Neutrophils in triple-negative breast cancer: an underestimated player with increasingly recognized importance

Over the past decade, tumor-associated neutrophils (TANs) have been increasingly recognized as a key functional identity with the potential to impact breast cancer prognosis, especially the TNBC subtype [10‐12]. TANs can exist in the circulation (circulating neutrophils) as well as in the microenvironment (tumor-infiltrating neutrophils).

Circulating tumor cells (CTCs) enter the vasculature from the primary tumor site and contain metastasis-initiating cells that are able to extravasate and colonize secondary organs [13]. It is now clear that circulating immune cells can form clusters with CTCs and help them survive the harsh hemodynamic conditions and metastasize. Neutrophils could attach to CTCs, which leads to the adhesion of tumor cells to the blood vessels, ultimately leading to transendothelial migration and extracellular matrix remodeling [14]. In a 2014 report, clustered CTCs were found to be more competent than individual circulating breast cancer cells at surviving and metastasizing [15]. To investigate the type of leukocytes in determining the fate of CTCs, the same group analyzed blood samples from 70 breast cancer patients and 5 breast cancer-bearing mouse models. They found that the majority (75% in human patients and 80.5–91.7% in mice) of leukocytes that escort circulating breast cancer cells are neutrophils. When neutrophils are depleted in breast cancer-bearing mice, the formation of lung metastasis is delayed. In addition, injection of CTCs isolated from the CTC-neutrophil clusters into the circulation gives rise to a considerably higher number of metastases to the lung than unclustered CTCs [16].

Why do CTCs that are associated with neutrophils metastasize more readily than those that are not? Several mechanisms account for the enhanced ability of CTCs clustered with neutrophils. First, in the neutrophil–CTC cluster, neutrophil-released cytokines such as IL-6 and IL-1β render CTCs more competent in cell cycle progression through a higher expression of proliferation-related genes such as Ki67 [16]. Second, granulocyte colony-stimulating factor (G-CSF) secreted by cancer cells stimulates the pro-tumor function of neutrophils in invasive breast cancer [17] (Fig. 1). Breast cancer patients, including TNBC patients, having CTC–neutrophil clusters show worse progression-free survival than those having unclustered CTCs. Circulating neutrophils endow CTCs with higher malignant potency [16].

Tumor-infiltrating neutrophils (TINs) are the neutrophils that interact with cancer cells in the tumor microenvironment (TME) [11, 18]. Wang et al. demonstrated that high levels of TINs are associated with advanced histologic grade and tumor stage as well as the TNBC subtype. Mechanistically, TINs induce migration, invasion, and epithelial-to-mesenchymal transition (EMT) of breast cancer cells through production of tissue inhibitor of metalloproteinase 1 (TIMP1) and subsequent induction of CD90 [19].

TINs can play quite different functions in the TME through a process called polarization. Polarization of neutrophils was first reported in 2009 by Fridlender and colleagues, who proposed the name N1 (anti-tumor) versus N2 (pro-tumor) to describe a process of morphological and functional reprogramming of TANs in a similar way to M1 versus M2 tumor-associated macrophages (TAMs) [20]. However, recent studies have revealed the complexity of neutrophil subpopulations and their functions in the TME, making it difficult to simply classify them as N1 or N2. For the purpose of generalization and clarity in this review, neutrophils are divided into the anti-tumor and pro-tumor types. Three tumor-derived cytokines, i.e., transforming growth factor-β (TGF-β), G-CSF, and interferon-β (IFN-β), are the most studied molecules involved in polarization of neutrophils. While TGF-β and G-CSF activate a tumor-promoting program of neutrophils, i.e., pro-tumor polarization [21], IFN-β promotes the reverse process, i.e., anti-tumor polarization [22] (Fig. 2). In TNBC, tumor cell-derived conditioned medium induces a polarized morphology and activation of neutrophils, mediated by cytokines such as TGF-β and the ligands for the chemokine receptor CXCR2, i.e., CXCL1/2/3 [23, 24]. In addition to these well-known cytokines, studies in other cancer types have shown that tumor can regulate neutrophil polarization through other factors. For example, IFN-γ combined with GM-CSF also drives neutrophils toward an anti-tumor state [25]. Whether these factors also contribute to neutrophil polarization in TNBC needs to be validated.

The anti-tumor TANs have higher levels of the C–C motif chemokine CCL3, tumor necrosis factor-α (TNF-α), and intercellular adhesion molecule 1 (ICAM1). The anti-tumor TANs also have higher levels of reactive oxygen species (ROS) such as hydrogen peroxide (H₂O₂) and superoxide anion (O₂‧) as well as reactive nitrogen species (RNS) such as nitric oxide (NO). In contrast, the pro-tumor TANs have higher levels of Arginase 1 (Arg1), CD206, Ym1, IL-6, CCL17, CXCR4, neutrophil elastase (NE), matrix metalloproteinase 9 (MMP9), vascular endothelial growth factor (VEGF), and the neuropeptide prokineticin 2 (PROK2, also known as Bv8) [4, 20, 26, 27]. Although neutrophils in their pro-tumor form also produce ROS, the level of ROS is relatively low (Fig. 2). It should be noted that the separation of the neutrophil functions as being anti-tumor or pro-tumor is probably oversimplified. Moreover, it is assumed that the differential roles of neutrophils may depend on the different stages of cancer development, with an anti-tumor role at a relatively early stage and a pro-tumor role at late stages of carcinogenesis [1]. However, this assumption needs to be further validated using multiple experimental systems.

The neutrophil extracellular trap (NET) is a peculiar web-like structure containing DNA strands in a complex with histones and neutrophil-derived enzymes such as NE, MMP9, myeloperoxidase (MPO), and peptidylarginine deiminase 4 (PAD4) [28, 29]. Neutrophils contribute to metastasis of a variety of tumors through the formation of NET. In mouse metastatic seeding models, formation of NETs has been found to be an important function of neutrophils that promotes breast cancer liver [30] and lung [31] metastasis. Notably, the number of NETs varies across different breast cancer subtypes, with the highest number observed in TNBC [31].

In the circulation, NETs can be formed by two cancer cell-related mechanisms. First, increased mobilization of neutrophils from the bone marrow, stimulated by tumor cell-secreted G-CSF, results in an increased count of circulating neutrophils called neutrophilia. Second, the factors carried in tumor cell-derived extracellular vesicles (EVs) stimulate the formation of NETs in the circulation (Fig. 3A). In turn, circulating NETs favor cancer progression through (1) stimulating the formation of cancer-associated thrombosis; (2) promoting endothelial damage and organ dysfunction; (3) facilitating cancer cell survival; and (4) promoting cancer cell metastasis [32, 33].

In the TME, NETs can awaken breast cancer cells by helping them exit the dormancy program (Fig. 3B). In doing this, the proteolytic enzymes NE and MMP9 contained in the NET cleave laminin in the extracellular matrix, resulting in the exposure of a cryptic epitope within laminin. In turn, dormant cancer cells trapped by NETs can sense the newly formed laminin epitopes through an α3β1 integrin-mediated mechanism, leading to activation of the proliferative program and, thus, awakening dormant breast cancer cells [34].

Since the first proposal by Kaplan et al. [35], the concept of the “pre-metastatic niche” has been well accepted and demonstrated in several types of cancer, including TNBC. A pre-metastatic niche can be defined as a supportive and permissive microenvironment within a target organ undergoing a series of cancer cell-induced molecular and cellular changes favoring colonization and growth of cancer cells before their arrival [36, 37]. The characteristic features of the pre-metastatic niche involved in TNBC metastasis include: (1) inflammation as manifested by recruitment of bone marrow-derived cells, including neutrophils; (2) immunosuppression; (3) increased vascular permeability and angiogenesis; and (4) extracellular matrix remodeling [37‐39].

In TNBC, tumor cells could regulate neutrophils by releasing EVs to facilitate pre-niche formation. miR-200b-3p derived from TNBC exosomes induces the upregulated expression of CCL2 by alveolar epithelial type II cells, and these chemokine ligands recruit neutrophils and promote pre-metastatic niche formation in the lung [40]. Lin28B, a conserved RNA-binding protein, is particularly upregulated in the TNBC subtype. In mouse models of TNBC, tumor-derived EVs containing the pluripotent factor Lin28B were found to upregulate the expression of CXCLs (CXCL1, CXCL2, CXCL3, and CXCL5) in the lung tissue, resulting in neutrophil infiltration. Lin28B increased IL-6 and IL-10 production, which could convert infiltrated neutrophils from tumor-suppressive to tumor-promoting. These pro-tumor neutrophils can inhibit T cell growth and activation and Th1 cell differentiation through various mechanisms. IL-10 can stimulate pro-tumor neutrophils to produce PD-L2, which inhibits T cells, while IL-6 can enhance this process. Additionally, pro-tumor neutrophils can inhibit IL-12, a crucial factor in Th1 cell differentiation [41].

In addition to EVs, tumor-derived secreted factors (TDSFs) play a critical role in the formation of the pre-metastatic niche. TDSFs include IL-6, IL-8, IL-1β, G-CSF, GM-CSF, and chemokines such as CCL2, CXCL12/SDF-1, and CXCL1 [42]. G-CSF derived from TNBC cells recruiting Ly6G⁺ Ly6C⁺ neutrophils initiate a pre-metastatic environment. These TDSF-primed granulocytes travel to the pre-metastatic lung to facilitate angiogenesis, colonization by cancer cells, and subsequent metastasis through activation of one of the Bv8 receptors, prokineticin receptor (PKR)-1 [43]. Downregulation of CTNND1, a member of the subfamily of armadillo (ARM) repeat-containing proteins, increases neutrophil infiltration in the bone and enhances cancer metastasis to the bone in TNBC. Because the knockdown of CTNND1 can upregulate CXCR4 through the PI3K/AKT/HIF-1α pathway, a large number of neutrophils are infiltrated into the bone where CXCL12, the ligand for CXCR4, is high. At the same time, knockdown of CTNND1 can make TNBC cells release a large amount of cytokines such as GM-CSF and IL-8 and promote neutrophils to produce an inhibitory effect on cytotoxic T lymphocytes, which could promote formation of the pre-metastatic niche in the bone [44] (Fig. 4). These factors act in an autocrine or paracrine fashion to render the secondary organ’s microenvironment favorable for colonization of metastatic cancer cells. Specifically, these factors function to (1) induce phenotypic changes in breast cancer cells; (2) recruit bone marrow-derived cells; and (3) form an inflammatory milieu in the microenvironment.

Chemicals such as nicotine can also promote the formation of the pre-metastatic niche through neutrophils in TNBC. Exposure of mice to nicotine leads to pre-metastatic niche formation in the lung by recruiting pro-tumor neutrophils, which induces a mesenchymal-to-epithelial transition (MET) process of breast cancer cells, thus promoting cancer cell colonization and metastasis [45]. Within the pre-metastatic niche, pro-tumor neutrophils inhibit the proliferation and activity of cytotoxic T lymphocytes, promote angiogenesis, and stimulate cancer cell survival, thereby promoting tumor invasion and metastasis. There is mounting evidence that although multiple types of immune cells are involved, neutrophils play a major role in generation of the pre-metastatic niche and, thus, contribute to the metastatic process of TNBC [30, 46].

Neutrophil polarization and NET secretion also contribute to pre-metastatic niche formation. TNBC patients expressing a high level of the nuclear protein high mobility group box 1 protein (HMGB1) are more likely to metastasize. Further investigations revealed that HMGB1 from the primary breast tumor could direct neutrophils to become a NET-producing immunosuppressive type. These data show that CD62L^dim neutrophils accelerate lung metastasis by promoting the development of NETs [47].

Through circulation, tumor-derived EVs and TDSFs reach the pre-metastatic site within the target organ. This results in establishment of a cancer-friendly milieu in the target organ prior to cancer cell arrival. In addition, iNOS, PD-L2, and NETs from pro-tumor neutrophils suppress T cells and promote pre-metastatic niche formation in breast cancer.

An increase in the number of neutrophils is traditionally regarded as an indicator of acute inflammation in pathological conditions. Increasing evidence has revealed the association between neutrophils and cancer prognosis. Patients of different types of cancer frequently show a remarkable increase in the number of circulating neutrophils, often manifested by an increase in the neutrophil-to-lymphocyte ratio (NLR). The NLR is a simple and easily obtainable marker that reflects systemic inflammation and is often used as a prognostic indicator in various types of cancer, including TNBC. Higher NLR values have been associated with poor prognosis, including shorter disease-free survival, overall survival, and higher risk of recurrence in patients with TNBC [48‐54].

Except for predicting survival outcomes, changes in NLR are associated with therapeutic response. NLR is associated with a response to neoadjuvant chemotherapy in a cohort consisting of 120 TNBC patients [55]. Furthermore, a high NLR predicts worse disease-free survival in TNBC patients who have failed neoadjuvant chemotherapy or non-metastatic TNBC patients [56‐60]. Interestingly, changes in NLR at different time points relative to therapy show different clinical associations, with changes before the start of radiotherapy (odds ratio: 1.115) and one year after surgery (odds ratio: 1.196) associated with increased risk of recurrence or death [61]. Similarly, a cohort study found a strong correlation between the dynamic changes in NLR and disease recurrence as well as the time of death in TNBC patients [62].

In addition to using NLR as a prognostic indicator, the absolute count of neutrophils and the level of neutrophil-derived factors are regarded as being able to predict prognosis in TNBC patients. In single-cell RNA sequencing datasets of TNBC (GSE118389 and GSE75688), higher infiltrating levels of neutrophils and M2 macrophages were reported to be the most efficient signature to predict poor survival compared to other immune cells in TNBC cases [63]. A recent study found that high levels of CXCR2, a chemokine receptor abundantly expressed by neutrophils in TNBC [64], were associated with a subgroup of TNBC patients characterized by better prognosis [65]. CXCL8, a neutrophil-attracting chemokine associated with brain metastasis in TNBC, can be used as a biomarker of poor prognosis of TNBC [66]. Despite these associations, the exact relationship between NLR and breast cancer prognosis is still under investigation and further studies are needed to establish the clinical utility of NLR as a reliable prognostic marker in breast cancer.

Given their currently known roles in TNBC development, particularly in the metastasizing cascade, neutrophils have been evaluated as a valid target for cancer therapy. In TNBC, approaches targeting neutrophils have shown great promise in slowing down or halting tumor progression in both preclinical (Table 1) and clinical (Table 2) studies [67].

Inhibition of chemokine-related pathways, such as CXCR2, leads to the reduction of neutrophil recruitment and, thus, reduces immunosuppression of neutrophils in the immune microenvironment and the circulation [68]. After treatment with CXCR2 inhibitor AZD5069, neutrophil infiltration in brain metastases of TNBC is significantly reduced and, thus, tumor metastasis is significantly inhibited [69].

Some key enzymes of neutrophils, such as Arg1 and Nos2, are directly or indirectly inhibited to reduce their tumor-promoting effects [70, 71]. Interestingly, histone deacetylases (HDACs) are essential enzymes that modify the functions of other enzymes that are important in neutrophils such as Arg1. Therefore, HDAC inhibitors have been reported to increase the efficacy of immune checkpoint inhibition by downregulation of Arg1 in pro-tumor neutrophils [72, 73]. There are several ongoing clinical trials combining HDAC inhibitors and immunotherapy in treating TNBC, and the outcomes from these trials will shed light on the feasibility of this approach.

Promotion of apoptosis of pro-tumor neutrophils or differentiation from pro-tumor into anti-tumor neutrophils to reduce their absolute number weakens the inhibitory effect of neutrophils on other immune cells, thus increasing the sensitivity of breast cancer cells to immunotherapy [74]. In 4T1 tumor bearing mouse models, targeting persistent pro-tumor neutrophils with gemcitabine after primary tumor resection decreases metastatic growth in the lungs [75].

Targeting NETs in breast cancer has also been examined in both preclinical and clinical studies [76‐78]. In a mouse model of TNBC metastasis, elimination of the DNA component of the NET using DNase I leads to reduction in the capacity of cancer cells to metastasize [32]. Furthermore, treatment with DNase I or an NE inhibitor (GW311616) significantly decreases the formation of metastases in the liver [30] or in the lung [47] following injection of 4T1 TNBC cells in mice. In the clinical setting, two FDA-approved drugs, aspirin and hydroxychloroquine, have been repurposed as anti-NET treatments. Aspirin, a non-steroidal anti-inflammatory drug, has been shown to halt NET formation and, consequently, patients taking aspirin daily have significantly reduced mortality and decreased risk of distant metastasis of breast cancer [79, 80]. Hydroxychloroquine, an autophagy inhibitor approved to treat many types of cancer [81], has been shown to inhibit the formation of NETs, thus potentiating the responsiveness of TNBC patients to clinical therapy [82].

Apart from these relatively classic targeted treatment methods, a more recent study has reported a new treatment method that targets neutrophils. Linde et al. demonstrate that neutrophils can be harnessed through the combined action of TNF, CD40 agonists, and tumor-binding antibodies to induce tumor eradication and reduce metastasis. It is interesting to note that complements, which are also an important part of the innate immune system, play a key role in this novel mechanism. Complement component C5a activates neutrophils to produce leukotriene B4, further leading to the production of ROS via xanthine oxidase, resulting in oxidative damage. Notably, this mechanism is effective in eliminating multiple tumor types, including TNBC, without requiring the involvement of T cells [83].