Glioblastoma (GBM) Cell Models for Research

Glioblastoma (GBM) is the most common and aggressive primary malignant brain tumor in adults, accounting for approximately 48.6% of all malignant brain tumors (CBTRUS 2023). The global incidence is about 3.2 per 100,000 person-years (WHO Classification of Tumors of the Central Nervous System, 5th edition, 2021). Despite standard-of-care therapy (maximal surgical resection, radiotherapy, and temozolomide), the median overall survival is only 14-16 months, and the 5-year survival rate is less than 7% (NCI SEER data, 2024). Risk factors include exposure to ionizing radiation and rare genetic syndromes (e.g., Li-Fraumeni, neurofibromatosis type 1), but the majority of cases are sporadic with no identifiable cause. The disease is universally fatal, and there is an urgent need for novel therapeutic targets and precision medicine approaches.

GBM is characterized by extensive intratumoral heterogeneity, including diverse cell states (neural progenitor-like, oligodendrocyte progenitor-like, astrocyte-like, mesenchymal-like) and a highly immunosuppressive tumor microenvironment. This complexity makes GBM an ideal model for studying tumor evolution, therapy resistance, and the role of cancer stem cells. Public datasets such as The Cancer Genome Atlas (TCGA) and the Chinese Glioma Genome Atlas (CGGA) provide extensive multi-omics data, enabling researchers to correlate molecular subtypes with clinical outcomes. Key open questions include: How do specific genetic alterations drive tumor initiation and progression? What mechanisms underlie resistance to temozolomide and radiotherapy? How can we target the immunosuppressive microenvironment? Gene-edited cell models are essential tools to address these questions by enabling precise manipulation of candidate genes in isogenic backgrounds.

GBM pathogenesis involves the dysregulation of several core signaling pathways. The most frequently altered pathways are:

• • RTK/RAS/PI3K signaling: Activation of receptor tyrosine kinases (EGFR, PDGFRA) leads to downstream activation of RAS and PI3K/AKT/mTOR, promoting cell proliferation and survival.
• • TP53 pathway: Mutations in TP53 or amplification of MDM2/MDM4 disrupt cell cycle arrest and apoptosis.
• • RB pathway: Loss of RB1 or amplification of CDK4/6 leads to uncontrolled cell cycle progression.
• • Wnt signaling: Aberrant activation of the Wnt/β-catenin pathway contributes to stemness and therapy resistance.

These pathways are not mutually exclusive; many GBM tumors harbor alterations in multiple pathways, reflecting the molecular heterogeneity of the disease.

Based on TCGA and COSMIC data, the following genetic alterations are most common in GBM:

Data from TCGA (Cancer Genome Atlas Research Network, Nature 2008; 455:1061-1068) and COSMIC (v100, 2024).

The molecular pathogenesis of GBM is driven by the interplay of multiple signaling networks:

• • PI3K/AKT/mTOR pathway: Key nodes include PI3K (PIK3CA, PIK3R1), AKT, PTEN, and mTOR. This pathway regulates cell growth, survival, and metabolism.
• • MAPK/ERK pathway: RAS (KRAS, NRAS, HRAS) activates RAF/MEK/ERK, promoting proliferation and differentiation.
• • p53 pathway: TP53, MDM2, MDM4, and ATM/ATR are central to DNA damage response and apoptosis.
• • RB pathway: RB1, CDK4/6, and cyclin D1 control the G1/S transition.
• • Wnt/β-catenin pathway: CTNNB1, APC, and GSK3β regulate stemness and invasion.
• • Notch pathway: NOTCH1-4 and DLL/JAG ligands are involved in cancer stem cell maintenance.

These networks are highly interconnected, and their dysregulation contributes to the aggressive phenotype of GBM. Gene-edited models targeting these nodes are critical for dissecting their functional roles.

Common GBM cell lines used in research include:

These cell lines have been extensively characterized by the Broad Institute's Cancer Cell Line Encyclopedia (CCLE) and DepMap. However, 2D cell lines do not fully recapitulate the 3D tumor microenvironment. Patient-derived organoids (PDOs) and glioblastoma stem-like cells (GSCs) are increasingly used to better model tumor heterogeneity and drug response. Organoids can be cultured from patient tumor samples and retain key genetic and phenotypic features, making them valuable for preclinical drug testing.

Animal models are essential for studying GBM in vivo. The main types include:

• • Patient-derived xenografts (PDX): Tumor cells from patients are implanted into immunodeficient mice. PDX models preserve the genetic and phenotypic heterogeneity of the original tumor and are used for drug efficacy testing.
• • Genetically engineered mouse models (GEMM): Mice with specific genetic alterations (e.g., EGFRvIII, TP53 loss, PTEN loss) develop GBM-like tumors. GEMMs allow study of tumor initiation and progression in an immunocompetent environment.
• • Induced models: Use of viral vectors (e.g., RCAS/tv-a) or CRISPR to introduce oncogenic mutations in specific brain regions. These models offer temporal and spatial control of tumor development.

Each model has advantages and limitations. PDX models are more clinically relevant but lack a functional immune system. GEMMs are useful for studying immune interactions but are time-consuming to generate.

CRISPR-based gene editing has revolutionized the generation of isogenic cell models for GBM research. By introducing precise knockouts (e.g., TP53, PTEN, EGFR) or knock-ins (e.g., EGFRvIII, IDH1 R132H) into a common parental cell line, researchers can directly assess the functional impact of specific mutations. These isogenic pairs eliminate confounding genetic background effects, enabling robust genotype-phenotype correlations. Commercially available, sequence-verified gene-edited cell lines are now widely used for:

• • Validating driver genes identified in genomic studies.
• • Studying drug resistance mechanisms (e.g., MGMT knockout to mimic methylation status).
• • Developing reporter lines (e.g., GFP-tagged proteins) for live-cell imaging.
• • Creating synthetic lethal models for targeted therapy screening.

These models are essential for translating genomic discoveries into therapeutic strategies.

Gene-edited cells are used to systematically validate genes implicated in GBM. For example, CRISPR knockout of TP53 in U87MG cells (which are TP53 wild-type) leads to increased proliferation and resistance to apoptosis, confirming its tumor suppressor role. Similarly, knock-in of EGFRvIII (a constitutively active EGFR mutant) into U87MG cells enhances cell migration and invasion, recapitulating the aggressive phenotype. These models allow researchers to study gene function in a controlled environment and to identify downstream effectors via transcriptomic or proteomic profiling.

Isogenic cell pairs are invaluable for drug screening. For instance, a PTEN-null U251MG cell line can be compared to PTEN-wild-type U251MG to identify compounds that selectively kill PTEN-deficient cells (synthetic lethality). Similarly, MGMT knockout cells are used to study temozolomide resistance, as MGMT promoter methylation is a key predictor of response. By generating resistance models through chronic drug exposure, researchers can identify mechanisms of acquired resistance and test combination therapies.

CRISPR screens using gene-edited cells can identify novel biomarkers and therapeutic targets. For example, a genome-wide CRISPR knockout screen in GBM cells treated with temozolomide can reveal genes whose loss sensitizes cells to the drug, providing potential biomarkers for patient stratification. Additionally, synthetic lethality screens can identify vulnerabilities in specific genetic backgrounds, such as EGFR-amplified tumors, leading to targeted therapy opportunities.