Image credit: Neuroscience News
Original Research: Closed access. “Fitness Screens Map State-Specific Glioblastoma Stem Cell Vulnerabilities” by Graham MacLeod et al., published in Cancer Research. DOI: 10.1158/0008-5472.CAN-23-4024
Abstract
Fitness Screens Map State-Specific Glioblastoma Stem Cell Vulnerabilities
Glioblastoma (GBM) is driven by self-renewing glioblastoma stem cells (GSCs) that persist after therapy and seed treatment-refractory recurrent tumors. GBM tumors display extensive intra- and inter-tumoral heterogeneity, which extends to GSCs existing along a transcriptional gradient between developmental and injury-response states. Effective treatments require drug targets specific to each state.
To identify conserved and subtype-specific genetic dependencies across a heterogeneous panel of GSCs, the authors designed the GBM5K targeted gRNA library and performed fitness screens in 30 patient-derived GSC cultures. The focused CRISPR screens identified the most conserved subtype-specific vulnerabilities and elucidated the functional dependency gradient between developmental and injury-response states.
Developmental-specific fitness genes were enriched for neurodevelopmental transcriptional regulators, whereas injury-response-specific fitness genes highlighted genes involved in integrin and focal adhesion signaling. These context-specific vulnerabilities produced differential sensitivity to inhibitors of β1 integrin, FAK, MEK and OLIG2. The screens also revealed subtype-driven differences in cyclin D dependency (CCND1 vs. CCND2).
These data provide biological and mechanistic insight into GBM heterogeneity and suggest routes for precision targeting of defined GBM and GSC subtypes to overcome therapeutic resistance and tumor recurrence.