Targeting the nervous system: Mechanistic insights into neuro–tumor communication and therapeutic opportunities

Abstract

The nervous system has emerged as an active regulator of the tumor ecosystem. Neural inputs from the central, peripheral, and enteric nervous systems influence tumor initiation, progression, metastasis, and therapeutic response through innervation and signal transmission. In addition to regulating tumor cell states, neural signaling reshapes stromal and immune compartments, promoting stemness, invasion, angiogenesis, metabolic adaptation, and immune evasion. Tumors, in turn, actively remodel their neural niche through neurogenesis, axonal remodeling, and tumor–nerve feedforward circuits. The biological effects of neuro–tumor crosstalk are highly context-dependent and are jointly determined by neural origin, receptor distribution, tumor state, and the local microenvironment. Accordingly, neural signaling has become a promising axis for therapeutic intervention. In this review, we summarize key advances in the molecular mechanisms of neuro–tumor communication over the past five years. We also highlight neurotransmitter and neuropeptide receptor axes with pharmacological and translational relevance, related targeted agents, and therapeutic strategies for targeting these pathways, as well as recent progress in neuro-immune combination therapy and clinical studies. Together, these advances provide an updated framework for understanding the role of neural signaling in tumor biology and its therapeutic potential.

Graphical abstract

Emerging evidence highlights that neural signaling modulates the tumor microenvironment to promote cancer progression, offering new therapeutic strategies through pharmacological intervention.

1. Introduction

The tumor microenvironment (TME) is a dynamic ecosystem that critically shapes tumor progression, metastasis, and therapeutic resistance¹. Beyond malignant cells, the TME comprises immune, stromal, vascular, and neural components that collectively influence disease evolution. Among these, the nervous system has emerged as an active and previously underappreciated regulator of tumor biology, giving rise to the rapidly expanding field of cancer neuroscience. Increasing evidence indicates that neural inputs are not merely secondary responses to tumor growth, but integral components of the tumor ecosystem that influence multiple stages of cancer development.

Bidirectional communication between tumors and the nervous system has been documented across a broad range of solid malignancies. Both the central nervous system (CNS) and the peripheral nervous system (PNS) participate in neuro–tumor crosstalk. This includes the sympathetic, parasympathetic, sensory, and enteric branches. They influence tumors through direct innervation, release of neurotransmitters and neurotrophic factors, neuroendocrine regulation, and indirect effects on immune and stromal cells2, 3, 4.

Through these pathways, neural signaling can shape proliferation, invasion, metastatic dissemination, angiogenesis, stemness, immune evasion, and therapeutic response. Functionally, the CNS mainly acts through systemic neuroendocrine programs⁵, whereas the PNS exerts more direct local effects within the tumor niche^6,7. In gastrointestinal malignancies, the enteric nervous system (ENS) provides an additional regulatory layer by integrating neural, immune, and microbiota-derived signals within the local tissue environment⁸. Together, these observations support neural regulation as a fundamental dimension of tumor biology (Fig. 1).

Figure 1.

An overview of the nervous system and its role in tumor progression. The central nervous system (CNS), peripheral nervous system (PNS), and enteric nervous system (ENS) are the three main divisions of the nervous system. The CNS includes the brain and spinal cord, while the PNS consists of cranial nerves, somatic nerves (sensory and motor), and autonomic nerves (sympathetic and parasympathetic). The ENS is an intrinsic part of the PNS, responsible for gut function. Neural innervation impacts tumor progression through multiple mechanisms, including tumor proliferation, tumor metastasis, angiogenesis, metabolic reprogramming, epigenetic reprogramming and immune evasion.

Despite the rapid progress of cancer neuroscience, the field remains conceptually fragmented. Although the importance of neural signaling in tumor biology is now widely recognized, related findings are often presented separately across tumor types, neural branches, and signaling molecules. This fragmentation makes it difficult to explain why similar neural inputs may exert distinct biological effects in different tumor ecosystems. Thus, a central challenge is no longer whether nerves regulate tumors, but how the biological consequences of neural signaling are shaped within specific tumor ecosystems.

In this review, we examine neuro–tumor communication from an integrated and context-aware perspective. We first summarize the anatomical and functional interfaces between the nervous system and the TME. We then discuss the major molecular mediators that govern neuro–tumor crosstalk. Next, we examine how tumors remodel their neural niche through neurogenesis, axon guidance, and feedforward innervation circuits. Finally, we highlight the therapeutic implications of these interactions, including receptor-targeted pharmacology, drug repurposing, neuroimmune combination strategies, and emerging translational opportunities. Through this review, we aim to provide an integrated view of how the nervous system shapes tumor biology and therapeutic response.

2. Anatomical interactions between the nervous system and the tumor microenvironment: from neural innervation to tumor evolution

2.1. CNS-mediated tumor regulation: neuroendocrine, neuroimmune, and barrier-associated interfaces

The CNS regulates tumor progression through several interconnected routes, including direct synaptic and neurotransmitter-mediated communication in brain tumors, systemic neuroendocrine and neuroimmune pathways, and specialized CNS interfaces with the cerebrospinal fluid (CSF). Through these mechanisms, the CNS influences both primary brain tumors and peripheral malignancies by shaping tumor proliferation, dissemination, immune surveillance, and therapeutic accessibility (Fig. 2).

Figure 2.

Schematic illustration of the mechanisms of the CNS and CSF-associated mechanisms in tumor regulation. In the CNS compartment, Glu signaling, BDNF-mediated neurotrophic support, neuron–glial interactions, and immune remodeling characterized by macrophage polarization toward TAMs and reduced T-cell activity collectively promote tumor proliferation. In the CSF compartment, upregulation of GAD67 and Bestrophin-1, associated with increased GABA signaling, facilitates metastatic progression. In parallel, MDK-driven macrophage polarization and downregulation of VLA-4 on T cells contribute to an immunosuppressive microenvironment.

2.1.1. Synaptic and neurotransmitter-mediated CNS regulation of tumor progression

The CNS regulates tumor biology through both direct and indirect mechanisms. In brain tumors, neurons and glial cells directly modulate the tumor microenvironment via neural circuits, neurotransmitters, and neurotrophic factors, thereby promoting tumor proliferation and invasion. In peripheral tumors, CNS influences are more often exerted indirectly through systemic neuroendocrine signaling.

In gliomas, tumor cells form functional synapses with neurons, receiving excitatory inputs such as glutamate through synaptic signaling, which enhances tumor proliferation and invasion^9,10. In addition, specific neuronal subpopulations, such as callosal projection neurons in the contralateral hemisphere, can further promote tumor malignancy by driving activity-dependent infiltration and hyperactive brain networks through bidirectional synaptic communication with tumor cells¹¹. This synaptic integration drives malignant progression and establishes a “proliferation–synaptic integration–hyperactive brain” vicious cycle¹². Neurotrophic support further strengthens tumor responsiveness to neural input. For example, brain-derived neurotrophic factor (BDNF) activates CAMKII through TrkB, promotes AMPA receptor trafficking to the glioma cell membrane, and increases the amplitude of glutamate-induced currents, thereby enhancing malignant synaptic strength and glioma proliferation¹³. Together, these findings indicate that CNS neural activity can directly shape tumor behavior by enabling functionally integrated neuron–tumor communication.

2.1.2. CNS-associated neuroendocrine and neuroimmune regulation of tumor progression

Beyond direct synaptic communication, the CNS also influences tumor progression through neuroendocrine and neuroimmune pathways. Within brain tumors, CNS-associated immune regulation contributes to the formation of an immunosuppressive niche, characterized by macrophage polarization¹⁴ and reduced T-cell activity¹⁵, which supports tumor growth and immune escape.

A representative systemic route is the HPA axis, through which the CNS modulates peripheral immunity. Intracranial tumors increase plasma corticosterone, activate glucocorticoid-receptor signaling, and upregulate CCR8 in T cells, while also elevating CCL1 and CCL8 levels in the bone marrow. Together, these changes retain T cells in the bone marrow and reduce their infiltration into brain tumors. Blocking the glucocorticoid receptor–CCR8 axis restores T-cell distribution and antitumor immunity¹⁶. These findings illustrate how CNS-derived neuroendocrine signals can reshape systemic immune allocation and thereby influence antitumor responses.

2.1.3. CSF as a key interface in CNS-associated tumor regulation

CSF serves as a CNS-specific regulatory interface that shapes tumor dissemination, immune surveillance, and therapeutic accessibility. In brain metastasis, CSF is not merely a transport medium but an active microenvironmental component that supports tumor survival and metastatic progression. In leptomeningeal metastasis (LM) from breast or lung cancer, tumor cells interact with the choroid plexus epithelium, upregulating key proteins such as GAD67 and Bestrophin-1. This results in elevated levels of GABA in the CSF, which helps sustain tumor cell survival and facilitates metastatic spread¹⁷.

Beyond its role in tumor dissemination, CSF also functions as a distinct immune niche that can influence antitumor activity and metastatic progression. For example, in lung cancer-associated LM, lipid-associated macrophages are regulated by Midkine (MDK), which is significantly elevated in both CSF and plasma, driving macrophage polarization towards an M2 phenotype and creating an immunosuppressive environment¹⁸. Similarly, in breast cancer LM models, CD8⁺ T cells in the CSF show senescence/exhaustion markers (e.g., downregulation of VLA-4), which reduces their anti-tumor activity and accelerates LM progression¹⁹. These findings indicate that CSF and barrier-associated interfaces are not merely anatomical constraints, but active components of CNS-associated tumor regulation.

The mechanisms discussed above indicate that CNS-associated tumor regulation is mediated through several interconnected routes, including neuroendocrine signaling, synaptic communication, and CSF- or barrier-related regulation. In contrast, peripheral neural regulation is distinguished by its closer anatomical proximity to the tumor ecosystem. Local neural projections and neurotransmitter release can act more directly on tumor cells, stromal components, and the immune microenvironment and can therefore influence tumor progression and microenvironmental remodeling.

2.2. Peripheral neural programs in the tumor ecosystem

The PNS, comprising motor, sensory, autonomic nerves (sympathetic and parasympathetic) and the ENS, has traditionally been regarded as a passive conduit for signal transmission. However, accumulating evidence indicates that the PNS functionally interacts with the TME through neural projections and the release of neurotransmitters and neurotrophic factors. This interaction modulates tumor progression. Neural–TME crosstalk involves neurotransmitter-driven oncogenic signaling, immune microenvironment remodeling, and tumor cell-intrinsic epigenetic and metabolic reprogramming, with these mechanisms potentially exhibiting specificity among different PNS components.

2.2.1. Sensory nerves

2.2.1.1. Sensory nerves in tumor progression

Sensory neurons detect external and internal stimuli through diverse receptors and convert these cues into electrical or chemical signals, allowing them to regulate pain perception, mechanosensation, and signal transmission within the tumor microenvironment (Fig. 3). Sensory neurons use several receptor types to detect and transmit signals. Transient receptor potential (TRP) channels (TRPV1, TRPM8, and TRPA1) detect thermal and chemical cues. PIEZO1 and PIEZO2 sense extracellular matrix (ECM) tension and pressure. CGRP receptors regulate proliferation and inflammatory signaling. CB1/CB2 and other ion channels control neuronal excitability and immune responses. Through these pathways, sensory neurons not only perceive environmental cues but also actively contribute to tumor progression, metastasis, and microenvironment remodeling.

Figure 3.

Sensory nerves contribute to tumorigenesis. The somatic nervous system contributes to tumor progression through sensory nerve pathways. PIEZO1 pathway: In breast cancer, PIEZO1 activation induces Ca²⁺ influx, driving EMT signaling and promoting EMT. CGRP–RAMP1 pathway: In melanoma, CGRP acts via RAMP1 to inhibit TCR signaling, leading to CD8⁺ T-cell exhaustion and immunosuppression; in gastric cancer, the CGRP–RAMP1 axis facilitates tumor proliferation. NLRP3⁺ macrophage and CAF-related pathway: In PDAC, NLRP3⁺ macrophages contribute to immunosuppression; CGRP and NGF inhibit IL-15 in CAFs, impairing NK cell function and enhancing immunosuppression. TRPV1 pathway: In breast cancer models, TRPV1⁺ sensory neurons promote tumor metastasis via immunosuppression. CCL21/CXCL10 pathway: In PDAC, this chemokine axis drives sensory nerve infiltration and promotes tumor proliferation.

During bone cancer growth and colonization, upregulation of TRPV1 receptors on sensory neurons promotes persistent and movement-evoked pain, and their antagonism or deletion significantly alleviates these pain behaviors^20,21. This suggests that sensory denervation or inhibition of sensory neuron excitation may reduce both bone pain and cancer progression. TRPM8 channels in prostate cancer cells similarly regulate Ca²⁺ permeability, influencing tumor cell function^22,23. Mechanosensory neurons expressing PIEZO1 and PIEZO2 detect ECM tension and pressure changes, and in breast cancer, elevated PIEZO1 expression correlates with malignant phenotypes, promoting epithelial-to-mesenchymal plasticity through Ca²⁺ influx-mediated activation of pro-migratory signaling pathways²⁴.

Neuropeptides released from sensory neurons also play critical roles in tumor progression. Calcitonin gene-related peptide (CGRP) density correlates with tumor incidence in head and neck squamous cell carcinoma (HNSCC), suggesting a direct tumor-promoting effect²⁵. In gastric cancer, tumor cells recruit CGRP ⁺ sensory neurons via NGF–Trk signaling and facilitate tumor dissemination through the CGRP–RAMP1 axis, while CGRP antagonists markedly suppress tumor growth and extend survival²⁶.

The endocannabinoid/exocannabinoid system (CB1/CB2) induces apoptosis and inhibits invasion in osteosarcoma²⁷. Sensory neurons further modulate host immunity by regulating intestinal barrier integrity and microbiota composition²⁸, while stimulator of interferon genes (STING) controls nociception in sensory neurons via type I interferon (IFN-I) signaling and counteracts tumor-promoting chronic inflammation²⁹.

2.2.1.2. Sensory neuroimmune interactions in the tumor ecosystem

Sensory neurons promote tumor immune evasion not only through direct neuro–immune interactions but also by secreting factors that regulate immune cell functions, shaping an immunosuppressive microenvironment and further exacerbating immune escape and tumor progression (Fig. 3).

In breast cancer models, TRPV1⁺ sensory neurons promote metastasis through neuroimmune signaling. They enhance neuro–immune crosstalk in the tumor microenvironment and accelerate tumor cell migration and invasion³⁰. Mechanistically, these neurons secrete substance P and activate the extracellular RNA–TLR7 axis. This signaling cascade directly drives breast cancer metastasis³¹.

In HNSCC, increased densities of CGRP⁺ sensory nerves correlate with immunosuppressive TME profiles; conversely, sensory denervation enhances T cell activity and inhibits tumor growth³².

In melanoma, CGRP inhibits T cell receptor (TCR) signaling through RAMP1, leading to CD8⁺ T cell exhaustion; sensory denervation or pharmacological blockade of CGRP restores T cell function and significantly prolongs survival in mouse models³³.

In PDAC, regions with high nerve infiltration form an immunosuppressive microenvironment composed of NLRP3⁺ macrophages, CAFs, and TGF-β-induced, migration-promoting TGFBI⁺ Schwann cells at the nerve–tumor interface, and are associated with poor patient prognosis³⁴. Nociceptor neurons interact with CAFs via CGRP and NGF. This interaction downregulates IL-15 expression in CAFs, reduces the infiltration and cytotoxicity of NK cells, and ultimately promotes PDAC progression and cancer-related pain³⁵. Hirth et al.³⁶ identified the key signaling axes CCL21–CCR7 and CXCL10–CXCR3 in sensory neuron–cancer cell interactions through unbiased screening³⁶. This axis drives the chemotaxis and invasion of PDAC cells toward sensory nerves, serving as a crucial mechanism mediating PDAC-related pain and providing a novel potential target for targeted analgesic therapy. These findings suggest that sensory neurons not only contribute to immune evasion in tumors but may also serve as potential therapeutic targets in immunotherapy.

2.2.2. Sympathetic nerves

2.2.2.1. Adrenergic signaling in tumor progression and microenvironmental remodeling

The autonomic nervous system, composed of the sympathetic nervous system and parasympathetic nervous system, plays a pivotal role in tumor initiation, progression, and immune regulation through multilayered neurotransmitter signaling and cell–cell interactions. The SNS exerts its tumor-promoting effects primarily through the release of norepinephrine (NE) and epinephrine (Epi), which activate adrenergic receptors (ARs) expressed on tumor and stromal cells, triggering downstream oncogenic signaling pathways. These pathways promote tumor cell proliferation, angiogenesis, EMT, and immunosuppressive remodeling of the TME through α- and β-adrenergic signaling (Fig. 4).

Figure 4.

Sympathetic nervous system regulates tumor progression. Schematic illustration of the sympathetic nervous system's role in regulating tumor progression. α-AR pathway: In CRC, NE induces CAFs to secrete NGF, activating the ADRA2A/PI3K/AKT pathway in tumor cells to promote proliferation; NE also maintains myeloid cell maturation via α-AR, and its blockade leads to MDSC/Treg accumulation and tumor progression. β-AR pathway: In CRC, Epi activates β₂-AR/PKA/CREB1 to enhance glycolysis and tumor growth; In gastric cancer, depressive states drive β-AR/MACC1-mediated neuroendocrine transdifferentiation, which enhances tumor invasiveness. In breast cancer, Epi activates USP22–ATGL to accelerate lung metastasis. NE regulates tumor cell secretion of CXCL9, inducing immunosuppression. NE activates IL-6/JAK/STAT3 to expand MDSCs and CXCL5/CXCR2/ERK to recruit MDSCs, inhibiting T cells to induce immunosuppression. NE activates β₁-AR on exhausted CD8⁺ T cells to exacerbate T cell exhaustion in pancreatic cancer and reduces ICB efficacy. In HCC, NE-induced SFRP1 activates WNT16B/β-catenin, promoting EMT/CSC-related invasion.

In neuroblastoma (NB), isoliquiritigenin or α1B receptor antagonists enhance the antiproliferative and prodifferentiation effects of 13-cis-retinoic acid, effectively suppressing tumor growth and improving control of residual disease by targeting the α1B adrenergic receptor³⁷. Pharmacological α₁-AR blockade has been shown to inhibit angiogenesis and EMT, effectively suppressing tumor growth³⁸.

In colorectal cancer (CRC), Epi upregulates β-AR expression, activating the CEBPB/TRIM2 axis to promote epithelial-to-mesenchymal transition and stemness³⁹. Epi promotes enhanced glycolysis and tumor growth in CRC through the β2-adrenergic receptor/PKA/CREB1 pathway⁴⁰. In breast cancer, chronic stress-induced Epi promotes tumor progression and lung metastasis. It activates AKT-mediated phosphorylation and stabilizes USP22. USP22 then stabilizes FOXO1, which enhances adipose triglyceride lipase (ATGL)-mediated lipolysis⁴¹. In gastric cancer, depressive states can drive β-AR/MACC1-mediated neuroendocrine transdifferentiation, increasing tumor aggressiveness⁴². In oral cancer, sympathetic nerves activate tumor cell β-AR by releasing NE, promoting TNF-α secretion and worsening pain perception. Meanwhile, the use of the β-blocker propranolol or chemical sympathectomy not only significantly alleviates cancer-induced pain but also inhibits tumor growth⁴³. NE promotes the survival and micro-metastasis of fallopian tube epithelial pre-neoplastic lesions to the ovary via β-AR, enhancing spheroid formation and hypoxia resistance. These changes collectively support the early progression of high-grade serous ovarian cancer⁴⁴.

Within the TME, NE acts not only on tumor cells but also on stromal components such as cancer-associated fibroblasts (CAFs) to create a pro-metastatic network. In breast cancer, sympathetic innervation contributes to bone remodeling and regulation of the bone vasculature, which increases the ability of tumor cells to colonize the bone niche⁴⁵. In CRC, NE induces CAFs to secrete NGF. NGF activates ADRA2A and the PI3K/AKT pathway in tumor cells, which promotes tumor growth. NE also increases sympathetic innervation and intratumoral norepinephrine accumulation. Together, these events form a CAF-nerve positive feedback loop that may be targeted with TRK inhibitors⁴⁶. In hepatocellular carcinoma (HCC), NE induces SFRP1 expression, activating a WNT16B/β-catenin positive feedback loop that drives EMT and CSC-associated gene expression, ultimately promoting tumor invasion and migration⁴⁷.

2.2.2.2. Adrenergic regulation of tumor immunity

In the TME, immune cells expressing AR, including T cells, MDSCs, and tumor-associated macrophages (TAMs), respond to signals from NE and Epi, forming potent immunosuppressive circuits⁴⁸ (Fig. 4).

In CRC, NE maintains myeloid cell maturation via α-AR; disruption of this signaling leads to the accumulation of immature MDSCs and Tregs in the tumor microenvironment, with marked suppression of anti-tumor immunity and accelerated tumor progression⁴⁹. Furthermore, NE promotes the expansion of MDSCs by activating the IL-6/JAK/STAT3 signaling pathway. This signaling event reinforces the immunosuppressive tumor microenvironment and facilitates breast cancer metastasis⁵⁰. In HCC, NE drives MDSC recruitment through the CXCL5/CXCR2/ERK pathway, further inhibiting T-cell proliferation and promoting immune escape⁵¹. This series of effects underscores the complex role of NE in regulating immune cell behavior and shaping the immunosuppressive nature of the TME. Catecholamines released by the sympathetic nerves directly drive CD8⁺ T cells into a state of functional exhaustion via the β1-adrenergic receptor ADRB1. Blocking this signaling pathway not only reverses T cell exhaustion and enhances their anti-tumor function but also synergizes with immune checkpoint blockade to improve therapeutic efficacy⁵². This discovery establishes a direct link between the sympathetic stress response and T cell exhaustion, providing a novel target for improving tumor immunotherapy.

Targeting NE may help overcome resistance to anti-PD-1 monoclonal antibody therapy. In lung adenocarcinoma (LUAD), NE regulates the secretion of CXCL9 and adenosine by tumor cells. This suppresses the chemotaxis and effector functions of CD8⁺ T cells and promotes resistance to anti-PD-1 therapy through the WNT7A/β-catenin pathway⁵³. In renal clear cell renal cell carcinoma, NE drives an immunosuppressive microenvironment that restricts the effector function of CAR T cells; inhibition of NE signaling alleviates CD8⁺ T cell exhaustion and M2 macrophage polarization. This enhances the antitumor efficacy of CAR T cells⁵⁴.

From a therapeutic perspective, emerging strategies targeting neuroimmune pathways are gaining traction. For instance, NE antagonists and immunosympathectomy exhibit significant antitumor effects. These findings highlight the promise of targeting neuroimmune interactions in cancer therapy, offering new strategies to overcome tumor immune tolerance.

2.2.3. Parasympathetic nerves

2.2.3.1. Acetylcholine signaling in tumor progression

The PNS primarily exerts its effects through acetylcholine (ACh), which activates nicotinic (nAChRs) or muscarinic (mAChRs) acetylcholine receptors on target cells and contributes to tumor proliferation and metastasis (Fig. 5).

Figure 5.

Parasympathetic nervous system regulates tumor progression. Schematic illustration of parasympathetic nervous system regulation of tumor progression through ACh-mediated signaling. Parasympathetic neurons release ACh, which activates acetylcholine receptors on tumor cells. In LUAD, ACh regulates the DNMT1/FHIT axis via α5-nAChR, enhancing tumor invasiveness. In NSCLC, α7-nAChR is activated by tobacco components (e.g., NNK), promoting tumor metastasis via GSK3β/β-catenin and invasion via JAK2/STAT3. In gastric cancer, ACh activates M3-mAChR to induce NGF expression, promoting tumor proliferation via YAP; in prostate cancer, ACh activates M1/M3-mAChRs to regulate the FAK–YAP pathway, driving castration-resistant tumor growth. VAChT-mediated pathway: In renal cell carcinoma, upregulated SLC18A3 enhances ACh uptake, activating the cAMP/PKA/CREB pathway to promote tumor proliferation. TAM-related pathway: In PDAC, ACh reduces TAM infiltration and serum pro-inflammatory cytokine levels, exerting an antitumor effect.

In tumors, nAChRs are frequently overexpressed, activating downstream oncogenic pathways to drive tumor cell proliferation, EMT, and CSC properties^55,56. Specifically, in LUAD, ACh engages α5-nAChR to modulate the DNMT1/FHIT axis, promoting migratory and invasive capabilities⁵⁷. In NSCLC, α7-nAChR is activated by tobacco component-induced signaling, such as nicotine- or NNK-induced signaling, promoting tumor progression⁵⁸. In esophageal squamous cell carcinoma (ESCC), nicotine activates α7-nAChR and subsequently upregulates the JAK2/STAT3 signaling pathway, promoting tumor malignancy and tumor-initiating capacity⁷. In breast cancer, α9-nAChR activation promotes angiogenesis⁵⁹.

In gastric cancer, DCLK1⁺ tuft cells and nerves release ACh, which induces NGF expression through activation of ACh muscarinic receptor-3 (m3-AChR). This response expands enteric nerve innervation and promotes YAP-dependent tumorigenesis⁶⁰. In prostate cancer, ACh activates m1 and m3-AChRs to regulate the FAK/YAP pathway, promoting castration-resistant growth of prostate cancer⁶¹. In renal cancer, upregulation of SLC18A3 increases ACh uptake, activating the protein kinase A (PKA)/cAMP-response element binding protein (CREB) pathway to promote tumor growth⁶².

Notably, the tumor-promoting effects of parasympathetic ACh signaling are often cancer type-specific, highly dependent on receptor subtype expression and the TME state. This suggests that the ACh-driven neuro–tumor interaction network is emerging as a promising target for precision therapeutic intervention.

2.2.3.2. Acetylcholine signaling in tumor immunity

α7nAChR on macrophages plays a key role in mediating cholinergic anti-inflammatory effects through the downregulation of NF-κB signaling. Recent studies further show that cholinergic signaling via α7nAChR regulates the migration of monocyte-derived macrophages during acute inflammation, contributing to the control of inflammation⁶³. Similarly, in PDAC, ACh has been reported to inhibit the ERK signaling pathway, reduce TAM infiltration, and lower serum pro-inflammatory cytokine levels, which exerts a tumor-suppressive effect⁶⁴.

Collectively, these mechanisms underscore the pivotal role of parasympathetic neural pathways in regulating immunosuppression, EMT, stemness maintenance, and tumor metastasis, providing a solid theoretical basis for targeting cholinergic signaling in therapeutic interventions for tumors.

2.2.4. Enteric nervous system

2.2.4.1. ENS-mediated gut–nerve–tumor interactions

The ENS, often referred to as the “second brain”, forms a complex neural network within the submucosal and myenteric plexuses, orchestrating gut motility, secretion, barrier integrity, and immune homeostasis⁶⁵. Beyond its autonomous control of gastrointestinal physiology^66,67, the ENS exerts multiple pro-tumorigenic effects in colorectal cancer by modulating the intestinal microenvironment and neurotransmitter signaling (Fig. 6).

Figure 6.

Enteric nervous system and the gut–neuro–tumor axis. Schematic illustration of ENS and extrinsic neural input in regulating intestinal tumor progression. The diagram highlights the anatomical and molecular interactions between the central nervous system, sympathetic nerves, and components of the ENS, including the myenteric and submucosal plexuses. Nerve invasion: tumor- and stroma-derived neurotrophic factors (Netrin-1, NGF, BDNF, and GDNF) drive nerve infiltration and perineural invasion. Inflammatory damage: EGCs under hyperglycemia release ROS and proinflammatory mediators, damaging the mucosal barrier, increasing tumor susceptibility and establishing a proinflammatory and protumor microenvironment. Stemness: Enterochromaffin-derived 5-HT reinforces CSC self-renewal. Sympathetic input and ENS-resident neurons regulate Notch signaling and SOX10 expression, promoting CSC plasticity and metastatic dissemination. Immunosuppression: metabolites (isovaleric acid and indole-3-lactic acid) stimulate proliferation and immunosuppression. Probiotic indole-3-lactic acid enhances CD8⁺ T-cell responses and inhibits colorectal tumorigenesis. Proliferation: CAFs secrete NGF and upregulate Netrin-1 under norepinephrine, forming a positive feedback loop for tumor growth. Metastasis and invasion: Schwann-cell-like glia activated by ENS signals enhance CRC cell migration and invasion via NF-κB/IL-8, sustaining a protumor neurogenic niche. In CRC, substance P-NK1R promotes tumor cell proliferation and survival.

Developmental defects of the ENS can disrupt gut structure and function, creating a microenvironment conducive to tumorigenesis. As a conserved regulator of cell fate, aberrant activation of the Notch signaling pathway impairs intestinal homeostasis, induces structural and functional abnormalities of the ENS, and further increases the risk of tumorigenesis⁶⁸. In contrast, systemic iron overload can trigger oxidative stress and ferroptosis; by suppressing the activity of the Notch signaling pathway, it impedes the proliferation of intestinal stem cells, alters their differentiation trajectory, and consequently perturbs intestinal homeostasis to a greater extent⁶⁹. In Hirschsprung disease (HSCR), NOTCH1 and its ligand JAG2 function as the core signaling mediators for ZEB2 to promote the proliferation and differentiation of enteric neural precursor cells, which alleviates HSCR-associated enterocolitis⁷⁰. Clinically, in CRC patients, upregulation of NOTCH4 correlates positively with proliferating cell nuclear antigen (PCNA) expression, vascular invasion, and poor differentiation, and negatively with overall survival, indicating its role as a prognostic marker⁷¹. Embryonic ENS developmental defects, such as mutations in the transcription factor Sox10, can lead to disrupted ganglion distribution and congenital megacolon-like structures, potentially fostering a pro-tumorigenic microenvironment⁷². In adulthood, pathological activation of enteric glial cells (EGCs) can release ROS and pro-inflammatory mediators under hyperglycemic stress, damaging the mucosal barrier and increasing tumor susceptibility. Upon carcinogen exposure, such dysfunction markedly elevates gastrointestinal cancer risk^73,74. Loss of NDRG4 in the enteric nervous system promotes intestinal adenoma formation and colorectal cancer progression by releasing neuron-derived factors enriched in Nidogen-1 and Fibulin-2, which enhance the migratory capacity of CRC cells⁷⁵.

The ENS also directly promotes malignant phenotypes by releasing neurotransmitters that act on tumor cell surface receptors. These neurotransmitters activate downstream signaling pathways and enhance proliferation. Specifically, enteric serotonergic neurons produce 5-HT that activates HTR1B/1D-mediated Wnt/β-catenin signaling to sustain colorectal CSC self-renewal, a process further enhanced by CRC-enriched microbiota-derived isovalerate and ultimately promoting colorectal tumorigenesis⁷⁶. In CRC, the substance P (SP)–neurokinin 1 receptor (NK1R) signaling axis promotes tumor cell growth and survival by inducing ER stress, triggering calcium release and thereby representing a key pro-tumorigenic mechanism⁷⁷.

Beyond direct tumor–neuron interactions, the ENS collaborates with CAFs and the gut microbiota to remodel the TME. In CRC, NGF secreted by CAFs under the influence of NE enhances intratumoral sympathetic innervation, forming a CAF–nerve feedforward loop that promotes tumor cell proliferation and cancer progression⁴⁶. In CRC, Netrin-1 is highly expressed in tumor cells and is also upregulated in CAFs. It promotes CAF–tumor cell signaling and enhances cancer stemness, changes that drive tumor progression and plasticity⁷⁸. Enteric glial cells indirectly shape the gut microbiota by modulating intestinal stem cell behavior and local immune responses, thereby cooperatively regulating the tumor microenvironment⁷⁹.

2.2.4.2. Pro-tumor immune–neural circuits in the ENS

Aberrant activation of the ENS can initiate chronic intestinal inflammation via the Eph/ephrin signaling pathway, triggering the release of pro-inflammatory cytokines such as IL-1β and TNF-α. This inflammatory milieu fosters the accumulation of genetic mutations and induces EMT, creating a pro-inflammatory and tumor-promoting microenvironment⁸⁰.

Moreover, the ENS can shape gut microbiota composition via the gut–brain axis, where microbial metabolites such as short-chain fatty acids and tryptophan derivatives modulate ENS neuronal development and function, establishing a “microbiota–neural–tumor” positive feedback loop^81,82. For instance, indole-3-lactic acid derived from Lactobacillus plantarum can modulate the proportion of CD8⁺ T cells, fine-tuning immune responses and inhibiting colorectal tumorigenesis⁸³.

In addition, the gut microbiota can indirectly influence tumor progression by regulating ENS activity. Metabolites from Lactobacillus plantarum OC01 modulate CRC cell secretions (TGF-β) to inhibit tumor-associated M2 macrophage polarization, promote a pro-inflammatory macrophage phenotype, and consequently suppress CRC proliferation and migration⁸⁴. Schwann cells, a glial cell type within the PNS, can be activated by ENS-derived signals in the TME. Once activated, Schwann cells enhance colorectal cancer cell migration and invasion through the NF-κB/IL-8 signaling axis⁸⁵ (Fig. 6). These findings indicate that tumor-associated neural remodeling is not solely dependent on neuronal axonal guidance but also requires the cooperative involvement of non-neuronal components such as Schwann cells and microbiota.

3. Molecular mediators of neuro–tumor communication: neurotransmitters, neuropeptides, neurexins and neuroligins

Beyond classical neurotransmitters covered in CNS and PNS projections, additional molecular mediators such as glutamate, GABA, dopamine, neuropeptides, and synaptic adhesion molecules including neurexins and neuroligins further regulate tumor cell behavior and the TME.

3.1. Neurotransmitters

Neurotransmitters are better understood as dynamic signaling mediators within broader neural communication networks, in which receptor-dependent and context-specific signaling shapes tumor progression, immune regulation, and therapeutic response.

3.1.1. Glutamate

Glutamate is primarily synthesized in the brain by astrocytes via the glutamine pathway, where glutamine synthetase (GS) catalyzes the conversion of glutamate and ammonia into glutamine. Neurons then take up this glutamine and convert it back into glutamate⁸⁶. Glutamate receptors include ionotropic (iGluRs) and metabotropic (mGluRs)⁸⁷ types. Glutamate plays a critical role in remodeling the TME. In tumor cell metabolism, glutamate serves as both an essential carbon source and a TCA-cycle intermediate. It supports antioxidant defense, energy production, and biosynthesis. Together, these functions promote tumor growth⁸⁸ (Table 189, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132).

Table 1.

Neurotransmitters.

Neurotransmitter	Target/receptor	Signaling pathway/axe	Function	Cancer type	Ref.
Glutamate	CD147	p-AKT/HIF-1α/DRP1	Promote glycolysis while inhibiting mitochondrial function under Gln deprivation	GBM	89
	SLC7A11 (xCT)	Cystine/glutamate antiporter	↑ glutamate efflux → seizures, shorter survival; inhibit T-cell antitumor activity	GBM	90,91
	IL-1β	IL-1β–CCL5 circuit	Glutamate from injured neurons → IL-1β by oligodendrocytes→ microglial CCL5 → optic glioma	Glioma	92
	–	Cytosolic glutamate efflux	↑ nucleotide synthesis to sustain rapid growth	HCC	93
	GDH	Glutamate→α-KG	↑ mitochondrial metabolism → promote liver fibrosis	Liver fibrosis	94
	GRIN2D	Neuron-cancer pseudo-synapse	Ca2+ signaling enhances invasion & metastasis	PDAC	95
	mGluR3	Rab27-EV release	↑ extracellular vesicle release → pro-invasive niche	BC	96
	–	PD-L1+ exosomes	xCT-mediated PD-L1+ exosomes drive M2 polarization → reduce PD-1 blockade efficacy	Melanoma	97
	NetG1	AKT/4E-BP1; GS/GLUT1	CAF-derived NetG1 induces immunosuppression via glutamate metabolism	PDAC	98
	NMDAR2B	CCL2–TAM recruitment	↑ TAM infiltration → remodel immunosuppressive microenvironment	HCC	99
	NMDAR	Ca2+–ROS–NF-κB	↑ TAM immunosuppression → inhibit CD8+ & NK cells	Hepatic sarcoma/Fibrosarcoma	100
	GRM4	cAMP–CREB	GRM4 loss → ↑ IFN-γ → CD8+ activation → tumor suppression	Melanoma	101
GABA	GAD1	–	High GAD1 → ↑ GABA → promote invasion & metastasis	CRC	102
	GABAA	PKC–CREB–PODXL/MMP3/9	↑ Ca2+ influx → CREB activation → metastasis	BC	103
	–	FOXA2/ABAT/NF-κB	↑ brain-colonization fitness via astrocyte interaction	NSCLC brain metastases	104
	–	IL-10	B cell-derived GABA → anti-inflammatory macrophages → inhibit CD8+ cytotoxicity	BCC/RCC	105
	GABAB	β-Catenin	Activate GABAB → inhibit GSK-3β → ↑ β-catenin → proliferation + CD8+ exclusion	NSCLC/CRC	106
5-HT	SERT	RHOA–ROCK–YAP	5-HT uptake → activate YAP → promote CRC growth	CRC	107
	HTR2B	ERK	Inhibition → ↓ ERK → suppress proliferation	CRC	108
	HTR2B	AKT/mTOR–CREB–ZEB1	↑ EMT → enhance migration & metastasis	CRC	109
	HTR2B	PI3K/AKT/mTOR–HIF1α	↓ lipid peroxidation → anti-ferroptosis → promote gastric cancer	Gastric	110
	–	NETs-H3Q5ser/H3cit	5-HT → NET formation → facilitate liver metastasis	NEPC	111
	–	PI3K/Akt/mTOR	↑ glycolysis → exacerbate immunosuppression	NSCLC	112
Dopamine	D1DR	NF-κB/AKT–EMT	QAP14 activates D1DR → anti-CSC & anti-metastatic effects	BC	113
	DRD1	cAMP/PI3K/AKT/CREB	↑ proliferation & metastasis	HCC	114
	DRD2	VHL–HIF-1α	DRD2 competes with VHL → ↑ HIF1α → EMT	Melanoma	115
	DRD1	LRP5–CCN4	Mechanical loading → DRD1 → ↓ CCN4 → suppress brain tumors	Brain tumors	116
	DRD3	IL-2–CD25	DRD3 → ↑ IL-2 & CD25 → CD8+ expansion	Melanoma	117
	DRD4	TGFBR–SMAD2/4–SNAI1/JAG1	↑ EMT → enhance metastatic potential	CRC	118
	DRD5	JAK1–STAT1	Enhance B-cell activation → ↑ tumor-specific T-cell cytotoxicity	BC	119
Histamine	HRH1	Ca2+–NF-κB	GSC-secreted histamine → promote angiogenesis	GBM	120
	HRH1	VISTA	HRH1 → M2-like macrophages → T-cell dysfunction	TNBC	121
	HRH3	PI3K/AKT/mTOR; MEK/ERK	↑ EMT → promote migration & invasion	NSCLC	122
	H4R	Src-TGF-β–SMAD2/3	↑ tumorsphere formation & EMT	BC	123
	–	CaMKII–CA12 endocytosis	Histamine → CA12 mislocalization (potential anti-cancer process)	NSCLC	124
NO	–	S-Nitrosylation–PFKM	Stabilize PFKM tetramer → ↑ glycolysis → growth & metastasis	OC	125
	eNOS	NO–sGC–cGMP–PKG	↑ stem-like properties & anti-androgen resistance	CRPC	126
	iNOS	–	Andrographolide ↓ iNOS → anti-proliferative & pro-apoptotic	CC	127
	–	↓ JNK/c-Jun nitrosylation	NOS blockade → ↓ S-nitrosylation of JNK/c-Jun → ↑ differentiation & chemosensitivity	MpBC	128
	–	S-Nitrosylation–NF-κB/Snail/RKIP	High NO → reverse EMT + induce ICD → anti-metastatic	BC	129
	–	Inhibit TCIPA	NO ↓ platelet activation → attenuate metastasis	BC	130
	–	↑ CRT/HMGB1; ↓ HIF-1α	NO improves hypoxia & immunity → prevent metastasis/recurrence	BC	131
	–	↑ IFN-γ/TNF-α; ↓ IL-6/IL-10	Low-dose NO donor → ↑ CD8+ & NK → tumor inhibition	Melanoma/NSCLC	132

In GBM, glutamate not only maintains redox homeostasis through glutathione (GSH) synthesis, but also induces CypB secretion and activates p-AKT signaling. This promotes glycolysis and suppresses mitochondrial function, helping tumor cells adapt to glutamine deprivation⁸⁹. Glutamate also promotes immune evasion in gliomas. The acidic microenvironment impairs immune cell function, creating a pro-tumor immune niche⁹¹. Glutamate released from nerve injury can activate oligodendrocytes to secrete IL-1β, inducing CCL5 expression, accelerating tumor progression, and driving optic glioma formation⁹².

Glutamate also plays key roles in other solid tumors. In HCC, approximately 30% of glutamine is metabolized to glutamate in the cytosol, supporting ATP production via the TCA cycle to sustain tumor growth⁹³. Liver fibrosis is a critical process in the development of hepatocellular carcinoma. Glutamate dehydrogenase (GDH) catalyzes the conversion of glutamate to α-ketoglutarate, enhancing mitochondrial energy metabolism to promote fibrosis, whereas GDH inhibition reduces ATP production and slows fibrosis and tumor progression⁹⁴. SLC7A11-mediated glutamate export acidifies the extracellular environment, which can also exacerbate the occurrence of seizures⁹⁰. In PDAC, sensory nerve endings form pseudo-synapses with cancer cells. GRIN2D-type NMDAR on tumor cells senses neuron-derived glutamate and activates calcium signaling. This enhances tumor growth, invasion, and metastasis. These pseudo-synapses also convert some co-cultured cancer cells into calcium-responsive cells, creating a feedforward loop that further promotes tumor progression⁹⁵.

Tumor cells export glutamate via the cystine-glutamate antiporter (xCT) to take up cystine for GSH synthesis. In breast cancer, xCT upregulation similarly increases extracellular glutamate, activating mGluR3 to enhance Rab27-dependent extracellular vesicle (EV) release, remodeling endosomal trafficking and generating a pro-invasive microenvironment⁹⁶. In melanoma, xCT-mediated PD-L1⁺ exosomes drive M2 macrophage polarization, reducing the efficacy of PD-1/PD-L1 checkpoint blockade therapy⁹⁷.

Glutamate further regulates immune cell phenotypes and cytokine release, reshaping the TME to promote immune evasion. In PDAC, CAF-derived Netrin G1 (NetG1) induces immunosuppression through regulation of glutamate metabolism, and NetG1 inhibition restores anti-tumor immunity⁹⁸. In HCC, glutamate promotes TAM infiltration by upregulating CCL2 expression in hepatoma cells via the NMDAR2B receptor. This reshapes the immune microenvironment and facilitates tumor progression⁹⁹. In hepatocellular sarcoma and fibrosarcoma tumors, activation of metabotropic NMDAR also induces an immunosuppressive phenotype, inhibiting T cell and NK cell activity; NMDA antagonists combined with anti-PD-1 therapy can enhance anti-tumor immunity¹⁰⁰. In melanoma, GRM4 signaling modulates anti-tumor immunity, and its inhibition enhances NK and T cell responses by promoting their activation, proliferation, and functional differentiation and strengthening anti-tumor immune efficacy¹⁰¹.

Collectively, these findings indicate that glutamate and its receptors influence the TME across multiple tumor types. Pharmacological inhibition of glutamate receptors has been shown to suppress tumor growth and metastasis, highlighting glutamate receptors as potential therapeutic targets.

3.1.2. γ-Aminobutyric acid

γ-Aminobutyric Acid (GABA) is primarily synthesized from glutamate via glutamate decarboxylase (GAD), a reaction that requires the coenzyme pyridoxal phosphate (PLP), and functions as the principal inhibitory neurotransmitter in both the CNS and peripheral tissues¹³³. GABA exerts its effects through three classes of receptors: GABA_A, GABA_B and GABA_C, which mediate either rapid or prolonged inhibitory signaling¹³⁴. Emerging evidence indicates that GABA signaling exhibits multifaceted pro-tumorigenic roles, including intracellular signaling regulation, invasion and metastasis, immune modulation, and metabolic adaptation (Table 1).

Tumor cells can upregulate GAD1 or GPT2 to reprogram glutamine metabolism, resulting in enhanced autocrine or paracrine GABA production. Activation of GABA receptors subsequently activates the Ca²⁺/PKC/CREB signaling pathway. This promotes tumor growth and metastasis. The metabolic and signaling axis has been closely associated with invasion and poor prognosis in CRC¹⁰² and breast cancer¹⁰³. In NSCLC brain metastases, GABA enhances brain colonization adaptability through the FOXA2/ABAT/GABA/NF-κB pathway and interactions with astrocytes¹⁰⁴.

GABA signaling also profoundly influences the tumor immune microenvironment. In renal cancer, B cells synthesize and secrete GABA, which drives monocytes to differentiate into IL-10⁺ anti-inflammatory macrophages and suppresses CD8⁺ T-cell cytotoxicity. This creates an immunosuppressive microenvironment¹⁰⁵. In parallel, tumor-intrinsic GABA–β-catenin signaling reduces CD8⁺ T-cell infiltration and weakens responses to immune checkpoint blockade¹⁰⁶.

Although some in vitro studies have reported that exogenous GABA inhibits CRC cell proliferation, in vivo evidence predominantly supports its pro-tumorigenic role, which likely reflects differences in receptor subtype expression and tissue microenvironment¹³⁵. Overall, GABA coordinates a pro-tumorigenic network that links intracellular metabolism, invasion, and immune regulation. Targeting key nodes such as GAD1 or specific GABA receptor subtypes may enhance immunotherapy efficacy and inhibit tumor progression across multiple cancer types¹³⁶.

3.1.3. 5-Hydroxytryptamine

In the CNS, 5-HT is primarily synthesized by dorsal raphe nucleus (DRN) neurons through the rate-limiting enzyme tryptophan hydroxylase 2 (TPH2), regulating mood, cognition, and neuronal activity¹³⁷. In the periphery, enterochromaffin (EC) cells in the gastrointestinal mucosa are the principal source of 5-HT production¹³⁸. Beyond its role as a neurotransmitter in CNS function, 5-HT participates in tumor initiation, progression, and immune regulation¹³⁹ through multiple mechanisms (Table 1).

In tumors, 5-HT binds to its GPCRs and activates multiple oncogenic signaling pathways. For instance, in CRC, the serotonin transporter (SERT) mediates cellular uptake of 5-HT, triggering the RHOA/ROCK/YAP signaling axis to drive CRC initiation and invasion. Inhibition of SERT or blockade of YAP significantly suppresses tumor growth and metastasis¹⁰⁷. Also in CRC, upregulation of the HTR2B receptor enhances ERK pathway activation, leading to increased tumor-cell proliferation and migration and faster tumor progression¹⁰⁸. Elevated HTR2B expression activates the AKT/mTOR/S6K1/CREB1/ZEB1 axis to drive EMT and metastasis, which further promotes tumor progression¹⁰⁹. In gastric cancer, 5-HT binds to the HTR2B receptor, activating the PI3K/AKT/mTOR pathway to promote tumor proliferation and invasion¹¹⁰. In neuroendocrine prostate cancer (NEPC), elevated secretion of 5-HT induces neutrophil extracellular trap (NET) formation and promotes liver metastasis¹¹¹. In NSCLC, NGF-driven neural infiltration increases intratumoral 5-HT levels. This activates the PI3K/AKT/mTOR pathway, induces metabolic reprogramming and immune suppression, and accelerates tumor progression¹¹².

At the epigenetic level, 5-HT mediates serotonylation of histone H3 at glutamine 5 (H3Q5ser), promoting the conversion of CAFs into an inflammatory phenotype (iCAFs). This process enhances tumor cell invasiveness, macrophage M2 polarization, and other immunosuppressive features. Notably, silencing SLC22A3 can reverse these 5-HT-driven tumor-promoting effects¹⁴⁰.

3.1.4. Dopamine

Dopamine synthesis depends on the sequential activity of tyrosine hydroxylase (TH) and aromatic l-amino acid decarboxylase (AADC) and occurs predominantly in the substantia nigra pars compacta, which regulates motor control, and the ventral tegmental area, which is involved in reward and emotional processing^141,142. Increasing evidence indicates that dopamine exerts complex, context-dependent effects in cancer, functioning as both a tumor promoter and an immune activator (Table 1).

Dopamine can exert anti-tumor effects through specific receptor subtypes. In breast cancer, D1DR activation inhibits the NF-κB–AKT pathway, downregulating EMT programs and CSC phenotypes and suppressing metastatic dissemination¹¹³. In HCC, D1DR activation enhances tumor cell proliferation and metastasis by activating the cAMP/PI3K/AKT/CREB signaling pathway, which accelerates cancer progression and predicts poor prognosis¹¹⁴. In melanoma, DRD2 activates the oxygen-independent HIF-1α pathway, enhances EMT in tumor cells, and promotes malignant progression¹¹⁵. DRD1/2 signaling has also been implicated in long-range regulation of brain tumors through the LRP5–CCN4 axis, increasing their growth capacity¹¹⁶. In melanoma, high DRD3 expression in CD8⁺ T cells stimulates IL-2 secretion, boosting anti-tumor immunity and enabling effective clearance of tumor cells¹¹⁷. In CRC, DRD4 activation induces transcription of SNAI1 and JAG1, driving EMT and enhancing migratory and invasive potential¹¹⁸. DRD5 activation enhances the JAK1–STAT1 pathway in B cells, promoting antigen presentation and cytotoxic T cell responses, which collectively inhibit tumor progression¹¹⁹.

3.1.5. Histamine

Histamine is synthesized primarily by neurons of the tuberomammillary nucleus (TMN) in the hypothalamus via histidine decarboxylase (HDC) and is released in a circadian rhythm^143,144. Acting through four distinct histamine receptors (HRH1–HRH4), histamine participates in diverse physiological processes, including sleep–wake regulation, immune-inflammatory responses, and cancer modulation (Table 1).

In GBM, high expression of histamine promotes angiogenesis through the HRH1/Ca²⁺/NF-κB pathway, with elevated HRH1 expression correlating with pro-angiogenic tumor microenvironment remodeling and poor patient prognosis^120,145. In breast cancer and melanoma, histamine promotes M2 polarization of tumor-associated TAMs through the HRH1 signaling pathway, suppressing CD8⁺ T cell function and promoting immune evasion and tumor progression¹²¹. In NSCLC, HRH3 overexpression activates the PI3K–AKT–mTOR and MEK–ERK pathways, enhancing tumor cell invasiveness¹²². In breast cancer, HRH4 activation promotes EMT and tumorsphere formation via TGF-β signaling, supporting cancer stemness¹²³. Additionally, histamine signaling regulates pH-dependent metastatic and drug-resistant phenotypes in lung cancer throughw carbonic anhydrase 12 (CA12)-mediated endocyto jit3is, which reverses acidity via CaMKII signaling¹²⁴.

Pharmacological inhibition of histamine receptors has demonstrated anti-tumor effects: HRH1 antagonists can inhibit tumor cell proliferation¹⁴⁶, HRH2 antagonists can influence cancer progression in various tumor types by modulating multiple mechanisms, including proliferation, apoptosis, immunity, and the tumor microenvironment¹⁴⁷, while HRH3 antagonists effectively suppress the growth of triple-negative breast cancer¹⁴⁸.

3.1.6. Nitric oxide

Nitric oxide (NO) is an atypical gaseous neurotransmitter synthesized from l-arginine via nitric oxide synthase (NOS), and is expressed in diverse cell types including neurons, endothelial cells, immune cells, and tumor cells¹⁴⁹. NOS comprises three isoforms: neuronal nitric oxide synthase (nNOS), endothelial nitric oxide synthase (eNOS) and inducible nitric oxide synthase (iNOS). NO transduces signals primarily through activation of soluble guanylyl cyclase (sGC) or by inducing protein S-nitrosylation, and exhibits concentration-, spatiotemporal-, and tissue-specific dual effects in cancer biology (Table 1).

nNOS mainly regulates tumor metabolic reprogramming via S-nitrosylation and other mechanisms, promoting cancer cell proliferation, metabolic adaptation, and metastasis. In ovarian cancer, nNOS-mediated S-nitrosylation of PFKM at Cys351 stabilizes its tetramer and promotes glycolytic metabolic reprogramming, which enhances tumor-cell proliferation and metastasis¹²⁵. eNOS mainly promotes angiogenesis, improves tumor blood supply, and modulates the redox state of the tumor microenvironment, supporting tumor growth and invasion. In prostate cancer, for example, the upregulation of eNOS activates the NO/sGC/cGMP/PKG signaling axis, driving CSC self-renewal and promoting tumor growth¹²⁶. iNOS may induce DNA damage, promote tumor progression, or mediate immunosuppression¹⁵⁰. In cervical cancer, iNOS is upregulated to promote tumor proliferation, migration, and survival, and andrographolide inhibits iNOS, suppressing cell proliferation, inducing G1–S cell cycle arrest, and promoting apoptosis¹²⁷. In metaplastic breast cancer, iNOS promotes S-nitrosylation of the JNK/c-Jun complex, maintaining tumor stemness and EMT, enhancing chemoresistance and malignant progression¹²⁸. Additionally, in breast cancer, NO activates the NF-κB/Snail signaling pathway, inducing EMT and enhancing the metastatic potential of tumor cells¹²⁹.

Despite its role in promoting tumor progression, NO can exert anti-tumor effects under certain conditions. Endogenous NO inhibits tumor cell-induced platelet aggregation (TCIPA), reducing the adhesion and extravasation of circulating tumor cells and consequently suppressing metastasis. Moreover, the co-delivery of NO donors and chemotherapy drugs via nanomaterial-based platforms has been shown to effectively inhibit metastatic colonization¹³⁰.

As a gasotransmitter, NO can enhance tumor thermosensitivity. NO-releasing micro/nanoplatforms downregulate HIF-1α, improve antigen presentation, and induce immunogenic cell death and prevent tumor metastasis and recurrence¹³¹. In melanoma and lung carcinoma models, low-dose NO donors modulate T cell and macrophage activity to inhibit tumor growth¹³². Broad-spectrum NO pathway inhibitors, such as L-NMMA, have shown anti-metastatic potential in clinical studies, supporting NOS inhibition as a viable therapeutic strategy for tumor metastasis¹⁵¹. Therefore, targeting NO signaling pathways may present a promising approach for combating tumor progression¹⁵².

3.2. Neuropeptides

Neuropeptides are better understood as components of broader neuropeptidergic systems, in which coordinated ligand–receptor networks shape tumor progression, immune regulation, and therapeutic response¹⁵³.

3.2.1. Gonadotropin-releasing hormone

Gonadotropin-releasing hormone (GnRH) is synthesized by hypothalamic neurons and released in a pulsatile manner into the pituitary portal circulation, stimulating the secretion of follicle-stimulating hormone (FSH) and luteinizing hormone (LH), which subsequently regulate the synthesis of sex steroids such as estrogen and testosterone¹⁵⁴. In hormone-related malignancies, GnRH exhibits organ-specific, signal-diverse, and rhythm-dependent oncogenic effects (Table 2155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175).

Table 2.

Neuropeptides.

Neuropeptide	Target/receptor	Signaling pathway/axe	Function	Cancer type	Ref.
GnRH	GnRH-R	–	Genetic variants in GnRH pathway predict NSCLC survival	NSCLC	155
	–	JNK↑/PI3K–PKB↓	Pro-apoptotic in androgen-independent prostate cancer	PCa	156
	GnRH-II receptor	Caspase-3	Apoptosis; inhibit xenograft growth	EOC	157
Kisspeptin	GPR54	–	Liposomal targeting via Kisspeptin/GPR54 → ↑ tumor accumulation	BC	158
	–	miR345–KISS1–E-cadherin	↓ KISS1 → ↑ E-cadherin → anti-brain metastasis	BC	159
	KISS1R	β-arr1/2–Raf–MEK–ERK–NF-κB	↑ growth, angiogenesis, EMT → pro-metastatic	PCa	160
Dynorphin	KOR	p38	Nuclear p38 → apoptosis & growth inhibition	Glioma	161
	KOR	KOR–ER–XBP1	↑ apoptosis → inhibit ER+ breast cancer	ER+ BC	162
	KOR	Urea cycle disruption- NH3 toxicity	KOR agonist U69593 → mitochondrial damage → anti-TNBC	TNBC	163
	DOR	JAK1/2–STAT3	↑ EMT → promote metastasis	BC	164
	MOR	–	↑ proliferation, colony formation, invasion, migration	HNC	165
VIP	VPAC2	VPAC2–PIWIL2–c-Myc; VPAC2–SP1–TGF-β1	↑ Clonogenic growth + inhibit T cells → pro-PDAC	PDAC	166
	VIP-R	PD1	VIP-R antagonist + anti-PD-1 synergize → ↑ tumor-specific T cells	PDAC	167
	VPAC1	RA–CCR9–ILC3	Maintain intestinal ILC3 & antibacterial immunity	Non-tumor	168
	VPAC2	IL-22-liver steatosis	ILC3–VPAC2 deletion → ↑ IL-22 → reduce fatty liver	Non-tumor	169
Neurotensin	NTSR1	IGF2BP3-m6A-NTSR1	↑ mRNA stability → proliferation, migration, irinotecan resistance	LUAD	170
	NTSR1	SLCO4A1–AS1–TOX4–NTSR1	↑ Cytoskeleton remodeling → migration & invasion	LUAD	171
	NTSR1	NTSR1–PI3K	↑ invasion + gemcitabine resistance	PDAC	172
Substance P	TACR1	SP–TACR1-ssRNA–TLR7	Neuronal SP → promote growth, invasion, metastasis; TACR1 antagonist aprepitant inhibits	BC	173
	Capsaicin (CAP)	SP–HSCs–HCC axis	CAP blocks SP-induced HSC activation → reverse EMT, ↓ drug resistance, ↓ angiogenesis, ↓ ECM deposition	HCC	174
	NK1R	ERK–c-Myc signaling	High NK-1R predicts poor survival; NK-1R antagonists induce ER stress → ↓ ERK–c-Myc → apoptosis & 5-FU sensitization	CRC	175

GnRH also exerts significant effects in other solid tumors. For instance, variants in GnRH signaling pathway genes have been associated with improved survival in patients with NSCLC, potentially involving altered ITPR3 expression and downstream GnRH-related intracellular signaling¹⁵⁵. In castration-resistant prostate cancer (CRPC), GnRH activates the Gαi/cAMP inhibitory signaling pathway through its receptor. This counteracts the persistently active androgen/AR axis and exerts anti-proliferative, pro-apoptotic, anti-metastatic, and anti-angiogenic effects¹⁷⁶. GnRH analogs also trigger apoptosis in androgen-independent prostate cancer cells by inhibiting the PI3K/PKB pathway and activating the c-Src/MLK3/JNK signaling cascade, which exerts a direct antitumor effect¹⁵⁶. In endometrial cancer, GnRH and its receptor regulate cell proliferation, cell cycle, and apoptosis through an autocrine system and inhibit tumor growth^157,177.

GnRH and its receptors are expressed in various solid tumors and regulate tumor cell proliferation, apoptosis, and metastasis through autocrine or paracrine mechanisms. Clinically, GnRH agonists are widely used for hormone-dependent tumors. Continuous stimulation of pituitary GnRH receptors causes receptor desensitization. This suppresses LH and FSH secretion, lowers circulating sex hormone levels, and indirectly inhibits hormone-dependent tumor growth. For example, GnRH-a can effectively provide reversible ovarian function suppression in premenopausal breast cancer patients, and clinical trials have shown that combining GnRH-a with aromatase inhibitors or tamoxifen improves patient survival¹⁷⁸. In prostate cancer, GnRH-a (such as the oral drug relugolix) can achieve rapid and complete suppression of testosterone levels, leading to a faster decline in prostate-specific antigen (PSA) and providing patients with a more controllable and optimized androgen-deprivation therapy¹⁷⁹.

In cancer therapy, the GnRH system is regarded as a promising therapeutic target, with applications extending beyond reproductive-related diseases to non-reproductive conditions. Future research directions may include combining GnRH-based treatments with other therapies, such as androgen receptor antagonists, to further enhance therapeutic efficacy and reduce adverse effects.

3.2.2. Kisspeptin

Kisspeptin is predominantly localized to the hypothalamic arcuate nucleus and the anteroventral periventricular nucleus (AVPV), functioning as an upstream regulator of GnRH rhythmicity^180,181. It exerts dual roles in tumor biology, with its effects largely dependent on tissue type, receptor expression status, and splice variant composition (Table 2).

In solid tumors, Kisspeptin signaling demonstrates pro-tumorigenic potential. In breast cancer, high KISS1R expression has been exploited for targeted liposomal drug delivery, prolonging median survival in murine models¹⁵⁸. In breast cancer brain metastases, KISS1 dysregulation is associated with impaired E-cadherin function, potentially facilitating loss of cell–cell adhesion and accelerating brain colonization¹⁵⁹. In metastatic prostate cancer, the KISS1/KISS1 receptor (KISS1R) axis recruits β-arrestin1/2 to activate ERK and NF-κB pathways, leading to proliferation, migration, invasion, and metastatic potential of prostate cancer cells¹⁶⁰.

Conversely, Kisspeptin has been reported to exhibit tumor-suppressive functions. Early studies in melanoma identified KISS1 as a metastasis suppressor gene, capable of inhibiting cell migration and preventing distant dissemination¹⁸². The functional duality of Kisspeptin is likely linked to its splice variant diversity, such as kisspeptin-10 and kisspeptin-54, which can activate distinct downstream pathways in different tissue contexts to elicit either oncogenic or tumor-suppressive effects¹⁸³.

3.2.3. Dynorphin

Dynorphin, a member of the endogenous opioid peptide family, is widely distributed across multiple regions of the CNS, including the ventral pallidum and hypothalamus¹⁸⁴. It is primarily involved in pain perception, stress response, and emotional regulation. Emerging evidence indicates that dynorphin and its cognate opioid receptors play pivotal roles in tumorigenesis, metastatic dissemination, and immune evasion. Dynorphin is primarily the endogenous ligand for the κ-opioid receptor (KOR), but it can also act on μ-opioid receptor (MOR), δ-opioid receptor (DOR) and non-classical receptors (Table 2).

KOR activation may inhibit proliferation and induce apoptosis. Specifically, KOR exerts tumor-suppressive effects in glioma by inducing receptor internalization, activating the p38 pathway, and regulating downstream gene expression to inhibit proliferation and promote apoptosis¹⁶¹. In estrogen receptor positive breast cancer, KOR agonists exert anti-tumor effects by activating the KOR/ER/XBP1 signaling axis, suppressing proliferation and promoting apoptosis¹⁶². In triple-negative breast cancer, the KOR agonist U69593 suppresses tumor growth and metastasis by disrupting urea cycle function, causing mitochondrial dysfunction and ATP depletion, which induces cancer cell apoptosis¹⁶³.

MOR and DOR promote tumor progression. In breast cancer, activation of the MOR triggers the JAK/STAT3 signaling cascade, inducing EMT and facilitating breast cancer metastasis¹⁶⁴. In HNSCC, the use of opioid analgesics, particularly MOR agonists, has been associated with accelerated disease progression¹⁶⁵. In urological malignancies, DOR expression correlates with poor prognosis¹⁸⁵.

Clinically, the administration of MOR agonists for analgesia in cancer patients may, while alleviating pain, inadvertently promote tumor progression in certain settings. From an immunomodulatory perspective, opioid peptides, including dynorphin, can suppress natural killer (NK) cell and T cell function, attenuating host antitumor immunity^186,187. This mechanism may contribute to immune escape and distant metastasis in malignancies. These findings highlight the need for cautious opioid use in oncology and support the development of selective, non-pro-tumorigenic opioid-based analgesics as a critical goal in cancer pain management.

3.2.4. Vasoactive intestinal peptide

Vasoactive Intestinal Peptide (VIP) is a neuropeptide secreted by neurons of the suprachiasmatic nucleus (SCN) in the hypothalamus, functioning as a master regulator of circadian rhythmicity. In the PNS, particularly within the ENS, VIP modulates smooth muscle contractility, epithelial secretion, and local immune homeostasis^188,189. Recent evidence indicates that VIP, via its receptors VPAC1 and VPAC2, exerts pro-proliferative and immunosuppressive effects in tumors (Table 2).

VIP acts on tumor-infiltrating T cells, inducing the expression of exhaustion markers and attenuating immune responses. In PDAC, VIP engages VPAC2 in an autocrine signaling loop to activate the TGF-β1 pathway, suppress T cell function and promote cancer cell growth¹⁶⁶. VIP/VIPR signaling can also suppress anti-tumor immunity by impairing T cell recruitment, activation, and function¹⁶⁷. Beyond T cell regulation, VIP also modulates group 3 innate lymphoid cell (ILC3) recruitment and function, stimulating IL-22 secretion to maintain intestinal barrier integrity and antimicrobial defense via VPAC1¹⁶⁸. Conditional deletion of Vipr2 in gut-resident ILC3 in mouse models significantly reduces IL-22 production, leading to hepatic steatosis and systemic metabolic dysregulation¹⁶⁹, further underscoring the central role of VIP as a neuro–immune–metabolic nexus.

3.2.5. Neurotensin

Neurotensin (NTS) is a tridecapeptide neuropeptide secreted primarily by neurons of the lateral hypothalamic area and by enteroendocrine cells of the gastrointestinal tract, acting as a key regulator of the neuro–gut–immune axis. In multiple solid malignancies, NTS exerts potent oncogenic effects through its high-affinity receptor neurotensin receptor 1 (NTSR1), orchestrating cell signaling and tumor immune microenvironment (TIME) remodeling (Table 2).

In LUAD, IGF2BP3-mediated m6A methylation stabilizes NTSR1 transcripts¹⁷⁰, promoting malignant progression and irinotecan resistance by enhancing proliferation, migration and invasion while inhibiting apoptosis. TOX4 also promotes cancer cell migration and invasion by mediating cytoskeletal remodeling downstream of NTSR1 in LUAD¹⁷¹. In PDAC, neuron-derived NTS promotes tumor invasiveness, recruitment, and gemcitabine resistance via NTSR1-mediated activation of the PI3K pathway¹⁷². Pharmacological blockade of NTSR1 has demonstrated robust anti-proliferative and anti-metastatic activity in diverse preclinical models, underscoring the NTS–NTSR1 axis as a pivotal regulatory hub in solid tumors and a promising target for translational intervention.

3.2.6. Substance P

Substance P (SP) is a prototypical neuropeptide synthesized and released by sensory neurons in the dorsal root ganglia (DRG), exerting its biological effects primarily via the NK1R. As a central mediator of pain transmission, inflammation, and stress responses, SP orchestrates a broad spectrum of protumorigenic processes within the TME by modulating neuro–immune–vascular interactions, including enhanced cell proliferation, migration and distant metastasis (Table 2).

In breast cancer, tumor-activated sensory neurons release SP, which induces ssRNA/TLR7 signaling in tumor cells to promote metastatic progression. SP also promotes tumor growth, invasion, and metastasis by activating TACR1 on breast cancer cells, inducing death of TACR1-high cells, and triggering ssRNA/TLR7-mediated pro-metastatic gene programs in neighboring tumor cells³¹. In HCC, peripheral nerve-derived SP activates hepatic stellate cells, promoting tumor invasion, metastasis, EMT, and drug resistance¹⁷⁴.

Given its multifaceted role in tumor progression, NK1R has emerged as a potential therapeutic target. Notably, certain NK1R antagonists (such as aprepitant, already clinically approved for chemotherapy-induced nausea and vomiting) are currently being evaluated for anticancer indications, underscoring the translational potential of targeting the SP–NK1R axis¹⁷⁵.

3.3. Neurexins and neuroligins

Neurexins (NRXNs) and neuroligins (NLGNs) are two closely related families of neuronal proteins that play a critical role in the nervous system. NRXNs are located on the presynaptic membrane and, as axonal adhesion molecules, pair with postsynaptic NLGNs to form the trans-synaptic neurexin–neuroligin signaling axis. These proteins are primarily involved in the formation and regulation of synaptic connections between neurons. However, their functions also extend to tumor biology, where they participate in processes such as tumor progression, metastasis, and differentiation. By mediating neuronal signaling and neuro–tumor interactions, they influence cancer cell growth, migration, and therapeutic resistance (Table 2).

In GBM, NRXN1 regulates tumor cell fate toward a neuron-like lineage, suppresses malignant phenotypes, and improves prognosis, serving as a key neuroligin-related factor in tumor differentiation control¹⁹⁰. In prostate and pancreatic cancers, NLGN1 cooperates with the neurotrophic factor GDNF to promote cancer cell migration and invasion along nerves. In GBM, NLGN3 functions as a Wnt/β-catenin-upregulated secreted oncogenic factor that induces and sustains cancer stem cell properties, promoting tumor progression and therapeutic resistance¹⁹¹. In gliomas, neuronal activity drives tumor progression through NLGN3. NLGN3 also drives tumor growth, proliferation, and migration by recruiting and activating Gαi1/3 to mediate downstream AKT/mTORC1 and ERK signaling¹⁹². In melanoma, downregulation of NLGN4X activates HIF1A signaling through suppression of its downstream target VBP1, which promotes invasion and metastasis and drives malignant progression¹⁹³.

ADAM10 inhibition by GI254023X abrogated all cellular sources of NLGN3 and attenuated tumor proliferation and growth in mouse models bearing pediatric glioblastoma or diffuse intrinsic pontine glioma orthotopic xenografts¹⁹⁴. These findings imply that anti-neurogenic drugs targeting NRXN–NLGN signaling may offer novel strategies to block tumor progression, representing a potential therapeutic approach for cancer treatment^195,196.

4. Tumor-induced modulation of neurons and neural circuits

Tumors actively shape their surrounding neural microenvironment, establishing pro-tumor neural circuits through multiple mechanisms, including induction of neurogenesis, modulation of neuronal and glial plasticity, secretion of neurotrophic and axon guidance factors, enhancement of neuronal excitability, and regulation of synaptic coupling. These processes remodel pre-existing nerve fibers and form supportive “neural niches” that promote tumor growth, immune evasion, and metastasis¹⁹⁷. They also enable tumor cells to acquire neuron-like phenotypes with electrophysiological excitability and synapse-dependent invasive behavior¹⁹⁸. Furthermore, long-range signaling and the gut microbiota–gut–brain axis provide systemic feedback that reinforces neural excitation, immune suppression, and tumor dissemination (Fig. 7). Collectively, these mechanisms highlight the central role of tumors in actively modulating neural circuits to drive their own progression and metastatic potential.

Figure 7.

Tumor-induced modulation of neurons and neural circuits. Schematic representation of how tumors influence neurogenesis through various molecular pathways. Neurogenesis and neural plasticity: in gastric and colorectal cancers, CSCs differentiate into SNs (TH⁺) and PSNs (VAChT⁺), forming neural networks for tumor neurogenesis. In oral cancer, TP53 deletion downregulates miR-34a, driving adrenergic transdifferentiation of trigeminal nerves to form NE-secreting fibers that promote tumor growth. In the NF1 mouse model, Nf1 mutations reduce HCN channel function, enhancing neuronal excitability and pro-tumor paracrine factor secretion to stimulate Schwann cell proliferation and tumor formation. In GBM, tumor cells regulate neuronal activity, form synaptic inputs to trigger Ca²⁺ signals, promote microtubule formation, and accelerate invasion. Neurotrophic and axon guidance: NGF, BDNF, ACh, and Netrin-1 regulate PNI, proliferation, and axonal growth. LIF, Gal-3, EVs, and miR-21-5p modulate immunosuppression, neural plasticity, and neural infiltration. Neural remodeling: In Schwann cells, Bap1 deletion leads to abnormal spread of H2AK119ub1 and H3K27me3, causing myelin instability and axonal degeneration. In PDAC, pCAFs produce high lactate levels; tumor cells undergo H3K18 lactylation to activate nerve invasion-related genes (e.g., L1CAM and SLIT1) and promote perineural invasion. In pediatric EPN, ETV5 inhibits NPY expression via repressive chromatin states, while NPY upregulation inhibits tumor progression via synaptic remodeling. Neuronal activity: electrical signals mediated by mTOR, NE, and AMPA receptors drive syncytial electrical propagation, and tumor cell electrophysiological excitability, which together regulate tumor proliferation. Microbiota–gut axis: SCFAs, tryptophan (via AMPK–SIRT1), and inflammatory genes contribute to tumorigenesis.

4.1. Tumor-driven neurogenesis and neural plasticity

Cancer progression drives neurogenesis and stimulates neural plasticity¹⁹⁹. In gastric and colorectal cancers, a subset of cancer stem cells can differentiate into tyrosine hydroxylase (TH)-expressing sympathetic neurons and vesicular acetylcholine transporter (VAChT)-expressing parasympathetic neurons, forming functional neural networks involved in tumor neurogenesis²⁰⁰. In oral cancer, studies have shown that TP53 loss drives adrenergic transdifferentiation of tumor-associated trigeminal sensory nerves by downregulating the microRNA miR-34a, resulting in the formation of newly generated adrenergic nerve fibers. These transdifferentiated nerve fibers can release signaling molecules such as noradrenaline, markedly promoting tumor growth²⁰¹. In GBM, tumor cells can regulate neuronal activity to induce synaptic input onto themselves, triggering Ca²⁺ signaling that promotes the de novo formation of tumor microtubes and accelerates invasion, indicating that tumors exert multi-level control over their own dissemination and heterogeneity by manipulating neurons²⁰². In Neurofibromatosis type 1 (NF1) mouse models, tumor-prone Nf1 mutations enhance neuronal excitability by reducing HCN channel function, leading to increased secretion of tumor-promoting paracrine factors that stimulate Schwann cell proliferation and tumor formation²⁰³. Epigenetic and metabolic reprogramming shape Schwann tumor cell states and remodel the immune microenvironment by activating neural crest or nerve injury-related pathways, and can be induced by radiotherapy to drive reversible conversion between tumor subtypes²⁰⁴.

4.2. Tumor-driven epigenetic and neural remodeling

Tumors can manipulate epigenetic and metabolic regulatory pathways to reprogram transcription and induce phenotypic changes in neurons or glial cells. This increases their plasticity to support tumor growth and invasion.

In Schwann cells, the loss of Bap1 leads to aberrant spreading of Polycomb-associated epigenetic marks such as H2AK119ub1 and H3K27me3, which prematurely activates nerve injury and neural crest-related gene programs, disrupts the myelination transcriptional network, and ultimately results in myelin instability and axonal degeneration²⁰⁵. In PDAC, perineural invasion-associated cancer-associated fibroblasts (pCAFs) enhance glycolysis to produce high levels of lactate, creating a lactate-rich tumor microenvironment. Lactate absorbed by pancreatic cancer cells induces histone H3K18 lactylation, an epigenetic modification that activates neural invasion-related genes such as L1CAM and SLIT1 and promotes perineural invasion. Meanwhile, acetylation of the glycolytic enzyme GAPDH in pCAFs increases its enzymatic activity and glycolytic flux, further amplifying lactate production and providing metabolic and epigenetic support for neural invasion²⁰⁶.

Furthermore, in pediatric ependymoma (EPN), histone serotonylation during tumorigenesis converts neuronal signals into epigenetic changes in tumor cells, activating core developmental transcription factors and remodeling chromatin states to promote tumor progression. ETV5 suppresses NPY expression by enhancing repressive chromatin states. In contrast, neuropeptide Y (NPY) upregulation inhibits tumor progression and excessive neural network activity through synaptic remodeling. These findings reveal how neuronal signaling, histone modification, and developmental programs interact to drive brain tumor malignancy²⁰⁷.

This perspective suggests that targeting specific epigenetic modifications could effectively inhibit tumor progression driven by neural plasticity, providing a potential strategy for precision therapy.

4.3. Tumor-driven neurotrophic and axon guidance signaling

Tumor cells can modulate intratumoral neural remodeling by secreting neurotrophic factors and axon guidance molecules, which enhance the growth, branching, and plasticity of pre-existing nerve fibers within the tumor microenvironment and form a supportive “neural niche” that facilitates tumor proliferation, immune evasion, and metastasis.

In oral and salivary gland tumors, NGF promotes tumor cell scattering and migration by activating the PI3K/AKT signaling pathway and enhances perineural invasion²⁰⁸. In pancreatic cancer, tumor cells secrete NGF, which promotes the growth and proliferation of pancreatic nerve plexus fibers, forming dense neural networks that exacerbate pancreatic neuropathy and facilitate tumor progression²⁰⁹. The HGF/c-Met signaling pathway promotes pancreatic tumor secretion of NGF through activation of the mTOR/NGF axis and TrkA and p75NTR receptors, recruits nerve fibers to form PNI, accelerates nerve growth, and facilitates tumor progression^210,211. In gliomas, BDNF secreted by neurons acts on TrkB receptors on the tumor cell surface, activating downstream signaling pathways such as CAMKII, regulating the membrane localization of AMPAR and synaptic plasticity. This enhances neuron-tumor synaptic function and promotes tumor proliferation and progression¹³. In intrahepatic cholangiocarcinoma (ICC), acetylcholine activates the CAMKII/GSK3β/β-catenin signaling pathway through the CHRNA5 receptor and simultaneously induces BDNF expression, promoting nerve infiltration and forming an acetylcholine/BDNF feedforward axis that drives tumor metastasis and enhances chemoresistance²¹².

In precancerous lesions of pancreatic cancer, the axon guidance molecule Netrin-1 is secreted by pancreatic cells and promotes sympathetic axon growth and branching through the DCC receptor, driving neural plasticity and reshaping local cellular dynamics, with major effects on the tumor microenvironment and early cancer progression²¹³.

Tumors also secrete neuroregulatory factors (Gal3 and LIF) to activate the HPA axis and PVN neurons, inducing systemic and local immunosuppression and reinforcing a positive feedback loop between nerves and the tumor. For example, in lung cancer, prostate cancer, colorectal cancer, and triple-negative breast cancer, tumors secrete LIF and Gal3 to activate the central nervous system and induce sympathetic nerve-mediated immunosuppression, thereby collectively exploiting the nervous system to promote tumor progression²¹⁴. Extracellular vesicles (EVs) serve as long-range signaling mediators, enhancing neural plasticity and facilitating tumor dissemination²¹⁵. In CRC, tumor-derived exosomal miR-21-5p induces Schwann cells to secrete NGF, and the tumor-derived NGF both promotes Schwann cell proliferation and nerve fiber expansion and activates tumor cells via the TrkA/ERK signaling pathway, forming a positive feedback loop that jointly drives neural infiltration, tumor growth, and metastasis²¹⁶.

4.4. Tumor-induced neuronal excitability and synaptic coupling

Increasing evidence indicates that tumors can modulate neuronal activity and electrical signaling, either by altering the plasticity of neighboring neurons or by acquiring intrinsic electrical excitability. These reciprocal interactions promote tumor proliferation, invasion, and metastasis.

In gliomas, tumors drive pathological plasticity in neighboring neurons by enhancing mTOR signaling, including decreased dendritic spine density, altered expression of synapse-related genes, and functional neuronal hyperexcitability, which exacerbates neural network abnormalities and may contribute to tumor-associated symptoms; these neuronal changes can be reversed by mTORC1/2 inhibitors²¹⁷. In SCLC, NE cells acquire electrical excitability, and their action potential firing directly drives tumor malignancy, accompanied by a reliance on oxidative phosphorylation and metabolic support from non-NE cells, forming a tumor–autonomous vicious cycle that enhances proliferative capacity and metastatic potential²¹⁸. In gliomas, neurons form bona fide neuron-glioma synapses with tumor cells mediated by AMPA receptors, allowing electrical signals to propagate synchronously through tumor microtube networks and promoting tumor proliferation and invasion⁹.

Overall, these results demonstrate the active role of tumors in regulating neuronal activity and electrical signaling, suggesting that targeting neuron–tumor electrical interactions may represent a novel strategy to inhibit cancer progression.

4.5. Gut microbiota–gut–brain axis feedback in tumorigenesis

The gut microbiota profoundly influences neuro–tumor interactions via the microbiota–gut–brain axis. Metabolites produced by gut microbes, such as short-chain fatty acids (SCFAs), glutamate, and glutamine, can cross the blood–brain barrier or act on the ENS to modulate neuroinflammation, immune responses, and TME formation.

For example, SCFAs can regulate glial cell maturation and influence neuroinflammation and nociceptive sensitivity^219,220. In dextran sodium sulfate (DSS)-induced colitis models, inflammatory gene expression in microglia is elevated, increasing colorectal cancer susceptibility²²¹. Furthermore, intestinal tryptophan maintains gut homeostasis by regulating metabolic pathways, preserving the integrity of the intestinal mucosal barrier, shaping the gut microbiota composition, and modulating AMPK–SIRT1-mediated autophagy, which confers anti-inflammatory and protective effects²²².

Overall, the gut microbiota modulates the inflammatory environment through neuro-immune regulatory mechanisms mediated by microbial metabolites. This highlights potential intervention points at the level of metabolite production, neural signaling, or immune modulation, suggesting that microbial metabolic pathways may serve as novel targets for precision cancer therapy²²³.

5. Therapeutic targeting of neurotransmitter and neuropeptide receptors in cancer

Pharmacological targeting of neurotransmitter and neuropeptide receptors is emerging as an important translational direction in cancer neuroscience (Table 3224, 225, 226, 227, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255 and 4256, 257, 258, 259, 260, 261, 262, 263, 264, 265, 266, 267, 268, 269, 270, 271, 272, 273, 274, 275, 276, 277, 278, 279, 280, 281, 282, 283). Current efforts follow two main strategies: repurposing established drugs for cancer therapy, and developing new small molecules, blocking antibodies, and radioligands against defined neurotransmitter or neuropeptide receptors.

Table 3.

Pharmacological agents targeting classical neurotransmitter signaling in cancer and translational implications.

Signaling axis	Target	Pharmacological agent	Therapeutic implication	Ref.
Norepinephrine/epinephrine	ADRB1 ADRB2	ADRB antagonists: propranolol (GC), atenolol (CRC), carvedilol (BC), ICI-118,551 (HNSCC)	Clinically advanced repurposing strategy.	224, 225, 226, 227
Acetylcholine	CHRM	CHRM antagonists: Atropine (CRC), darifenacin (GBM), oxybutynin (GC), tiotropium (GBM)	Druggable but highly context-dependent.	228, 229, 230
Glutamate	NMDA receptor	NMDA receptor antagonist: Memantine (BC), ketamine (CRC), MK-801 (LC), ifenprodil (PC), D-APV	Potential repurposing strategy in combination therapy.	231, 232, 233, 234, 235
GABA	GABA receptors GABA metabolic enzymes	GABAB agonist: Baclofen (HCC); GABAA modulating compounds: Muscimol, bicuculline (BC); GABA metabolic inhibitor: Vigabatrin (BC)	Requires receptor- and context-specific stratification.	236, 237, 238, 239
Dopamine	DRD2/3	DRD2/3 antagonists: Thioridazine (BC), trifluoperazine, pimozide (GBM), haloperidol (GBM), ONC201 (GBM), ONC206 (SEC)	Promising for repurposing neuropsychiatric agents.	240, 241, 242, 243, 244
Histamine	HRH1/HRH2	HRH1/HRH2 antagonists: Cimetidine (CRC), famotidine (CRC), ranitidine (BC), diphenhydramine	Potential adjunctive therapeutic strategy.	245, 246, 247, 248
Serotonin	5-HT receptors SERT	5-HT3 antagonists: Ondansetron, palonosetron (GC), granisetron (GC), tropisetron (LC); SERT inhibitors: Sertraline (BC), fluoxetine (CRC), imipramine (SCLC); 5-HT2A agonist: Psilocybin	Tractable pathway requiring subtype-specific optimization.	249, 250, 251, 252, 253, 254, 255

Table 4.

Pharmacological agents targeting neuropeptide signaling in cancer and translational implications.

Signaling axis	Target	Pharmacological agent	Therapeutic implication	Ref.
CGRP	RAMP1 CALCRL	CGRP receptor antagonists/blocking antibodies: Erenumab, galcanezumab, eptinezumab, olcegepant (AML), rimegepant	Promising for neuro-immune combination strategies.	256, 257, 258, 259, 260
Substance P	NK1R	NK1R antagonists: Aprepitant (OC), fosaprepitant (AML), netupitant, rolapitant, L-733,060, L-732,138 (HC)	Representative target for neuropeptide receptor antagonism.	261, 262, 263, 264
Neurotensin	NTSR1	NTSR1 antagonists/ligands: SR48692 (OC), 3BP-227 (PC), [225Ac]-FPI-2059, [111In]-FPI-2058	Relevant to therapy-resistant tumors.	265,266
VIP	VPAC	VPAC antagonists: VIPhyb (PC), PG 97-269 (NSCLC), PG99-465	Potential for immunotherapy-based combinations.	267, 268, 269
Dynorphin	KOR MOR	KOR agonists: U50,488 (NSCLC), U69,593; KOR antagonist: Nor-BNI; MOR antagonists: Naltrexone (OC), naloxone	Supports receptor-selective analgesic development.	270, 271, 272, 273, 274, 275
GnRH	GnRH receptor	GnRH receptor antagonists: Degarelix (PCa), TAK-385; GnRH receptor agonist: Leuprolide	Clinically established endocrine-targeting strategy, particularly in prostate cancer.	276, 277, 278
Kisspeptin	KISS1R	KISS1R agonists: Kisspeptin-10 (BC), kisspeptin-54, MVT-602; KISS1R antagonist: Peptide 234	Requires tumor-type-specific development.	279, 280, 281, 282, 283

Among classical neurotransmitter pathways, ADRB1/ADRB2 is currently the most extensively studied and the most advanced in translation. β-Blockers such as propranolol, originally used for cardiovascular disease, are now being re-evaluated for their antitumor potential²⁸⁴. Beyond adrenergic signaling, CHRM antagonists such as darifenacin²⁸⁵, 5-HT receptor- or transporter-targeting agents such as sertraline²⁸⁶, and modulators of GABA, histamine, and glutamate signaling have also provided pharmacological support. For pathways such as DRD2/3, development is not limited to repurposed neuropsychiatric agents, as newer compounds such as ONC201 have also entered investigation²⁸⁷.

A similar pattern is seen in neuropeptide pathways. NK1R antagonists such as aprepitant²⁸⁸, initially developed as antiemetics, have been incorporated into anticancer studies. Recently, Han et al.²⁸⁹ reported that magnolol acts as a novel natural NK1R inhibitor²⁸⁹. At the same time, receptor-directed small molecules, blocking antibodies, and radioligands targeting NTSR1, the CGRP axis, VPAC, KISS1R, GnRH receptor, and opioid receptor-related signaling are also under active development. Together, these findings indicate that both classical neurotransmitter receptors and neuropeptide receptors now have a substantial preclinical basis for therapeutic intervention.

From a clinical perspective, several agents targeting neurotransmitter and neuropeptide receptors have already entered clinical evaluation, as shown in Table 5, with data from https://clinicaltrials.gov. ONC201 has been tested in multiple clinical studies. Propranolol has also been evaluated in breast cancer, melanoma, and other solid tumors, and has advanced into combination studies with immune checkpoint inhibitors. GnRH receptor-targeted strategies have already established a relatively mature clinical role in prostate cancer.

Table 5.

Clinical trials targeting neurotransmitter and neuropeptide receptors in cancer.

NCT Number	Target	Agent	Trial status	Development strategy	Cancer type	Outcome
NCT02135445	GnRH receptor	TAK-385 Degarelix	Completed	New drug	PCa	Castrate testosterone levels were maintained.
NCT03085095	GnRH receptor	Relugolix Leuprolide	Completed	New drug	Advanced PCa	Rapid, sustained testosterone suppression.
NCT06631521	GnRH receptor	Darolutamide relugolix	Recruiting	Drug repositioning	Localized/locally advanced PCa	Neoadjuvant safety/feasibility ongoing.
NCT02525692	DRD2	ONC201	Completed	New drug	Recurrent GBM	Well tolerated; activity in some patients.
NCT03295396	DRD2	ONC201	Completed	New drug	Recurrent H3 K27M diffuse midline glioma	Well tolerated; durable activity reported.
NCT05009992	DRD2	ONC201	Ongoing	New drug	Diffuse midline glioma	Phase II ongoing.
NCT05580562	DRD2	ONC201	Ongoing	New drug	Newly diagnosed H3 K27M diffuse glioma	Phase III ongoing.
NCT02250781	DRD2	ONC201	Completed	New drug	Advanced solid tumors	Phase I safety shown
NCT02013492	ADRB1/ADRB2	Propranolol	Completed	Drug repurposing	Recurrent/metastatic solid tumors	Early-phase study
NCT02596867	ADRB1/ADRB2	Propranolol	Completed	Drug repurposing	BC	Neoadjuvant study
NCT05741164	ADRB1/ADRB2; PD-1	Propranolol Pembrolizumab	Recruiting	Drug repurposing	TNBC	Phase II ongoing.
NCT05968690	ADRB1/ADRB2; opioid receptor	Propranolol naltrexone	Recruiting	Drug repurposing	Advanced melanoma	Phase I ongoing.
NCT06789757	NMDA receptor; PD-L1; VEGF	Memantine Atezolizumab Bevacizumab	Recruiting	Drug repurposing	HCC	Phase II ongoing.
NCT05605522	NTSR1	[225Ac]-FPI-2059/[111In]-FPI-2058	Active, not recruiting	New drug	NTSR1+ solid tumors (PDAC, CRC, GC, PCa)	Phase I safety/PK study ongoing.
NCT04863950	SERT/NET	Imipramine hydrochloride Lomustine	Recruiting	Drug repurposing	Recurrent GBM	Phase II ongoing.

Overall, therapeutic targeting of neurotransmitter and neuropeptide receptors has progressed from mechanistic investigation to a translational stage supported by pharmacological intervention and clinical testing. At present, the ADRB1/ADRB2, DRD2, and GnRH receptor directions are advancing most rapidly, whereas pathways involving GABA, histamine, glutamate, and certain sensory neuropeptides remain less developed clinically. Nevertheless, these pathways already have corresponding pharmacological tools and therefore remain important directions for further investigation.

6. Neuro–tumor communication across the tumor ecosystems: toward precision intervention

6.1. Interpreting divergent neural effects across tumor ecosystems

Neurotransmitters and other neuroactive mediators do not exert uniform biological effects across different tumor types. Instead, their functions are shaped by the tumor ecosystem, and their biological effects depend on how neural inputs are decoded within specific cellular and microenvironmental contexts. Sympathetic adrenergic signaling is more consistently associated with tumor-promoting programs, mainly including enhanced stemness maintenance, increased angiogenesis, stromal activation, and immune suppression. In contrast, parasympathetic-, sensory-, and enteric nerve-associated signals, as well as neurotransmitter systems such as acetylcholine, GABA, dopamine, and histamine, more often display tumor type-specific, receptor-dependent, and microenvironment-dependent activities.

Neuropeptide signaling in tumors is likewise heterogeneous and can be organized into several relatively stable dominant output patterns (Table 3, Table 4 and 4). Specifically, the CGRP/VIP axis is more closely associated with neuro-immune suppression. The SP axis is more strongly linked to inflammation-driven invasion and metastatic dissemination. The NTS axis is more prominently involved in malignant progression and therapy resistance. In contrast, dynorphin and kisspeptin reflect the bidirectional and context-dependent nature of neuropeptide regulation in different tumor settings. Taken together, the functional output of neurotransmitters and neuropeptides is highly context-dependent, and their biological consequences are shaped less by ligand identity alone than by receptor subtype usage, the lineage identity of responding cells, and the local microenvironment.

6.2. Therapeutic stratification of neuro–tumor communication

These observations indicate that therapeutic targeting of neuro–tumor communication should be stratified according to the dominant mode of neural regulation, rather than applied uniformly across tumor types or neural pathways. This view is also consistent with recent efforts to frame neuro–tumor crosstalk as a therapeutically reprogrammable process, further supporting stratified intervention across distinct tumor ecosystems²⁹⁰. Pathways with relatively consistent tumor-promoting effects, particularly sympathetic adrenergic signaling, may represent broader candidates for intervention through pharmacological blockade, drug repurposing, or combination treatment²⁹¹. By contrast, parasympathetic, sensory, and neuropeptide-related programs show greater tumor type specificity and stronger dependence on receptor distribution and local microenvironmental architecture. These pathways are therefore more likely to require receptor-selective targeting, local denervation or neural ablation strategies, microenvironment-informed intervention, or the development of more precise inhibitors.

Accordingly, precision strategies should move beyond histological classification alone and incorporate neural features that reflect functional dependency, including nerve density, neurotransmitter receptor expression, neuropeptide pathway activation, and states of neural remodeling²⁹². Such features may help guide different therapeutic approaches, including direct pathway blockade, repurposing of clinically available neuroactive drugs, neuroimmune combination strategies, and the development of new agents targeting defined neural receptor axes²⁹³. In this framework, the key challenge is to identify which neural program is functionally dominant in a given tumor ecosystem, so that different modes of intervention can be matched to different forms of neural dependency.

6.3. Resolving neural dependency with emerging models and technologies

Meeting this challenge will require experimental models and analytical tools that can resolve neural dependency with greater spatial and temporal precision.

As noted above, Trace-n-Seq, which combines retrograde axon tracing with single-cell transcriptomics, provides a paradigm for dissecting the subtype-specific roles of neurons in the tumor microenvironment, although systematic neuronal atlases remain limited for most tumor types²⁹⁴. This is particularly important because neuro–tumor and neuro–immune interactions are intrinsically dynamic, involving neurotransmitter plasticity, axonal regeneration, synaptic remodeling, and circuit reprogramming, whereas many current in vivo and in vitro systems still capture only terminal states.

To address this limitation, newly developed platforms can be incorporated into a more mechanistic workflow. Brain organoid-based neural co-culture systems²⁹⁵ and microfluidic neural chips provide tractable platforms for modeling tumor–neural networks, monitoring neuronal migration, and analyzing synapse-like connectivity at higher resolution^296,297. In parallel, multimodal neuroimaging approaches integrating MRI²⁹⁸, DTI, computer-aided diagnosis (CAD)²⁹⁹, and metabolic tumor markers³⁰⁰ may help visualize neuro–tumor interactions in vivo for diagnosis, innervation mapping, and therapeutic monitoring, although their spatial resolution and dynamic tracking capacity remain limited, especially for deep-seated tumor-associated nerves.

High-resolution imaging innovations such as light-microscopy-based connectomics (LICONN)³⁰¹ and ARCHmap-blockface-VISoR³⁰² further expand the toolkit for reconstructing neural circuitry and whole-body neural architecture, creating new technical opportunities to study tumor-associated neural complexity.

Taken together, future progress in cancer neuroscience will depend on integrating mechanistic biology with high-resolution models and imaging platforms. Rather than treating neural regulation as a static background feature of the tumor microenvironment, the field should move toward defining when, where, and in which cell populations neural signaling becomes functionally dominant. Such a shift may improve biomarker-informed stratification, refine mechanism-based intervention, and strengthen the translational value of targeting neuro–tumor crosstalk.

7. Conclusions

Cancer neuroscience is redefining the role of the nervous system in the tumor ecosystem. Rather than functioning as a passive background component, neural regulation is increasingly recognized as an integral determinant of tumor progression, immune remodeling, and therapeutic response. Neuro–tumor communication does not conform to a single biological pattern, but varies across tumor contexts. This view provides a coherent basis for understanding the heterogeneity of neural effects in cancer and how distinct neural programs are incorporated into tumor evolution. Future progress will depend on more precise identification of neural dependencies across distinct tumor ecosystems and on translating this knowledge into actionable therapeutic strategies. Continued integration of mechanistic studies with emerging experimental models and spatiotemporally resolved technologies should further refine neuro–tumor dependency and support more precise neural-targeted intervention in cancer.

Author contributions

Tong Wu and Yaguang Zhang wrote the main manuscript text and Tong Wu prepared figures. Junhong Han provided valuable discussion and revised this manuscript. All authors have reviewed and approved the final manuscript.

Declaration of generative AI and AI-assisted technologies in the writing process

The authors declare that no generative AI or AI-assisted technologies were used to generate, draft, or substantively revise the content of this manuscript.

Conflicts of interest

The authors declare no conflicts of interest.