The role of tumor-associated macrophages in cancer development and their significance as prognostic markers

Abstract

Macrophages are a diverse group of immune cells which have key roles in immune defense, tumor homeostasis, and wound repair. During the last two decades, the role of macrophages as one of the most abundant tumor-infiltrating stromal cells has gradually emerged. The normal function of these tumor-associated macrophages (TAMs) in tumor microenvironment (TME) is to suppress tumor cells through triggering both direct cell cytotoxicity and antibody-mediated immune response. However, they have also been implicated in the progression of cancers. Tumor cells produce chemokines that polarize macrophages into tumor-promoting TAMs. This is the reason why the accumulation of TAMs in TME is correlated with poor prognosis in cancer patients. High plasticity of TAMs makes it feasible to regulate their polarization and adjust the balance between the anti-tumor TAMs and those with pro-tumor phenotypes. In this review, we aim to provide an overview about the origin and polarization of TAMs and their significance as biomarkers for the prediction and prognostication of various cancers.

Introduction

During the growth of tumors, cancer cells engage with the extracellular environment, forming a specialized environment, referred to as tumor microenvironment (TME). This TME provides a supportive condition for tumor cells, thereby promoting tumor development and metastasis [1]. In fact, TME is a complex and interactive network which is made up of a variety of cells and components, including the extracellular matrix (ECM), tumor-associated macrophages (TAMs), lymphocytes, dendritic cells, growth factors, chemokines, cytokines, natural killer cells, and myeloid-derived suppressor cells. Among these elements, TAMs are particularly significant, making up approximately 50% of the tumor’s mass [2].

Macrophages, a diverse group of immune cells known for their ability to phagocytose, have distinct phenotypes depending on the stimuli they encounter within their environment. These cells are key players in tissue homeostasis, immune defense, and wound repair, but are also implicated in the progression of various diseases, including cancer [3, 4]. In such pathological conditions, macrophages can polarize into two primary subtypes, M1 and M2, which have different functions and characteristics; the M1 macrophages are involved in pro-inflammatory responses and have antitumor activity, while the M2 ones contribute to tissue repair and also can promote tumor progression [5].

Tumor cells release chemokines and growth factors that attract macrophages and polarize them into the tumor-promoting M2 TAMs. These type of macrophages are essential for supporting tumor growth, invasion, and metastasis, and their presence within tumors is correlated with poor patient prognosis and resistance to standard therapies [6, 7]. Moreover, changes in the balance between M1 and M2 have been associated with immunotherapy outcomes, highlighting the growing focus on TAM modulations for therapeutic purposes [8–10]. One promising approach involves adjusting the balance between M1 and M2 macrophages to enhance antitumor immunity. Reprogramming TAMs into anti-tumor macrophages offers more precise TAM-specific targeting and may help to target human cancers. Inhibiting proteins like Signal regulatory protein α (Sirp1a), cluster of differentiation 40 (CD40), Toll-like receptor 7 (TLR7), and TLR9 reprograms macrophages, and antibodies or small molecules targeting these proteins are in early clinical trials [11]. Fortunately, due to the high plasticity of TAMs, it is feasible to regulate their polarization and change M1/M2 ratio with hope to diminish the pro-tumor characteristics of TME [12]. In our previous study, we reviewed the detailed mechanisms underlying the polarization of macrophages into TAMs and various strategies for modulation of this polarization [13]. Preclinical studies have demonstrated that targeting all types of macrophages in the TME, e.g. using colony stimulating factor 1 receptor (CSF-1R) inhibitors, may have significant side effects due to the broad actions these drugs may have [11]. Therefore, the exact balance between M1 and M2 TAMs is highly demanded. This progress will of course be achieved when a detailed understanding of TAM polarization is obtained. This review aims to provide a comprehensive overview of the origins of TAMs within the TME, their polarization into M1 and M2 subtypes, and the underlying mechanisms by which TAMs promote cancer metastasis. Additionally, we will explore the role of TAMs as biomarkers for various cancers, emphasizing their prognostic significance.

Origin and the heterogeneity of TAMs

TAMs have diverse origins and functions, distinguishing them from other myeloid cell types found in tumors [5]. They originate either from tissue-resident macrophages or from monocyte-derived macrophages recruited to different tumor sites [14–16] According to the studies on mouse models, it was previously believed that TAMs primarily arise from bone marrow-derived monocyte precursors that repopulate the TME [14, 15]. Current research supports that most macrophages originate from circulating monocytes in the peripheral blood. During early embryonic development, monocytes are recruited from the bone marrow into the bloodstream, from where they migrate to different tissues and organs, differentiating into tissue-specific macrophages (Fig. 1A) [17]. In tumors where TAMs mostly arise from monocytes, C-C motif chemokine ligand 2 (CCL2), a chemokine produced by tumor cells, is the primary recruitment factor acting through C-C chemokine receptor type 2 (CCR2) on classical monocytes. However, other chemoattractants may also play redundant roles in recruiting monocytes [14]. In contrast, some tissue-resident macrophages, such as alveolar macrophages in the lungs, microglia in the brain, and Kupffer cells in the liver, are not derived from blood monocytes (Fig. 1A). The origin as well as the mechanisms behind the self-renewal, proliferation, and replacement of these macrophages are not yet fully understood [18]. Recent research has demonstrated the coexistence of proliferating tissue-resident macrophages and blood monocyte-derived macrophages in organs such as the lungs, spleen, and brain, confirming their distinct phenotypes and functions [17]. At the early stages of tumor development, tissue-resident macrophages accumulate around the tumor to promote immune evasion, induce regulatory T-cell responses, and support epithelial-mesenchymal transition (EMT) and tumor invasion [19]. Meanwhile, monocyte-derived macrophages are attracted to the TME by cytokines and chemotactic proteins produced by circulating hematopoietic stem cells. Then, various stimulators trigger the differentiation of these monocyte-derived macrophages into TAMs. Factors such as interleukin 1β (IL-1β), vascular-endothelial growth factor (VEGF), CCL2, and Stromal cell-derived factors 1α (SDF-1α), which are produced within tumors, recruit pro-angiogenic macrophages to the tumor sites. While the exact timing of monocyte conversion into TAMs remains unclear, strong evidence suggests that tissue-specific signals trigger changes in the transcriptional profiles of these recruited monocytes [12].

Fig. 1.

The origin and polarization of macrophages. (A) According to current knowledge, most macrophages in adulthood originate from circulating monocytes in the peripheral blood. However, some tissue-resident macrophages, such as alveolar macrophages in the lungs, microglia in the brain, and Kupffer cells in the liver, are not derived from blood monocytes. Coexistence of such proliferating tissue-resident macrophages with blood monocyte-derived macrophages in such organs may reflect their distinct phenotypes and functions. (B) Polarization of M0 macrophages into M1 and M2 macrophages. Influenced by various extracellular signals from the surrounding environment, macrophages can polarize into two main subtypes: the pro-inflammatory M1 macrophages and the anti-inflammatory M2 macrophages. M2 macrophages are often considered pro-tumor macrophages. M2 macrophages can further classified into subtypes (M2a, M2b, M2c, and M2d), each activated by distinct cytokine and growth factors and has specific role in the microenvironment (for detailed description, see Sect. 3 in the text)

Multiple studies now indicate that the developmental origin of TAMs, whether embryonic-derived tissue-resident macrophages or monocyte-derived macrophages, can impose durable functional and transcriptional differences within the same tumor type. However, surprisingly few studies have directly examined the behavior of resident macrophages during cancer progression. One early report documented a progressive loss of tissue-resident macrophages accompanied by an increase in monocyte-derived TAMs in a breast cancer model, highlighting dynamic remodeling of the macrophage compartment during tumor formation and progression [20].

Whether macrophage origin intrinsically determines function remains unclear. Several transcriptional studies conducted under homeostatic conditions suggest that macrophages of distinct ontogeny exhibit largely overlapping gene expression profiles within the same tissue, and that recruited monocytes can adopt resident-like transcriptional programs after infiltration within the tissue. These findings argue that the local tissue environment, rather than developmental origin, is the dominant driver of macrophage identity. However, accumulating evidence indicates that this apparent convergence is highly context-dependent and can be disrupted in disease settings. In pancreatic ductal adenocarcinoma (PDAC), Zhu et al. used genetic fate-mapping to demonstrate that a substantial fraction of TAMs arises from embryonic hematopoiesis and expand locally through self-renewal rather than monocyte recruitment. Importantly, these embryonically derived TAMs exhibited a pro-fibrotic, tumor-supportive transcriptional program, characterized by extracellular matrix production and stromal remodeling, and their selective depletion reduced tumor growth. In contrast, monocyte-derived TAMs within the same tumors were enriched for antigen-presentation and immune regulatory pathways, indicating distinct functional specializations linked to macrophage developmental origin despite a shared tumor microenvironment [21].

Ontogeny-dependent TAM specialization has also been observed across additional tumor contexts, supporting its broader relevance while underscoring tissue-specific differences. In brain tumors, Bowman et al. demonstrated spatial and molecular segregation between embryonically derived microglia, which predominantly localize to tumor margins, and bone marrow–derived macrophages (BMDMs), which preferentially infiltrate the tumor core. These populations exhibited distinct transcriptional states, with recruited BMDMs enriched for inflammatory and immunosuppressive programs and distinguishable by lineage-associated markers such as Itga4 (CD49d) [22]. By extending these findings beyond the central nervous system, Loyher et al. showed in lung cancer models that embryonic and monocyte-derived macrophages coexist but differ in expansion dynamics, localization, and gene expression profiles, emphasizing how tissue context modulates ontogeny-driven TAM functions [23].

Recent advances in single-cell RNA sequencing, lineage tracing, and spatial profiling have further refined our understanding of TAM heterogeneity, resolving transcriptionally distinct macrophage states that align with inferred developmental origin and temporal recruitment [24]. In parallel with these technological advances, emerging conceptual frameworks propose refining TAM nomenclature by incorporating both ontogeny and timing, distinguishing macrophages that pre-exist in tissues prior to tumor formation (Pre-TAMs) from those recruited during tumor or metastatic progression, irrespective of developmental origin or spatial location [24]. These dimensions are not mutually exclusive; rather, integrating developmental origin, tissue context, and temporal dynamics provides a unifying framework to conceptualize the biology of TAM identity. Such an approach will be essential for systematically mapping TAM functions and for designing therapeutic strategies that selectively target macrophage populations at distinct stages of cancer development and progression.

It should be kept in mind that both recruited and resident macrophages can serve different functions depending on the cancer type. In pancreatic cancer, for example, resident macrophages promote tumor growth, while bone marrow-derived ones may have anti-tumoral functions. In humans, TAMs in different cancers like breast and endometrial show different transcriptional profiles from tissue-resident macrophages, suggesting tissue-specific activation. This highlights that TAMs should include all macrophages within the TME, including resident macrophages, not just bone marrow-derived ones [14].

Macrophage polarization

Macrophage polarization refers to the activation of macrophages at a given time and location, influenced by various extracellular signals from the surrounding environment e.g. microbes, damaged tissues, or tumor cells. This polarization is not a rigid and fixed mechanism, as macrophages are highly plastic and capable of integrating multiple signals. These signals may have effect on epigenetics [25].

Macrophages undergo specific differentiation in different tissue environments and are generally classified into two main polarization states: pro-inflammatory M1 macrophages and anti-inflammatory M2 macrophages [11, 26–29]. These subtypes reflect their roles in immune response, with M1 macrophages playing critical roles in innate host defense and killing tumor cells and M2 macrophages supporting tissue repair and immunosuppression (Fig. 1B) [30].

Classically activated macrophages, or M1 macrophages, are primarily activated by cytokines like interferon-γ (IFN-γ). These macrophages are involved in pro-inflammatory responses, producing cytokines such as IL-12, and IL-23, which are essential for helper T cell (Th1) immune responses [28]. In another words, Th1 cytokines such as IL-18 and IL-12 or activated TLRs promote macrophages to M1 polarization [30]. M1 macrophages are further induced by TLR ligands and Th1 cytokines, including interferon γ (IFN-γ), tumor necrosis factor α (TNF-α), and colony-stimulating factor 2 (CSF-2) [11]. These cells possess strong antigen-presenting capabilities. Moreover, due to their capacity to secrete reactive oxygen species (ROS), induced nitric oxide synthase (iNOS), and pro-inflammatory cytokines like IL-6, IL-1β, and TNF-α, M1 macrophages are highly effective at eliminating pathogens and tumor cells, making them integral to antitumor immunity [11, 31, 32].

While M1 macrophages contribute to pathogen clearance and tumor destruction, their potent pro-inflammatory activity can also lead to tissue damage. Nonetheless, their role in promoting immune responses makes them crucial players in cancer prognosis and immune defense [33].

TAMs, on the other hand, are characterized by their M2-like features within the TME, including elevated levels of anti-inflammatory cytokines, angiogenic factors, scavenger receptors, and proteases compared to M1 macrophages. TAMs play a crucial role in linking inflammation to cancer progression by promoting tumor cell proliferation, invasion, and metastasis. They also stimulate tumor angiogenesis and inhibit antitumor immune responses mediated by T cells, thereby facilitating tumor growth. TAMs play a pivotal role in cancer progression by promoting tumor cell proliferation, invasion, metastasis, and angiogenesis, while suppressing antitumor immune responses. This creates an immunosuppressive environment, driving tumor growth through TAM-derived factors. Importantly, TAMs are shaped by signals from their microenvironment rather than being inherently polarized by their location [26, 28].

M2 macrophages, also known as “repair” macrophages as they stimulate tissue repair, are polarized by anti-inflammatory cytokines such as IL-4, IL-10, IL-13, and transforming growth factor β (TGF-β) [30, 34]. Studies have demonstrated that macrophage polarization into the M2 phenotype can be triggered by factors such as CSF-1, IL-4, IL-13, IL-10, and parasite infections, as well as other types of stimuli [33]. M2 macrophages are involved in Th2 immune response including humoral immunity, wound healing, immune tolerance, tissue remodeling, and debris scavenging [30, 34]. In the context of cancer, M2 macrophages support angiogenesis by secreting adrenomedullin and VEGFs, and expressing immunosuppressive molecules like IL-10, PD-L1, and TGF-β, thereby promoting tumor growth [34]. Hence, M2 macrophages are often considered pro-tumor or “bad” macrophages. These macrophages are further classified into subtypes: M2a, M2b, M2c, and M2d (Fig. 1B). M2a is activated by IL-4 or IL-13 and associated with Th2 immune responses, while M2b is involved in immunoregulation through immune complexes and TLR ligands. M2c is polarized by IL-10, TGF-β, or glucocorticoid and contributes to tissue remodeling and immunosuppression, whereas M2d is activated by cytokine and growth factors in the TME, exhibiting pro-tumor and immunosuppressive characteristics [26, 35]. Additionally, the classical binary polarization model of macrophages has been expanded to a novel spectral polarization model, which recognizes the diverse roles of macrophage subtypes marked by specific surface markers such as CD169, TLR, MARCO, and IFN-γ. This model highlights the complexity of macrophage involvement in human pathologies [30].

Moreover, several cytokines and signaling pathways in the TME influence M1/M2 macrophage polarization. For instance, the JAK/STAT6 and PI3K/Akt pathways, activated by signals such as IL-4, IL-10, TGF-β, and bone morphogenetic protein 7 (BMP-7), promote M2 macrophage polarization [2].

It is noteworthy to mention that the polarization of macrophages is more complex than the simple M1/M2 classification, which only represents the extremes of macrophage functions. In reality, macrophages exhibit a high plasticity and exist along a spectrum of activation states, with many subsets displaying mixed characteristics. Several subtypes, such as CD169 + and TCR + macrophages, have been identified, and numerous others, especially TAMs, remain to be fully characterized [11, 33].

The polarization of macrophages toward the M2 phenotype is not governed by a single linear pathway but is instead orchestrated by a network of signals that operate in a hierarchical and temporal manner. Distinguishing between pathways that initiate the M2 program and those that maintain and amplify it is crucial for understanding TAM biology and developing targeted interventions [36].

• Initiating signals: the STAT6-centric core. The JAK/STAT6 signaling pathway serves as the primary transcriptional initiator of canonical M2 polarization. Upon binding of key cytokines, such as IL-4 or IL-13 to their receptors, JAK kinases are activated, leading to the phosphorylation and dimerization of STAT6. Then STAT6 translocates to the nucleus wherein it directly upregulates the expression of hallmark M2 genes, including Arg1, Mrc1 (CD206), and Retnla (Fizz1) [37]. This pathway is responsible for establishing the foundational M2 transcriptional profile. Recent single-cell RNA sequencing studies across breast, lung, and colorectal cancers consistently identify a STAT6-driven gene signature as a core feature of pro-tumor TAM subsets [37], confirming its role as a conserved initiating signal.

• Maintenance and amplification signals. Following initiation of polarization, other pathways are engaged to sustain the M2 phenotype, integrate pro-tumorigenic functions, and lock in the immunosuppressive state.

  • The PI3K/Akt pathway is a critical maintenance and amplifier node. It can be activated downstream of various TME signals, including CSF-1 and immune complexes. PI3K/Akt signaling reinforces M2 polarization by stabilizing HIF-1α under hypoxic conditions and promoting metabolic reprogramming towards oxidative phosphorylation, which supports long-term TAM survival and function within the TME [38]. Its role in enhancing M2 characteristics, for example by stimulating VEGF production for angiogenesis is experimentally validated in gastric, ovarian, and breast cancer models [38].
  • The TGF-β/NF-κB axis often functions as a contextual maintenance pathway. In the later stages of tumor progression, TGF-β is secreted by tumor cells and stromal elements and can suppress M1-related genes while promoting an immunosuppressive TAM phenotype [39]. Similarly, non-canonical NF-κB signaling can be activated to sustain expression of pro-survival and pro-invasive factors. These pathways are less about initiating the M2 state and more about adapting it to support advanced tumor processes like metastasis and immune evasion.

• Pathway cross-talk and experimental validation. These pathways do not operate in isolation. For instance, STAT6 can synergize with PI3K/Akt to maximize M2 gene expression, while TGF-β signaling can modulate both. In terms of experimental validation across cancer types, the STAT6 and PI3K/Akt pathways demonstrate the strongest and most consistent evidence [40]. Genetic knockout or pharmacological inhibition of these pathways in preclinical models of melanoma, breast, and pancreatic cancer robustly reduces M2 marker expression, impairs tumor growth, and improves anti-tumor immunity [40]. While the TGF-β pathway is ubiquitously active in cancer, its specific and essential role in initiating TAM polarization is less definitive than its broader role in TME maintenance.

The role of TAMs in promoting cancer metastasis

Metastasis occurs when the cancer cells spread from their primary origin to surrounding and/or distant organs; this process accounts for 90% of cancer-related mortality [41–43]. In this section, we first review the general mechanisms by which TAMs can promote cancer cell invasion and metastasis, and then we will specifically discuss the role of TAMs in metastasis of some human cancers.

An accumulating body of evidences suggests that TAMs are involved in almost every step of metastasis, including the detachment of tumor cells from their original tissue, preparing pre-metastatic niches, intravasation, survival of tumor cells within the circulation, as well as extravasation [41, 42, 44]. TAMs can degrade the matrix of endothelial cells through secreting matrix metalloproteinases (MMPs), cathepsins, and serine proteases, and disintegrate collagen and other extracellular matrix components, thereby promoting the cancer cell migration and metastasis (Fig. 2) [26, 42, 44, 45]. For example, it has been found that CD11b-CD18, integrins from M2 exosomes, accelerate cancer cell invasiveness and metastasis by increasing MMP-9 expression [43].

Fig. 2.

The role of TAMs in metastasis promotion in gastric cancer, breast cancer and ovarian cancer. (A) Studies show that the number of TAMs in TME of gastric cancer is highly increased. TAMs induce EMT in gastric cancer cells through PI3K/AKT/Snail/MMP-9 and P38 MAPK signaling pathways. In addition, TGF-β2, secreted by TAMs, stimulates gastric cancer progression and metastasis via the TGF-β2/NF-κB/Kindlin-2 axis by promoting Kindlin-2 expression. Polarization of M2 macrophages is stimulated through gastric cancer associated mesenchymal stem cells (GC-MSCs) by JAK/STAT pathway via the secretion of IL-6 and IL-8. TNF-α is secreted by gastric cancer cells and induces the release of CXCL1 and CXCL5 from macrophages, which can promote tumor migration through CXCR2/STAT3 in the tumor cells. CXCL1 and CXCL5 stimulate CXCR2 to induce EMT and metastasis in gastric cancer. (B) In breast cancer, TAMs induce the expression of MMPs, ​​cathepsins, and serine proteases in ECM, thereby allowing its disruption. In addition, the secretion of SPARC, CCL18 and EGF by TAMs mediates the attachment of tumor cells to fibronectin, increased tumor infiltration by administrative T cells, and destabilization of ECM by initiating E2F3 signaling in TAMs. The recruitment of inflammatory monocytes to metastatic sites is stimulated by the CCL2-CCR2 signaling pathway; the resulting monocytes are referred to as metastasis-associated macrophages (MAM). MAM-derived VEGF-A stimulates extravasation and dissemination of tumor cells. CCL2-CCR2 signaling can also activate CCL3-CCR1 signaling in MAMs to support their accumulation at the site of metastasis. (C) TAMs are capable to stimulate the dissemination of ovarian cancer cells by secretion of diverse soluble factors such as TNF-α, EGF, CCL18, IL-6 and TGF-β. TAMs can also trigger the invasive properties of ovarian cancer cells through the stimulation of JNK and NF-κB, which results in the downregulation of epithelial marker E-cadherin and upregulation of mesenchymal markers N-cadherin and vimentin. M2 macrophages secrete EGF, which promotes the cancerous phenotype and stimulates its receptor (EGFR) expression on cancer cells, thereby amplifying VEGF/VEGFR signaling. This amplification results in tumor cell proliferation, spheroid formation and, consequently, cancer dissemination. ASK1, a constituent of MAPK family, is critical for macrophage activation in ovarian cancer, and its deficiency or overexpression of its inhibitor (SOCS1) is linked to diminished spheroid formation and reduced ovarian cancer metastasis. (D) TAMs as potential targets for cancer prognosis. Interactions with strong experimental validations (consistent genetic/pharmacologic evidence across multiple cancer types) are shown with black lines, while those with lower experimental validation (clearly involved but often part of broader TME regulation) are shown with blue lines Strong. Consistent genetic/pharmacologic evidence across multiple cancer types Clearly involved but often part of broader TME regulation

Various studies demonstrated that TAMs are involved in the regulation of the EMT by producing or activating different factors such as nuclear factor kappa B (NF-κB) and inflammatory cytokines and growth factors, like IL-6, IL-8, TGF-β1, and TNF-α [46, 47]. EMT is a process during which epithelial cells acquire mesenchymal characteristics and induce invasion and metastasis [26, 42, 47]. Also, TAMs were shown to secrete EGF to activate EGFR/ERK1/2 signaling in cancer cells, thereby promoting EMT [47]. Moreover, apolipoprotein E (ApoE)-enriched TAM-secreted exosomes were found to mediate activation of the PI3K-Akt pathway in gastric cancer cells, facilitating EMT and tumor cell metastasis [48]. In addition, the differentiation of TAMs is strongly regulated by cytokines produced by tumor cells, leading to the formation of a positive feedback loop between TAMs and EMT [26]. Wei et al. indicated that in patients with colorectal cancer (CRC), TAMs disrupt the regulation of JAK2/STAT3/miR-506-3p/FoxQ1 and induce EMT, resulting in enhancement of cellular invasion and metastasis. Furthermore, activation of this axis could lead to the formation of CCL2 thereby facilitating macrophage recruitment [49]. It was shown that macrophage-derived IL-8 could promote EMT and enhance the migratory potential of hepatocellular carcinoma (HCC) cells [42]. Similarly, the presence of TAMs increased mesenchymal markers and decreased epithelial markers in pancreatic ductal adenocarcinoma cells [42]. TGF-β1 derived from M2 macrophages promoted EMT in CRC and HCC cells. The work of Chen et al. showed that M2 macrophages promote tumor metastasis by secreting chitinase 3-like protein 1 (CHI3L1), which interacts with the IL-13 receptor α2 (IL13Rα2) on tumor cells, leading to MMP upregulation [42].

NF-κB consists of a family of transcription factors that play pivotal roles in acquiring metastatic potential in cancer cells. The crosstalk between TAMs and cancer cells can lead to NF-κB activation [42, 46]. For example, TAMs stimulate NF-κB activation in both stromal and cancer cells by secretion of TNF-α, which upregulates the expression of Snail, thereby causing migration, invasion, and metastasis of breast cancer cells. Snail can also activate the expression of IL-8, which plays a vital role in monocyte recruitment, by binding to its E3/E4 E-box located in the promoter of IL-8. There is also a connection between IL-8 secreted by TAMs and EMT. For example, in HCC cells, EMT is induced by TAMs through IL-8-mediated activation of JAK2/STAT3/Snail signaling pathway [46]. A work by Han et al. showed that TAMs facilitate the invasive potential of osteosarcoma cells by increasing the expression of MMP-9, COX-2, and phosphorylated STAT3, leading to the induction of EMT [47].

TGF-β1 is also an essential inflammatory cytokine produced by infiltrating immune cells such as TAMs and play a crucial role in inducing EMT, facilitating the escape of tumor cells from the immune surveillance, and enhancing the dissemination and metastasis of cancer cells [46]. Moreover, Kominsky et al. demonstrated that TGF-β promotes renal cell carcinoma (RCC) bone metastasis [50], resulting in tumor-promoting paracrine interactions between tumor cells and the bone microenvironment [51].

In addition to contributing to the early EMT of tumor cells, TAMs contribute to the preparation of pre-metastatic niches (PMNs), the distant sites that support metastatic growth [42, 52]. Macrophages are recruited into PMNs by various tumor-secreted factors, including CCL2, CSF-1, VEGF, PlGF, TNF-α, TGF-β, exosomes, and tissue inhibitors of metallopeptidase (TIMP)−1. The presence of recruited and resident macrophages provides a roadmap for the homing of disseminated cancer cells into the PMNs, leading to sustained metastatic growth [42]. The myeloid chemoattractants S100A8 and A9, whose synthesis is induced by the primary tumor, are also among the tumor-produced factors required for PMN. S100 proteins induce the amyloid protein A synthesis, which signals via TLR4 in myeloid and endothelial cells. Moreover, lysyl oxidase crosslinks the collagen in PMN and is essential for recruiting myeloid cells that secrete MMP-9, which releases matrix-bound VEGF, whose function is required to increase metastatic efficiency [53].

Recently, CSF-1 and CCL2 were shown to play essential roles in monocytes/macrophages’ recruitment to metastasis sites [42, 45]. Using the CCL2-CCR2 pathway and engagement of a chemokine cascade that involves CCR1–CCL3 autocrine signaling, metastatic cells recruit monocytes and differentiate them into metastasis-associated macrophages (MAMs), which support the survival and metastasis of tumor cells by suppressing T cells [4, 43]. The inhibition of CCR2 or VEGFR has been shown to reduce the metastasis and extravasation of tumor cells by decreasing the infiltration of MAMs into the target organs. A study performed on non-neoplastic MCF10A human breast cancer cells showed that CCL2 secreted by TAMs can induce the expression of endoplasmic reticulum oxidoreductase (ERO)1-α and MMP-9, thereby leading to EMT in this cell line [54]. In addition, a positive paracrine feedback loop was found between TAMs-derived EGF and tumor cell-derived CSF-1, in which CSF-1 stimulates TAMs motility and EGF expression by TAMs. This crosstalk signals the tumor cells to migrate to the blood vessels [42, 45]. EGF is a pro-invasive factor secreted by M2 TAMs, which induces EMT in cancer cells by activating the EGFR-ERK pathway [55].

TAMs also abundantly secrete CCL18 which triggers the accumulation of integrin around human breast tumor cells and allows their attachment to ECM via interactions with the phosphatidylinositol transfer protein, the membrane-associated 3 receptor, resulting in increased intravasation of tumor cells and decreased patient survival [52]. Moreover, it has been shown that TAM-secreted CCL18 binds to a receptor on myofibroblasts, PIPTNM3, which further facilitates differentiation and the invasion of myofibroblasts. Various analyses showed that phosphatase of regenerating liver-3 (PRL-3) upregulated CCL26, leading to increased infiltration of TAMs as well as CRC invasion and metastasis [47]. Similarly, CCL5 secreted by TAMs can significantly enhance EMT and metastatic properties of prostate cancer cells by activating the β-catenin/STAT3 pathway [43, 47].

Macrophages are the primary producers of TNF-α and are also highly responsive to TNF-α, which induces MAPK activation in a c-Raf-1 and Raf-B-independent manner. Low doses of TNF-α, produced by RCC and its stromal cells, are involved in tumor growth and metastasis promotion [46, 51]. TNF-α can induce cancer invasion and metastasis associated with EMT by inhibiting E-cadherin, activating MMP-9, and upregulating vimentin. In RCC cells, the TNF-α-induced EMT can also be mediated by PI3K/Akt/GSK-3β signaling pathway [46].

TAMs not only help tumor cells to migrate away from the primary site but also involved in tumor extravasation and colonization in the target tissue [41]. Circulating tumor cells induce endothelial cells to express vascular cell adhesion molecule-1 (VCAM-1) and vascular adhesion protein-1 (VAP-1), which mediate the recruitment of macrophages and support metastatic cell survival [41, 46].

Gastric cancer

Gastric cancer is the fifth most common malignancy and is regarded as the fourth leading cause of cancer-related death worldwide [56, 57]. Given the inefficiency of the current therapies including surgery, radiotherapy, chemotherapy, discovering novel therapeutic targets for gastric cancer remains of great priority [58]. Regarding the advancements obtained for the treatment of gastric cancer, survival rate of patients still stands poor due to metastasis [59–61]. Distant metastasis can develop in the early stages of gastric cancer, making this malignancy difficult to manage [62].

Regarding the role of TAMs in the pathogenesis of gastric cancer, studies have shown that the number of TAMs in TME is highly increased [59]. However, the underlying mechanism of TAMs in gastric cancer has remained largely unknown. Increasing evidence proposes that macrophages are necessary for onset and development of gastric cancer [63]. In patients with gastric cancer, the number of M2 macrophages in tumor stroma is correlated with the peritoneal metastasis, angiogenesis and immune response. The expression of certain genes promotes the recruitment of immune cells into the TME through inflammatory signals which are secreted by gastric cancer cells [33]. Cadherin 11 (CDH11), a member of integral membrane proteins, is associated with cell-cell adhesion. CDH11 plays a critical role in EMT and macrophage polarization in the TME, leading to the progression of gastric cancer and poor prognosis. Dysregulation of CDH11 can contribute to various pathologic processes including inflammation, cell invasion, EMT, and carcinogenesis. Liu et al. disclosed that TAMs induced EMT through PI3K/AKT/Snail/MMP-9 pathway (Fig. 2A) [56]. A forkhead box-containing transcription factor (FOXQ1), facilitates EMT and metastasis in multiple types of malignancies including gastric cancer. Gue et al. showed that co-culture of tumor cells with TAMs can induce EMT in tumor cells and enhance their invasive properties through FOXQ [59]. High mobility group A (HMGA) family of proteins, including HMGA1A, HMGA1B, and HMGA2, regulate gene expression through altering chromatin. HMGA1B/2 was shown to transcriptionally stimulate POU class 1 homeobox 1 (POU1F1) which known as Pit1 [60]. POU1F1 was upregulated in gastric cancer, and its expression was associated with poor prognosis. High expression of POU1F1 induced gastric cancer through modulating macrophage proliferation, migration, and polarization via C-X-C motif chemokine ligand 12 (CXCL12)/C-X-C chemokine receptor type 4 (CXCR4) axis [60]. Wang et al. indicated that TGF-β2 secreted by TAMs have a functionality in gastric cancer progression and metastasis via the TGF-β2/NF-κB/Kindlin-2 axis by promoting Kindlin-2 expression. Kindlin-2 is overexpressed in gastric cancer and is significantly correlated with gastric cancer invasion and metastasis (Fig. 2A) [64]. Li et al. reported that polarization of M2 macrophages was stimulated through gastric cancer associated mesenchymal stem cells (GC-MSCs) by JAK/STAT pathway via the secretion of IL-6 and IL-8. GC-MSCs significantly induced the metastasis of gastric cancer by facilitating the EMT process [61]. Zhou et al. proposed a crosstalk between tumor cells and macrophages to promote cancer progression and metastasis in the tumor microenvironment; their study suggested that CXCR2 can be considered as a promising therapeutic target for gastric cancer therapy. Macrophages can promote the tumor migration through CXCR2/STAT3 in gastric cancer cells. TNF-α is secreted by tumor cells and induces the release of CXCL1 and CXCL5 from macrophages. CXCL1 and CXCL5 stimulate CXCR2 to induce EMT and metastasis in gastric cancer [65]. Songa et al. discovered the efficiency of EBV-miR-BART11 on EMT and metastasis of gastric cancer cells via targeting FOXP1 which can influence the differentiation from monocytes to macrophages. Their study demonstrated that EBV-miR-BART11 promoted EMT and metastasis. They showed that co-culture of TAMs with tumor cells significantly induced EMT in gastric cancer. High expression of IL-6 were shown to be a marker of low survival, high recurrence and metastasis [66]. Another study showed that hypoxia-injured macrophages lead to the invasion of gastric cancer cells via secreting mediators like VEGF. Culture medium collected from hypoxia treatment macrophages was shown to remarkably induce the invasion of gastric cancer cells. Furthermore, VEGF was found to be increased in culture medium of macrophages, and inhibition of VEGF/VEGFR signaling significantly decreased the invasion of gastric cancer cells. These observations suggest that hypoxia increases VEGF production through macrophages, which regulates the PI3K-Akt and p38 MAP kinase signaling pathways and promotes and invasion of gastric cancer cells by VEGFR [67]. Cui et al. compared the expression of PTX3 between gastric cancer cells and normal cells and demonstrated the role of PTX3 in the EMT [68]. PTX3 was shown to have potentiality in suppressing the stemness of gastric cancer cells and promoting the stimulation of TAMs through reducing the expression of IL-4 and IL-10 by the JNK1/2 signaling pathway [68].

Breast cancer

Breast cancer is a major health problem due to its high frequency in women. The five-year survival rate of metastatic breast cancer is less than 30%, even with adjuvant chemotherapy [69] or hormone therapy [70]. Recent GLOBOCAN 2020 data produced by International Agency for Research on Cancer (IARC) from 185 countries reported 2.3 million new cases of breast cancer (11.7% of all cancers) and a mortality rate of 6.9% [69].

In breast cancer, TAMs control all metastatic pathways, including invasion, blood vessel intravasation, extravasation to distant targets and metastatic colonization [53, 71, 72]. Local invasion is highly dependent on the properties of ECM. TAMs induce the expression of MMPs, ​​cathepsins, and serine proteases in ECM, thereby allowing its disruption [73, 74]. Also, the secretion of secreted, acidic, and cysteine-rich protein (SPARC) [75], CCL18 [76], and EGF [77] by TAMs has pro-tumor effects. These components mediated tumor cell attachment to fibronectin [76], increased tumor infiltration by administrative T cells, and destabilizes ECM by initiating E2F3 signaling in TAMs (Fig. 2B) [78]. Blockade of these pathways decreased breast cancer cell invasion and metastasis both in vitro and in vivo [75–77].

A subset of TAMs, TIE2-expressing perivascular TAMs, express VEGF-A to promote vascular penetration [79]. Intravascular penetration and metastasis have been shown to be inhibited by TIE2 kinase or angiopoietin-2 (Ang2), a TIE2-blocking ligand, in a PyMT mammary tumor model [80, 81]. Macrophages promote late metastasis by inducing EMT in precancerous lesions [82]. Early recruitment of inflammatory monocytes to pre-metastatic cells is stimulated by the CCL2-CCR2 signaling pathway, and the resulting monocytes become metastasis-associated macrophages (MAM). MAM-derived VEGF-A stimulates extravasation and dissemination of tumor cells [83]. In addition, CCL2-CCR2 signaling activates CCL3-CCR1 signaling in MAMs to support their accumulation at the site of metastasis (Fig. 2B). This process promotes the efflux and metastasis of breast cancer cells in several mouse models of breast cancer metastasis [84].

Additionally, CCL2 stimulates TAM to produce IL-1β, resulting in neutrophil-mediated promotion of mammary tumor metastasis in mice [85]. These data indicate that one or more CCL2-CCR2 signaling pathways mediate breast cancer progression. In mouse models of breast cancer lung metastasis, metastatic cell growth after tumor cell seeding required continuous macrophage recruitment [83, 86] and could be reduced by conditional macrophage suppression [86]. The Ang2-TIE2 pathway contributed to post-seeding metastatic growth. Blocking these signaling pathways significantly reduced metastatic outgrowth in mouse models [81]. Furthermore, the MARCO scavenger recognition receptor, co-expressed with M2 markers on TAMs, was implicated in promoting breast cancer metastasis [87]. Importantly, preclinical studies have demonstrated that MARCO antibody treatment of mice with 4T1 mammary carcinoma cell line repolarizes M2 TAMs to M1, thereby inhibiting metastasis. In addition, it increased the formation of germinal centers and the ratio of CD4+/CD8 + T cells in draining lymph nodes, thereby improving tumor immunogenicity [87].

Granulocyte-macrophage colony-stimulating factor (GM-CSF) and the CCL18 feedback loop also contributes to macrophage-stimulated metastasis. Inhibition of GM-CSF or CCL18 with antibodies breaks the feedback loop and reduces metastasis [88, 89]. Taken together, these results indicate that multiple signaling pathways in TAMs are likely involved in tumor progression and metastasis. Thus, TAMs may be potential targets for the treatment of breast cancer.

Ovarian cancer

Ovarian cancer is the most prevalent gynecological cancer and the fifth leading cause of cancer-related mortality in females worldwide [90]. Despite the remarkable progress made in treatment, the survival rate of these patients remains unclear due to the ineffectiveness of screening methods at the early stage [65, 91]. The 5-year survival rates of patients with ovarian cancer for stage III and IV are lower than 30%. Therefore, it is important to identify novel markers for the early detection of ovarian cancer, as well as to discover potential therapeutic strategies for advanced-stage ovarian cancer [92]. An outstanding characteristic of ovarian cancer metastasis is that in most cases, tumor cells are capable of surviving and disseminating without any need to scaffold and vascular structures. This transcelomic metastasis is, therefore, mediated through the movement of ascitic fluid and doesn’t require tumor cells to enter the circulation [90]. Detached tumor cells can evade the immunity and anoikis and resist chemotherapeutic agents by forming spheroids [93, 94]. Peritoneal metastasis is promoted by a crosstalk between malignant cells and other stromal cells in the TME [95]. In epithelial ovarian cancer, TAMs account for approximately 50% of cells in the peritoneal metastatic lesions and studies have also shown that the development of ovarian cancer is associated with the accumulation of TAMs in the surrounding ascites [95]. Determination of the underlying mechanism by which TAMs are involved in the metastasis of ovarian cancer may help inhibit the spread of ovarian cancer cells into the peritoneum, omentum and vasculature [96]. TAMs are capable of promoting tumor cell dissemination through secreting diverse soluble factors such as TNF-α, epidermal growth factor (EGF), CCL18 and TGF-β [63, 97]. A cell culture study indicated that macrophages could amplify the invasive properties of ovarian cancer cells through the stimulation of JNK and NF-κB (Fig. 2C). This study also showed that co-culture of ovarian cancer cells with TAMs downregulated the expression of E-cadherin while upregulated the N-cadherin and vimentin expression [98]. E-cadherin is a constituent of group of calcium-dependent adhesion molecules involved in the formation of tight and adherent junctions. Down-regulation of E-cadherin has been determined in response to hypoxia and related with invasive and low survival rates in epithelial ovarian cancer patients [90]. It was also found that TWEAK-mediated macrophages could secrete exosomes, which lead to over-expression of miR-7 in ovarian cancer cells, thereby repressing the EGFR/AKT/ERK1/2 pathway and ovarian cancer metastasis (Fig. 2C) [99]. One study showed that numerous macrophages were found in the ovarian cancer patients. These macrophages exhibited an upregulated level of colony-stimulating factor 1 (CSF-1). CSF-1 regulates macrophage differentiation by binding to its receptor, CSF-1R, on macrophages and monocytes [47]. Yihan Bai disclosed that collagen triple helix repeat containing 1 (CTHRC1), a regulator of M2 polarization of macrophages, promoted the progression and metastasis of ovarian cancer via integrinβ3/FAK signaling pathway. Upregulation of CCL2, CSF-1 and VEGF-A were shown to be associated with macrophage buildup in ovarian cancer, as well as breast and lung cancers. Polarization of macrophages into M2 TAMs is stimulated by IL-4 and IL-13 through STAT6 signaling pathway [100]. Mingzhu Yin indicated a correlation between TAM-associated spheroids and ovarian cancer. TAMs were shown to induce spheroid formation during the metastasis in epithelial ovarian cancer. M2 macrophages secrete EGF, which promotes the occurrence of cancerous phenotype. Furthermore, EGF stimulates its receptor (EGFR) on cancer cells which amplifies VEGF/VEGFR signaling, thereby resulting in tumor cell proliferation, spheroid formation and consequently, cancer dissemination [101]. Apoptosis signal-regulating kinase 1 (ASK1), a constituent of MAPK family, is critical for macrophage activation. Mingzhu Yin indicated that deficiency of ASK1, which is pivotal in tumor growth of ovarian cancer, is associated with declining TAM-mediated spheroid formation. In addition, overexpression of suppressor of cytokine signaling-1 (SOCS1), an ASK1 inhibitory protein, reduces ovarian cancer growth (Fig. 2C). SOCS1 also play critical role in regulation of TNF-induced inflammation and ASK1 degradation in ovarian cancer. The stimulation of ASK1-JNK signaling results in the degradation of E-cadherin [93]. TAMs also stimulate the C-X-C motif chemokine ligand 16 (CXCL16)/C-X-C motif chemokine receptor 6 (CXCR6) signaling pathway, which regulates the metastasis of ovarian carcinoma. CXCR6 expression is associated with TNF-α, IL-6, and CXCL16 expression in ovarian cancer. Co-culture of SKOV3 cells with macrophages or TNF-α remarkably promoted CXCL16, CXCR6 and p65 expression and upregulated the phosphorylated PI3K and Akt, thereby inducing the migration and invasion of SKOV3 cells. The CXCR6/CXCL16 axis in macrophages was shown to promote the migration and invasion of ovarian carcinoma cells through stimulating the PI3K/Akt signaling pathway [65]. The expression of MMPs is highly regulated in metastasis of ovarian cancer, because MMPs play a great role in the regulation of the primary steps of mesothelial adhesion [90]. Xing et al. reported that co-culture of ovarian cancer cell line SKOV3 and TAMs result in upregulation of the MMPs and invasion of SKOV3 cells. TAMs promoted cancer progression through stimulating TLRs, NF-κB and MAP kinases. Their study indicated that TAMs induced the high expression level of MMP-2, MMP-9, MMP-10 expression via TLR signaling pathway. The study suggested that TAMs could promote the cancer cell invasion and progression [102]. IL-6 was shown to induce JAK/STAT signaling pathway which promotes tumor progression in ovarian cancer. IL-6 also promotes invasion of other tissues through cancer cells via prominent EMT. Upregulation of IL-6/IL-6R results in increased expression of VEGF which is capable to promote angiogenesis. IL-6 can also contribute to the invasion of tumoral cell lines through the secretion of MMP-9 which is a zinc-dependent metalloproteinase that facilities tumor invasion and metastasis. Therefore, IL-6 plays a critical role in the degradation of ECM, cancer progression and metastasis through JAK/STAT3 and EMT [103]. Ubiquitin protein ligase E3 component n-recognin 5 (UBR5) is a homologous to E6AP C-terminus (HECT) domain-containing ubiquitin ligase which is critical for embryonic development. UBR5 is significantly upregulated in cancers such as breast cancer and ovarian cancer. Mei Song1 showed that UBR5 promotes metastasis of ovarian cancer via CCL2/CSF-1. UBR5 can also contribute to sustained β-catenin signaling to promote the formation of spheroid [104]. Taken together, the accumulating studies show that M2 macrophages cause tumor progression and metastasis in ovarian tumors.

Clinical translation of TAM-targeted therapeutics

While TAMs present a promising therapeutic target, translating preclinical strategies into effective clinical treatments has proven challenging [105]. This gap underscores the profound complexity and adaptability of TAM biology within the TME [105]. Below is a critical analysis of key therapeutic approaches, their clinical performance, and the underlying limitations.

CSF-1/CSF-1R axis Inhibition

several agents, such as pexidartinib (CSF-1R inhibitor) have been designed to deplete immunosuppressive TAMs. In preclinical models, they have shown successful reduction of TAM infiltration and slowed tumor growth. However, clinical trials in solid tumors (e.g., tenosynovial giant cell tumor) revealed significant limitations [106]. While showing efficacy in this specific niche, objective responses in common carcinomas like breast or pancreatic cancer have been minimal [106]. Major adverse effects include elevated liver enzymes (transaminitis), fatigue, and periorbital edema. The limited success is attributed to compensatory pathways, such as the upregulation of granulocyte-macrophage colony-stimulating factor (GM-CSF), which can recruit alternative myeloid populations and sustain the immunosuppressive TME, thereby fostering therapeutic resistance [107].

CD40 agonists

Agonistic CD40 antibodies aim to “re-educate” TAMs toward an immunostimulatory, M1-like phenotype and enhance antigen presentation [105]. Early-phase trials demonstrated immune activation and some antitumor activity. Nonetheless, clinical application has been constrained by substantial toxicity [106]. Dose-limiting adverse events, including cytokine release syndrome (CRS), hepatic toxicity, and thromboembolic events, are frequently observed. These toxicities stem from systemic immune activation, indicating that achieving a localized therapeutic effect within the TME without triggering systemic inflammation remains a major hurdle [106].

CD47/SIRPα “don’t eat me” signal Blockade

This strategy aims to block CD47 signal on tumor cells, thereby enhancing macrophage-mediated phagocytosis. Despite compelling preclinical data, clinical efficacy as a monotherapy has been modest [108]. A critical limitation is on-target, off-tumor toxicity. CD47 is broadly expressed on healthy cells, particularly erythrocytes, leading to dose-dependent anemia and thrombocytopenia due to phagocytosis of blood components [108]. Furthermore, the immunosuppressive TME can upregulate other “don’t eat me” signals (e.g., PD-L1, MHC class I), creating redundant pathways of immune evasion that limit the potency of single-axis blockade [39].

Emerging strategies and combinatorial approaches

Recent efforts focus on overcoming the above-mentioned limitations [106]. Strategies include nanoparticle-mediated delivery to selectively repolarize TAMs in situ, bispecific antibodies, and combination therapies [107]. Notably, combining TAM-targeting agents (e.g., CSF-1R inhibitors) with immune checkpoint inhibitors (for example anti-PD-1/PD-L1) is being actively explored in clinical trials [107]. The rationale is to simultaneously remove immunosuppressive barriers (TAMs) and reactivate cytotoxic T-cells, potentially overcoming resistance to single-agent immunotherapy. While promising, these combinations also risk compounded toxicities, underscoring the need for precise patient stratification based on TAM density, polarization status, and other TME biomarkers [39]. Emerging insights into metabolic reprogramming of TAMs also offer novel avenues for therapeutic intervention [109].

In conclusion, the clinical journey of TAM-directed therapies underscores a fundamental lesson: the TME is a dynamic and adaptive system. Future success will depend on developing more selective agents, leveraging combination regimens, and using integrated biomarker strategies to identify patients most likely to benefit [105, 106].

TAMs as potential markers for cancer prognosis

As the relationship between TAMs and malignant tumors becomes better understood, TAMs are increasingly being considered as potential biomarker for the prognosis, or even diagnosis, of various cancers, including breast cancer, prostate cancer, pancreatic cancer, lung cancer and esophageal squamous cell carcinoma [110–114]. This is somewhat challenging because TAMs are highly heterogeneous, making it difficult to develop therapeutics that specifically target pro-tumor TAMs without affecting other TAM populations [115, 116].

Patients with a higher density of TAMs in their tumor stroma have shorter overall survival (OS) rates compared to those with a lower density of TAMs. For example, osteopontin (OPN) is an extracellular matrix protein that promotes cell-mediated immune responses, acts as a cytokine, and controls cell migration. It has been suggested that OPN skews macrophages to M2 phenotype. Co-expression of OPN and CD204 in TAMs resulted in higher disease progression and poor survival, suggesting the possible use of OPN and TAMs as biomarkers in gastric cancer [110].

Clusters of differentiation receptors CD163, CD204, or CD206 are abundantly expressed on the surface of TAMs and can be used as a screening method for multiple advanced cancers [110, 112]. One approach in macrophage detection is to evaluate the density, polarization, and micro-localization of TAMs by staining macrophage markers in biopsy specimens to determine phenotypes depending on the context of the TME. For example, to detect macrophage subpopulations and calculate the M1/M2 ratio, it is possible to use the double immunohistochemical staining of M1 macrophage markers CD68/HLA-DR, and M2 macrophage markers CD68/CD163. CD206, another M2 marker, can also be helpful in this case [111]. According to the published dataset of human cancers, CD36 is highly expressed in human tumor tissues, and its expression is positively correlated with the density of TAMs and fatty acid oxidation gene signatures. These findings suggest that CD36 might play a vital role in regulating lipid metabolism in peritumoral cells to promote tumor growth and metastasis and suppress the immune system. Therefore, CD36 can be considered as a potential diagnostic and prognostic biomarker [116]. Moreover, evaluation of other markers such as VEGF, TFL, HIFs, PCNA, and CCL18 in combination with the mentioned pan-macrophage markers has also been beneficial [111].

Another approach is to evaluate the pre-surgery and post-surgery levels of TAM response factors in the serum samples, including those enhancing the recruitment of macrophages and their M2 polarization, such as CCL2, CCL5, CXCL12, CSF-1, and VEGF, and those released by macrophages and stimulate tumor progress, like CCL18, EGF, VEGF, CSF-1, IL-8, uPA, and MMPs [111]. Simultaneous measurement of CA-125, MMP-7, CCL18, and CCL11 in serum has been shown to improve the diagnosis of epithelial ovarian cancer [117]. It has also been suggested that TAM-derived CCL8 increases monocyte infiltration at the tumor site, produces more pro-tumoral TAMs, suppresses the immune system microenvironment, and increases tumor cell malignancy, all of which are associated with shorter disease-free and recurrence-free survival (RFS). As a result, measurement of these factors may be used as diagnostic and prognostic indicators [118].

In addition, several clinically appropriate imaging reagents have been developed for in vivo TAM imaging in breast cancers, some of which can specifically target M2 TAMs to improve the understanding of temporal and pathophysiological changes of this cell population in the TME [111, 119]. Moreover, evaluation of CSF-1 receptor by immunohistochemistry in human breast tumors is associated with nodal involvement and reduced survival in node-negative patients. Since this receptor can be an indicator of macrophages and tumor cells producing CSF-1, its presence can help identify patients who may benefit from macrophage-targeted therapy [115]. Some of TAM markers, which are helpful for prognosis/diagnosis of cancer, are given in Table 1.

Table 1.

TAM-related biomarkers for cancer prognosis

TAM marker	Type of cancer	Detection method	Function	Reference
Tim-3	Gastric	Fellow cytometry	Involved in a negative regulatory pathway in T cell activation	[122]
MR	Gastric	IHC and qPCR	Mannose receptor and C-type lectin	[123]
KRS	Gastric	IHC	A lysyl-tRNA synthetase	[125]
OPN	Gastric	IHC	An extracellular matrix protein	[121]
Serum macrophage MIF	Gastric	IHC	An important regulator of innate immunity	[129]
CD204	Gastric	IHC	Scavenger receptor class A [SR-A]	[125]
TP	Gastric	DNA sequencing	An angiogenic factor	[120]
CD11	Gastric	IHC	An important factor in antitumor immune response	[126]
CD163	Gastric	IHC	Monocyte scavenger receptor	[127]
Macrophage M1 infiltration	Gastric	IHC	Anti-tumor activity	[130]
Macrophage M2 infiltration	Gastric	IHC	Tumor promotion	[130]
TAM density in solid tumor	Gastric			[130]
NF-κB)p65(	Gastric	IHC	Ubiquitous transcription factor.	[125]
CCL2/CCR2	Gastric	IHC	Tumor-derived chemotactic factor	[125]
CD68	Breast, colorectal, lung, ovarian, prostate	IHC, Flow cytometry	Transmembrane glycoprotein	[152]
CD80	Colorectal, lung	IHC	Immunoglobulin superfamily	[171]
CD163	Breast, colorectal, lung	IHC, IF	A scavenger receptor	[172]
CD204 (MSR1)	Breast, colorectal, lung, prostate	IHC	Macrophage scavenger receptor	[173, 174]
CD206	Breast, colorectal, ovarian, prostate	IHC, flow cytometry, RNA-Seq	Mannose receptor and C-type lectin	[175]
B7-H4	Prostate	IF, flow cytometry	Co-stimulatory protein of antigen-presenting cells	[176]
COX-2	Breast	IHC, multiplex IF	An enzyme responsible for formation of prostanoids
HLA-DR	Breast, ovarian	Multiplexed IHC, IHC	MHC class II cell surface receptor	[177]
IGF1	Lung, ovarian	Gene chip analysis	Anabolic hormone	[178]
iNOS	Ovarian	IHC, IF	Enzymes catalyzing the production of NO from L-arginine	[179]
MARCO	Lung, ovarian	Multiplex IF, RNA-seq	Class A scavenger receptor	[180]
MMP-9	Breast, lung	IF	Matrix metalloproteinase	[181]
mTORC2	Colorectal	IF	Rapamycin-insensitive protein complex	[182]
PD-L1 (CD274)	Ovarian	IF	Immunosuppressive protein	[183]
SIGLEC1 (CD169)	Breast	RNA-seq, qPCR	Sialic binding receptor	[184]
OPN	Lung	IHC	A protein involved on angiogenesis and metastasis	[185]
Stabilin-1 (RS1)	Breast, colorectal	IHC, IF	Scavenger receptor or intracellular sorting receptor	[186]
TIE2	Breast	IF	Angiopoietin receptor	[187]
TREM-1	Lung	IF, ELISA, Western blotting	A receptor involved in the regulation of inflammatory response	[188]
VEGF	Colorectal, ovarian	IHC, qPCR	Growth factor	[189]
VSIG4	Lung	IF	Co-stimulatory protein of antigen-presenting cells	[190]
CHI3L2 (YKL-39)	Breast	IHC, qPCR	A chitinase-like protein, often involved in immune and inflammatory responses	[191]
CHI3L1 (YKL-40)	Breast, lung, prostate	IHC, qPCR, ELISA	A chitinase-like protein, pro-angiogenic	[192]
ZEB1	Ovarian	IHC	A transcription factor – driver of EMT	[193]

Tim3: T-cell immunoglobulin and mucin domain-containing molecule 3, MR: mannose receptor, KRAS: Kirsten rat sarcoma virus, OPN: osteopontin, TP: thymidine phosphorylase, COX2: cyclooxygenase, HLA-DR: Human Leukocyte Antigen – DR isotype, IGF1: insulin like growth factor 1, iNOS: induced nitric oxide synthases, MACRO: macrophage receptor with collagenous structure, MMP-9: matrix metalloproteinase 9, Mtorc2: mTOR complex 2, PDL1: Programmed death-ligand 1, SIGLEC: sialic acid-binding immunoglobulin-type of lectins, TIE2: Tyrosine-protein kinase receptor, TREM1: triggering receptor expressed on myeloid cells, VEGF: vascular endothelial growth factor, VSIG4: V-set and immunoglobulin domain containing 4, CHl3L1: chitinase-3-like protein 1, CHl3L2: chitinase-3-like protein 2, ZEB1: Zinc finger E-box binding homeobox 1, EMT: epithelial-mesenchymal transition, IHC: immunohistochemistry, IF: immunofluorescence

TAM-related biomarkers in gastric cancer

Various macrophage-related biomarkers have been recognized in gastric cancer that have prognostic value in gastric cancer. Thymidine phosphorylase (TP), an enzyme that promotes angiogenesis, is commonly expressed in TAMs. TP expression level, CD68+, and CD163 + infiltrating macrophage counts, and tumor microvessel density (MVD) were assessed by immunohistochemistry in 111 patients with gastric cancer. The analysis showed that TP expression was significantly correlated with the number of infiltrating macrophages and MVDs in intestinal gastric cancer. The number of infiltrating macrophages was also correlated with MVD in the intestinal and diffuse types. An increased number of CD68 + macrophages was significantly associated with poor outcomes in patients with CRC, but not diffuse cancer. TP may be a specific marker enzyme expressed in tumor-infiltrating macrophages and involved in tumor angiogenesis and its expression is correlated with poor prognosis in patients with gastric intestinal cancer [120].

OPN is an extracellular matrix protein which functions like a cytokine, and involved in macrophage recruitment in gastric cancer. Lin et al. measured OPN level and TAM count in 170 gastric cancer samples, and showed that OPN can redirect macrophages to M2 TAMs [121]. They also demonstrated that simultaneous expression of OPN and CD204 in TAMs has a correlation with disease progression and poor 5-year survival. Therefore, they pointed out that OPN and TAMs may have values as biomarkers in gastric cancer [121].

T-cell immunoglobulin and mucin domain-containing molecule 3 (Tim-3) is expressed in macrophages and affects the cancer-immune interaction. Gastric cancer patients showed upregulation of Tim-3 in their monocytes compared to healthy subjects, which is associated with lymph node metastasis, depth of tumor invasion, and advanced clinical stages [122]. Hence, Tim-3 may play an important role in gastric cancer progression.

Mannose receptor (MR) is an adhesive molecule mainly found on the surface of macrophages and plays an important role in phagocytosis and endocytosis. The level of MR expression in the tumor microenvironment is significantly high and there is a significant correlation between the increase in the amount of MR in gastric cancer stroma and tumor size, tumor stage and shortened survival [123].

Lysyl tRNA synthetase (KRS) is an aminoacyl tRNA synthetase essential for protein synthesis and is part of RAS/MAPK signaling pathway. About 43.3% of gastric cancers have shown higher levels of KRS, and in more than 37% of the patients, macrophages and monocytes have been shown to express large amounts of this protein [124]. In addition, here has been an association between KRS expression in gastric cancer stroma and clinico-pathological parameters [124]. Therefore, KRS may be an independent prognostic marker for gastric cancer. Studies have also reported that CD204-positive TAMs, presented in the stroma of gastric tumors, may be considered as a risk factor for the development of gastric adenocarcinoma adenoma [125]. This finding suggests that there may be promises for screening high-risk patients who are at the risk of developing stomach cancer [125].

Tumor-infiltrating CD11bþ-positive antigen-presenting cells including TAMs and dendritic cells are another possible independent prognostic factor in gastric cancer. Okita et al. reported that patients with a high amount of CD11bþ cells in their tumor stroma, had a lower survival rate compared to patients with low amount of CD11bþ cells [126].

Infiltration of CD163þ-expressing TAMs substantially differs compared to other TAMs. A study on 178 gastric cancer patients showed that tumor tissue expressed high CD163þ infiltration rates in 52.8% of cases. Normal tissue expressed high amounts of CD163þ only in 21.3% of cases. Patients with high expression of CD163þ had a shorter OS compared to those with low expression of CD163þ. This study implied that high expression of CD163þ, together with TGF-β, is associated with an aggressive character of cancer, and therefore, could be used as an independent prognostic factor in gastric cancer [127].

Macrophage migration inhibitory factor (MIF) is a pro-inflammatory cytokine that participates in the innate immune response by affecting macrophages. This factor has shown to have high potential in the prognosis of gastric cancer in patients with dyspepsia [128]. A lower 5-year survival rate was observed in gastric cancer patients with high serum MIF values. The expression of MIF in gastric cancer tissues was higher compared to adjacent normal non-cancerous tissues, and there was a significant correlation between the high level of MIF and lymph node metastasis, poor tumor differentiation, advanced tumor stage and poor survival [129].

A meta-analysis about the significance of TAMs in gastric cancer was performed based on 20 studies that included 1388 patients. This study demonstrated that there is a negative correlation between the rate of TAM infiltration and OS [130]. This meta-analysis indicates that total TAMs and infiltrating M2 macrophages might be a negative prognostic factor for patients with gastric cancer. The analysis also showed a correlation between elevated TAM infiltration and decreased risk for lymph node metastasis in gastric and ovary cancers, i.e. patients with a high density of TAMs have a lower probability to have lymph node metastasis [130]. Taken together, it seems that TAMs may have great potential for prognostic purposes in gastric cancer; they may have promises for prediction of tumor size, depth of invasion, and lymph node metastasis as well as OS. More studies, however, are needed to evaluate whether TAM could be used as a biomarker for gastric cancer in clinical setting.

TAM-related biomarkers in breast cancer

Studies suggest that a high density of cells expressing macrophage-related markers in primary breast cancer was generally associated with worse patient prognosis and increased metastatic risk [45, 131, 132]. TAM abundance has also been correlated with unfavorable clinico-pathological features such as tumor grade, lack of hormone-receptor expression, and diminished survival. Furthermore, TAM density positively correlates with advanced-stage disease in breast cancer patients [133, 134]. Some of TAM-associated markers include CD163, VEGF, HIFs, PCNA, FTL and CCL18 [111].

The most commonly used M2 markers for the analysis of TAM phenotype in human breast cancer include CD68 [45], CD163, CD204, CD206, and stabilin-1 [133]. Several clinical studies showed that expression of CD163 in breast tumor stromal cells is positively correlated with disease recurrence, lymph node involvement, poor histological grade, larger tumor size, Ki67 positivity and poor survival [132, 133]. CD68 + macrophage infiltration was also associated with poor prognostic characteristics, such as larger tumor size, lymph node metastasis, higher tumor grade, vascular invasion, hormone receptor negativity, HER2 expression, and basal phenotype [131]. Additional markers expressed on other cell types, including CD47, COX-2, MMP-9, TIE2, YKL-39, YKL-40, and PD-L1, can also be used for characterizing functional TAM phenotype. Although CD163 can be used as an independent biomarker that indicates a poor prognosis for patients with breast cancer, it has been found that the combination of markers can also be used to identify the correlation of TAM rate/phenotype with clinical parameters and metastasis in these patients. For example, high number of COX-2/CD163 in the tumor nest and tumor stroma and high number of CD68+/COX-2 TAMs in the tumor stroma were observed to be correlated with poor survival in tumors of breast cancer patients [113, 133, 135].

The prognostic value of CD163 + and CD68 + macrophages may be dependent on the breast cancer subtype. It has been reported that the intensity of macrophage infiltration is strongly associated with estrogen receptor (ER)/progesterone receptor (PR)-negativity and high mitotic rates [111, 131]. High infiltration of CD68 + macrophages was associated with shorter disease-free survival (DFS) and/or OS in triple-negative breast cancer (TNBC) and ER + breast cancer. In comparison, high infiltration of CD163 + macrophages can be used as a prognostic factor for worse DFS and/or OS in TNBC and HER2 + breast cancer patients [131]. For example, high expression of MMP-9 in CD68+/CD163 + TAMs was associated with worse OS in ER + breast tumors. On the other hand, the amount of stabilin-1 + TAMs in the breast cancer patients was mainly abundant in stage I. High infiltration of CD68 + TAMs in TNBC had a significantly higher risk of developing distant metastasis as well as lower rates of OS and DFS. A high number of intra-tumoral CD68 + TAMs was an independent prognostic factor for poor DFS in hormone receptor-positive breast cancer subgroup, but not in hormone-receptor-negative subgroup [133]. In addition, since CD36, an essential receptor for lipid uptake and TAM generation, plays a vital role in the occurrence and development of tumors, this factor has been suggested as a prognostic marker in various cancers, mainly of epithelial origin, such as breast cancer [116].

Localization of TAMs in tumor stroma and tumor nest has suggested a controversial clinical value of TAMs in the progression and prognosis of tumors. High density of CD68 + TAMs in tumor stroma was significantly correlated with larger tumor size. Also, high infiltration of CD68 + and CD163 + TAMs in tumor stroma was significantly associated with lymph node metastasis. The numbers of CD163 + macrophages in tumor nest could be used as an independent prognostic marker, mirroring reduced OS and DFS in metastatic breast cancer patients [111, 133]. Overexpression of CSF-1, the central lineage regulator for macrophages, and CCL2 chemokine have been correlated with poor prognosis of breast cancer patients [45, 53]. CCL-2 and CSF-1 are overexpressed in a wide range of cancers and are associated with poor prognosis in breast cancer [53, 136].

AnxA2 has also been identified as a significant biomarker for breast cancer diagnosis. Pathological investigation of breast tissue specimens showed that AnxA2 was upregulated in breast cancer tissues compared with adjacent non-tumor tissues. It was also upregulated in 68% of patients with lymph node metastases, 65% of patients with distant metastases, and 32% of those with non-lymph node metastases. These findings confirmed that AnxA2 was highly expressed in most breast cancer patient tissues and can be regarded as a prognosis indicator of lymph node and distant metastasis [44]. Increased levels of VEGF lead to higher recruitment of macrophage into the TME; macrophages, in turn, increase tissue and serum VEGF levels in human breast cancer, where VEGF scores in macrophages and serum predict patients’ outcomes [111].

Hypoxia activates TAMs to display an angiogenic phenotype, which could explain the correlation between TAM HIF-2a and angiogenesis; conversely, in less hypoxic regions, which still depend on angiogenesis, alternative nonhypoxia-dependent TAM angiogenic pathways, like TP-mediated axis may exist. According to these results, VEGF-, HIF-, TP-expressing, or hypoxic macrophages can be potential prognostic indicators of angiogenesis, metastasis, and high tumor grade [111]. Moreover, TAM-associated FTL, which is stored in M2 (CD68+/CD163+) TAMs, has been shown as a prognostic biomarker in lymph node-negative breast cancer [111].

Chen et al. found that breast cancer TAMs abundantly produced CCL18, and its expression in blood or cancer stroma has been shown to be associated with metastasis and reduced survival [137]. The count of CCL18 + TAMs can significantly predict histological grade, disease stage, lymph node, and distant metastasis. Therefore, the density of stromal CCL18 + TAMs and serum level of CCL18 are novel prognostic biomarkers independent of other clinicopathological parameters [111].

TAM-related biomarkers in ovarian cancer

In ovarian cancer, TAMs are involved in tumor progression through angiogenesis, stromal remodeling, cell invasion and immune escape. The M1/M2 ratio correlates with clinical outcomes in ovarian tumors [138, 139].

A better OS rate is generally correlated with increased immune cell infiltration. Yafei et al. studied 42 patients with ovarian cancer at all disease stages to determine the predictive significance of CD68 + and CD163 + positive macrophages. According to their research, a high percentage of CD163+ (M2 phenotype) in whole CD68 + macrophages is a biomarker for a poor prognosis [140]. Liu et al. investigated publicly accessible datasets and evaluated 13 separate studies involving 2218 HGSOC patients. According to the results, having more M1 macrophages is linked to a better prognosis, while having more M2 macrophages is linked to a poor prognosis [17].

The results of a meta-analysis of 9 studies comprising 794 individuals with ovarian cancer revealed that tumors with a high M1/M2 ratio had a better prognosis [141]. The M1/M2 and M2/whole TAM ratios are positively linked with progression-free survival (PFS) and OS, although the overall TAM density in ovarian malignancies showed poor prognosis [142].

A clinical study revealed that high tumor infiltration with M2 phenotype was associated with poor prognosis and unfavorable OS, supporting the harmful involvement of TAMs in ovarian cancer. It assists in the clinical screening of patients with a high probability of survival. Moreover, it has great clinical significance in boosting the prognosis of ovarian cancer patients.

TAMs-related biomarkers in colorectal cancer

The development and progression of CRC involve complex processes, such as genetic alterations in cancer cells and various elements of the TME that surround the tumor cells. In CRC, macrophages are the most abundant immune cells in TME, and play a critical role in tumor formation, cancer progression and metastasis [143, 144]. TAM may affect the effectiveness of cytoreductive therapies and antagonize or synergize with the antitumor activity of these therapies. Therefore, depending on the treatment, the prognosis of TAM in CRC may change accordingly [145–149]. Furthermore, re-educating the polarization of TAMs may aid tumor immunotherapy [143].

Yamei Zhao et al. showed that, unlike other solid tumors, high density CD68 + macrophage infiltration could be an excellent prognostic marker for CRC. However, when macrophages act as targets of combination therapy in the treatment of CRC, it may be more effective for CRC patients with high CD8 + T cell infiltration with high microsatellite instability, no distant metastasis, no lymph node metastasis and non-mucinous cancer [150]. Nevertheless, Waniczek et al. examined 89 patients with CRC and showed that infiltration of CD68 + macrophages into the tumor stroma was an adverse prognostic factor [151]. Pinto et al. demonstrated that stage III CRC sufferers with reduced CD68 + macrophage infiltration and increased CD80/CD163 ratios have higher OS rates [152]. Yang et al. showed that, in CRC patients, increased CD163 + and CD68 + ratios at the tumor infiltration front, positively correlated with shortened RFS, and OS. This finding highlights the importance of assessing TAM polarization in evaluating the prognosis of CRC patients [153, 154]. Even though M2 TAMs play an adverse influence in a poorer prognosis for most malignancies, it is less obvious in the case of CRC.

Feng et al. included two separate cohorts of patients with stage II colon cancer following radical resection. They demonstrated that patients may be efficiently divided into groups with a low and high risk of tumor recurrence using the CD206/CD68 ratio of TAMs. The CD206/CD68 ratio could also help physicians identify patients who would benefit from postoperative adjuvant chemotherapy. This indicates that CD206/CD68 ratio is probably a better biomarker for stage II colon cancer prognosis and predictions following adjuvant treatment [155].

As stated above, in CRC, the prognostic significance of TAM infiltration has been reported as both favorable and unfavorable, reflecting substantial biological and methodological heterogeneity. One major source of discrepancy is the microanatomical localization of TAMs within the tumor. A comprehensive meta-analysis of over 6,000 CRC patients found that high densities of CD68⁺ TAMs in the tumor stroma (TS) were associated with improved 5-year overall survival, whereas CD68⁺ TAMs in tumor islets (TI) showed no significant correlation with survival. Interestingly, elevated TAM densities in regional lymph nodes also predicted better CRC prognosis [156]. These findings indicate that stromal TAMs may exert anti-tumor or immune-supportive functions, whereas macrophages in other compartments may have neutral or context-dependent roles, highlighting the importance of spatial resolution when assessing TAM prognostic impact.

Another important contributor to conflicting outcomes is the choice of macrophage markers used to define TAM populations. CD68, a pan-macrophage marker, does not distinguish between functionally distinct macrophage states, potentially masking opposing activities within the TAM compartment. In contrast, CD163, often interpreted as an M2-like or immunosuppressive marker, has been more frequently associated with poor prognosis, including reduced overall and disease-free survival in CRC cohorts [157]. However, even CD163-based studies report variable outcomes, reflecting the fact that CD163 expression alone does not capture the full functional spectrum of TAMs. Studies combining marker expression with spatial analysis have shown that high CD163⁺ or high CD163⁺/CD68⁺ ratios at the invasive front correlate with tumor invasion and adverse outcomes, whereas CD68⁺ macrophages in intraepithelial or stromal compartments may associate with reduced invasiveness and improved survival [158, 159].

Beyond localization and phenotype, the broader immune context critically shapes the prognostic impact of TAMs in CRC. Recent large-scale analyses demonstrated that high TAM infiltration predicts favorable relapse-free survival only in tumors with robust T-cell infiltration, whereas TAM accumulation in immune-cold tumors correlates with poorer outcomes [158]. These findings suggest that TAMs can support anti-tumor immunity in immunologically active tumors but may instead reinforce immune suppression in T-cell–poor environments. Together, these observations indicate that TAM prognostic associations in CRC cannot be interpreted in isolation but must be evaluated in the context of spatial distribution, phenotypic definition, and immune ecosystem. Collectively, these studies argue for integrative approaches combining spatial profiling, multiplex marker analysis, and immune contexture to resolve the apparent contradictions in TAM prognostic studies in CRC.

TAMs-related biomarkers in lung cancer

Lung cancer is the primary cause of cancer-related death for men and women all over the world. Non-small cell lung cancer (NSCLC) accounts for approximately 85% of reported lung cancer cases, with most new cases diagnosed at advanced stages [160–163]. According to estimates, up to 80% of patients with NSCLC are diagnosed only after their cancer has already spread to regional lymph nodes or metastasized to distant organs [164]. Findings suggest that TME is a predictive biomarker for lung cancer and plays important role in the disease progression [165, 166]. In the TME of NSCLC, TAMs constitute the predominant cellular component. They not only act as immunosuppressive cells enabling immune evasion of NSCLC cells, but also directly contribute to cancer cell proliferation, survival, invasion, and metastasis [147]. Due to differences between previous studies on TAM infiltration and lung cancer prognosis, the prognostic relevance of TAMs in NSCLC is still under debate. In 2016, Mei et al. examined 20 separate studies involving 2,572 NSCLC patients. Their results indicated that, although the total CD68 + TAM density was not associated with OS, TAM location and the M1/M2 ratio may serve as prognostic predictors of NSCLC [167]. Patient survival was correlated with both islet and stromal CD68 + TAM density in lung cancer. Poor survival was indicated by low islet and high stromal CD68 + TAM density. Also, it was possible to link low islet M1 count or high stromal M2 count to tumor progression [167]. In 2020, Hwang et al. reported that patients with high M1 macrophage expression had significantly better OS compared to patients with low infiltration of M1 macrophages, while patients with high M2 ratio (CD163+/CD68+) had significantly worse OS compared to that of patients with low M2 ratio [164]. Zheng et al. investigated TAM subtype density and distribution between tumor center and invasive margin in lung tumor tissue from 104 patient samples. They demonstrated that higher proximity of tumor cells to M2 invasive margin and lower proximity to M1 macrophages (which are mostly localized at the center of the tumor) were linked to poor survival [168]. In a recent study by Kawaguchi et al., it was shown that high expression of M2 marker, CD206, were associated with tumor recurrence and low OS in patients with lung cancer [169]. These contradictory reports in NSCLC regarding the role of TAMs in tumor suppression or progression, may relate to the choice of TAM marker. Various TAM subtypes, each with different markers, may have distinct and even opposite functions. Other reasons may include low statistical power, homogeneous cohorts (using a particular tumor stage), and discrepancies in the method used to assess patterns of macrophage infiltration [170].

Conclusion

Tumor-associated macrophages (TAMs) play a pivotal role in the progression and metastasis of cancer. Their diverse origins, plasticity, and ability to polarize into distinct M1 and M2 subtypes highlight their complexity and the significant impact they have on tumor biology. While M1 macrophages are typically associated with anti-tumor responses, M2 macrophages contribute to tumor progression, immune suppression, and metastasis, making the balance between these subtypes a critical factor in cancer prognosis. The presence and distribution of TAMs in the tumor microenvironment have been identified as significant prognostic markers, with high levels of M2 macrophages often correlating with poor survival outcomes. Additionally, the plasticity of TAMs offers a promising therapeutic avenue, with strategies aimed at reprogramming macrophages to enhance anti-tumor immunity. As our understanding of TAM biology continues to evolve, their potential as both prognostic biomarkers and therapeutic targets will likely play an increasingly important role in cancer treatment strategies. Further research into the mechanisms governing TAM polarization and function is essential for improving patient outcomes and developing targeted therapies that can disrupt the pro-tumor functions of TAMs.

TAMs have emerged as valuable prognostic biomarkers, with their presence, density, and polarization status correlating with disease progression and patient outcomes. In gastric cancer, breast cancer, and ovarian cancer, specific TAM-related biomarkers have shown promise in predicting prognosis and response to therapy. These biomarkers offer potential for individualized cancer treatment strategies, highlighting the need for further exploration of TAMs as predictive tools in clinical practice.

Targeting TAMs or their specific pathways has the potential to improve cancer prognosis and enhance therapeutic efficacy. Modulating the M1/M2 polarization balance, reprogramming TAMs, or disrupting their interactions with tumor cells could provide novel therapeutic approaches to limit tumor growth and metastasis. As research into TAMs continues, a deeper understanding of their diverse roles across different cancers will be essential in developing more effective prognostic biomarkers and therapeutic interventions, ultimately leading to improved patient outcomes.

Author contributions

AS and MY: conceptualization, original draft writing, supervision, editing. SML, ZF, SG, FP, SK: investigation, original draft writing. PSF: figure preparation, review, editing. All authors reviewed the manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare that they have no conflict of interest.