Endometrial Stromal Cell Senescence: A Non‐Negligible Factor in Recurrent Pregnancy Loss

Endometrial Stromal Cell Senescence: A Non-Negligible Factor in Recurrent Pregnancy Loss Shuang Wu, Cenlan Bu, and Qinzheng Xu contributed equally to this work. ABSTRACT Recurrent pregnancy loss (RPL) affects approximately 1%–2% of couples of reproductive age, imposing a substantial physiological and psychological burden on patients. Concurrently, against the backdrop of global population aging, the association between female reproductive aging and infertility is increasingly evident. Cellular senescence is typically characterized by an essentially irreversible cell-cycle arrest and widespread cellular injury, culminating in tissue dysfunction. Aberrant senescence of endometrial stromal cells (EnSCs) may contribute to RPL by compromising endometrial receptivity and impeding embryo implantation and subsequent development, among other mechanisms. However, the interplay between cellular senescence and tissue or organismal function is complex and interdependent. In the endometrium, senescence of EnSCs engages in bidirectional crosstalk with the tissue-specific process of decidualization. Based on existing literature, this article aims to: (1) elucidate how cellular aging is both a component of normal decidualization and may be involved in the progression of RPL; (2) speculate on the cellular event sequence involved in EnSCs aging during normal and damaged decidualization processes. In-depth analysis of the interaction between EnSCs aging and decidualization will help propose targeted anti-aging therapies to eliminate aging cells, as a potential strategy for restoring damaged endometrial receptivity and improving the success rate of in vitro fertilization. Summary - Endometrial stromal cell (EnSC) senescence is a key maternal factor contributing to recurrent pregnancy loss by impairing endometrial receptivity. - Physiological, transient EnSC senescence supports normal decidualization and implantation, but persistent accumulation of senescent cells creates chronic inflammation and tissue dysfunction. - Oxidative stress, hormonal and metabolic disturbances, and stem/progenitor cell exhaustion synergistically drive premature EnSC senescence in RPL patients. - Targeting senescent cells and related pathways offers a potential therapeutic strategy to restore a healthy endometrial environment and improve pregnancy outcomes. 1 Introduction According to guidelines from the European Society for Human Reproduction and Embryology (ESHRE) and the American Society for Reproductive Medicine (ASRM), recurrent pregnancy loss (RPL) is defined as two or more clinically confirmed pregnancy losses (documented by ultrasound or histopathology). It has an incidence of approximately 1%–2% [1]. About 15% of clinically recognized pregnancies end in miscarriage, and 50%–60% of sporadic cases are attributable to chromosomal abnormalities in the embryo [2]. However, the proportion of embryonic chromosomal abnormalities is relatively low in RPL. And with each additional miscarriage, the incidence of normal chromosomal miscarriage increases, while the probability of a successful pregnancy decreases [3]. This phenomenon indicates that, in addition to embryonic factors, maternal uterine determinants contribute substantially to elevated miscarriage rates [4, 5]. With the delay of childbearing age, women's risks of miscarriage and other reproductive disorders rise markedly after the age of 30, and features of reproductive aging often precede chronological age [6, 7]. Although research on reproductive aging has historically centered on the ovary, accumulating evidence now underscores that the endometrium also plays a critical role in determining pregnancy outcomes. Despite advances, effective interventions for RPL remain scarce, and in a substantial proportion of patients, the underlying cause is still unclear [8, 9]. Whereas earlier work focused largely on embryonic quality and endocrine or immunologic anomalies, increasing evidence highlights the central role of the endometrial environment in supporting embryo implantation and sustaining pregnancy [10-12]. Among determinants of implantation, the functional state of endometrial stromal cells (EnSCs) is particularly critical. Stromal-cell aging has been posited as an important pathological mechanism in RPL and has gained wide attention in basic research and clinical practice [13, 14]. EnSCs are key endometrial constituents and the principal mediators of decidualization. During the luteal phase, hormonal signals drive EnSCs to differentiate into decidual stromal cells, establishing a supportive micro-environment for embryo implantation and placental development [15]. Conversely, premature senescence of EnSCs, or the aberrant accumulation of senescent cells, disrupts decidualization, diminishes endometrial receptivity, and compromises the maternal–fetal interface, thereby increasing the risk of preterm birth, fetal growth restriction, preeclampsia, and miscarriage [16, 17]. Abnormal EnSCs senescence has been documented in the decidua of women with recurrent miscarriage, implicating it in the pathogenesis of RPL [17, 18]. This review aims to provide a systematic account of EnSCs' senescence in physiological and pathological pregnancy, delineate its putative mechanistic contributions to RPL, and evaluate emerging interventions targeting EnSCs' senescence, thereby providing a possible new theoretical basis and research direction for the treatment of RPL in the future. 2 Main Text 2.1 The Function and Physiological Characteristics of EnSCs 2.1.1 The Structural Composition and Function of Endometrium The endometrium comprises a functionalis and a basalis and contains epithelium, glands, stroma, and vasculature [19-21]. The endometrial epithelium includes ciliated and secretory cells, corresponding to the luminal and glandular epithelia, respectively. Ciliated cells generate fluid flow and facilitate embryo transport, whereas secretory cells produce growth factor–rich secretions that support embryonic development and implantation and are crucial for early pregnancy maintenance [22]. The endometrial stroma is composed primarily of fibroblasts but also contains blood vessels and diverse immune cells, including natural killer (NK) cells, T cells, macrophages, and dendritic cells. Throughout the human menstrual cycle, the stroma is periodically exposed to steroid hormones, particularly 17β-estradiol and progesterone [23]. EnSCs, the principal stromal component, are fibroblast-like cells located around glands and blood vessels, providing structural support and mediating intercellular signaling (Figure 1). The endometrium, the innermost layer of the uterus and a dynamic multicellular tissue, performs three fundamental functions: preparation for implantation, support of pregnancy following successful implantation, and menstruation in the absence of conception. Single-cell RNA sequencing (scRNA-seq) analyses have refined the staging of the menstrual cycle [24], delineating four principal phases based on gene-expression profiles across distinct cell populations: early proliferative, late proliferative, early secretory, and mid/late secretory. Throughout the cycle, periodic exposure to ovarian steroids induces transitions between proliferative and receptive states across multiple cell types, thereby preparing the endometrium for embryo implantation within a narrow window of implantation [25, 26]. Phase 1 corresponds to the early proliferative phase [27]. The expression of specific genes by distinct cell types defines the early proliferative phase—matrix metallopeptidase 1 (MMP1), thrombospondin 1 (THBS1), and PAEP by epithelial cells and MMP1, secreted frizzled-related protein 1, and Wnt family member 5 (WNT5A) by stromal cells. The expression patterns of these cells in this phase agree with the proliferative state of the endometrium [21]. Phase 2 is the late proliferative phase, characterized by expression of secretoglobin family 1D member 1 (SCGB1D1) in epithelial cells and WNT5A in stromal cells. Genes critical for cell proliferation and the cell cycle, already expressed in the preceding phase, reach peak levels at this stage. Toward the end of the early and late proliferative phases, specific cells express metalloproteinase-encoding genes that remodel the extracellular matrix in anticipation of the implantation window. In addition, expression of estrogen receptor 1 (ESR1) declines during this stage [21]. During the estrogen-stimulated phase, increased proliferation of stromal and epithelial cells precedes the subsequent decline in estrogen and the rise in progesterone dominance, which marks the onset of the secretory phase. The third stage is the early secretion phase. In the early stage of secretion, the epithelium overexpresses member 2 of the secretory globin family 1D and exhibits elevated levels of metalloproteinase gene expression. At this stage, the endometrium prepares for the window period of implantation and embryo reception. At the same time, stromal cells begin to express genes necessary for decellularization, and the number of lymphocytes increases [21]. The fourth stage is the mid/late stage of secretion. A sudden, discontinuous activation of implantation-related gene programs in the primary ciliated epithelium is a defining feature of the mid/late secretory phase. Concurrently, epithelial expression trajectories help pinpoint the precise moment of endometrial receptivity: genes that are moderately expressed throughout phase 4 become highly expressed in phase 1 of the next cycle (e.g., PAEP and glutathione peroxidase 3). By the end of phase 4, other genes decline in expression—chemokine ligands, monoamine oxidase A, dipeptidyl peptidase 4, and metallothionein genes. The latter two gene groups demarcate the mid-to-late secretory transition, whereas the continued presence of metalloproteinase gene expression indicates an important role in regulating the implantation window. Analysis of stage 4 stromal fibroblasts showed that transcription factors related to the initiation of decidualization, such as forkhead box transcription factors and the decidual matrix marker interleukin-5, gradually overexpressed, indicating that decidualization began before the implantation window opened [21]. Stromal cells then transition into secretory decidual cells. Decidualization involves glandular maturation, spiral-artery remodeling, conversion of energy stores into glycogen, and establishment of an optimal environment for embryo implantation. The epithelial glands also undergo secretory transformation, generating a receptive endometrium and shifting the tissue from a proliferative to an implantation-ready state in anticipation of the narrow window during which implantation can occur. During phase 4, lymphocytes migrate into the stroma and display increased expression of genes associated with uterine NK cell function that are known to be upregulated in pregnancy (e.g., integrin α1 subunit and cluster-of-differentiation markers). In addition, genes encoding ligand–receptor pairs that support intercellular communication between stromal fibroblasts and lymphocytes are upregulated (e.g., interleukin-14, interleukin-2 receptor subunits β and γ, MHC class I genes, and NK-receptor genes) [21]. During the window of implantation, the endometrium achieves a structural and biochemical state that permits embryonic adhesion and the initiation of implantation. Without implantation, the corpus luteum in the ovary will regress, leading to a decrease in estradiol and progesterone at the end of the secretion period, contraction of the spiral artery, and shedding of the endometrial functional layer. Under the influence of the ovarian cycle, stromal cells are highly sensitive to sex steroids. In the proliferative phase, stromal cell proliferation and secretion of growth factors and cytokines increase, thickening the endometrium; the secretory phase is progesterone-dominant, during which stromal cells initiate decidualization, morphologically shifting from elongated to polygonal forms and secreting decidual factors (e.g., interleukins, placental growth factor) to prepare for implantation. If implantation occurs, stromal cells further differentiate into decidual cells that form a compact layer surrounding the embryo and supporting placental development; if not, stromal cells undergo apoptosis or are shed during menstruation, and the endometrium enters the next cycle. Recent work shows that scRNA-seq identifies multiple stromal subpopulations within the endometrium [28]. Four stromal subsets participate in proliferation, tissue repair, and regeneration, and at the transcriptomic level express genes that regulate the actin–myosin machinery essential for these processes (e.g., ACTA2 and myosin light chain), together with genes associated with myofibroblast differentiation and epithelial–mesenchymal transition. Another four stromal subsets express genes that regulate epithelial proliferation and differentiation, epithelial–stromal cross-talk, and innate immune responses during wound healing, as well as genes involved in interferon signaling, innate immunity, and cell differentiation. These subsets likely play key roles after menstruation and may be functionally related to pericytes. The remaining two subsets identified by scRNA-seq express markers of regulators of G-protein signaling (RGS), which inhibit G-protein signaling and have been recognized as biomarkers of perivascular pericyte populations [29]. These perivascular cells are not stromal cells in the strict sense but cooperate with stroma in regeneration: one subset expresses vascular-lineage genes, and the other expresses vascular smooth-muscle genes. Based on these findings, single-cell gene-expression analyses by scRNA-seq have helped define a cellular atlas of human EnSCs. Thus, scRNA-seq and similar tools, together with the gene-expression datasets they generate, may offer ways to address previously unresolved questions. These results remain crucial for improving our understanding of the functions of each cell population, how distinct cell types cooperate, and how both shape the menstrual cycle. In the context of endometrial disease, insights from scRNA-seq can help pinpoint root causes and affected pathways, thereby deepening our understanding of stromal-cell-related physiology and pathophysiology and enabling the development of new therapeutic approaches. 2.1.2 Physiological Functions of Stromal Cells EnSCs, the principal cellular component of the endometrial stroma, are distributed around glands and blood vessels. Under precise regulation by ovarian hormones, these cells undergo cyclical remodeling to prepare the tissue for embryo implantation. Recent single-cell studies reveal that EnSCs constitute a functionally heterogeneous population comprising multiple subclusters with distinct biological roles [30]. DIO2+ stromal cells, which highly express type II iodothyronine deiodinase (DIO2), convert thyroxine (T4) to the active triiodothyronine (T3). These cells are enriched early at the embryo anchoring sites during the window of implantation and secrete factors such as WNT5A and HOXA10 to regulate epithelial–mesenchymal transition (EMT), thereby constructing a plastic microenvironment permissive to trophoblast invasion [31]. By contrast, PLA2G2A+ stromal cells, which are marked by high expression of secretory phospholipase A2 group IIA (PLA2G2A), are key effectors of decidualization. They can secrete IL-15 to recruit and modulate uterine natural killer (uNK) cells, promoting vascular remodeling and immune tolerance that are essential for maintenance of pregnancy [32]. Pregnancy is a complex process where the fertilized egg develops into a fetus that can survive outside the uterus. Any stage of damage can lead to adverse consequences. Several critical windows carry elevated risks of miscarriage or preterm birth. The earliest is the “window of implantation” (WOI), typically 4–6 days in the mid-luteal phase, the only period during which the maternal endometrium permits embryonic attachment. Moreover, spontaneous loss is relatively common before 12 weeks of gestation. Decidualization of the endometrium is a prerequisite for establishing the WOI and the formation of the maternal–fetal interface. Decidualization is a remodeling process of the intrinsic layer of the endometrium during the luteal phase, driven by hormone signals, mainly by progesterone produced by the corpus luteum and an increase in intracellular cAMP, which does not require embryo implantation [33]. In terms of morphology, EnSC transforms from spindle shaped to polygonal epithelioid cells. At the molecular level, classic decidual markers such as insulin-like growth factor binding protein-1 (IGFBP1) and prolactin (PRL) are significantly upregulated [34]. These changes create conditions favorable for implantation. In the absence of implantation, falling progesterone triggers an inflammatory response in the decidua, recruiting and activating leukocytes, which leads to breakdown of the functional layer and menstruation. After menses, rising follicular estradiol (E2) drives rapid tissue growth, increasing endometrial thickness several-fold within 10 days to prepare for a potential subsequent pregnancy. Mesenchymal stem cell-like cells (MSCs) in the basalis and epithelial progenitors are activated to participate in the repair and regrowth of the endometrium [35]. In clinical practice, poor endometrial growth is closely related to implantation failure [36, 37]. The decidual stromal cells have multiple functions. Decidual stromal cells regulate endometrial receptivity and secrete nutritional factors for the embryo, such as IGFBP1 and progesterone-induced blocking factor PIBF. In addition, decidual stromal cells reshape the extracellular matrix to allow for controlled invasion of the trophoblast layer. Moreover, decidual stromal cells regulate immunity by recruiting and cultivating uNK cells, macrophages, and other cells to promote tolerance. Interruption of decidualization, such as incomplete decidualization or premature degeneration, can disrupt the stability of the maternal-fetal interface and increase the risk of miscarriage [38]. Single-cell analyses of decidual tissue indicate that EnSCs are the most abundant cell type and diversify into multiple subpopulations during the decidualization process [38]. The timely emergence of functionally diverse decidualized EnSCs is essential for the establishment and maintenance of pregnancy, contributing to trophoblast invasion and growth, prevention of maternal immune rejection, and promotion of angiogenesis. Decidualized EnSCs produce a variety of extracellular matrix (ECM) components, including fibronectin, type IV collagen, laminin, heparan sulfate proteoglycans, and decorin, that interact with invading extravillous trophoblasts to facilitate embryonic attachment and invasion while simultaneously limiting the depth of invasion to protect the decidua from damage [38-40]. Vascular remodeling is a hallmark of decidualization, enabling the placenta to exchange gases, blood, and nutrients with the mother. Decidualized EnSCs secrete pro-angiogenic signaling molecules, including VEGFA/B/D and ANGPTL2, which act on cognate receptors on uterine endothelial cells to establish angiogenic niches and drive vessel formation [40, 41]. Compared with healthy pregnancies, the decidua of women with miscarry exhibits fewer vessels with narrower lumens, suggesting that defective angiogenesis may contribute to pregnancy loss. EnSCs also exert potent immunological functions that are integral to establishing and maintaining maternal–fetal immune tolerance. Molecular phenotyping shows a close relationship between EnSCs and follicular dendritic cells (FDCs) [30]. Both EnSCs and FDCs originate from perivascular PDGFRβ+ adult stem/progenitor cells and differentiate in response to inflammatory cues, which are key mediators of local and systemic immune tolerance, respectively. Decidualized EnSCs secrete multiple chemokines, including CX3CL1, CXCL9, CXCL10, and CXCL11, to recruit peripheral NK cells, Th1 lymphocytes, and cytotoxic T cells [42]. Appropriate clustering of these immune cells is crucial for host defense and for managing the inflammatory phase associated with implantation. During the anti-inflammatory phase of decidualization and the maintenance of pregnancy, decidualized EnSCs attenuate the cytotoxicity of these immune cells, preserving decidual homeostasis and preventing an overly aggressive immune milieu at the maternal–fetal interface. In addition, EnSC-derived IL-15 acts on its receptors on decidual NK (dNK) cells to promote placental formation [43-45]. In sum, decidualized EnSCs interact with diverse immune cells within the decidua to establish a type-2–skewed immune response favorable to fetal immune tolerance and normal pregnancy progression. Taken together, EnSCs are far from a functionally uniform population. Rather, under the precise control of ovarian hormones, they constitute a core unit with pronounced functional heterogeneity and spatiotemporal dynamics. Their central physiological roles span the initiation, establishment, and maintenance of pregnancy, making them indispensable to successful pregnancy. 2.2 Cell Senescence Cellular senescence represents a dynamic, multifaceted program triggered by diverse intrinsic and extrinsic stressors. Its onset typically follows stress signals such as DNA damage, oxidative stress, oncogenic signaling, and telomere shortening. These cues activate tumor-suppressive pathways, most notably the p53/p21 axis, culminating in proliferative arrest [46]. The genetics of aging involves multiple factors from telomere dynamics to key regulatory genes such as Klotho and Angiotensin converting enzyme (ACE) [47, 48] that play pivotal roles in aging trajectories and lifespan control. A deeper understanding of how senescence is regulated will not only provide valuable insight into the aging of EnSCs but may also inform effective therapeutic strategies for recurrent pregnancy loss driven by EnSC senescence (Figure 2). 2.2.1 Mechanisms, Pathways, and Key Molecules of Cellular Aging 2.2.1.1 Aging Related Senescence-Associated Secretory Phenotype (SASP) Beyond proliferation arrest, senescent cells undergo chromatin and metabolic remodeling and secrete a complex milieu of cytokines, chemokines, growth factors, and matrix metalloproteinases termed the senescence-associated secretory phenotype (SASP) [49]. SASP factors reinforce senescence and contribute to tissue remodeling and repair, but their chronic production promotes persistent inflammation and age-related pathologies, motivating the development of senolytics that selectively eliminate senescent cells to delay or prevent age-associated disorders [50]. The SASP is highly context dependent, varying with cell type, microenvironment, and inducing stress, and is regulated by signaling pathways including NF-κB, mTOR, and p38 MAPK. In particular, mTOR integrates nutrient and stress signals to control protein synthesis, autophagy, and metabolic activity in senescent cells, linking nutrient-sensing networks to senescence and SASP regulation [51]. Acute senescence is typically transient and beneficial for wound healing and regeneration, whereas late-stage or chronic senescence is marked by persistent, apoptosis-resistant senescent cells, sustained SASP secretion, and chronic inflammation that drive tissue dysfunction, fibrosis, and age-related diseases such as osteoarthritis, atherosclerosis, and neurodegeneration [52, 53]. Immune-mediated clearance of senescent cells is essential for tissue homeostasis, but age-associated decline in immune surveillance permits their accumulation, further exacerbating chronic inflammation [52]. 2.2.1.2 Klotho Gene The Klotho gene encodes KL, a type I membrane protein expressed predominantly in the kidney, brain, and parathyroid glands. The Klotho gene functions as an anti-aging gene: its overexpression extends lifespan, whereas its deficiency produces a phenotype of accelerated aging—hence its name after Clotho, the Greek Fate who spins the thread of life [54]. The Klotho protein can be shed and released into the circulation, where it acts as a hormone influencing multiple physiological processes. KL modulates the insulin/IGF-1 signaling pathway, a key regulator of aging by reducing phosphorylation of insulin receptor substrates, thereby dampening insulin and IGF-1 signaling and lowering cellular glucose uptake and metabolic activity [55]. This action is thought to mimic the effects of caloric restriction, a well-known intervention that extends lifespan across species. Beyond glucose metabolism, KL also governs calcium and phosphate homeostasis by serving as a co-receptor for fibroblast growth factor 23. This interaction is essential for maintaining mineral balance and preventing vascular calcification, a hallmark of aging and chronic kidney disease. Recent studies indicate that Klotho confers protection against oxidative stress by upregulating antioxidant enzymes such as manganese superoxide dismutase and catalase, thereby mitigating the damaging effects of reactive oxygen species (ROS), which accumulate with age and contribute to cellular injury [56]. Klotho's therapeutic potential has attracted broad interest: in preclinical models, administration of recombinant KL improves cognitive function, reduces vascular calcification, and prolongs lifespan [57]. These findings suggest that supplementing or upregulating Klotho may represent a novel strategy to promote healthy aging and counteract age-related diseases. 2.2.1.3 Angiotensin Converting Enzyme (ACE) and Aging The acn-1 gene has been identified as the nematode homolog of ACE and has been used to examine the relationship between ACE-inhibitor use and lifespan [58, 59]. Administration of the ACE inhibitor captopril reduces acn-1 activity, thereby extending lifespan, enhancing stress resistance, and delaying age-related degenerative changes as assessed by pharyngeal pumping measurements. Mouse models have likewise been used to study the impact of ACE inhibitors on longevity. In one study, combined treatment with ramipril and statins (e.g., simvastatin) increased mean lifespan in isocalorically fed mice [60]. Enhanced stress resistance has been shown to be a contributor to lifespan extension across multiple species, including humans [61]. As noted above, ACE is an angiotensin-converting enzyme and a well-established drug target for the management of hypertension and its complications. In addition, ACE functions as an amyloid-degrading enzyme and has been employed to limit late-stage accumulation of β-amyloid. These distinct activities support the concept of staged aging from transitional states to more advanced pathological phases. At earlier stages, ACE's role as an angiotensin-converting enzyme may be more beneficial for overall health, whereas once β-amyloid begins to accumulate abnormally, its function as an amyloid-degrading enzyme becomes increasingly important [60]. 2.2.1.4 NF-κB Signaling Pathway The nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway is a key regulator of inflammation, immune responses, and cellular senescence, and it plays a pivotal role in the aging process [46]. NF-κB is typically inactive in the cytoplasm but becomes activated in response to stressors such as DNA damage, oxidative stress, and cytokines, then translocates to the nucleus to promote the expression of pro-inflammatory genes [62, 63]. With advancing cellular age, NF-κB activity increases, driving SASP secretion, a collection of pro-inflammatory cytokines, chemokines, and growth factors, which can lead to tissue dysfunction and chronic inflammation. This chronic activation contributes to age-related diseases, including osteoarthritis, cardiovascular disease, and neurodegeneration. Inhibiting NF-κB has shown promise for reducing cellular senescence and age-related inflammation. For example, the small molecule SR12343 blocks NF-κB activation and extends lifespan and healthspan in mouse models [64, 65]. Natural compounds such as curcumin, resveratrol, and epigallocatechin gallate (EGCG) also inhibit NF-κB signaling, suggesting that dietary interventions may beneficially modulate aging. The role of NF-κB in specific age-related conditions is well documented: it drives neuroinflammation in Alzheimer's disease, vascular inflammation in cardiovascular disease, and insulin resistance in metabolic disorders [66]. Targeting NF-κB thus represents a promising strategy to mitigate inflammation and delay age-associated pathology, with the potential to extend both healthspan and lifespan [67]. As research progresses, modulation of NF-κB may prove to be a key approach to counteracting the deleterious effects of aging. 2.2.1.5 Other Pathways or Aging Key Molecules The insulin/IGF-1 signaling (IIS) pathway regulates aging by controlling processes such as protein synthesis, glucose metabolism, cell proliferation, oxidative stress, and inflammation [68]. The role of IIS in longevity was first discovered in Caenorhabditis elegans, where loss of daf-2, which encodes the insulin/IGF-1 receptor, doubled lifespan [69]. IGF-1 is essential during development and declines with age, yet its sustained activation promotes aging by acting through pathways such as FOXO, NF-κB, and MAPK. IGF-1 also regulates autophagy via Nrf2/Sirt3 signaling [70]. Although reduced IGF-1 signaling can extend lifespan, outright deficiency leads to sensorineural hearing loss and heightened inflammation, underscoring its complex role in aging. In humans, age-related declines in IGF-1 correlate with worsening hearing loss, suggesting that modulation of IGF-1 signaling warrants careful consideration in age-related hearing loss (ARHL) and other disorders [71]. mTOR kinase is a central regulator of cell growth and metabolism, forming two distinct complexes: mTORC1 and mTORC2. mTORC1 responds to nutrient availability, growth factors, and cellular energy status to promote protein synthesis, lipid biogenesis, and cell proliferation while suppressing autophagy [72]. Chronic activation of mTORC1, however, is linked to accelerated aging and age-associated pathologies [73]. Inhibition of mTORC1 by caloric restriction or pharmacologically with rapamycin extends lifespan across multiple species, an effect attributed to enhanced autophagy, improved proteostasis, and reduced cellular senescence. Notably, rapamycin mitigates several age-related conditions, including neurodegenerative and cardiovascular diseases, by modulating mTOR activity [74]. AMPK is a key energy sensor that is activated in response to increased AMP/ATP and ADP/ATP ratios, indicating cellular energy deficiency [74]. Whereas AMPK activation under low-energy conditions induces autophagy, mTORC1 activity depends on positive inputs, such as high energy, normoxia, amino acids, or growth factors, that suppress autophagy. Upon activation, AMPK phosphorylates targets that enhance catabolic pathways to generate ATP (e.g., glucose uptake and fatty-acid oxidation) while inhibiting ATP-consuming anabolic processes such as lipid and protein synthesis [75, 76]. This dual action restores energy balance and promotes survival under stress. In the context of aging, AMPK activation confers multiple benefits: it stimulates autophagy by phosphorylating ULK1, thereby clearing damaged organelles and proteins, and it enhances mitochondrial biogenesis via activation of PGC-1α, improving mitochondrial function and reducing ROS production [77-79]. Collectively, these actions contribute to extended healthspan and delayed onset of age-related disease. Sirtuins are a family of NAD-dependent deacetylases and ADP-ribosyltransferases that regulate metabolic homeostasis, stress responses, and genome stability [80]. Among the seven mammalian sirtuins (SIRT1–7), SIRT1 and SIRT3 are particularly prominent in aging. SIRT1 deacetylates substrates including PGC-1α, FOXO transcription factors, and NF-κB, thereby enhancing mitochondrial function, bolstering antioxidant defenses, and reducing inflammation [81]. SIRT3, localized to mitochondria, deacetylates and activates enzymes involved in oxidative metabolism and antioxidant defense, reducing ROS production and oxidative damage [82]. Activation of sirtuins has been associated with lifespan extension in model organisms and improved healthspan, suggesting their potential as therapeutic targets for age-related diseases. 2.2.2 Changes in the Ultrastructure and Functional Defects of Senescent Cells Another sign of aging cells is abnormal intracellular signaling, accompanied by changes in cell morphology, function, and organelle quality. These changes may be directly or indirectly caused by a mismatch between increased organelle biosynthesis and organelle dysfunction. When organelles are damaged, the clearance efficiency of proteins and aggregates decreases. Although aging cells may continuously produce new organelles to compensate, as aging progresses, the newly generated organelles can exacerbate damage through increased degradation stress and oxidative damage [83]. Morphologically, senescent cells become abnormally enlarged and flattened, with a disproportionate increase in the ratio of cytoplasm to nucleus. Lysosomes are the main degradation organelles that undergo upregulation of biosynthesis in senescent cells. Cellular aging leads to an increase in lysosomal mass [84]. In contrast, mitochondrial changes are one of the most prominent age-related features. During the process of cellular aging, mitochondria have larger mass and size, and their function decreases. Dysfunctional mitochondria can produce excessive ROS, which can damage DNA, lipids, and proteins, thereby disrupting mitochondrial dynamics and leading to significant mitochondrial enlargement [85, 86]. Mitochondrial damage also evokes retrograde signaling to the nucleus, altering transcriptional reprogramming and thereby influencing cell proliferation and senescence. The Golgi apparatus, a membrane-bound organelle, has been reported in senescent cells to exhibit an enlarged, dilated, and disorganized architecture with functional alterations [87, 88]. The endoplasmic reticulum (ER) is another membranous network in which the unfolded protein response (UPR) is activated in response to ER stress. Both dysregulated ER stress and UPR activation occur during cellular senescence. In addition to these well-recognized senescence-associated organelles, dysfunction of peroxisomes and the cytoskeleton is closely linked to senescence as well [89-91]. Peroxisome-derived ROS crosstalk with mitochondria to coordinate cellular ROS homeostasis, and peroxisomal deficiency provokes oxidative stress that triggers early stages of senescence. Senescent cells also display cytoskeletal alterations that impact cell shape, division, motility, and intracellular transport. These features act synergistically and reinforce one another to drive the senescence program, collectively shaping the senescence landscape. A comprehensive understanding of these interconnected mechanisms lays the groundwork for developing targeted therapeutic strategies to modulate EnSC senescence. 2.3 EnSCs and Cellular Aging An age-related endometrial transcriptomic analysis published in 2022 showed that women ≥ 35 years of age exhibit altered expression of genes involved in ciliary motility and epithelial cell proliferation, suggesting molecular mechanisms that may drive endometrial aging [92]. Moreover, a 2023 systematic review and meta-analysis of > 11,000 euploid embryo transfers found that even after controlling for embryonic chromosomal status, women < 35 years had significantly higher ongoing pregnancy and live-birth rates than older women, implicating non-ovarian factors as important determinants of reproductive outcomes in advanced maternal age [93]. On this basis, the present review focuses on EnSCs, synthesizing their associations with senescence and underlying mechanisms to inform clinical assessment and targeted interventions. 2.3.1 Characteristics of Aging of EnSCs Cellular senescence exhibits paradoxical biological roles in many respects. On the one hand, cells exposed to stresses such as DNA damage, oxidative stress, or telomere shortening enter an essentially irreversible state of cell-cycle arrest [52]. Senescent cells display canonical features, including enlarged and flattened morphology, elevated senescence-associated β-galactosidase (SA-β-gal) activity, and upregulation of the cell-cycle inhibitors p16 and p21 [94]. They also accumulate markers of DNA damage and secrete a spectrum of pro-inflammatory cytokines, chemokines, and proteases collectively termed the SASP [95]. In moderation, senescence helps remove damaged cells, suppress tumorigenesis, and participate in tissue repair. However, excessive senescent cells that are not cleared can provoke chronic inflammation and impair tissue function. A similar “double-edged sword” operates in the endometrium: controlled, localized stromal senescence during normal pregnancy (e.g., regulated release of SASP factors during decidualization) fosters a receptive endometrial milieu, whereas excessive accumulation of senescent stromal cells diminishes endometrial receptivity [96]. These features are largely shared between “acute” and “late (chronic)” senescence; what differs is context and duration. Acute, stress-induced senescence is typically transient, spatially restricted, and tightly controlled, contributing to tissue remodeling and repair. In contrast, chronic, age-related senescence reflects the long-term persistence and accumulation of senescent cells due to impaired clearance, leading to sustained SASP exposure, chronic low-grade inflammation, and progressive tissue dysfunction [52]. In female reproduction and placentation, these two forms are often discussed together because they draw on the same canonical senescence machinery but operate on different timescales: controlled, acute senescence is harnessed as a physiological tool for decidual remodeling and trophoblast invasion, whereas chronic senescence reflects cumulative reproductive aging and becomes a source of pathological inflammation and impaired endometrial function [52, 96]. In reproductive medicine, premature senescence of EnSCs has been closely linked to adverse outcomes. In a 2020 study, Tomari et al. assessed stromal senescence using endometrial biopsies obtained on the oocyte pick-up day (2 days before embryo transfer) [97]. Compared with stromal cells from receptive endometrium, those isolated from non-receptive endometrium exhibited higher SA-β-gal activity, a classic senescence marker. Consistently, stromal cells from patients with successful pregnancies were more often in S phase (proliferative), whereas those from implantation-failure patients were more frequently arrested in G0/G1, reflecting the cell-cycle arrest characteristic of senescence. In line with this, expression of senescence-related genes CDKN1A and CDKN2A was significantly elevated in stromal cells from the implantation-failure group [36]. These findings indicate a negative correlation between the degree of stromal senescence in human endometrium and embryo implantation competence, suggesting that implantation failure may be attributable, at least in part, to excessive stromal senescence (Table 1). | State | Cell cycle profile | Senescence markers | Decidualization capacity | Transcriptomic features | |---|---|---|---|---| | Young reproductive age | Balanced G0/G1 and S phases | Low p21, p16, SA-β-Gal | High (IGFBP1, PRL) | ECM remodeling genes active | | Advanced reproductive age | G0/G1 arrest predomina-nt | p21, p16, SA-β-Gal | Reduced | Inflammatory gene enrichment | | RPL-associated | Persistent G0/G1 arrest | p21, p16, SASP cytokines | Severely impaired | Downregulation of decidual genes | | Chronic endometritis | Irregular cycle arrest | p21, p16, IL-1β, TNF-α | Impaired | Pathogen-response genes enriched | | Hypoxia-induced | G0/G1 arrest | HIF-1α, p21 | Reduced | Hypoxia-response signature | | State | Cell cycle profile | Senescence markers | Decidualization capacity | Transcriptomic features | Beyond marker-based assessments, genomic approaches have been used to delineate the impact of age on EnSCs. Erikson et al. compared the transcriptomes of endometrial stromal fibroblasts from premenopausal (or younger) and perimenopausal women and identified differential expression in more than 1000 genes [98]. Pathway analyses showed enrichment in cell-cycle–related processes (e.g., “cytoskeletal organization,” “fiber formation,” “actin stress-fiber formation”) and fibroblast functions (e.g., “fibroblast proliferation,” “fibroblast migration”), as well as multiple signaling pathways commonly implicated in cellular senescence. qPCR validation further indicated that progesterone receptor (PR) and estrogen receptor β (ESR2) levels were relatively lower in the younger group, whereas estrogen receptor α (ESR1) expression was higher [99, 100]. The study also compared transcriptomic features of endometrial mesenchymal stromal/stem cells (eMSCs) between age groups and performed an integrated analysis across four cell sets (younger and older fibroblasts, younger and older eMSCs). Notably, gene-expression profiles of stromal fibroblasts from perimenopausal women more closely resembled those of eMSCs from both age groups, implying that higher-age fibroblasts are less differentiated than premenopausal stromal cells—that is, aged stromal cells display a more stem-like, undifferentiated state [101]. Another study further supports the detrimental impact of age on stromal function. Berdiaki et al. performed in vitro functional assays on stromal cells from women aged 25–46, sampling in the proliferative phase and assessing proliferation by fluorescence-based methods [102, 103]. Results showed significantly higher proliferation in the 25–35-year group compared with the 36–46-year group. Expression of BMP2 and STAT3, which are key regulators of proliferation and differentiation, was likewise higher in the younger group. Upon in vitro decidualization, stromal cells from younger women exhibited significantly higher levels of decidual markers prolactin (PRL) and insulin-like growth factor-binding protein-1 (IGFBP-1) than those from older women [104, 105]. These data indicate age-related dysfunction in stromal cells, including dysregulated cell-cycle control, reduced proliferative capacity, and impaired decidualization. In parallel, aged stromal fibroblasts undergo transcriptomic shifts toward a more stem-like profile. Such age-associated changes may be mediated, at least in part, by alterations in steroid-hormone receptor expression (e.g., PR and ERα/ERβ). It is well known that stress-induced premature cellular senescence is accompanied by the emergence of the senescence-associated secretory phenotype, whose secreted factors can promote senescence of neighboring normal cells through autocrine and paracrine routes and can modulate the senescence response. Researchers have shown that extracellular insulin-like growth factor binding protein 3 plays an important role in paracrine induction of senescence in young MESC by regulating the activity of the PI3K and AKT pathway [106]. Previous studies have identified IGFBP7 as a key factor for decidualization [107], and together with other members of the IGFBP family, it plays a crucial role in senescence-related signaling. Siraj and colleagues [108] found that signaling mediated by reactive oxygen species and prostaglandins drives the release of IGFBP7. A neutralizing antibody against IGFBP7 can reduce senescence that is induced by the senescence-associated secretory phenotype, whereas exposure to IGFBP7 causes cells to enter a senescent state. IGFBP7 can bind insulin, which may suppress the antiaging and growth-promoting actions of insulin [109]. In addition, IGFBP7 may enhance IGF2 signaling by blocking the IGF1 receptor and by increasing the interaction with the IGF2 receptor, which promotes senescence [108]. These effects depend on the extracellular signal-regulated kinase pathway and the AKT pathway. IGFBP7 and activin A appear to regulate each other, which suggests the presence of a compensatory mechanism that guards against excessive senescence. IGFBP7 not only inhibits activin A but also interacts with its receptor, which may induce senescence through the SMAD pathway. The mechanisms that regulate cellular senescence are related to the fates of stromal and glandular epithelial cells, whose disruption is associated with infertility and other reproductive disorders. The delicate balance between cellular differentiation and senescence is essential for decidualization, whose disturbance may lead to abnormalities [110]. In summary, the aging of human endometrial stromal cells (HESCs) is defined by the core phenotype. The core phenotype of aging is irreversible G0/G1 phase arrest, elevated SA-β-gal, upregulation of p16/p21, and SASP release. These are closely related to non-receptive endometrium and implantation failure. Physiologic, transient, localized senescence may contribute to establishing the normal decidual microenvironment, but excessive or persistent senescence and SASP accumulation disrupts endometrial receptivity and becomes a key limiting factor for favorable reproductive outcomes. Future work urgently requires age-stratified longitudinal, single-cell, and multi-omics studies with functional validation to clarify causality and threshold effects, thereby enabling the development of actionable molecular diagnostics and individualized therapeutic strategies (Figure 3). 2.3.2 Factors That Make EnSCs Susceptible to Aging EnSCs play a pivotal role in reproduction, contributing to endometrial proliferation, decidualization, and embryo implantation. However, with advancing age or exposure to external insults, these cells are prone to senescence. Senescence of EnSCs not only impairs endometrial function but may also lead to adverse outcomes such as infertility and recurrent pregnancy loss. Multiple factors, including hormonal fluctuations, oxidative stress, gene mutations, inflammatory responses, and environmental pollutants, can accelerate this senescent process. Defining the factors that render EnSCs susceptible to senescence and elucidating their molecular mechanisms is therefore crucial for improving reproductive health and preventing senescence-related fertility problems. Endocrine hormones such as estrogen and progesterone are key regulators of EnSC decidualization and senescence. In vitro, the canonical decidualization cocktail leads to an accumulation of senescence markers [111]. Wang et al. showed that progesterone receptor B (PR-B) is essential for inducing senescence through interaction with FOXO1. MPA treatment produces G1 cell-cycle arrest and upregulates p21 and p16, indicating that PR-B/FOXO1 signaling is a critical driver of endometrial stromal senescence [112]. Similarly, Tsuru et al. demonstrated that knockdown of progesterone receptor membrane component-1 (PGRMC1) enhances FOXO1 expression and increases senescence in EnSCs treated with a cAMP analog and progesterone [113]. These findings identify FOXO1 as a downstream target through which hormonal signals promote senescence during decidualization. FOXO1 activation drives a subset of EnSCs into acute senescence via interleukin-8 (IL-8) signaling, helping establish the inflammatory milieu required for endometrial receptivity and tissue remodeling [14]. Prostaglandin E2 (PGE2) is another key inducer of decidualization [114]. The COX2/PGE2 axis can trigger senescence in human fibroblasts and lung cells, and PGE2-induced decidualization in vitro is accompanied by increased expression of senescence-associated molecules [115-117]. However, PGE2 signals through its receptor PTGER2 to activate protein kinase A (PKA) and induce decidualization without overt senescence, in contrast to the cAMP-driven pathway [111]. Kusama et al. investigated the EPAC2–calreticulin (CALR) signaling axis in the regulation of PGE2 production and found that knockdown of EPAC2 or CALR increases senescence markers, indicating that these pathways are required for appropriate, modest levels of senescence during decidualization. cAMP/PKA is a common downstream pathway of estrogen, progesterone, and PGE2 in inducing decidualization [118]. EPAC2 (exchange protein directly activated by cAMP) is a non-PKA cAMP effector; evidence indicates that EPAC2 prevents aberrant senescence during decidualization by modulating p21 and p53. Silencing EPAC2 increases SA-β-Gal activity, which is the classic senescence marker, underscoring EPAC2's role in restraining cAMP-induced senescence [119, 120]. Oxidative stress is also a widely used trigger of senescence in vitro, including exposure to ultraviolet radiation, ethanol, hydrogen peroxide (H2O2), and tert-butyl hydroperoxide (t-BHP) [121]. A common consequence of these stimuli is increased reactive oxygen species (ROS). ROS can induce senescence through multiple routes, including activation of the p38/MAPK, IL-1α/NF-κB, mTORC1, and IGF-1 pathways [122-124]. Moderate oxidative stress is essential for normal pregnancy, but excessive stress hampers the progression of decidualization. Single-cell RNA-seq of EnSCs has identified a decidualization-stage subpopulation with high oxidative stress and active senescence [125, 126]. Jin et al. showed that cigarette-smoke extract induces oxidative stress at the fetal–maternal interface in amnion mesenchymal cells (AMCs) and chorionic cells, with the resulting ROS activating p38/MAPK and promoting senescence; co-treatment with antioxidants or p38 inhibitors attenuates these effects [127]. UVB/UVA irradiation likewise increases senescence markers in primary human EnSCs. Antiphospholipid antibodies (aPL) promote ROS generation and activate the p38/MAPK pathway, thereby increasing senescence—a process that can be effectively suppressed by antioxidants or specific p38/MAPK inhibitors [18, 128]. Metabolic homeostasis shapes senescence both in vivo and in vitro, while the senescent state, in turn, feeds metabolic dysfunction (mitochondrial metabolism, glycolysis, lipid metabolism), creating a vicious cycle implicated in tumorigenesis and progression [129-131]. Evidence for metabolic drivers of decidual senescence is emerging. Recently, branched-chain amino acids (BCAAs), particularly leucine, were shown to induce decidual stromal cell (DSC) senescence, with marked upregulation of the leucine transporter SLC3A2 in senescent DSCs [111]. Leucine supplementation increased senescence markers via activation of the p38 MAPK pathway, whereas leucine depletion reduced senescence, highlighting a role for BCAAs in promoting decidual senescence [111]. Lipid-metabolic changes in senescent cells center on fatty acids, phospholipids, and cholesterol [132]. Lipid-metabolism reprogramming is linked to the establishment and maintenance of cellular senescence, and interventions that target lipid pathways may modulate senescent-cell burden [133-135]. Lanekoff et al. emphasized the role of lipid metabolism in decidual senescence by studying Trp53-deficient mice, where lipid alterations were associated with precocious decidual senescence and adverse pregnancy outcomes [136]. At implantation sites in these mice, changes in diacylglycerols (DGs) and oxidized phosphatidylcholines (Ox-PCs) were observed, indicating that lipid dysregulation contributes to uterine instability during pregnancy [136]. These findings underscore the importance of lipid metabolism in maintaining a supportive decidual environment and preventing premature senescence. Sirtuins (SIRTs) help cells regulate energy output by sensing cytosolic NAD+ levels [137]. Associations between SIRTs and aging have been demonstrated in disorders such as osteoarthritis and Alzheimer's disease. SIRT1 levels are reduced in senescent decidua, and mice with uterine SIRT1 deletion exhibit premature aging phenotypes and impaired decidualization [138, 139]. Resveratrol, a natural SIRT activator, reduces SA-β-Gal activity and p53 expression in HESCs [140, 141]. As noted, a pro-inflammatory phenotype is a hallmark of senescent cells, and inflammatory signaling can also promote senescence [142]. In particular, IL-1β acts via the JNK pathway as a key driver of EnSC senescence. For example, IL-17RB stimulation enhances NF-κB signaling and increases secretion of SASP factors (including IL-1β, IL-6, and IL-8). IL-1β has been shown to suppress endometrial organoid growth and induce senescence through JNK signaling [143, 144]. Exposure of EnSCs to IL-1β accelerates senescence via JNK, elevating pro-inflammatory cytokines such as IL-6, IL-8, and TNF-α; conversely, JNK inhibition mitigates these effects, underscoring the role of IL-1β-mediated inflammation in disrupting decidualization and impairing fertility [145]. In sum, EnSC senescence is a multifactorial, complex process involving hormonal regulation, oxidative stress, inflammatory signaling, and other biological mechanisms. As research advances, our understanding of these pathways is deepening, opening new avenues to intervene in senescence and mitigate its impact on reproductive health. Continued investigation will not only benefit women's reproductive health but also inform prevention and therapy for other senescence-related diseases. 2.3.3 The Impact of EnSCs Aging on Decidualization in Physiological and Pathological Conditions Based on existing data on EnSC senescence and decidualization, two scenarios can be posited: a physiological state and a pathological state. The proliferating, cyclic EnSC compartment is heterogeneous, containing cells of different replicative “ages”, including youthful cells and cells in a pre-senescent state [146]. In a healthy endometrium, the EnSC pool is expected to contain only a very small fraction of fully senescent cells, because such cells are rapidly recognized and cleared by immune cells in response to their SASP. The mid-luteal hormonal switch initiates decidualization. During this process, undifferentiated EnSCs pass through several stages with distinct gene-expression and secretory profiles, ultimately yielding two divergent subpopulations—mature decidual cells and senescent decidual cells. The combined secretome of these differentiating cells elicits an initial pro-inflammatory decidual response that renders the endometrium receptive and “opens” the window of implantation to permit further blastocyst invasion. Notably, the proportion of pre-senescent cells within the undifferentiated EnSC pool modulates the intensity of this inflammatory priming. As decidualization progresses, temporal shifts in the secretome recruit uNK cells, which then eliminate the senescent decidual cells generated during differentiation, allowing the tissue to complete maturation and support successful implantation. Mechanistically, the transcription factor FOXO1 induces acute senescence in a subset of EnSCs via an IL-8–dependent pathway, driving the transient pre-implantation inflammatory response required for receptivity [147, 148]. Subsequently, under the influence of IL-15, uNK cells selectively clear these senescent decidual cells through granzyme-mediated cytotoxicity, effecting endometrial remodeling and “renewal” and ensuring proper opening of the WOI [105]. By contrast, undifferentiated EnSC pools may harbor residual, uncleared fully senescent cells. These terminal aging EnSCs no longer proliferate, their differentiation potential is significantly reduced, and they continue to secrete SASP factors, maintaining a chronic pro-inflammatory microenvironment, disrupting extracellular matrix structure, and promoting aging of adjacent healthy cells through paracrine effects [105]. When cyclical hormones prompt young and pre-senescent EnSCs to differentiate and acquire decidual traits, the fully senescent cells remain largely unresponsive and arrested. The persistent inflammation maintained by senescent EnSCs perturbs the finely tuned coordination of secretory programs among decidual subpopulations. Consequently, the composition and timing of factors that normally confer receptivity become deranged, creating an endometrial milieu unfavorable to implantation. Consistent with this model, clinical observations show a higher proportion of senescent cells and significantly elevated expression of pro-inflammatory SASP factors (e.g., IL-6, IL-8) in EnSCs from patients with implantation failure compared with those with normal implantation capacity [105, 149]. Importantly, even with the involvement of uNK, both senescent cells generated during decidualization and pre-existing senescent cells in tissues cannot be completely eliminated under pathological conditions [105]. The resulting accumulation and long-term persistence of senescent cells diminishes endometrial plasticity and impairs the sensing and selection of embryonic signals. In sum, excessive EnSC senescence coupled with impaired clearance disrupts normal decidualization, reduces endometrial receptivity, and increases the risk of implantation failure or miscarriage (Figure 4). Studies indicate that decidual senescence increases as pregnancy progresses, and that accelerated decidual aging is associated with preterm birth [150, 151]. Decidual senescence may also play an important role in the initiation of labor and thus help determine the timing of parturition. Here, we focus on decidual senescence in early pregnancy and explore its potential impact on embryo implantation. As decidualization proceeds, a suite of senescence-associated molecules changes dynamically, indicating ongoing interplay in senescence levels [152, 153]. The balance between senescence and decidualization is further regulated by key mediators of cell-cycle arrest and cellular aging, including p53, p16, and p21 [154, 155]. These proteins are upregulated under stress and govern cell growth and senescence. Accumulation of senescent cells reduces endometrial receptivity, leading to impaired implantation and adverse pregnancy outcomes. It has been reported that p53 deficiency causes a marked increase in p21 expression [156, 157]. However, targeting mammalian target of rapamycin complex 1 (mTORC1) or deleting the p21 gene in mice substantially slows the occurrence of preterm birth and improves fetal survival [157]. Notably, p53 deficiency exerts little effect on early gestational events—ovulation, fertilization, embryonic development, and implantation—compared with wild-type mice; after implantation, however, p53-deficient uteri exhibit markedly restricted proliferation of decidual cells, resulting in terminal differentiation and premature senescence of the decidua [157]. Recent studies have shown that during decidualization some ESCs undergo irreversible cell cycle arrest, which leads to cellular senescence [158, 159]. These senescent decidual cells secrete a variety of proteins, including pro inflammatory cytokines and chemokines such as interleukin IL 6 and IL 8, which play a crucial role in regulating the local cellular environment. As noted, senescent cells typically display an inflammatory profile known as the SASP, which includes cytokines, chemokines, growth factors, matrix metalloproteinases (MMPs), and tissue inhibitors of metalloproteinases (TIMPs) [160]. In controlled amounts, the SASP supports tissue integrity and remodeling [154, 155]. Excessive SASP production, however, perturbs the maternal–fetal interface, promotes local inflammation, and increases miscarriage risk. Senescent decidual cells also generate elevated levels of reactive oxygen species (ROS), leading to DNA damage, mitochondrial dysfunction, and tissue degeneration [16, 161]. SASP driven signaling, which can induce senescence in healthy cells, also alters immune function and promotes cancer progression. Interleukin 1, which accelerates cellular senescence through JNK signaling, impairs the decidualization of EnSCs [162], and mice in which the tumor suppressor p53 is deleted in a uterus specific manner exhibit aberrant cellular senescence in the decidua that leads to preterm birth [106, 163]. This highlights a delicate balance in which senescent decidual cells regulate immunity and tissue homeostasis while posing risks when dysregulated. Through proteomic analysis, investigators identified plasminogen activator inhibitor 1 as a key mediator within the SASP and demonstrated with CRISPR Cas9 that endogenous plasminogen activator inhibitor 1 plays a central role in paracrine induction of senescence, and the study further proposed that elevated levels of this inhibitor may interfere with trophoblast invasion and embryo implantation, which is associated with clinical problems such as recurrent miscarriage and implantation failure. Decidual cells also play a vital role in uterine hemostasis, menstrual regulation, and placental development by modulating plasminogen activator inhibitor-1 (PAI-1) [164]. PAI-1 is not only a biomarker of cellular and organismal aging but also a key mediator that contributes to age-related morbidity and represents a potential therapeutic target for delaying aging and age-associated diseases [165]. Although direct evidence for regulatory crosstalk between decidual senescence and PAI-1 is currently insufficient, PAI-1 is a senescence marker that tends to increase during decidualization and thus warrants clarification in future experiments. Moreover, research in 2024 showed that CDC42 plays a crucial role in regulating the senescence of EnSCs in such a way that decidualization is maintained, while also suggesting that inhibitors of Wnt signaling could serve as potential therapeutics to mitigate endometrial senescence [13]. Several groups have likewise shown that activation of DIO2 promotes physiological senescence of decidual cells and functions in the early stages of implantation [166-168]. Elevated DIO2 expression has been detected in endometrial and placental samples from women with RPL, implicating DIO2 in pathological decidual senescence. Rapamycin significantly reduces DIO2 and SA-β-gal expression and lowers the proportion of senescent cells during decidualization-particularly in senescence-induction models—thereby facilitating normal embryonic development by diminishing SASP secretion and cellular debris [168, 169]. Single-cell sequencing studies indicate that a subset (but not all) of cells undergoes pronounced senescence characterized by DIO2 and SCARA5 expression. The FOXO1-DIO2 signaling axis is thought to trigger decidual senescence in early pregnancy and during implantation. FOXO1 directly regulates DIO2, and FOXO1 knockdown suppresses DIO2 expression and decidual senescence in EnSCs, whereas FOXO1 activation elevates DIO2 and promotes a senescent decidual phenotype [168]. This regulatory mechanism is essential for the acute senescence that supports tissue remodeling and implantation; when dysregulated, it may lead to pathological outcomes such as RPL and implantation failure. In summary, physiological decidualization is also accompanied by an increase in aging related molecules. A widely accepted view is that decidualization entails the emergence of a senescent decidual stromal cell (DSC) subpopulation. Eliminating decidual senescence with dasatinib, a senolytic agent, does not benefit blastocyst expansion in co-culture systems, indicating that senescence is necessary [167]. The proportion of this senescent subset within DSCs appears particularly important. On the contrary, Deryabin et al. observed the opposite pattern, with stronger aging and weaker physiological decidualization. This provides indirect support for the notion that maintaining a steady state balance of aging is necessary [17, 154]. When this balance and clearance mechanism are disrupted, a pathological chain of decreased endometrial receptivity and adverse pregnancy outcomes occurs (Figure 4). 2.4 The Relationship Between EnSC Aging and Recurrent Miscarriage 2.4.1 Enscs and Rpl Recent studies suggest that excessive senescence of EnSCs may be closely linked to recurrent pregnancy loss (RPL) [170-172]. In the normal luteal phase cycle, when EnSCs differentiate into decidual stromal cells (DSCs), the cells enter an acute aging state, forming senescent DSCs (snDSCs) that secrete pro-inflammatory factors to help the endometrium receive them. Subsequently, uterine natural killer (uNK) cells are activated to clear these senescent cells, thereby maintaining decidual tissue homeostasis. If snDSCs accumulate excessively during decidualization, or if uNK-mediated clearance declines, the balance of the decidual environment is disrupted, potentially leading to abnormalities at the embryo–maternal interface and adverse pregnancy outcomes. Single-cell transcriptomic evidence supports this model. Luo et al. analyzed mid-secretory endometrium and identified type II iodothyronine deiodinase (DIO2) and scavenger receptor class A member 5 (SCARA5) as markers of senescent versus non-senescent decidual cells, respectively [173]. In peri-implantation endometrium from women with RPL, DIO2 expression was significantly increased, whereas SCARA5 was decreased, indicating a higher proportion of senescent DSCs compared with normal pregnancy [174, 175]. Researchers further demonstrated that the aging scores of DSCs in RPL patients were elevated, while the expression of typical aging regulatory factors p53 and p16 was increased, and the expression levels of various inflammatory and metabolic related genes were elevated [176, 177]. Mechanistically, tumor necrosis factor-α (TNF-α) and its receptor TNFR1 were upregulated in RPL decidua. In vitro, excess TNF-α activated the p53/p16 pathway through TNFR1, inducing excessive senescence of DSCs during decidualization [178]. These findings point to an aberrant, pro-senescent decidual response as one pathological mechanism underlying miscarriage. Intriguingly, a measured degree of cellular senescence also appears necessary for normal pregnancy. Insufficient senescent cells in the endometrium may likewise predispose to implantation failure or miscarriage. In a clinical study of 311 patients, immunohistochemistry for p16 (a senescence marker) in mid-secretory endometrium showed that the proportion of p16-positive cells correlated with outcomes: women who ultimately delivered had significantly higher p16 positivity in luminal and glandular epithelium than those who miscarried (e.g., luminal epithelium 35.2% vs 11.7%; glandular epithelium 9.3% vs 2.9%; both p < 0.001). Conversely, low p16 positivity predicted implantation failure or miscarriage; proposed cut-offs were < 12.5% (luminal) and < 3.2% (glandular). Thus, too few senescent cells may signal suboptimal receptivity and increased miscarriage risk. Taken together, both excessive decidual senescence and insufficient physiological senescence can negatively impact pregnancy, underscoring the need to maintain an appropriate balance [179]. Beyond cell-level phenomena, the endometrium also exhibits time-dependent “aging”. Epigenetic-clock studies provide evidence: using the Horvath clock (genome-wide DNA methylation), Olesen et al. estimated the “biological age” of the endometrium. In nine women aged 18–38, biopsied on LH+ across two consecutive cycles, epigenetic age correlated with chronological age, yet endometrial tissue often appeared “older” than the individual—one extreme case exceeded chronological age by 13 years. Although sample size was small and pregnancy outcomes were unavailable, the study demonstrated age-related epigenetic change in human endometrium and raised the hypothesis that premature endometrial aging may underlie some unexplained infertility, particularly when endometrial factors predominate [180]. Addressing this hypothesis, Yang et al. conducted a large-scale transcriptomic analysis linking premature endometrial aging to recurrent implantation failure (RIF). Mid-secretory endometrial transcriptomes from 245 RIF patients 40. Unsupervised clustering revealed two molecular subtypes among younger RIF patients: an immune-active cluster and a metabolism-active cluster. Cluster 1 (immune-active) showed upregulated immune pathways and higher expression of receptivity-associated genes; cluster 2 (metabolism-active) more closely resembled the older control group, with upregulation of mitochondrial function, lipid metabolism, and steroidogenesis. Clinically, cluster 2 exhibited a higher rate of window-of-implantation displacement, poorer receptivity, and lower implantation success than cluster 1. In other words, despite their youth, these patients displayed endometrial molecular features akin to advanced age—a premature-aging phenotype—closely associated with RIF. The study concluded that a subset of young women indeed harbors prematurely aged endometrium, and that this state is tightly linked to poor reproductive outcomes. For some younger patients with RIF, the clinical problem may thus reside not in the embryo, but in a “prematurely aged” endometrial environment [181]. In summary, the degree of EnSC/decidual senescence must be held within an optimal range: excessive senescence elevates miscarriage risk, whereas an insufficient physiological senescence response may impair receptivity and implantation. The concepts of endometrial biological age and premature aging constitute a relatively new but important research frontier that may explain a fraction of unexplained infertility and miscarriage. Nevertheless, our understanding of the causes and features of premature endometrial aging remains incomplete. Further mechanistic work and exploration of potential interventions are urgently needed to improve pregnancy outcomes in affected patients. 2.4.2 Stem Cell Depletion and Accelerated Aging The endometrium has a remarkable capacity for cyclic regeneration, underpinned by a stem-cell pool in the basalis, particularly HESCs. These stem cells are activated after menstruation to drive rapid repair and reconstruction of the functionalis, thereby maintaining endometrial structural integrity and functional adaptability. In women with RPL, however, depletion and dysfunction of this stem-cell pool have increasingly emerged as key constraints on endometrial regenerative potential [182-184]. Investigators have reported a marked reduction of HESCs in the endometrium of women with repeated miscarriages, together with abnormal retention of epigenetic marks (e.g., H3K27me3), indicating severely compromised stem-cell activity and possible functional exhaustion. Stem-cell depletion not only weakens endometrial regeneration but may also precipitate a vicious cycle of stromal-cell senescence. Each pregnancy loss injures the endometrium and compels the remaining stem cells to proliferate continuously to meet repair demands. Prolonged hyperproliferation accelerates telomere attrition; once telomeres fall below a critical threshold, the p53/p21 pathway is activated, driving irreversible growth arrest—i.e., replicative senescence [185, 186]. At the same time, aging stromal cells secrete a powerful SASP, which is rich in IL-6, IL-8, and MMP10 and can activate NF-κB signaling, recruit immune cells such as macrophages, and consolidate the pro-inflammatory microenvironment, thereby amplifying local tissue damage and aging cues [187, 188]. Ultimately, this triad—stem-cell depletion, stromal senescence, and inflammatory amplification—forms a positive feedback loop that directly diminishes the endometrium's responsiveness to pregnancy signals. In vitro, EnSCs from RPL patients exhibit pronounced functional deficits: even under exogenous hormonal stimulation, expression of decidual markers (PRL, IGFBP1) reaches only 30%–50% of that in controls. These cells also display reduced mitochondrial activity and a metabolic shift from oxidative phosphorylation to less efficient glycolysis, failing to meet the high energy demands of decidualization and early embryonic development—thereby providing a metabolic basis for pregnancy failure. In sum, by eroding regenerative capacity, inducing stromal senescence, and remodeling the inflammatory milieu, stem-cell depletion establishes a self-reinforcing pathological process that restricts endometrial plasticity and substantially elevates RPL risk, making it a critical target for intervention [188, 189]. 2.4.3 Chronic Inflammation and Microenvironment Imbalance Chronic inflammation and microenvironmental imbalance are thought to play key roles in early pregnancy pathology. Studies show that the decidualization competence of EnSCs is tightly linked to their senescent state. Both impaired decidualization and premature EnSC senescence can lead to adverse outcomes, including RPL and RIF. RPL is commonly defined as two or more consecutive clinical pregnancy losses [190]. RIF refers to failure of implantation after at least three consecutive IVF cycles in which 1-2 high-quality embryos were transferred each time. Successful implantation depends on both embryonic developmental potential and endometrial receptivity. In patients with RPL or RIF, however, the cyclical differentiation of EnSCs into decidual cells is often compromised; suboptimal decidualization prevents the endometrium from achieving an optimal implantation state and markedly increases the risk of pregnancy failure. Aberrant pro-inflammatory responses during decidualization further support this view: primary EnSCs from RPL patients exhibit persistent, highly dysregulated secretion of inflammatory mediators following decidualization induction [191]. By contrast, in women with successful implantation, post-decidualization inflammatory secretion by EnSCs is tightly time-regulated, typically showing a sharp cytokine peak on Day 2 after induction that rapidly subsides [192]. In implantation failure, this dynamic rhythmicity is lost: there is no clear peak, and inflammation remains elevated for a prolonged period. This sustained pro-inflammatory state closely resembles the SASP. On this basis, authors have hypothesized that sporadic implantation failure may stem from cycle-to-cycle fluctuations in the ratio of mature decidual cells to senescent decidual cells, whereas RIF may involve damage or deficiency in the progenitor EnSC pool [193, 194]. Work on chronic endometritis (CE) likewise provides strong evidence linking long-standing inflammatory dysregulation, impaired decidualization, and RPL/RIF. CE is characterized by persistent, low-grade inflammation of the endometrium and is increasingly recognized as a risk factor for RPL and RIF [195, 196]. Although the causal relationship remains to be fully resolved, abnormal secretion of cytokines and paracrine factors within the endometrium, together with compromised EnSC decidualization, are widely believed to be important contributors [197]. In short, chronic, inappropriate inflammation can erode endometrial receptivity and thereby adversely affect pregnancy. A representative study further delineated a causal link between premature EnSC senescence and susceptibility to RPL. Researchers cultured primary EnSCs from healthy donors and RPL patients, and analyzed mid-luteal endometrial biopsies from RPL and subfertile women. The findings suggest that RPL may originate from a deficiency of MSCs and premature EnSC senescence [198]. Crucially, EnSCs driven into premature senescence in vitro mounted a sustained, high-level pro-inflammatory secretory response during decidualization, closely mirroring the secretion profile of EnSC cultures from RPL patients. Thus, premature EnSC senescence and the associated inflammatory tissue state substantially increase RPL risk. A unifying hypothesis for RPL and RIF posits that chronically dysregulated secretion of pro-inflammatory factors driven by premature EnSC senescence, prolongs the “window of implantation,” allowing embryos to attach outside the normal time frame. The consequences are twofold: low-quality embryos may implant during an abnormally extended window, and high-quality embryos may implant into an unfavorable endometrial micro-environment. Either scenario restricts embryonic development and increases miscarriage risk. Such dysregulation of endometrial receptivity is considered a major mechanistic basis for RPL and RIF. Notably, EnSC senescence is often accompanied by chronic inflammation and local microenvironmental imbalance. Senescent EnSCs secrete large amounts of SASP factors that remodel the local immune milieu, markedly perturbing the distribution, phenotype, and function of immune cells and thereby disrupting homeostasis at the maternal-fetal interface [199, 200]. This senescence- and inflammation-driven microenvironmental shift is viewed as a key contributor to embryonic developmental failure-in other words, sterile chronic inflammation initiated by EnSC senescence steers the interface in a direction unfavorable to the embryo. Under normal decidualization, uterine natural killer (uNK) cells are the dominant population of uterine innate lymphocytes and are essential for a healthy gestational environment [111, 155, 172]. The phenotypic changes of uNK cells and the polarization imbalance of macrophages are the two most core and functional consequences of the interaction between senescent cells and the immune microenvironment, directly determining the inflammation tolerance balance at the maternal-fetal interface. They are not only passive responders, but also regulate the fate of decidual stromal cells through the secretion of cytokines, chemokines, and protease feedback. uNK cells support placental angiogenesis by secreting pro-angiogenic and growth-promoting factors such as VEGF and help maintain early-pregnancy immune tolerance [201]. They also regulate trophoblast invasion of the decidua and remodeling of spiral arteries, thereby providing the immune and hemodynamic support required for implantation and development. In a pathological uterine microenvironment dominated by senescent EnSCs, however, uNK-cell function becomes markedly dysregulated [202]. On one hand, expression of key receptors such as KIR2DL1/S1 and LILRB1 on uNK cells is significantly reduced in women with a history of reproductive failure [203]. These receptors recognize HLA-C and HLA-G on trophoblasts. Their downregulation weakens uNK activation and function, impairing normal uNK-trophoblast interactions. Consequences include insufficient secretion of angiogenic factors by uNK cells, reduced cytotoxic fine-tuning, and inadequate support for decidual vascular remodeling and placental development. On the other hand, pro-inflammatory uNK subsets increase abnormally [204]. In the endometrium of RPL and RIF patients, a higher proportion of NK cells expressing cytotoxic markers has been observed, implying an overall shift toward heightened cytotoxic potential. Over-activated NK cells secrete large amounts of pro-inflammatory cytokines (e.g., TNF-α and IFN-γ), which may directly injure trophoblasts and disrupt normal trophoblast invasion [166]. In RPL, inhibitory receptor Tim-3 is also significantly downregulated on peripheral NK cells, biasing them toward a cytotoxic phenotype and depriving them of proper inhibitory cues [166]. This undermines maternal immune tolerance and further weakens the protection of the embryo. Overall, quantitative and functional disturbances of uNK cells represent a major facet of how a senescent, inflammatory environment derails immune regulation at the maternal–fetal interface. Concurrently, macrophage polarization is also aberrantly skewed in this senescent, inflamed microenvironment. Senescent EnSCs release abundant damage-associated molecular patterns (DAMPs), notably high-mobility group box 1 (HMGB1) [205, 206]. By engaging TLR2/4 and activating NF-κB signaling on macrophages, HMGB1 and related DAMPs drive polarization toward the classically activated M1 phenotype [18]. In the decidua of RPL patients, macrophages preferentially adopt M1 polarization, while reparative M2 polarization is suppressed. Increased M1 macrophages secrete high levels of pro-inflammatory cytokines, including IL-1β, IL-6, and TNF-α, exacerbating local inflammation and potentially inducing inflammatory cell death in neighboring cells, thereby directly damaging placental structures [207, 208]. Conversely, M2 macrophages, which in normal pregnancy secrete anti-inflammatory mediators such as IL-10 and TGF-β, facilitate spiral-artery remodeling, and promote the induction of regulatory T cells (Tregs) to maintain tolerance, which are markedly reduced [209]. In a senescence-associated, imbalanced inflammatory milieu, these critical anti-inflammatory signals are insufficient: decidual vascular remodeling is hindered, Treg induction is blunted, and immune homeostasis at the interface is disrupted. In summary, chronic inflammation and micro-environmental imbalance caused by EnSC aging disrupt the fine coordination between maternal-fetal interface vascular remodeling, immune regulation, and trophoblast invasion, mainly through multi-layer effects on uNK cells and macrophages, significantly increasing the risk of developmental failure and miscarriage after implantation. From an immunological perspective, this describes a pathogenic pathway where interface dysregulation accelerates embryo death. Each abnormal step amplifies the next step, ultimately forming a vicious cycle of “aging inflammation immune dysfunction,” which explains the clinical pattern of recurrent miscarriage in RPL patients. 2.4.4 Telomere Shortening and Mitochondrial Dysfunction Telomere attrition and mitochondrial dysfunction are core features of EnSC senescence. Acting in concert, they drive progressive deterioration of the endometrial milieu and constitute a key pathological basis for RPL [210]. With each additional pregnancy failure, these cellular insults worsen, markedly impairing decidualization capacity and embryonic support. Abnormal telomere dynamics are a hallmark of the senescent state. In a prospective cohort, EnSCs from women with RPL had telomeres 40% shorter on average than those of age-matched healthy controls, alongside significantly reduced expression of telomere-protective proteins [210]. The degree of telomere shortening correlated positively with the number of prior miscarriages, indicating cumulative endometrial injury and a steadily intensifying senescence phenotype. At the mitochondrial level, quality-control systems are compromised, further deranging cellular energetics. Decline in mitophagy is particularly critical: reduced activity of the PINK1/Parkin pathway leaves damaged mitochondria uncleared, leading to their intracytoplasmic accumulation and excessive reactive oxygen species production [211]. Persistent ROS activates HIF-1α signaling and forces a metabolic shift from oxidative phosphorylation (OXPHOS) to glycolysis. However, ATP demand rises sharply during decidualization, and glycolytic flux cannot meet this load, precipitating an energy crisis that stalls the decidual program. In parallel, mitochondrial DNA (mtDNA) integrity deteriorates over the course of repeated losses: in endometrium from women with ≥ 3 miscarriages, mtDNA copy number declines by 30% on average, and common deletions, such as the 4977 bp lesion, are detectable [212, 213]. These mutations impair the synthesis of respiratory chain complex I subunits, reduce electron transport efficiency, increase electron leakage, and further amplify ROS, creating a vicious cycle that gradually weakens metabolism and stress response capabilities. Taken together, telomere shortening and mitochondrial dysfunction synergize to drive deep, structural-to-functional senescence of EnSCs. This sustained damage not only disrupts the decidualization and embryo support niche, but also lays the molecular foundation for the periodic recurrence of miscarriage, which is an urgently needed and precisely located node in RPL intervention (Figure 4). 2.5 Diagnostic Strategies Based on Endometrial Stromal Cell Aging As recognition of endometrial stromal cell senescence in the pathogenesis of RPL has deepened, early screening strategies oriented toward endometrial function are becoming a research focus. Traditional risk assessment for miscarriage relies largely on maternal age and prior miscarriage history. Although informative, these metrics struggle to identify individuals with covert endometrial dysfunction in a timely manner. To this end, emerging molecular biomarkers are being introduced with the aim of earlier and more precise risk stratification (Table 2). | Pathological features | Molecular markers/functional alterations | Detection techniques | Diagnostic value | |---|---|---|---| | Cell cycle arrest | p16/INK4a ↑, p21/CIP1↑ | IHC/Western Blot/qPCR | Directly reflects the irreversible senescent state | | Aberrant SASP secretion | IL-6 ↑, IL-8 ↑, MMP-3/9 ↑, CCL2↑ | ELISA | Assess the degree of inflammatory microenvironment imbalance | | Accumulation of DNA damage | γH2AX ↑, 53BP1↑ | IF/IHC | Indicates genomic instability and senescence persistence | | Impaired decidualization | PRL ↓, IGFBP1 ↓, FOXO1↓ | qPCR/Western Blot | Predicts decreased embryo implantation potential | | Epigenetic dysregulation | H3K27me3 ↓, Aberrant DNA methylation | ChIP-seq/WGBS | Reveals the underlying regulatory mechanisms of senescence | | Immune microenvironment imbalance | dNK cell activity ↓, Th1/Th17 Cells ↑, Treg Cells↓ | Flow cytometry analysis/scRNA-seq | Associated with maternal-fetal immune tolerance failure | | Angiogenesis iImpairment | VEGFA ↓, ANGPTL ↓, Reduced vascular density | IHC/RNA-seq | Reflects impaired placental support function | Among the candidate biomarkers, the PLA2G2A/DIO2 mRNA ratio in EnSC has been proposed as an indicator of cellular aging status and decidualization potential. In fact, 7 days after the surge in luteinizing hormone, endometrial biopsy was performed and the expression of matrix genes was quantified using digital droplet PCR. Then, calculate the AIIA type phospholipase A2 (PLA2G2), which is a progesterone-induced immediate early gene, and its transcription ratio to DIO2. DIO2 is a progesterone-inhibited thyroid hormone-activating enzyme, and its high expression indicates progesterone resistance [214, 215]. Thus, this ratio serves as a readout of stromal progesterone responsiveness and progression of the decidual transition. Prospective cohort data support this concept—when the ratio drops below 25%, it indicates that the decidual response is stagnant or obstructed, leading to a significant increase in the risk of miscarriage and a sharp decrease in live birth rate. After excluding miscarriages due to embryonic aneuploidy, the predictive value strengthens further: women with ratios < 25% have substantially lower odds of live birth than those with higher ratios. Collectively, these findings indicate that the PLA2G2A/DIO2 ratio is a useful indicator of endometrial functional aging and impaired decidual capacity, providing a tool to identify high-risk individuals [31]. To improve accessibility and patient adherence, investigators are developing noninvasive or minimally invasive alternatives. One direction is molecular analysis of uterine cavity fluid (endometrial lavage). Studies comparing lavage from early pregnancy controls with samples from RPL patients show significantly elevated inflammatory mediators, such as LPS, TNF-α, IL-2, and IFN-γ, in RPL, suggesting that this fluid reflects local endometrial pathophysiology. Parallel efforts are exploring mitochondrial DNA copy number and oxidative-stress markers such as 8-hydroxy-2'-deoxyguanosine (8-OHdG) in lavage. Early results indicate high concordance (85%) between lavage measurements and paired endometrial tissue assays, supporting lavage as a repeatable, noninvasive option for early screening of endometrial dysfunction and miscarriage risk without repeated biopsies [216, 217]. In sum, next-generation early screening strategies that integrate molecular biology with regenerative medicine are moving beyond traditional, clinic-only metrics to deliver more precise, forward-looking risk prediction and personalized intervention for RPL. Diagnostic methods based on EnSC aging, such as key gene ratio determination, non-invasive biomarker analysis of endometrial lavage, and organ-based functional testing, can help identify high-risk patients early and lay the foundation for targeted therapy. As these technologies mature and are validated clinically, they are poised to play a greater role in practice—reducing unexplained recurrent miscarriage at its source and improving pregnancy outcomes and reproductive health. 2.6 Intervention Strategies: Delaying or Reversing Endometrial Aging 2.6.1 Senolytics and Pharmacologic Modulation of Senescence Because senescent cells accumulate with age and secrete pro-inflammatory SASP factors that disrupt tissue homeostasis, senolytics have been developed to eliminate these cells and reduce their deleterious effects selectively. Preclinical studies show that senolytics can improve tissue function and extend lifespan in animal models; for example, the combination of dasatinib and quercetin reduces senescent-cell burden and ameliorates age-associated phenotypes [218-220]. Ongoing clinical trials are evaluating the safety and efficacy of senolytic therapies in humans. Several drugs with potential anti-aging effects are under investigation. Metformin and rapamycin can attenuate senescent phenotypes in EnSCs prior to decidualization and suppress SASP expression, improving embryo expansion impeded by senescent cells in in vitro implantation models [17, 221]. However, their effects on functional restoration differ: only metformin reverses the inhibitory effect of senescent EnSCs on decidual signaling in a healthy uterine environment, whereas rapamycin shows no significant efficacy in this respect, suggesting a narrower therapeutic window for rapamycin. Rapamycin, an mTOR inhibitor, has been shown to extend lifespan across species by modulating pathways involved in cell growth and metabolism [222, 223]. In addition, the tyrosine-kinase inhibitor dasatinib can selectively clear abnormally activated senescent ESC subsets by inducing apoptosis, demonstrating strong senolytic capacity [224]. Metformin, a widely used antidiabetic agent, has shown potential to extend lifespan and delay age-related diseases in preclinical studies. It exerts anti-senescent and anti-inflammatory effects by inhibiting NF-κB signaling and SASP release [225]. Low-dose interleukin-2 (IL-2), as an immunomodulatory therapy, selectively expands regulatory T cells (Tregs), thereby alleviating immune imbalance induced by a senescent microenvironment [226, 227]. Although these agents are promising, rigorous clinical trials are needed to establish efficacy and safety in humans. 2.6.2 Stem-Cell Based Regenerative Strategies Stem-cell therapy, a burgeoning regenerative strategy, holds broad promise for improving EnSC function, alleviating cellular senescence, and enhancing pregnancy outcomes [228]. By restoring the endometrial stem-cell pool, strengthening stromal regeneration, and modulating the SASP, stem cells and their derivatives offer novel therapeutic avenues for RPL [228]. Preclinical work shows that autologous bone-marrow-derived MSCs exhibit good homing capacity, migrate to the endometrium, and differentiate into functional HESCs, thereby improving renewal and regenerative potential. In a phase I trial of 10 refractory RPL patients, MSC transplantation doubled colony-forming units (CFUs) within 6 months, and 60% achieved live birth—providing preliminary evidence of safety and efficacy. Beyond homing and differentiation, MSCs act via paracrine signaling: MSC-conditioned-medium-derived exosomes, enriched in miR-21, miR-29a, and other effectors, can reverse telomere attrition and suppress aberrant SASP secretion, ameliorating the senescent micro-environment [229]. In animal models, intrauterine perfusion of exosomes raised live-birth rates from 20% to 75%, underscoring strong reproductive support potential. In sum, whether by cellular replacement or extracellular signaling, stem-cell-based approaches offer significant prospects for improving endometrial aging, enhancing decidualization, and increasing live-birth outcomes. With ongoing mechanistic insights and clinical trials, this strategy may become a breakthrough for precision intervention in RPL. 2.6.3 Herbal and Natural Products: Focus on Quercetin Given the profound impact of EnSC aging on endometrial receptivity and decidualization, there is growing interest in drug interventions, particularly herbal and natural products, to improve uterine receptivity, enhance implantation, and alleviate related diseases. Among these, the natural flavonoid quercetin has attracted attention due to its multitarget actions and relatively low toxicity. Widely present in fruits, vegetables, and grains, quercetin exhibits antioxidant, anti-inflammatory, and immunomodulatory activities and has recently been shown to possess senolytic properties. As the role of cellular senescence in endometrium-related conditions has become clearer, the potential of quercetin to modulate EnSC senescence and dysfunction has gained support, particularly for correcting decidualization defects and improving the adverse uterine environment associated with endometriosis. Experimental data indicate that quercetin can selectively induce apoptosis of EnSCs with a senescent phenotype, thereby reducing the accumulation of senescent cells in the endometrium. At the same time, quercetin suppresses secretion of multiple SASP factors (e.g., IL-6, MMP-3) and downregulates canonical senescence markers p16 INK4a and p21 Cip1/Waf1, thereby attenuating chronic inflammation and tissue dysfunction driven by senescent EnSCs. Quercetin also restores EnSC function. By inhibiting aberrant activation of AKT and ERK1/2 and enhancing the expression and activity of p53, it promotes decidualization. In EnSCs from both healthy donors and patients with endometriosis, quercetin significantly increases secretion of decidual markers IGFBP1 and PRL, indicating recovery of decidualization capacity [230]. Notably, in EnSCs derived from endometriosis patients, quercetin improves decidualization to near-normal levels. Given that endometriosis is a major cause of infertility and miscarriage—with mechanisms including anatomic distortion, chronic pelvic inflammation, reduced ovarian reserve, immune dysregulation, and declining endometrial receptivity—quercetin's multitarget profile and favorable side-effect spectrum confer unique advantages for improving the endometrial environment. Animal studies support this potential: in a rat model of endometriosis, quercetin reduced lesion burden and improved the decidual response (Table 3). | TCM | Primary target/pathway | Mechanism of action | Evidence type | Potential relevance to EnSCs | References | |---|---|---|---|---|---| | Resveratrol | SIRT1 activation | Enhances deacetylation, reduces ROS | In vitro, animal | Delays EnSCs senescence | [231] | | Curcumin | NF-κB inhibition | Suppresses SASP factor transcription | In vitro, clinical | Reduces inflammatory microenvironment | [232] | | Berberine | AMPK activation | Improves mitochondrial function, induces autophagy | In vitro, animal | Restores metabolic homeostasis | [233] | | Ginsenoside Rg1 | p53/p21 modulation | Prevents premature senescence | Animal | Supports decidualization | [234] | | Astragaloside IV | Telomerase activation | Extends telomere length, delays senescence | In vitro, animal | Preserves proliferative potential | [235] | | Danshensu | p38 MAPK inhibition | Reduces oxidative stress-induced senescence | In vitro | Protects ECM structure | [236] | Telomerase-activating strategies include small-molecule activators and gene-therapy approaches. Resveratrol, a natural polyphenol, has been identified as a potential telomerase activator with anti-aging activity. Gene therapy delivering the TERT gene has shown promise in extending lifespan in mouse models [237]. However, because unchecked cell proliferation may increase cancer risk, development in this area must proceed cautiously [238]. Gene therapy offers the potential to directly modify genetic factors involved in aging—either by delivering genes that promote cellular rejuvenation or by inhibiting genes that drive senescence [55]. In animal studies, upregulating Klotho expression has been associated with lifespan extension. While gene therapies targeting senescence-related genes have been explored, they remain experimental and require extensive validation [239]. In addition, emerging data indicate that RNA-binding proteins and RNA methylation regulators, such as m6A “writers” and “erasers” (e.g., the methyltransferase METTL3 and the demethylase FTO), govern the stability and translation of transcripts involved in inflammation, metabolism, and senescence. This epitranscriptomic regulation may act as a fine-tuning mechanism for adaptive stress responses during aging [218, 240]. Finally, noncoding RNAs, including microRNAs and long noncoding RNAs, are increasingly recognized as key modulators of senescent phenotypes, influencing chromatin remodeling, SASP regulation, and telomere maintenance [240]. These insights provide novel therapeutic targets for aging-related interventions and highlight the value of integrative approaches to unravel the complexity of senescence. 2.6.4 Challenges Overall, interventions targeting endometrial aging and EnSC senescence are still at an early stage of transition from proof-of-concept to clinical translation, and the current evidence shares several general limitations. First, the extrapolability of existing models is limited: many studies are based on in vitro cultured EnSCs or mouse models, which cannot fully recapitulate the cyclic remodeling of the human endometrium, its endocrine regulation, or the unique immune tolerance required for pregnancy; therefore, caution is needed when directly extrapolating these findings to clinical practice. Second, most available clinical and translational studies are single-center with small sample sizes, and enrolled patients differ markedly in age, number of RPL events, prior treatments, and comorbidities, resulting in limited statistical power and making it difficult to reliably identify the subgroups most likely to benefit. Third, outcome measures are heterogeneous and lack unified standards: different studies have used endpoints ranging from cellular senescence markers and histological improvement of the endometrium to embryo implantation rates and live-birth rates, and there is still no standardized, comparable evaluation system that spans the full continuum from “endometrial aging–decidualization–reproductive outcomes.” Fourth, data on long-term safety are scarce. For senolytic agents, immunomodulators, and stem cell- or gene/epigenetic-based interventions alike, follow-up periods are mostly on the scale of months to a few years, and evidence remains insufficient regarding potential risks such as tumorigenesis, immune dysregulation, and effects on offspring growth and development. Finally, individualized strategies remain largely conceptual. Although multiple studies suggest that both “senescence phenotypes” and “senescence drivers” differ substantially among patients, there is still no widely validated “endometrial aging classification” or “EnSC senescence score,” and precision stratification and personalized intervention remain mostly at the level of theory and small exploratory cohorts. Looking ahead, there is a need for multicenter, large-sample, prospective cohorts and randomized controlled trials that tightly link findings from cellular and animal models to reproductive outcomes, and for the development of standardized senescence assessment systems to clarify the indications and risk boundaries of different interventions. On this basis, the rational integration of pharmacologic therapies, regenerative strategies, and molecularly targeted approaches within a precision-medicine framework may ultimately yield effective and safe interventions to delay or reverse endometrial aging and improve reproductive outcomes in patients with RPL and related disorders. 3 Summary and Prospect Endometrial stromal cell senescence is increasingly recognized as an important pathogenic mechanism in RPL. Hallmark features include depletion of the stromal stem cell pool, telomere attrition, mitochondrial dysfunction, and an imbalance between DIO2+ and PLA2G2A+ functional stromal subpopulations. This perspective reframes conventional paradigms of miscarriage pathophysiology, emphasizing the central roles of endometrial plasticity and microenvironmental adaptability in sustaining pregnancy. Recent systems biology and multi-omics methods have expanded our understanding of aging, revealing novel regulatory networks beyond classical pathways and driving progress in aging research. High-throughput transcriptomics, proteomics, metabolomics, and single-cell sequencing have identified emerging participants in regulating age-related processes. Therefore, future research on the aging mechanism of EnSCs should focus on the following key directions. Firstly, a high-resolution single-cell spatiotemporal map is constructed to systematically depict the dynamic changes of stromal cell subsets in patients with recurrent miscarriage at various stages of the menstrual cycle, identifying key pathological subtypes and potential therapeutic targets. Secondly, the endometrial organoid model should be further optimized, especially based on patient-derived autologous tissue, to accurately simulate the pathological state of blocked decidual response, and applied to high-throughput drug screening to search for candidate molecules that can reverse the aging phenotype. In addition, the integration of interdisciplinary technologies is also crucial, and future research should combine epigenomics and metabolomics to analyze the multidimensional regulatory network of EnSCs' aging micro-environment and reveal potential intervention pathways. At the clinical level, randomized controlled trials stratified by stromal functional states are warranted to evaluate the efficacy and safety of interventions targeting EnSC aging, such as stem cell-based therapies, mitochondrial targeted protectants, or sentherapies, can increase live birth rates. With deeper mechanistic insight and maturation of these strategies, precise preconception restoration of endometrial function may become feasible, breaking the vicious cycle of RPL and offering renewed reproductive hope to affected families. Author Contributions Shuang Wu: conceptualization, literature synthesis, visualization, writing – original draft, writing – review and editing. Cenlan Bu: methodology, quality assessment, figure preparation. Qinzheng Xu: writing – original draft, writing – review and editing. Jinling Chen: supervision, resources, funding acquisition, writing – review and editing. Yun-Zhao Xu: funding acquisition, resources, supervision, writing – review and editing. Acknowledgments This review was supported by the National Natural Science Foundation of China (No. 8247062947). Ethics Statement The authors have nothing to report. Conflicts of Interest The authors declare no conflicts of interest. Data Availability Statement The authors have nothing to report.