Perturbations of the endometrial immune microenvironment in endometriosis and adenomyosis: their impact on reproduction and pregnancy

Despite strong evidence indicating an increased incidence of infertility in women with endometriosis, as well as higher rates of infertility, miscarriage, and pregnancy complications in women with adenomyosis, research on the underlying mechanisms has not kept pace. Immunological adaptation is critical for the success of each stage during reproduction and pregnancy. Women with endometriosis and adenomyosis exhibit significant alterations in endometrial immune cells, compared to healthy controls. Furthermore, these changes are not always consistent between endometriosis and adenomyosis, which may explain clinically different impacts they have on reproduction and pregnancy. Nevertheless, the alterations in endometrial immune cells are somewhat comparable to those in women with RIF and RM. Thus, it is plausible to hypothesize that these changes may contribute to a higher risk of reproductive disorders and pregnancy complications. To advance this field, further studies are needed to characterize endometriosis and adenomyosis-specific endometrial immune cell profiles and to identify immune targets for the development of targeted therapies in the future.

Endometriosis and adenomyosis are two of the common gynaecological disorders, which are often considered together due to similar symptoms and pathophysiology. Both conditions are characterized by the growth of ectopic endometrial tissues: in endometriosis, this tissue is located outside the uterus, while in adenomyosis, it is found within the myometrium. The precise aetiology of both conditions is not yet fully understood, although various theories have been proposed attempting to explain them. Among these, the most widely accepted include Sampon’s theory of retrograde menstruation, which suggests that shedding endometrial tissue travel through the fallopian tubes and enters the peritoneal cavity in cases of endometriosis. In contrast, adenomyosis is thought to involve the infiltration of endometrial basal layer deep into the myometrium due to an absent or altered junctional zone. Importantly, a high degree of association between the two conditions further suggests these two diseases may share a common origin in abnormal eutopic endometrium [ 1 ]. Indeed, several studies have shown cellular and molecular differences in the eutopic endometrium between women with endometriosis or adenomyosis and healthy controls [ 2 , 3 ]. However, most studies did not clearly exclude one condition from the other or specify the comorbidity, making it uncertain how similar the changes in eutopic endometrium are between endometriosis and adenomyosis. Emerging evidence suggests that women with endometriosis and/or adenomyosis have a higher incidence of infertility and pregnancy complications, which may be partially attributed to impaired decidualization due to changes in the eutopic endometrium [ 4 ]. In addition to the defective decidualization of stromal cells, these women exhibit significant alterations in endometrial immune cells, which not only facilitate the disease progression, as previously reviewed [ 5 ], but also negatively impact reproduction and pregnancy [ 6 ]. In healthy women, endometrial immune cells dynamically adapt to the biological events during the preparation of the endometrium for embryo implantation and subsequent decidualization, thereby supporting embryo growth thereafter. Proper immunomodulation is crucial for angiogenic remodeling, maternal immune tolerance to the semi-allogenic fetus, and successful placentation. Disruption and dysfunction of this intricate system can result in infertility, miscarriage, and other pregnancy complications [ 7 ]. However, whether the changes in endometrial immune cells in women with endometriosis increase the risk of infertility and pregnancy complications remains largely unknown. In this review, we aim to examine the endometriosis and adenomyosis-related alterations in endometrial immune cells and discuss their potential contributions to infertility and pregnancy complications. Endometriosis has been reported to affect approximately 5–10% of reproductive-aged women worldwide and increase the risk of female infertility twofold [ 8 ]. The pathophysiology of endometriosis impacts reproduction at nearly every stage, including gametes, fertilization, embryo implantation, and placentation. These effects are further supported by clinical data. Women affected by endometriosis, particularly endometrioma, often exhibit significantly lower ovarian reserve function and fecundity [ 9 ]. Similarly, multiple meta-analyses consistently show that women with endometriosis undergoing in vitro fertilization-embryo transfer (IVF-ET) tend to have significantly lower number of oocytes retrieved, as well as decreased fertilization and implantation rates (Table  1 ). This suggests endometriosis adversely affects both the quantity and quality pf oocytes. In contrast to the effects on oocytes, the influence of endometriosis on pregnancy rate, miscarriage rate and live birth rate is not consistent across these meta-analyses (Table  1 ). Similarly, the effects of endometriosis on pregnancy complications, such as preterm birth, small for gestational age, preeclampsia, cesarean delivery, stillbirth and neonatal death, show variability. However, the increased risk of placenta previa in women with endometriosis is relatively well recognized. A meta-analysis conducted by Gasparri et al. suggested that endometriosis alone may act as an independent risk factor for placenta previa [ 10 ]. Nevertheless, these meta-analyses did not conduct subgroup analysis based on specific types of endometriosis [ 11 , 12 ], such as deep infiltrating endometriosis, which may have a more severe impact on reproductive and pregnancy outcomes. Table 1 Summarized meta-analysis of pregnancy complications in women with endometriosis Reference Publication year Study dates No. of studies No. of cases Mode of conception No. of oocytes retrieved Fertilization rate Implantation rate Pregnancy rate Miscarriage rate Live birth rate Placenta previa Other obstetric complications Barnhart et al. [ 140 ] 2002 1980–1999 22 6760 IVF ↓ a , ↓ d ↓ a , ↓ d ↓ a , ↔ d ↓ a , ↓ d - - - - Harb et al. [ 11 ] 2013 ~ 2012 27 8984 IVF - ↓ b , ↔ c ↓ c , ↔ b ↓ c , ↔ b - ↔ c , ↔ b - - Hamdan et al. [ 12 ] 2015 1980–2014 36 529,454 IVF/ICSI ↓ a , ↓ c - - ↓ a , ↓ c ↔ a ↔ a , ↓ c - - Hamdan et al. [ 20 ] 2015 1980–2014 33 - IVF/ICSI ↓ a - - ↔ a ↔ a ↔ a - - Horton et al. [ 141 ] 2019 1980–2018 63 - ART/ SC ↓ a , ↓ a , ↓ b ↓ b - ↑ a - - Preterm birth ↑ a , cesarean section ↑ a , neonatal unit admission following delivery ↑ a Wang et al. [ 142 ] 2021 ~ 2019 28 2,409,064 ART/ SC - - - - ↑ a - ↑ a Preterm birth ↑ a , gestational hypertension ↑ a , cesarean section ↑ a , preeclampsia ↑ a , placental abruption ↔ a Huang et al. [ 143 ] 2020 ~ 2020 - - ART/ SC - - - - ↑ a (SC) , ↔ a(ART) - ↑ a Preterm birth ↑ a , stillbirth ↑ a , antepartum hemorrhage ↑ a , postpartum hemorrhage ↑ a , placental abruption ↔ a , pre-eclampsia ↔ a , gestational diabetes ↔ a , low birthweight ↔ a , intrauterine growth restriction ↔ a Bruun et al. [ 144 ] 2018 1950–2017 17 - - - - - - - - - Preterm birth ↑ a , small for gestational age ↑ a Lalani et al. [ 145 ] 2018 1990–2017 33 3,280,488 ART/ SC - - - - - - ↑ a (ART, SC) Preterm birth ↑ a (ART, SC) , cesarean section ↑ a (SC) , low birth weight ↑ a (SC) , pre-eclampsia ↑ a (mixed) , gestational diabetes ↑ a (mixed) , gestational cholestasis ↑ a (mixed) , antepartum hemorrhage ↑ a (mixed) Gasparri et al. [ 10 ] 2018 ~ 2018 5 8007 ART - - - - - - ↑ a Placental abruption ↔ a Qu et al. [ 146 ] 2022 ~ 2021 70 - IVF/ICSI ↓ a - ↓ a - - ↔ a , ↔ c ↑ a Postpartum hemorrhage ↑ a , preterm birth ↔ a , preeclampsia ↔ a postpartum hemorrhage ↑ a , small for gestational age ↔ a Nagase et al. [ 147 ] 2022 ~ 2021 28 4,719,258 - - - - - - - - Instrumental delivery ↑ a , cesarean section ↑ a , postpartum hemorrhage ↔ a No., number; ↓, risk significantly decreased; ↑, risk significantly increased; ↔, no difference in risk; -, not applicable; a , endometriosis vs. control; b , mild endometriosis (stage I-II) vs. control; c , severe endometriosis (stage III-IV) vs. control; d , severe endometriosis (stage III-IV) vs. mild endometriosis (stage I-II); IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; ART, assisted reproduction technology; SC, spontaneous conception Summarized meta-analysis of pregnancy complications in women with endometriosis Preterm birth ↑ a , cesarean section ↑ a , neonatal unit admission following delivery ↑ a ↑ a (SC) , ↔ a(ART) No., number; ↓, risk significantly decreased; ↑, risk significantly increased; ↔, no difference in risk; -, not applicable; a , endometriosis vs. control; b , mild endometriosis (stage I-II) vs. control; c , severe endometriosis (stage III-IV) vs. control; d , severe endometriosis (stage III-IV) vs. mild endometriosis (stage I-II); IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; ART, assisted reproduction technology; SC, spontaneous conception The estimated prevalence of adenomyosis is hard to obtain and probably underestimated, ranging from 8.8 to 61.5% of hysterectomy patients [ 13 ]. Women with adenomyosis also experience a higher incidence of infertility (about 24.4% of infertile women over 40 years old and 22% in infertile women less than 40) [ 14 ]. Multiple meta-analyses (Table  2 ) indicate that women with adenomyosis undergoing IVF-ET retrieve a similar number of oocytes compared to those without the condition. However, affected women exhibit significantly lower pregnancy and live birth rates, along with notably higher incidences of miscarriage, preterm birth, preeclampsia, fetal malpresentation, cesarean section, and postpartum hemorrhage [ 15 – 18 ]. These adverse effects are also observed in women with spontaneous conception [ 19 ]. It is clear that adenomyosis poses a substantial obstacle to achieving a healthy pregnancy and warrants increased attention of obstetric care providers. Table 2 Summarized meta-analysis of pregnancy complications in women with adenomyosis Reference Publication year Study dates No. of papers No. of cases Mode of conception No. of oocytes retrieved Implantation rate Pregnancy rate Miscarriage rate Live birth rate Preterm birth Preeclampsia Small for gestational age Other obstetric complications Vercellini et al. [ 15 ] 2014 1999–2012 9 1865 IVF - - ↓ ↑ ↓ - - - - Younes et al. [ 16 ] 2017 ~ 2012 11 2054 IVF - ↓ ↓ ↑ ↓ - - - - Bruun et al. [ 144 ] 2018 1950–2017 4 - - - - - - - ↑ - ↑ - Razavi et al. [ 17 ] 2019 2006–2018 6 9742 - - - - - - ↑ ↑ ↑ Fetal malpresentation ↔ Horton et al. [ 141 ] 2019 1980–2018 11 - ART/ SC - - ↓ ↑ ↓ ↑ ↑ ↑ Cesarean section ↑ Huang et al. [ 143 ] 2020 ~ 2020 - - ART - - - ↑ - - - - - Nirgianakis et al. [ 18 ] 2021 ~ 2020 17 - ART - - ↓ ↑ - ↑ ↑ ↑ Cesarean section↑, fetal malpresentation ↑, postpartum hemorrhage ↑ Cozzolino et al. [ 148 ] 2022 ~ 2020 22 - IVF ↔ - ↓ ↑ ↓ - - - - No., number; ↓, risk significantly decreased; ↑, risk significantly increased; ↔, no difference in risk; -, not applicable; IVF, in vitro fertilization; ART, assisted reproduction technology; SC, spontaneous conception Summarized meta-analysis of pregnancy complications in women with adenomyosis No., number; ↓, risk significantly decreased; ↑, risk significantly increased; ↔, no difference in risk; -, not applicable; IVF, in vitro fertilization; ART, assisted reproduction technology; SC, spontaneous conception Unfortunately, surgical and hormonal treatments do not seem to significantly reverse these adverse effects of endometriosis and adenomyosis on reproduction and pregnancy [ 20 , 21 ]. Therefore, it is crucial to uncover the mechanisms underlying these negative effects to promote the development of novel therapeutic strategies. Macrophages, derived from monocytes, are essential components of the immune system to adapt and respond to various physiological and pathological stimuli. The term “macrophage” comes from the ancient Greek words “makros” (large) and “phagein” (to eat), reflecting their role in engulfing and digesting cellular debris, pathogens, and foreign substances. Macrophages can be broadly classified into two main types based on their ontogeny and function: M1 and M2. M1 macrophages, also known as classically activated macrophages, are pro-inflammatory, and involved in pathogen elimination and tissue damage. They secrete a range of pro-inflammatory cytokines contributing to inflammation and pathogen clearance. In contrast, M2 macrophages, or alternatively activated macrophages, are associated with tissue repair, immune regulation, and the resolution of inflammation. Various markers are used to identify the functions and phenotypes of macrophages, as summarized in Table  3 . These cells are distributed throughout the body, where they maintain tissue homeostasis, regulate inflammation, and orchestrate immune responses. These versatile cells are highly adaptable and can adjust their functions based on the local microenvironment. In the female reproductive tract, they are pivotal for ensuring the proper functioning of the reproductive system, including menstrual cycle changes, embryo implantation, pregnancy maintenance, and the initiation of parturition. Changes in macrophages associated with conditions such as endometriosis and adenomyosis may impact reproduction and pregnancy. Table 3 The markers of human and mouse macrophages Marker Species Structure Function Expression Profile M1/M2/Pan CD68 Human, Mouse Transmembrane glycoprotein Phagocytosis, lysosomal marker Cell surface, intracellular Pan CD80 Human, Mouse Transmembrane protein Co-stimulatory molecule for T-cell activation Cell surface M1 CD86 Human, Mouse Transmembrane protein Co-stimulatory molecule for T-cell activation Cell surface M1 iNOS Human, Mouse Enzyme Produces nitric oxide in response to inflammation Intracellular M1 TNF-α Human, Mouse Cytokine Pro-inflammatory cytokine Soluble M1 CD206 Human, Mouse Transmembrane protein Mannose receptor, endocytosis Cell surface M2 Arg1 Human, Mouse Enzyme Involved in polyamine synthesis, M2 marker Intracellular M2 CD163 Human, Mouse Transmembrane protein Scavenger receptor, anti-inflammatory Cell surface M2 IL-10 Human, Mouse Cytokine Anti-inflammatory cytokine Soluble M2 CD11b Human, Mouse Integrin Phagocytosis, cell adhesion, migration Cell surface Pan F4/80 Mouse Transmembrane glycoprotein Macrophage-specific marker Cell surface Pan MHC II Human, Mouse Transmembrane protein Antigen presentation Cell surface M1 CD14 Human, Mouse Co-receptor LPS recognition, part of the TLR4 complex Cell surface Pan The markers of human and mouse macrophages During the menstrual cycle, macrophages exhibit distinct fluctuations of population and activity within the endometrium. In the proliferative phase, CD68 + or CD163 + macrophages constitute 1–2% of endometrial cells, increasing to 1–5% in the early-to-mid secretory phase, then rising dramatically to 7% in the late secretory phase, and peaking at 6–15% during the pre-menstrual phase [ 22 , 23 ]. The majority of endometrial macrophages are of the M2 phenotype [ 24 ]. These accumulated macrophages are typically found as single cells or in clusters within the glandular lumens of the superficial endometrium [ 25 ], where they play crucial roles in tissue repair and angiogenesis. Furthermore, the remarkable increase in macrophages during the secretory phase supports the endometrium’s transitions into the decidua by promoting stromal cell differentiation into decidual cells (decidualization) through the secretion of factors such as IL-10 and TGF-β. M2 macrophages also play a role in regulating extracellular matrix remodeling, which is essential for successful decidualization and placentation. In contrast, the frequency of CD68 + IL-10 − iNOS + M1 macrophages in the secretory endometrium (lower than 20%) is significantly lower compared to the proliferative phase (about 30%) [ 26 ]. This reduction in M1 macrophages during the secretory phase is crucial for fostering endometrial receptivity by creating a favorable immune environment essential for successful implantation (Fig.  1 ). Fig. 1 Dynamics of endometrial immune cell counts during menstrual cycle in endometriosis and adenomyosis. Compared to healthy controls, women with endometriosis exhibit a significant increase in NK cells, uNK progenitor cells, M1 macrophages, CD8 + T cells, Th1 cells, Th17 cells and γδT cells in the endometrium. Similarly, women with adenomyosis also showed a significant increase in NK cells, M2 macrophages, CD8 + T cells, Th1 cells, Th17 cells and γδT cells in the endometrium. KIRs: killer-immunoglobulin-like receptors. Created with Biorender. com Dynamics of endometrial immune cell counts during menstrual cycle in endometriosis and adenomyosis. Compared to healthy controls, women with endometriosis exhibit a significant increase in NK cells, uNK progenitor cells, M1 macrophages, CD8 + T cells, Th1 cells, Th17 cells and γδT cells in the endometrium. Similarly, women with adenomyosis also showed a significant increase in NK cells, M2 macrophages, CD8 + T cells, Th1 cells, Th17 cells and γδT cells in the endometrium. KIRs: killer-immunoglobulin-like receptors. Created with Biorender. com Decidual macrophages exhibit a spectrum of activation states shaped by the dynamic microenvironment. They are classified into decidual M1-like and M2-like subtypes based on their functional tendencies rather than strict polarization. Decidual M1-like macrophages are primarily involved in immune defense against potential infections, while decidual M2-like macrophages dominate from early pregnancy through the third trimester, playing crucial roles in tissue remodeling, immune tolerance, and promoting trophoblast invasion. During parturition, recognized as an inflammatory event, M1-like macrophages become the predominant subset, secreting pro-inflammatory cytokines and chemokines that promote uterine contractions and the breakdown of decidual tissues, facilitating the expulsion of the fetus and placenta (Fig.  2 ). Fig. 2 Roles of endometrial immune cells in establishing and maintaining early pregnancy. At the maternal-fetal interface, endometrial immune cells contribute to pregnancy in four main ways: ( a ) NK cells, M2-like macrophages and CD8 + T cells regulate trophoblast invasion during early pregnancy through KIRs and cytokine secretion, respectively; ( b ) M2-like macrophages facilitate placental development by regulating tissue remodeling via cytokine secretion; ( c ) NK cells and M2-like macrophages contribute to vascular remodeling through producing cytokines; ( d ) NK cell cytotoxicity is reduced, and there is simultaneous down-regulation of the inflammatory response by macrophages and T cells, thereby promoting immune tolerance. HLA: human leukocyte antigen; KIRs: killer-immunoglobulin-like receptors; VEGF: vascular endothelial growth factor; SDF: stromal cell-derived factor. Created with Biorender. com Roles of endometrial immune cells in establishing and maintaining early pregnancy. At the maternal-fetal interface, endometrial immune cells contribute to pregnancy in four main ways: ( a ) NK cells, M2-like macrophages and CD8 + T cells regulate trophoblast invasion during early pregnancy through KIRs and cytokine secretion, respectively; ( b ) M2-like macrophages facilitate placental development by regulating tissue remodeling via cytokine secretion; ( c ) NK cells and M2-like macrophages contribute to vascular remodeling through producing cytokines; ( d ) NK cell cytotoxicity is reduced, and there is simultaneous down-regulation of the inflammatory response by macrophages and T cells, thereby promoting immune tolerance. HLA: human leukocyte antigen; KIRs: killer-immunoglobulin-like receptors; VEGF: vascular endothelial growth factor; SDF: stromal cell-derived factor. Created with Biorender. com This cyclical and adaptive behavior of macrophages underscores their versatile and indispensable roles in female reproduction, from regulating endometrial receptivity and supporting fetal development to orchestrating the complex process of pregnancy. Endometriosis is associated with significant alterations in the number and function of macrophages in eutopic endometrium [ 27 ]. Research has demonstrated an increased density of macrophages, particularly during the proliferative phase, in women with endometriosis compared to controls [ 28 ]. Additionally, the functional phenotypes undergo significant shifts. In women with endometriosis, the M2 subpopulation is reduced across all phases compared with controls [ 28 ]. The reduced presence of CD163 + macrophages, with an area under the curve (AUC) of 0.833, suggests its potential as a diagnostic marker for endometriosis [ 28 ]. Consistently, RNA sequencing analysis of fluorescence-activated cell-sorted macrophages has revealed an increased presence of pro-inflammatory M1 macrophages in the endometrium of women with endometriosis, compared to healthy controls [ 29 ]. Utilizing mass cytometry, researchers have identified an increased abundance of CD91 + macrophages in the eutopic endometrium of individuals with endometriosis. These macrophages play a crucial role in the efferocytosis of apoptotic cells during endometrial shedding through the calreticulin/CD91 pathway. Although their numbers are elevated in endometriosis, these macrophages exhibit defective phagocytic capacity, potentially due to the altered inflammatory microenvironment, which contributes to the survival of endometrial cells during shedding [ 30 ]. Furthermore, changes in the endometrial macrophages correlate with the stage of endometriosis. M1 macrophages are more prevalent in early-stage endometriosis (stages I–II), while M2 macrophages increase in advanced-stage disease (stages III–IV) [ 31 ]. Paradoxically, despite their typical anti-inflammatory nature, M2 macrophages in the endometriotic endometrium exhibit a pro-inflammatory phenotype, likely driven by the altered microenvironment in endometriosis [ 30 ]. The persistent inflammatory milieu of endometriosis seems to override their typical immunosuppressive nature, driving them to adopt a paradoxical pro-inflammatory behavior [ 30 ]. Pathways enriched in advanced stages, such as TGF-β, PI3K/AKT/mTOR, IFN-γ signaling, along with metabolic reprogramming, may further support this polarization [ 31 ]. However, the underlying mechanism and remain unclear. Altered macrophages in both eutopic and ectopic endometrium may impact infertility in several mechanisms. The increase in M1 macrophages in the eutopic endometrium may contribute to defective endometrial receptivity for embryo implantation as the dominance of M2 macrophages plays an essential role in this process through cytokine regulation and immune cell interactions. However, research specifically addressing the issue is limited. In contrast, more evidence suggests that activated macrophage and their cytokine secretions, including TNF-α, GM-CSF, IL-1, and IL-6, in the peritoneal fluid are pivotal in negatively affecting oocyte quality and impairing embryo development [ 32 , 33 ]. These inflammatory cytokines may impact the environment in the fallopian tube and endometrial cavity, leading to oxidative stress that disrupts fertilization and early embryo development, though the precise mechanisms remain unclear. Additionally, macrophages from the peritoneal fluid of infertile patients with endometriosis had higher sperm phagocytosis compared to those from infertile women without endometriosis, suggesting a direct detrimental effect on fertilization [ 34 ]. This heightened inflammatory status within the pelvic cavity likely plays a central role in reducing fertility [ 35 ]. In addition to infertility, excessive pro-inflammatory M1 macrophages, combined with a weakened anti-inflammatory M2 profile may increase the risk of miscarriage in women with endometriosis by impairing immune tolerance [ 36 ]. This imbalance in macrophages disrupts the modulation of the local immune environment, leading to an overactive inflammatory response that may interfere with crucial processes such as trophoblast invasion and placental development. However, decidual macrophages, different from endometrial macrophages, have distinct roles in supporting early pregnancy [ 37 ]. The specific changes in decidual macrophages in women with endometriosis have yet to be fully revealed (Fig.  3 a). Fig. 3 Possible mechanisms by which alterations in endometrial immune cells in endometriosis/adenomyosis contribute to reproductive failure. ( a ) In endometriosis, embryo implantation may fail due to the downregulation of NK cell activation receptors, resulting in diminished interaction with trophoblast cells. Concurrently, an abnormal increase in the number of uNK progenitor cells, NK cells, and M1 macrophages may lead to defective endometrial receptivity. Additionally, enhanced NK cell cytotoxicity and increased secretion of pro-inflammatory cytokines by M1 macrophages and T cells combine to lead to an excessive inflammatory response. ( b ) In adenomyosis, there is an abnormal increase in M2 macrophages, which may lead to implantation failure by regulating the activation of T cells and B cells. Furthermore, the elevated number of M2 macrophages and NK cells can impair endometrial receptivity. Moreover, increased NK cell cytotoxicity, their recruitment of immune cells, and increased secretion of proinflammatory cytokines by M2 macrophages and T cells collectively drive a pro-inflammatory response that disrupts maternal immune tolerance. HLA: human leukocyte antigen; ESC: endometrial stromal cell; LIF: leukemia inhibitory factor. Created with Biorender. com Possible mechanisms by which alterations in endometrial immune cells in endometriosis/adenomyosis contribute to reproductive failure. ( a ) In endometriosis, embryo implantation may fail due to the downregulation of NK cell activation receptors, resulting in diminished interaction with trophoblast cells. Concurrently, an abnormal increase in the number of uNK progenitor cells, NK cells, and M1 macrophages may lead to defective endometrial receptivity. Additionally, enhanced NK cell cytotoxicity and increased secretion of pro-inflammatory cytokines by M1 macrophages and T cells combine to lead to an excessive inflammatory response. ( b ) In adenomyosis, there is an abnormal increase in M2 macrophages, which may lead to implantation failure by regulating the activation of T cells and B cells. Furthermore, the elevated number of M2 macrophages and NK cells can impair endometrial receptivity. Moreover, increased NK cell cytotoxicity, their recruitment of immune cells, and increased secretion of proinflammatory cytokines by M2 macrophages and T cells collectively drive a pro-inflammatory response that disrupts maternal immune tolerance. HLA: human leukocyte antigen; ESC: endometrial stromal cell; LIF: leukemia inhibitory factor. Created with Biorender. com Several studies have confirmed that macrophages are highly enriched in the eutopic and ectopic endometrium of patients with adenomyosis, aggregating within the superficial endometrial glands [ 38 , 39 ]. Recently, a single-cell RNA sequencing study revealed that immune cells made up 24% of the total cell population in the eutopic endometrium of adenomyosis patients during the follicular phase, with macrophages and monocytes comprising 4% of the total cell population, identified by CD74 and human leukocyte antigen (HLA)-DRA expression [ 40 ]. Compared to women without adenomyosis, the density of CD163 + M2 macrophages in the eutopic endometrium of patients with severe diffuse or local adenomyosis is significantly increased [ 38 , 41 ]. In contrast, immune cells represent only 13% of the cell population in the ectopic endometrial tissue, where macrophages and monocytes are the most abundant, contributing to 6% [ 40 ]. Genes associated with immune response and macrophage activation, such as CD74, HLA-DRB1, HLA-DRA, S100A6, and NFKBIA, are significantly overexpressed in the adenomyotic epithelium [ 40 , 42 ]. These findings indicate that macrophages are activated and may lead to inflammation in adenomyosis, but this issue remains underexplored. The increased presence of M2 macrophages in adenomyosis appears to play a pivotal role in driving the condition’s associated infertility. While M2 macrophages are typically involved in tissue repair and anti-inflammatory processes, their excessive polarization in adenomyosis leads to pathological tissue fibrosis and chronic inflammation, which undermines normal endometrial function. This dysregulated activity in adenomyosis is primarily driven by signals from the eutopic endometrial cells. For instance, the study conducted by Yang et al. [ 43 ] demonstrated that endometrial stromal cells (ESCs) from adenomyotic tissue exhibited elevated IL-6 mRNA expression, leading to M2 polarization. Similarly, Hu et al. illustrated how adenomyotic tissue directly influences macrophage behavior through extracellular vesicles, which also induce M2 polarization [ 44 ]. The polarized macrophages could, in turn, facilitate epithelial-mesenchymal transition, a biological process where epithelial cells lose their cell-cell adhesion properties and gain migratory and invasive characteristics typical of mesenchymal cells [ 44 ]. This transition enables the epithelial cells to break through the basement membrane, promoting the spread of ectopic endometrial tissue into the myometrium, and enhancing the invasiveness of adenomyotic lesions [ 39 , 44 ]. Moreover, M2 macrophages activate signaling pathways such as TGF-β1/Smad3 and IL-6/JAK2/STAT3, promoting cell growth and proliferation within adenomyotic lesions, further contributing to the progression and severity of the condition [ 45 ]. The overactivity of M2 macrophages disrupts the key signaling molecules necessary for embryo implantation, such as leukemia inhibitory factor (LIF), which is crucial for creating an implantation-receptive endometrium [ 46 ]. In a healthy endometrium, the proper immune environment supports optimal LIF production, which regulates surface glycan structures on epithelial cells. However, in adenomyosis, the overactivation of M2 macrophages, combined with the ongoing inflammatory response, may impair LIF production or its regulatory functions, indirectly compromising the endometrial receptivity needed for successful implantation [ 46 , 47 ]. Research has suggested that macrophage-derived-LIF significantly decreased in both the endometrium and uterine flushing fluid during the window of implantation in women with adenomyosis-related infertility [ 46 , 47 ]. Additionally, endometrial macrophages may influence embryo implantation through interactions with other immune cells, such as T cells and B cells, further exacerbating the inflammatory milieu. In adenomyotic endometrium, high expression of HLA class II can activate macrophages, which in turn stimulate T cells to secrete IL-6, IL-8, and IL-10 [ 5 , 48 ]. These cytokines then stimulate B cells to produce immunoglobulins (Ig), creating an immune response that potentially hinder embryo implantation [ 5 ]. These findings suggest that abnormal cytokine secretion and immune cell interactions involving endometrial macrophages may create an immunological ‘vicious circle’ that exacerbates adenomyosis-related infertility [ 49 ]. However, the precise mechanisms by which endometrial macrophages contribute to infertility in adenomyosis remain poorly understood and warrant further investigation (Fig.  3 b). Natural killer (NK) cells are defined as effector innate lymphoid cells produced by progenitor cells in the bone marrow, with the capacity to recognize and eliminate distressed cells [ 50 ]. Human NK cells are typically classified based on the expression of two surface molecules: CD56 and CD16. Consequently, human peripheral NK (pNK) cells can be divided into two main subpopulations: CD56 dim CD16 + pNK cells, which comprise approximately 90% of the population, and CD56 bright CD16 − pNK cells, which account for about 10% [ 51 ]. Unlike the majority of pNK cells, uterine NK (uNK) cells generally express CD56 at high levels while lacking CD16. Furthermore, uNK cells express adhesion molecules that facilitate tissue residency, which phenotypically differ from those of CD56 bright CD16 − pNK cells. Recent studies using single-cell RNA sequencing demonstrate that the heterogeneity of uNK cells is much more complex than a simple classification based on CD56 and CD16 expression [ 52 , 53 ]. However, further studies are needed to validate these new categories. As the predominant immune cells during the implantation window and early pregnancy [ 54 ], uNK cells play a crucial role in embryo implantation and placentation. Disruptions in their density or function have been linked to immune etiology of reproductive failure [ 55 ]. Emerging studies have reported significant alterations in uNK cells in women with endometriosis and adenomyosis, which may increase the risk of reproductive failure. The morphology of uNK cells changes with menstrual cyclicity. In the proliferative phase, uNK cells are characterized by small granules; while in the secretory phase after ovulation, they gradually enlarge and possess larger granules [ 56 ]. Throughout the menstrual cycle, the density of uNK cells among total leukocytes reaches 30–40% in the proliferative phase, increases further to 60% in the secretory phase, peaks at 70% of the total leukocytes in early gestation, and then declines from mid-pregnancy [ 51 , 57 , 58 ]. This variation highlights their importance in embryo implantation and early pregnancy (Fig.  1 ). NK cell receptors (NKRs) regulate NK cell functions, primarily through killer-immunoglobulin-like receptors (KIRs), which are transmembrane receptors featuring both activating and inhibitory isoforms that can interact with fetal HLAs. It is crucial to maintain a functional balance between activating and inhibitory KIRs for embryo implantation and placentation in early pregnancy [ 59 ]. Most uNK and decidual NK (dNK) cells share a unique repertoire of NKR, with a notable preference for KIR2D expression [ 60 ]. However, emerging evidence indicates that dNK cells differ from uNK cells by expressing significantly lower frequencies of KIRs, especially KIR2DS1, KIR2DL2L3S2 and KIR2DL2S2, while exhibiting considerably higher frequencies of activation receptors NKG2D, NKp30, NKp46 and CD244 [ 61 ]. This reduction in KIR expression may be attributed to increased interactions between these receptors and their respective HLA ligands on trophoblasts [ 62 ]. Conversely, the upregulation of receptors on dNK cells could be a result of elevated IL-15 levels derived from stromal cells [ 63 ]. The activation of these upregulated receptors can lead to elevated production of chemokines, cytokines and angiogenic factors, such as vascular endothelial growth factor (VEGF) and stromal cell-derived factor (SDF), thereby facilitating vascular growth and trophoblast invasion [ 64 ] (Fig.  2 ). Taken together, the plasticity of NK cells leads to the heterogeneity of various subtypes, as well as the diversity of functions, which lays an important foundation for successful pregnancy. Similar to healthy women, those with endometriosis exhibit a significant increase in uNK cells during the secretory phase [ 65 ]. The dramatic increase of uNK cells might from three sources: (1) in situ proliferation of mature uNK cells in uterus [ 66 ]; (2) migration of peripheral NK cells [ 67 ]; and (3) maturation and differentiation of NK progenitor cells [ 68 ]. However, patients with severe endometriosis show significantly higher numbers of CD56 + uNK cells in the mid-luteal phase compared to healthy women [ 69 ]. Additionally, the number of uNK progenitor cells (i.e. uNK cells at developmental stages) is noticeably increased in women with endometriosis [ 68 ], indicating a dysfunction in the in situ development of mature uNK cells and potential dysregulation of their functions [ 68 , 70 ]. Accumulated evidence demonstrates that women with repeated implantation failure (RIF), typically defined as failure to implant after three consecutive transfers of high-quality embryos, and those with recurrent miscarriage (RM), usually defined as two or three consecutive miscarriage before 20 or 24 weeks of gestation, both show an increased percentage of uNK cells. However, further studies are needed to determine whether the increase in uNK cells contributes to the high incidence of infertility and miscarriage in women with endometriosis. The expression of NKp46 and the NKp46 + /CD56 + cell ratio is both found significantly decreased in patients with severe endometriosis compared with healthy women [ 69 ]. Meanwhile, low expression of NKp46 in uNK cells has also been demonstrated in women with reproductive disorders such as RIF and RM [ 71 , 72 ]. Given the involvement of NKp46 in uNK cell activation/maturation and angiogenic functions during pregnancy in mice [ 73 ], its inadequacy may imply a deficiency in the number and function of uNK cells in patients with severe endometriosis, potentially leading to their inadequate pregnancy support. There are also data suggesting that the expression of NKp30 (a natural cytotoxicity receptor of uNK cells) and its ligand BAG6 is upregulated in uNK cells of the eutopic endometrium of endometriosis [ 65 ]. It may lead to the increased cytotoxicity of these uNK cells, which is detrimental to reproduction (Fig.  3 a). At present, there was a paucity of publication describing the relationship between uNK cells and adenomyosis. No significant difference in the number of CD56 + uNK cells has been reported in the eutopic endometrium of patients with adenomyosis compared with controls in two previous studies. However, a subgroup analysis of the adenomyosis cohort revealed that the number of uNK cells did not increase in women with mild adenomyosis while it was elevated in the eutopic endometrium of women with severe adenomyosis (either diffuse or adenomyoma type) during the late luteal phase (cycle days 22–26) [ 5 , 38 , 74 , 75 ]. As stated before, the increased concentration of uNK cells in adenomyosis may also affect endometrial receptivity thereby contributing to infertility and miscarriage, which warrants further studies to substantiate. Yang et al. recruited 10 women with adenomyosis (5 in the luteal phase) and 12 women without adenomyosis (6 in the luteal phase) to compare the expression of KIRs. They found reduced expression of KIRs, including NKB1 and GL183, on uNK cells in the eutopic endometrium from patients with adenomyosis [ 75 ], which may be a compensated response to increased cytotoxicity of uNK cells, potentially resulting in lower fertility in these patients. Additionally, a prospective case-control study found that the expression of monocyte chemoattractant protein-1 (MCP-1), a cytokine produced by uNK cells that triggers the migration of immune cells to target tissues, was significantly increased in endometrial tissues of patients with adenomyosis compared with controls during the implantation window after ovarian stimulation [ 76 ]. The aggregation of immune cells may lead to excessive proinflammatory response at the maternal-fetal interface, potentially resulting in low clinical pregnancy rates and high miscarriage rates in patients with adenomyosis [ 5 ] (Fig.  3 b). Overall, the current understanding of the impact of uNK cell alterations on reproductive failure in endometriosis and adenomyosis remains limited, warranting further investigation. Throughout menstrual cycle and pregnancy, fluctuations in female hormones orchestrate changes in T cell numbers and functions to meet various physiological needs, balancing immune tolerance–essential for a successful pregnancy–with immune activation needed for protection against infection [ 77 ]. However, this balance can be disrupted in endometriosis and adenomyosis, potentially leading to reproductive disorders and pregnancy complications. A variety of uterine T cell subsets regulate immune responses during the menstrual cycle, embryo implantation and pregnancy. These endometrial T cells primarily include CD4 + helper T cells, CD8 + T cells, regulatory T cells (Treg), and γδ T cells. The composition of T cells in the endometrium shows a similar proportion of CD4 + T cells and CD8 + T cells, unlike peripheral blood, where CD4 + T cells comprise two-thirds of the T cell population [ 78 ]. Decidual T cells exhibit a more differentiated state, with a lower proportion of CD45RA + naive T cells compared to peripheral and endometrial T cells [ 79 ]. The number of endometrial T cells do not significantly fluctuate during the menstrual phase [ 80 ], but the increasing numbers of NK cells and macrophages contribute to a reduced proportion of T lymphocyte [ 58 ]. The majority of endometrial CD8 + T cells are CD45RO + memory T cells [ 78 ]. During the proliferative phase, CD8 + T cells exhibit high cytolytic activity to recognize and eliminate abnormal or infected cells [ 81 ]. In the luteal phase, CD8 + T cells lose their cytolytic function to avoid attacking the allogeneic embryo [ 81 ]. In early pregnancy, decidual CD8 + T cells are capable of promoting the invasive capacity of trophoblast cells [ 82 ]. Unlike CD8 + T cells, CD4 + T helper cells modulate the immune response and contribute to endometrial homeostasis by secreting cytokines. Based on their cytokine profiles, endometrial CD4 + T helper cells can be divided into Th1, Th2 and Th17 subsets. Th1 cells produce pro-inflammatory cytokines, Th2 cells secrete anti-inflammatory cytokines, and Th17 cells uniquely produce the pro-inflammatory cytokine IL-17 [ 83 , 84 ]. Additionally, Tregs, characterized by the transcription factor Foxp3 and surface markers CD25, play a crucial role in maintaining immune tolerance by suppressing excessive immune responses [ 84 ]. The dynamic balance between local Th1 and Th2 cells, as well as Th17 and Tregs reflects the inflammatory response, which is crucial for maintaining immune tolerance at the maternal-fetal interface [ 83 ]. During the luteal phase, endometrial Tregs significantly increase compared to the proliferative phase, indicating a shift to an anti-inflammatory state that is crucial for inducing immune tolerance and preventing inflammatory damage to the embryo [ 85 ]. After embryo implantation, Tregs respond to fetal antigens, proliferate significantly, and become enriched in the decidua [ 86 ]. During late pregnancy, Treg clonal expansion increases compared to early stage, suggesting a strong requirement for antigen-specific tolerance during late gestation [ 86 ]. In contrast, Th17 cells remain stable throughout pregnancy [ 87 ]. Another important T cell subset is γδ T cells, identified by the γδ T-cell receptor and divided into two main sub-populations: Vδ1 and Vδ2 [ 88 ]. Compared to the proliferative phase, γδT cell are also significantly higher in the luteal phase and during pregnancy [ 89 ]. The Vδ1 cells are known to secrete Th2-type cytokines, thereby creating a specialized immune microenvironment that supports embryo implantation and early pregnancy (Fig.  2 ). Various endometrial T cells exhibit significant changes in women with endometriosis compared to healthy or non-endometriosis controls. Firstly, endometrial CD3 + T cells significantly increase during the proliferative phase, although they are similar in the luteal phase in patients with endometriosis [ 90 ]. Secondly, the proportion of CD8 + T EM cells, primarily to recognize and eliminates infected cells, is significantly higher in the eutopic endometrium across all menstrual phases in endometriosis [ 91 , 92 ]. Additionally, increased secretion of IL-17a from Th17 cells leads to the accumulation of neutrophils and proliferation of endometriotic cell, resulting in persistent inflammation [ 93 ]. Consistently, single-cell transcriptomic analysis of endometrium from women with endometriosis revealed elevated levels of proinflammatory cytokines [ 94 ]. Furthermore, one study demonstrates a reduction of Foxp3 + cells in the peri-implantation endometrium of patients with endometriosis [ 95 ], a finding supported by recent studies [ 96 ] (Fig.  3 a). Research on the local immune microenvironment in adenomyosis is relatively limited compared to endometriosis, but the pattern of pro-inflammatory response is similar. Increased numbers of CD3 + T cells, CD4 + and CD8 + T cells and γδT cells have been reported in the eutopic endometrium of patients with adenomyosis compared to controls [ 90 , 97 , 98 ]. Additionally, decreased Foxp3 and increased IL-17 A expression levels are observed in the eutopic endometrium of women with adenomyosis, and these changes are positively associated with the severity of dysmenorrhea [ 99 ] (Fig.  3 b). Accumulated evidence suggests that disruptions in endometrial T cell dynamics, both before and after pregnancy, are correlated with reproductive failure. Excessive CD8 + T cell activity in the mid-luteal phase endometrium and during pregnancy have also been found in women with unexplained RM [ 100 ]. Additionally, an imbalance in endometrial Th cell populations, particularly a shift towards Th1 or Th17 dominance and insufficient Tregs has been linked to unexplained RIF and RM [ 85 , 101 , 102 ]. Consistently, decreased Treg cells have been observed in the decidual of spontaneous abortion compared to induced abortion [ 103 ]. Furthermore, decreased Tregs are predictive markers for early pregnancy failure in infertile patients [ 104 ]. Similarly, women with endometriosis and adenomyosis exhibit elevated CD8 + T cells [ 92 , 97 ], γδT cells [ 98 ], Th1 and Th17 cytokine profile in the mid-luteal phase endometrium [ 93 , 99 ], alongside decreased Tregs [ 96 , 103 ]. It is plausible to hypothesize that these changes may contribute to reproductive failure. However, direct evidence of this causal effect is lacking, as many studies have not conducted subgroup analysis based on the history of reproductive failure or reproductive outcomes. Additionally, no animal studies have been conducted to demonstrate the causal effects. Further investigation is warranted to understand their contributions to reproductive failure in endometriosis and adenomyosis (Fig.  3 ). In addition to NK cells, macrophages, and T cells, other endometrial immune cells, including dendritic cells (DC), B cells, and neutrophils also play a role in shaping the immune microenvironment for embryo implantation and placentation. However, these cells are rarely studied in women with endometriosis and adenomyosis. Limited evidence suggests that DC maturation is defective in women with endometriosis, with a higher density of immature DC [ 105 ]. While the changes in endometrial B cells in women with endometriosis are controversial, serum antibody titers against the endometrium are significantly higher in these women [ 106 ]. Additionally, the number of neutrophils and their attractant IL-8 have been reported to be substantially increased in women with endometriosis [ 29 , 107 ]. In contrast, neutrophils are debated in limited reports of adenomyosis, with most studies indicating decreased levels of IL-8 in the eutopic endometrium of women with adenomyosis compared to controls [ 5 ]. Overall, it remains unclear whether these changes affect reproduction and pregnancy. The underlying causes of endometrial immune cell alterations that contribute to the inflammatory response in endometriosis and adenomyosis remain unclear due to a lack of direct evidence. We speculate that factors driving chronic inflammation in the microenvironment may act as upstream mediators. Chronic endometritis (CE) is a persistent inflammatory condition of the endometrium, characterized by the presence of plasma cells. It is usually asymptomatic and been reported to be related with infertility and miscarriage due to chronic inflammation [ 108 ]. Interestingly, the prevalence of CE was significantly higher in patients with endometriosis and adenomyosis compared with those without these conditions [ 109 – 112 ]. Furthermore, in infertile patients with endometriosis, CE is linked to decreased cumulative pregnancy rate and live birth rate [ 113 ], and increased risk of pregnancy complications [ 114 ]. These findings reveal a strong association between CE and endometriosis/adenomyosis. The local chronic inflammation environment in women with CE induces significant alterations in the composition of endometrial immune cells. These changes are characterized by an increase in CD68 + macrophages [ 115 ], elevated CD56 + uNK cells during the mid-luteal phase [ 116 , 117 ], and higher levels of CD3 + T cells, CD8 + T cells, Th1 cells, and Foxp3 + Treg cells [ 116 ]. In contrast, Th2 cells are decreased [ 115 , 118 ]. Hence, it is plausible that CE in the context of endometriosis or adenomyosis contributes to the disruption of endometrial immune cells. The most common cause of CE is infection triggered by pathogenic microorganisms, often occurring alongside dysbiosis. The Lactobacillus predominates in a healthy uterus and inhibits the colonization and infection of pathogens, while non- Lactobacillus- dominant microbiotas have been associated with adverse reproductive outcomes [ 119 ]. Compared with healthy women, patients with endometriosis exhibit reduced Lactobacillus abundance in the uterus, along with increased microbiota diversity [ 120 , 121 ]. Similarly, women with adenomyosis are less likely to show a Lactobacillus- dominant endometrial microbiota [ 122 ]. The pathogenic microbiome can initiate immune response mediated by antigen presenting cells. Additionally, microbiota metabolites may regulate immune cell function [ 123 ]. Therefore, the changed endometrial microbiota in CE might be another driving factor affecting endometrial immune cells, which requires further studies. Both endometriosis and adenomyosis are characterized by elevated estrogen levels in the endometriotic lesions [ 124 – 126 ]. Estrogen receptors are not expressed on most endometrial leukocytes [ 127 ], with the exception of γδT cells [ 128 ]. Studies have shown that estrogen can directly enhance IL-17 expression in endometrial γδT cells [ 128 ]. Although research on the impact of estrogen on endometrial immune cells is limited, some studies have demonstrated that estrogen can regulate the development and function of lymphocytes either through direct interaction or indirectly via products from stromal and epithelial cells that express hormone receptors [ 129 , 130 ]. Several in vivo studies have highlighted estrogen’s effects on immune cells, including promoting the proliferation of Foxp3 + Treg cells [ 131 ], and enhancing the activity of IFN-producing Th1 cells through estrogen receptor α on hematopoietic cells [ 132 ]. Therefore, the elevated estrogen concentration in adenomyosis and endometriosis may contribute to immune cell alterations, although direct evidence supporting this hypothesis remains lacking. Epigenetics is the study of how genetic information is transmitted to offspring without changes to DNA sequence, through processes like DNA methylation, histone modifications, chromatin remodeling, RNA transcriptional changes [ 133 ]. Abnormal gene expression in immune cells resulting from epigenetic alterations can lead to a loss of immune tolerance, inflammation, and autoimmunity [ 134 ]. Current evidence strongly suggests that epigenetic mechanisms play a significant role in the pathogenesis of endometriosis, including hypermethylation at the ends of chromosomes in endometriotic stromal cells, increased activity of histone deacetylases (HDACs) in endometriotic cells, altered miRNA expression patterns, and nuclear receptor modulation [ 135 ]. Similarly, the etiology of adenomyosis is linked to epigenetic dysregulation, with notable changes such as elevated expression of HDAC1 and HDAC3 in both eutopic and ectopic endometria, as well as promoter hypermethylation of the progesterone receptor B isoform (PR-B) [ 136 ]. Abnormal expression of inflammation-related factors due to epigenetic alterations has been reported in both endometriosis and adenomyosis. For example, excessive production of prostaglandin E 2 and cytokines such as TNFα, IL-1β, IL-6, IL-8, IFNγ, and MCP-1 is observed in endometriosis lesions [ 137 ], while low expression of PR-B (acting as an anti-inflammatory agent) is found in both conditions [ 138 , 139 ]. These epigenetic changes and the resulting inflammation in the endometrium suggest that epigenetic mechanisms may contribute to the inflammatory environment characteristic of endometriosis and adenomyosis. Although direct studies linking epigenetic alterations to endometrial immune cell dysfunction in these diseases are limited, this hypothesis provides a promising direction for further research.