Skip to main content
. 2022 Dec 19;45(1):3–16. doi: 10.1007/s00281-022-00976-y

Table 2.

Cell identity approaches used in glioblastoma brain tumors

Identifying category Feature(s) Implications Proof type
Morphology and location* SVZ contact Poor prognosis Correlative, not functional [108]
Surface epitopes

CD44, CD133,

EGF

GBM stem cells Good evidence, not definitional [109, 110]
Transcription factors POU3F2, SOX2, SALL2, and OLIG2 GBM stem-like, tumor propagating cells Good evidence, functional, not definitional [21]
Functional test or assay Self-renewal, neurosphere formation GBM stem cells Functional, not definitional [111]
Lineage tracing assay Serial xenotransplantation GSCs as tumor initiators Functional [18, 111]
Phospho-protein signaling response

▲ Basal p-STAT5

▲ Basal p-S6

Negative prognostic GBM cells Definitional, functional [25]
DNA mutations IDH1 or IDH2 mutation Metabolically reprogrammed glioma cells Correlative, not functional [112]
Transcription factor mRNA and genetics Signatures including PDGFRA, IDH1, EGFR, and NF1 Classical, mesenchymal, and proneural subtypes ** Correlative, not functional [113]

*Additional features to consider here could include location in tumor core or periphery, association with vasculature, degree of pathology-defined necrosis, and ability to form gap junctions with tumor cells or synapses with neurons. **Classification system based originally on a set of 200 + transcripts, DNA mutation, and copy number status