Modeling glioma intratumoral heterogeneity with primary human neural stem and progenitor cells.

Return to Grants

Publication Year:
2025
Authors:
PubMed ID:
40749669
Funding Grants:
Public Summary:
Gliomas are a type of brain tumor notorious for being difficult to treat, largely because of a phenomenon called intratumoral heterogeneity — meaning that within a single tumor, the cancer cells aren't all alike. Different cells within the same tumor can behave differently, respond differently to treatment, and even resemble different stages of normal brain development. This internal diversity is a major reason gliomas tend to resist therapy so effectively. Scientists have observed that glioma cells often mimic the natural hierarchy of neural stem cells, the cells responsible for generating different brain cell types during normal development. But it hasn't been fully clear how a tumor's diversity is shaped by two key factors: which specific cell type the cancer originally arose from, and which genetic mutations are driving its growth. To investigate this, researchers built gliomas from the ground up in a controlled laboratory setting. They started with three types of neural stem and progenitor cells purified from human brain tissue at the mid-point of fetal development: "tripotent" neural stem cells, which can become several different types of brain cells; "bipotent" glial progenitor cells, which can become two cell types; and "unipotent" oligodendrocyte progenitor cells, which are more specialized and can only become one particular cell type. Because all three cell types came from the same genetic source, the researchers could isolate the effect of the starting cell type itself, separate from any genetic differences between individuals. They then introduced specific combinations of well-known cancer-driving genetic mutations (including TP53, NF1, CDK4, EGFR, and PDGFRA) into each cell type and transplanted the modified cells into mice to see what kinds of tumors would develop. The results revealed a clear connection between where a tumor starts and how it develops. Tumors that originated from the more specialized oligodendrocyte progenitor cells contained a higher proportion of mature, oligodendrocyte-like cancer cells, resembling a lower-grade, less aggressive form of glioma seen in patients. Meanwhile, specific genetic mutations pushed tumors toward distinct identities regardless of starting cell type: the CDK4 mutation drove tumors to develop a neuron-like character, while the EGFR mutation pushed tumors toward resembling glial progenitor cells. Beyond these specific findings, the researchers highlight that their experimental system itself is a valuable new tool. Because it's built in a modular, adaptable way, other scientists can use it to systematically test different combinations of starting cell types and mutations, helping to map out, piece by piece, how gliomas develop their notorious internal diversity. Understanding these origins matters because a tumor's diversity is a major driver of why treatments often fail or why cancers recur. By clarifying how specific starting cells and mutations shape a tumor's makeup, this research lays groundwork that could eventually help doctors better predict how a given patient's glioma will behave, and could inform the development of more precisely targeted treatments.
Scientific Abstract:
Gliomas are notorious for their intratumoral heterogeneity, which drives therapy resistance. Glioma tumor cells mimic a neural stem cell (NSC) hierarchy reminiscent of normal brain development. How intratumoral heterogeneity is shaped by cell-of-origin and various driver mutations is not fully understood. We develop a model of glioma initiation using neural stem and progenitor cells (NSPCs) purified from midgestational human brain tissue, including tripotent NSCs, bipotent glial progenitor cells (GPCs), and unipotent oligodendrocyte progenitor cells (OPCs). We transduced these isogenic lines with defined combinations of oncogenic drivers (TP53, NF1, CDK4, EGFR, and PDGFRA) and transplanted them into mice. We find that OPC-derived tumors harbored a higher proportion of differentiated oligodendrocyte-like cells, reminiscent of low-grade oligodendrogliomas. CDK4 drove a neuron-like subtype, while EGFR drove a GPC-like subtype. Our platform is highly adaptable and allows for modular and systematic interrogation of how cell-of-origin and specific driver mutations shape the tumor landscape.