The most striking of these discoveries concerns synapses, the specialised junctions through which neurons communicate. It is now well established that gliomas, the most common primary brain tumours, form bona fide functional synapses with neighbouring neurons. In landmark work published in Nature in 2019, Venkataramani and colleagues showed that glutamatergic synaptic input drives calcium fluxes in glioma cells and accelerates tumour progression. Subsequent studies revealed that gliomas hijack both excitatory signalling through AMPA and NMDA receptors and, remarkably, inhibitory signalling through GABA-A receptors, as demonstrated by Barron and colleagues in 2025 for diffuse midline gliomas. In other words, the tumour does not merely sit near neural circuitry; it integrates into it, exploiting the brain’s own communication hardware to sustain its proliferation.

What makes the new review consequential is its argument that this phenomenon extends far beyond the brain. Sakthivelu, Savchuk, Barron and their co-authors reported in Nature in 2025 that small cell lung cancer, a tumour with neuroendocrine origins, forms functional synapses with cortical and vagal neurons. A companion paper in the same issue described neuronal activity-dependent mechanisms of small cell lung cancer pathogenesis, showing that the electrical behaviour of the nervous system itself can shape tumour behaviour. For a disease that kills most patients within two years of diagnosis, the suggestion that tumour cells literally plug into neural circuits opens an entirely new axis of investigation, from electrophysiology in the clinic to drugs originally designed for neurological disease.

Brain metastases add a further layer of complexity. Breast and lung cancers that spread to the brain exploit glutamatergic signalling in ways that echo the glioma literature, suggesting that the neural niche is a generalised feature of cerebral tumour colonisation rather than a peculiarity of primary brain tumours. Work by Nejo and colleagues published in Nature Communications in 2025 showed that gliomas remodel neuronal circuitry, inducing regional functional connectivity that is associated with tumour invasion. The tumour, in this view, is not simply a lesion within the brain but a participant in brain networks, capable of co-opting the plasticity mechanisms that normally underlie learning and memory. Taylor and colleagues made this explicit in 2023, showing that glioma synapses recruit mechanisms of adaptive plasticity.

Synaptic communication is only one channel in a broader neuro-tumour dialogue. The review emphasises paracrine signalling, in which neurotransmitters released into the tumour microenvironment act as growth factors. A classic example is neuroligin-3, a synaptic adhesion protein whose secretion by neurons in response to neural activity promotes glioma proliferation, as shown by Venkatesh and colleagues in Cell in 2015. Beyond the central nervous system, tumours actively recruit innervation. In pancreatic cancer, neural invasion is a hallmark of the disease and a driver of its notorious pain and aggressiveness, as reviewed by Demir and colleagues. In prostate cancer, Magnon and colleagues demonstrated in Science in 2013 that autonomic nerve development contributes to tumour progression, with sympathetic and parasympathetic fibres playing distinct roles at different stages. Ayala and colleagues had earlier documented cancer-related axonogenesis and neurogenesis in prostate tumours, establishing that new nerve growth within tumours is not incidental but tumour-promoting.

The autonomic nervous system emerges as a particularly clinically relevant player. The balance between beta-adrenergic tone, the fight-or-flight arm of the autonomic system, and vagal tone, its calming counterpart, appears to carry prognostic weight across multiple cancers. Chronic stress, which floods the body with catecholamines, has been shown in experimental models to accelerate tumour growth: Thaker and colleagues reported in Nature Medicine in 2006 that chronic stress promotes tumour growth and angiogenesis in a mouse model of ovarian carcinoma, while Partecke and colleagues showed that adrenergic signalling drives experimental pancreatic cancer growth. On the human side, a population-based cohort study and meta-analysis by Løfling and colleagues in 2022 examined beta-blockers and breast cancer survival by molecular subtype, and Giese-Davis and colleagues found in 2015 that higher vagal activity was associated with survival in patients with advanced breast cancer. A 2025 systematic review and meta-analysis by Huang and colleagues in BMC Cancer consolidated the evidence linking vagal nerve activity to cancer prognosis.

Perhaps the most provocative section of the review concerns the intersection between neural signalling and classical drug-resistance pathways. The authors argue that stress-related and neural signalling cooperates with canonical resistance circuits, including the p53-EGFR-ERK signalling axis, P-glycoprotein-mediated drug efflux, reactive oxygen species-dependent redox programmes, and adipokine-Hsp90 axes, to blunt the efficacy of chemotherapy, targeted agents and immune checkpoint inhibitors. The molecular logic is plausible: beta-adrenergic signalling activates cAMP-dependent pathways that can modulate ERK activity and redox balance, while obesity-associated adipokines such as leptin and resistin have been shown by Malvi and colleagues to impair the efficacy of dacarbazine in melanoma under obese conditions. If neural inputs feed into the same downstream circuits that tumours use to survive cytotoxic pressure, then the nervous system is not merely promoting growth but actively teaching tumours how to resist our drugs.

Immune therapy may be particularly vulnerable to this crosstalk. Baruch and colleagues reported in Nature in 2025 that cancer-induced nerve injury promotes resistance to anti-PD-1 therapy, a finding that directly links neural damage to the failure of modern immunotherapy. Sympathetic signalling is also known to regulate the tumour microenvironment’s immune composition, as reviewed by Cole and colleagues in Nature Reviews Cancer in 2015, and beta2-adrenergic agonists have been shown to bias dendritic cell responses towards an IL-17 immune profile. These observations suggest that the nervous system modulates not only tumour cells but also the immune cells charged with eliminating them, adding a neuroimmunological dimension to cancer treatment that has been largely absent from clinical trial design.

The therapeutic implications are tantalising but the review is careful about the caveats. Drugs that modulate neural signalling already exist: memantine, an NMDA receptor antagonist used for dementia, has been tested as adjunctive therapy in metastatic colon cancer in a pilot randomised trial, and riluzole, an AMPA receptor modulator used in amyotrophic lateral sclerosis, has been proposed as a potential anti-cancer agent. Beta-blockers are cheap, widely available and supported by observational data in breast cancer. But the brain and peripheral nerves are not dispensable organs, and indiscriminate suppression of neurotransmission carries obvious neurological costs. The authors stress that translation will require attention to specificity, neurological safety, and rational combination with cytotoxic, targeted and immune therapies rather than repurposing neurological drugs as standalone cancer treatments.

What the review ultimately delivers is a reframing. Cancer, in this synthesis, is a systemic disease whose behaviour is shaped by reciprocal interactions with the nervous system, from synaptic contacts in gliomas and small cell lung cancer to autonomic tone in breast and pancreatic malignancies. The 2022 update of the hallmarks of cancer by Douglas Hanahan acknowledged neuronal interactions among the emerging dimensions of tumour biology, and the evidence base has grown rapidly since. Molecular imaging of neurotransmitter receptors in oncology is advancing, and clinical trials of neuromodulatory agents are beginning to test whether the nervous system can be disconnected from the tumours it feeds. If even a fraction of the preclinical promise survives clinical scrutiny, the oncologist’s toolkit of the coming decade may include drugs borrowed from the neurologist, prescribed not for the mind but for the malignant conversations taking place between nerves and tumours.

Subject of Research: Neuron-tumour interactions, including synaptic and autonomic neural signalling that promotes cancer growth and drug resistance

Article Title: Central nervous system and cancer: mechanistic insights and therapeutic opportunities CNS and cancer: a review

Article References: Petrelli, F., Callea, A., Ghidini, A., Carpo, M., Barbuto, M., Gambini, D., Mailland, E., & Ferraro, B. (2026). Central nervous system and cancer: mechanistic insights and therapeutic opportunities CNS and cancer: a review. Medical Oncology, 43(10), Article 266. https://doi.org/10.1007/s12032-026-03365-9

Image Credits: AI Generated

DOI: 10.1007/s12032-026-03365-9

Keywords: neuron-tumour interactions, glioma, small cell lung cancer, synapses, glutamatergic signalling, autonomic nervous system, beta-blockers, drug resistance, tumour microenvironment, brain metastases, vagal tone, immunotherapy resistance