The authors organise the resistance problem into four interlocking dimensions. The first is physical and cellular: the blood–brain barrier and the blood–brain tumour barrier restrict what enters the brain, while once inside, tumour cells deploy ATP-binding efflux pumps that expel chemotherapeutics, sequester drugs in subcellular compartments where they cannot act, and metabolically inactivate agents before they reach their targets. The net effect is that intracellular concentrations of effective drug fall below therapeutic thresholds even when patients tolerate full systemic doses. Temozolomide, the backbone of glioblastoma care, illustrates the point vividly: the DNA-repair enzyme MGMT can directly reverse the drug’s methylation damage, and many tumours either express MGMT abundantly or acquire that capacity under treatment pressure.

The second dimension is molecular escape. Tumour genomes are not static; they repair DNA damage, mutate drug targets, amplify oncogenic drivers on extrachromosomal DNA — circular fragments of genetic material that exist outside the chromosomes and can carry dozens of copies of resistance genes — and rewire RNA splicing to produce protein variants that no longer recognise the drug. Extrachromosomal DNA has attracted particular attention because it can rapidly change copy number as selective pressure shifts, allowing tumour cell populations to dial oncogene expression up or down within a few cell divisions. Aberrant splicing similarly generates isoforms of kinases and receptors that lack the binding domains targeted by small-molecule inhibitors, rendering otherwise potent drugs blind to their intended substrate.

Third, the review catalogues intrinsic cellular drivers of resistance: cancer stem cells that persist in a dormant, quiescent state impervious to agents that kill dividing cells; subtype switching, in which tumour populations shift their transcriptional identity toward a more aggressive or less drug-sensitive state; epigenetic remodelling that silences pro-apoptotic genes; metabolic reprogramming that reroutes biosynthetic and energy pathways; and a broad resistance to multiple modes of cell death, not merely apoptosis. A glioblastoma cell that has downregulated its apoptotic machinery will survive a therapy designed to trigger it, which is why the authors emphasise that future treatments must engage alternative death programmes such as ferroptosis — iron-dependent lipid peroxidation — and pyroptosis, an inflammatory form of regulated necrosis.

The fourth dimension is the tumour microenvironment and its neural connections. Hypoxic regions reduce the efficacy of radiotherapy and many drugs; immunosuppressive cell populations, including myeloid-derived suppressor cells and polarised tumour-associated macrophages, blunt immune-based approaches; and exosomes shuttle resistance-conferring proteins and nucleic acids between cells. Perhaps most strikingly, the review highlights intertumoural neural-like physical networks — tumour cells connected by membrane tubes that share cytoplasmic content — and functional connections with host neurons, which facilitate both the spread of toxic signals and malignant regeneration after treatment. Brain tumours, in this view, are not merely collections of resistant cells but integrated tissues wired into the nervous system itself.

Against this formidable backdrop, the authors argue that nanomedicines offer a uniquely multidimensional counter-strategy. Nanoparticles can be engineered to cross the blood–brain barrier through receptor-mediated transcytosis, using ligands such as apolipoprotein E, transferrin or cyclic RGD peptides that hijack natural transport routes. Once across, they can be decorated with targeting moieties that promote uptake by tumour cells while bypassing efflux pumps, since particles are internalised intact rather than recognised as soluble substrates. Co-delivery is another central advantage: a single carrier can ferry a small-molecule inhibitor together with nucleic acid drugs — small interfering RNAs, microRNAs or even CRISPR components — to silence resistance pathways such as MGMT, PI3K signalling or Bcl2L12 at their genetic source, achieving combinations that would be pharmacologically difficult with free drugs.

The review then dissects five categories of nanosystems. Lipid-based carriers, including liposomes, solid lipid nanoparticles and nanostructured lipid carriers, are the most clinically mature; nanoliposomal irinotecan has already reached patients with glioma, and rhenium-186 nanoliposomes have been explored for brachytherapy-like localised irradiation. Polymeric nanoparticles built from PLGA, polycaprolactone or polyethyleneimine offer controlled release and robust nucleic acid loading. Inorganic platforms — mesoporous silica, gold nanostructures, carbon dots and magnetic iron oxide nanoparticles — enable physicochemical interventions such as photothermal therapy and photodynamic therapy, which use near-infrared light to generate lethal heat or reactive oxygen species and can inactivate thermotolerant, drug-resistant proteins that ordinary chemotherapy cannot touch. Hybrid lipid–polymer nanoparticles combine the stability of polymer cores with the biocompatibility of lipid shells, while bioderived and biomimetic systems — extracellular vesicles, cell-membrane-coated particles and bacterial outer membrane vesicles — exploit natural stealth properties to evade immune clearance and home to tumour tissue.

The authors are careful to temper enthusiasm with a candid accounting of defects. Carrier toxicity remains a concern, particularly for cationic materials such as polyethyleneimine; in vivo stability is undermined by protein corona formation and clearance by the reticuloendothelial system; scale-up from laboratory synthesis to Current Good Manufacturing Practice production is non-trivial for complex, multi-component particles; and clinical translation has been slowed by inconsistent targeting efficiency in human tumours, heterogeneity of the blood–brain tumour barrier across patients, and the difficulty of demonstrating intratumoural drug delivery non-invasively. Strategies for improvement discussed in the review include biomimetic camouflage, stimuli-responsive release triggered by tumour acidity or glutathione levels, microenvironment remodelling to normalise vasculature and relieve hypoxia, and switching cell death modality from apoptosis toward ferroptosis and pyroptosis to defeat apoptosis-resistant clones. Convection-enhanced delivery, which infuses therapeutic agents directly into brain tissue under pressure gradients, is presented as a complementary route that sidesteps the barrier entirely.

Ultimately, the review positions nanomedicine not as a single magic bullet but as a platform for rational, multi-pronged assault on resistance — breaching barriers, evading pumps, silencing escape pathways, applying physical energy, remodelling the microenvironment and reprogramming how tumour cells die. The authors frame their synthesis as a reference for both basic researchers designing the next generation of carriers and clinicians weighing which nanotherapeutic strategies are closest to meaningful clinical impact. For patients with resistant CNS tumours, whose options remain tragically narrow, the message is cautiously hopeful: the tools to systematically reverse drug resistance are being assembled, and the decisive question now is whether the field can carry them across the final, most difficult barrier of all — the distance from the laboratory bench to the clinic.

Subject of Research: Nanomaterial-based strategies to overcome therapeutic resistance in central nervous system tumours

Article Title: Nanomedicines for resistant tumours in the central nervous system: novel materials and mechanisms of action

Article References: Li, G., Chan, Y.-T., Xiong, F., Xin, Y., Hong, R., Xie, Z., Zhou, W., Zhang, C., Zhang, R., Wang, N., & Huang, Q. (2026). Nanomedicines for resistant tumours in the central nervous system: novel materials and mechanisms of action. Molecular Cancer. https://doi.org/10.1186/s12943-026-02783-7

Image Credits: AI Generated

DOI: 10.1186/s12943-026-02783-7

Keywords: nanomedicine, CNS tumours, blood-brain barrier, drug resistance, glioblastoma, ferroptosis, pyroptosis, lipid nanoparticles, biomimetic nanosystems, tumour microenvironment, clinical translation, chemoradiotherapy resistance