Interrupting glioblastoma malignant circuitry: plasticity traps, synaptic blockade, and myeloid reprogramming

Purpose of review Despite advances in targeted therapy and immunotherapy, glioblastoma (GBM) remains a therapeutic challenge because of its intrinsic adaptability, driven by intratumoural heterogeneity, cell-state plasticity, tumour–stroma crosstalk, and an immunosuppressive tumour microenvironment. The aggressive mesenchymal subtype, linked to treatment resistance and poor prognosis, exemplifies these adaptive mechanisms. Recent studies identify tractable oncogenic dependencies within these processes, opening routes to subtype-informed, multiaxis therapies. Recent findings GBM exploits extrachromosomal DNA (ecDNA) and structural variants to rewire enhancer networks, reinforcing therapy-resistant mesenchymal states with distinct kinase dependencies (e.g. p38 MAPK signalling, STAT3). Tumour cells further hijack neuronal activity, glutamate/GABA, and long-range neuromodulatory (e.g. cholinergic) inputs, to promote growth, with synaptically enriched regions exhibiting immune suppression and mesenchymal enrichment. Myeloid-derived ligands (e.g. TNF and Oncostatin M) can drive proneural to mesenchymal transition, while Thrombospondin-1-mediated synaptic remodelling suppresses T-cell function, mechanistically coupling neural connectivity to immune evasion. Summary The resilience of GBM arises from the interplay of epigenetic plasticity, neural circuit co-option, and myeloid-skewed immunosuppression. A coordinated strategy, kinome-guided targeting plus circuit disruption and myeloid reprogramming, offers a credible path to contain adaptation and improve outcomes.