From Mice to Primates: Assessing Hormone-Based Endometriosis Models for Preclinical and Therapeutic Insights

Endometriosis remains a complex and poorly understood gynecological disorder, necessitating reliable animal models to elucidate its pathogenesis and evaluate potential therapies. This review analyzed hormone-based animal models of endometriosis, comparing induction methods and their translational relevance. Our novel scoring system highlighted the strengths and limitations of various models, with non-human primates—particularly the unopposed estrogenicity model in baboons—emerging as the most physiologically relevant due to their close hormonal and genetic resemblance to humans. However, practical constraints such as cost, ethical considerations, and feasibility continue to make rodent models indispensable for preliminary research. The findings underscore the need for improved preclinical models that better replicate human endometriosis, particularly in hormonal and immunological contexts. Future research should prioritize genetically modified models and refined hormonal induction techniques to enhance translational outcomes. By optimizing model selection based on our scoring criteria, researchers can improve the reliability of preclinical studies, ultimately accelerating the development of targeted therapies for endometriosis. This comprehensive evaluation provides a valuable framework for guiding future investigations and bridging the gap between experimental findings and clinical applications.

Endometriosis is a non-cancerous condition characterized by the growth of tissue resembling the endometrium in areas outside the uterine cavity. It commonly leads to symptoms such as ongoing pelvic discomfort, pain associated with the menstrual cycle, and difficulties with fertility. It is estimated that over 10% of women are affected by this condition. 1 , 2 While various mechanisms of endometriosis generation have been thoroughly studied, the processes leading to its development and maintenance are not yet completely understood, and significant gaps in research remain. The development of new surgical methods and medications for the prevention and elimination of disease or its recurrence necessitates defining the pathophysiologic mechanisms. Endometriosis is a multifactorial disease, and various models and approaches have been proposed to explain its pathophysiology. 3 Widely accepted pathophysiological aspects can be classified as inflammation, 4 immunological factors, 5 environmental factors, 6 genetics, 7 pain, 8 infertility, 9 and hormonal. While several theories are described in each category, it is not certain which one is the primary reason for the occurrence of endometriosis. Still, one of the main theories for the ectopic formation of endometrial tissue in the peritoneal cavity is considered to be the old “retrograde menstruation”. 10 Retrograde menstruation, a theory proposed over 100 years ago, suggests that menstrual blood flows backward through the fallopian tubes into the pelvic cavity. 11 While this theory has historical significance, it is important to note that more recent research has proposed alternative mechanisms for the development of endometriosis, including the stem cell origin theory. 12 This newer theory suggests that stem cells from the bone marrow or endometrial tissue itself may contribute to the formation of endometriotic lesions. 13 Estrogen-dependency of endometriosis has been well documented, 14 and drugs that reduce or suppress sex steroid production in ovaries for endometriosis treatment 15 or relief of its symptoms, such as pain, 16 have been proposed. One of the key modulators in endometriosis pathogenesis is the increasing local production of estrogens. 17 Many effects of sex steroids are mediated by their actions at their nuclear receptors, but growing evidence suggests that some effects may occur independently of these receptors. 18 Due to the delay in diagnosing the disease and its progression by the time it is diagnosed, conducting experiments to detect predisposing factors related to synthesis or involved in sex steroid effects in endometriosis is not feasible. Due to ethical considerations, controlled experiments such as disease progression monitoring by repeated laparoscopies are limited. Therefore, animal models of endometriosis can be valuable for conducting extensive research on the effects of sex steroids and their mechanisms, including ectopic endometrial tissue adherence, invasion, and vascular establishment. They also facilitate the development of new approaches to study the initial onset, recurrence prevention, and the treatment of lesions. The closest models to humans for evaluating endometriosis have been non-human primates. However, their use raises serious ethical concerns due to their high cognitive abilities, social structures, and the challenges associated with their care and handling. Due to their high cost and handling limitations, small laboratory animals have been used as alternative models of endometriosis. This literature review aims to provide comprehensive insight into hormone-related animal models of endometriosis and compare different induction methods, along with their advantages, disadvantages, and applications for evaluating sex steroid hormones (SSH) in the pathogenesis of endometriosis. Additionally, a summary of all published studies is provided, and a novel scoring table is designed to provide a full understanding of the most important aspects of available animal models. We used PubMed Central/Medline databases to find research on sex hormone-related endometriosis. All included studies were published between 2010 and 2024. The Mendeley Desktop application 1.19.8 (Mendeley Ltd., London, UK) was used to screen the results. The following search query was utilized to discover all related papers, resulting in 158 related papers within our desired time interval: “endometriosis”[Title/Abstract] AND (“animal”[Title/Abstract] OR “model”[Title/Abstract]) AND (“hormone”[Title/Abstract] OR “sex”[Title/Abstract] OR “steroid”[Title/Abstract]). We included studies published between 2010 and early 2024 that proposed an endometriosis induction method in animals as the first criterion. An assessment of SSH or hypothalamic-pituitary-gonadal (HPG) axis hormones (receptors or concentration assessment) was the second inclusion criterion. We summarized all the studies that met our inclusion criteria in table 1 , and articles that represented a novel induction method were classified separately. The more traditional and widely used approaches in these studies were also reviewed and explained separately. Hormone-based treatment of endometriosis on previously approved animal models of endometriosis. Instead of the referenced method, we mentioned the source article that provided the utilized method. The assessed article has mentioned modification of the source method. We considered sex hormones in terms of gonadal steroid hormones: testosterone, androgens (including testosterone), estradiol, estriol, estrone (estrogens), progesterone, progestogens (including progestins) as SSH. 19 Furthermore, luteinizing hormone, follicle-stimulating hormone, gonadotropin-releasing hormone (GnRH), and melatonin were considered hormones of the HPG axis. We excluded non-English articles and review articles. We also did not consider studies that utilized steroid supplementation while inducing endometriosis in murine models as hormone-related studies. This section includes all sex steroid hormone-related studies of endometriosis and provides comprehensive insight into assessed therapeutics for endometriosis using animal models ( figure 1 ). Additionally, this section outlines targets of interest in the endometriosis state, mechanisms, and potential drugs that significantly ameliorate the condition. At the end of this section, a review of more frequently cited induction methods was prepared to contrast older methods with newer ones. This figure illustrates the applications of different types of animal models in endometriosis research. ¶ Non-human primate model of endometriosis is performed in species including baboons, rhesus monkeys, marmoset monkeys, and cynomolgus monkeys. § Spontaneous endometriosis is reported in baboons, rhesus monkeys, and cynomolgus monkeys Hormone-based endometriosis studies have explored both the role of hormones in endometriosis treatment and their influence on the generation of endometriosis models. These studies aim to understand how hormonal fluctuations can affect the development and progression of endometriosis, as well as how hormone therapy can be used to manage its symptoms. By investigating both aspects, researchers can gain a deeper understanding of the hormonal dynamics involved in endometriosis and develop more effective therapeutic strategies. Sex steroid hormone-related studies, HPG axis hormones, and their role in endometriosis, and the frequently used methods are shown in table 1 . The effectiveness of therapeutics in endometriosis models is influenced by the method used to induce lesions. Different methods, such as suturing, injection, gluing, and auto-transplantation, have varying effects on the inflammatory response and lesion characteristics. This section provides a comparison of these methods, highlighting their impact on study outcomes and the evaluation of treatments. Affects the inflammatory response significantly, leading to higher immune cell infiltration and more pronounced lesion development. Commonly used in studies evaluating the inflammatory pathways and immune responses to treatments. 29 , 48 Allows for precise control over lesion size and location, but can induce variable inflammatory responses depending on the injection site and technique. 63 , 64 Provides consistent lesion induction with moderate inflammatory response, suitable for studies focusing on lesion growth and treatment efficacy. 65 Mimics the natural process of endometriosis more closely, with moderate to high inflammatory response. Used in long-term studies to evaluate chronic effects and treatment outcomes. 61 We separated studies with new approaches toward endometriosis induction to assess them exclusively with a novel scoring system. This scoring system ( table 2 ) represents a modified combination of previously proposed scoring models to enhance the prediction of drug assessment while utilizing animal models. Detailed sections including auto-transplantation models of endometriosis related to SSH and HPG hormones, xenotransplantation models of endometriosis related to SSH and HPG hormones, allotransplantation models of endometriosis, and spontaneous endometriosis are discussed, as shown in figure 2 . 75 Comparison of newly proposed methods of endometriosis induction for hormone-related studies Scoring for each study is considered as follows: Species: Non-human primate=2, Non-human mammal=1 Induction method: spontaneous disease occurrence=3, xenotransplantation=2, autotransplantation /allotransplantation=1 Pharmacological effects: Consistent drug effects with human=+1, No pharmacological evaluation=0, Non-consistent drug effects with human=-1 Hormonal evaluation: More than one similar hormonal marker regulation to human=+2, Single marker regulation similarity to human=+1, Not mentioned=0, Non-consistent hormonal marker(s) regulation(s) with human=-1 Genetic evaluation: More than one similar gene expression with human=+2, Single gene expression similarity with human=+1, No genetic evaluation=0, Non-consistent gene(s) regulation(s) with human=-1 Histological evaluation: Ectopic lesion(s) detection=+1, Not mentioned=0, No ectopic lesion(s) detection=-1 Feasibility: High feasibility=2, Moderate feasibility=1, Low feasibility=0 High feasibility (2): Models that are easy to source, maintain, and handle in most research settings (e.g., common rodents). Moderate feasibility (1): Models that require some specialized resources or care but are still manageable (e.g., specialized rodent strains or larger mammals). Low feasibility (0): Models that are difficult to source, require specialized environments, or are impractical for many labs (e.g., non-human primates). Ethics: Low ethical concerns = 2, Moderate ethical concerns=1, High ethical concerns=0 Low ethical concerns (2): Models with well-established ethical guidelines and lower welfare concerns (e.g., invertebrates or lower-order mammals like mice and rats). Moderate ethical concerns (1): Models that require more stringent oversight and ethical considerations (e.g., larger mammals). High ethical concerns (0): Models subject to significant ethical scrutiny and regulatory requirements (e.g., non-human primates). Financial Cost: Low cost=2, Moderate cost=1, High cost=0 Low cost (2): Models that are relatively inexpensive to acquire and maintain (e.g., common laboratory rodents). Moderate cost (1): Models with higher acquisition or maintenance costs but still within reach for many research budgets (e.g., some specialized rodents or larger mammals). High cost (0): Models that are very expensive to acquire, maintain, and use in research (e.g., non-human primates or very specialized animal models). This figure presents different types of animal models used in endometriosis research. * Uterine tissue fragments or biopsies are used without enzymatic cell isolation. † Orange circles represent different species as the source of tissue, while white circles indicate the same species as the source of tissue. ¶ The non-human primate model of endometriosis has been established in species such as baboons, rhesus monkeys, marmoset monkeys, and cynomolgus monkeys. § Spontaneous endometriosis has been reported in baboons, rhesus monkeys, and cynomolgus monkeys. We classify the induction methods into ‘earlier’ and ‘new’ categories to underscore the evolution of animal models in endometriosis research. Earlier methods, such as auto-transplantation and hormone-induced models, have been fundamental in exploring the basic mechanisms of the disease. However, these methods do not fully replicate the complexities of human endometriosis, often due to oversimplification in the model design. On the other hand, new induction methods, including genetically modified animal models and more refined hormonal treatments, better reflect the natural progression of endometriosis, allowing for a more detailed exploration of its molecular and cellular mechanisms. This classification helps highlight the advancements in research methodologies and their implications for understanding the pathogenesis of endometriosis. We used a scoring table ( table 2 ) for model assessment, which was inspired by Ferreira and colleagues, 76 Denayer and colleagues, 77 and Sams-Dodd and others. 78 We also utilized proven hormonal markers (ER1. PR, CYP17A1) and other genetic markers (BDNF, AGTR1, CCL2, C3, CD40, TIMP2, SERPINE1, CYP17A1, IGF1, IGF2, IL10, MMP1, MMP7, and MMP9) 79 to evaluate the models. The scoring was designed in a way that contrasts between the suitability of available studies would be easily observable. Our scoring rationale was considered as follows: 1. Non-human primate models gained higher scores than non-human mammals due to more resemblance to humans physiologically and pathologically: (Non-human primate=2, Non-human mammal=1) 2. Spontaneous disease occurrence, which is exclusive to non-human primates, gained the highest scores. Xenotransplantation models gained higher scores than autotransplantation/allotransplantation models due to utilizing cells or tissues from humans, which can be perceived as closer to the human pathological state: (Spontaneous disease occurrence=3, Xenotransplantation=2, Autotransplantation /allotransplantation=1) 3. Consistent pharmacological effects in humans by available verified drugs can be a promising criterion for better future translational animal modeling: (Consistent drug effects with human=+1, No pharmacological evaluation=0, Non-consistent drug effects with human=-1) 4. Relevant life stage in animal models is a key factor for suitability due to the occurrence of different outcomes with drug treatments in different stages of life: (Identical life stage=+1, Not mentioned=0, Non-identical life stage=-1) 5. Previous studies have suggested extensive hormonal dysregulations in endometriotic animal models, which can comply with the human physiologic state. The more similar the dysregulation, the better the translational animal model would be: (More than one similarity of hormonal markers regulations to human=+2, Single marker regulation similarity to human=+1, Not mentioned=0, Non-consistent hormonal marker(s) regulation(s) with human=-1) 6. Numerous genes get disrupted in the state of endometriosis. Gene regulations can be consistent with human body dysregulations or not: (More than one similarity of genes expression with human=+2, Single gene expression similarity with human=+1, No genetic evaluation=0, Non-consistent gene(s) regulation(s) with human=-1) 7. One of the traditional ways of validating endometriosis existence is laparoscopy. The evaluation of engaged tissues by staining or observation of ectopic lesions is still the gold standard for disease diagnosis: (Ectopic lesion(s) detection=+1, Not mentioned=0, No ectopic lesion(s) detection=-1) Table 2 provides novel insight for future studies to choose the most suitable model. It is worth mentioning that despite achieving fewer scores for non-human mammals than non-human primates, they benefit from wide availability, easy handling, cost-effectiveness, faster growth, and breeding. In the segment of non-human mammals, the study by Burns and colleagues, 69 Burns and coworkers, 64 and Greaves and others 29 gained the highest score by providing key elements and resemblance to human endometriosis. 29 Greaveset and others proposed their model in non-immunocompromised mice, which can be considered an advantage for studies involving immunologic topics. 29 This simple selection can produce inaccurate results for clinical trials and major analyses. Despite providing a considerable approach for xenotransplantation, Huang and colleagues did not provide an adequate animal modeling assessment. 68 Although the mouse model by Kulak and others 35 and Naqvi and colleagues 71 gained a considerable grade, it is not recommended to use the outbred stock (CD-1) due to genetic variation and the availability of other inbred mouse strains. Transgenic animals are gaining increasing attention for further investigation in endometriosis studies. Some of these transgenic mice exhibit features of endometriosis, while others serve primarily to facilitate the investigation of the disease. The endometriotic state is induced using the previously developed methods. Transgenic animal models in this field are among the most promising tools for precisely examining the functions of various genes. Non-human primates are considered to be very similar to humans genetically and physiologically. Additionally, they can develop spontaneous endometriosis, which can be a significant advantage for research. In addition to similarities to humans, endometriosis is confirmed more accurately considering hormonal assessment; therefore, they achieved higher scoring results. In this segment, Nair and colleagues have proposed the best model, scoring-wise. 74 Unlike other models of non-human primates, this model utilized anti-progestin to develop endometriotic lesions in baboons. Therefore, it can be proposed as the best available animal model for hormone-related endometriosis assessment. Utilizing novel animal models to assess endometriosis pathophysiology is not a new approach. 26 , 80 There has been extensive research using proposed animal models to assess etiology, 81 diagnosis, 82 resemblance, 83 genetics, 84 biomarkers, 85 and therapeutics 86 for endometriosis disease. Sometimes these studies have reported promising results in encountering endometriosis as a multifactorial gynecological disease. 87 However, managing the challenges associated with endometriosis diagnosis remains a significant concern. Despite extensive research efforts, non-invasive, definitive, and reliable diagnostic methods for endometriosis have yet to be established. 88 Consequently, women suspected of having endometriosis still endure pain and the risks associated with invasive procedures such as laparoscopy. Animal models have made significant contributions to understanding this disease and overcoming related challenges. However, due to limited data, comprehensive reviews such as systematic reviews or meta-analyses have struggled to fully address these issues. 89 Therefore, it is essential to revisit and revise preclinical procedures, including animal modeling, to improve research quality and generate more reliable and authentic data.