Macrophages in endometriosis: key roles and emerging therapeutic opportunities-a narrative review

Endometriosis (EM) is a chronic estrogen-dependent inflammatory disorder characterized by the ectopic implantation, proliferation, infiltration, and cyclic shedding of the functional layer of endometrial glandular and stromal tissue outside the uterine cavity [ 1 , 2 ]. This condition affects approximately 10% of women of reproductive age worldwide, significantly impairing their quality of life and fertility [ 1 , 3 , 4 ]. Common symptoms include pelvic pain, dysmenorrhea, and infertility, but diagnosis is frequently delayed because of the presence of asymptomatic lesions and the need for surgical confirmation [ 5 ]. Current treatments primarily rely on surgical intervention and hormonal therapies. However, surgery carries high recurrence rates, while pharmacological interventions often induce hypoestrogenic side effects, underscoring the urgent need for novel therapeutic approaches [ 6 , 7 ]. Despite extensive research, the precise etiology of endometriosis remains unresolved. Several theories have been proposed, including Sampson’s retrograde menstruation theory, coelomic metaplasia, and lymphatic or vascular dissemination [ 8 , 9 ]. Among these, the retrograde menstruation theory, suggested by Sampson in 1921, posits that viable endometrial cells reflux into the peritoneal cavity during menstruation, leading to implantation [ 10 ]. However, 76–90% of healthy women also experience retrograde menstruation without developing the disease, indicating that additional factors, such as immune dysregulation, a hyperestrogenic environment, and enhanced inflammatory responses, contribute to disease initiation and progression [ 1 , 11 ]. Recent research increasingly highlights immune dysfunction as a key contributor to the pathogenesis of endometriosis, with macrophages playing a central role in disease progression. A compromised immune response, characterized by impaired cytotoxicity and phagocytic capacity, fails to effectively eliminate ectopic endometrial cells, creating a permissive environment for lesion persistence [ 12 ]. Dysregulated activation of both innate and adaptive immunity leads to excessive production of inflammatory mediators and cytokines, further supporting the establishment and growth of ectopic lesions [ 13 ]. Compared to normal endometrium, endometriotic lesions exhibit an altered immune landscape, prominently marked by an increased presence and altered function of macrophages [ 14 ]. These macrophages display abnormal distribution and activation, encompassing subsets with pro-inflammatory features—such as secretion of monocyte chemoattractant protein-1 (MCP-1) and production of interleukin-1β (IL-1β), IL-6, and tumor necrosis factor-α (TNF-α)—that amplify local inflammation, as well as subsets with immunoregulatory profiles that facilitate tissue repair and immune evasion [ 15 , 16 ]. This functional plasticity allows macrophages to participate in multiple stages of endometriosis pathophysiology, from initiating inflammatory cascades to sustaining lesion survival through angiogenesis, fibrosis, and immune modulation [ 12 ]. Recent single-cell RNA sequencing analyses have further highlighted the prominence of macrophages within the endometriotic microenvironment, identifying them as significant contributors to ectopic lesion formation through pro-angiogenic and fibrotic activity [ 17 ]. This review examines the multifaceted role of macrophages in the endometrium under physiological conditions, as well as their pathological alterations in endometriosis. We examine how macrophages contribute to chronic inflammation, impaired phagocytosis, angiogenesis, and fibrosis, ultimately facilitating disease progression. Notably, the review highlights emerging macrophage-targeted therapeutic strategies, including inhibition of macrophage recruitment, repolarization of macrophages, and modulation of key immune checkpoints such as CD47-Signal Regulatory Protein α (SIRPα) axis. By integrating recent advances in macrophage biology with novel therapeutic approaches, this review aims to provide new insights into macrophage-mediated mechanisms in endometriosis and assess the potential of immunomodulatory interventions as innovative treatment strategies for this chronic disease. Macrophages are versatile innate immune cells essential for development, tissue homeostasis, repair, and host defense [ 18 ]. Advances in fate-mapping, single-cell omics, and epigenetic profiling indicate that most macrophages arise from two hematopoietic waves - erythro–myeloid progenitors (EMPs) and hematopoietic stem cells (HSCs) - that differ in temporal emergence, developmental routes, and functional potential [ 19 – 21 ], (see Fig. 1 A). The earliest hematopoietic progenitors emerge from the posterior lateral plate mesoderm of the extraembryonic yolk sac [ 22 , 23 ]. These primitive progenitors generate erythrocytes and megakaryocytes, and microglia are thought to arise from yolk sac–derived primitive macrophages [ 24 ]. The subsequent, “classical” EMP wave arises from yolk sac hemogenic endothelium, seeds the fetal liver, and gives rise to pre-macrophages that populate developing organs as long-lived, self-renewing tissue-resident macrophages (TRMs)—for example, microglia, Langerhans cells, and Kupffer cells—supporting immune quiescence and tissue homeostasis [ 25 – 27 ]. By contrast, HSCs originate in the aorta–gonad–mesonephros (AGM), expand in the fetal liver, and ultimately home to the bone marrow to sustain lifelong hematopoiesis. HSC-derived monocytes enter the circulation at steady state and during inflammation, infiltrate tissues, and differentiate into shorter-lived macrophages that replenish or partially replace TRMs in a tissue- and context-dependent manner [ 28 , 29 ]. While most fate-mapping studies support EMP dominance in TRM ontogeny, contributions of HSC-derived monocytes—or their fusion with EMP-derived cells—have been documented at specific sites (e.g., osteoclasts; intestinal lamina propria), underscoring ontogeny-dependent plasticity [ 30 ]. This dual-lineage paradigm—early self-renewing TRMs and stress-expanded bone marrow–derived macrophages—appears evolutionarily conserved, being observed across mammals and in invertebrate models such as Drosophila [ 31 ]. Fig. 1 Macrophage ontogeny and activation spectrum. A Dual developmental origins and tissue seeding—erythro–myeloid progenitors (EMPs) vs. hematopoietic stem cells (HSCs); B M1–M2 activation continuum with transcriptional, metabolic, and epigenetic regulation; C M2 subtypes: induction cues and functional characteristics (Created by Figdraw ( http://www.figdraw.com )) Macrophage ontogeny and activation spectrum. A Dual developmental origins and tissue seeding—erythro–myeloid progenitors (EMPs) vs. hematopoietic stem cells (HSCs); B M1–M2 activation continuum with transcriptional, metabolic, and epigenetic regulation; C M2 subtypes: induction cues and functional characteristics (Created by Figdraw ( http://www.figdraw.com )) The M1/M2 scheme introduced by Mills et al. (2000) remains a useful historical touchstone for macrophage activation. In vivo, however, macrophage states lie along a continuum; “M1-like” and “M2-like” are best viewed as poles of that spectrum [ 32 ] (see Fig. 1 B). M1-like (classically activated) macrophages, induced by lipopolysaccharide (LPS) or T helper 1 (Th1) cytokines such as interferon-γ (IFN-γ), produce interleukin-12 (IL-12), TNF-α, and IL-1β and up-regulate inducible nitric oxide synthase (iNOS/NOS2), CD80/CD86, and major histocompatibility complex class II (MHC II), supporting antimicrobial and antitumor activity [ 33 ]. M2-like (alternatively activated) macrophages, induced by Th2 cytokines (IL-4/IL-13), participate in immunoregulation and tissue repair [ 33 , 34 ]. Based on stimuli and functions, M2-like states have been subdivided into M2a/M2b/M2c/M2d [ 35 ], (see Fig. 1 C): M2a (IL-4/IL-13) expresses IL-10, transforming growth factor-β (TGF-β), and CCL17/18/22 and promotes phagocytosis and repair [ 36 , 37 ]; M2b (immune complexes + TLR ligands + IL-1β) co-produces pro- and anti-inflammatory cytokines (e.g., TNF/IL-1β/IL-6 with IL-10), modulating response magnitude and duration [ 38 ]; M2c (glucocorticoids/IL-10/TGF-β) secretes IL-10, TGF-β, CCL16/18 and mediates efferocytosis and remodeling [ 39 , 40 ]; M2d (adenosine + TLR agonists) produces IL-10, vascular endothelial growth factor (VEGF), and CCL18, favoring angiogenesis and tumor immune escape [ 41 ]. At the molecular level, M1-like polarization depends on transcription factors such as nuclear factor-κB (NF-κB), signal transducer and activator of transcription 1 (STAT1), STAT5, and interferon regulatory factor 3 (IRF3). M2-like states rely on STAT6, IRF4, peroxisome proliferator–activated receptor δ/γ (PPARδ/γ), and the histone demethylase JMJD3 [ 42 – 46 ]. Metabolically, M1-like macrophages predominantly utilize glycolysis and the pentose phosphate pathway, whereas M2-like macrophages rely on oxidative phosphorylation and fatty-acid oxidation [ 47 , 48 ]. Epigenetically, HDAC3 promotes M1-like polarization, whereas HDAC4/5, STAT6 signaling, and TET2 activity facilitate M2-like polarization; conversely, DNMT3b represses M2-associated genes, thereby driving polarization toward M1-like programs [ 49 – 51 ]. Post-transcriptional regulators also contribute: in M1-like states, miR-155 amplifies NF-κB signaling, miR-125b inhibits IRF4, and miR-146a restrains inflammation by targeting IRAK1/TRAF6; in M2-like states, miR-21 promotes IL-10 via PDCD4 suppression, miR-124 induces ARG1/FIZZ1 via C/EBPα, and miR-223 stabilizes M2-like programs by inhibiting Pknox1 [ 52 – 55 ]. Although the M1/M2 model has provided an important framework for macrophage biology, its limitations have become increasingly apparent. First, genetic background shapes macrophage metabolism and effector functions (e.g., differences in arginine flux after Slc7a2 promoter deletion in C57BL/6 vs. BALB/c)—knowledge not incorporated when the paradigm was proposed [ 56 ]. Second, macrophages generated under simplified in vitro conditions—often stimulated with a single factor and assessed by a limited set of markers—cannot fully capture the phenotypic complexity observed in vivo. In vivo, lymphocytes profoundly influence macrophage polarization through the secretion of various cytokines—an adaptive immune regulation that was not fully accounted for in the original model [ 57 ]. Third, in complex pathological contexts such as cancer, autoimmune diseases, and endometriosis, macrophage functions cannot be fully captured by a binary M1/M2 classification [ 58 ]. Single-cell transcriptomics and CyTOF (cytometry by time-of-flight) analyses reveal a continuum of activation states, with some cells co-expressing M1 and M2 markers, and even in endometriosis, macrophages may simultaneously express pro-inflammatory and tissue-repair genes [ 58 – 60 ]. This spectrum concept, shaped by microenvironmental, metabolic, and epigenetic cues, indicates that reliance on a binary M1/M2 framework may limit precision therapeutics; accordingly, the field is moving toward refined subpopulation delineation and spectrum-based analyses, enabling more accurate characterization of macrophage heterogeneity and providing a stronger basis for targeted interventions. To improve reproducibility and cross-study comparability, Murray et al. proposed activation-based nomenclature and experimental standards [ 57 ]. Key elements include: (i) using CSF-1–induced BMDMs (mouse) and PBMC-derived macrophages (human) as references with explicit stimuli labels, e.g., M(IFN-γ), M(IL-4); (ii) fully documenting cell source, culture conditions, stimuli, and readouts; (iii) defining subpopulations by marker combinations rather than laboratory-specific names; (iv) adopting an activation spectrum framework; and (v) distinguishing CSF-1– from GM-CSF–derived macrophages. Regarding marker expression, it is important to recognize species-specific differences. In mice, M1 macrophages typically express iNOS (NOS2), CD86, and MHC II, whereas M2 macrophages express Arg1, CD206 (Mrc1), and mouse-specific markers such as Fizz1 (Retnla) and Ym1 (Chil3). In humans, M1 macrophages are characterized by costimulatory molecules such as CD80, CD86, and HLA-DR, while M2 macrophages are defined by CD163 and CD206 expression [ 61 – 63 ]. These distinctions are critical when comparing results across species and underscore the need to integrate marker combinations with species context, rather than directly extrapolating from one species to another. Notably, ARG1 is an unreliable M2 marker in human macrophages, reinforcing the need to interpret marker panels in a species-aware manner. In conclusion, the M1/M2 classification has provided a foundational framework for understanding macrophage immune functions, but its simplification cannot fully encompass the complexity of macrophages in diverse pathological states. Advances in single-cell omics, metabolomics, and epigenetics demand a shift toward viewing macrophage polarization as a continuous and multidimensional spectrum. Integrating human–mouse marker differences, standardized nomenclature, and functional frameworks will enable a more comprehensive understanding of macrophage roles in health and disease, ultimately facilitating the development of precise diagnostics and targeted therapeutics. Macrophages are essential in the normal endometrium and throughout the female reproductive tract: they provide immune surveillance, mediate tissue remodeling, and fine-tune inflammation to support the menstrual cycle, embryo implantation, maintenance of pregnancy, and the onset of parturition [ 64 , 65 ]. Their abundance gradually increases from 1 to 2% of endometrial cells in the proliferative phase to 3–5% in the secretory phase and peaks at 6–15% during menstruation, when M2-like macrophages (CD163 + CD14^low) predominate [ 66 – 68 ]. In the proliferative phase, macrophages express markers such as CD54 and CD69, supporting endometrial regeneration [ 69 ]. In the secretory phase, they regulate glandular remodeling and angiogenesis through the secretion of VEGF [ 69 , 70 ]. During menstruation, macrophages release matrix metalloproteinases (MMPs), such as MMP-9, aiding in the shedding of the functional layer [ 71 , 72 ]. They also phagocytose apoptotic cells and tissue debris during menstruation, contributing to orderly clearance and repair. Additionally, the marked increase in macrophages during the secretory phase promotes the differentiation of endometrial stromal cells into decidual cells through the secretion of factors such as IL-10 and TGF-β [ 73 ]. In contrast, the frequency of CD68 + IL-10 − iNOS + M1 macrophages in secretory endometrium (< 20%) is lower than in the proliferative phase (~ 30%), a shift that is thought to favor endometrial receptivity by helping establish a permissive immune milieu for implantation [ 74 , 75 ]. Notably, endometrial macrophages express high levels of Toll-like receptor 4 (TLR4) [ 76 ], thereby enhancing innate immune defense by recognizing pathogen-associated molecular patterns (e.g., LPS), with studies citing increased protection against Gram-negative bacteria (e.g., Neisseria gonorrhoeae , Chlamydia trachomatis ) [ 77 ]. Additionally, M2-like macrophages express co-stimulatory receptors such as CD40 and the mannose receptor CD206, which contribute to pathogen recognition [ 77 ]. M1-like macrophages predominantly release pro-inflammatory mediators such as TNF-α, IL-12, and IL-1β, whereas M2-like macrophages secrete anti-inflammatory cytokines including IL-10. This dynamic balance between inflammation and immune tolerance is essential for maintaining uterine homeostasis and supporting reproductive function [ 11 ] (see Fig. 2 ). Fig. 2 The role of macrophages in normal endometrium The roles of macrophages in the endometrium of healthy women can be categorized into three key aspects: (1) promotion of angiogenesis and endometrial repair, achieved through the secretion of growth factors such as VEGF and TGF-β; (2) phagocytosis of menstrual debris and endometrial fragments, facilitating the clearance of the uterine cavity; (3) immune defense mechanisms, wherein macrophages express high levels of TLR4 and CD40 to maintain a sterile uterine environment (Created by Figdraw ( http://www.figdraw.com )) The role of macrophages in normal endometrium The roles of macrophages in the endometrium of healthy women can be categorized into three key aspects: (1) promotion of angiogenesis and endometrial repair, achieved through the secretion of growth factors such as VEGF and TGF-β; (2) phagocytosis of menstrual debris and endometrial fragments, facilitating the clearance of the uterine cavity; (3) immune defense mechanisms, wherein macrophages express high levels of TLR4 and CD40 to maintain a sterile uterine environment (Created by Figdraw ( http://www.figdraw.com )) Recent studies primarily employ flow cytometry, immunohistochemistry, and immunofluorescence to analyze macrophage quantity and phenotype across menstrual cycle phases. However, due to methodological heterogeneity, reader subjectivity, and technical complexity, consistent conclusions regarding macrophage changes in the eutopic endometrium of endometriosis patients remain elusive [ 78 – 81 ]. Berbic et al. reported that women with endometriosis exhibited an increased density of total CD68 + macrophages in the eutopic endometrium during the proliferative phase, whereas the cyclical rise in macrophage numbers observed during menstruation in controls was absent in patients. Notably, a relative decrease in macrophage numbers during the mid-menstrual phase, compared to healthy controls, may impair clearance of endometrial fragments, potentially promoting retrograde menstruation and contributing to disease pathogenesis [ 78 ]. Another study also showed increased macrophage populations in the proliferative and secretory phases in endometriosis patients [ 80 ]. Conversely, Braun et al. found reduced macrophage numbers during early proliferative phase and no significant differences at other menstrual phases [ 81 ]. Additionally, reduced apoptosis in the eutopic endometrium was noted, potentially enhancing the survival of refluxed endometrial fragments at ectopic sites [ 81 ]. Takebayashi et al. observed an overall increase in macrophage numbers but a decreased proportion of CD163 + macrophages, consistent with a shift toward an M1-like polarization pattern [ 79 ] (Table 1 ). Table 1 Studies comparing the number of macrophages in the eutopic endometrium The number of macrophages in the eutopic endometrium（endometriosis vs control） Study Method Marker Menstrual Cycle Age Cases Controls Proliferative phase Secretory phase Menstrual phase Berbic et al. [ 37 ] Standard immunohistochemical techniques CD68-PGM1 (phosphoglucomutase 1) Increase - ns（Decreases in the mid-menstrual phase） 16-54 42（21,Proliferative samples+21,Menstrual samples） 34（20,Proliferative samples+14, Menstrual samples） Braun et al. [ 39 ] Standard avidin-biotin-peroxidase complex (ABC) CD68 Decrease in the early- proliferative phase ns ns Reproductive age 51 24 Khan et al. [ 38 ] Immunohistochemistry CD68 （KP1） Increase Increase - 26-36 20(8,Proliferative samples+ 12,Secretory samples) 20(10,Proliferative samples+ 10,Secretory samples) Takebayashi et al.[ 33 ] Immunohistochemistry, Immunofluorescence CD68、CD163 Increase Increase Increase 44-48 36 37 ns no statistical difference, - The studies did not cover Studies comparing the number of macrophages in the eutopic endometrium Decrease in the early- proliferative phase Immunohistochemistry, Immunofluorescence ns no statistical difference, - The studies did not cover Vallvé-Juanico et al. utilized mass cytometry combined with immune phenotyping to reveal that the eutopic endometrium in endometriosis patients recruits increased numbers of immune cells, particularly classical and intermediate monocytes, which subsequently differentiate into macrophages [ 67 ]. In the proliferative phase of the eutopic endometrium, patients with endometriosis exhibited a higher overall proportion of macrophages compared with controls, with a notable enrichment of CD91 + macrophages. These CD91 + macrophages displayed elevated expression of the phagocytosis-inhibitory molecule SIRPα and the pro-inflammatory marker CD64, indicating reduced capacity for clearance of cellular debris and the maintenance of a local pro-inflammatory microenvironment [ 67 , 82 ]. This functional alteration was more pronounced in early-stage (I/II) disease, suggesting that CD91 + macrophages may play an important role in sustaining inflammation and promoting lesion progression during the initial stages of endometriosis [ 67 ]. Notably, accumulating evidence indicates that endometriosis does not maintain a uniformly pro-inflammatory profile throughout its course; instead, the immune landscape appears to undergo dynamic changes as the disease progresses. As reported by Poli-Neto et al., the eutopic endometrium in early-stage (I/II) disease is characterized by a predominantly pro-inflammatory immune composition, enriched in M1-like macrophages, activated dendritic cells, and CD4 + effector memory T cells, reflecting heightened local inflammatory activity. In contrast, advanced-stage (III/IV) disease exhibits a more anti-inflammatory microenvironment, with a higher proportion of M2-like macrophages and other immunosuppressive components, alongside enrichment of signaling pathways such as PI3K-AKT-mTOR and TGF-β [ 68 ].This shift from a pro-inflammatory to an anti-inflammatory state may influence lesion persistence and tissue remodeling, and suggests that the mechanisms of immune regulation vary across disease stages. Xie et al. demonstrated that peritoneal macrophages from women with endometriosis exhibit markedly elevated SIRP-α and reduced CD36 expression compared with healthy controls. Importantly, THP-1-derived macrophages exposed to homogenates of eutopic endometrium from endometriosis patients showed the same pattern of SIRPα upregulation and CD36 downregulation, indicating that eutopic endometrium can directly impair the phagocytic function of peritoneal macrophages [ 83 ]. This mechanism provides a plausible explanation for the defective clearance of shed endometrial cells and debris following retrograde menstruation, thereby contributing to local immune evasion and persistent inflammation in the peritoneal cavity [ 83 ]. Liu et al. found that in patients with endometriosis, key molecules in the thrombospondin-1 (TSP1)–CD47–SIRPα signaling pathway were significantly upregulated [ 84 ]. Functionally, TSP1 not only promotes the proliferation and fibrosis of normal endometrial stromal cells (NESCs) but also induces M2-like polarization of macrophages, increases IL-6 secretion, and downregulates p22^phox expression, which is associated with phagocytic function, thereby potentially impairing macrophage phagocytic capacity [ 84 ]. Additionally, studies suggest that infertile patients with endometriosis exhibit a higher proportion of M1-like macrophages, which is associated with NF-κB pathway activation and increased production of pro-inflammatory factors, which may impair embryo implantation and contribute to infertility and pain in affected individuals [ 85 ]. Macrophages are central to the pathogenesis of ectopic lesions in endometriosis, with increased abundance, context-dependent shifts toward repair/immunoregulatory (M2-like) programs, coexistence of pro-inflammatory outputs, and impaired phagocytic activity [ 17 ]. In the early stages, these features likely create a pro-inflammatory yet immune-evasive microenvironment that supports lesion proliferation, invasion, angiogenesis, fibrosis, and pain. Emerging evidence also indicates that the immune milieu may transition over time from a pro-inflammatory to a more immunoregulatory/anti-inflammatory state, reflecting the dynamic plasticity of macrophages. This dual role influences local inflammation and tissue remodeling, positioning macrophages as compelling targets for therapeutic intervention (see Fig. 3 ). Fig. 3 The role of macrophages in endometriosis. Phagocytic capacity of macrophages is reduced in eutopic endometrium; In ectopic endometrium, macrophages promote angiogenesis and lymphangiogenesis through VEGF and bFGF; macrophages regulate fibrosis by facilitating EMT and fibroblast differentiation, with S1P and TGF-β1 enhancing these processes; macrophages interact with nerve fibers in ectopic lesions, secreting neurotrophic factors like NGF and BDNF that increase neurogenesis and pain sensitivity. (Created by Figdraw ( http://www.figdraw.com )) The role of macrophages in endometriosis. Phagocytic capacity of macrophages is reduced in eutopic endometrium; In ectopic endometrium, macrophages promote angiogenesis and lymphangiogenesis through VEGF and bFGF; macrophages regulate fibrosis by facilitating EMT and fibroblast differentiation, with S1P and TGF-β1 enhancing these processes; macrophages interact with nerve fibers in ectopic lesions, secreting neurotrophic factors like NGF and BDNF that increase neurogenesis and pain sensitivity. (Created by Figdraw ( http://www.figdraw.com )) In a mouse model of induced endometriosis, lesion-resident macrophages were found to originate from eutopic endometrial tissue, infiltrating large peritoneal macrophages (LpM), and blood-derived monocytes. This model also demonstrated that endometriosis triggers continuous recruitment of monocytes and expansion of CCR2 + LpM [ 86 ]. Notably, Hogg et al. reported that, in mice, endometrium-derived macrophages promote disease progression, whereas monocyte-derived peritoneal macrophages exert an inhibitory effect on lesion development [ 86 ]. In the peritoneal cavity, two macrophage subsets are distinguished by residency and phenotype: large peritoneal macrophages (LpMs; embryo-derived, tissue-resident; F4/80^hi CD11b^hi, MHC class II^lo) and small peritoneal macrophages (SpMs; largely monocyte-derived, infiltrating; F4/80^lo CD11b + , MHC class II^hi) [ 87 , 88 ]. In a murine model of endometriosis, Yuan et al. demonstrated that large peritoneal macrophages (LPMs) rapidly infiltrate sterile injury sites within hours and contribute to early tissue repair, whereas small peritoneal macrophages (SPMs) accumulate at later stages and progressively acquire LPM-like phenotypic features. Importantly, their study further revealed that LPMs preferentially polarize toward an M1-like (pro-inflammatory) phenotype, while SPMs tend to adopt an M2-like (pro-resolving/immunosuppressive) phenotype. These findings suggest that the dynamic and complementary changes of LPMs and SPMs play a pivotal role in shaping the peritoneal inflammatory microenvironment. Owing to their distinct polarization tendencies, the overall peritoneal macrophage population exhibits a bidirectional polarization state, which may represent a key mechanism driving the onset and progression of endometriosis [ 89 ]. Macrophages play a crucial role in innate immunity by clearing pathogens and cellular debris. Reduced phagocytic activity in retrograde menstrual blood and peritoneal fluid has been implicated in ectopic lesion formation [ 11 ]. Xie et al. reported that diminished CD36 expression on peritoneal macrophages in endometriosis is associated with impaired clearance of apoptotic cells and oxidized lipoproteins [ 83 , 90 ]. Chuang et al. further demonstrated that blocking CD36 reduces macrophage phagocytosis, while its overexpression restores phagocytic function [ 91 ]. They proposed that prostaglandin E2 (PGE2) suppresses CD36 expression through the EP2-dependent signaling pathway, thereby diminishing phagocytic capacity and promoting ectopic lesion formation [ 92 , 93 ]. Additionally, Wu et al. suggested that PGE2 impairs macrophage function via the EP2/EP4-PKA pathway [ 94 ], and PGE2-mediated suppression of annexin A2 (ANXA2) further compromises the phagocytic capacity of peritoneal macrophages in women with endometriosis [ 95 ]. Elevated CD200 and heme levels in peritoneal fluid have been shown to inhibit macrophage phagocytosis, facilitating immune evasion in endometriosis [ 96 , 97 ]. Additionally, CD47 is markedly upregulated in ectopic endometrial stromal cells, acting as a “don’t eat me” signal by engaging SIRPα on peritoneal macrophages and thereby suppressing their clearance of shed endometrial cells and debris. Aberrantly activated ERβ in ectopic lesions directly binds the CD47 promoter, enhancing its transcription and further strengthening macrophage inhibition. Blockade of the CD47–SIRPα pathway restores macrophage phagocytosis and induces apoptosis of ectopic stromal cells, indicating a dual therapeutic potential [ 98 – 101 ]. Similarly, Liu et al. found that compared with NESCs, ectopic endometrial stromal cells (EESCs) more markedly upregulate SIRPα expression on macrophages, further inhibiting their phagocytic activity [ 84 ]. Menstrual blood with endometrial fragments adheres to the peritoneum or ovaries during retrograde menstruation, causing inflammation and forming ectopic lesions. Adhesion and invasion, driven by stromal cells, are key in lesion development [ 102 ]. Wang et al. found that fractalkine (FKN/CX3CL1) levels in peritoneal fluid and endometrial stromal cells (ESCs) correlate with disease progression. FKN recruits and activates immune cells, and promotes macrophage polarization towards the M2 phenotype, enhancing ESC invasiveness via P38-MAPK and integrinβ1 signaling [ 103 , 104 ]. Tan et al. demonstrated that co-culture with ESCs derived from ectopic endometriotic lesions stimulates macrophages to produce CCL20, which in turn activates CCR6 signaling in ESCs, thereby promoting their proliferation and migration and impairing lysosomal function [ 105 , 106 ]. High ERβ expression in stromal cells from ectopic endometriotic lesions activates NF-κB signaling, leading to increased CCL2 secretion and subsequent recruitment of macrophages, which in turn promote the proliferation of these stromal cells [ 107 ]. Women with endometriosis have higher TGF-β1 levels in peritoneal fluid, serum, ectopic endometrium, and peritoneum [ 108 ]. The expression of TGF-β by peritoneal macrophages may also regulate integrin expression in ESCs and peritoneal mesothelial cells, facilitating the attachment of ectopic endometrial cells to the peritoneum [ 108 ]. Immunohistochemistry (IHC) shows that, compared with controls, the expression of IL-24, IL-20R1, and IL-20R2 is reduced in both eutopic and ectopic endometrium. In vitro, IL-24 suppresses the viability and invasiveness of ESCs; these inhibitory effects can be reversed either by neutralizing IL-24 with specific antibodies or by macrophage co-culture, which downregulates IL-24/IL-20R1 (but not IL-22R1) [ 109 ]. Together with the estrogen-mediated repression of IL-24, these findings suggest that macrophage–ESC crosstalk removes an IL-24–mediated brake on ESC growth and invasion; however, in vivo confirmation is still needed [ 109 ]. Endometriotic lesions consist of complex cellular components, where angiogenesis plays a crucial role in the establishment, survival, and growth of ectopic lesions [ 2 ]. Bacci et al. found that depleting peritoneal macrophages in mice with endometriosis reduces angiogenesis in endometriotic lesions, highlighting their key role in this process [ 110 ]. Yosuke et al. found that CD206 + macrophage depletion in mice reduced VEGFA and TGFβ1 mRNA levels and CD31 expression, suggesting that CD206 + macrophages promote angiogenesis in endometriosis-like lesions [ 111 ]. Kazuki et al. elucidated through experiments that host-derived cells, particularly CD11b + macrophages, use VEGFR1 signaling to secrete basic fibroblast growth factor (bFGF), which plays a critical role as a driving factor in the growth and angiogenesis of endometrial implants [ 70 ]. Increased VEGF-C and VEGF-D expression in ectopic lesions correlates with lymphangiogenesis in endometriosis patients. Kyoko et al. found that transplanting endometrial fragments lacking VEGFR1’s tyrosine kinase domain into host mice of the same genotype (TK→TK) reduced ectopic lesion size and lymphangiogenesis [ 112 ]. Endometriosis patients exhibit iron overload in the peritoneal cavity, with higher levels correlating with disease severity [ 113 ]. Iron overload is linked to oxidative stress, inflammation, cellular damage, reduced fertility, and malignant transformation [ 114 ]. Iron-laden macrophages, containing hemosiderin, are prevalent in endometriosis and derive iron from phagocytosis of senescent red blood cells or hemoglobin–haptoglobin complexes [ 115 , 116 ]. Heme oxygenase metabolizes hemoglobin, releasing iron, leading to oxidative stress, tissue injury, and ectopic lesion proliferation [ 117 , 118 ]. ROS regulates NF-κB, inducing cytokines and factors that promote adhesion, proliferation, and angiogenesis of ectopic lesions [ 114 , 119 ]. Interestingly, macrophages can secrete IL-33, which upregulates SLC7A11 in ESCs through the p38/JNK/ATF3 pathway [ 120 ]. This upregulation inhibits ferroptosis in ESCs, thereby contributing to the development of endometriosis [ 121 ]. It is noteworthy that IL-33 can also induce anti-inflammatory polarization of macrophages and collaborate with hemoglobin to promote the generation of mature splenic red pulp macrophages (RPMs) through activation of the MyD88 adaptor protein and ERK1/2 kinase downstream of IL-33 receptor IL1RL1, which is important for iron cycling and iron homeostasis [ 122 ]. Ectopic lesions in endometriosis are considered chronic wounds that undergo epithelial-mesenchymal transition (EMT), fibroblast differentiation, myogenesis, and fibrosis, contributing to pain and infertility [ 123 , 124 ]. In endometriosis, elevated sphingosine-1-phosphate (S1P) levels in peritoneal fluid have been associated with a shift toward an M2-like macrophage phenotype accompanied by IL-6 and COX-2 upregulation. This seemingly paradoxical pro-inflammatory profile emphasizes the conceptual limitations of the binary M1/M2 paradigm and suggests that M2-like macrophages within the endometriotic microenvironment can also acquire pathogenic roles, contributing to both inflammation and fibrosis [ 125 , 126 ]. In a C57BL/6J background CD11b-DTR transgenic mouse model of endometriosis, depletion of macrophages by diphtheria toxin significantly attenuates lesion fibrosis, whereas adoptive transfer of M2a macrophages restores and further exacerbates fibrotic progression [ 127 ]. IL-4 activates M2a macrophages, triggering TGF-β1 production and the TGF-β1/Smad3 signaling pathway, promoting EMT [ 124 , 127 ]. Similarly, IL-33, a damage-associated alarmin markedly upregulated in the oxidative-stress-rich and estrogen-high microenvironment of ectopic lesions, signals via its receptor ST2 in endometrial cells to activate Wnt/β-catenin, thereby promoting EMT and fibrotic phenotypes [ 128 – 130 ]. IL-33 promotes the polarization of peritoneal macrophages toward an M2 phenotype, during which these macrophages upregulate IL-1β expression. In turn, IL-1β stimulates endometrial stromal cells to produce more IL-33, establishing an inflammatory–immune positive feedback loop that may exacerbate lesion progression [ 130 ]. In mouse models, IL-33 deletion significantly reduces lesion volume [ 131 ]. Macrophages in endometriosis also produce key mediators like TGF-β1, IL-6, and IL-10 that enhance fibrosis [ 132 ]. Recent studies indicate that patients with endometriosis have elevated levels of IL-10 in their peritoneal fluid and serum [ 133 ]. Consistently, in vitro co-culture of ectopic endometrial stromal cells (ESCs) isolated from endometriotic lesions with macrophages markedly increases IL-10 secretion, which subsequently attenuates NK-cell cytotoxicity [ 16 , 133 , 134 ]. In a syngeneic surgical endometrial transfer model of endometriosis in C57BL/6 mice, perilesional administration of recombinant IL-10 significantly enhanced angiogenesis and increased lesion size, whereas local blockade of IL-10 activity with neutralizing antibodies attenuated these effects [ 134 ]. IL-10 promotes fibrosis through sustained STAT3 activation, inducing a pro-fibrotic phenotype in ESCs, marked by increased proliferation, collagen synthesis, and gel contraction [ 135 ]. Additionally, IL-6-induced STAT3 activation is linked to fibrosis in endometriosis [ 136 ]. The presence of nerve fibers in endometriotic lesions correlates with pain and the severity of dysmenorrhea in patients with peritoneal endometriosis [ 137 ]. Peritoneal lesions exhibit increased sensory, sympathetic, and parasympathetic nerve fibers compared to normal peritoneum [ 138 ]. Macrophages can migrate towards nerve fibers, and estradiol has been shown to enhance this interaction in peritoneal endometriosis [ 139 ]. Colony-stimulating factor 1 (CSF-1) and MCP-1, CCL-2, secreted by nerve fibers, can induce macrophage migration [ 139 ]. Additionally, leukemia inhibitory factor (LIF), IL-1α, IL-1β, and pancreatitis-associated protein III (PAP-III) have also been shown to mediate the recruitment of macrophages to nerve fibers [ 140 ]. Macrophages in endometriosis secrete elevated levels of neurotrophic factors such nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), neurotrophin-3 (NT-3), and NT-4, promoting neurogenesis and contributing to lesion progression [ 141 , 142 ]. Additionally, macrophage-derived insulin-like growth factor-1 (IGF-1) enhances neurogenesis and neural sensitization; blocking its receptor with linsitinib reverses pain behaviors in mouse models [ 143 ]. Macrophage polarization also influences nerve fiber development, with M2-like macrophages secreting VEGF-A to facilitate Schwann cell migration and peripheral axon regeneration [ 144 ]. Single-cell RNA sequencing (scRNA-seq) and spatial transcriptomics (ST) have begun to reshape our understanding of macrophage biology in endometriosis (EM). Unlike conventional bulk RNA-seq, scRNA-seq provides high-resolution delineation of rare subpopulations, differentiation trajectories, and putative transcriptional regulators, while ST preserves the in situ architecture needed to infer cell–cell communication and niche effects. Across studies, macrophages (MΦ) in EM display marked heterogeneity that defies the binary M1/M2 framework, instead occupying a dynamic continuum spanning immunosuppressive, pro-inflammatory, and tissue-repair states. Treatment exposure, lesion compartment, and sampling window can modulate these states and complicate cross-study comparisons. In this framework, ontogeny (resident vs. blood-derived), lesion type (e.g., ovarian, peritoneal, deep infiltrating), and spatiotemporal context jointly shape MΦ phenotypes that may contribute to lesion formation, immune escape, chronic inflammation, and tissue remodeling (Table 2 ). Table 2 Single-cell and spatial transcriptomic studies revealing macrophage subtypes and functional characteristics in endometriosis Study (Author, Year) Sample Source and Type (n) Cell Number & Platform Macrophage Subtypes (Markers + Function) Key Functions/Pathways Main Findings Tan et al., 2022 Human, n=14 (Ctrl, EuE, EcP, EcPA, EcO; ASRM II–IV; most on hormonal therapy) 108,497 cells; 10X scRNA-seq • MΦ1-LYVE1 (FOLR2, LYVE1; immune tolerance/angiogenesis); • MΦ2-peritoneal (FOLR2, ICAM2; peritoneal resident); • MΦ3-APOE (FOLR2, APOE; lipid metabolism); • MΦ4-infiltrated (CLEC5A, CCR2, VEGFA; blood-derived); • MΦ5-activated (mixed; activated phenotype) Immune tolerance with down-modulation of antigen presentation; ↑ angiogenesis; ↑ neuroimmune remodeling Perivascular MΦ1-LYVE1 dominates ovarian lesions; MΦ4-infiltrated scarce in EcO; lesion-proximal niches may dampen recruitment/activation of inflammatory blood-derived MΦ Hou et al., 2025 Human, n=6, 9 tissue samples (PEM, OEM, DIE; hormone-naïve) 65,185 cells; 10X scRNA-seq • MΦ-C1q (C1QA/B/C, CD163; munosuppressive M2-like); • MΦ-CD3 (CD3D/E; TCR⁺ interaction); • MΦ-activated (SPP1, SPARC; nflammation/remodeling); • MΦ-NRN1 (NRN1; neuroregulation); • MΦ-derived myofibroblast (APOE, COL1A1; fibrotic phenotype) Immunosuppression; ↑ fibrosis; TCR⁺ interaction OEM shows strong immune suppression; discovery of myofibroblast-like macrophages Wu et al., 2025 Human, n=8 (3 OE cyst wall + 2 healthy endometrium; +3 normal endometrium from GSE179640 ) 54,000 cells; 10X scRNA-seq • SELL⁺ Mono (migratory precursor); • CX3CR1⁺ Mono (circulating precursor); • OLR1⁺ Mono (inflammatory precursor); • TREM2⁺ Macro (immunosuppressive/repair); • CCL4⁺ Macro (pro-inflammatory); • MHCII^hi Macro (antigen presentation); • IGFBP5⁺ Macro (CXCL2/3/8↑; pro-inflammatory & repair); • MKI67⁺ Macro (proliferative) IGFBP5⁺ pro-inflammation; ↑ CXCL2/3/8; promotes endothelial migration and epithelial proliferation IGFBP5⁺ macrophages expanded, driving inflammation and epithelial/endothelial repair Zou et al., 2021 Human, n=2 (peritoneal fluid:EMs vs Ctrl; 1 each for scRNA-seq) 17,530 cells; 10X scRNA-seq • Phagocytic (MARCO, CD163); • Pro-inflammatory (VCAN, S100A8/A9); • Chemotactic (CCL2/13/18, CXCL12); • Antigen-presenting (MHC-II genes); • Adhesion/fibrotic (SPP1, CD9); • TCR⁺ (TRBC1/2, CD3); • Proliferative (MKI67, CDK1) Phagocytosis defect; pro-inflammation; ↑ chemotaxis; ↑ fibrosis Defective phagocytosis → impaired clearance of retrograde debris, promoting lesion persistence Henlon et al., 2024 Mouse model (experimental EMs), CD45⁺ cells from lesion, peritoneal lavage, menstrual-like endometrium Pooled samples; 10X scRNA-seq • Peritoneal macrophages: LpM1, LpM2, Proliferating LpM, Transitory LpM, SpM • Lesion-resident macrophages (LRM1–5): ∙ LMono (monocyte-like) ∙ LRM2 (immune tolerance/neuroinflammation) ∙ LRM3 (SAM-like: Spp1⁺, Trem2⁺) ∙ LRM4 (TAM-like: Folr2⁺, Ccl8⁺) ∙ LRM5 (recently infiltrated) • Endometrial-derived clusters: ∙ EM (monocyte-like) ∙ AP (antigen-presenting) ∙ EM/LRM1 (proliferating) ECM remodeling; angiogenesis/neurogenesis; immune tolerance Folr2⁺ TAM-like macrophages promote disease; monocyte-derived LpM protective; TAM/SAM conserved across species Burns et al., 2025 Human, superficial peritoneal lesions & paired eutopic endometrium scRNA-seq + spatial transcriptomics No new subtypes; emphasis on epithelium–macrophage axis ↑ Inflammatory signaling (TNFα/NF-κB, IL2–STAT5); ↑ Complement C3; ↑ MHC-II antigen presentation; ↑epithelium-macrophage crosstalk significantly enhanced epithelium–macrophage signaling; drives immune reprogramming & lesion persistence MΦ Macrophage, Ctrl Control endometrium, EuE Eutopic endometrium, EcP Peritoneal ectopic lesion, EcPA Adjacent peritoneal tissue, EcO Ovarian ectopic lesion, OE Ovarian endometrioma, EMs Endometriosis, PEM Peritoneal endometriosis, OEM Ovarian endometriosis, DIE Deep infiltrating endometriosis, LpM Large peritoneal macrophages, SpM Small peritoneal macrophages, LRM Lesion-resident macrophage, TAM Tumor-associated macrophage, SAM Scar-associated macrophage, scRNA-seq Single-cell RNA sequencing, TCR T-cell receptor, ECM Extracellular matrix, ↑ Upregulated Single-cell and spatial transcriptomic studies revealing macrophage subtypes and functional characteristics in endometriosis • MΦ1-LYVE1 (FOLR2, LYVE1; immune tolerance/angiogenesis); • MΦ2-peritoneal (FOLR2, ICAM2; peritoneal resident); • MΦ3-APOE (FOLR2, APOE; lipid metabolism); • MΦ4-infiltrated (CLEC5A, CCR2, VEGFA; blood-derived); • MΦ5-activated (mixed; activated phenotype) Immune tolerance with down-modulation of antigen presentation; ↑ angiogenesis; ↑ neuroimmune remodeling Perivascular MΦ1-LYVE1 dominates ovarian lesions; MΦ4-infiltrated scarce in EcO; lesion-proximal niches may dampen recruitment/activation of inflammatory blood-derived MΦ • MΦ-C1q (C1QA/B/C, CD163; munosuppressive M2-like); • MΦ-CD3 (CD3D/E; TCR⁺ interaction); • MΦ-activated (SPP1, SPARC; nflammation/remodeling); • MΦ-NRN1 (NRN1; neuroregulation); • MΦ-derived myofibroblast (APOE, COL1A1; fibrotic phenotype) Immunosuppression; ↑ fibrosis; TCR⁺ interaction • SELL⁺ Mono (migratory precursor); • CX3CR1⁺ Mono (circulating precursor); • OLR1⁺ Mono (inflammatory precursor); • TREM2⁺ Macro (immunosuppressive/repair); • CCL4⁺ Macro (pro-inflammatory); • MHCII^hi Macro (antigen presentation); • IGFBP5⁺ Macro (CXCL2/3/8↑; pro-inflammatory & repair); • MKI67⁺ Macro (proliferative) IGFBP5⁺ pro-inflammation; ↑ CXCL2/3/8; promotes endothelial migration and epithelial proliferation • Phagocytic (MARCO, CD163); • Pro-inflammatory (VCAN, S100A8/A9); • Chemotactic (CCL2/13/18, CXCL12); • Antigen-presenting (MHC-II genes); • Adhesion/fibrotic (SPP1, CD9); • TCR⁺ (TRBC1/2, CD3); • Proliferative (MKI67, CDK1) Phagocytosis defect; pro-inflammation; ↑ chemotaxis; ↑ fibrosis • Peritoneal macrophages: LpM1, LpM2, Proliferating LpM, Transitory LpM, SpM • Lesion-resident macrophages (LRM1–5): ∙ LMono (monocyte-like) ∙ LRM2 (immune tolerance/neuroinflammation) ∙ LRM3 (SAM-like: Spp1⁺, Trem2⁺) ∙ LRM4 (TAM-like: Folr2⁺, Ccl8⁺) ∙ LRM5 (recently infiltrated) • Endometrial-derived clusters: ∙ EM (monocyte-like) ∙ AP (antigen-presenting) ∙ EM/LRM1 (proliferating) ECM remodeling; angiogenesis/neurogenesis; immune tolerance Folr2⁺ TAM-like macrophages promote disease; monocyte-derived LpM protective; TAM/SAM conserved across species ↑ Inflammatory signaling (TNFα/NF-κB, IL2–STAT5); ↑ Complement C3; ↑ MHC-II antigen presentation; ↑epithelium-macrophage crosstalk significantly enhanced epithelium–macrophage signaling; drives immune reprogramming & lesion persistence MΦ Macrophage, Ctrl Control endometrium, EuE Eutopic endometrium, EcP Peritoneal ectopic lesion, EcPA Adjacent peritoneal tissue, EcO Ovarian ectopic lesion, OE Ovarian endometrioma, EMs Endometriosis, PEM Peritoneal endometriosis, OEM Ovarian endometriosis, DIE Deep infiltrating endometriosis, LpM Large peritoneal macrophages, SpM Small peritoneal macrophages, LRM Lesion-resident macrophage, TAM Tumor-associated macrophage, SAM Scar-associated macrophage, scRNA-seq Single-cell RNA sequencing, TCR T-cell receptor, ECM Extracellular matrix, ↑ Upregulated Yuliana Tan and colleagues profiled eutopic and ectopic tissues from 14 participants (most receiving hormonal therapy at sampling)—control eutopic endometrium (Ctrl), eutopic endometrium from patients (EuE), peritoneal endometriotic lesions (EcP), adjacent peritoneum (EcPA), and ovarian lesions (EcO)—generating 108,497 single-cell transcriptomes and resolving five major cell classes with 58 subclusters [ 17 ]. Within the myeloid compartment, MΦ were partitioned into five subsets: three tissue-resident–like states (MΦ1-LYVE1, MΦ2-peritoneal, MΦ3-APOE), characterized by high expression of the embryonic tissue-resident marker FOLR2 [ 145 , 146 ]; MΦ2-peritoneal was restricted to the peritoneum and expressed ICAM2 [ 147 , 148 ]; MΦ4-infiltrated (blood-derived; CLEC5A, CCR2, VEGFA); and MΦ5-activated, likely arising from both resident and infiltrating lineages. Functionally, MΦ1-LYVE1 and MΦ5-activated were enriched in EuE and lesions. MΦ1-LYVE1, dominant in ovarian lesions, localized perivascularly and expressed immune-tolerance/angiogenic programs (e.g., VSIG4, IL10, THBS1, IGF1) with down-modulation of antigen presentation, consistent with roles in immune tolerance, neovascularization, and neuroimmune remodeling [ 146 ]. By contrast, MΦ4-infiltrated was scarce in EcO and showed more immune-tolerant signatures in patient EuE than in Ctrl EuE—suggesting that lesion-proximal niches may dampen recruitment or activation of inflammatory blood-derived MΦ and thereby favor local persistence [ 149 ]. Similarly, Hou and colleagues performed scRNA-seq on 65,185 cells from six untreated patients with peritoneal (PEM), ovarian (OEM), and deep infiltrating endometriosis (DIE), revealing pronounced MΦ heterogeneity and defining five subsets: an immunosuppressive C1q + cluster (MΦ-C1q; C1QA, C1QC, CD163, MRC1), TCR + macrophages (MΦ-CD3), an activated SPP1 + subset (MΦ-activated), a neuro-modulatory NRN1 + subset (MΦ-NRN1), and a myofibroblast-like subset (MΦ-derived myofibroblast; COL1A1+, APOE+) [ 150 ]. In OEM, the microenvironment showed reduced NK and CD8 + T cells and increased plasma cells, consistent with enrichment of immunosuppressive and tissue-remodeling MΦ across lesion types in this untreated cohort [ 150 ]. Focusing on ovarian endometriotic cyst (OE) tissue, Wu and colleagues profiled samples from eight subjects (OE cyst walls n  = 3; normal endometrium n  = 2; plus integrated public data GSE179640 , n  = 3) and identified eight myeloid states: SELL + monocytes, CX3CR1 + monocytes, OLR1 + monocytes, TREM2 + MΦ, CCL4 + MΦ, MHC-II^hi MΦ, IGFBP5 + MΦ, and MKI67 + MΦ [ 151 ]. Except for IGFBP5 + MΦ, the relative abundance of all subsets decreased in OE, suggesting a dysregulated MΦ compartment. IGFBP5 + MΦ showed robust pro-inflammatory modules (e.g., CXCL2/3/8) together with TNFRSF12A, linking inflammatory initiation to wound-healing programs. Trajectory analysis indicated monocyte-to-IGFBP5 + MΦ differentiation accompanied by up-regulation of NUPR1, MDK, HTRA1, SERPINF1/2, and ID4, with GO terms implicating endothelial migration and epithelial proliferation—consistent with a context-dependent pro-inflammatory yet reparative role in OE. Compared with Tan and Hou, the differences in Wu’s findings may reflect distinct temporal windows and anatomical contexts, underscoring the high spatiotemporal plasticity of the EM MΦ lineage and suggesting that their immune phenotypes shift with disease stage. To characterize free-floating peritoneal MΦ, Zou and colleagues performed scRNA-seq on peritoneal fluid from EM patients and controls and identified seven functionally distinct MΦ subsets: phagocytic (MARCO, CD163, C1QA/B), pro-inflammatory (VCAN, LYZ, S100A8/A9), chemotactic/lipid-regulatory (CCL2, CCL13, CXCL12, APOE), antigen-presenting (MHC-II genes), adhesion/fibrosis-associated (SPP1, CD9), TCR + MΦ (TRBC1/2, CD3D/E, ZAP70), and proliferative (MKI67, CDK1) [ 58 ]. Pseudotime analyses reconstructed a continuum from antigen-presenting (monocyte-like) to pro-inflammatory states, culminating in chemotactic and phagocytic phenotypes, with adhesion/fibrosis-associated MΦ persisting across the trajectory and enriched in later stages. Functional assays showed impaired phagocytosis alongside heightened inflammatory, antigen-presenting, and fibrotic programs in EM, supporting a model in which defective clearance of retrograde menstrual debris contributes to lesion establishment and maintenance.Notably, the TCR + and proliferative MΦ represented newly identified subsets in this study. Using a mouse EM model, Henlon and colleagues performed scRNA-seq on CD45 + immune cells from lesions, peritoneal fluid, and “menstrual-like” endometrium, mapping lineage relationships and functional evolution among endometrium-derived macrophages (MΦ), lesion-resident macrophages (LRM), and large/small peritoneal MΦ (LpM/SpM) [ 152 ]. Endometrium-derived MΦ comprised three groups: a predominant Il1b+/Ly6c2+/S100a8/a9 + pro-inflammatory cluster (monocyte-like), an antigen-presenting cluster (Cd209a, Slamf7, Cd74) of mixed tissue origin, and a proliferative EM/LRM1 cluster expressing cell-cycle genes present in both lesion and eutopic sites. In the peritoneum, MΦ segregated into SpM (H2-Aa, Ccr2, Retnla) and four LpM subsets (LpM1, LpM2, proliferating LpM, transitory LpM); LpM1 increased and LpM2 decreased in Endo-Ovx mice, indicating microenvironmental remodeling. LRMs further resolved into five subsets: LMono (pro-inflammatory, monocyte-like), LRM2 (immune tolerance/neuroinflammation), LRM3 (SAM-like; scar/repair; Spp1+), LRM4 (TAM-like; angiogenesis/hypoxia/immune suppression; Gas6+; the most abundant pro-disease subset), and LRM5 (recent peritoneal influx; early repair/pain modulation). Lineage tracing and qPCR confirmed that SAM-like and TAM-like LRMs arise predominantly from endometrium and monocytes rather than resident LpM, and cross-species immunofluorescence detected Gas6+ (TAM-like) and Spp1+ (SAM-like) MΦ in human lesions. Together, these data support the view that precursor origin is a determinant of LRM fate. While conceptually informative, interspecies differences warrant cautious extrapolation to human disease. Integrating scRNA-seq with ST, Burns and colleagues compared superficial peritoneal lesions with paired EuE and highlighted a central role for MΦ in lesional immunity [ 153 ]. Lesion MΦ exhibited up-regulation of inflammatory pathways (e.g., TNFα/NF-κB and IL2–STAT5) together with down-modulation of EMT/myogenesis programs, yielding a mixed pro-inflammatory and reparative transcriptional state. ST-guided ligand–receptor inference (e.g., CellChat) indicated the strongest autocrine MΦ–MΦ signaling within lesions, followed by markedly enhanced epithelial–MΦinteractions relative to EuE, dominated by epithelial-derived complement (C3) and MHC-II signaling. These data are consistent with a model in which lesional epithelium actively reprograms MΦ immunity to sustain chronic inflammation while maintaining tissue-repair/tolerance circuits. Macrophages play a critical role in endometriosis development, making them an attractive target for immune-based therapies. Strategies to modulate macrophage function include: [ 1 ] enhancing efferocytosis/phagocytosis by targeting immune checkpoints like the CD47-SIRPα axis [ 2 ], inhibiting macrophage recruitment via CCL-CCR2 pathways [ 3 ], repolarizing macrophages from the M2-like tissue-repairing phenotype to the pro-inflammatory M1-like phenotype, and [ 4 ] blocking pro-inflammatory cytokines such as IL-1β and IL-33 (see Fig. 4 ). Fig. 4 Strategies for immunotherapy targeting macrophages The main strategies currently in clinical use or under evaluation can be categorized into four types: ( a ) repolarization of macrophages from a tissue-repairing (M2) to a pro-inflammatory (M1) phenotype; ( b ) inhibition of pro-inflammatory cytokines, such as IL1β and IL33; ( c ) inhibition of macrophage recruitment, for example, by blocking CCL2 or CCR2; and ( d ) modulation of key immune checkpoints, such as the CD47-SIRPα axis, to enhance macrophage phagocytic capacity. (Created by Figdraw ( http://www.figdraw.com )) Strategies for immunotherapy targeting macrophages The main strategies currently in clinical use or under evaluation can be categorized into four types: ( a ) repolarization of macrophages from a tissue-repairing (M2) to a pro-inflammatory (M1) phenotype; ( b ) inhibition of pro-inflammatory cytokines, such as IL1β and IL33; ( c ) inhibition of macrophage recruitment, for example, by blocking CCL2 or CCR2; and ( d ) modulation of key immune checkpoints, such as the CD47-SIRPα axis, to enhance macrophage phagocytic capacity. (Created by Figdraw ( http://www.figdraw.com )) The reduced phagocytic capacity of macrophages is a key factor in the pathogenesis and progression of endometriosis, and various mechanisms can impair this function. One notable example is the interaction between the immune checkpoint CD47 and SIRPα [ 154 , 155 ]. This “don’t eat me” signal allows ectopic endometrial cells to evade macrophage-mediated clearance and promoting ectopic endometrial cell survival and proliferation [ 101 , 156 ]. Blocking this interaction enhances macrophage-mediated clearance and promotes apoptosis of endometrial cells [ 100 ]. Clinically, the humanized anti-CD47 antibody Hu5F9-G4 (magrolimab) disrupts the CD47–SIRPα axis in oncology; however, on-target hematologic toxicities (e.g., anemia, thrombocytopenia) and the lack of endometriosis-specific trials warrant caution when extrapolating to EM [ 157 ]. In endometriosis, CCL2 is upregulated in the lesion microenvironment—especially in stromal cells. ERβ–NF-κB signaling drives CCL2 to recruit CCR2 + monocytes/macrophages, which in turn enhance stromal proliferation and clonogenicity. In parallel, MCP-1 activates integrin-linked kinase (ILK) at Ser246 within endometriotic cells, directly promoting migration, adhesion, colonization, and invasion and amplifying inflammation with greater resident/circulating macrophage infiltration; these effects are reversed and T-regulatory/T-helper balance restored by the ILK inhibitor CPD22. Additionally, mast-cell Mas-related G protein–coupled receptor X2 (MRGPRX2)–CCL2–CCR4 signaling promotes EMT [ 107 , 158 , 159 ]. These mechanisms provide a biological rationale for targeting the CCL2 axis. Early oncology trials also support feasibility: the small-molecule CCR2 antagonist PF-04136309 showed safety and activity when combined with FOLFIRINOX in a phase 1b pancreatic cancer study, and the CCR2 monoclonal antibody MLN1202 has been evaluated in phase I/II trials, including cohorts with bone metastases [ 160 ]. Importantly, murine endometriosis models indicate non-equivalent functions by macrophage origin: endometrium-derived macrophages promote lesions, whereas monocyte-derived macrophages can constrain lesion development [ 86 ]. Thus, indiscriminate blockade of CCL2/CCR2 may be counterproductive in endometriosis. A more nuanced approach would suppress CCL2 at its source or downstream (e.g., ERβ–NF-κB or ILK) and/or modulate mast-cell–driven EMT, while preserving—and ideally harnessing—the potentially protective monocyte-derived macrophage compartment. In endometriosis, macrophages often display M2-like features with attenuated efferocytosis, facilitating lesion persistence [ 11 ]. Reprogramming toward pro-resolving, high-efferocytosis states (with selective inflammatory activation as needed) may be beneficial. Studies in cancer models demonstrate that re-education of tumor-associated macrophages (TAMs) can slow tumor growth by activating NK and T cells [ 161 ]. In the study by Xue et al., poly (lactic-co-glycolic acid) (PLGA) nanoparticles coated with bacterial outer membrane vesicles (OMVs) (OMV-NPs) preserved the intrinsic immunostimulatory properties of OMVs and were efficiently internalized by peritoneal macrophages. This uptake reprogrammed macrophages from the pro-fibrotic M2 phenotype to the M1 phenotype, leading to reduced TGF-β1 production and suppression of myofibroblast activation, which collectively attenuated fibrosis and significantly inhibited the progression of endometriosis [ 162 ]. Members of the IL-1 cytokine family (including IL-33 and IL-1β) and their receptors appear to be effective targets for immunotherapy in endometriosis. Evidence from mouse models suggests that monoclonal antibodies against these ligands and inhibitors of their downstream MyD88 signaling pathway can reduce the volume of ectopic lesions [ 131 ]. In a rat model of endometriosis, recombinant IL-2 reduces lesion size by inducing the recruitment of activated leukocytes, including lymphocytes, macrophages, NK cells, and dendritic cells, to endometriotic-like lesions [ 163 ].

Endometriosis is a multifactorial, estrogen-dependent condition in which hormonal signaling, inflammation, fibrosis, oxidative stress, angiogenesis, and immune dysregulation intersect [ 1 , 2 , 164 , 165 ], Although retrograde menstruation remains a central etiologic hypothesis, increasing evidence highlights the critical contribution of immune alterations—particularly within the macrophage compartment—to lesion establishment and persistence [ 160 , 166 ]. Single-cell and spatial transcriptomic studies now refine this view, showing that macrophages occupy a continuum of states rather than a binary M1/M2 pattern, and that phenotype is shaped by ontogeny (resident vs. blood-derived), lesion type, and local niches [ 167 , 168 ]. In healthy women, macrophages in the eutopic endometrium facilitate tissue repair and maintain a sterile environment [ 76 ]. In contrast, in endometriosis, these macrophages exhibit impaired phagocytic capacity, leading to inadequate clearance of menstrual debris and enhanced survival of refluxed endometrial cells that seed ectopic lesions [ 83 ]. Although this dysfunction has previously been ascribed to “M1-like” activity, the strict M1/M2 dichotomy is overly simplistic. Consistent with the continuum model, recent single-cell studies demonstrate that macrophages occupy a continuum of activation states, with altered expression of phagocytosis-related molecules (e.g., SIRPα, CD36, CD91) observed in patients, and increased production of pro-inflammatory cytokines such as TNF-α and IL-1β in both human and mouse models [ 67 ]. These alterations weaken immune surveillance while sustaining local inflammation. Furthermore, macrophage phenotypes appear to shift with disease progression, from predominantly pro-inflammatory profiles in early disease to more immunosuppressive, alternatively activated states in advanced stages. Such dynamic remodeling highlights the multifaceted roles of macrophages in maintaining inflammation, impairing fertility, and driving tissue remodeling in endometriosis [ 68 , 74 ]. Within ectopic lesions, single-cell atlases describe heterogeneous macrophage subsets, including perivascular, immunosuppressive/tissue-remodeling, and proliferative states [ 17 ]. Spatial analyses highlight epithelial–macrophage crosstalk (e.g., complement/MHC-II signaling) and perivascular niches associated with mixed inflammatory and reparative transcriptional programs [ 153 ]. Functionally, macrophages can promote angiogenesis and fibrosis, integrate iron/heme handling with oxidative stress, and interact with nerve fibers through neurotrophic factors—mechanisms that plausibly contribute to pain, remodeling, and lesion maintenance. Mouse lineage-tracing studies further demonstrate the importance of precursor origin: endometrium-derived macrophages promote lesion growth, whereas monocyte-derived peritoneal macrophages may constrain it [ 86 ]. However, interspecies differences—such as CCR2 dependence of macrophage recruitment in mice versus more redundant pathways in humans—should be considered when extrapolating findings. Current research has identified significant alterations in the immune microenvironment of endometriosis patients, but several questions remain unanswered. The precise drivers of macrophage polarization within lesions and the regulators of their phenotypic transitions are still unclear. Furthermore, although the importance of macrophages in disease progression is increasingly recognized, their interactions with other immune populations, including T cells and dendritic cells, remain underexplored despite their likely role in shaping immune responses and influencing outcomes [ 68 ]. Interpretation of existing data is further complicated by treatment status, menstrual phase at sampling, tissue dissociation artifacts, and interspecies differences across models. Despite advancements in understanding the immune landscape of endometriosis, current treatment options remain limited. Thus, targeted immunotherapies, particularly those aimed at modulating macrophage function, have emerged as a promising alternative. By selectively targeting immune cells and specific signaling pathways involved in disease progression, immunotherapy could potentially reduce or eliminate endometriotic lesions without the adverse effects of conventional treatments. However, developing these therapies requires precise mapping of the immune microenvironment and identification of key macrophage subsets, receptors, and cytokines that drive disease. These insights inform but do not yet validate macrophage-targeted strategies. Conceptually attractive approaches include: (i) restoring phagocytosis by interrupting “don’t-eat-me” signaling (e.g., CD47–SIRPα); (ii) modulating recruitment through the CCL2/CCR2 axis; (iii) reprogramming macrophages toward pro-resolving states; and (iv) attenuating IL-1 family signaling. Most evidence remains preclinical or early-phase. Notably, anti-CD47 antibodies in oncology trials have caused hematologic toxicities such as anemia and thrombocytopenia, and the non-equivalence of macrophage lineages cautions against indiscriminate blockade [ 169 ]. Future therapeutic design will likely need to be lesion- and lineage-specific (e.g., preserving protective monocyte-derived subsets) and incorporate pharmacodynamic biomarkers to confirm on-target immune effects. Integration of spatial multi-omics with longitudinal sampling may enable precise mapping of macrophage states and their transitions, providing a stronger foundation for rational immunotherapy design.