Colorectal cancer (CRC) is among the most prevalent malignant tumors worldwide, with its incidence ranking fourth and mortality ranking third, according to the latest global cancer data (1). The CRC encompasses hereditary, sporadic, and colitis-associated colorectal cancer (CAC) forms (2). Indeed, extensive epidemiological and experimental studies have consistently revealed that chronic inflammation plays a pivotal role in the initiation and progression of CRC (3–6). Depending on the stage at which inflammation affects CRC development, it can be categorized into three types: pre-tumor chronic inflammation, tumor-triggered inflammation, and treatment-induced inflammation. These processes activate tumor-promoting innate immune cells, inhibit anti-tumor adaptive immune cells, and establish an immunosuppressive tumor microenvironment (TME) (5).

Neutrophils serve as the primary line of defense against microbial infection by employing various mechanisms: (1) phagocytosis, wherein they engulf pathogenic microorganisms to form phagocytic vacuoles and generate toxic chemicals for microbial eradication (7); (2) degranulation, involving the controlled release of cytotoxic enzymes and proteases from intracellular vesicles to eliminate microorganisms (8); and (3) the formation of neutrophil extracellular traps (NETs), which are protein-DNA structures that ensnare and neutralize microorganisms (9). These processes encompass both oxidative and non-oxidative pathways that drive neutrophils to exert their antimicrobial effects with potent killing capabilities (10, 11). It has been traditionally believed that circulating neutrophils cannot proliferate after reaching maturity, leading to rapid depletion under steady-state conditions with a half-life of less than 1 day (12). However, it is important to note that neutrophils exhibit high plasticity and can transition into different functional states in various disease contexts, consequently impacting their lifespan (13). Recent studies have demonstrated that neutrophils recruited to tissues can undergo reverse migration back into the bloodstream after functioning and can even migrate to distant organs to induce inflammatory responses (14, 15).

Neutrophils not only play a crucial role in the body’s defense against infection, but also exert significant influence in tissue repair, autoimmune diseases, and tumor development (16–18). Due to the heterogeneity and plasticity of the tumor-associated neutrophils (TANs) within the TME, the precise subtype characterization of TAN has become a crucial research focus. Neutrophils can be polarized into TAN1 or TAN2 subtypes in the TME. TAN1 exhibits increased expression of immune-activating cytokines and chemokines, along with reduced arginase. While TAN2 demonstrates elevated expression of angiogenic factors, stroma-degrading enzymes, and substantial arginase, exerting immunosuppressive effects (19). The absence of widely recognized subgroup markers has hindered comprehension of neutrophil heterogeneity. Consequently, higher-dimensional techniques are needed for investigation. A study employed time-of-flight mass spectrometry to identify seven neutrophil clusters in human melanoma: the CD117⁺CD66b⁺ progenitor population and Cneut1 to Cneut6. It shows the heterogeneity in neutrophil differentiation. Notably, Cneut2 has the highest expression of CD101, CD10, and CD16, indicating its maturity (20). Furthermore, recent advancements in single-cell RNA sequencing (scRNA-seq) analysis have identified distinct TAN subsets in cancers. The four TAN subgroups in pancreatic ductal adenocarcinoma are designated as TAN1 to TAN4. TAN1 is characterized by elevated expression of VEGFA, PLAU, and LGALS3. TAN2 expresses inflammation-related genes NLRP3 and PDE4B. TAN3 is a transitional stage with high expression of transendothelial migration genes. TAN4 preferentially expresses interferon-stimulated genes, such as IFIT1, IFIT2, and ISG1 (21). The six TAN subgroups identified in primary liver cancer are Neu_01_MMP8, Neu_07_APOA2, Neu_08_CD74, Neu_09_IFIT1, Neu_10_SPP1, and Neu_11_CCL4. Subgroups Neu_09/10/11 are associated with a poorer prognosis. The expression of CD274, encoding PD-L1, is elevated in TANs, highest in the Neu_09_IFIT1 subgroup (22). To further investigate the tissue and phenotype plasticity of neutrophils, researchers integrated transcriptome data from 225 samples across 17 cancers. The analysis revealed substantial heterogeneity, characterized by ten distinct cell states, including S100A12⁺, HLA-DR⁺CD74⁺, VEGFA⁺SPP1⁺, TXNIP⁺, CXCL8⁺IL1B⁺, CXCR2⁺, IFIT1⁺, ISG15⁺, MMP9⁺, NFKBIZ⁺, HIF1A⁺, and ARG1⁺ neutrophils (23).

All these indicate that neutrophils are not single-function cells with terminal differentiation, but rather their differentiation and maturation trajectories can be significantly altered under specific conditions, leading to diverse functional outcomes. Here, we discuss neutrophils’ multifaceted contributions to CRC development, encompassing NET formation, interactions with gut microbiota, and intricate crosstalk within the TME. Furthermore, we discuss the inhibitory effect of neutrophils on CRC progression and their potential to guide the prognosis and treatment of CRC. We hope to provide novel perspectives for optimizing CRC treatment strategies.

The classical process of intravascular neutrophil recruitment during inflammation involves five coordinated stages: tethering, rolling, adhesion, crawling, and transmigration (24). The entire process is regulated by a variety of inflammatory-related molecules, including chemoattractant subfamilies (e.g., chemokines and their receptors, chemotactic lipids, complement anaphylatoxins), cytokines, integrins, and cell adhesion molecules (24–26). In addition to the classical molecules mentioned above, a recent study discovered that neutrophil surface RNAs with glycan modifications facilitate neutrophil recruitment and promote neutrophil-endothelial interactions by binding to P-selectin on endothelial cells (27).

Tumor progression is always regarded as a chronic inflammatory process (28, 29), during which the recruitment pattern of neutrophils is consistent with physiological inflammation. Chemotactic compositions present in the TME mainly include CXCL8 family chemokines (e.g., CXCL1/2/3/5/6/7/8) (30), cytokine interleukin (e.g. IL-1β, IL-8, IL-17) (31–33), and the associated signaling pathways (e.g., TGF-β/Smad3) (34), responsible for neutrophil recruitment and polarization. In CRC, chemokines (e.g., CXCL1/2) and cytokines (e.g., IL-8, and TGFβ) (35–37) are also involved in regulating neutrophil enrichment to the TME and facilitating their polarization towards TAN2 phenotype, while IL-22 plays a role in recruiting beneficial neutrophils (38).

Neutrophils are increased in peripheral blood and tumors of patients with CRC, as compared to healthy donor peripheral blood and paired adjacent non-tumor colon tissue, and are associated with higher cancer stage and histological grade (39). The phenomenon has been found that ulcerative colitis patients with more neutrophil infiltration are at a higher risk of developing CAC (40). In both colitis and CAC models, neutrophil infiltration increased as the disease progressed, and neutrophils may promote the transition from colitis to CAC (40). The above studies suggest neutrophils are important in the development and progression of CRC.

Most studies on the role of neutrophils in tumors have primarily reported their promotion of tumor development; however, they indeed possess anti-tumor functionality in CRC ( Figure 4 ). This is intricately associated with the complex TME, tumor stage, and neutrophil heterogeneity.

Neutrophils possess high plasticity and can transit into different anti-tumor roles. Studies have demonstrated a significant increase of CD177⁺ neutrophils in CAC and CRC tissues. Furthermore, patients with high infiltration of CD177⁺ neutrophils (>8 cells/HPF in CRC sections) exhibited improved 5-year overall survival (OS, 80.4% versus 69.3%, P = 0.007) and 5-year disease-free survival (DFS, 81.7% versus 71.6%, P = 0.010). Subsequent investigation confirmed the ability of CD177⁺ neutrophils to inhibit epithelial cell tumorigenesis (97). Additionally, a study reported that TANs with high levels of ICAM-1 and CD95 displayed an anti-tumor phenotype and were found to infiltrate the invasive edge of early colon tumors, suggesting their potential role in combating the disease at an early stage (98). The strong colocalization of HLA-DR⁺ neutrophils and CD8⁺ T cells in COAD indicates a spatial association between antigen-presenting neutrophils and T cells. HLA-DR⁺ neutrophils can widely trigger T cell activation, antigen reactivity, and cytotoxicity, promoting T cell response and playing a synergistic role in immunotherapy (23). Transcriptome analysis revealed dysregulation of genes involved in antimicrobial and inflammatory processes in neutrophil-deficient colon tumors. The mechanism study demonstrated that polymorphonuclear neutrophils in early colon tumors inhibited the proliferation and aggressiveness of tumor cells by suppressing IL-17-mediated bacteria-dependent inflammatory response and the expansion of B cells in the colon (99).

IL-22 is generated by various immune cells in innate and adaptive immune systems, including group 3 innate lymphoid cells, NK cells, and CD8⁺ T cells (100, 101). Combined survival analysis revealed that the protective performance of IL-22 in CRC relied on the presence of neutrophils. IL-22 can stimulate the secretion of chemokines CXCL1, CXCL2, and CXCL3 by colon tumor cells for recruiting neutrophils. IL-22 augmented the T cell response through the recruitment of beneficial neutrophils (CD66b⁺ cells) and contributed to favorable 5-year OS (58% versus 43%, P = 0.004) (38). Co-cultivation with CD66b⁺ neutrophils enhanced the response of CD8⁺ T cells to the T cell receptors, leading to increased activation and proliferation of CD8⁺ T cells, and expanding the population of cells exhibiting the “central memory” phenotype characterized by CD45RO/CD62L expression (102). A recent study has demonstrated that neutrophil-deficient mice (Csf3r^−/− mice) showed higher susceptibility to carcinogenesis. Neutrophils may confer protection against intestinal inflammation and CAC by modulating the gut microbiome and initiating activation of an γδ T cells-derived IL-22-dependent tissue repair approach (103). IL-22 has been demonstrated to accelerate the development of CRC (104–106), which contradicts the previously mentioned association between IL-22⁺ cell infiltration and improved prognosis. This indicates that the interaction between neutrophils and IL-22 within the established TME exerts multifaceted effects on tumor progression.

Given the multifaceted roles of neutrophils in the pathogenesis and progression of CRC, this review provides several novel insights for developing clinical treatment strategies for CRC patients. Neutrophils are essential components of the host’s defense against infection (24), and their depletion could result in significant immunosuppression. Although existing studies indicate that neutrophils contribute to tumor progression, clinical trials targeting neutrophils remain limited due to concerns about inducing neutropenia. The lack of widely recognized subgroup markers hinders a comprehensive understanding of the heterogeneity of neutrophils, making therapies targeting specific subgroups of neutrophils difficult. Therefore, most research has focused on inhibiting neutrophil-associated molecules known to promote tumor growth and aggressiveness (107). Therefore, we mainly review the clinical studies of key molecules or pathways related to neutrophil function. While not directly depleting neutrophils, these interventions can elucidate the effects of neutrophil-associated targeted or combination therapies on CRC, highlighting their potential to guide clinical therapies.

The importance of the CXCR1/2-IL8 axis in neutrophil chemotaxis and the multiple pro-tumor functions mediated by this axis (35–37) make CXCR1/2-IL8 a desirable therapeutic target. Other classical cytokines such as G-CSF, TGF-β, and VEGF are closely related to the functional status of neutrophils during tumor development (68, 108, 109) and are also important therapeutic targets in CRC treatment. Researchers have found that neutrophils from CRC samples strongly express Bv8/PROK2, while inhibition of G-CSF or Bv8/PROK2 can improve the efficacy of anti-VEGF antibodies and prevent the emergence of drug resistance. The G-CSF/Bv8/PROK2 axis holds potential as a therapeutic target for CRC when combined with anti-VEGF agents (110). In both colitis and CAC models, neutrophil infiltration and pSTAT3 expression increased as the disease progressed. This suggests that neutrophils may promote the transition from colitis to CRC through the JAK/STAT pathway (40). Several clinical trials have explored these targets in depth ( Table 1 ).

Studies have shown neutrophils participate in the inhibitory immune microenvironment through the CD80/CD86-CTLA4 pathway, resulting in poor response to immune checkpoint inhibitors (ICIs) in MSI-H patients. In a mouse CRC model, blocking the CD80/CD86-CTLA4 axis under inflammatory conditions improves tumor response to PD-1 blocking (86). The PD1/PD-L1 axis is commonly associated with immune escape and tumor progression. Studies of mouse breast cancer have revealed that tumor cells expressing PD-1 inhibit the cytotoxicity of neutrophils and enhance their metastatic potential through the PD-L1/PD-1 axis. PD-L1⁺ neutrophils are less cytotoxic than PD-L1 neutrophils, and blocking the PD-L1/PD-1 interaction can enhance the anti-tumor cytotoxicity of neutrophils (111). PD-L1⁺ neutrophils in HCC patients can effectively inhibit the proliferation and activation of T cells, and blocking PD-L1 can partially reverse this inhibition (112). In a mouse model of CRC, tumor-derived G-CSF induces neutrophils to express PD-L1 through the STAT3 pathway, and neutrophils promote tumor by inhibiting anti-tumor immunity of NK cells dependent on PD-L1/PD-1. Neutrophil depletion combined with anti-PD-L1 can improve the survival of the mice with CRC (113). These strongly suggest that the PD1/PD-L1 axis is an important therapeutic target under conditions of neutrophil infiltration, and targeting TANs may be an important strategy to prevent PD1/PD-L1-associated tumor immune evasion.

Other functions of TANs are currently under investigation in preclinical studies. PAD4 expressed in neutrophils plays an important role in forming NETs (114). Studies on nude mouse xenograft models have shown that GSK484 (PAD4 inhibitor) can improve the radiosensitivity of CRC, inhibit NET formation, and inhibit tumor growth (115). F. nucleatum infection can induce PD-L1 expression by activating the NF- κB/STAT3 signaling in neutrophils, and CX3CR1⁺PD-L1⁺ neutrophils infiltration promotes CRC metastasis and weakens the efficacy of immunotherapy. Treatment with doxycycline eliminated F. nucleatum intracellular, thereby reducing CX3CR1⁺PD-L1⁺ neutrophils populations and slowing F. nucleatum-promoted tumor growth and metastasis in mice (116). This suggests that the combination of antibiotics and ICIs may benefit such patients.

It should be noted that it has been reported that the combination of anti-TGF-β and anti-PD-L1 therapies may lead to tumor resistance, which has been linked to the upregulation of CR-related metabolic pathways in mice (117). It is recommended that researchers and clinicians should employ a multifaceted approach to assess patient status and develop a precise medical strategy when formulating combination drug regimens targeting TAN.

Multivariate survival analysis found that the high levels (more than 60 per TMA spot) of intratumoral CD66b⁺ neutrophil was an independent risk factor for adverse overall patient survival (hazard ratio [HR]: 2.040; 95% confidence interval [CI]: 1.186–3.843; P = 0.010) ( 118). Multivariate Cox analysis demonstrated that high levels of intra-tumoral CD66b⁺ neutrophils are significantly associated with decreased OS and DFS (119). Interestingly, a separate study reported that co-infiltration of CD66b⁺ neutrophils and CD8⁺ T cells is associated with a more favorable prognosis compared to sole infiltration by CD8⁺ T cells in CRC, potentially due to the complex interplay between these immune cells (102). However, the investigation into the prognostic role of neutrophil infiltration at intra-tumoral subsites in CRC revealed that low levels (semi-quantitatively score, no/sporadic) of neutrophil infiltration in the front of the tumor is an independent prognostic factor for the poor prognosis of patients with early colon cancer (HR: 2.32, 95% CI: 1.45-3.73, P <.001) (120). These ostensibly contradictory findings suggest that the prognostic role of neutrophils is multifaceted and may be influenced by their infiltration sites within the TME and their functional status.

The neutrophil-to-lymphocyte ratio (NLR) has been utilized in various types of tumors as a prognostic indicator for poor survival, and it can be employed to inform immune-related treatment strategies and predict clinical outcomes (121–124). Colon cancer patients with a high NLR (> 3.0) exhibit poor 5-year OS (87.0% versus 94.5%, P = 0.042) and 5-year relapse-free survival (RFS) (77.9% versus 87.8%, P = 0.032) (125). The high NLR serves as a crucial prognostic indicator for advanced colon cancer, particularly in the case of left-sided colon cancer (5-year OS: 86.4% versus 95.2%, P = 0.014; 5-year RFS:79.2% versus 87.3%, P = 0.047) (125). Multivariate analysis showed that increased NLR (> 5.0) in CRC liver metastasis was associated with poor 5-year OS (27% versus 47%, P < 0.01) and 5-year DFS (6% versus 37%, P < 0.01) (126). Low baseline NLR (<5.11, HR: 0.42, 95% CI: 0.21-0.84, P = 0.014) and early NLR reduction after two courses of immunotherapy (HR: 0.33, 95% CI: 0.18-0.61, P < 0.001) were significantly associated with better outcomes in CRC patients. Among them, patients with a low baseline NLR and an early decline in NLR exhibited the longest median OS (n = 53, median OS = 29.63 months) (127). Further analysis showed that a combination of NLR and tumor mutation burden provided additional predictive capacity (127). Currently, several clinical studies have investigated the practical clinical significance, including the exploration of NLR (NCT05673343, NCT06495827), combination of NLR with C-reactive Protein (NCT05129046), and neutrophils-circulating tumor cells analysis (NCT05793775) in diagnosis and prediction of advanced CRC.

Further research on neutrophil-related genes in CRC patients identified 17 genes significantly associated with OS. Based on these 17 genes, a prognostic risk score (PRS) system was developed. Patients in the high-PRS group were more likely to experience tumor recurrence and metastasis. However, high-PRS group patients showed improved response to immunotherapy, possibly due to increased tumor mutation burden and MSI. Therefore, the PRS can serve as a prognostic model for guiding individualized treatment for colon cancer patients (128).

Neutrophil infiltration can also predict the response of CRC patients to the drug therapies. A study revealed that neutrophil infiltration increases in stage I-III CRC but decreases significantly in stage IV, possibly indicating immune escape in advanced disease. In stage III CRC, high TAN infiltration is linked to a favorable response to a 5-fluorouracil-based chemotherapy regimen (better DFS, HR: 0.42, p = 0.01), suggesting that evaluating TAN infiltration may help identify patients who would benefit from this therapy (129). Due to neutrophil-associated immunosuppression, local inflammatory conditions in MSI-H CRC can hinder tumor response to ICIs, potentially leading to ICI resistance. It has been demonstrated that inflammation infiltration and high NLR (> 3.0) are clinical features of adverse ICI reactions in MSI-H CRC (86).

CRC is characterized by chronic inflammation, which is thought to function in tumor progression and metastasis. The neutrophils are vital for forming the inflammatory TME via the action of NETs, their interaction with the gut microbiota, and crosstalk with other cells such as tumor cells, T cells, and macrophages. However, neutrophils exhibit a dual role in cancers, which may be achieved through two mechanisms (13, 18, 23): (1) Neutrophils themselves possess heterogeneity, with different subtypes performing distinct functions; (2) Neutrophils undergo reprogramming within the TME. The function of neutrophils might result from a combination of these two mechanisms, which requires more evidence.

In CRC, TANs play a crucial role in the immune evasion mechanism during therapy by shaping an immunosuppressive microenvironment and promoting tumor resistance to immune checkpoint therapy through inflammation response. Targeting the molecules of the pathways involved in TANs-mediated tumor progression and therapeutic resistance can lead to the development of multiple synergistic therapeutic strategies. Several approaches have been explored for targeting TANs (107, 130): (1) Inhibition of TANs enrichment, proliferation, and polarization (131); (2) Suppression of neutrophil-induced immune escape, especially enhancing reaction of tumor cells to ICIs (132); (3) Inhibition of other key molecules associated with neutrophil function (133); (4) Utilization of neutrophils as drug delivery carriers based on their innate inflammatory response sensitivity and ability to cross physical barriers (134, 135). These pathways alone or in combination may bring breakthroughs in treating CRC.

The anti-tumor effect of neutrophils in CRC should not be disregarded, and it is worth considering amplifying the antagonistic effect of specific neutrophils subtypes to control the progression of CRC. Furthermore, a study has demonstrated that in successful immunotherapy for mice with lung tumors, neutrophils characterized by interferon-stimulated genes rapidly accumulate in tumor tissues and exhibit an anti-tumor phenotype (136). Another perspective suggests that T cells can eliminate tumor cells with high antigen density, while neutrophils kill tumor antigen escape variants, and immunotherapy with the combination of the two can eliminate highly heterogeneous tumors (137). Therefore, immunotherapy that induces anti-tumor T cells can be combined with therapy that optimizes the anti-tumor function of neutrophils, which may lead to more durable tumor control following treatment.

Neutrophils remain technical challenges in study due to their fragility, short lifespan, and low RNA content. Although mouse models are commonly employed for studying human tumors, they do not accurately reflect the in vivo conditions. Research has demonstrated that human neutrophils selectively release active neutrophilic elastase to eliminate various cancer cells while sparing non-cancer cells; however, this property is absent in mouse neutrophils (138). Therefore, there is a pressing need to develop reliable research techniques for studying human-derived TANs to obtain more credible insights into the mechanisms by which TANs regulate cancers. To simulate the in vivo environment more accurately, we propose that organ-on-a-chip represents an advanced model for in vitro research of neutrophils. It is feasible to recreate a TME on a microfluidic chip, presenting significant application potential (139). Intravital microscopy (IVM) combined with fluorescent neutrophil reporter mice enabled real-time visualization of neutrophil dynamics in tumors. Commonly used in vivo imaging techniques include confocal and multiphoton microscopy. Imaging techniques can broaden our understanding of how neutrophils are involved in cancer development, providing a powerful tool for neutrophilic research (140). The microbiome is involved in the establishment of the TME. The microbiome can trigger the immune system response directly or influence the functional state of immune cells via metabolic processes and metabolites (141). The combination of microbiome and immunotherapy is also a direction that researchers should keep exploring (142). Proteomic, biomechanical, and functional analyses can elucidate neutrophil heterogeneity in systemic lupus erythematosus, offering another possible method to investigate neutrophil heterogeneity in CRC (143).

In recent years, scRNA-seq technology has been utilized for the analysis of TAN characteristics in tumors, revealing the heterogeneous subtypes within the TME and elucidating the role of each subtype in tumor development and response to therapy (21, 22). Despite extensive studies on consensus molecular subtype classification of CRC (144–146), there is a lack of comprehensive analysis on TAN subtypes using scRNA-seq, which could provide valuable insights into assessing CRC prognosis and treatment based on TAN heterogeneity. The following directions can be considered for scRNA-seq of neutrophils from CRC patients: (1) Identify the TAN subtypes in human CRC; (2) Characterize molecular markers specific to each TAN subtype, with a focus on the pro-tumor subtype; (3) Investigate the differentiation pathways that lead to the development of pro-tumor neutrophil subtype; (4) Examine the metabolic reprogramming in tumor-promoting TANs and evaluate how these metabolic changes significantly influence immune cell function; (5) Analyze the interactions between specific TAN subtypes and other immune cells such as T cells and NK cells; (6) Based on these findings, explore strategies to block key molecules involved in the differentiation pathway of pro-tumor neutrophils. For instance, inhibiting upstream regulators and disrupting the active metabolic programs of TANs can suppress the pro-tumor functions and immunosuppressive activities of neutrophils, thereby identifying potential TAN-related therapeutic targets for CRC.

ScRNA-seq research of neutrophils poses several challenges due to their biological characteristics, functional heterogeneity, and technical limitations. Neutrophils are fragile and short-lived cells (with a half-life of 7 to 10 hours) (147) and have low mRNA content(0.33 μg per million cells) (148), challenging the isolation of neutrophils and the subsequent extraction of RNA. In a recent study, the lack of crucial biomarkers could restrict the accuracy of identifying the cells among neutrophil subsets with similar transcriptomic characteristics (149). ScRNA-seq is also insufficient to fully capture the spatiotemporal characteristics of neutrophils. The transcriptomes of neutrophils can vary depending on their activation states and microenvironment, making it difficult to comprehensively describe their transcriptional landscapes at the single-cell level (21). Existing datasets lack spatial information regarding tissue distribution and cell-cell interactions; fortunately, rapidly evolving spatial profiling technologies may address these limitations (150). Other inherent limitations of scRNA-seq technology should not be overlooked, including its limited sensitivity, scale, accuracy, and noise introduced by single-cell RNA preamplification (151). Overcoming these challenges requires a robust experimental design, advanced computational methodologies, and a profound understanding of neutrophil characteristics.