Abstract
Successful embryo implantation relies on embryo quality, endometrial receptivity, and the synchronized development of both. In recent years, methods for assessing endometrial receptivity have considerably advanced, shifting from traditional imaging techniques to modern molecular biology approaches. The evaluation of endometrial thickness, morphology, subendometrial blood flow, and peristalsis before implantation has become less prominent, whereas molecular markers such as pinopodes, integrin αvβ3, its ligand osteopontin, and homologous box gene A10 are widely used in clinical practice for assessing endometrial receptivity. To provide a comprehensive overview, this narrative review systematically searched PubMed and Web of Science for studies on endometrial receptivity assessment in infertile women over the past two decades, with a focus on both clinical and experimental research. This narrative review has examined recent advancements in morphological markers, endometrial receptivity–related genes, endometrial microbiota, and other emerging indicators, emphasizing their potential in optimizing embryo implantation and improving the success rates of assisted reproductive technologies. Notably, endometrial receptivity array testing and uterine microbiome analysis have gained substantial attention for their potential in personalized diagnosis and treatment. This review aims to provide scientific insights for precise infertility diagnostics and individualized treatment strategies while offering recommendations for future research development.
Keywords
Introduction
Infertility is a widespread health problem that impacts the lives of millions of couples trying to conceive. As per a report from the World Health Organization in 2023, approximately 12.6%–17.5% of couples of reproductive age are struggling with infertility globally. 1 Although the quality of fertilized eggs has always been a focus in clinical practice and has played an important role in assisted reproductive technology, with the continuous development and improvement of embryo in vitro culture technology, the quality of embryos has been greatly improved. However, the success rate of embryo implantation has not yet achieved the desired improvement. 2 Some experts have suggested that the reason for failed implantation lies in the inadequate receptivity of the endometrium, making it difficult for embryos to thrive. This has led to a new topic in the field of reproductive medicine on how to improve the success rate of embryo implantation and enhance endometrial receptivity (ER). 3
ER, a key focus in reproductive medicine research, refers to the unique state of the endometrium that allows for embryo attachment, embryo penetration, and induction of stromal changes, ultimately leading to pregnancy. 4 When evaluating ER status with clinical precision, endometrial biopsy is considered the most accurate approach. Under electron microscopy, microscopic structures such as pinopodes are observed, which are considered the gold standard for assessing the morphological aspect of ER. 5 Research on various regulatory factors on the surface of the endometrium (e.g. integrin αvβ3 subtype and osteopontin) as well as hallmark molecules (e.g. homeobox gene A10) is also gaining increasing attention from experts. 6 Researchers have gradually delved into the microscopic level by studying a multitude of macroscopic indicators, especially focusing on ER chip testing, endometrial microbiota, and endometrial microRNA examination in patients with repeated implantation failure (RIF) during assisted reproduction processes, which can alter ER to some extent. 7 This review primarily focuses on the molecular biology markers that alter the receptivity of the endometrium.
This study systematically searched PubMed and Web of Science databases for studies related to ER assessment in infertile women over the past two decades, with a specific emphasis on clinical and experimental research. This study summarized key molecules and biomarkers affecting ER in infertile patients (including those with natural infertility, those undergoing ovulation induction, and those undergoing in vitro fertilization (IVF) treatment), such as phagocytosis, integrin αvβ3 including osteopontin, homologous box gene A10 (HOXA10), leukemia inhibitory factor (LIF), endometrial receptivity array (ERA), and endometrial microbiota (Table 1). This review is guided by the Scale for the Assessment of narrative review articles. 8
Summary of molecular and biological markers for assessing ER.
LIF: leukemia inhibitory factor; ERA: endometrial receptivity array; HOXA10: homologous box gene A10; RIF: repeated implantation failure; ER: endometrial receptivity.
Phagocytosis
During the luteal phase of mammalian menstruation, the microvilli structures on the surface of the endometrial epithelium fuse together to form an enlarged, smooth membrane protrusion resembling a “balloon-like” structure; hence, they are named as pinopodes. 5 During a regular menstrual cycle, the implantation window for the endometrium lasts usually approximately 6–7 days post-ovulation, approximately on days 20–24. Simultaneously, trophoblasts undergo morphological changes through three stages—development, maturation, and regression, as observed under an electron microscope. Endometrial pinopodes undergo distinct morphological changes across development. In early stages, the cells expand as microvilli reduce and fuse, forming smooth membrane protrusions. Maturation pinopodes exhibit characteristic “blister-like” swelling with smooth surfaces containing few or no microvilli and occasional fine apical wrinkles, indicating optimal ER. During regression, prominent surface folds develop and microvilli reappear, indicating closure of the implantation window. Typically, microvilli begin to alter and fuse around day 17 of the menstrual cycle, disappearing around day 24, aligning with the timing of the implantation window. 9
Individuals exhibit variations in the morphology and quantity of endometrial pinopodes, which in turn affect the receptivity of the endometrium and consequently impact embryo implantation. 10 The fully developed endocervical polyp appears with a swollen and plump membrane status, indicating good tolerance of the endometrium. 11 Aunapu et al. found that in assisted reproduction, patients with a lack of cilia had a lower clinical pregnancy rate than those with abundant cilia. Additionally, patients who experienced repeated IVF and embryo transfer (IVF-ET) failures showed a reduction or disappearance of microvilli structures in their endometrium as well as varying degrees of degeneration or disappearance of cilia. 12 Although the abundance of cilia in the endometrial surface does not necessarily indicate fertility, a randomized controlled study found that patients undergoing assisted reproductive technology with incomplete expression of cilia and a count <85 had a considerably higher rate of miscarriage and RIF than normal patients (count >85). This suggests that cilia have significant implications in patients with recurrent miscarriages and RIF. 13 Research demonstrates that L-selectin ligands, which are key glycoproteins involved in embryonic implantation, are primarily concentrated on the pinopodes of endometrial epithelial cells, with particularly high expression at initial blastocyst attachment sites. The absence or significant reduction of these ligands is strongly correlated with dramatically decreased or completely eliminated pregnancy potential, highlighting their essential role in successful embryo–endometrium interaction during implantation. 14
Additionally, the expression of the corpus luteum is influenced by the levels of serum estrogen and progesterone. Fluctuations in progesterone levels in the bloodstream can impact the development and regression of the corpus luteum. Stavreus-Evers et al. conducted a study involving 27 healthy women with normal menstrual cycles to measure hormone levels and hormone receptor levels using immunohistochemical methods. They found that the increase in serum progesterone levels and downregulation of progesterone receptors play a crucial role in the development of the corpus luteum, whereas the levels of serum estrogen and expression of estrogen receptors do not affect the development of the corpus luteum. 15
Although substantial research suggests that pinopodes play an important role in assessing ER, there are still limitations in clinical practice that need further discussion and research. First, pinopodes require an invasive biopsy of the endometrium, which limits their widespread use in clinical settings. Second, biopsies can only detect a small amount of endometrial tissue and may not reflect the overall condition of the entire endometrium. Early literature has suggested that simply counting pinopodes cannot fully describe the implantation window or accurately assess ER.15,16 Third, the morphology of podocyte foot processes is obtained through scanning electron microscopy. The evaluation of podocyte foot processes in the stages of development, maturation, and decline lacks an objective assessment standard, making it somewhat subjective. This leads to a slightly biased assessment of podocyte foot processes. However, as research on organoid and other in vitro culture systems progresses, a deeper understanding of the formation mechanism and structure of podocyte foot processes can be achieved, thereby shedding light on factors contributing to implantation failure.5,17 Another future direction is utilizing noninvasive methods to detect cell membrane protrusions combined with artificial intelligence (AI) technology to analyze electron microscopy results.
Integrin α vβ 3 including osteopontin
Integrins are a major group of transmembrane glycoprotein receptors present in various cell types within the human body. Among them, integrin αvβ3 stands out as the prime molecular marker for the embryo implantation window. Integrin αvβ3, along with its ligand osteopontin, engages in a calcium-dependent ligand–receptor interaction, playing a role in endometrial decidualization and the growth and development of embryos. It is associated with genes related to ER and is involved in the adhesion and implantation process of blastocysts. 18
During their study on the endometrium during the implantation period, Franchi et al. found that the expression levels of integrin αvβ3 and its ligand osteopontin were higher in the epithelial tissue of the “implantation window” phase than in stromal cells, with high expression within epithelial cells closely associated with embryo implantation. 19 Moreover, bone morphogenetic protein can be found in endometrial and placental nourishing cells during the implantation period and is easily influenced by estrogen levels, facilitating placental decidualization and subsequently regulating early embryo implantation and placental development. 20 The dysfunction of integrin αvβ3 subtype and osteopontin expression can cause damage to ER and can disrupt the expression of ovarian hormones as well as various cytokines and growth factors, consequently leading to infertility. 21 During the clinical IVF-ET superovulation process, the increase in progesterone levels on the day of human chorionic gonadotropin (hCG) injection can lead to abnormal secretion and expression of osteopontin, based on microRNA arrays and microarrays. This can also result in decreased ER, ultimately leading to a decrease in clinical pregnancy rates and live birth rates. 22 Therefore, many studies have focused on how to improve the expression of integrin αvβ3 and its ligand osteopontin in patients with RIF, thereby enhancing the receptivity of damaged endometrium and increasing the live birth rate in patients with RIF.23,24 In patients with polycystic ovary syndrome (PCOS), a decrease in integrin αvβ3 expression was observed on the endometrium during the implantation window, leading to abnormal mid-secretory endometrial histological development. This decreased expression of integrin αvβ3 in the endometrium causes asynchrony with embryo development, affecting its normal implantation. This could be one of the reasons for the decreased pregnancy rate and increased miscarriage rate in patients with PCOS. 25 The impact of integrin αvβ3 and osteopontin on ER has been demonstrated in the abovementioned studies, thus indicating their role in the process of embryo implantation.
HOXA10
The homeobox gene family, including HOXA10, plays a crucial role in regulating the proliferation and differentiation of endometrial cells in the human body. It also has the ability to control the development and differentiation of embryos. 26 The endometrium of adult women experiences a cycle of growth, secretion, periodic tissue breakdown, reconstruction, and reorganization, which is similar to the developmental process of embryos. Studies have indicated that the expression of HOXA10 is downregulated during the proliferative phase of the endometrium, is upregulated during the secretory phase, and is higher in the stroma than in the glands. 27
Abnormal expression of HOXA10 can affect embryo implantation. Animal experiments have shown that the lack of HOXA10 in a rat model of PCOS can affect embryo implantation. Treatment with estrogen can promote the proliferation and differentiation of endometrial cells, increasing the endometrial thickness and the ratio of glandular/stromal area as well as enhancing ER. The main potential mechanism is the regulation of HOXA10 by sex hormone levels. Patients with PCOS have elevated androgen levels, which can decrease the binding of estrogen and its receptor, as well as exhibit inherent low estrogen status, all of which can affect the expression of HOXA10 in the endometrium. These factors may contribute to the low pregnancy rates and high miscarriage rates in patients with PCOS. 28
The expression pattern of HOXA10 aligns with that of integrin αvβ3. As the 5′-end of the integrin αvβ3 gene contains a binding site for HOXA10, it can be concluded that integrin αvβ3 is a downstream target gene regulated by HOXA10. 29 Through the HOXA10 pathway, estrogen and progesterone alter the expression of integrin αvβ3 on the uterine lining. Researchers in China have discovered that treatment with kidney-nourishing and fetus-nurturing pills can alter hormone levels in mice and increase the expression of HOXA10, integrin αvβ3, and other related proteins such as LIF, calcitonin, and mucin 1, leading to enhanced endometrial shedding, improved ER, and higher implantation success rates. 30 A significant decrease in the expression levels of HOXA10 and its protein was observed in the endometrium of 12 patients with RIF and 20 patients with recurrent miscarriage compared with the normal population. HOXA10 and its protein play a potential role in the embryo implantation process, influencing the pregnancy outcomes of women with recurrent pregnancy failure. 31 Research on the mechanisms underlying HOXA10 revealed that the collaboration between long noncoding RNA nuclear-enriched abundant transcript 1 (lncRNA NEAT1) and CCCTC-binding factor (CTCF) could boost HOXA10 promoter activity and expression, leading to increased proliferation of endometrial epithelial cells and the development of ER, ultimately supporting embryo implantation. 6 A study by an Italian scholar Stefano Palomba revealed that changes in the endometrial structure of patients with PCOS is the main reason for miscarriage or implantation failure. Treatment options such as lifestyle adjustments, medication (such as metformin and aspirin), or even surgical interventions (such as laparoscopic ovarian drilling) can improve hormonal imbalances in patients, affecting the expression of HOXA10 and integrin αvβ3, thus improving endometrial dysfunction.32,33
LIF
LIF, a multifunctional cytokine from the interleukin-6 (IL-6) family, is abundantly present in uterine and embryonic tissues. It plays a crucial role in embryonic growth, development, and implantation, serving as a marker for ER. 34
In mice with LIF deficiency, ovulation and fertilization occur normally, but embryo implantation does not occur. However, if the embryos are implanted into female mice with normal LIF expression, embryo implantation and development occur normally. 35 Concurrently, research has revealed that transmembrane glycoprotein 130 (gp130) affects signal transduction and transcription activation factor (STAT) pathways as well as the binding of LIF. Knockout mice lacking gp130 and STAT3 genes showed rapid degeneration and death of embryos on days 6–7 after fertilization. 36 However, upon exogenous supplementation of LIF to female mice with LIF knockout, a gradual improvement was observed in the inability of the uterus to shed its lining as well as the inability of the embryos to implant, indicating the indispensable role of LIF in embryo implantation. 37
During the implantation window, patients undergoing IVF-ET for assisted reproduction in the pregnant group showed higher levels of LIF protein in the endometrial secretions than those in the nonpregnant group. This suggests that the levels of LIF secreted by the endometrium could serve as a predictive marker for ER. 38 Examination of the endometrial lining in women experiencing RIF revealed a significant decrease in LIF expression levels. Treatment with medications, such as 65 kDa mycobacterial heat shock protein (HSP65) or tacrolimus, resulted in elevated LIF levels, thereby improving the clinical implantation rate in patients with RIF.39,40 These findings indicate that LIF plays a crucial role in regulating ER, embryo implantation, and endometrial shedding processes.
ERA
ERA is based on recent advances in whole-genome analysis technology. It involves measuring the expression patterns of 238 genes in the female endometrium and analyzing the differences in gene expression levels between the implantation and preimplantation phases. Developed by Spanish scholars Carlos Simon and Díaz-Gimeno in 2011, this technique categorizes the endometrium into receptive and nonreceptive phases based on the analysis of the implantation window. The nonreceptive phase is further divided into prereceptive and postreceptive stages, aiding in the determination of the optimal window for embryo implantation. 41 Upon further investigation by Díaz-Gimeno et al., it was discovered that the transcriptional genomic features of endometrial gene expression profiles remained unchanged between cycles or over a period of 29–40 months. This finding offers a potential testing method for assessing ER in patients with RIF. 42 Hashimoto et al. conducted an analysis of uterine factors in 44 of the 50 patients who experienced RIF. Through endometrial testing on LH + 7D (7 days after the luteinizing hormone surge), they identified that 24% of the patients were in a nonreceptive phase. By tailoring the transfer protocol based on ERA results, they found a 50% pregnancy rate in patients who were previously nonreceptive (n = 10). The utilization of ERA testing to pinpoint the appropriate embryo implantation window holds significance for patients with unexplained RIF. 43
However, some studies have found that although ERA testing can help determine the optimal timing for embryo transfer, this adjustment does not always lead to significant improvements in clinical pregnancy outcomes. This indicates that ERA testing cannot universally alter the transplant strategy or guarantee higher success rates in all cases. Although it has shown potential in certain situations, such as for patients with RIF, its benefits are not consistent across all patient populations. Therefore, further research is needed to better understand its limitations, refine its application, and identify the specific groups of patients who are most likely to benefit from this approach. 44
With the development of the second-generation sequencing technology, ERA is not the only genetic testing method available. Utilizing high-throughput sequencing technology for transcriptome sequencing (RNA sequencing) and analyzing the expression levels of relevant genes in endometrial tissue, combined with AI machine learning, the endometrial receptivity test (ERT) has been developed. This method involves constructing a model for ER gene testing of over 10,000 genes. ERT can more accurately differentiate between prereceptive, receptive, and postreceptive endometrium than ERA, with a detection accuracy of 93.3%–100%. It can detect endometrial window displacement in 69.67% of patients with RIF and provide personalized implantation strategies, resulting in a pregnancy rate of 55.6%. 45 With the advancement of technology, AI and machine learning models can be used to analyze patients’ basic information and develop low-cost, noninvasive, and precise tools for assessing ER.
Li et al. investigated 155 patients with RIF, categorizing them into two groups: (a) 60 patients underwent RNA sequencing–based endometrial receptivity testing (rsERT), revealing that 60% (36/60) of the patients had implantation window displacement and (b) 95 patients received standard frozen embryo transfer without molecular guidance. The rsERT-guided group showed significantly higher pregnancy rates (β-hCG positive: 56.3% vs. 30.5%, P = 0.003; clinical pregnancy: 43.8% vs. 24.2%, P = 0.017), demonstrating that inadequate ER, which accounts for approximately two-thirds of implantation failures, can be effectively addressed through precision-timed embryo transfer based on transcriptomic analysis. 46
In pathology, the traditional HE staining method combined with either ERA or ERT is a highly valuable method for assessing ER due to its high accuracy and good repeatability. However, there is an urgent need for evaluating ER in patients with RIF using large samples and simpler detection methods, which warrants clinical research on large sample data.
Certainly, it is important to recognize that there are some limitations, as both ERA and ERT are expensive and invasive tests in clinical practice. This limits their routine application in normal ovulation induction or controlled ovarian stimulation, making widespread adoption challenging in clinical settings. 47
Endometrial microbiota
Previously, humans relied on cultivation techniques to investigate microbial communities, but this approach was limited in detecting only a small number of microorganisms. Consequently, it was widely believed that the presence of a cervical mucus plug indicates that the uterus is sterile in a healthy state. Nevertheless, advancements in new sequencing technologies and whole-genome shotgun sequencing technologies have led to a new understanding and discovery of endometrial microbiota, revealing the presence of microbes within the uterine cavity. Research has shown an increasing presence of a uterine microbiome in the human body. However, the presence of harmless endometrial microbiota in the normal human body, known as normal flora or stable ecological flora, is crucial for maintaining the stable physiological state of the uterine cavity. The endometrial microbiota may regulate key pathways necessary for successful embryo implantation and pregnancy development. Imbalance in the proportion of endometrial microbiota may lead to reproductive tract infections or increase the risk of endometrial infection, potentially affecting embryo implantation. 48 Recent studies have shown a growing amount of evidence linking the composition of endometrial microbiota to early miscarriage. 49
The composition of the endometrial microbiota is currently a topic of debate. An analysis of the microbial communities in the endometrial fluid and vagina of healthy women of reproductive age and infertile women revealed differences in the microbial composition between the endometrial and vaginal microbiota. The endometrial microbiota is primarily composed of Bacillus and Gardnerella, with the presence of Lactobacillus species being linked to pregnancy outcomes. A predominance of Lactobacillus species in the endometrial microbiota increases the embryo implantation rate (Lactobacillus genus >90%), whereas a microbial community dominated by non-Lactobacillus species decreases the embryo implantation rate, ongoing pregnancy rate, and live birth rate. 50 Compared with the uterine microbiota in a healthy pregnancy, the microbial spectrum of uterine samples before a miscarriage exhibits higher bacterial diversity and lower abundance of lactobacilli. Recent studies have shown that the bacterial communities differ in various parts of the uterus in women, with lactobacilli mainly found in the vagina and cervix. However, they are less prevalent on the uterine lining, where nonmotile rods and Pseudomonas are predominant. It was also observed that the abundance of lactobacilli decreases gradually from the cervix and vagina to the uterine lining. 51
Imbalance in the microbiota of the endometrium could lead to chronic inflammation of the endometrium, which in turn can affect its receptivity and ultimately impact embryo implantation and development, potentially causing infertility. 52 The supplementation of controlled ovarian hyperstimulation and progesterone during the IVF-ET–assisted reproduction process for infertile women can lead to improvements in the composition of the vaginal and endometrial microbiota, ultimately influencing ER and placental formation. 53 Cicinelli et al. conducted hysteroscopy in 106 women with RIF, with 70 diagnosed with endometritis. Analysis of the endometrial flora revealed the presence of Escherichia coli in 33% of the patients, Klebsiella pneumoniae in 23%, and Chlamydia trachomatis in 30%. Following a standardized 6-month antibiotic treatment, there was a significant increase in clinical pregnancy and live birth rates. The possible mechanism is that dysbiosis of the endometrial microbiota leads to inflammatory changes in the uterine cavity, delaying the window of implantation and reducing ER, thus resulting in RIF. 54 Accumulating evidence has indicated that the failure of repeated planting and recurrent miscarriages are linked to an increase in microbial diversity and a decrease in lactobacilli in the female reproductive system. 48 A possible mechanism is the impact of uterine endometrial lactobacilli, Gardnerella, and Mobiluncus abundance on the expression levels of integrin αvβ3 and LIF. 55
Although research on the endometrial microbiota has been flourishing in recent years, leading to a greater understanding of the microbial community in the female uterine cavity, there is still a need for further exploration into the potential causes of RIF in the field of assisted reproductive medicine. However, the assessment methods for endometrial microbiota still have certain limitations. For instance, it remains unclear how to standardize operations, reduce sample contamination, and lower the cost of examination methods, which are key areas for our future research.
Conclusions
Infertility represents a significant global health challenge, and despite marked improvements in embryo culture techniques within assisted reproductive technology, the rate of successful embryo implantation remains suboptimal. This review highlights that impaired ER is a central factor contributing to implantation failure. ERA expression profiling has proven valuable in identifying implantation windows for patients with RIF. The relationship between endometrial microbiota dysregulation and implantation failure, particularly in patients with PCOS, highlights the importance of comprehensive evaluation approaches. Combining immune regulation, molecular biomarkers (integrin αvβ3, HOXA10, and LIF), and microbiome interventions may improve implantation success rates. Given the complex nature of embryo implantation and current knowledge limitations, clinical practice requires consideration of individual factors and treatment protocols through a comprehensive scoring system for future research guidance.
Footnotes
Author contributions
JFL and XHF planned the paper. JFL, HGL, and CY drafted the manuscript. JL, SZY, and HX edited the draft. HX created the table. All authors discussed the manuscript and approved the final version of the article for publication.
Data availability statement
All data utilized in this review are sourced from publicly available literature, accessible through the PubMed and Web of Science databases. All references cited in the text have been included in the reference list. This study did not create any new datasets.
Conflicts of interest
The authors declare no conflicts of interest.
Institutional review board statement
Not applicable.
Funding
This research was funded by the Central Funds Guiding the Local Science and Technology Development, 2022, China (grant number Z20221343023).
