Roles of neutrophil extracellular traps in cancer immunotherapy resistance and therapeutic targeting

Abstract

Resistance to cancer immunotherapy markedly limits clinical benefit, and neutrophil extracellular traps (NETs) have emerged as critical mediators in shaping the immunosuppressive tumor microenvironment. The molecular components of NETs, including cell-free DNA, myeloperoxidase (MPO)-DNA complexes, and citrullinated histones (H3Cit)—have demonstrated significant clinical utility as diagnostic indicators and prognostic biomarkers for predicting disease progression and immunotherapy outcomes. Regulated by a complex network of tumor-derived signals and exosomes, NETs form dense physical barriers and promote T-cell exhaustion. This review elucidates the heterogeneity of NET formation and dissects the core mechanisms by which NETs mediate therapeutic resistance along two interrelated dimensions: spatial physical exclusion and biological functional modulation. Furthermore, we synthesize emerging strategies to target NETs—including inhibition of their formation, degradation of established barriers, and blockade of downstream interactions—with the aim of providing conceptual frameworks and translational directions for reshaping anti-tumor immunity and overcoming immunotherapy resistance.

Introduction

Cancer immunotherapy, particularly the application of immune checkpoint inhibitors (ICIs), has revolutionized the landscape of cancer treatment over the past decade. Although ICI use has achieved favorable clinical outcomes in various malignancies, such as lung cancer [1–3], metastatic colorectal cancer [4] and hepatocellular carcinoma(HCC) [5–7], a substantial proportion of patients either fail to respond or develop acquired resistance. This limited efficacy is primarily attributed to the immunosuppressive network established within the complex tumor microenvironment (TME) [8–11]. Within this ecosystem, neutrophils-the most abundant immune cells in peripheral blood—were long regarded as effector cells of acute inflammation. However, their role as a double-edged sword in tumor progression is increasingly recognized [12–19].

In recent years, a distinct mode of neutrophil death—neutrophil extracellular trap (NET) formation or NETosis—has emerged as a research hotspot in tumor immunology. NETs are web-like DNA–protein structures released under specific stimuli and function as a critical nexus linking inflammation and immune tolerance within the TME [20–23]. Neutrophil extracellular traps (NETs) eliminate pathogens through the synergistic effect of physical entrapment and enzymatic degradation, a unique immune defense mechanism that was first systematically characterized by the Brinkmann group in 2004 [24]. NETs are not only web-like DNA-protein complexes released by neutrophils under specific stimuli but also serve as a key nexus connecting inflammatory responses with immune tolerance within the TME [25]. Studies have indicated that NETs in the TME are not simple defensive structures but are tightly regulated across multiple dimensions by tumor-derived factors, intercellular interactions, metabolic reprogramming, and physicochemical stressors [26, 27]. Importantly, NETs impede immune cell infiltration by forming dense physical barriers and promote T-cell exhaustion through the presentation of inhibitory molecules such as PD-L1. Consequently, they block anti-tumor immune responses across physical spatial and biological functional dimensions, leading to immunotherapy resistance [28]. Targeting NETosis was demonstrated to overcome therapeutic resistance in the preclinical cancer model [29–33].

Although the mechanisms of NETs in tumor metastasis and immune evasion have become increasingly clear, the precise molecular pathways through which they cause immunotherapy resistance and the corresponding therapeutic strategies require systematic elucidation [34]. This review examines the regulatory mechanisms controlling NET formation in the TME, delineates the core mechanisms by which NETs mediate immunosuppression and therapeutic resistance, and summarizes emerging strategies targeting NETs to remodel the immune microenvironment. The aim is to provide new theoretical foundations and translational perspectives for overcoming cancer immunotherapy resistance.

NET formation and regulation in the immunosuppressive microenvironment

TME is a complex ecosystem in which the establishment of an immunosuppressive network is a central mechanism of tumor immune evasion and therapeutic resistance [16, 35–39]. NETs are not merely defensive structures but have become integral components of the immunosuppressive TME [16, 40]. The formation and regulation of NETs are subject to precise, multidimensional regulation by tumor-derived cytokines, intercellular interactions, metabolic reprogramming, and physicochemical stresses [26, 41]. Elucidating their regulatory networks is essential for identifying targeted strategies to dismantle NET structures and restore anti-tumor immunity.

NET formation and heterogeneity

NETs are composed of decondensed chromatin decorated with granular proteins and were originally described as a distinct form of cell death, termed NETosis, which enables extracellular pathogen capture and killing [42, 43]. Subsequent studies have revealed marked heterogeneity in NET formation [44–46], primarily categorized into lytic (suicidal) NETosis, which accompanies cell death, and non-lytic (vital) NETosis, which maintains cell survival [47, 48]. Distinguishing these two modes elucidates the specific roles of NETs in tumor immune evasion and provides a theoretical basis for the development of differentiated inhibitory drugs.

Lytic NETosis: a nuclear Genome-driven cell death Program

Lytic NETosis represents the classical mode of NET formation and is characterized by complete nuclear disintegration, cell membrane rupture, and eventual neutrophil death [45, 46] (Fig. 1). This process typically occurs slowly (over several hours) and constitutes a regulated death program distinct from apoptosis and necrosis [49, 50].

Fig. 1.

The two heterogeneous modes of neutrophil extracellular trap (NET) formation and their underlying mechanisms. Left: lytic NETosis (suicidal NETosis) is a nuclear genome-driven cell death program. This process depends heavily on reactive oxygen species (ROS) produced by NADPH oxidase (NOX), which activate the Raf-MEK-ERK signaling pathway. Neutrophil elastase (NE) and myeloperoxidase (MPO) translocate to the nucleus to aid chromatin decondensation. Simultaneously, peptidyl arginine deiminase 4 (PAD4) catalyzes histone citrullination, and calpain synergizes with PAD4 to degrade lamin B1, leading to nuclear envelope disassembly. Ultimately, Gasdermin D (GSDMD) pores mediate cell membrane rupture, releasing nuclear DNA-rich NETs and resulting in neutrophil death. Right: Vital/Non-lytic NETosis allows neutrophils to release NETs while maintaining membrane integrity and viability. Stimulation via TLR4 or complement C5a receptors triggers mitochondrial membrane hyperpolarization and mitochondrial ROS generation. NETs released through this pathway primarily contain mitochondrial DNA (mtDNA), which can activate the type I interferon (IFN) response via the STING pathway

Mechanistically, lytic NETosis is highly dependent on reactive oxygen species (ROS) generated by NADPH oxidase (NOX) [31, 51–56]. ROS production activates downstream signaling pathways, including the Raf–MEK–ERK pathway, which upregulates anti-apoptotic proteins and further amplifies NOX activity [57] (Fig. 1). Neutrophil elastase (NE) is then released from azurophilic granules and translocates to the nucleus, where it partially degrades specific histones to initiate chromatin decondensation. During this process, myeloperoxidase (MPO) can synergize with NE to further loosen chromatin, even independently of its enzymatic activity [47].

Histone citrullination, catalyzed by peptidylarginine deiminase 4 (PAD4), represents a pivotal step in chromatin decondensation during lytic NETosis [58]. Concurrently, calpain synergizes with PAD4 to degrade nuclear lamins, such as Lamin B1, thereby increasing nuclear envelope permeability [59]. Following the breakdown of nuclear and granular membranes, nuclear components mix with cytoplasmic granule proteins [50]. Ultimately, NETs rich in nuclear DNA are released through cell membrane rupture, a process that depends on pore formation by Gasdermin D (GSDMD), activated by neutrophil proteases. This finding highlights mechanistic overlap between NETosis and pyroptosis [60].

Non-lytic NETosis: preservation of mitochondrial DNA (mtDNA) source and function

In contrast to lytic NETosis, non-lytic NETosis enables neutrophils to release NETs while maintaining cell membrane integrity and cellular viability [26, 42, 46]. After priming with granulocyte/macrophage colony-stimulating factor (GM-CSF), surviving neutrophils can rapidly generate NETs upon short-term stimulation of Toll-like receptor 4(TLR4) or complement C5a receptors [61].

A defining feature of non-lytic NETosis is the origin of the released DNA [26, 46]. Unlike the nuclear DNA released by the lytic pathway, NETs released by vital neutrophils mainly contain mtDNA and lack nuclear DNA [62]. This process does not compromise neutrophil lifespan, allowing their continued execution of immune functions such as phagocytosis and chemotaxis (Fig. 1).

Regarding signal transduction, non-lytic NETosis exhibits differences in NOX dependence [26, 46] (Fig. 1). Although some studies point out that ROS are still essential, evidence also suggests the existence of NOX-independent NET formation pathways [63]. For example, aluminum adjuvants induce rapid NET release independently of NOX through mitochondrial membrane hyperpolarization and enhanced mitochondrial ROS production [64]. Moreover, oxidized mtDNA released during non-lytic NETosis has strong interferogenic activity and can activate type I interferon responses via the STING pathway, a mechanism of particular relevance to autoimmune diseases and to inflammatory regulation within the TME [62].

Most critically, the structural architecture and origin of vital NETs are markedly distinct. The extruded scaffold consists predominantly of highly oxidized mtDNA, which is packaged and released through vesicular export, rather than nuclear chromatin [61, 62]. Within the TME, this structural distinction has profound immunological consequences. Oxidized mtDNA functions as a potent danger-associated molecular pattern (DAMP) [65]. Rather than trapping pathogens, it robustly and persistently activates the cGAS-STING signaling pathway in surrounding myeloid and stromal cells, thereby establishing a chronic [66], interferogenic, and pro-tumorigenic inflammatory niche that supports tumor proliferation and metastasis.

Pathogenic and sterile oncogenic stimuli

Physiologically, NETosis is triggered by pathogen-associated molecular patterns (PAMPs) such as lipopolysaccharide (LPS) to entrap microbes. However, in malignancies occurring in the absence of acute infection, sterile damage-associated molecular patterns (DAMPs) and cytokines, including IL-8, granulocyte colony-stimulating factor (G-CSF), Cathepsin C, and HMGB1, sustain chronic NETosis via receptors such as CXCR1/2 and RAGE [67].

In pathogen-associated cancers, the etiological mechanisms overlap. For example, Helicobacter pylori infection in gastric cancer(GC) and human papillomavirus (HPV) in cervical cancer induce chronic oxidative stress, modulate local ferroptosis, and continuously activating neutrophils [68]. This persistent pathogen-driven NET deposition causes sustained DNA damage in epithelial cells, thereby linking antimicrobial inflammation to malignant transformation.

Mechanistic convergence: NETosis and pyroptosis

A critical mechanistic overlap exists between NETosis and pyroptosis. Pyroptosis is traditionally defined as an inflammatory lytic cell death driven by Caspase-1/11-mediated cleavage of Gasdermin D (GSDMD), resulting in membrane pore and the release of IL-1β and IL-18 [69, 70]. Strikingly, GSDMD also serves as the terminal effector of lytic NETosis. However, in neutrophils, GSDMD is cleaved and activated independently of caspases by highly abundant neutrophil proteases, predominantly NE [71]. These GSDMD pores permeabilize both granular and plasma membranes, facilitating DNA extrusion and positioning GSDMD oligomerization as a shared therapeutic target for simultaneously mitigating pyroptotic cytokine storms and immunosuppressive NET barriers.

Paracrine regulation by tumor-derived factors

Tumor cells or immune cells actively induce neutrophil NETosis by secreting various soluble factors [72–75] (Fig. 2). Owing to tumor heterogeneity, the molecular mechanisms inducing NETs differ across cancer types, providing a rationale for personalized interventions targeting specific cancers.

Fig. 2.

Driving mechanisms of NET formation and pro-tumorigenic functions within the tumor microenvironment. Multifactorial components within the complex tumor microenvironment—including tumor cell secretions, cancer-associated fibroblasts, platelet aggregates, the metabolic milieu, and physical stress—synergistically activate critical signaling pathways in neutrophils (such as ROS- and PAD4-dependent pathways). These stimuli culminate in NET release, which forms physical barriers and promotes tumor progression by impeding T-cell and NK cell infiltration, enhancing metastasis, and inducing metabolic exhaustion in T-cells

Chemokine and cytokine networks

The chemokine axis is central to mediating neutrophil recruitment and activation [73, 76] (Fig. 2). In pancreatic ductal adenocarcinoma (PDAC), KDM6A deletion upregulates CXCL1 expression in tumor cells, recruiting neutrophils and promoting NET formation via the CXCR2 receptor [77]. Similarly, diffuse large B-cell lymphoma (DLBCL) secretes IL-8, which binds to its receptor CXCR2 to induce NET formation through Src, p38, and ERK signaling pathways [78]. Furthermore, collagen-induced activation of DDR1 signaling can stimulate pancreatic cancer cells to produce CXCL5, representing another critical recruitment and activation pathway [79]. Clinically, CXCR1/2 inhibitors, such as reparixin, are being explored to block neutrophil infiltration and reduce NET deposition at the tumor site.

Proteases and signaling proteins

Beyond classical chemokines, specific tumor-secreted proteins play pivotal roles in NET induction [80–83] (Fig. 2). Cathepsin C from breast cancer cells enzymatically activate membrane-bound proteinase 3 on neutrophils, thereby promoting IL-1β processing and NF-κB activation to induce ROS-dependent NET formation [83]. In HCC, tumor-derived SPP1 (osteopontin) binds to CD44 on alveolar epithelial cells, inducing CXCL1 production, and thus creating a NET-rich pre-metastatic niche in the lungs [84]. Tissue inhibitor of metalloproteinases-1 (TIMP1) can also directly trigger NETs by interacting with the CD63 receptor and activating ERK signaling [85]. Targeting these tumor-secreted proteases or their receptors may represent an effective adjuvant approach for combination immunotherapy.

Neurotransmitters and metabolic enzymes

Neuro-immune interactions are emerging as important regulators of NET formation [86, 87] (Fig. 2). Neuroendocrine prostate cancer cells secrete serotonin, which induces chromatin decondensation via histone H3 serotonylation in neutrophils [87]. Additionally, NAD(P)H:quinone oxidoreductase 1 (NQO1) expression in breast cancer stabilizes and promotes PPIA secretion, which activates CD147 on neutrophils to stimulate NET release [88].

Tumor-derived Exosomes (TDEs) modulating NETs

Extracellular vesicles play a critical role in determining neutrophil fate. Tumor-Derived Exosomes (TDEs) carrying specific non-coding RNAs (e.g., miR-1260b, miR-92a-3p) and oncogenic proteins are directly transferred into neutrophils, where they activate TLR signaling and trigger rapid NET extrusion [89, 90]. These exosome-induced NETs subsequently release MMP-9, which enhances VEGF release and promotes angiogenesis. Conversely, mesenchymal stem cell-derived exosomes delivering CD59 inhibit complement C5b-9 activation, thereby suppressing NETosis [91].

Metabolic reprogramming and oxidative stress drive

Metabolic features of the TME, such as hypoxia, oxidative stress, and metabolite accumulation, act as endogenous drivers of NET formation while simultaneously imposing metabolic constraints on effector T-cells [22, 92, 93] (Fig. 2). NETs orchestrate a coordinated and sequential metabolic reprogramming program that establishes a substantial barrier to effective anti-tumor immunity and markedly compromises the efficacy of immunotherapeutic interventions.

Metabolic mode switching and product accumulation

Tumor-infiltrating neutrophils undergo metabolic reprogramming, shifting toward glycolysis and the pentose phosphate pathway, which promotes NET generation in a ROS-dependent manner [94] (Fig. 2). This highly glycolytic phenotype implies that neutrophils compete with T-cells for limited glucose in the microenvironment, exacerbating T-cell metabolic exhaustion. In the pre-metastatic niche, aberrant metabolite accumulation further induces NETs. For example, in a breast cancer lung metastasis model, hydroxyacid oxidase 1 upregulation in alveolar epithelial cells triggers oxalate accumulation, which induces NET formation via NOX activation [95]. Furthermore, plasma albumin oxidation disrupts redox balance, inducing non-inflammatory NETosis via intracellular ROS accumulation in neutrophils [96]. The hyperglycolytic state generates large quantities of lactate. The accumulation of extracellular lactate not only creates an acidic and hostile microenvironment but also directly suppresses T-cell proliferation and downregulates interferon-gamma (IFN-γ) secretion, thereby driving the immune response toward a state of profound exhaustion.

Subsequently, alterations in lipid and amino acid metabolism synergize with this glycolytic exclusion. In metastatic niches, specific cytokines (e.g., IL-33) and dysregulated signaling networks induce extensive lipid droplets synthesis within low-density neutrophils through DGAT1/2-dependent pathways [25, 97]. When these pathologically altered neutrophils undergo NETosis, the resulting lipid-rich NETs exert dual pro-tumorigenic effects. First, they provide a direct bioenergetic fuel source through sustained β-oxidation, thereby reactivating dormant cancer cells. Second, they release potent immunosuppressive lipid mediators, such as prostaglandin E2 (PGE2), which strongly inhibit natural killer (NK) cell cytotoxicity [98]. Furthermore, the metabolic environment surrounding NETs is characterized by severe amino acid depletion, particularly of arginine and tryptophan. The upregulation of indoleamine 2,3-dioxygenase 1 (IDO1) restricts tryptophan availability, triggering apoptosis in nutrient-deprived T-cells and further reinforcing the immunosuppressive barrier [99].

Mitochondrial dysfunction and metabolic interactions

Environmental stressors, such as exposure to airborne particulate matter, can disrupt neutrophil peroxisome homeostasis and induce NET formation via the ROS–NET–peroxisome axis [100] (Fig. 2). Mitochondrial dysfunction acts as a “signal amplifier” in this process.

The release of mtDNA is governed by precise molecular regulation, with SIRT1 identified as a pivotal “molecular switch”. SIRT1 downregulation facilitates mitochondrial permeability transition pore opening in neutrophils, resulting in mtDNA release and “vital NETs” formation that preserve cell viability [101]. Moreover, the released mtDNA exhibits distinct biochemical modifications. In neutrophils from patients with HCC, elevated mitochondrial ROS levels trigger a highly oxidized state of the released mtDNA (oxidized mtDNA). This oxidative modification significantly potentiates the pro-inflammatory activity of NETs, establishing a pro-metastatic microenvironment through a robust activation of the cGAS-STING signaling pathway and inflammatory responses [102].

A malignant metabolic positive feedback loop exists between NETs and tumor mitochondria. Active components released by NETs such as NE and HMGB1 engage TLR4 on the tumor cell surface and intracellular Toll-like receptor 9 (TLR9), respectively. This interaction significantly upregulates genes associated with mitochondrial biogenesis in tumor cells, increasing mitochondrial density and ATP production to sustain the energy demands of rapid proliferation [103]. Additionally, NET-derived HMGB1 promotes mitochondrial fission and autophagy in tumor cells via RAGE-mediated signaling pathways, further inducing chemotherapy resistance [104]. This bidirectional “neutrophil-tumor mitochondria” metabolic interaction offers novel therapeutic targets for intercepting tumor energy supply.

Thus, by integrating glucose starvation, lipid-driven metastasis, amino acid depletion, and enhanced mitochondrial bioenergetics, NETs establish a coordinated metabolic defense framework that markedly reduces the efficacy of immune checkpoint inhibitors.

Synergy of intercellular interactions and physical microenvironment

In addition to chemical signals, physical contact, mechanical stress, and intercellular cooperative networks within the TME strongly influence NET formation [21, 22, 80, 105–107]. These factors collectively create a physical immune barrier that restricts immune effector cell infiltration and immunotherapeutic drug penetration. This “physical immune privilege” is a critical, often overlooked mechanism underlying primary resistance to ICIs.

Synergistic activation by cells and matrix

Cancer-associated fibroblasts (CAFs) promote intratumoral and systemic NET formation in a ROS-dependent manner by secreting amyloid-beta [108] or cytokines [80, 109, 110]. This interaction often exhibits a bidirectional promoting effect: the DNA backbone and proteases within NETs can conversely activate CAFs to produce more collagen, leading to matrix stiffening in tumor tissues [111]. Additionally, single-cell sequencing has revealed that MFAP5+ cancer-associated fibroblasts (CAFs) are specifically enriched in GC peritoneal metastases. These CAFs establish a NET-rich immune microenvironment via secreted factors, thereby driving pro-metastatic gene programs [112].

In addition to CAFs, mesenchymal stem cells (MSCs) have been confirmed as key stromal components inducing NET formation [113–116]. Recent single-cell transcriptome sequencing analysis have revealed that GC-associated MSCs (GC-MSCs) exhibit elevated NET-associated scores. These GC-MSCs not only expand in number but also remodel the tumor immune microenvironment by recruiting neutrophils and promoting NET formation [116].

Physical injury and physiological stress

Physical trauma itself is a potent NET inducer. The implantation trauma of continuous glucose monitoring sensors rapidly induces tissue NETosis [117]. Surgical tumor resection-induced stress response triggers local hypoxia and HMGB1 release, which activates the TLR9 pathway on neutrophils, thereby promoting NET formation and subsequent tumor recurrence [118]. Furthermore, surgery facilitates platelet–tumor cell aggregation mediated by the TLR4–ERK5 axis, significantly enhancing the capture efficiency of NETs for circulating tumor cells (CTCs) [118]. This physically hinders the cytotoxic contact of NK cells and T-cells and may potentially enable immune surveillance evasion by camouflaging with normal platelet surface antigens. Therefore, the perioperative use of DNase I to disrupt NET structures may serve as an adjuvant strategy to prevent early micrometastases triggered by physical stress, thereby improving subsequent adjuvant immunotherapy response rate. In murine glioblastoma models, the injection of NET-inhibiting agents into the surgical cavity effectively reduces postoperative recurrence [119]. Physiologically, ovulation can similarly facilitate cancer cell capture through follicular fluid via a G-CSF–mediated ROS/NOX/PAD4 pathway, promoting pre-metastatic niche formation [120].

Special cell subsets

Under specific physical and metabolic pressures, neutrophils differentiate into subsets with unique phenotypes and functions, posing challenges for precise immune profiling [121]. For example, under specific conditions such as the early stages of breast cancer lung metastasis, a subset of immunosuppressive CXCR4(hi)CD62L(lo) aged neutrophils (Naged) emerges; these cells form a unique “vital” NET via SIRT1-mediated mitochondrial DNA release [101]. Unlike typical nuclear DNA NETs, these structures enriched in oxidized mitochondrial DNA possess stronger pro-inflammatory capabilities while simultaneously inducing the exhaustion of surrounding immune cells.

Key transcription factors and epigenetic regulation Mechanisms

Key Regulatory axes

The presence of transcription factors enables NETs to function as a “signaling pool,” exerting a sustained influence on the malignant phenotypes of surrounding tumor cells. The ROS-p38-cFOS-PAD4 axis is a critical pathway for NET formation in triple-negative breast cancer, where the transcription factor c-FOS directly promotes PAD4 transcription by binding to its promoter region [122] (Fig. 2). Beyond this classic ROS-p38 pathway, components of the DNA replication complex participate in the transcriptional regulation of NETs. GINS complex subunit 2 (GINS2) in oral squamous cell carcinoma promotes proliferation and acts as an upstream regulator, remodeling neutrophil phenotypes via the GINS2–PTP4A1–PKM2 axis [123]. Conversely, the long non-coding RNA MIR503HG negatively regulates NET-induced metastasis: NET stimulation downregulates MIR503HG, relieving its inhibition on the NF-κB/NLRP3 inflammasome pathway [124]. Therefore, in addition to direct PAD4 inhibition, targeting upstream transcription factors such as c-FOS or utilizing lncRNA mimetics represents a viable strategy to suppress NET formation. Apart from the transcriptional regulatory axes, regulation by non-transcription factors is also involved; for example, the S100A9-induced RAGE/TLR4-ROS signaling pathway is a significant mechanism underlying the explosive NET formation in hepatitis B virus-related HCC [55].

Metabolic-epigenetic reprogramming and modification

Lactylation is central to NET generation and immunosuppressive functions [125] (Fig. 2). Tumor-intrinsic molecular characteristics can drive this process. For example, the loss of the tight junction protein claudin-7 in colorectal cancer activates the NF-κB/CXCL1 signaling axis, enhancing neutrophil recruitment and glycolysis. Lactate from glycolysis serves as a substrate for histone lactylation in neutrophils, promoting a pro-tumor phenotype, NET release, and CD8+ T-cell functional exhaustion [126]. In microsatellite instability-high colorectal cancer, myeloid nuclear differentiation antigen (MNDA) recruits the acetyltransferase EP300 to the promoter region of the CXCR2 gene, specifically catalyzing lactylation at the H3K18 site (H3K18la). This enhances PMN-MDSC recruitment, NET formation, and resistance to anti-PD-1 immunotherapy [127].

Acetylation also regulates NETs. In esophageal squamous cell carcinoma, the loss of acetyl-CoA carboxylase 1 increases histone acetylation, activating c-FOS transcription and promoting CXCL8 secretion. This establishes a positive feedback loop via the CXCL8-NET axis, enhancing tumor invasive and metastatic capabilities [128]. Targeting glycolysis, lactylation/acetylation modifications, or the MNDA/EP300 axis may disrupt this vicious cycle and restore immunotherapy sensitivity.

Role of the microbiota

Gut microbiota dysbiosis can systemically promote NET formation in the liver by enhancing neutrophil-mediated inflammation [129] (Fig. 2). Deoxycholic acid produced by the gut flora can inhibit neutrophil migration and NET formation, whereas microbiota lacking bile salt hydrolase activity causes accumulation of conjugated bile acids such as TCA (taurocholic acid). TCA promotes neutrophil recruitment and NETosis via TIMP1 upregulation in tumor cells and CXCL5/CXCR2 axis activation, contributing to liver metastasis and immune evasion [130]. These findings link the gut-liver axis with NET-mediated immunotherapy resistance, suggesting that microbiota modulation via probiotics or fecal microbiota transplantation could serve as an adjuvant therapy to improve the tumor immune microenvironment and reduce NET burden.

Mechanisms by which NETs mediate the tumor immunosuppressive microenvironment

Neutrophils within tumors are traditionally regarded as a line of defense for tumor surveillance and killing [8, 131–136]. However, as integral components of the immune microenvironment, their interactions with other cells can also promote tumor progression [26, 28, 134]. Myeloid-derived suppressor cells (MDSCs), which originate from the bone marrow, are influenced by NETs during immune regulation. They inhibit tumor destruction within the immune microenvironment and contribute to pre-metastatic niche formation, maintaining an inseparable relationship with tumor-associated neutrophils (TANs) [137–140]. Once recruited by TANs, MDSCs can inhibit T-cell function; meanwhile, NETs can also independently interact with T-cells, driving T-cell exhaustion within the immune microenvironment [28, 141]. (Table 1)

Table 1.

Regulatory effects and molecular mechanisms of NETs on different immune cell subsets

Target Cell Subset	Overall Functional Outcome	Consolidated Regulatory Effect & Specific Molecular Mechanism
CD8+ T Cells	Immune Exhaustion & Immune Exclusion	Functional Inhibition & Steric Hindrance: NET-DNA binds TMCO6 to block TCR/NF-κB signaling [142]. NETs mediate YTHDF2-dependent degradation of SLC2A3 mRNA, impairing T cell metabolism [143]. NETs act as scaffolds for PD-L1, binding PD-1 to directly induce T cell exhaustion [28, 144]. Additionally, tumor-induced NETs construct dense physical barriers that prevent CD8+ T cell infiltration into the tumor parenchyma [75].
Regulatory T Cells (Tregs)	Immune Tolerance	Induced Differentiation & Recruitment: NETs reprogram naive CD4+ T cell metabolism (via Akt/mTOR/SREBP2 or oxidative phosphorylation) to drive Treg differentiation [145, 146]. The NET-associated MPO-ENO1 complex activates the ERK pathway to enhance Treg function [147], while the DDR1/CXCL5 axis promotes targeted Treg infiltration [148, 149].
Macrophages (TAMs)	Immune Tolerance & Immune Exclusion	Pro-tumor Polarization (M2) & Phagocytosis Inhibition: NETs induce a CD206+ M2 macrophage phenotype and inhibit the costimulatory molecule CD80 [150, 151]. Neutrophil elastase (NE) within NETs inactivates PAR2 signaling, downregulating the CD24 phagocytosis checkpoint and impairing macrophage-mediated tumor clearance [18].
Natural Killer (NK) Cells	Immune Exclusion	Reduced Infiltration & Impaired Killing: The CD276/CXCL1–CXCR2 axis triggers NETs to limit local NK cell infiltration [152]. In circulation, NETs collaborate with platelets to enwrap circulating tumor cells (CTCs), creating a physical shield that prevents NK cell recognition and cytolysis [153].
B Cells	Immune Tolerance	Induction of Immunosuppressive Cytokines: NETs inhibit SHP-1 phosphatase and generate ROS, prompting innate-like B cells to secrete IL-10, which further drives Treg expansion [20].
Dendritic Cells (DCs)	Immune Tolerance	Impaired Antigen Presentation & Maturation Blockade: MPO-driven lipid peroxidation severely impairs the antigen cross-presentation capacity of DCs [154]. The persistence of NETs in the TME blocks DC maturation, which can be reversed by NET clearance (e.g., via DNASE1L3) 186.

Evolution and alterations of neutrophils in NETs

Origins and connections between neutrophils and MDSCs

Within the immunosuppressive network of the TME, neutrophils and their pathological subset, PMN-MDSCs, exhibit complex relationships [155–159] (Fig. 3). Their phenotypic similarity and functional heterogeneity provide key entry points for current research on cancer immunotherapy resistance [137, 160]. In tumors, abnormal bone marrow hematopoiesis, driven by tumor-derived factors, results in blocked differentiation or pathological activation of myeloid cells, generating immunosuppressive PMN-MDSCs. Neutrophils and PMN-MDSCs express shared antigens—CD11b, CD15, and CD66b in humans or CD11b and Ly6G in mice—prompting debate over whether PMN-MDSCs represent an independent cell lineage or a neutrophil activation state induced by specific environments [161–163].

Fig. 3.

Mechanisms of the tumor immunosuppressive microenvironment mediated by NETs. Tumor-derived factors, such as IL-8, stimulate neutrophils or PMN-MDSCs to release NETs, reshaping the tumor immune microenvironment. NETs suppress anti-tumor immunity through multiple mechanisms: directly inhibiting CD8+ T-cell function and promoting their exhaustion; inducing the differentiation and accumulation of regulatory T-cells (Tregs); modulating B cells to produce IL-10 to facilitate Treg expansion; polarizing macrophages toward a pro-tumor phenotype and establishing a positive feedback loop; impairing dendritic cell (DC) antigen cross-presentation via MPO-driven lipid peroxidation; and cooperating with platelets in the circulation to coat tumor cells, forming a physical shield against NK cell-mediated killing. Green solid arrows indicate promotion, activation, or recruitment; red dashed lines indicate inhibition, blockade, or failed killing

Advances in single-cell sequencing and high-dimensional flow cytometry have identified Lectin-like oxidized low-density lipoprotein receptor-1 (LOX-1) as a specific marker for human PMN-MDSCs. High LOX-1 expression is closely associated with endoplasmic reticulum stress and abnormal lipid metabolism, distinguishing PMN-MDSCs with potent T-cell-suppressive functions from ordinary neutrophils [164–166]. High-resolution transcriptomic trajectory analyses (pseudotime modeling) reveal that PMN-MDSCs follow a fundamentally distinct developmental trajectory that originates early within the bone marrow and is clearly separated from normal neutrophil granulopoiesis [167]. ScRNA-seq studies shown that PMN-MDSCs comprise a continuum of maturation states, including precursor PMN-MDSCs, classical PMN-MDSCs, and late PMN-MDSCs [168]. In murine tumor models, the transition among these populations can be tracked; as tumor progression occurs, PMN-MDSCs progressively replace classical PMNs in the spleen and peripheral blood and are distinguished by differential CD14 expression [169].

Chromatin immunoprecipitation sequencing (ChIP-seq) analyses of histone modifications further demonstrate that PMN-MDSCs possess a distinct and stable epigenetic landscape. Activating chromatin marks, including H3K4me3 and H3K27ac, are enriched at the transcription start sites (TSSs) of genes associated with immunosuppressive and angiogenesis [170]. This epigenetic reprogramming promotes chromatin accessibility at loci associated with immune tolerance, thereby maintaining these cells in an immunosuppressive state. Consequently, this extensive epigenetic remodeling distinguishes PMN-MDSCs from transiently activated classical neutrophils and establishes them as a pathologically programmed cell population that contributes to the cellular foundation of primary immunotherapy resistance.

Functional neutrophil remodeling

Functional remodeling is the core mechanism underlying neutrophil transformation toward an immunosuppressive phenotype [14, 17]. Tumor-derived cytokines and chemokines induce “emergency granulopoiesis” in the bone marrow, expanding immature myeloid cells, and directly remodel neutrophil function in the local microenvironment. This remodeling endows neutrophils with PMN-MDSC-like immunosuppressive properties, inhibiting T-cell and NK-cell anti-tumor activity [164, 171, 172]. Activated neutrophils or PMN-MDSCs mediate T-cell immune tolerance through high ROS production and lipid peroxidation [173] (Fig. 3). Notably, the complement system (such as C5a) also participates in PMN-MDSC recruitment and functional activation, promoting their conversion to a pro-tumor phenotype [174, 175].

In this process of functional remodeling, NET formation serves as a critical nexus linking inflammation with immune suppression. Tumor-derived factors (such as IL-8) recruit MDSCs and induce PMN-MDSCs to undergo NETosis and release NETs [176, 177]. PMN-MDSCs in the TME can also restrict dendritic cell (DC) antigen cross-presentation via MPO-driven lipid peroxidation, thereby blocking anti-tumor immune response initiation at the antigen-recognition stage [154]. This TME-dominated functional remodeling, involving NET release and metabolic reprogramming, ultimately establishes a highly immunosuppressive state, constituting a cellular basis for ICI therapy failure. (Table 1)

Impact of NETs on T cells

Direct inhibition of CD8+ T-Cell function and exhaustion

NET-derived DNA fragments (NET-DNA) can bind the transmembrane protein TMCO6 on CD8+ T-cells, blocking T-cell receptor signaling and inhibiting NF-κB activation, thereby attenuating anti-tumor immunity [142] (Fig. 3). NETs also interfere with cell metabolism by promoting YTHDF2-mediated SLC2A3 mRNA degradation, which subsequently suppresses CD8+ T-cell activity and enhances tumor resistance to ferroptosis [143]. Consequently, CD8+ T-cell function is diminished or lost. In non-small cell lung cancer models, CD8+ T-cells exposed to NETs exhibit downregulation of effector genes and marked upregulation of exhaustion-associated genes and cytosolic DNA-sensing pathways [178]. This exhaustion is particularly pronounced post-radiotherapy, where high NET deposition correlates with elevated PMN/CD8 ratios, contributing to tumor radioresistance [179].

Induction of Regulatory T cell (Treg) differentiation and recruitment

NETs promote Treg-mediated immune tolerance through mechanisms including chemotactic recruitment and metabolic reprogramming [20, 180] (Fig. 3). In sepsis-related immunosuppression, NETs enhance cholesterol metabolism in naive CD4+ T-cells via the Akt/mTOR/SREBP2 pathway, thereby promoting their conversion to Tregs [145]. A similar mechanism occurs in non-alcoholic steatohepatitis (NASH)-driven liver cancer, where NETs drive Treg differentiation by promoting mitochondrial respiration and oxidative phosphorylation in naive CD4+ T-cells [146]. Mechanistically, specific protein interactions further specify this process: the MPO–ENO1 complex on NETs interacts with CD4+ T-cells, recruits IFITM2, and activates the downstream ERK signaling pathway, enhancing Treg differentiation and function [147]. In sterile liver inflammation, T helper (Th) 17 cells are also affected, maintaining an inflammatory state that favors tumorigenesis [181].

Beyond inducing differentiation, NETs recruit Tregs to the immune microenvironment, reinforcing immune tolerance. In breast cancer, the collagen-activated DDR1/CXCL5 axis induces NET formation, which promotes Treg infiltration [148]. In NASH-associated liver cancer, elevated Trem2 expression positively correlates with NET accumulation and Treg infiltration; NET clearance can downregulate CXCR4 expression on Tregs, thereby reducing their aggregation within the tumor [149].

Regulation of other myeloid cells and remodeling of the stromal microenvironment

NETs remodel the tumor immune microenvironment beyond T-cells, extensively reprogramming various immune subpopulations—including macrophages, B-cells, NK cells, and DCs—to synergistically construct a comprehensive immunosuppressive barrier [46, 182] (Fig. 3). This multicellular regulation induces an “immune desertification” of the stromal microenvironment, further exacerbating tumor immune escape and therapeutic resistance.

Macrophage polarization

In pancreatic cancer, NETs polarize macrophages toward a pro-tumor CD206+ phenotype and suppress the expression of the T-cell costimulatory molecule CD80 [150] (Fig. 3). Conversely, extracellular vesicles containing FGL2 secreted by tumor-associated macrophages, or CXCL5 secreted via the adenosine/A2AR pathway, can feed back to promote NET formation, thereby establishing a malignant positive feedback loop [183]. Furthermore, asporin, a small leucine-rich proteoglycan highly expressed in GC cells, activates PI3K/AKT signaling to induce NET formation and promote M2 macrophage polarization [151]. Consequently, the immune microenvironment is remodeled, and tumor cell escape is further exacerbated.

B cell Regulation

In the TME, NETs affect T-cells and reshape the functional phenotype of B-cells through specific metabolic and signaling mechanisms, transforming them from anti-tumor immunity participants into immune tolerance promoters [20, 184] (Fig. 3). In the omental metastasis microenvironment of ovarian cancer, NETs induce innate-like B-cells to produce IL-10 by inhibiting SHP-1 phosphatase and generating ROS. This cytokine subsequently promotes Treg expansion and tumor colonization [20].

NK cell Regulation

Although NK cells constitute the body’s first line of defense against tumor metastasis, NETs severely impair NK cell surveillance and killing functions by constructing physical barriers and inducing spatial exclusion [81, 185] (Fig. 3). At the physical spatial level, NETs synergize with platelets in circulation to enwrap CTCs, forming dense cell aggregates [153]. This “shield” physically prevents NK cells from recognizing and lysing tumor cells, thereby increasing cancer cell survival and metastatic potential.

DC Regulation

DCs bridge innate and adaptive immunity, but NETs block anti-tumor T-cell immunity initiation at the source by interfering with their maturation and antigen presentation capabilities (Fig. 3). Mechanistic studies have revealed that MPO carried by NETs exerts a core inhibitory effect, as MPO drives lipid peroxidation in DCs [177]. This state of oxidative stress specifically disrupts antigen cross-presentation in DCs, rendering them unable to activate CD8+ T cells and causing the anti-tumor immune response to enter a “disabled” state. Conversely, NETs clearance within the microenvironment by DNASE1L3-positive DCs alleviates this inhibition, promotes DC maturation, and restores their ability to recruit and activate T-cells [21].

Cancer stem cells (CSCs) and EMT Crosstalk

The dynamic interaction between cancer stem cells (CSCs) and NETs establishes a strong framework for metastasis and therapy resistance. NETs actively induce epithelial-to-mesenchymal transition (EMT) in neighboring epithelial cells. This process is mediated primarily by NET-associated proteases, including neutrophil elastase (NE) and matrix metalloproteinase-9 (MMP-9), which degrade intercellular junctions and activate surface receptors such as TLR4. Consequently, this signaling cascade upregulates mesenchymal transcription factors and markers, including vimentin, Twist, and Snail, while downregulating E-cadherin [186].

This EMT induction synergizes closely with the Wnt/β-catenin signaling pathway. Disruption of E-cadherin complexes releases membrane-bound β-catenin and promotes its nuclear translocation. This transcriptional reprogramming not only drives the mesenchymal phenotype but also enriches the CD44+/ALDH+ CSC population, thereby enhancing tumor phenotypic plasticity [187]. Endowed with increased stem-like properties, these CSCs actively establish an immunosuppressive and protective niche. They secrete chemokines, predominantly CXCL1 and IL-8 (CXCL8), which bind to CXCR2 receptors on circulating polymorphonuclear cells [176]. This persistent chemokine gradient serves a dual function: it continuously recruits neutrophils to the tumor microenvironment and directly induces NETosis. Ultimately, this creates a self-perpetuating positive feedback loop that sustains CSC maintenance, immune evasion, and long-term therapy resistance.

Mechanisms by which NETs promote resistance to cancer immunotherapy

NETs in the TME extend beyond traditional roles in cell death, forming an “immunological shield” by constructing dense physical networks and carrying inhibitory ligands. This dual barrier—physical and functional—impedes anti-tumor immune responses. High-risk NET signatures often predict poor immunotherapy response and adverse clinical outcomes [188] (Fig. 4).

Fig. 4.

Mechanisms by which neutrophil extracellular traps (NETs) promote tumor immunotherapy resistance. Tumor-derived factors (e.g., CXCL1, Chi3l1, and TRAPs) or specific therapies (e.g., cryoablation) induce neutrophils to release NETs. These NETs create a dense physical network that blocks infiltration of CD8+ T-cells and NK cells into the tumor parenchyma. Furthermore, as functional scaffolds, NETs harbor PD-L1 to induce T-cell exhaustion and impair macrophage phagocytosis through signaling modulation. Together, these effects establish an “immune shield” that hinders anti-tumor immune responses across physical and functional dimensions

The Synergistic “physical-biochemical-metabolic” hierarchical model of resistance

The establishment of a therapeutic resistance barrier by NETs does not occur through isolated or random mechanisms. Rather, extensive multi-omics and spatial transcriptomic analyses indicate that this process follows a highly coordinated hierarchical model. Within this framework, physical occlusion, biochemical suppression, and metabolic deprivation are not merely additive; instead, they act sequentially and synergistically to markedly reduce the efficacy of immunotherapies, including ICIs. Understanding this hierarchical interplay is essential for the rational design of combination therapies.

Direct physical blockade of immune cell infiltration

The physical network formed by NETs is a key factor driving the “immune-excluded” tumor phenotype. Tumor cells can specifically induce TANs to release NETs by secreting CXCR1 and CXCR2 chemokine receptor agonists (Fig. 4). These web-like structures tightly wrap tumor cells, forming a physical barrier that impedes contact between immune cells and surrounding target cells [81]. Upstream, the immune checkpoint CD276 (B7–H3) promotes NET formation via the CXCL1–CXCR2 signaling axis, reducing NK cell infiltration within the tumor [40, 152]. In triple-negative breast cancer, the cytokine Chi3l1 promotes neutrophil recruitment and induces NET formation, directly blocking CD8+ T-cell infiltration into the tumor parenchyma [75]. This spatial restriction is observed in other solid tumors as well, and NET degradation by DNASE1L3-positive DCs can restore cytotoxic T-cell (CTLs) clustering [21]. Excessive NET formation may also cause vascular occlusion and tissue hypoxia, triggering tumor necrosis; this physical vascular restriction further deteriorates the microenvironment, indirectly hindering the effective transport of immune cells [134].

This physical barrier constructed by NETs not only restricts immune cell infiltration to the poorly vascularized and hostile tumor periphery but also reinforces its defensive capacity through the “entrainment” (or education) of surrounding myeloid cells [18] (Fig. 4). On one hand, neutrophil elastase (NE), a core component of NETs, alters PAR2 signaling, leading to the downregulation of the phagocytic checkpoint CD24 and thereby impairing macrophage clearance functions. NET clearance can reverse the pro-tumor polarization state of macrophages, as evidenced by decreased expression of the pro-tumor M2 marker CD206 and increased expression of the anti-tumor M1 marker CD80 [150]. On the other hand, in specific tumors such as bladder cancer, NETs upregulate the anti-phagocytic checkpoint STC1 through activation of the TLR2-MAPK-FosL1 signaling axis in tumor cells, thereby suppressing antigen presentation [30].

The establishment of this barrier involves complex intercellular communication (Fig. 4). For example, DMBT1 secreted by hepatotropic tumor cells induces CD62L expression in Kupffer cells, which subsequently activates neutrophils to produce NETs through CCL8 signaling, thereby forming a multicellular cooperative barrier at metastatic sites [80].

Synergistic elimination via metabolic starvation and biochemical checkpoints

Once anti-tumor immune cells, such as T cells, are trapped by this impenetrable physical and cellular network, the metabolic barrier takes dominance. The trapped immune cells suffer severe nutritional deprivation orchestrated by tumor cells and polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs). The highly glycolytic nature of PMN-MDSCs, combined with the local hypoxia induced by NETs, relentlessly deprives the trapped T cells of critical nutrients necessary for survival and proliferation (such as glucose and tryptophan) [189–191]. Meanwhile, this metabolic rewiring exposes immune cells to extremely high local concentrations of toxic metabolites, such as lactic acid and reactive oxygen species (ROS) [192]. This brutal metabolic starvation rapidly paralyzes the proliferative capacity of immune infiltrating cells, pushing them into an unresponsive state even before they can truly engage the tumor.

Ultimately, immune cells that remain physically confined and metabolically compromised undergo further suppression through biochemical checkpoint signaling. NETs are not inert DNA structures; rather, they function as highly bioactive molecular scaffolds. Although PD-L1 is primarily expressed on the cell surface or within exosomes, evidence suggests that, under specific TME conditions, NETs can integrate and present high densities of inhibitory immune checkpoint ligands, particularly PD-L1, within their chromatin networks [81, 193–195]. Furthermore, hypersialylated structures on NETs can bind Siglec receptors on macrophages and NK cells [196, 197]. Due to the severe spatial confinement and metabolic dysfunction imposed previously, T cells and NK cells become highly sensitive to these inhibitory biochemical signals. Engagement of these checkpoint ligands acts as a final suppressive mechanism, driving residual immune cells toward terminal exhaustion or apoptosis.

Therapy-induced dynamic barriers and acquired resistance

Clinical treatments can trigger NET-mediated barriers, promoting acquired resistance post-therapy. Following cryoablation, intracellular calcium levels significantly rise in neutrophils, triggering PAD4-dependent NET formation; this burst-like barrier construction further promotes tumor immune evasion [198] (Fig. 4). Similarly, in studies on resistance in EGFR-mutant non-small cell lung cancer, spatial analysis revealed that CXCR1+ neutrophils predominantly infiltrate the tumor core regions of resistant samples. Moreover, the average distance between neutrophils and tumor cells was significantly shortened to 19 μm, suggesting that this close physical contact is a key spatial feature mediating resistance [199].

Standard chemotherapeutic agents (e.g., anthracyclines) can paradoxically contribute to therapeutic resistance. Cytotoxic cell death releases large amounts of ATP and DAMPs into the TME, thereby hyperactivating the NLRP3 inflammasome in neutrophils and triggering burst-like NETosis [200]. These therapy-induced NETs form a physical “DNA sponge” that sequesters cationic chemotherapeutic agents, thereby reducing drug penetration into the tumor core and diminishing local drug efficacy [201]. Furthermore, this chemotherapy-induced NET barrier activates TGF-β signaling pathways in surviving tumor cells, thereby promoting acquired resistance [202].

Metastatic physical barriers and immune evasion: from circulatory voyage to niche awakening

NETs exert significant spatiotemporal effects during metastasis. During vascular circulation, intravascular NETs rapidly capture circulating tumor cells (CTCs) through interactions involving β1-integrin and von Willebrand factor [203]. This dense physical aggregation protects CTCs from hemodynamic shear stress and prevents Natural Killer (NK) cells from forming effective cytolytic synapses [81].

Upon arrival at metastatic niches, NETs can reactivate dormant disseminated cancer cells. NET-associated proteases sequentially remodel laminin, thereby exposing cryptic epitopes that activate integrin α3β1 signaling in tumor cells [25]. Concurrently, NE directly activates TLR4 signaling in these cells, initiating a p38-mediated signaling cascade that upregulates PGC1α [103]. This process enhances mitochondrial biogenesis, thereby alleviating metabolic constraints in metastatic cells and promoting secondary tumor growth.

Relative contribution of NET-Mediated resistance mechanisms across different Cancers

Although NETs consistently promote therapeutic resistance through a synergistic “physical-biochemical-metabolic” hierarchy, emerging evidence from spatial transcriptomics and in vivo imaging demonstrates that the relative contribution of each mechanism varies substantially according to the histological and anatomical context of the tumor [204]. Treating NETosis as a uniform barrier may limit the precision of therapeutic interventions; therefore, defining the dominant mechanistic dependencies within distinct malignancies is essential.

In highly desmoplastic malignancies, such as PDAC, spatial physical exclusion appears to be the predominant resistance mechanism. In this context, NETs cooperate closely with cancer-associated fibroblasts (CAFs). NET-derived proteases stimulate CAFs to upregulate discoidin domain receptor 1 (DDR1), thereby accelerating collagen remodeling and extracellular matrix (ECM) stiffening [79]. This process creates a dense fibrotic barrier that promotes immune exclusion, in which cytotoxic T cells remain confined to the tumor margin and are unable to efficiently access tumor islets [81]. Consequently, overcoming physical steric barriers may represent a critical prerequisite for successful immunotherapy in PDAC.

Conversely, in malignancies such as non-small cell lung cancer (NSCLC) and hepatocellular carcinoma (HCC), functional immune exhaustion appears to predominate over physical exclusion. In NSCLC, infiltrating NETs act as biochemical scaffolds embedded within the tumor parenchyma. They present high densities of immune checkpoint ligands, particularly PD-L1, directly to infiltrating CD8+ T-cells, thereby promoting premature exhaustion and reducing the efficacy of standard ICIs [28]. Furthermore, in triple-negative breast cancer (TNBC), NETs promote an immunosuppressive microenvironment. Tumor-secreted factors, including Chi3l1 and Cathepsin C, continuously recruit neutrophils and induce metabolic deprivation, ultimately impairing antigen cross-presentation and contributing to an “immune-desert” phenotype [83, 205].

In metastatic niches such as the liver, NETs orchestrate metabolic reprogramming. Through bile acid metabolites, including taurocholic acid (TCA), and lactate accumulation driven by LDHA upregulation, NETs suppress natural killer (NK) cell recruitment and promote macrophages polarization toward the pro-tumor M2 phenotype [206, 207]. Understanding these context-dependent mechanisms highlights an important clinical principle: the selection of adjuvant therapies, such as DNase I or immune checkpoint blockade) should be tailored to the spatial and metabolic characteristics of individual cancer types (Table 2).

Table 2.

Emerging NET-mediated resistance mechanisms across various malignancies

Cancer Type	Primary Therapy Resisted	Core NET-Mediated Resistance Mechanism	Key Signaling Axis/Components	Representative Basic Research Evidence (Full Citation)
Ovarian Cancer	Chemotherapy (Doxorubicin/Platinum)	NETs construct a dense physical DNA meshwork that directly impairs the diffusion and penetration of chemotherapeutic agents, shielding cancer cells from drug-induced apoptosis.	Dense DNA Scaffold/Physical Steric Hindrance	[201]
Glioblastoma (GBM)	Radiotherapy & Microenvironment Resistance	NET-derived HMGB1 specifically binds to RAGE receptors on glioma cells, activating NF-κB and inducing IL-8 secretion. This establishes a malignant positive feedback loop driving radioresistance and progression.	HMGB1/RAGE/IL-8 Axis	[208]
Colorectal Cancer (CRC)	Targeted Therapy & Chemotherapy	Internalization or surface binding of NET-derived Neutrophil Elastase (NE) by CRC cells aberrantly hyperactivates the ERK signaling cascade, reversing drug-induced cellular apoptosis and promoting metastasis.	Neutrophil Elastase (NE)/ERK Pathway	[209]
Renal Cell Carcinoma (RCC)	Anti-angiogenic Immunotherapy	Anti-angiogenic induced local hypoxia triggers explosive NETosis. NETs compensate for VEGF blockade by driving alternative metabolic reprogramming and promoting immune evasion mechanisms.	Hypoxia-HIF-1α/Metabolic Reprogramming	[210]
Bladder Cancer	Radiotherapy	Radiation triggers burst-like NET formation within the tumor bed. This creates a therapy-induced physical and biochemical barricade that severely hinders CD8+ T-cell infiltration, predicting radiotherapeutic failure.	Therapy-induced NETosis/PMN-CD8 Spatial Exclusion	[179]
Oral Squamous Cell Carcinoma (OSCC)	Chemotherapy & Cellular Death Resistance	NET-derived Neutrophil Elastase (NE) penetrates OSCC cells and specifically inhibits the NLRP3 inflammasome, effectively preventing cancer cells from undergoing pyroptosis under therapeutic stress.	Neutrophil Elastase (NE)/NLRP3/Pyroptosis Inhibition	[211]
Gastric Cancer	Chemotherapy (5-FU)	Hypoxia-driven NETs aggressively capture chemotherapeutic agents and stimulate protective signaling in gastric epithelial cells, augmenting tumor growth and inducing profound 5-FU resistance.	Hypoxia/Drug Sequestration/HIF-1α	[212]
Esophageal Cancer	Systemic Chemotherapy	Surgery and inflammation-induced leukocytosis orchestrate heavy NET infiltration at the invasive margin, triggering epithelial-mesenchymal transition (EMT) to evade chemotherapy-induced cytotoxicity.	Surgical Stress/EMT Activation	[213]
Multiple Myeloma (MM)	Proteasome Inhibitors (Bortezomib)	PAD4-dependent NETs deposited within the bone marrow niche physically and chemically shield myeloma cells, severely blunting the apoptotic effects of standard care like Bortezomib.	PAD4/Bone Marrow Niche Shielding	[214]
Diffuse Large B-Cell Lymphoma (DLBCL)	Targeted Therapy & Systemic Clearance	Lymphoma-derived IL-8 continuously induces systemic NETosis. The NET-DNA complexes heavily engage TLR9 on malignant B-cells, hyperactivating NF-κB and STAT3 to resist targeted eradication.	IL-8/TLR9/STAT3 Axis	[78]

The double-edged sword role of NETs in immunotherapy resistance across solid and liquid tumors

Within the context of cancer immunotherapy, NETs have traditionally been regarded as key mediators of immune evasion. However, emerging evidence indicates that NETs function as a complex “double-edged sword,” exerting both pro-tumorigenic and anti-tumorigenic effects that vary substantially between solid tumors and hematological malignancies, as well as across different therapeutic settings [131, 184].

Solid tumors: physical barriers vs. Acute cytotoxicity

In solid tumors, including non-small cell lung cancer, breast cancer, and colorectal cancer, the pathological accumulation of NETs predominantly contributes to immunotherapy resistance. As discussed previously, NETs form physical barriers that restrict cytotoxic T-cell and NK-cell infiltration and act as functional scaffolds for immune checkpoints ligands, including PD-L1^{18 28}.

Paradoxically, under certain conditions, NETs may also exert anti-tumor effects. Recent in vivo colorectal cancer models have demonstrated that chemotherapy-induced NETs can directly inhibit tumor growth and induce malignant cell death rather than promote tumor progression [19]. Additionally, during the early stages of tumorigenesis, the dense network of chromatin and cytotoxic enzymes, including NE and MPO, may function analogously to antimicrobial NETs by trapping and eliminating malignant cells [131]. In particular, the high local concentrations of ROS and granular proteases retained within these NET structures can directly induce irreversible membrane damage and apoptosis in nascent tumor cells, thereby contributing to innate immune surveillance [215–217]. This context-dependent plasticity suggests that complete abrogation of NETs may inadvertently suppress their intrinsic anti-tumor cytotoxic functions, thereby highlighting the double-edged nature of NET-targeted therapies in solid tumors.

Liquid tumors (hematological malignancies): niche protection and infection vulnerability

Although solid tumors are characterized by distinct spatial barriers, NETs in hematological malignancies, including diffuse large B-cell lymphoma (DLBCL), multiple myeloma (MM), and acute myeloid leukemia (AML), play a major role in systemic immune resistance [218].

In these malignancies, NETs frequently establish an immunosuppressive and protective niche for circulating and bone marrow-resident malignant cells. In DLBCL, tumor-derived IL-8 strongly induces NETosis, which subsequently upregulates TLR9 signaling in malignant cells, thereby accelerating proliferation and promoting resistance to targeted therapies and immune-mediated clearance [78]. Similarly, in multiple myeloma, PAD4-driven NETs protect myeloma cells from immune surveillance and therapy-induced apoptosis [219]. Recent transcriptomic analyses in AML have further demonstrated that elevated NET scores and associated markers, including LTF, correlate with an immunosuppressive bone marrow microenvironment, leading to T-cell exhaustion and poor responses to systemic therapies [220].

Despite these pro-tumorigenic effects, the double-edged role of NETs in hematological malignancies involves an important clinical trade-off. Patients receiving intensive immunotherapies, such as CAR-T therapy or bispecific antibodies, are often profoundly immunocompromised. Because neutrophils and NETs serve as frontline defenses against opportunistic infections, systemic NET degradation as a strategy to overcome immunotherapy resistance may increase risk of severe or potentially fatal sepsis.

Disease-specific NET signatures in solid tumors

Advances in high-resolution transcriptomics have revealed that the prognostic and predictive significance of NETs cannot be uniformly generalized across different cancer types. Instead, individual tumors rely on distinct disease-specific molecular signatures to interact with NETs and evade immune surveillance.

In PDAC, exposure to NETs activates the ITGB1–CCDC25-ILK signaling axis, which regulates integrin-actin remodeling. This signaling pathway promotes EMT and pseudopodia formation, thereby contributing to poor overall survival and therapy resistance [221]. In contrast, resistance mechanisms in non-small cell lung cancer (NSCLC) are more closely associated with altered cellular metabolism and programmed cell death (PCD) pathways than with mechanical invasiveness [222]. Multi-omics analyses have identified the mitochondrial pathway signature (MitoPS), highlighting genes such as NDUFB10 as important regulators of NET-associated oxidative stress. In triple-negative breast cancer (TNBC), immunotherapy response is closely linked to the tumor immune microenvironment gene expression signature (TIME-GES) and the natural killer cell immunotherapy predictive model (NKCIPM) [223].

Recognition of these disease-specific signatures is essential for the development of precision NET-targeted therapies. For example, therapeutic strategies in PDAC may need to prioritize disruption of the integrin-fibroblast signaling axis, whereas improving therapeutic sensitivity in lung and breast cancers may depend more heavily on reversing metabolic exhaustion and immune checkpoint-mediated suppression.

NET-Mediated immune checkpoint resistance

Beyond simple physical isolation, NETs act as bioactive scaffolds for immunosuppressive molecules. Although PD-L1 is not universally associated with NETs and is predominantly localized to cell membranes or tumor-derived exosomes, direct PD-L1 decoration of NETs can occur under specific context-dependent conditions [28, 195] (Fig. 4). For example, tumor-released autophagosomes activate the TLR4-Myd88-ERK/p38 signaling axis in neutrophils, inducing “PD-L1-decorated” NET formation that directly inhibits T cell function in vitro and in vivo, thereby promoting the immunosuppressive state formation within the pre-metastatic niche (PMN) of the lungs [144]. In “cold tumor” models such as pancreatic cancer, high UBE2H expression is closely associated with NET formation. This microenvironmental alteration is accompanied by the downregulation of antigen presentation pathways (such as H2Kb and I-Ab) in macrophages and neutrophils, further exacerbating PD-1+ T-cell exhaustion [224]. PD-L1+ TANs and their formed NETs are also observed in HCC ascites, which is closely related to tumor progression [225].

TMB, neoantigens, and antigen presentation blockade

A high tumor mutational burden (TMB) theoretically generates abundant neoantigens that may sensitize tumors to ICIs. However, NETs can counteract this advantage by impairing the antigen presentation machinery (APM). First, NET-associated MPO induces lipid peroxidation in dendritic cells (DCs), disrupting endosomal processing pathways and impairing neoantigen cross-presentation to CD8+ T cells [226]. Second, chronic metabolic stress associated with NETs transcriptionally downregulates critical APM components within tumor cells, including TAP1/2, tapasin, and β2-microglobulin, thereby reducing MHC-I surface expression. Finally, even when neoantigens are presented, dense NET strcutures may sterically hinder T-cell receptor (TCR) interactions with MHC-I complexes, limiting effective immunological synapse formation [227]. These findings suggest that evaluation of TMB as a predictor of ICI responsiveness may benefit from integration with assessments of NET burden [228].

Targeting NETs to overcome cancer immunotherapy resistance

NETs not only promote distant tumor metastasis through metabolic interference and signaling transduction but also induce resistance to conventional therapies by forming physical barriers and fostering an immunosuppressive microenvironment [30, 229–232]. Developing strategies to target NETs has thus become pivotal for overcoming cancer immunotherapy resistance and remodeling the immune microenvironment [231]. Current intervention strategies primarily encompass three dimensions: inhibiting formation, degradation/clearance, and blocking downstream interactions. With advancements in nanotechnology and materials science, novel delivery systems are gradually addressing the limitations of traditional drugs. This section summarizes novel drugs and delivery systems targeting NETs and the upstream signaling of NETs and explores the potential and challenges of their clinical translation (Tables 3 and 4, Fig. 5).

Table 3.

Emerging therapeutics and delivery systems Targeting NETs

Therapeutic Category	Target	Specific Drug/Carrier	Cancer Model	Mechanism of Action/Resistance Overcoming	Translational Readiness	Ref.
Small Molecules	PAD4	BMClRu (Oral metal-based inhibitor)	Triple-negative breast cancer	Inhibits NET formation and induces immunogenic cell death (ICD), activating anti-tumor immunity.	Preclinical	[233]
	PAD4	BMS-P5	Multiple myeloma	Blocks NET formation via PAD4 inhibition, effectively delaying disease progression.	Preclinical	[214]
	NE	Sivelestat/GW311616A	DLBCL, Breast cancer	Prevents NE nuclear translocation; prevents NET proteases from cleaving laminin (maintaining cancer cell dormancy).	Preclinical/Early Clinical	[78, 214]
	Autophagy	Chloroquine/Hydroxychloroquine	Pancreatic cancer	Inhibits autophagy-dependent NET formation; reduces platelet aggregation and reverses hypercoagulability.	Advanced Clinical (Repurposed)	[234]
Nanomaterials	PAD4	YW4-03 @ LDH (Layered double hydroxides)	Metastatic tumors	Combines PAD4 inhibition with photodynamic therapy (PDT) to promote DC maturation and T cell infiltration.	Preclinical	[235]
	PAD4	ZD-E-1 (pH-responsive nanodrug)	General Model	Binds PAD4 in acidic TME, downregulates LAG3, upregulates IFN-γ, improving the immune microenvironment.	Preclinical	[236]
	MPO/Iron	TIN (Transformable iron nanochelator)	Anti-PD-L1 resistant model	Regulates intracellular ferrous iron in neutrophils, inhibiting NET formation and enhancing anti-PD-L1 efficacy.	Preclinical	[219]
	NET-DNA	DNase I (pH-responsive hydrogel)	Hepatocellular carcinoma	Neutralizes tumor acidity and releases DNase I, degrading NETs and enhancing adoptive NK cell therapy.	Preclinical	[237]
	ROS & DNA	CNPs (Cerium oxide nanozymes)	General Model	Possesses dual DNase I-like activity to degrade NETs and antioxidant properties to prevent NET regeneration.	Preclinical	[238]
	Feedback Loop	Nanoparticles (Dual nanostrategy)	Metastasis model	Degrades NETs and simultaneously disrupts tumor mitochondrial metabolism, severing the hypoxia-NET feedback loop.	Preclinical	[92]
Biologics & Peptides	CCDC25	R5, R7, R9 (Cationic oligopeptides)	Metastasis model	Competitively bind to NET-DNA, blocking its interaction with the CCDC25 receptor on tumor cells.	Preclinical	[239]
	CCDC25	VNP20009-shCCDC25 (Oncolytic bacteria)	General Model	Delivers shRNA to specifically silence tumor CCDC25, severing downstream pro-metastatic signaling.	Preclinical	[240]
	β1-Integrin	Anti-β1-integrin antibody	Lung to liver metastasis	Blocks physical adhesion between circulating tumor cells and NETs in circulation.	Preclinical	[241]
Traditional & Natural Products	PADI2	Icaritin	Urothelial carcinoma	Inhibits PADI2 expression and histone citrullination, blocking NET-induced metastasis.	Preclinical	[242]
	PI3K/AKT	Pi Ji Pills (TCM Formula)	Pancreatic cancer	Inhibits the PI3K/AKT pathway to reduce NET formation, enhancing gemcitabine efficacy.	Preclinical	[243]
	ROS/CitH3	Resveratrol	Breast cancer	Targets SIRT1 to scavenge ROS and significantly inhibit histone H3 citrullination.	Preclinical	[244]

Table 4.

Clinical trials Targeting upstream NET signaling and microenvironment regulation

Drug	Target	Targeting Modality	Cancer Type	Clinical Trial ID	Phase
Reparixin	CXCR1/2	Indirect (Chemokine Blockade)	TNBC/HER2-/Pancreatic Cancer	NCT02370238/NCT01861054/NCT02996125	Phase 1b/2
AZD5069	CXCR2	Indirect (Chemokine Blockade)	mCRPC/HCC/Head & Neck Cancer	NCT03177187/NCT02583477	Phase 1/2
Navarixin (MK-7123)	CXCR1/2	Indirect (Chemokine Blockade)	Non-Small Cell Lung Cancer (NSCLC)	NCT03473925	Phase 2
SX-682	CXCR1/2	Indirect (Chemokine Blockade)	Metastatic Melanoma/PDAC	NCT03161431/NCT04599140	Phase 1/2
BMS-986253	IL-8 (CXCL8)	Indirect (Cytokine Neutralization)	Melanoma/Solid Tumors	NCT02536469	Phase 1/2
Canakinumab	IL-1β	Indirect (Inflammatory Signaling)	Non-Small Cell Lung Cancer (NSCLC)	NCT03631199 (CANOPY-1)	Phase 3
Anakinra	IL-1 R	Indirect (Inflammatory Signaling)	Pancreatic Cancer/Colorectal Cancer	NCT02550327	Phase 2
Tocilizumab	IL-6 R	Indirect (Inflammatory Signaling)	Ovarian Cancer/Solid Tumors	NCT03126110	Phase 1/2
Motixafortide	CXCR4	Indirect (Chemokine Blockade)	Pancreatic Cancer (PDAC)	NCT02907099	Phase 2
Danvatirsen	STAT3	Indirect (Transcriptional Blockade)	Advanced Solid Tumors/Lymphoma	NCT01839604	Phase 1
Hydroxychloroquine	Autophagy	Direct/Indirect (NETosis pathway)	Pancreatic Cancer/Colorectal Cancer	NCT04524702	Phase 2

Fig. 5.

Neutrophil extracellular traps (NET)-targeting mechanisms for tumor immune microenvironment remodeling and immunotherapy resistance mitigation. NETs confer immunotherapy resistance by constructing physical barriers and establishing an immunosuppressive microenvironment. Key strategies to overcome this resistance include inhibiting NET formation at the source (targeting PAD4, NE, and ROS pathways), degrading established NET physical barriers via DNase I or nanozymes, blocking critical signaling crosstalk between NETs and tumor cells (involving TLR9, CCDC25 receptors, and integrins), and disrupting the sialic acid-Siglec immunosuppressive axis through glycosylation editing. Combined with therapies such as immune checkpoint inhibitor (ICI) use, these NET-targeting approaches aim to remodel the immune microenvironment and restore the cytotoxic capacity of CD8+ T cells and NK cells against tumors, thereby generating synergistic therapeutic effects

Inhibiting NETs formation (source blockade)

Blocking NET generation at the source is currently the most extensively studied strategy. Core targets include PAD4, NE, autophagy pathways, and upstream inflammatory cytokine signaling (Fig. 5, Table 2).

Targeting the PAD4 Enzyme

PAD4-mediated histone citrullination is a critical step in NET formation [245–247] (Fig. 5). However, first-generation pan-PAD inhibitors (such as Cl-Amidine) are limited in clinical application owing to their short half-lives, low bioavailability, and the risk of compromising the anti-infective function of normal physiological neutrophils. To overcome these issues, a novel oral metal-based PAD4 inhibitor BMClRu was developed, which inhibits NET formation and induces immunogenic cell death (ICD), thereby activating anti-tumor immunity [233]. Nanotechnology has further optimized PAD4 inhibitor pharmacokinetics. Loading YW4-03 onto defect-rich layered double hydroxides combined with photodynamic therapy promotes DC maturation and T-cell infiltration while inhibiting NETs [235]. By addressing the poor targeting of traditional small-molecule inhibitors, a pH-responsive self-assembling nanodrug (ZD-E-1) enables specific release in the acidic TME, enters the nucleus to bind PAD4, downregulates LAG3, and upregulates IFN-γ, thereby improving the immune microenvironment [236].

Regulating MPO and iron metabolism

MPO is highly expressed in neutrophils, and its oligomeric state is crucial for chromatin decondensation: dimeric MPO promotes nucleosome disassembly leading to NET formation, whereas monomeric MPO does not disrupt DNA contacts [248]. Although MPO inhibitors are effective in experimental models, completely blocking MPO activity may weaken the organism’s oxidative killing capacity against pathogens. Therefore, more nuanced regulation strategies have emerged. For example, a peptide-drug-conjugated transformable iron nanochelator can regulate ferrous ions specifically in neutrophils, thereby efficiently inhibiting NET formation and enhancing anti-PD-L1 therapy efficacy [219] (Fig. 5, Table 3).

Targeting NE

NE is another key enzyme in the NET formation process, synergizing with matrix metalloproteinase 9 (MMP9) to remodel the extracellular matrix. NE inhibitors, such as Sivelestat or GW311616A, block NETosis by preventing NE nuclear translocation and chromatin decondensation [214] In DLBCL models, GW311616A significantly inhibits tumor-induced NET formation, delaying tumor growth and lymph node dissemination [78]. Notably, as NE is also an important host defense enzyme, long-term systemic inhibition may lead to impaired immune defense; thus, future research should improve tumor-specific NE inhibitor delivery systems.

Targeting autophagy pathways

Autophagy is an intracellular recycling mechanism that plays a key role in cancer-associated NET formation [26, 249, 250]. Research indicates that the autophagy inhibitor chloroquine can effectively inhibit NET formation (Fig. 5). In pancreatic cancer mouse models, chloroquine treatment not only reduced NET-mediated platelet aggregation and tissue factor release but also reversed the hypercoagulable state, thereby providing a novel strategy for leveraging autophagy inhibition to block NETs in the prevention of cancer-associated thrombosis and tumor progression [234]. However, the dual effect of chloroquine, a broad-spectrum autophagy inhibitor, on tumor and immune cells (such as potentially affecting T-cell survival) make dosage control in combination with immunotherapy challenging [251].

Blocking Chemokine signaling axes

Tumor-derived inflammatory factors are major upstream signals for neutrophil recruitment and NETosis. In DLBCL, tumor-derived IL-8 binds to its receptor CXCR2, inducing NET formation via the Src, p38, and ERK signaling pathways; using anti-CXCR2 antibodies or small molecule inhibitors (SB225002) can considerably block this process and inhibit tumor progression [78]. In ovarian cancer, tumor-secreted IL-8, GROα/β, and G-CSF stimulate neutrophil chemotaxis and NET formation; neutralizing these factors reduces NET levels [252]. However, chemokine redundancy in the TME implies that blocking a single target often causes resistance, thereby necessitating multi-target combined interventions.

Blocking neutrophil recruitment through the IL-8/CXCR1/2 signaling axis represents a promising therapeutic strategy. Reparixin, an investigational non-competitive allosteric inhibitor of CXCR1 and CXCR2, has demonstrated encouraging efficacy in preclinical studies and clinical trials (e.g., NCT01861054) [253]. By disrupting CXCL8-mediated signaling, Reparixin not only reduces intratumoral neutrophil infiltration and NET deposition but may also decrease the CSC population through interruption of FASL/FAS survival signaling [254]. Combination therapy involving Reparixin and ICIs therefore represents a potential strategy for reducing NET-mediated immunosuppression at its source.

Traditional Medicine and signaling pathway Regulation

Several traditional medicines and natural products inhibit NETs (Fig. 5). The use of the Traditional Chinese Medicine formula “Pi Ji Wan” enhances gemcitabine efficacy by inhibiting the PI3K/AKT pathway and regulating NET formation in the TME [243]. Treatment with the monomer icaritin blocks histone citrullination and ROS generation by inhibiting PADI2 expression in neutrophils, thereby suppressing NET-induced metastasis [242]. Additionally, the use of cryptotanshinone, an active component of Salvia miltiorrhiza, inhibits neutrophil infiltration by downregulating endothelial E-selectin (CD62E) expression; when combined with ginsenoside Rg1, it significantly reduces NET generation in lung tissue [255].

Preventing NET release (release blockade)

Beyond enzymatic inhibition, interfering with intracellular redox balance is another crucial approach for NET inhibition. ROS accumulation is a key trigger for NETosis [55, 56]. Scavenging ROS using albumin supplementation or antioxidants, such as N-acetylcysteine (NAC), can effectively block oxidative stress–induced NET formation (Fig. 5, Table 3). However, ROS acts as a double-edged sword in tumor immunity: excessive antioxidant therapy may weaken the killing efficiency of CTLs against tumor cells. Therefore, achieving ROS scavenging that specifically targets neutrophils remains a key challenge.

Inhibiting histone H3 citrullination

Histone citrullination is essential for chromatin decondensation. PAD4 inhibitors (e.g., BMS-P5, GSK484, and BB-Cl-Amidine) prevent chromatin decondensation and NET release by specifically blocking the conversion of arginine to citrulline [87, 246] (Fig. 5). This reduces physical DNA meshwork formation and alters TME biochemistry; for example, in colorectal cancer liver metastasis, inhibiting citrullination reverses the epithelial–mesenchymal transition features of tumor cells, thereby inhibiting metastatic growth [256].

Targeting ROS pathways

ROS accumulation is a critical signal for NETosis [55, 72, 134, 251]. Plasma albumin provides thiols to maintain redox balance; when albumin is oxidized or deficient, neutrophils accumulate ROS, triggering non-inflammatory NET formation. Supplementation with albumin or treatment with antioxidants (such as NAC) can scavenge ROS and effectively block redox imbalance-induced NETosis, thereby inhibiting lung metastasis [96].

Degrading formed NETs (barrier clearance)

For already formed NETs, inhibiting upstream generation is often insufficient to remove existing physical barriers. NETs form dense DNA meshwork in the TME, which not only impede CD8+ T-cells and NK cells from contacting and killing tumor cells (“immune exclusion”) but also shield tumor cells from drug effects. Degradation using DNase I is a direct approach to clear this barrier.

Direct application and modification of DNase I

DNase I hydrolyzes the DNA backbone of NETs, disrupting their physical structure. In a breast cancer dormancy awakening model, DNase I digested the DNA scaffold of NETs, preventing the associated proteases (NE and MMP9) from cleaving laminin, thereby preventing the awakening and metastasis of dormant cancer cells [25]. Furthermore, in a pancreatic cancer model, DNase I treatment reduced NET-induced platelet aggregation and alleviated hypercoagulability [234].

New applications for enhancing DNase I delivery efficiency

Currently, the clinical application of DNase I is limited by its short half-life in the blood and poor penetration in dense tumor matrices. Smart delivery systems have been developed to overcome these challenges.

Application of smart delivery systems

Dual pH-responsive hydrogel systems have been designed to load DNase I and tumor acidity neutralizers, thereby releasing enzymes in response to the TME to degrade NETs and enhancing adoptive NK cell therapy efficacy [237] (Fig. 5). Injectable fibrin-alginate hydrogels co-delivering DNase I and propranolol have been used to prevent postoperative recurrence by disrupting NETs, blocking immunosuppression, and antagonizing stress-induced neurotransmitters, thereby reducing lung metastasis [86].

Multifunctional nanozyme strategies

Nanomaterials with enzymatic activity also show promise in enhancing DNase I activity delivery efficiency. Bifunctional cerium oxide nanoparticles not only exhibit DNase I-like activity to degrade NETs but also possess antioxidant properties that lower intracellular ROS levels, thereby preventing NET reformation [238]. In colorectal cancer research, solid lipid nanoparticles encapsulating galunisertib and DNase degraded NETs and inhibited TANs, synergizing with ICIs to remodel the immune microenvironment [257] (Fig. 5).

Blocking interactions between NETs and tumor cells

Interactions between NETs and tumor cells extends beyond simple chemical signal transduction; they form a complex “physical contact and cellular collaboration network.” The DNA scaffolds and associated proteins (such as NE and MMP9) released by NETs act as traps capturing CTCs and provide physical scaffolds for tumor cell adhesion and migration, thereby influencing tumor biological behavior through mechanical force transduction (Fig. 5).

Targeting the CCDC25 Receptor

The transmembrane protein CCDC25 is a key receptor allowing tumor cells to sense NET-DNA [34]. Cationic oligopeptides (such as R5, R7, and R9) can competitively bind to NET-DNA with higher affinity than MPO or citrullinated histones, thereby blocking its interaction with CCDC25 and inhibiting metastasis [239]. Similarly, polyaspartic acid-based cationic nanomaterials (cANP) also interfere with NET-DNA binding to CCDC25, reducing NET infiltration in the liver [258]. Additionally, oncolytic bacterium VNP20009 can deliver shRNA targeting CCDC25, specifically blocking downstream pro-metastatic signaling pathways at the tumor site [240]. This approach preserves normal neutrophil function, but further evaluation of CCDC25 expression and potential toxicity in normal tissues is needed.

Blocking integrin-mediated signaling

Proteases in NETs (NE and MMP9) can sequentially cleave laminin in the extracellular matrix, exposing new antigen epitopes. This epitope serves as a ligand to activate the integrin α3β1 signaling pathway on the tumor cell surface, subsequently activating the FAK/ERK/MLCK/YAP signaling axis and awakening dormant cancer cells. Developing specific antibodies against this NET-remodeled laminin can block its binding to integrins, thereby effectively preventing cancer recurrence in both in vivo and in vitro models [25]. β1-integrin also mediates interactions between CTCs and NETs; blocking it reduces cancer cell adhesion in the liver [241].

Blocking TLR9 signaling pathways and metabolic feedback

NET DNA complexes activate TLR9 on tumor cells. In DLBCL, NETs promote tumor cell proliferation and migration by upregulating TLR9 expression and activating downstream NF-κB, STAT3, and p38 signaling pathways. Using TLR9 antagonists (such as ODN-TTAGGG) or TLR9 knockdown can substantially block the pro-tumorigenic effects of NETs [78]. However, systemic TLR9 blockade may produce complex immunomodulatory consequences in contexts where TLR9 agonists are used to activate anti-tumor immunity Furthermore, a malignant “metabolic collaboration” exists between NETs and tumor mitochondria: hypoxia induces NETs, which in turn enhance mitochondrial metabolism to exacerbate hypoxia. Utilizing nanostrategies to simultaneously target NET degradation and disrupt mitochondrial metabolism can sever this self-amplifying feedback loop [92].

Future perspectives on neutrophil glycosylation editing therapy

Mechanisms between NETs and hypersialylation-driven Immunosuppression

Beyond the physical obstruction imposed by chromatin networks, the biochemical interactions that regulate immune cell fate and immunotherapy resistance within the TME are strongly influenced by surface glycosylation profiles. In particular, the Siglec-sialic acid axis has emerged as a key mechanism linking altered tumor glycans to NET-mediated immunosuppression. A hallmark of malignant transformation is the dysregulation of cellular glycosylation machinery. One major consequence is the pathological upregulation of sialic acid-containing glycans, a phenomenon termed hypersialylation), which extensively decorates tumor cell surfaces and secreted proteins [259]. These terminal sialic acid residues, commonly present in α2–3 and α2–6 linkages, are not merely structural components. Instead, they function as high-affinity ligands for sialic acid-binding immunoglobulin-like lectins (Siglecs), a family of immunomodulatory receptors. Siglec-9 in humans and its murine ortholog Siglec-E are highly expressed on infiltrating neutrophils, PMN-MDSCs, and tumor-associated macrophages (TAMs) [260].

The downstream consequences of this glycan-receptor interaction are strongly immunosuppressive and contribute to impaired anti-tumor immunity. Recent single-cell transcriptomic studies have identified a long-lived tumor-associated neutrophil subset in lung adenocarcinoma characterized by elevated expressions of SiglecF, a sialic acid-binding receptor [261]. Binding of hypersialylated tumor antigens to the extracellular V-set domain of Siglec-9 on neutrophils induces phosphorylation of intracellular immunoreceptor tyrosine-based inhibitory motifs (ITIMs), leading to recruitment and activation of SHP-1 and SHP-2 tyrosine phosphatases [262]. This phosphatase activity suppresses downstream kinase signaling pathways, including MAPK and PI3K cascades, thereby dampening neutrophil oxidative burst responses and reducing neutrophil elastase activity. Consequently, this interaction impairs the ability of neutrophils and NK cells to mediate ADCC against tumor cells and limits the induction of anti-tumor cytotoxic NETosis [197, 260]. Instead, Siglec-engaged neutrophils promote secretion of transforming growth factor-β (TGF-β) and IL-10, thereby establishing a feed-forward immunosuppressive network that shifts myeloid infiltrate toward a pro-tumorigenic and immunotolerant phenotype. This immunosuppressive state further synergizes with the expansion of regulatory T cells (Treg), thereby limiting effective immune surveillance.

Therapeutic explorations Targeting the sialic acid-Siglec Axis

Recognition of this mechanism suggests that physical NET clearance alone may be insufficient to reverse glycosylation-mediated immunosuppression. Consequently, the concept of “precision glycocalyx editing” has gained increasing attention as a potential therapeutic strategy. Approaches such as targeted antibody-sialidase conjugates, including Trastuzumab-Sialidase, aim to enzymatically remove sialic acids from tumor cell surfaces and infiltrating myeloid cells. Similarly, blocking monoclonal antibodies directed against Siglec-9 represent another promising strategy [196, 263]. By disrupting these inhibitory glycan-mediated interactions, such therapies may relieve ITIM-mediated immune suppression, restore neutrophil and NK cell-mediated antibody-dependent cellular cytotoxicity (ADCC), and potentially overcome this mechanism of immunotherapy resistance.

Furthermore, bispecific antibodies such as REGN4018, which target highly glycosylated tumor antigens including MUC16, can facilitate T-cells-mediated tumor cell killing while partially overcoming the “antigen sink” effect caused by abundant soluble mucins [264]. This approach provides an additional strategy for targeting the glycoprotein-rich NET-associated tumor microenvironment. Overall, therapeutic targeting of the Siglec-sialic acid axis represents a promising avenue for reversing NET-mediated immune tolerance.

Combination therapy to remodel the immune microenvironment

Targeting NETs alone is often insufficient to completely reverse resistance [29, 232, 265–267]; combining NET-targeted strategies with ICIs or other therapies can achieve synergistic effects [32, 33, 198, 268, 269]. DLL4-targeted CAR-T therapy eliminates DLL4-positive tumor cells, reducing soluble DLL4-mediated Notch signaling in neutrophils and subsequent NET release, thereby sensitizing tumors to neoadjuvant chemotherapy [270]. In triple-negative breast cancer, metal-based PAD4 inhibitors combined with ICD inducers can activate anti-tumor immunity and overcome immune escape [233]. Additionally, a sonodynamic therapy nanoplatform loaded with PAD4 inhibitors, under ultrasound excitation generates ROS to kill tumor cells, downregulates PD-1 expression, and promotes M1 macrophage polarization, enabling efficient control of metastatic tumors [33]. In EGFR/PD-L1/CTLA-4-driven colorectal cancer, a nanoplatform co-delivering valosin-containing protein inhibitors and bispecific aptamers, together with NET degraders, synergistically enhances immune reprogramming and dual checkpoint blockade [257].

Translational challenges and the systemic risks of Targeting NETs

Although dismantling the NET barrier represents an attractive strategy for re-sensitizing “cold” tumors and overcoming immunotherapy resistance, the clinical translation of NET-targeted therapies faces substantial pharmacological and safety challenges. These limitations arise largely from the dual role of neutrophils and NETs in tumor progression and host defense [271]. NETs are not exclusively pathological structures; they are also essential components of innate immunity required for protection against bacterial, fungal, and protozoan pathogens. Consequently, systemic pharmacological suppression of NET formation may carry significant clinical risks that require careful therapeutic management.

One of the primary translational challenges is the risk of iatrogenic immunosuppression and impaired tissue homeostasis. Systemic administration of broad-spectrum, non-selective PAD4 inhibitors, such as Cl-amidine, or potent NE inhibitors, including high-dose Sivelestat, suppresses chromatin decondensation and NET formation globally. Although this strategy may reduce the physical tumor-protective barrier, it can simultaneously impair antimicrobial NETosis. Preclinical studies have demonstrated that PAD4 knockout mice or animals treated with systemic PAD inhibition exhibit reduced bacterial clearance and increased mortality in models of necrotizing fasciitis and polymicrobial sepsis [58, 214]. Because patients with advanced cancer receiving chemotherapy or immunotherapy are frequently immunocompromised, broad inhibition of NET formation may increase susceptibility to opportunistic pulmonary infections and sepsis. Furthermore, neutrophils and NETs contribute to extracellular matrix repair following tissue injury; therefore, systemic NET inhibition may impair postoperative wound healing and delay tissue repair [24, 272].

Conversely, therapeutic approaches aimed at degrading pre-existing NETs within the TME, particularly through systemic recombinant DNase I administration, may also carry significant toxicity risks. Although DNase I effectively degrades the extracellular DNA scaffold, rapid NET dissolution can release large amounts of previously sequestered cytotoxic granular proteins, including CitH3, MPO, and NE, into the systemic circulation. Free histones, particularly CitH3, can directly damage endothelial cells, activate coagulation pathways, and promote widespread microvascular thrombosis [273]. Simultaneously, liberated proteases may increase endothelial permeability and contribute to systemic inflammatory response syndrome (SIRS) and multi-organ injury [274].

To address these translational barriers, current therapeutic strategies are increasingly focused on TME-responsive delivery systems rather than systemic NET blockade. Intelligent nanocarrier systems, including pH-responsive polymer micelles, hypoxia-activated prodrugs, and ROS-responsive sulfoxide-containing polymers such as PMeSEA, may enable localized PAD4 inhibition or DNase I release within the acidic tumor stroma [107, 236]. Spatially restricted NET-targeting approaches may therefore reduce immunotherapy resistance while preserving systemic antimicrobial defense mechanisms. In addition, future translational studies should focus on distinguishing antimicrobial NETs from pro-tumorigenic NETs. Pro-tumorigenic NETs may possess distinct molecular features, including enrichment of highly oxidized mtDNA and epigenetic modifications such as hypercitrullination and lactylation [275–277]. The development of targeted capable of selectively recognizing these pathological NET signatures, rather than broadly inhibiting all forms of chromatin decondensation, may represent an important step toward safer and more effective NET-directed therapeutic strategies.

Clinical monitoring value of NETs as biomarkers

NETs not only drive tumor progression but also serve as potential biomarkers owing to their unique molecular components, such as citrullinated histone H3 and MPO-DNA complexes. Quantifying NETs—in situ within tumor tissues and in peripheral blood—can complement traditional TNM staging, offering a more precise biological basis for patient prognostic stratification and treatment response prediction (Table 5).

Table 5.

Clinical monitoring value of NETs as biomarkers

Tumor Type	Disease Stage	Sample Type	Detection Marker/Method	Clinical Significance & Prognostic Value	Ref.
Early-Stage Breast Cancer	Early Stage (Localized)	Plasma	NE-DNA Complexes	Circulating NET levels show NO significant association with recurrence risk or clinicopathological features in early stages.	[278]
Pancreatic Neuroendocrine Tumors	Early-Stage (Resectable)	Tumor Tissue	Neutrophil & Macrophage NETs	Positive in situ expression indicates poorer Recurrence-Free Survival (RFS); an independent prognostic risk factor.	[279]
Cervical Cancer	Locally Advanced	Tumor Tissue	Stromal NETs Density (MPO+/H3Cit+)	High stromal NET density is an independent predictor for shortened RFS; improves accuracy when combined with TNM staging.	[280]
Rectal Cancer	Locally Advanced	Pre-treatment Biopsy	NETs Density (MPO/CitH3)	High NET density predicts poor pathological complete response (non-pCR) to neoadjuvant therapy and poorer RFS.	[281, 282]
Bladder Cancer	Locally Advanced	Tumor Tissue	NETs Deposition, PMN/CD8 Ratio	Post-radiotherapy NETs hinder CD8+ T cell infiltration; a high PMN/CD8 ratio correlates with poorer Overall Survival (OS).	[179]
Gastric Cancer	Resectable to Advanced	Serum & Tissue	Serum MPO-DNA, Tissue CitH3	Elevated preoperative serum MPO-DNA correlates with tissue CitH3+ cells and independently predicts poor DFS and OS.	[212]
Colorectal Cancer	Resectable to Metastatic	Tissue & Serum	Tissue Cit-H3, Serum MPO-DNA	High tissue Cit-H3 and preoperative serum MPO-DNA independently predict poorer RFS and metastasis probability.	[209]
Pancreatic Ductal Adenocarcinoma	Resectable to Advanced	Tumor Tissue	Neutrophil NETs (MPO+/CitH3+)	NETs positivity correlates with poorer PFS and OS, especially marking worse survival in the PD-L1 low-expression subgroup.	[283]
Pan-Cancer	Advanced/Terminal	Plasma	H3Cit, cfDNA, Nucleosomes	High circulating H3Cit and cfDNA strongly associate with increased all-cause mortality, independent of coagulation parameters.	[284, 285]

Prognostic value of in situ NETs detection in tumor tissue

Within the TME, high-density NET infiltration is closely associated with poor prognosis across various solid tumors. Furthermore, the NET structure often provides stronger prognostic significance than simple neutrophil counts. Multiplex immunofluorescence has shown that high-density NET infiltration (MPO+/H3Cit+) in the stromal region, rather than within the tumor nests, is an independent prognostic factor for shortened recurrence-free survival (RFS) in cervical cancer [280]. In non-functional pancreatic neuroendocrine tumors, positive expression of neutrophil- and macrophage-derived NETs indicates poorer RFS, and serve as independent prognostic risk factors [279]. Such in situ detection can also reveal the spatial distribution characteristics of NETs; for instance, in bladder cancer, the physical barrier formed by NETs at the tumor-stroma interface hinders CD8+ T-cell infiltration, thereby worsening overall survival (OS) [209].

Incorporating NETs detection into existing clinical staging systems can significantly enhance the accuracy of prognostic assessment. Combining stromal NET density with the TNM staging system in cervical cancer yields significantly better prediction accuracy for 5- and 6-year survival rates than using the TNM system alone. Similarly, in non-functional pancreatic neuroendocrine tumors, a nomogram model integrating WHO grade, TNM stage, and NET status demonstrated a remarkably high concordance index (0.866) [280]. These lines of evidence suggest that tissue-level NET detection reflects the immune status of the TME and serves as a vital supplement to traditional anatomical staging.

Detection of circulating NETs markers and non-invasive monitoring

As part of liquid biopsy, peripheral blood NET markers (e.g., MPO-DNA complexes and H3Cit-DNA) provide non-invasive assessment of tumor burden and prognosis. In GC, preoperative serum MPO-DNA levels were shown to be significantly higher than those in healthy controls, and high MPO-DNA levels significantly correlated with poor disease-free survival and OS [212]. Similarly, in colorectal cancer, elevated preoperative serum MPO-DNA levels independently predict poor RFS [209].

For patients with advance-stage and terminal cancer, circulating NET markers also possess significant risk stratification value. In terminal cancer cohorts, elevated levels of plasma H3Cit-DNA and NE are significantly associated with increased all-cause mortality [284]. Although these patients are typically in a hypercoagulable state, NET markers such as H3Cit-DNA predict prognosis independently of coagulation-fibrinolysis indicators such as D-dimers in this specific cohort, suggesting that NETs may influence prognosis via non-coagulation-dependent pathways [285].

However, the clinical application of circulating NETs as a marker requires consideration of tumor staging. A prospective study on early-stage breast cancer found no significant association between plasma NETs levels at diagnosis and recurrence risk or clinicopathological characteristics [278]. This contrasts with the high NETs levels observed in advanced or metastatic breast cancer, suggesting that circulating NETs may primarily reflect systemic inflammatory responses associated with heavy tumor burden or late-stage disease, and their sensitivity in early tumor screening warrants further verification.

Beyond serological assays, advanced molecular imaging techniques are expanding the monitoring dimension from “content determination” to “activity visualization.” Recent research has developed a ratiometric near-infrared-IIb fluorescent nanoprobe specifically responsive to the overexpressed NE activity within the TME [286]. Unlike traditional static biomarkers, this technology enables quantitative imaging of dynamic changes in TANs, distinguishing immunotherapy responders from non-responders in vivo. This visualization strategy provides a more intuitive, non-invasive tool for the real-time assessment of NET-associated immune status.

Guiding significance of NETs markers in predicting treatment response

NET detection is not only useful for prognostic assessment but also predicts response to chemoradiotherapy and neoadjuvant therapy [30, 179, 287–289]. In locally advanced rectal cancer, high density of NETs in pre-treatment biopsy tissues significantly correlates with low pathological complete response to neoadjuvant therapy, suggesting that NETs may represent potential indicators of therapeutic resistance [281, 282]. In PDAC, patients with low neutrophil infiltration or negative NET status achieve better survival from adjuvant chemotherapy, aiding in identifying patients most likely to benefit from chemotherapy [283].

Furthermore, NET presence is associated with immune checkpoint molecule expression and potential immunotherapy targets. In pancreatic cancer, NET positivity correlates with B7–H4 expression, and among patients with low PD-L1 expression, NET-positive individuals exhibit poorer survival, suggesting that targeting NETs may provide an alternative strategy for patients insensitive to immunotherapy [283]. In bladder cancer radiotherapy models, radiation-induced NET formation hinders CD8+ T-cell infiltration, and the intratumoral PMN/CD8 ratio may serve as a predictive marker for radiotherapy resistance and prognosis [179]. Therefore, incorporating NET detection into routine pathology can guide personalized combination therapies, such as DNase I or NE inhibitor co-administration, enhancing translational and clinical decision-making.

Standardization of NET biomarker panels and clinical thresholds

To fully integrate NETs into precision oncology and support their implementation in early-phase clinical trials, including those guided by the Society for Immunotherapy of Cancer and Prostate Cancer Working Group 4, the translation of preclinical findings into standardized and quantifiable biomarker panels is essential. Current evidence indicates that measurement of individual NETosis-associated components provides limited predictive accuracy; therefore, the development of comprehensive biomarker panels and rigorous assay standardization is necessary.

Panel selection: specificity versus sensitivity

An optimal clinical biomarker panel for monitoring immunotherapy resistance should incorporate circulating citrullinated histone H3 bound to cell-free DNA (H3Cit-DNA), myeloperoxidase (MPO)-DNA complexes, and neutrophil elastase (NE). However, the diagnostic significance of these markers differs substantially and should be interpreted accordingly. MPO-DNA complexes and generic cell-free DNA (cfDNA) demonstrate high analytical sensitivity and often correlate with tumor burden; however, their clinical specificity remains limited. MPO and NE may be released into the systemic circulation through conventional neutrophil degranulation in the absence of true NETosis, whereas cfDNA levels may also increase because of tumor necrosis and apoptosis. In contrast, H3Cit-DNA is considered one of the most specific biomarkers of NETosis. Histone citrullination is catalyzed by PAD4 [290], a critical enzymatic step required for chromatin decondensation. Consequently, detection of H3Cit-DNA provides stronger evidence for the occurrence of structured NETosis within the TME and may more accurately reflect the physical and biochemical barriers associated with immunotherapy resistance.

Addressing assay standardization and technological transition

The lack of assay standardization remains a major challenge in the clinical implementation of NET biomarkers. Traditionally, NET quantification has relied on manual fluorescence microscopy and laboratory-developed enzyme-linked immunosorbent assays (ELISAs) targeting MPO-DNA or H3Cit-DNA complexes. However, these conventional methodologies are associated with substantial inter-laboratory variability, observer-dependent bias, cross-reactivity with non-specific nucleosomes, and instability of enzymatically modified in vitro standards.

To address these limitations, diagnostic approaches are increasingly transitioning toward automated, high-throughput chemiluminescent immunoassays (ChLIA). Recent advances involving highly specific monoclonal antibodies combined with semi-synthetic nucleosome controls, including synthetically generated H3.1-nucleosomes containing defined arginine-to-citrulline substitutions at positions 2, 8, and 17, have significantly improved assay reproducibility [291]. Measurement of circulating H3.1-nucleosomes enables the generation of more reproducible calibration curves that are less dependent on variable enzymatic efficiencies, thereby improving inter-laboratory consistency and facilitating multicenter clinical trial comparisons [292].

Establishing clinical thresholds for patient selection

With the development of standardized detection methods, establishing objective clinical thresholds represents a critical step in therapeutic patient stratification. Accumulating prognostic evidence across multiple solid tumors suggests that biomarker cutoff values may help distinguish distinct clinical outcomes. Advanced analytical models have proposed thresholds based on the upper quartiles of baseline distributions. For example, circulating H3Cit levels exceeding 44.9 ng/mL and systemic cfDNA levels above defined thresholds (e.g., 419 ng/mL in gastric cancer studies), have been associated with therapeutic failure, immune exclusion, and increased short-term mortality [284, 293]. Patients exceeding these thresholds may possess TMEs characterized by extensive NET-associated barriers, potentially reducing the effectiveness of PD-1/PD-L1 monotherapy. Consequently, incorporation of validated H3Cit-DNA and MPO-DNA thresholds into early-phase ICI clinical trial protocols may help identify patients who could benefit from adjunctive NET-degradation therapies, such as localized DNase I delivery or PAD4 inhibition, prior to immune checkpoint blockade.

Conclusions

The elucidation of NETs has fundamentally reshaped our understanding of the TME. NETs are no longer viewed merely as innate defense mechanisms but as a central hub “hijacked” by malignancies to create a profound immunosuppressive barrier. As detailed in this review, NETs orchestrate a “tripartite” resistance mechanism encompassing physical, biochemical, and metabolic dimensions: they form dense physical barriers that exclude T-cell infiltration, serve as scaffolds for checkpoint ligands (e.g., PD-L1) to induce exhaustion, and rewire the metabolic landscape via mitochondrial crosstalk. This multidimensional defense network explains why ICI monotherapy often fails in “cold” or immune-excluded tumors. Shifting the therapeutic focus from simply activating T-cells to dismantling these physical and metabolic obstacles—particularly by targeting NETs—represents a key frontier in overcoming immunotherapy resistance.

However, translating NET-targeting therapies into clinical practice faces a “double-edged sword” challenge. Neutrophils and NETs are essential for pathogen clearance and tissue repair; hence, systemic abrogation (e.g., through the use of broad-spectrum PAD4 inhibitors or high-dose DNase I) poses significant risks of iatrogenic immunodeficiency or sepsis. Future interventions require precise discrimination between “anti-microbial NETs” and “pro-tumorigenic NETs” using molecular signatures, such as oxidized mtDNA or specific post-translational modifications like citrullination. Furthermore, leveraging TME-specific features—acidity, hypoxia, or elevated enzymatic activity—through smart nanodelivery systems can enhance local efficacy while minimizing systemic toxicity.

Effective NET-targeted strategies must go beyond physical disruption and deep immune ecosystem reprogramming. Glycocalyx editing shows that neutrophil hypersialylation acts as an “invisible cloak” maintaining their malignant phenotype; thus, targeting the Siglec-sialic acid axis could revolutionize the reversal of NET-mediated immune tolerance. Moreover, the metabolic feedback loop between NETs and tumor mitochondria reveals a profound therapeutic vulnerability; severing this “metabolic collaboration” may yield more durable effects than direct cytotoxicity. In clinical translation, establishing a companion diagnostic framework based on circulating NET biomarkers (e.g., serum MPO-DNA) is essential for identifying patients with high NET burden. Ultimately, incorporating NET blockade as a “priming” strategy to convert immune-excluded tumors into “hot” environments sensitive to ICI treatment constitutes a vital step toward reshaping the cancer immunotherapy landscape.

Abbreviations

CAFs

Cancer-associated fibroblasts

CTLs

Cytotoxic T lymphocytes

DCs

Dendritic cells

DLBCL

Diffuse large B-cell lymphoma

GC

Gastric cancer

GC-MSCs

Gastric cancer–associated mesenchymal stem cells

HCC

Hepatocellular carcinoma

ICIs

Immune checkpoint inhibitors

mtDNA

Mitochondrial DNA

NAC

N-acetylcysteine

NETs

Neutrophil extracellular traps

PMN-MDSCs

Polymorphonuclear myeloid-derived suppressor cells

RFS

Recurrence-free survival

ROS

Reactive oxygen species

TANs

Tumor-associated neutrophils

TME

Tumor microenvironment

OS

Overall survival

Author contributions

XL Li and DY Shan researched the literature and drafted the manuscript. LM Zheng, CH Zuo and L Zhou contributed to data collection and figure preparation. M Yan and JW Li provided critical revisions and domain-specific insights. M Yan and JW Li conceived and supervised the project. All authors discussed the content and contributed to the final version of the manuscript.

Funding

This study was supported by the National Natural Science Foundation of China (82500253, 82170192, 82300116), The Science and Technology Innovation Program of Hunan Province (No. 2024RC1025) and the Key R&D plan of Hunan Province (No. 2025JK2131) and the Key project of State Key Laboratory of Neurology and Oncology Drug Development (SKLSIM-20250169).

Data availability

No datasets were generated or analyzed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable

Consent for publication

All authors consent to publication.

Competing interests

The authors declare no competing interests.