The endometriosis–adenomyosis spectrum: shared pathophysiology and microenvironment-driven disease divergence

Abstract

Endometriosis and adenomyosis are common gynecologic disorders associated with dysmenorrhea, chronic pelvic pain, and infertility. Although they share several molecular features, the mechanisms by which endometrium-derived tissues develop distinct pathological phenotypes in different tissue environments remain incompletely understood. This review summarizes shared and divergent pathogenic mechanisms, focusing on lesion-specific microenvironments. This narrative review was based on a PubMed literature search from the year of the first publication through December 2025 using terms related to endometriosis, adenomyosis, mitochondrial function, oxidative stress, fibrosis, mechanical stress, and calcium signaling. Both disorders develop in the context of repetitive tissue injury, estrogen-dependent repair responses, chronic inflammation, oxidative stress, and mitochondrial dysfunction. However, differences in lesion location and microenvironment appear to drive distinct pathological phenotypes. In superficial peritoneal endometriosis and ovarian endometrioma, mitochondrial adaptation primarily supports hypoxia tolerance, oxidative stress responses, angiogenesis, cellular survival, and metabolic reprogramming. In contrast, deep infiltrating endometriosis and adenomyosis are characterized by fibrosis, extracellular matrix remodeling, tissue stiffening, and adaptation to mechanical stress. In adenomyosis, mitochondrial regulation of calcium homeostasis, smooth muscle contractility, reactive oxygen species production, and TGF-β–related fibrotic signaling may play important roles in disease progression. We propose a proliferation–fibrosis divergence model in which common pathogenic stimuli are integrated through mitochondria-dependent responses to distinct local microenvironments. Mitochondria may act as central regulators linking hypoxic adaptation, inflammation, metabolism, fibrosis, and mechanotransduction, thereby influencing whether disease progression favors proliferative expansion or fibrotic remodeling. This framework may provide a basis for future mechanism-based precision therapeutic strategies.

Introduction

Endometriosis [1] and adenomyosis [2, 3] are estrogen-dependent disorders that commonly affect women of reproductive age and are strongly associated with clinical symptoms such as dysmenorrhea, chronic pelvic pain, and infertility. The two conditions exhibit clear differences in their anatomical locations. In endometriosis, endometrium-like tissue develops in extrauterine sites such as the peritoneal cavity and ovaries, whereas adenomyosis is characterized by the invasion and proliferation of endometrial tissue within the uterine myometrium [4]. Nevertheless, these two diseases share many similarities in their clinical manifestations and epidemiological features, and they frequently coexist in the same patient [1, 5, 6]. Consequently, recent perspectives have proposed that these disorders should not be regarded as completely independent entities but rather as related diseases that share a common biological basis [6].

Several etiopathogenetic theories have been proposed for both endometriosis and adenomyosis. For endometriosis, the most widely accepted theory is retrograde menstruation, in which viable endometrial cells reflux through the fallopian tubes into the peritoneal cavity and subsequently establish ectopic lesions [7]. Additional mechanisms, including coelomic metaplasia, stem/progenitor cell involvement, and Müllerian remnant theories, have also been proposed to explain lesions occurring at sites not readily accounted for by retrograde menstruation alone [8]. In contrast, adenomyosis has traditionally been explained by direct invasion of the basal endometrium into the underlying myometrium, potentially facilitated by repeated tissue injury and repair at the endometrial–myometrial interface [9]. Although these theories suggest distinct routes of lesion initiation, both disorders ultimately involve ectopic or misplaced endometrium-like tissue that develops within markedly different tissue environments.

Furthermore, the substantial overlap in current therapeutic strategies for endometriosis and adenomyosis reflects the shared biological characteristics of these disorders, including their estrogen dependence [10, 11]. Consequently, hormonal therapies such as oral contraceptives, progestins, and gonadotropin-releasing hormone (GnRH) analogs are widely used for the management of both conditions. These treatments primarily target common upstream drivers of disease activity rather than lesion-specific pathological processes.

Indeed, accumulating evidence indicates that endometriosis and adenomyosis share several molecular and cellular characteristics, including enhanced local estrogen production, chronic inflammation, immune dysregulation, and apoptosis resistance [4, 12–14]. In addition, increased levels of reactive oxygen species (ROS) [15, 16], mitochondrial dysfunction [17], and dysregulation of the phosphatidylinositol 3-kinase (PI3K)–protein kinase B (AKT)–mechanistic target of rapamycin (mTOR) signaling pathway have been identified as common pathogenic factors in both conditions [2, 16, 18–21]. Furthermore, the Tissue Injury and Repair (TIAR) theory, which proposes that repeated cycles of tissue injury and repair promote disease development, has also been suggested as a shared pathogenic model for these disorders [22, 23]. Based on these findings, increasing attention has been directed toward the possibility that endometriosis and adenomyosis may form a continuous disease spectrum.

Although these molecular characteristics exhibit substantial overlap, endometriosis and adenomyosis are known to differ in their patterns of disease progression and histopathological features, reflecting differences in lesion location and modes of tissue involvement [24]. In particular, endometriosis comprises several distinct subtypes, including superficial peritoneal endometriosis (SUP), ovarian endometrioma (OMA), and deep infiltrating endometriosis (DIE), each of which develops and progresses while adapting to its unique microenvironment. These endometriotic lesions arise within ectopic peritoneal or pelvic microenvironments and are characterized by tumor-like biological properties, including enhanced cellular proliferation, tissue invasion, inflammatory activation, and angiogenesis [25]. However, certain lesion subtypes, particularly DIE, also exhibit extensive fibrotic remodeling and extracellular matrix accumulation. In contrast, adenomyosis is histologically characterized by the presence of ectopic endometrial glands and stroma embedded within the myometrium, often forming irregular island-like structures surrounded by smooth muscle hypertrophy, fibrosis, and collagen deposition. Accordingly, structural remodeling of the myometrium represents a defining pathological feature of adenomyosis [26]. Therefore, although these disorders may share common origins and molecular pathways, differences in the tissue microenvironments in which lesions develop are likely to exert a substantial influence on disease progression and phenotypic manifestation. Nevertheless, despite their common endometrial origin, the mechanisms by which distinct tissue microenvironments shape lesion-specific biological behaviors remain incompletely understood.

According to this background, the present review proposes a conceptual framework for integrating and distinguishing endometriosis and adenomyosis from the perspective of microenvironment-dependent pathological differences. Specifically, we focus on lesion-specific microenvironments, mitochondrial adaptation, calcium homeostasis, fibrosis, and mechanobiological signaling, and examine how the distinct tissue microenvironments of the peritoneal cavity and the myometrium influence the pathophysiology of these two disorders. Ultimately, this review aims to integrate existing evidence and provide a conceptual framework for understanding the relationship between endometriosis and adenomyosis. By elucidating both the shared mechanisms and the disease-specific molecular pathways underlying these conditions, we seek to establish a theoretical foundation for a deeper understanding of their pathophysiology.

Materials and methods

A literature search was performed using PubMed, a bibliographic database provided by the National Library of Medicine. The search covered publications from the year of the first publication through December 2025. The search terms included “endometriosis,” “adenomyosis,” “mitochondria,” “mitochondrial dysfunction,” “oxidative stress,” “reactive oxygen species,” “fibrosis,” “mechanical stress,” “calcium signaling,” “myometrial remodeling,” “mTOR,” and “transforming growth factor-beta (TGF-β) signaling,” combined using Boolean operators (AND, OR). These search terms were selected to comprehensively identify studies addressing molecular mechanisms, intracellular signaling pathways, tissue remodeling, and microenvironmental factors relevant to both diseases.

For study selection, we included studies addressing the pathophysiology of endometriosis and adenomyosis. Specifically, basic research, clinical studies, and review articles investigating molecular mechanisms, mitochondrial function, oxidative stress, fibrosis, calcium signaling, mechanical stress, and tissue remodeling in these disorders were considered eligible. Eligible studies were limited to articles published in English and included human studies, animal model studies, and in vitro experimental studies. In contrast, studies focusing exclusively on surgical techniques, imaging diagnostics, or treatment outcomes without discussion of molecular mechanisms or disease pathophysiology were excluded. In addition, case reports with insufficient mechanistic information, conference abstracts, and articles for which the full text was not available were excluded.

The selection process involved an initial screening of titles and abstracts to assess relevance, followed by full-text evaluation when necessary. Furthermore, the reference lists of selected articles were reviewed to identify additional relevant publications related to the topic of this study. The selected literature was analyzed with particular attention to the study objectives, methodologies, major molecular mechanisms, and pathophysiological implications. The findings were then synthesized from the perspectives of shared biological mechanisms and microenvironment-dependent disease divergence.

This study is a literature-based review that relied exclusively on previously published data and did not involve the collection of new human samples or the analysis of patient-level data. Therefore, approval by an institutional ethics committee and the acquisition of informed consent were not required. This narrative review was conducted and reported with consideration of the reporting standards of the Scale for the Assessment of Narrative Review Articles (SANRA) [27].

The author used a large language model (LLM, ChatGPT) solely to assist with English language editing and improvement of readability. The author reviewed and edited all generated text and takes full responsibility for the accuracy, interpretation, and integrity of the manuscript. The LLM was not used for literature searching, data analysis, scientific interpretation, or generation of conclusions.

Results

Although endometriosis and adenomyosis arise in different anatomical locations, they share many similarities in clinical manifestations, hormone dependence, and pathophysiology [1, 2]. Both disorders are commonly associated with symptoms such as dysmenorrhea, chronic pelvic pain, dyspareunia, and infertility [1, 2]. Regarding the mechanisms underlying infertility, endometriosis has been associated with alterations in the peritoneal environment, ovarian dysfunction, chronic inflammation, and impaired decidualization caused by epigenetic changes [28]. In contrast, in adenomyosis, abnormal uterine contractility, reduced endometrial receptivity, and imbalance of the endocrine–immune regulatory axis have been suggested to affect the establishment of pregnancy [29, 30]. Epidemiological studies also demonstrate a considerable coexistence rate between the two diseases, with adenomyosis reported in approximately 15–89% of patients with endometriosis [31]. Consequently, recent perspectives propose that these conditions should not be viewed as entirely independent disorders but rather as estrogen-dependent uterine disorders that share a common biological background [15, 32, 33]. This review summarizes both the shared pathophysiological mechanisms and the distinguishing features of endometriosis and adenomyosis.

The TIAR theory in the pathogenesis of endometriosis and adenomyosis

As one hypothesis proposed to explain the pathophysiology of these disorders, Leyendecker and colleagues introduced the TIAR theory [23, 34] (Fig. 1). Uterine peristalsis refers to the rhythmic contractions of the junctional zone and represents an important physiological function of the uterus, facilitating sperm transport toward the uterine fundus, embryo transport within the uterine cavity, and menstrual blood expulsion. This contractile activity has also been implicated in the process of retrograde menstruation.

Fig. 1.

Shared pathophysiological foundations and microenvironment-dependent divergence within the endometriosis–adenomyosis spectrum. Endometriosis and adenomyosis share common pathogenic foundations, including estrogen dependence, Tissue Injury and Repair (TIAR), chronic inflammation, and recurrent tissue remodeling, all of which contribute to disease progression. Both disorders are further characterized by oxidative stress, metabolic reprogramming, apoptosis resistance, and mitochondrial stress. Increased ROS production, mitochondrial dysfunction, and activation of cell survival signaling pathways are key factors supporting lesion maintenance and growth. However, the adaptive mechanisms that drive disease progression differ according to the specific microenvironment in which lesions reside. In Superficial Peritoneal Endometriosis (SUP), hypoxia is the predominant microenvironmental stressor, leading to activation of HIF-1α, VEGF-dependent angiogenesis, and the PI3K–AKT–mTOR and NF-κB signaling pathways. In Ovarian Endometrioma (OMA), repeated hemorrhage results in iron accumulation and oxidative stress, which promote NF-κB activation, mitochondrial dysfunction, and adaptation to oxidative stress. In DIE, chronic tissue injury, fibrosis, and tissue stiffening drive activation of TGF-β/Smad signaling, Epithelial–Mesenchymal Transition (EMT), Fibroblast-to-Myofibroblast Transdifferentiation (FMT), Smooth Muscle Metaplasia (SMM), and Extracellular Matrix (ECM) remodeling. In adenomyosis, persistent mechanical stress within the myometrium activates both the TGF-β/Smad pathway and Hippo–YAP/TAZ-mediated mechanotransduction signaling, resulting in progressive fibrosis and tissue stiffening. Taken together, endometriosis and adenomyosis can be viewed as a disease spectrum that shares common molecular foundations but diverges through microenvironment-specific adaptive responses. Specifically, hypoxia adaptation predominates in SUP, oxidative stress adaptation in OMA, and mechanical stress- and fibrosis-associated adaptation in DIE and adenomyosis, ultimately giving rise to their distinct pathological phenotypes

Although originally proposed as a hypothesis, subsequent studies have accumulated indirect evidence supporting several key components of the TIAR concept [22, 23]. Magnetic resonance imaging (MRI) studies have demonstrated abnormal uterine peristalsis and structural alterations of the junctional zone in patients with endometriosis and adenomyosis [9]. In addition, molecular biological investigations have reported enhanced local estrogen production, activation of inflammatory responses, and increased tissue remodeling at the endometrial–myometrial interface [22, 23, 35].

Based on these observations, the TIAR theory explains the pathogenesis of endometriosis and adenomyosis as follows. Chronic uterine peristalsis or hyperperistalsis induces repetitive microtrauma at the endometrial–myometrial interface, particularly near the uterine cornual raphe [23, 34, 35]. This injury activates the TIAR mechanism, resulting in increased local estrogen production [9, 22, 23, 35]. Persistent hyperperistalsis further amplifies tissue injury and estrogen synthesis, while paracrine estrogen signaling disrupts the physiological ovarian regulation of uterine contractility, thereby establishing a self-perpetuating cycle of uterine hyperperistalsis [22, 23, 35]. Consequently, fragments of the basal endometrium may be displaced into the peritoneal cavity, giving rise to endometriotic lesions, whereas direct invasion of the basal endometrium into the myometrium may lead to the development of adenomyosis.

Nevertheless, although the TIAR theory provides an attractive and comprehensive framework for explaining disease pathogenesis, its validity remains a subject of ongoing debate. Current evidence supports the possibility that repetitive mechanical injury and subsequent repair processes contribute to the development of both endometriosis and adenomyosis; however, the precise causal relationships have yet to be fully elucidated and require further investigation [24].

Site-specific microenvironmental adaptation mechanisms and signaling networks

Endometriosis and adenomyosis share numerous molecular mechanisms as estrogen-dependent and chronic inflammatory disorders [36]. However, differences in lesion location and tissue context result in distinct relative contributions of microenvironmental stressors and adaptive mechanisms that drive disease pathogenesis.

Several pathological processes are commonly involved in both disorders, including chronic inflammation, oxidative stress, angiogenesis, epigenetic alterations, metabolic adaptation, and resistance to apoptosis [15, 16, 37] (Fig. 1). During disease initiation and progression, ROS production is increased through the activation of inflammatory cells, hypoxic conditions, mitochondrial dysfunction, and the elevated metabolic demands associated with tissue repair and remodeling [38, 39]. Increased ROS not only induces oxidative stress but also promotes metabolic reprogramming through alterations in mitochondrial function and membrane potential. In concert with estrogen signaling, these changes establish a microenvironment that supports persistent tissue remodeling and cellular proliferation [38, 39]. As a consequence, cells undergo Warburg-like metabolic reprogramming characterized by an increased reliance on glycolysis, thereby enhancing their capacity to survive under stressful conditions [40]. Furthermore, apoptosis resistance is reinforced through increased expression of B-cell lymphoma 2 (Bcl-2), reduced expression of Bcl-2-associated X protein (Bax), and suppression of caspase activity, enabling the survival and persistence of ectopic endometrial cells that would otherwise be eliminated [36, 41, 42].

Collectively, these findings support the concept that endometriosis and adenomyosis represent a disease spectrum sharing common pathogenic foundations, including estrogen dependence, chronic inflammation, metabolic reprogramming, evasion of apoptosis, and mitochondrial stress [6, 15, 17, 20, 21, 26, 33, 37, 38, 41]. Nevertheless, the dominant stress-response pathways and signaling networks differ according to the specific microenvironment in which lesions reside, thereby shaping the distinct biological characteristics of each disease.

Despite sharing these common pathogenic foundations, endometriosis comprises several distinct subtypes, including SUP, OMA, and DIE, each of which maintains lesion survival and progression through adaptation to its specific microenvironment.

First, SUP develops within the peritoneal cavity, where lesions are exposed to an immune cell-rich environment and, particularly during the early stages of lesion establishment, experience relative hypoxia due to limited vascular supply [43]. This hypoxic microenvironment activates Hypoxia-Inducible Factor-1 alpha (HIF-1α) [43, 44], which subsequently induces vascular endothelial growth factor (VEGF)-dependent angiogenesis [45]. In addition, activation of the PI3K–AKT–mTOR pathway [46] and the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway [47] promotes metabolic adaptation and cell survival signaling, thereby facilitating lesion implantation and growth. Accordingly, hypoxia-responsive and angiogenic signaling pathways play central roles in the pathophysiology of SUP [43, 45].

In contrast, OMA is characterized by a unique microenvironment shaped by repeated hemorrhage-associated iron accumulation and oxidative stress [48]. Because lesion cells are chronically exposed to excessive iron-derived ROS, the NF-κB pathway remains persistently activated [49], coordinately regulating chronic inflammation, apoptosis evasion, angiogenesis, invasive capacity, and metabolic reprogramming. Moreover, selective expansion of clones harboring somatic mutations in Kirsten Rat Sarcoma Viral Oncogene Homolog (KRAS) and Phosphatidylinositol-4,5-Bisphosphate 3-Kinase Catalytic Subunit Alpha (PIK3CA) has been reported, suggesting that OMA represents a dynamic tissue undergoing clonal evolution [50]. Dysfunction of Sirtuin 3 (SIRT3), a key regulator of mitochondrial redox homeostasis, has also been described, implicating mitochondrial dysfunction and impaired oxidative stress responses in disease pathogenesis [51]. Therefore, adaptation to oxidative stress, rather than hypoxia itself, appears to play a predominant role in the pathobiology of OMA [48].

In contrast, DIE is characterized by extensive fibrosis and tissue stiffening, making adaptation to mechanical stress a central driver of disease progression [52]. Fibrosis in endometriosis, particularly in DIE and OMA, is thought to develop as an aberrant wound-healing response associated with chronic inflammation and repeated cycles of tissue injury and repair [53]. In this context, the TGF-β1/Smad pathway promotes myofibroblast differentiation and extracellular matrix production. In addition, Notch, Wnt/β-catenin, and Sphingosine-1-phosphate (S1P) signaling pathways contribute to fibrotic progression [53]. A hallmark of DIE is the induction of epithelial–mesenchymal transition (EMT), fibroblast-to-myofibroblast transdifferentiation (FMT), and smooth muscle metaplasia (SMM) by TGF-β1 and inflammatory cytokines, resulting in extensive tissue remodeling and fibrosis [53]. However, epithelial cells within DIE lesions have been reported to retain their epithelial phenotype despite residing in a stiff tissue environment, suggesting that TGF-β1/Smad signaling is not constitutively activated and may instead be subject to precise spatiotemporal regulation [54]. Although localized and transient hypoxic signaling may also contribute to DIE, its dependence on sustained hypoxia is considerably less pronounced than that observed in SUP. Rather, the defining features of DIE appear to be tissue injury and repair, smooth muscle hyperactivity, SMM, tissue stiffening, and fibrosis [20, 23, 32, 33, 48, 53].

The myometrium is a dynamic smooth muscle tissue that undergoes repeated cycles of contraction and remodeling throughout the reproductive lifespan. The uterus participates in uterine peristalsis, which facilitates sperm transport toward the uterine fundus, embryo transport within the uterine cavity, and menstrual blood expulsion [55]. Owing to abnormalities in this contractile activity, adenomyosis can be regarded as a mechanobiology-related disorder in which mechanical stress, myometrial contraction, and tissue stiffening contribute to disease progression [24, 56]. Progressive myometrial remodeling, including smooth muscle hypertrophy, fibrosis, inflammation, and nerve fiber proliferation, has been documented in adenomyosis and is associated with abnormal uterine peristalsis and hypercontractility [33, 57]. Furthermore, fluid shear stress generated by uterine contractions has been shown to increase cytoskeletal components such as F-actin in human endometrial epithelial cells and myometrial smooth muscle cells, suggesting that mechanical stimuli may directly influence cellular architecture and tissue remodeling [58].

Thus, similar to DIE, adenomyosis is strongly influenced by fibrosis and mechanical stress. Cells are known to sense the stiffness of their surrounding tissue through adhesion complexes and the actin–myosin contractile apparatus and subsequently regulate cellular function and cell fate in response to mechanical cues [59]. Such mechanosensing mechanisms are critical for understanding pathological conditions characterized by abnormal tissue stiffening and fibrosis. In particular, myometrial stiffness in adenomyosis has been reported to be maintained through interactions among tissue stiffness, viscoelasticity, and water content, suggesting that the mechanical properties of the tissue play an important role in disease pathogenesis [60]. Within this mechanical environment, lesions reside in the highly contractile myometrium and are therefore continuously exposed to uterine hyperperistalsis as well as repetitive cycles of tissue injury and repair.

As a consequence, in addition to the TGF-β/Smad pathway, mechanotransduction pathways centered on Hippo-Yes-associated protein (YAP)/Transcriptional co-activator with PDZ-binding motif (TAZ) are activated, promoting tissue remodeling and fibrosis [2, 61]. Moreover, fibrosis-induced tissue stiffening is thought to further amplify the abnormal mechanical environment, thereby establishing a positive feedback mechanism that supports lesion progression.

Although fibrosis is often emphasized as a hallmark of adenomyosis, substantial evidence indicates that it is also a central pathological feature of endometriosis [62]. Recurrent tissue injury, cyclic bleeding, chronic inflammation, and activation of EMT and FMT pathways promote extracellular matrix accumulation within endometriotic lesions. In advanced disease, fibrosis directly contributes to adhesion formation, distortion of pelvic anatomy, chronic pelvic pain, infertility, and organ dysfunction. Therefore, fibrosis should be regarded as a major driver of disease progression in both endometriosis and adenomyosis; however, the underlying mechanical environments and the extent of tissue remodeling differ between the two disorders.

Taken together, fibrosis represents a shared pathological feature of both endometriosis and adenomyosis. However, its relative importance in disease pathogenesis appears to vary according to lesion type. In SUP, adaptation to hypoxia predominates as a major pathogenic mechanism, whereas adaptation to oxidative stress plays a more prominent role in OMA. In contrast, adaptation to mechanical stress and fibrotic responses become more pronounced in DIE and adenomyosis, serving as key mechanisms that support lesion maintenance and progression. Thus, although these disorders share common molecular foundations, including estrogen dependence, chronic inflammation, mitochondrial stress, and evasion of apoptosis, they selectively activate distinct signaling networks in response to their respective microenvironments, thereby giving rise to characteristic pathological phenotypes.

Lesion site-specific mitochondrial function and metabolic adaptation

The various subtypes of endometriosis and adenomyosis share the ability to maintain lesion survival through mitochondrial functional alterations and metabolic reprogramming in the context of chronic inflammation. However, the distinct microenvironments associated with different lesion types are thought to induce unique mitochondrial adaptations, thereby leading to differences in the direction of metabolic reprogramming and the resulting pathological phenotypes [63, 64] (Fig. 2).

Fig. 2.

Microenvironment-dependent mitochondrial adaptation and metabolic reprogramming: a model of pathological divergence in endometriosis and adenomyosis. The various subtypes of endometriosis and adenomyosis share the common feature of maintaining lesion survival through mitochondrial functional alterations and metabolic reprogramming in the setting of chronic inflammation. However, differences in the microenvironment surrounding individual lesions induce distinct mitochondrial adaptations, thereby contributing to the diversity of pathological phenotypes. In Superficial Peritoneal Endometriosis (SUP), adaptation to hypoxia leads to activation of HIF-1α, which induces the expression of glycolysis-associated molecules, including GLUT1, HK2, LDHA, and PDK1. As a result, a hypoxia-adapted Warburg-like metabolic phenotype is established, characterized by increased reliance on glycolysis rather than oxidative phosphorylation. In Ovarian Endometrioma (OMA), adaptation to iron accumulation and a high-ROS environment caused by repeated hemorrhage places particular importance on mitochondrial ROS regulation and antioxidant responses. Antioxidant mechanisms centered on SOD2 maintain mitochondrial function and energy metabolism, thereby supporting cellular proliferation and migratory capacity. In Deep Infiltrating Endometriosis (DIE), chronic inflammation, tissue stiffening, and mechanical stress promote mitochondrial homeostasis and energy metabolic regulation through SIRT3, thereby supporting lesion cell survival. These adaptations contribute to the maintenance of fibrosis, proliferation of smooth muscle cells, and adaptation to mechanical stress, resulting in a metabolic phenotype characterized by fibrosis- and mechanobiology-driven metabolic reprogramming. In contrast, adenomyosis is characterized by selective mitophagy mediated through the PINK1–Parkin pathway, AMPK activation associated with reduced GRIM-19 expression, and enhanced mitochondrial fission. Furthermore, mitochondria-derived ROS amplify TGF-β/Smad signaling, thereby promoting EMT, tissue remodeling, and fibrosis. Because adenomyotic lesions reside within the highly contractile myometrium, a mechanobiological environment is established in which mechanical stress, tissue stiffening, and mitochondrial dysfunction interact with one another. Collectively, these observations suggest that distinct metabolic phenotypes predominate according to lesion type: hypoxia-adapted Warburg-like metabolism in SUP, antioxidant and redox homeostasis-maintaining metabolism in OMA, fibrosis- and mechanotransduction-adapted metabolism in DIE, and mitophagy–fibrosis-associated metabolism in adenomyosis. These microenvironment-specific metabolic adaptations may ultimately contribute to the development of distinct pathological phenotypes

In SUP lesions, hypoxic adaptation is a central driver of metabolic alteration because early lesions become established within the relatively avascular peritoneal cavity. Activation of HIF-1α induces the expression of glycolysis-related molecules, including glucose transporter 1 (GLUT1), hexokinase 2 (HK2), lactate dehydrogenase A (LDHA), and pyruvate dehydrogenase kinase 1 (PDK1), thereby shifting cellular metabolism toward glycolysis and away from oxidative phosphorylation [40]. Although mitochondrial function is partially suppressed, mitochondria continue to support lesion survival through ROS production and HIF-1α stabilization. Thus, hypoxia-adapted Warburg-like metabolism represents the most characteristic metabolic feature of SUP lesions [40]. Studies directly comparing the role of metabolic reprogramming, including the Warburg effect, among the various endometriosis subtypes (SUP, OMA, and DIE) and adenomyosis remain extremely limited. Therefore, how differences in the microenvironment influence the direction of metabolic adaptation and the development of pathological phenotypes remains an important subject for future investigation.

In OMA lesions, iron accumulation and a high-ROS environment resulting from repeated hemorrhage constitute the defining microenvironmental features. Ectopic endometrial stromal cells derived from OMA exhibit increased mitochondrial ROS production and enhanced energy generation [15]. Consequently, mitochondrial ROS regulation, antioxidant responses, and mitochondrial quality control become particularly important. In these cells, expression of the mitochondrial antioxidant enzyme superoxide dismutase 2 (SOD2) is increased, and SOD2-mediated regulation of oxidative stress helps maintain mitochondrial function and energy metabolism, thereby promoting cellular proliferation and migratory capacity [15].

In DIE lesions, progressive fibrosis and tissue stiffening, together with repetitive cycles of tissue injury and repair driven by hyperperistalsis and smooth muscle hyperactivity, represent major sources of stress. Under these conditions of chronic inflammation and mechanical stress, increased SIRT3 expression has been reported in peripheral blood mononuclear cells from patients with DIE [65]. Because SIRT3 regulates mitochondrial oxidative stress while maintaining redox homeostasis and energy metabolism, its upregulation may contribute to the adaptive survival of DIE cells. In general, SIRT3 deacetylates multiple metabolic enzymes within mitochondria, thereby maintaining electron transport chain activity and ATP production [66]. In a canine kidney epithelial cell line, SIRT3 activates SOD2, preserves mitochondrial redox homeostasis, and reduces ROS levels [67]. Accordingly, increased SIRT3 expression in DIE may support lesion cell survival through preservation of mitochondrial function and may contribute to the maintenance of fibrosis, proliferation of smooth muscle cells, and adaptation to mechanical stress. These findings suggest that the metabolic phenotype of DIE differs fundamentally from the hypoxia-adapted phenotype of SUP and may instead be characterized as a mechanobiology-driven metabolic reprogramming associated with fibrosis and adaptation to mechanical stimuli.

In adenomyosis, activation of the PI3K–AKT–mTOR pathway, which suppresses autophagy, has been reported, whereas mitochondrial damage and oxidative stress induce selective mitophagy through the phosphatase and tensin homolog (PTEN)-induced putative kinase 1 (PINK1)–Parkin pathway [68]. Furthermore, activation of AMP-activated protein kinase (AMPK) associated with reduced expression of gene associated with retinoid-IFN-induced mortality 19 (GRIM-19) also promotes mitophagy [69]. These observations suggest that suppression of autophagy through mTOR activation and enhancement of selective mitophagy may coexist in adenomyosis. Such alterations are likely to contribute to myometrial remodeling and fibrosis.

Fibrosis in adenomyosis is primarily regulated by the TGF-β–Smad pathway [17, 70–72], while ROS enhance TGF-β signaling, thereby promoting fibroblast proliferation and extracellular matrix accumulation [73]. Similar mechanisms linking mitochondrial dysfunction to fibrosis have been reported in cardiac fibrosis [74] and pulmonary fibrosis [75], suggesting that comparable processes may also operate in adenomyosis. Indeed, studies using human adenomyotic tissues and tamoxifen-induced mouse models have demonstrated that mitochondrial dysfunction induces oxidative stress and mitophagy, which in turn promote EMT and fibrosis through activation of TGF-β1 signaling [17]. These findings suggest that mitochondrial dysfunction may function as an upstream regulator of tissue remodeling and fibrosis.

Moreover, because adenomyotic lesions reside within the highly contractile myometrium, they are exposed to cyclic mechanical stress in a manner similar to DIE [9]. As lesions progress, smooth muscle hyperplasia, EMT-associated remodeling, and fibrosis become increasingly prominent [2, 76], leading to increased tissue stiffness and the establishment of a mechanobiological environment characterized by the coexistence of persistent mechanical stimulation and tissue stiffening [56]. In addition, mechanical stress and extracellular matrix stiffening induce the expression of genes involved in cellular proliferation and angiogenesis, thereby promoting tissue repair and remodeling [58, 77]. Therefore, mitochondrial dysfunction, fibrosis, and mechanical stress are likely to interact synergistically to drive the pathogenesis of adenomyosis. Collectively, these observations suggest that mitochondria play a critical role not only in metabolic adaptation but also in the development and maintenance of fibrosis.

Mitochondria-mediated regulation of calcium homeostasis and fibrosis: a comparison between endometriosis and adenomyosis

Mitochondria are central cellular organelles that regulate not only ATP production but also ROS generation, cell death, metabolic reprogramming, and calcium homeostasis [15, 17, 19, 20, 75, 78–81]. Consequently, alterations in mitochondrial function influence not only the survival and proliferation of lesion cells but also fibrosis, tissue remodeling, and smooth muscle contractility. In this subsection, we compare the roles of mitochondria in endometriosis and adenomyosis, with particular emphasis on calcium homeostasis and its contribution to fibrosis and contractile function (Fig. 3).

Fig. 3.

Distinct roles of mitochondrial function in endometriosis and adenomyosis: from metabolic adaptation to the calcium–fibrosis axis. Mitochondria are central cellular organelles responsible not only for ATP production but also for ROS generation, regulation of cell death, metabolic reprogramming, and maintenance of calcium homeostasis. This figure schematically illustrates the distinct roles of mitochondria in endometriosis and adenomyosis. In endometriosis, mitochondria primarily support lesion cell survival, proliferation, and metabolic adaptation. Specifically, they facilitate adaptation to hypoxia in SUP, oxidative stress in OMA, and mechanical stress in DIE. In contrast, a key aspect of adenomyosis pathophysiology is the regulation of calcium homeostasis by mitochondria. Intracellular Ca²⁺ homeostasis is maintained through coordinated interactions between the Endoplasmic Reticulum (ER) and mitochondria. However, chronic Ca²⁺ influx can lead to mitochondrial Ca²⁺ overload, resulting in increased ROS production and activation of TGF-β–Smad signaling, thereby promoting fibrosis and tissue remodeling. Furthermore, enhanced oxytocin receptor signaling and prostaglandin-related signaling induce abnormalities in Ca²⁺ dynamics through the PLC–IP3 pathway, further increasing mitochondrial Ca²⁺ influx and ROS generation. Consequently, disruption of ER–mitochondrial communication may contribute to disease pathogenesis in adenomyosis through a pathological cascade involving calcium dysregulation, mitochondrial dysfunction, and fibrosis

In adenomyosis, smooth muscle cells, in addition to endometrial glandular and stromal cells, play a critical role in disease pathogenesis, contributing to fibrosis, tissue remodeling, and myometrial hypertrophy [76]. Smooth muscle cells within adenomyotic lesions exhibit ultrastructural features distinct from those of normal myometrium, and their hypertrophy and structural alterations are associated with impaired contractile function [82]. Furthermore, alterations in mitochondrial density and function [17, 68, 69, 83, 84], increased ATP production [83–86], and enhanced ROS generation [83] have been reported, suggesting the presence of metabolic adaptations that support smooth muscle hypertrophy and fibrosis [2, 26, 76]. These characteristics contrast with endometriosis, in which glycolysis-dependent metabolic reprogramming is considered a predominant feature [38, 40, 63, 86].

A key factor in understanding this pathology is the regulation of calcium homeostasis by mitochondria [87]. Intracellular calcium homeostasis is maintained through coordinated interactions among the endoplasmic reticulum (ER), plasma membrane calcium channels, various transporters, and mitochondria [88]. In particular, mitochondria and the ER are functionally interconnected, and this ER–mitochondrial coupling plays a central role not only in calcium signaling but also in the maintenance of cellular homeostasis. The ER serves as a critical intracellular organelle responsible for protein folding, lipid synthesis, and intracellular calcium storage, and ER stress can be induced by various cellular stressors, including hypoxia, oxidative stress, inflammation, and metabolic disturbances [89]. Although the ER stress response functions as an adaptive mechanism to maintain homeostasis, it is also closely linked to mitochondrial function, autophagy, apoptosis, and calcium signaling. Because the ER and mitochondria are structurally and functionally interconnected, their interaction serves as a critical regulatory mechanism governing the balance between cell survival and cell death. Consequently, disruption of ER–mitochondrial communication may lead to impaired calcium homeostasis, mitochondrial dysfunction, and dysregulated cell fate determination, thereby contributing to disease pathogenesis.

Furthermore, myometrial contraction is initiated by increases in intracellular Ca²⁺ concentrations, which activate myosin light chain kinase (MLCK) and subsequently induce actin–myosin interactions [90]. In addition to supplying ATP, mitochondria function as calcium buffers through Ca²⁺ uptake mediated by the mitochondrial calcium uniporter (MCU) [78, 91]. However, chronic Ca²⁺ influx may result in mitochondrial Ca²⁺ overload, leading to electron transport chain dysfunction and increased ROS production [80, 81]. Elevated ROS can further activate TGF-β signaling and amplify fibrotic responses through the ROS–TGF-β–Smad pathway, thereby promoting myometrial stiffening and tissue remodeling [17, 70–73]. It should be noted, however, that some of the evidence supporting this mechanism is derived from studies of human non-pregnant and pregnant uterine smooth muscle cells [80, 87], and direct evidence in adenomyosis remains limited.

In addition, increased expression of oxytocin receptors and enhanced prostaglandin- and leukotriene-related signaling have been reported in adenomyosis [92, 93]. These signaling pathways are known to augment intracellular Ca²⁺ dynamics through the phospholipase C (PLC)–inositol 1,4,5-trisphosphate (IP3) pathway [94]. Although mitochondria serve as important intracellular calcium buffers, excessive Ca²⁺ influx can induce mitochondrial Ca²⁺ overload, resulting in disruption of calcium homeostasis and increased ROS generation [79]. Therefore, enhanced Ca²⁺ signaling and mitochondrial dysfunction may interact in adenomyosis to promote ROS-mediated, TGF-β-dependent fibrosis.

Accordingly, calcium dysregulation in adenomyosis should be viewed not as an isolated mitochondrial abnormality but rather as a consequence of dysfunction within the broader intracellular calcium regulatory network, including ER–mitochondrial interactions. In this context, a pathogenic cascade consisting of mechanical stress, calcium dysregulation, mitochondrial dysfunction, and fibrosis may represent a plausible disease model for adenomyosis.

In contrast, in endometriosis, mitochondria primarily regulate cellular proliferation, survival, hypoxic responses, and metabolic reprogramming, thereby supporting lesion establishment and progression [63, 64]. In particular, mitochondria play central roles in adaptation to hypoxic and oxidative stress conditions in SUP and OMA lesions [15, 40], supporting lesion cell survival and metabolic adaptation. In DIE, mitochondria-derived ROS are thought to amplify TGF-β-dependent fibrotic signaling and mechanotransduction pathways, thereby contributing to extracellular matrix accumulation and tissue stiffening [53].

Thus, mitochondrial function in endometriosis primarily supports lesion growth, survival, and adaptation to the local microenvironment, and functional interactions with the ER appear to be important in these processes. Dysregulation of ER stress responses has also been implicated in the pathogenesis of endometriosis. Indeed, aberrant expression of ER stress-related molecules has been reported in endometriosis [89], and ER stress is thought to interact synergistically with inflammation, oxidative stress, and hypoxia to promote lesion formation and progression. Moreover, because the ER plays a central role in maintaining calcium homeostasis, ER stress is closely linked to mitochondrial function, autophagy, apoptosis, and metabolic regulation.

According to this background, ER–mitochondrial interactions represent an important molecular framework integrating mitochondrial function, metabolism, calcium homeostasis, and cell fate regulation in endometriosis. However, direct evidence demonstrating mitochondrial involvement in calcium homeostasis or contractile abnormalities across the various endometriosis subtypes remains limited. Therefore, the effects of mitochondria in endometriosis are currently understood primarily as secondary consequences mediated through ROS production, metabolic dysregulation, and activation of inflammatory signaling pathways. Nevertheless, given that ER–mitochondrial communication is a major regulator of intracellular calcium dynamics, elucidating how disturbances in calcium homeostasis contribute to the pathogenesis of endometriosis remains an important challenge for future research.

Taken together, the primary role of mitochondria in endometriosis appears to be the maintenance of metabolic adaptation and cellular survival, whereas in adenomyosis, mitochondria are involved not only in energy metabolism but also in the regulation of calcium homeostasis, smooth muscle contraction, fibrosis, and tissue remodeling. Thus, mitochondria in endometriosis function predominantly as metabolic organelles that support lesion proliferation and survival, whereas in adenomyosis they appear to play a more prominent role as integrative regulators of calcium signaling and fibrotic remodeling.

A proliferation–fibrosis divergence model in endometriosis and adenomyosis: integration of somatic mutations, microenvironmental stress, and mitochondrial reprogramming

Endometriosis and adenomyosis are both estrogen-dependent and chronic inflammatory disorders that share cellular proliferation and fibrosis as common pathological foundations; however, their relative contributions and resulting tissue phenotypes differ substantially (Fig. 4). Similar divergence between proliferative and fibrotic outcomes has been observed in other organs. In the liver, chronic inflammation may lead to either hepatocellular carcinoma or cirrhosis; in the lung, to lung cancer or pulmonary fibrosis; and in the pancreas, to pancreatic cancer or pancreatic fibrosis [95–97]. In these diseases, tissue injury, chronic inflammation, and increased ROS represent common initiating events, whereas the signaling pathways subsequently activated and the local tissue microenvironment determine whether tumor-like proliferation or fibrotic remodeling becomes the dominant outcome [95–97]. In this subsection, we propose that this concept can be applied to endometriosis and adenomyosis and conceptualized as a proliferation–fibrosis divergence model.

Fig. 4.

A proliferation–fibrosis divergence model in endometriosis and adenomyosis: an integrative concept linking somatic mutations, the microenvironment, and mitochondrial metabolic adaptation. Endometriosis and adenomyosis share common pathogenic foundations, including estrogen dependence, chronic inflammation, tissue injury, and increased Reactive Oxygen Species (ROS) production. Somatic mutations and clonal expansion involving genes such as ARID1A, KRAS, PIK3CA, PTEN, and CTNNB1 have been identified in both disorders; however, these genetic alterations alone are insufficient to fully explain their distinct pathological phenotypes. This figure illustrates a conceptual framework in which a common tissue stress response diverges into either a proliferation-dominant or fibrosis-dominant phenotype through the influence of the local microenvironment and mitochondria-dependent metabolic adaptation. As central upstream signaling hubs, the PI3K–AKT–mTOR and RAS–MAPK pathways interact with HIF-1α, NF-κB, AMPK, Wnt/β-catenin, TGF-β, and Hippo–YAP/TAZ signaling networks to regulate cellular proliferation, metabolic reprogramming, autophagy, mitophagy, ROS production, and fibrogenesis. Under hypoxic and inflammatory conditions, signaling pathways centered on HIF-1α, NF-κB, and mTOR are activated, promoting angiogenesis, enhanced glycolysis, apoptosis resistance, and the survival of lesion cells. In contrast, in environments characterized by tissue stiffening and mechanical stress, activation of the TGF-β–Smad pathway and YAP/TAZ promotes Epithelial–mesenchymal Transition (EMT), Fibroblast-to-Myofibroblast Transdifferentiation (FMT), Extracellular Matrix (ECM) accumulation, and tissue remodeling. Mitochondria function as central mediators of this divergence by regulating ROS generation, metabolic reprogramming, autophagy/mitophagy, and energy metabolism. In hypoxic and inflammatory microenvironments, proliferation-oriented metabolic adaptation is favored, contributing to the recurrent, invasive, and tumor-like characteristics of endometriosis. Conversely, in high-stiffness and mechanically stressed microenvironments, ROS–TGF-β–Smad and YAP/TAZ signaling become predominant, driving fibrosis, tissue stiffening, and uterine enlargement, which are characteristic features of an adenomyosis-like phenotype. Collectively, this model suggests that although endometriosis and adenomyosis share a common molecular foundation, differences in the local microenvironment and mitochondrial metabolic adaptation direct disease progression toward either a proliferation-dominant or fibrosis-dominant phenotype, thereby supporting the concept of an endometriosis–adenomyosis disease spectrum

Recent studies have highlighted the roles of somatic mutations and signaling networks as key determinants of this divergence. In endometriosis, somatic mutations in AT-Rich Interaction Domain 1 A (ARID1A), KRAS, PIK3CA, PTEN, and Catenin Beta 1 (CTNNB1) are detected at relatively high frequencies [4, 50, 98, 99] and are implicated in tumor-like characteristics, including cellular proliferation, invasion, angiogenesis, apoptosis resistance, immune evasion, and fibrosis. In contrast, although KRAS and PIK3CA mutations have been reported in adenomyosis [4, 100], mutations in tumor suppressor genes such as ARID1A and PTEN appear to be uncommon. These observations suggest that clonal expansion of mutation-bearing endometrial epithelial cells represents a shared biological foundation of both disorders [4]. However, genetic alterations alone are insufficient to explain the differences in disease phenotype, indicating that the site of lesion formation and the surrounding tissue microenvironment act as important modifying factors.

Indeed, these driver mutations are interconnected with multiple signaling networks centered on the PI3K–AKT–mTOR and RAS–mitogen-activated protein kinase (MAPK) pathways, including Wnt/β-catenin, HIF-1α, NF-κB, AMPK, Hippo–YAP/TAZ, and TGF-β signaling. The PI3K–AKT–mTOR and RAS–MAPK pathways promote HIF-1α expression and stabilization, thereby inducing VEGF and glycolysis-related genes and regulating hypoxic adaptation, angiogenesis, and metabolic reprogramming [101]. In addition, PI3K–AKT activation suppresses Glycogen Synthase Kinase 3 Beta (GSK3β), leading to activation of Wnt/β-catenin signaling through β-catenin accumulation. This process promotes cellular proliferation, EMT, invasion, and fibrosis while simultaneously inducing c-Myc-dependent glycolytic activation [102]. Furthermore, KRAS and PIK3CA mutations, together with mitochondria-derived ROS, activate the NF-κB pathway, thereby sustaining chronic inflammation, apoptosis resistance, and lesion establishment [103]. In contrast, AMPK functions as an energy sensor that antagonizes mTOR signaling and promotes autophagy, mitophagy, and fatty acid oxidation, thereby facilitating adaptation to metabolic stress [104]. These mechanisms are likely to be particularly important in SUP and OMA.

By contrast, fibrosis-dominant disease progression is strongly influenced by YAP/TAZ and TGF-β signaling, which are activated in response to tissue stiffening and mechanical stress. YAP/TAZ activity is further enhanced by PI3K–AKT and RAS–MAPK signaling, thereby promoting both cellular proliferation and fibrosis [105]. Moreover, ROS and TGF-β mutually reinforce one another, inducing EMT, FMT, and ECM production [73]. Because KRAS and PIK3CA mutations also enhance responsiveness to TGF-β signaling, the ROS–TGF-β–Smad network is likely to represent a major driver of fibrosis and tissue remodeling, particularly in DIE and adenomyosis.

Notably, the PI3K–AKT–mTOR and RAS–MAPK pathways function as critical hubs within these signaling networks and contribute to the regulation of mitochondrial metabolism, ROS production, autophagy, mitophagy, metabolic reprogramming, and fibrosis [19, 46]. Therefore, the pathological significance of somatic mutations should not be viewed merely as genetic abnormalities but rather as determinants of mitochondria-dependent metabolic adaptation to local environmental stresses, including inflammation, hypoxia, mechanical stress, and tissue stiffening. However, studies directly examining the relationship between somatic mutations and activation of the PI3K–AKT and MAPK pathways in endometriosis and adenomyosis remain limited. Consequently, further experimental investigations are required to clarify the precise roles of these pathways in disease pathogenesis.

This proliferation–fibrosis divergence model may also explain several important clinical characteristics of the two disorders. The high recurrence rate and invasive behavior of endometriosis are consistent with a proliferation-dominant phenotype [43, 45], whereas uterine enlargement and tissue stiffening in adenomyosis can be interpreted as manifestations of a fibrosis-dominant phenotype [2, 59–61]. Furthermore, the frequent coexistence of these disorders may be explained as the consequence of divergent pathological trajectories arising from a shared etiological foundation.

Taken together, endometriosis and adenomyosis can be viewed as related disorders that originate from common pathogenic mechanisms but diverge toward either proliferation-dominant or fibrosis-dominant phenotypes as a result of differences in local microenvironmental conditions and mitochondrial metabolic reprogramming. This conceptual framework may provide a theoretical basis for the development of metabolism- and immune-targeted therapies for endometriosis, as well as antifibrotic and mitochondria-protective therapeutic strategies for adenomyosis.

Discussion

The concept of an endometriosis–adenomyosis spectrum has gained increasing attention in recent years. Rather than representing completely distinct disease entities, these disorders are increasingly viewed as a continuum because of their frequent clinical coexistence and shared molecular characteristics [6, 15, 17, 20, 21, 26, 33, 37, 38, 41]. In this review, we propose an integrative model in which endometriosis and adenomyosis arise from a common endometrium-derived stress response but diverge as a consequence of adaptation to distinct tissue microenvironments.

Although both disorders share common molecular foundations, the tissue environments in which lesions develop are fundamentally different. In SUP and OMA, which arise within the peritoneal cavity, lesion cells must adapt to hypoxia, iron overload, oxidative stress, and immune surveillance. Consequently, metabolic adaptation programs centered on HIF-1α, NF-κB, mTOR, and AMPK are preferentially activated, promoting survival and proliferative programs characterized by enhanced glycolysis and angiogenesis. In contrast, in DIE and adenomyosis, which develop within or adjacent to the myometrium, persistent contractile activity, mechanical loading, tissue stiffening, and fibrosis represent the major selective pressures. Under these conditions, TGF-β, YAP/TAZ, and mechanotransduction pathways are likely to play more prominent roles, favoring tissue remodeling and fibrotic programs.

Within this framework, mitochondria may serve as critical integrators of disease divergence. In endometriosis, mitochondria are primarily involved in adaptation to hypoxic conditions, ROS production, metabolic reprogramming, and maintenance of lesion cell survival. In adenomyosis, however, mitochondria may function not only in energy metabolism but also as central regulators of calcium homeostasis, smooth muscle contractility, and fibrogenesis. In particular, abnormalities in Ca²⁺ transport and mitochondrial Ca²⁺ overload may provide a mechanistic basis for fibrosis through activation of the ROS–TGF-β–Smad signaling axis. This perspective is important because it supports the view that adenomyosis should not simply be regarded as an intrauterine manifestation of endometriosis but rather as a disorder of smooth muscle remodeling and fibrosis.

Furthermore, the PI3K–AKT and RAS–MAPK pathways may function as common upstream regulators capable of activating HIF-1α, NF-κB, Wnt/β-catenin, and Hippo–YAP/TAZ signaling pathways. However, the pathological differences observed between SUP/OMA and DIE/adenomyosis may reflect distinct microenvironmental contexts. In the hypoxic and inflammatory peritoneal environment, the HIF-1α–NF-κB axis may predominate, whereas in the mechanically stressed and high-stiffness myometrial environment, the Wnt/β-catenin–YAP/TAZ axis may become dominant, thereby driving either proliferation-dominant or fibrosis-dominant disease phenotypes. Mitochondria are known to adaptively remodel their respiratory activity in response to environmental changes and, through metabolic and structural reorganization, regulate cellular and tissue functions [106]. Therefore, elucidating how microenvironment-dependent mitochondrial adaptation mediates signaling divergence and phenotypic specification represents an important direction for future research.

From this perspective, the proliferation–fibrosis divergence model proposed in this review may help explain the distinct clinical manifestations of endometriosis and adenomyosis. Endometriosis is characterized predominantly by tumor-like proliferative properties, including a high recurrence rate and invasive behavior, whereas adenomyosis is characterized by progressive tissue stiffening and uterine enlargement, reflecting a more fibrosis-dominant phenotype. These differences may be determined less by the genetic alterations present within lesion cells themselves than by the mechanical and metabolic characteristics of the microenvironment in which the lesions reside.

Nevertheless, several important limitations of this model should be acknowledged. First, although KRAS and PIK3CA mutations have been reported in both endometriosis and adenomyosis, direct evidence demonstrating the extent to which these mutations activate the PI3K–AKT–mTOR or RAS–MAPK pathways within lesion tissues remains limited. Much of the current understanding is extrapolated from cancer biology studies [107–109], and it remains unclear whether the same molecular mechanisms operate in benign disorders such as endometriosis and adenomyosis.

Second, the boundary between endometriosis and adenomyosis is not always clearly defined. In particular, DIE appears to exhibit characteristics intermediate between the metabolically adaptive phenotypes of SUP and OMA and the fibrosis-dominant phenotype of adenomyosis. DIE is characterized by activation of TGF-β and YAP/TAZ signaling, ECM stiffening, and SMM, suggesting that its molecular pathology may more closely resemble that of adenomyosis. These observations support the notion that these disorders should not be viewed as dichotomous entities but rather as components of a continuous spectrum extending from proliferation-dominant to fibrosis-dominant disease states.

In addition, recent advances in single-cell and spatial transcriptomic analyses have revealed substantial cellular heterogeneity even within individual lesions [110, 111]. Consequently, future investigations should move beyond disease-based classifications and instead redefine these disorders in terms of cellular states. Understanding how the microenvironment and mitochondrial adaptation influence cell fate decisions will be essential for advancing this field. Such insights may facilitate the development of a unified pathogenic framework for endometriosis and adenomyosis and ultimately contribute to the design of more precise and effective therapeutic strategies.

Future perspectives

Future research should prioritize molecular pathological analyses that take lesion-specific microenvironments into account. Although endometriosis and adenomyosis have traditionally been regarded as single disease entities, SUP, OMA, DIE, and adenomyosis are exposed to distinct tissue environments and are therefore subject to different selective pressures that shape disease progression. Consequently, it is necessary to comprehensively evaluate somatic mutations, metabolic states, mitochondrial function, mechanical properties, and microenvironmental factors within each lesion type and to clarify how these factors collectively contribute to disease pathogenesis.

To advance our understanding of lesion-specific disease mechanisms, comprehensive investigations focusing on mitochondria will be particularly important. Mitochondria are not merely sites of energy production but are central organelles that regulate cellular metabolism, oxidative stress responses, calcium homeostasis, and cell fate determination. Therefore, future studies should extend beyond metabolic profiling to include multidimensional analyses of mitochondrial dynamics, such as mitochondrial fusion and fission, calcium flux, autophagy, and mitophagy, in order to elucidate their roles in microenvironmental adaptation.

A deeper understanding of these pathogenic mechanisms is also expected to facilitate the development of novel therapeutic strategies. Future treatment approaches may shift from uniform disease-based management toward precision medicine guided by dominant pathogenic drivers. For example, lesions characterized by predominant metabolic adaptation may be particularly amenable to mTOR inhibition or metabolism-targeted therapies [112, 113]. In contrast, in adenomyosis, the Ca²⁺–mitochondria–TGF-β axis may contribute to fibrogenesis and therefore represent a promising disease-specific therapeutic target. Furthermore, an integrated understanding of mitochondrial metabolism and mechanotransduction will be essential. Because YAP/TAZ and TGF-β function as key molecular mediators linking mechanical stimuli to metabolic reprogramming, they may represent particularly attractive therapeutic targets in DIE and adenomyosis [2, 72, 114].

Ultimately, integrating knowledge of somatic mutations, the microenvironment, mitochondrial adaptation, and cellular states will facilitate validation of a disease model that conceptualizes endometriosis and adenomyosis as a continuum. Such advances are expected to promote the development of more precise and effective diagnostic approaches and therapeutic interventions.

Conclusion

Endometriosis and adenomyosis may originate from a common foundation of endometrium-derived cells and shared somatic mutations, yet diverge into distinct pathological phenotypes as a consequence of differences in the microenvironments in which lesions develop. In endometriosis, metabolic reprogramming adapted to hypoxia, inflammation, and immune evasion predominates, thereby promoting lesion proliferation and survival. In contrast, adenomyosis is characterized by the predominance of mechanical stress, dysregulated calcium homeostasis, and profibrotic signaling, resulting in progressive tissue remodeling and fibrosis.

Mitochondria function as central integrators of these environmental signals by regulating metabolic adaptation, ROS production, calcium homeostasis, and fibrotic responses. In particular, mitochondria may primarily act as regulators of metabolism that support cellular proliferation and survival in endometriosis, whereas in adenomyosis they may serve as key remodeling regulators that integrate calcium signaling and fibrotic pathways.

Taken together, endometriosis and adenomyosis can be viewed as related disorders that arise from a common pathogenic foundation but diverge toward either proliferation-dominant or fibrosis-dominant phenotypes under the influence of distinct local microenvironments. The proliferation–fibrosis divergence model proposed in this review provides a conceptual framework for the unified interpretation of these diseases and may serve as a theoretical basis for the development of novel personalized therapeutic strategies, including metabolism-targeted therapies, antifibrotic interventions, and approaches aimed at modulating mitochondrial function.

Abbreviations

AKT

Protein kinase B

AMPK

AMP-Activated Protein Kinase

ARID1A

AT-Rich Interaction Domain 1A

Bax

Bcl-2-associated X protein

Bcl-2

B-cell lymphoma 2

CTNNB1

Catenin Beta 1

DIE

Deep infiltrating endometriosis

EMT

Epithelial–mesenchymal transition

ER

Endoplasmic reticulum

FMT

Fibroblast-to-myofibroblast transdifferentiation

GLUT1

Glucose Transporter 1

GRIM-19

Gene associated with Retinoid-IFN-induced Mortality 19

HIF-1α

Hypoxia-Inducible Factor-1 alpha

HK2

Hexokinase 2

IP3

Inositol 1,4,5-trisphosphate

KRAS

Kirsten Rat Sarcoma Viral Oncogene Homolog

LDHA

Lactate Dehydrogenase A

MAPK

Mitogen-Activated Protein Kinase

MCU

Mitochondrial calcium uniporter

MLCK

Myosin light chain kinase

MRI

Magnetic resonance imaging

mTOR

Mechanistic target of rapamycin

NF-κB

Nuclear Factor Kappa-light-chain-enhancer of Activated B Cells

OMA

Ovarian endometrioma

PDK1

Pyruvate Dehydrogenase Kinase 1

PI3K

Phosphatidylinositol 3-kinas

PIK3CA

Phosphatidylinositol-4,5-Bisphosphate 3-Kinase Catalytic Subunit Alpha

PINK1

PTEN-induced putative kinase 1

PLC

Phospholipase C

PTEN

Phosphatase and Tensin Homolog

ROS

Reactive oxygen species

S1P

Sphingosine-1-phosphate

SANRA

Scale for the Assessment of Narrative Review Articles

SIRT3

Sirtuin 3

SMM

Smooth muscle metaplasia

SOD2

Superoxide Dismutase 2

SUP

Superficial peritoneal endometriosis

TAZ

Transcriptional Co-Activator with PDZ-Binding Motif

TGF-β

Transforming Growth Factor-β

TIAR

Tissue Injury and Repair

VEGF

Vascular Endothelial Growth Factor

YAP

Yes-Associated Protein

Authors’ contributions

HK designed the study, developed the literature search strategy, conducted comprehensive literature searches on relevant scientific publications, and drafted the manuscript. This author read and approved the final version of the manuscript.

Funding

No funding was received for this study.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.