Cancer Neuroscience: The Hidden Network Between Tumors and the Nervous System
In this article
An Unexpected Discovery: Neuron-Cancer Synapses
A central emphasis of the review was the neuron-cancer synapses demonstrated in glioblastoma and diffuse midline glioma (DMG). These were bona fide chemical synapses — neurons released glutamate, while glioma cells received this signal through calcium-permeable AMPA receptors (GluA2Q) expressed on their postsynaptic membranes and converted it into intracellular calcium waves. Neuronal activity thus became a direct growth stimulus within the tumor cell network.
Subsequent studies revealed a much broader picture. Gliomas were found to integrate with corticostriatal and cholinergic projections in the brain (via the muscarinic M3 receptor); GABAergic interneurons paradoxically produced depolarization in DMG (because of the high intracellular chloride concentration in tumor cells — a neurodevelopmental feature); and long-range projections were mapped using retrograde rabies-virus tracing technologies. Serotonergic neurons provided an exception to the rule: in a rare ependymoma subtype (RELA fusion-positive), histone serotonylation triggered tumor differentiation and slowed growth — meaning neural signals do not always favor the tumor.
Even more strikingly, this synaptic phenomenon is not confined to the brain. Small cell lung cancer (SCLC), breast cancer cells that metastasize to the brain, and melanoma cells also receive glutamatergic synaptic input from neurons. In pancreatic cancer, putative synapses between sensory neurons and tumor cells operating through NMDA-type receptors (Glun1/Glun2D) were described; disrupting Glun2D markedly reduced metastasis in mouse models.
interaction mechanisms
(5 established + 5 emerging)
synapses
(a turning point for the field)
drugs — repurposing
candidates
sleep disorders
in cancer patients
A Roadmap of Ten Mechanisms
The review divides the field into two main axes: five established mechanisms and five emerging areas. The diagram below summarizes these ten headings and their interrelationships.
Figure 1. A Ten-Mechanism Map of Cancer Neuroscience: Established and Emerging Areas
Source: compiled according to the Cell 2026 review by Winkler F. et al.; the diagram is original.
How Does the Nervous System Feed Cancer?
In tumors outside the brain (extracranial cancers), autonomic and sensory fibers accelerate growth through neurotransmitters and neuropeptides released by paracrine signaling. In prostate cancer, sympathectomy or loss of β2/β3-adrenergic receptors prevents tumor initiation, while cholinergic M1 signaling triggers spread. In breast cancer, β-adrenergic signaling increases metastasis; in gastric cancer, vagal inputs activate the tuft-cell acetylcholine (ACh) → nerve growth factor (NGF) axis, fueling YAP-driven growth. In pancreatic cancer, β-adrenergic signaling increases axonogenesis and lethality through a β2-adrenergic–neurotrophin feedback loop — a loop that can be attenuated by β-blockers or Trk inhibitors.
Perhaps one of the most surprising findings is that neurons also provide direct metabolic support to cancer cells. In pancreatic cancer, sensory and sympathetic axons were shown to supply serine to tumor cells in a serine-poor environment; in breast cancer, functional mitochondria were transferred from central nervous system (CNS) and dorsal root ganglion neurons to cancer cells, increasing oxidative phosphorylation and metastatic potential. In other words, neurons do not merely send signals: they also share physical organelles.
The Nervous and Immune Systems: An Unexpected Partnership in Crime
One of the review's most striking sections addresses neuronal effects on immune control. Most findings point in the same direction: neuronal signals suppress antitumor immunity and reduce the efficacy of immune-checkpoint inhibitors (ICI). In cutaneous squamous cell carcinoma, cancer-induced nerve injury (CINI) produces resistance to PD1 blockade. In melanoma, CGRP (calcitonin gene-related peptide) secreted by nociceptive neurons leads to exhaustion of intratumoral CD8+ T cells. In head and neck squamous cell carcinoma, nociceptive neurons in tumor-draining lymph nodes create local immunosuppression — targeting CGRP can enhance the ICI response.
Perhaps the most practical implication in this field concerns β-blockers. Exhausted CD8+ T cells overexpress the β1-adrenergic receptor ADRB1; catecholamine exposure suppresses their TNF and IFN-γ production and proliferation. The nonselective β-blocker propranolol shows synergy with the anti-CTLA4/anti-PD1 combination in an orthotopic pancreatic cancer model. A drug used for decades in cardiology and thyroid disease could therefore be an adjunct awaiting reconsideration in clinical oncology — particularly in patients under chronic stress.
💡 A PARADIGM SHIFT
Cancer is no longer viewed solely as an intracellular signaling disorder, but as a systemic disease living within a “macroenvironment” and centrally regulated by the nervous system. This offers a new lens for understanding why immunotherapy-resistant cases do not produce the expected response.
Stress, Sleep, and Circadian Rhythm: The Neglected Trio
The review's section closest to the patient experience may be brain-body-cancer communication. Extracranial tumors release systemic cytokines such as IL-6 and TNF-α, causing neuroinflammation, fatigue, apathy, and cognitive impairment — which can occur even before chemotherapy. Chronic restraint stress was shown years ago to markedly increase distant breast cancer metastases in mouse models; the mechanism has now been elucidated: tumors can recalibrate the hypothalamic-pituitary-adrenal (HPA) axis in their favor, opening a window for spread to the lungs through neutrophil extracellular traps (NET).
Circadian disruption — shift work, chronic jet lag — has been associated with cancer risk, tumorigenesis, and metastasis. Core clock genes (BMAL1, PER2, CLOCK) suppress epithelial-mesenchymal transition (EMT), immune evasion, and entry into the cell cycle; their loss leads to expansion of cancer stem cells. More strikingly, circulating tumor cell numbers peak during sleep — clear evidence that metastasis is a process active at night. The finding that immune-checkpoint inhibitor efficacy varies with the time of day of administration (chrono-immunotherapy) suggests that timing in treatment planning may be an efficacy variable, beyond simple convenience.
🇹🇷 The Turkish Perspective
β-blockers such as propranolol and metoprolol are widely used in cardiology and endocrinology practice in Türkiye; this accessibility provides a natural basis for research into adjunctive cancer treatment. Caution is needed, however: the review emphasizes that β-blocker benefits vary dramatically by cancer type and stage — so the most appropriate approach is evaluation in clinical trials rather than initiation in an individual patient on the assumption that “it will help.” Gabapentin and perampanel are likewise already used for epilepsy in patients with glioma, but prospective evidence of antitumor efficacy has not yet been generated. Vagal nerve stimulation (VNS) is available in Türkiye for refractory epilepsy; international studies examining quality of life in breast and ovarian cancer have begun.
NiCT: Neuroscience-Instructed Cancer Therapy — Candidate Drug Classes
The review's clinical promise is brought together under “NiCT.” The approach has two pillars: repurposing more than 100 neuroactive drugs approved in neurology, psychiatry, and internal medicine, and developing new molecules specific to cancer neuroscience. The visual below summarizes the candidate drug classes currently being evaluated and the mechanisms they target.
Figure 2. Candidate Drug Classes for NiCT and the Neuro-Cancer Mechanisms They Target
Source: schematic summary based on the Cell 2026 review. Bar lengths conceptually represent evidence levels; they are not numerical data.
There is a critical nuance here: the most mature message from the review is that NiCT is gaining strength not as monotherapy , but as a layer added to standard cytotoxic treatments and immunotherapies. A β-blocker alone does not cure breast cancer; however, when added to an anthracycline regimen, it can markedly improve metastatic control. Perampanel likewise shows potential to sensitize glioblastoma cells to temozolomide and radiotherapy — the results of the ongoing PerSurge phase II trial are therefore critical.
Open Questions and the Next Decade
In a section titled “Box 1,” the review lists the field's critical questions for the next decade. Among the most striking: do bona fide neuron-cancer synapses truly exist in cancers outside the brain, or are they merely “putative”? Are tumor subpopulations with stem-cell properties the primary partners in neural interactions? Do calcium signals and membrane depolarization constitute a “final common pathway” across all cancer types? And perhaps most practically: could NiCTs in precancerous syndromes (such as neurofibromatosis type 1) be used preventively ? Lamotrigine was shown to prevent the initiation of NF1-mutant optic pathway gliomas in mouse models — this could evolve into a prophylactic paradigm for people with tumor predisposition.
🎓 DROZDOGAN Academy — Commentary and Discussion
The Significance of This Topic
Cancer neuroscience was a collection of observations that might have been called “marginal” a decade ago; today, it has matured enough to receive its own review in Cell's Leading Edge category. The issue is not simply “a new signaling pathway” — it concerns how oncology conceptualizes cancer. Moving away from seeing a tumor as an isolated intracellular signaling disorder toward recognizing it as a “systemic disease” intertwined with the nervous system, immunity, endocrine axes, and circadian rhythms helps explain why clinical practice fails in some patients despite powerful therapeutic weapons.
Strengths
The review's most striking strength is that it has settled the debate over whether the field's central phenomenon exists. Neuron-cancer synapses have now been functionally demonstrated in four independent cancer types (glioblastoma, DMG, SCLC, pancreas), supported by electrophysiological recordings, optical/chemogenetic manipulations, and retrograde viral tracing. A second strength is the integrated mapping of ten mechanisms as a “nervous system-cancer connectome,” rather than disconnected subfields. Third, the NiCT framework focuses on repurposing — making clinical testing straightforward and inexpensive while expanding our knowledge of existing drugs even if it fails.
Limitations and Open Questions
The review's optimistic tone needs some balance. First, most evidence comes from mouse models; much of the human evidence is retrospective and correlational. Better glioblastoma survival in gabapentin users represents association, not causation — other selection biases may be hidden behind better symptom control. Second, β-blocker data are heterogeneous: benefit has been shown in some cancer types, no effect in others, and even potential harm in a subgroup. Generalizing that “β-blocker = good” would be dangerous. Third, extracranial “synapses” are largely “putative” — meaning they are morphologically and ultrastructurally similarto synapses but do not meet classic electrophysiological criteria for a synapse. Finally, many NiCT drugs are not approved for their proposed cancer indication, which will create issues with prescribing outside SUT coverage and reimbursement in countries such as Türkiye — this is a health policy issue as well as a scientific one.
Implications for Clinical Practice
For now, the evidence is insufficient to change routine clinical practice — but awareness is critical in some respects. In patients with glioblastoma, GABAergic benzodiazepines (diazepam, midazolam) should be considered a drug class that may warrant avoidance because of their effects on tumor behavior; alternatives such as perampanel or levetiracetam can be considered when selecting antiepileptic treatment. Patients with breast or pancreatic cancer under high chronic stress should be informed that psycho-oncological support and physical activity may matter not only for quality of life but potentially for tumor biology. Routinely asking immunotherapy patients about sleep quality and circadian disruption could be an important step if chrono-immunotherapy becomes established in the future. In Türkiye, widespread access to neuroactive drugs is an advantage; however, this accessibility should not outpace scientific evidence and become empirical prescribing.
Questions for the Future
The critical questions for the next five years are: to what extent do neuron-cancer synapses in extracranial cancers show bona fide synaptic transmission, or are they paracrine-like “synapse-like structures”? In which patient subgroups (genetic profile, tumor biology, stress biomarkers) does NiCT offer the greatest benefit? Can practically feasible protocols be established for chrono-immunotherapy and chrono-chemotherapy? Do nonpharmacological interventions such as sleep modulation, exercise, and psychotherapy measurably prevent metastasis? And perhaps the biggest question: could neurally targeted prophylaxis become a clinically viable option in precancerous syndromes (Li-Fraumeni, NF1, Lynch)? The answers will determine what kind of discipline oncology becomes over the next decade.
Conclusion: A Silent Revolution
Cancer neuroscience is a quiet sixth pillar being added to oncology's traditional pillars (surgery, radiotherapy, chemotherapy, targeted therapy, immunotherapy). The review clearly shows that the field has moved from “discovery” to “consolidation and translation.” It is unrealistic to expect a single NiCT to cure cancer in the coming years; however, a well-timed package of neuroactive interventions, added to standard treatments and selected for the right patient subgroup, could meaningfully improve treatment responses. At the same time, symptoms affecting patients' quality of life — fatigue, apathy, cognitive fog, sleep disturbance, chronic pain — will increasingly be viewed as part of tumor biology, beyond merely “side effects.” This perspective requires oncology practice to collaborate much more closely with neurologists, psychiatrists, anesthesiologists, and psycho-oncologists. This Heidelberg-based review serves as a compass for those seeking to understand where the field is heading toward 2035.
Scientific References
- Winkler F, Heuer S, Althammer F, et al. Cancer neuroscience: The past, the present, and the road ahead. Cell. 2026 Apr 16;189:2464–2489. DOI: 10.1016/j.cell.2026.03.018
- Venkatesh HS, Morishita W, Geraghty AC, et al. Electrical and synaptic integration of glioma into neural circuits. Nature. 2019;573:539–545.
- Venkataramani V, Tanev DI, Strahle C, et al. Glutamatergic synaptic input to glioma cells drives brain tumour progression. Nature. 2019;573:532–538.
- Monje M, Borniger JC, D'Silva NJ, et al. Roadmap for the Emerging Field of Cancer Neuroscience. Cell. 2020;181:219–222.
- Winkler F, Venkatesh HS, Amit M, et al. Cancer neuroscience: State of the field, emerging directions. Cell. 2023;186:1689–1707.
- Hanahan D, Monje M. Cancer hallmarks intersect with neuroscience in the tumor microenvironment. Cancer Cell. 2023;41:573–580.
- Globig AM, Zhao S, Roginsky J, et al. The β1-adrenergic receptor links sympathetic nerves to T cell exhaustion. Nature. 2023;622:383–392.
- Diamantopoulou Z, Castro-Giner F, Schwab FD, et al. The metastatic spread of breast cancer accelerates during sleep. Nature. 2022;607:156–162.
- Heuer S, Burghaus I, Gose M, et al. PerSurge (NOA-30) phase II trial of perampanel treatment around surgery in patients with progressive glioblastoma. BMC Cancer. 2024;24:135.
- Bernstock JD, Mehari M, E Gerstl JV, et al. Gabapentinoids confer survival benefit in human glioblastoma. Nat Commun. 2025;16:4483.
- Jones G, Anderson JL, Nguyen PTT, et al. Novel approaches to clinical trial design in cancer neuroscience. Neuron. 2025;113:2791–2813.
- Pan Y, Hysinger JD, Barron T, et al. NF1 mutation drives neuronal activity-dependent initiation of optic glioma. Nature. 2021;594:277–282.
Editorial note: This content was prepared by the drozdogan.com editorial team, restructuring the comprehensive review published in the April 2026 issue of Cell for Turkish readers and adding commentary, a Turkish perspective, and critical discussion sections. AI-assisted language tools were used to edit and summarize the text; all clinical comments and practical recommendations underwent editorial review before publication. This article is for general information; individual diagnostic and treatment decisions must be discussed with the relevant specialist physician.
Was this article helpful?
Your feedback helps us improve our content.
Reader comments
Share your comment
Your name and comment are published only after administrator approval. Your email address is never shown to visitors.
Be the first to comment.