Endometrial cancer: from clinical reality to molecular treatment.

Inherited genetic mutations, such as those found in Lynch syndrome, significantly increase the risk of EC. Patients with this syndrome generally inherit a mutation in one of the genes involved in DNA mismatch repair ( MLH1 , MSH2 , MSH6 and PMS2 ) (Fig.  1 ), and these mutations predispose individuals to earlier onset of EC [ 12 – 14 ]. Fig. 1 Schematic representation of the DNA mismatch repair mechanism. Here, the ratios of the MSH2/MSH6 and MLH1/PMS2 complexes and their roles in DNA repair and the search for mismatched bases are presented Schematic representation of the DNA mismatch repair mechanism. Here, the ratios of the MSH2/MSH6 and MLH1/PMS2 complexes and their roles in DNA repair and the search for mismatched bases are presented MSH2 and MSH6 form the heterodimeric complex MutSα [ 15 ]. MSH2 leads to the stabilization of MSH6, which interacts with double-stranded DNA and scans for mismatched bases [ 16 ]. At the end of the scanning process, MutSα associates with MLH1 to repair damaged DNA [ 17 ]. MLH1 regulates protein‒protein interactions during the recognition of mismatches, followed by the discrimination of strands and their removal [ 18 ]. The formation of MutSα accommodates the creation of the MutLα heterodimer [ 19 ], which is formed of MLH1 and PMS2 [ 20 ]. PMS2 is involved in endonuclease activity; it is the active site of MutLα. This leads the complex to nick into the discontinuous DNA strand [ 21 ]. MLH1 interacts with PMS2 at residues 506–756 [ 22 ]. The interaction between MutSα and MutLα occurs upon mismatch binding. When MutSα reaches the mismatch recognition stage, MutLα has weaker ATPase activity. The mismatched DNA strand is then degraded at the end of the process by EXO1 [ 23 ]. Cowden syndrome is caused by inherited mutations in the PTEN tumor suppressor gene. The product of this gene is a protein that antagonizes the PI3K/AKT/mTOR signaling pathway, resulting in reduced cell proliferation and survival. In this syndrome, loss of function of this gene leads to increased cell proliferation and survival, promoting tumorigenesis. The appearance of numerous benign hamartomas is characteristic of this disease, and their transformation into malignant tumors is often observed in the endometrium, leading to a 5–26% increase in the risk of developing EC [ 14 , 24 – 26 ]. Polymerase proofreading-associated polyposis (PPAP) is caused by germline mutations in the exonuclease domain of the catalytic subunit of the DNA polymerase epsilon and delta 1, which are encoded by the POLE and POLD1 genes, respectively. The exonuclease domain of these enzymes is responsible for detecting and correcting nucleotide errors to ensure the fidelity of DNA replication. The increased risk of EC in PPAP patients is due to a significant increase in the somatic mutation rate caused by dysfunctional corrective activity in pathogenic variants of the exonuclease domain of the POLE and POLD1 genes [ 20 , 24 – 26 ]. Most endometrial carcinomas (ECs) arise from sporadic somatic mutations rather than from hereditary causes, which account for only 5–10% of cases. Although the specific mutations involved can vary between tumors, certain genes, such as those mentioned above, are recurrently altered across many EC cases, contributing to the molecular heterogeneity observed. This variability is reflected in the wide range of clinical presentations and histopathological subtypes that will be addressed later in this review[ 27 , 28 ]. Estrogens are a group of steroids that act primarily as female sex hormones. They are mostly produced by the developing follicles of the ovaries or by the placenta. Some estrogens are also produced in small quantities by other tissues, such as the liver, adrenal glands, breasts and adipose tissue. The three types of estrogens are estradiol, estriol and estrone. Estradiol is produced from testosterone, and estrone is produced from androstenedione. The main estrogen receptors are intracellular proteins encoded by two distinct genes in animals that possess both types of receptors: estrogen receptor alpha ( ERα ) and estrogen receptor beta ( ERβ ). Two types of effects can occur upon binding to the receptor: genomic (mainly gene expression) and nongenomic (protein) [ 29 ]. Progesterone (P4) is synthesized from cholesterol by luteinizing hormone (LH). This hormone can then be converted into androstenedione, testosterone or estradiol by granulosa cells surrounding the ovarian follicle. The concentration of progesterone varies during the menstrual cycle. Progesterone is secreted in greater quantities from ovulation onward by the granulosa cells of the corpus luteum. It helps to maintain and thicken the uterine mucosa, promotes endometrial vascularization, and promotes the formation of the uterine glands responsible for the serrated appearance of the uterine wall. If fertilization does not occur, progesterone levels return to normal, and the cells of the corpus luteum in which it is produced immediately decrease and disappear [ 29 ]. The vast majority of lesions in EC (approximately 80%) are associated with hyperestrogenism, which is characteristic of endometrioid-type neoplasms. Conversely, estrogen-independent neoplasms are generally associated with poor differentiation and are not associated with previous hyperplasia or dysplasia [ 29 ]. P4 can also participate in EC growth; unlike E2, exposure to P4 is a “protective” factor. This is caused by the transcription of FOXO1 and the upregulation of IGFBP-1 and its targets, which has antiproliferative effects on EC cells. Thus, P4 has been chosen as a primary treatment for premenopausal patients with EC, with a response rate of 60%. The expression of PRs in the glands of the endometrium is induced by E2 or progesterone. P4 also regulates the TGF-β pathway by decreasing SMAD and E-cadherin levels and increasing vimentin levels, leading to epithelial‒mesenchymal transition (EMT), which greatly impacts cancer cell viability. Therefore, PR expression in tumors is a good prognostic indicator because the receptor is required to respond to P4 treatment. Restoring these receptors and sensitizing tumors to progestin therapy could be potential therapeutic options [ 30 ]. As mentioned previously, prolonged unopposed estrogen exposure is a primary risk factor for EC, i.e., a hormonal imbalance exposing the endometrium to more estrogen than progesterone [ 31 ]. Several circumstances can lead to such an imbalance and thus increase the risk of developing EC. These include early menarche and/or late menopause, which extends the period of estrogen secretion by the ovaries; the absence of pregnancy; infertility; hormone replacement therapy for menopause consisting solely of estrogen; and tamoxifen, which acts as an antiestrogen in the breast tissue but as an estrogen in the uterus [ 29 , 31 ]. Importantly, the benefits of tamoxifen as a treatment for and prevention of breast cancer outweigh the risks associated with EC [ 31 ]. Prolonged unopposed exposure of the endometrium to estrogen can lead to the development of endometrial hyperplasia, i.e., excessive proliferation of endometrial glandular cells [ 31 – 33 ]. There are different types of endometrial hyperplasia: “simple” or “complex” and “with” or “without atypia”. Although this condition is mainly benign, the presence of atypia, i.e., abnormal cells, increases the risk of progression to EC in the absence of treatment [ 31 , 33 ]. Other conditions, such as endocrine disruptors, can cause estrogen/progesterone imbalance. According to the WHO, an endocrine disruption chemical is “ an exogenous substance or mixture that alters the function(s) of the endocrine system and consequently causes adverse effects in an intact organism, or its progeny, or (sub)populations ”. They can be found in persistent contaminants, pesticides, and industrial or natural substances; they can also be found in the environment or in nutrient intake environments. They mostly have estrogenic activities and can be considered oncogenic factors [ 34 ]. As a result, inappropriate responses to these activities lead to breast, ovarian and uterine cancers. Endocrine disruptors usually have delayed and heterogeneous effects, leading to various carcinogenic mechanisms. Thus, determining the dangerous dosage is challenging since it is impossible to define each patient’s exposure during her entire life and obtain epidemiological evidence of a carcinogenic effect. The possible mechanisms are oxidative stress, hormone-sensitive action, and proinflammatory mechanisms. These compounds mostly alter epigenetic processes such as methylation to promote or reduce gene expression, which induces a phenotype, rather than inducing mutations directly in the DNA sequence. Another point to consider is the “cocktail effect”; individuals are usually exposed to a vast array of endocrine disruptors at the same time. Currently, their interactions are unpredictable. It is necessary to implement measures to limit exposure during the most vulnerable periods of life, i.e., the embryonic period and puberty [ 34 ]. Overall, the individual effects of these molecules are known, but their combined impacts on ECs are still unclear and need to be further studied. For example, bisphenol A (BPA) (Fig.  2 ) is a completely synthetic substance. It is produced by mixing two phenols and acetone. BPA is similar to E2 and is considered an ERα and ERβ agonist. Low doses of BPA for a long period alter the estrogenic cycle; these effects are comparable to those of estrogenic therapies. It favors epithelial mesenchymal transition (EMT) and COX2 production, leading to invasion and metastasis. The correlation between BPA levels in the environment and tumor development in humans is now well established. [ 33 , 34 ]. Fig. 2 Schematic representation of the impact of BPA. It affects the estrogen cycle (E2 and P4 agonists), promotes EMT and COX2 production, deregulates miRNAs and activates ERK1/2 phosphorylation Schematic representation of the impact of BPA. It affects the estrogen cycle (E2 and P4 agonists), promotes EMT and COX2 production, deregulates miRNAs and activates ERK1/2 phosphorylation miRNAs regulate estrogen signaling by modulating ERα expression. Several miRNAs, such as miR-18a, miR-22, miR-206, miR-221, and miR-222, can directly bind to the 3’UTR of ERα mRNA, reducing its stability and translation. Dysregulation of these miRNAs by treatment with tamoxifen or exposure to endocrine-disrupting chemicals can alter estrogen-responsive gene expression, potentially contributing to endometrial carcinogenesis [ 35 – 37 ]. Polycystic ovary syndrome (PCOS) is another major risk factor for EC. This syndrome is characterized by excessive androgen production (hyperandrogenism), irregular or absent menses and an accumulation of immature follicles (polycystic ovaries) [ 29 , 31 , 38 , 39 ]. Owing to hyperandrogenism, PCOS is one of the most common causes of anovulation and infertility in patients [ 34 , 39 – 42 ]. As a result, the chronic anovulation observed in these patients is central to prolonged unopposed estrogen exposure of the endometrium, which is directly associated with the risk of endometrial hyperplasia and EC [ 34 , 39 , 40 ]. Other common female hormonal diseases are uterine fibroids and endometriosis. Uterine fibroids are outgrowths inside the uterus and benign tumors growing around the myometrium, whereas endometriosis is the migration of the endometrium outside the uterus. These lesions can adhere to other organs, and their activation depends on many factors, including estrogen. Like cancer cells, endometriotic cells can invade and metastasize. These two diseases are now associated with an increased risk of gynecological cancer development [ 42 ] . Although it is not linked with EC, endometriosis has been shown to increase the risk of developing ovarian endometriomas, with a particular predominance of ovarian clear cell carcinoma and ovarian endometrioid carcinoma [ 43 , 44 ]. Metabolic factors and lifestyle habits, such as obesity, type 2 diabetes, a poor diet and a sedentary lifestyle, are directly linked to the development of EC. This is due mainly to the considerable levels of estrogen produced by adipose tissue, particularly after menopause, when the ovaries stop producing it [ 33 , 45 – 47 ]. Thus, as a person's body fat increases, so do estrogen levels; weighing more than 200 lbs increases the risk of developing EC by approximately sevenfold [ 33 , 45 , 47 ]. Indirectly, a sedentary lifestyle and a poor diet are risk factors for EC, since these factors promote the development of obesity [ 33 , 47 , 48 ]. Type 2 diabetes is a chronic disease characterized mainly by high levels of glucose (hyperglycemia), insulin (hyperinsulinemia) and insulin-like growth factor 1 (IGF-1) [ 46 , 47 ]. First, excessive blood glucose provides a sugar-rich environment that can contribute to the growth and invasiveness of cancer cells [ 49 , 50 ]. Furthermore, increased insulin and IGF-1 levels promote overactivation of the PI3K-AKT-mTOR and MAPK signaling pathways, which are frequently altered in EC cells. These pathways, discussed later in this review, play a central role in tumor development since they are involved in regulating mechanisms such as cell proliferation, growth and survival [ 51 – 56 ]. EC has had numerous classifications since its discovery. It can be classified into two main categories: Type I, also known as endometrioid, which is associated with excess estrogen, is often diagnosed at an early stage and has a favorable prognosis, and Type II, nonendometrioid, which is more aggressive, is not associated with excess estrogen, includes serous and clear-cell carcinomas, and has a less favorable prognosis [ 53 ]. We have produced tables to summarize the existing classifications: the FIGO staging system, the histological classification by the WHO, and the molecular classification. FIGO staging system (updated in 2023) [ 54 ] Stage 1: Cancer confined to the body of the uterus and ovary. 1A: Confined to the endometrium or nonaggressive histological type (low-grade endometrioid type with invasion of less than half the myometrium, without invasion of the lymphovascular space or localized invasion (LVSI)) or good prognosis. 1A1 : Nonaggressive histological type limited to an endometrial polyp or confined to the endometrium. 1A2 : Nonaggressive histological type involving less than 50% of the myometrium with no or focal lymphovascular space invasion (LVSI) as defined by the WHO criteria. 1A3: Low-grade endometrioid carcinoma confined to the uterus and ovary. 1B : Nonaggressive histological types with invasion of half or more of the myometrium and negative or LVSI. 1C: Aggressive histological types limited to a polyp or confined to the endometrium. 1A: Confined to the endometrium or nonaggressive histological type (low-grade endometrioid type with invasion of less than half the myometrium, without invasion of the lymphovascular space or localized invasion (LVSI)) or good prognosis. 1A1 : Nonaggressive histological type limited to an endometrial polyp or confined to the endometrium. 1A2 : Nonaggressive histological type involving less than 50% of the myometrium with no or focal lymphovascular space invasion (LVSI) as defined by the WHO criteria. 1A3: Low-grade endometrioid carcinoma confined to the uterus and ovary. 1A1 : Nonaggressive histological type limited to an endometrial polyp or confined to the endometrium. 1A2 : Nonaggressive histological type involving less than 50% of the myometrium with no or focal lymphovascular space invasion (LVSI) as defined by the WHO criteria. 1A3: Low-grade endometrioid carcinoma confined to the uterus and ovary. 1B : Nonaggressive histological types with invasion of half or more of the myometrium and negative or LVSI. 1C: Aggressive histological types limited to a polyp or confined to the endometrium. Stage 2: Invasion of the cervical stroma without ectopic expansion or with significant LVSI or aggressive histological types with invasion of the myometrium. 2A: Invasion of the cervical stroma by nonaggressive histological type. 2B: Significant LVSI according to nonaggressive histological type. 2C: Aggressive histological types with any myometrial invasion. 2A: Invasion of the cervical stroma by nonaggressive histological type. 2B: Significant LVSI according to nonaggressive histological type. 2C: Aggressive histological types with any myometrial invasion. Stage 3: Local and/or regional spread of a tumor of any histological subtype. 3A: Invasion of the uterine serosa, uterine adnexa (ovaries/fallopian tubes) or both by direct spread or metastasis. 3A1: Spread to ovaries or fallopian tubes (unless stage 1 A criteria are met). 3A2: Invasion of the uterine submucosa or spread through the uterine serosa. 3B: Direct or metastatic spread to the vagina and/or parametrium (connective tissue surrounding the uterus) or pelvic peritoneum. 3B1 : Direct or metastatic spread to the vagina and/or parametrium. 3B2: Metastases to the pelvic peritoneum. 3C: Metastases to the pelvic and/or para-aortic lymph nodes. 3C1: Metastases to pelvic lymph nodes. 3C1i : Micrometastasis 3C1ii: Macrometastases 3C2: Metastases to para-aortic lymph nodes to renal vessels, with or without metastases to pelvic lymph nodes 3C2i : Micrometastasis 3C2ii: Macrometastases 3A: Invasion of the uterine serosa, uterine adnexa (ovaries/fallopian tubes) or both by direct spread or metastasis. 3A1: Spread to ovaries or fallopian tubes (unless stage 1 A criteria are met). 3A2: Invasion of the uterine submucosa or spread through the uterine serosa. 3A1: Spread to ovaries or fallopian tubes (unless stage 1 A criteria are met). 3A2: Invasion of the uterine submucosa or spread through the uterine serosa. 3B: Direct or metastatic spread to the vagina and/or parametrium (connective tissue surrounding the uterus) or pelvic peritoneum. 3B1 : Direct or metastatic spread to the vagina and/or parametrium. 3B2: Metastases to the pelvic peritoneum. 3B1 : Direct or metastatic spread to the vagina and/or parametrium. 3B2: Metastases to the pelvic peritoneum. 3C: Metastases to the pelvic and/or para-aortic lymph nodes. 3C1: Metastases to pelvic lymph nodes. 3C1i : Micrometastasis 3C1ii: Macrometastases 3C2: Metastases to para-aortic lymph nodes to renal vessels, with or without metastases to pelvic lymph nodes 3C2i : Micrometastasis 3C2ii: Macrometastases 3C1: Metastases to pelvic lymph nodes. 3C1i : Micrometastasis 3C1ii: Macrometastases 3C1i : Micrometastasis 3C1ii: Macrometastases 3C2: Metastases to para-aortic lymph nodes to renal vessels, with or without metastases to pelvic lymph nodes 3C2i : Micrometastasis 3C2ii: Macrometastases 3C2i : Micrometastasis 3C2ii: Macrometastases Stage 4: Spread to the bladder mucosa and/or intestinal mucosa and/or distant metastases. 4A: Invasion of the bladder mucosa and/or intestinal mucosa. 4B: Abdominal peritoneal metastases beyond the pelvis. 4C: Distant metastases, including metastases to intra- or extra-abdominal lymph nodes located above the renal vessels, lungs, liver, brain or bones. 4A: Invasion of the bladder mucosa and/or intestinal mucosa. 4B: Abdominal peritoneal metastases beyond the pelvis. 4C: Distant metastases, including metastases to intra- or extra-abdominal lymph nodes located above the renal vessels, lungs, liver, brain or bones. Classification by morphology, immunohistochemistry (IHC) and from the classification of The Cancer Genome Atlas program (TCGA). Endometrioid: Most frequent type (80%), variable architecture (glandular, papillary or solid), composed of cells with endometrioid differentiation and usually with atypical hyperplasia. Recognition of TCGA molecular classification to enable better prediction of clinical outcome than later classification on the basis of morphology, stage and presence or absence of LVSI. The second most common type (approximately 10% of cases), high-grade carcinoma (aggressive tumors = 40% of EC deaths), is characterized by nuclear pleomorphism and papillary and/or glandular architecture (sometimes solid), usually with atrophy and the potential to metastasize via the tubules. Almost all samples were classified as TP53 abnormal (copy high-number) or had aberrant p53 expression on IHC with nuclear overexpression or complete absence of staining. The exact frequency of clear cells is unknown but possibly accounts for approximately 2% of ECs (diagnostic criteria imprecise). These cells are composed of pleomorphic cells that may be polygonal, cuboid, flat or nail shaped with clear or eosinophilic cytoplasm (affinity for eosin), papillary, tubulocystic and/or solid architecture. The vague definition leads to overlap with the endometrioid and serous types. The highly heterogeneous molecular profile consists of a mix of TCGA subtypes, but TP53 is generally WT and negative for estrogen receptor and progesterone receptor expression. Undifferentiated and dedifferentiated: Aggressive carcinomas with no specific linear differentiation and a solid sheet-like architecture. The sudden appearance of keratinization foci is possible and is often accompanied by necrosis and high mitotic activity. In 40% of cases, tumors are composed of a mixture of endometrioids of stage 1 or 2 but rarely stage 3 or serous type. Molecular profiling revealed that in 50–75% of cases, tumors are MMR deficient, and a large proportion contain inactivating mutations in core genes of the SWI/SNF complex ( SMARCA4, SMARCB1, ARID1A, ARID1B ). Mixed: Represents less than 10% of ECs, composed of two distinct types morphologically and by proteins (IHC), at least one of which is either a serous or a clear cell. Dedifferentiated and carcinosarcoma types, which are mixed by definition, are excluded from the diagnosis. Tumor behavior is generally dictated by the worst histotype. Mesonephric adenocarcinoma: Very rarely derived from mesonephric remnants (embryonic-stage renal ducts), tumors are generally composed of small tubules with dense eosinophilic colloid material and diverse architectures (papillary, ductal, retiform, solid, fusiform). Molecular profile of WT TP53 and p53, which are negative for ER and PR. Mesonephric adenocarcinoma type: Very rare type with an absence of mesonephric remnants; otherwise, morphologically, proteins were detected via immunohistochemistry (IHC) (GATA3 +, ER-, TTF1 +, CD10 +, p53 WT) and molecularly very similar to mesonephric adenocarcinoma. Epidermoid: Very rare type in the body of the uterus with a poor prognosis, originating from squamous cells (epithelial cell type). It is associated with risk factors such as chronic inflammatory conditions, previous irradiation and HPV infection. Morphologically identical to squamous cell carcinomas seen elsewhere on the body and may present with broad invasive fronts. Mucinous gastrointestinal type: Aggressive tumors that are most common in the cervix and have a glandular architecture with mucin-secreting epithelium and may include caliciform cells (mucus synthesis). Negative molecular profiles for ER and PR differentiated them from the endometrioid type with mucinous differentiation. Carcinosarcoma: Aggressive tumors with LVSI and recategorization as endometrial carcinoma, since sarcomatous elements (malignant cells originating from connective tissue) have been shown to be transdifferentiated from carcinoma during tumor evolution. Molecular profiling revealed that 78% or more are TP53 abnormal, 22–38% are TP53 WT, and fewer than 5% are POLE -mutated or MMR-deficient. Neuroendocrine carcinoma (poorly differentiated = NEC) is very uncommon (> 1% of NECs) and has a poor prognosis. Two distinct morphological variants have been identified: small-cell and large-cell variants. Frequent distant metastases, nuclear pleomorphism, a leaf-shaped architecture and high mitotic activity (highly invasive) are common. Neuroendocrine tumors (well-differentiated NETs), which are very uncommon and have stages I and II disease, are associated with a poor prognosis. Composed of small, uniform cells with nesting, trabecular or serpentine architecture. In general, we have summarized the general information in Table  1 . Table 1 Summary of WHO histological classifications Stage Characterization Low grade Endometrioid (depending on proportion of solid zone) Stage I:  50% solid zone) Serous Clear-Cell Mesonephric like Mixed Mucinous gastrointestinal type Undifferentiated/differentiated Squamous Carcinosacorma Neuroendocrine (NEC) Summary of WHO histological classifications Over time, similarities between endometrioid and serous tumors have highlighted the importance of genomic-based classification to manage the treatment of patients [ 7 ]. Since the identification of different subtypes can be difficult, TCGA, supported by the ProMisE (Proactive Molecular Risk Classifier for EC) project, distinguished in 2017 four different types of EC on a molecular basis, with different prognoses. These 4 groups are summarized in Table  2 . In summary, the current classification is applied clinically to 3 parameters: p53 status, MSI, and mutated POLE . These subtypes are analogous to those used in TCGA: mismatch repair defect mismatch repair (MMR), which involves one or more gaps in repair proteins (corresponding to the hypermutated group); mutated POLE (DNA polymerase epsilon), which has a mutation in the domain of the exonuclease 9–14 domain (exons); abnormally mutated p53 protein, which corresponds to the high-copy-number subtype; and finally, the TP53 wild-type, which corresponds to the low-copy-number subtype [ 63 ]. Table 2 Summary of molecular and physiopathological classifications Classification Hypermutated (MSI) [ 10 ] POLE mutated [ 7 ] p53 mutated [ 58 – 61 ] p53 wild-type [ 62 ] Classical features Mismatch repair defect (MMR-D), one or more gaps in DNA repair proteins DNA polymerase epsilon with mutation in exonuclease domain 9–14 (exons) Corresponds to the high number of copies subtype Corresponds to the low number of copies subtype Patient prognosis Good prognosis Excellent prognosis. Very rarely relapses (if it occurs, very aggressive) Worst rate and prognosis Intermediate prognosis Prognosis at 5 years [ 57 ] Approximately 75% Approximately 90% Approximately 30% Approximately 85% Molecular characteristics of the patient Unstable microsatellites Tumor heterogeneity Mutations in the p53 gene present in endometrial adenocarcinomas Low copy rate. Low mutation rate Important mutations in MMR proteins: MSH2, MSH6, MLH1, PMS1, PMS2 Preponderance of transitions between C and A. Amino acid substitution Similar to serous uterine tumors, Low KRAS alteration, low PI3K alteration Hypermethylation of the MLH1 promoter is very common MSS Basal breast cancers or HGSOCs p53 WT Most frequently associated histotypes Endometrioid high grade. Undifferentiated and dedifferentiated Endometrioid high grade in younger, nonobese women Endometrioid low grade, serous cases, clear cell or mesonephric Serous, carcino-sarcoma, endometrioid, or mixt Summary of molecular and physiopathological classifications Classical features Corresponds to the low number of copies subtype Patient prognosis Molecular characteristics of the patient Tumor heterogeneity Preponderance of transitions between C and A. Amino acid substitution Most frequently associated histotypes Endometrioid high grade. Undifferentiated and dedifferentiated MSI-high ECs have important mutations in the MMR proteins MSH2, MSH6, MLH1, PMS1 and PMS2. Associations between MMR status and adverse clinical effects: Hypermethylation of the MLH1 promoter is very common in MSI patients, with a mutation rate of 18*10^6 mutations/Mb. Reijnen et al. demonstrated that this mutation status is a predictive marker for radiotherapy, which was confirmed through the ProMise study (stage IB-II) [ 10 , 64 ]. EC is also comparable to colorectal carcinoma because of its MSI and POLE classification, which is associated with extremely high mutation rates [ 58 ]. POLE mutants have an excellent prognosis [ 65 ] and are high-grade endometrioid type, with tumor heterogeneity and invasion of the lymphovascular space. They have other features, such as a preponderance of transitions between nucleotides C and A, amino acid substitutions and microsatellite stability groups [ 59 ], with mutation rates that can exceed 232*10^6 mutations/Mb [ 10 ]. In addition, other studies have shown that ECs frequently have very high TIL levels and are often accompanied by a patient with a reaction similar to that of Crohn's disease [ 59 ]. In most cases, POLE mutants are MSS (microsatellite stable), but there are reports of cases with loss of MMR proteins [ 60 ]. POLE mutants rarely relapse. This type of mutation tends to occur in younger nonobese patients, and despite greater aggressiveness, relapse is rare [ 61 ]. Because POLE -mutated ECs have many T-cell responses intratumorally, they have excellent prognosis for these cancers [ 59 ]. In 1990, TP53 mutations were diagnosed in 90% of serous carcinomas, and loss of p53 led to the majority of serous carcinomas [ 10 ]. Grade 3 mutated p53 patients with endometrioid carcinoma relapsed more frequently than did those without the mutation [ 66 ]. The role of p53 in EC and hyperplasia has shown that mutations in the TP53 gene are present in endometrial adenocarcinomas but absent in hyperplasia. Currently, a greater proportion of patients with mutated p53 mutants have nonendometrioid histology (leaning toward serous) and have the worst survival rates [ 57 ]. On the other hand, ECs, with frequent p53 mutations, resemble serous uterine tumors, basal breast cancers and high-grade serous ovarian carcinomas (HGSOCs). TP53 is mutated in 96% of HGSOCs [ 58 ]. As has been demonstrated, EC can be compared mutationally with HGSOCs or breast carcinomas, despite a much higher mutation rate. However, high mutation rates of FBXW7 , PPP2R1A and ARIDA in serous carcinomas of the uterus are not found in basal breast cancer or HGSOCs [ 59 ]. This classification currently shows promising clinical results [ 61 ]. ProMisE can be applied to diagnostic samples and to choose surgical procedures or adjuvant therapies [ 7 ]. With that in mind and its potential, clinicians can reduce overtreatment and undertreatment, leading both to complications [ 58 ].

Currently, treatment for EC primarily involves surgery, particularly hysterectomy, followed by adjuvant radiotherapy and/or chemotherapy, depending on the stage and tumor characteristics. For advanced or recurrent cancers, targeted therapies—such as those targeting the PI3K–AKT–mTOR pathway—and immunotherapy are being explored, although their efficacy remains variable and patient dependent. While inhibitors of this pathway show promise, they face challenges such as managing side effects and overcoming tumor resistance. Research continues to focus on further personalizing treatments, improving therapeutic options, and reducing inequalities in access to care. In summary, while significant advances have been made in the management of EC, there is still a need for new therapeutic strategies and more personalized approaches to optimize outcomes.

EC can cause a variety of signs and symptoms as the disease progresses. Other medical conditions, such as uterine fibroids and PCOS, can cause similar symptoms, making diagnosis challenging. The most common symptom is abnormal vaginal bleeding. This includes menstrual changes (heavier, longer or more frequent periods), bleeding between periods, postmenopausal bleeding and light vaginal bleeding. Other signs and symptoms include unusual vaginal discharge, which may be foul-smelling, pus-like or tinged with blood, pain during intercourse, pain or pressure in the pelvis, lower abdomen, back or legs, pain during urination, difficulty urinating or blood in the urine, pain during defecation, difficulty defecating or blood in the stool, bleeding from the bladder or rectum, accumulation of fluid in the abdomen (called ascites) or legs (called lymphedema), weight loss, loss of appetite and difficulty breathing [ 67 , 68 ]. First, the patient is offered a pelvic exam, where a professional observes the external appearance of the vulva and vaginal opening for signs such as redness, discharge, masses (such as cysts or genital warts) and other abnormal conditions. After this step, a speculum is inserted into the vagina to widen it to visualize the cervix and the wall of the vagina (searching for masses, inflammation and abnormal discharge). A Pap test may also be performed. Finally, for bimanual examination, healthcare professionals place one or two gloved fingers inside the vagina, place the other hand on the lower part of the abdomen, and apply little pressure. This allows them to evaluate the texture, size and shape of the uterus and ovaries. It also allows the detection of any sensitive areas or masses. If uncertain, a professional can perform a rectovaginal examination. After these first exams, the ultrasound probe or transducer is inserted into the patient’s vagina (pelvi-genital tract) to produce images of pelvic structures and organs, including the vagina, uterus, ovaries, fallopian tubes and bladder. Hysteroscopy is a procedure involving the use of an endoscope to examine or treat the uterus and fallopian tubes. The cells or tissue can be removed for microscopic examination. Doctors may also use hysteroscopy to remove polyps, uterine fibroids or tumors. If a biopsy is performed, the biological sample (tissue or tumor removed for examination) is analyzed to diagnose cancer or to determine whether an abnormality is cancerous. Typically, a sample is taken from the endometrium. As a less invasive method, blood samples can be taken. They establish the stage of uterine cancer and perform the following biochemical blood tests: BUN and creatinine levels can also be measured to assess kidney function. Higher-than-normal levels could indicate that the cancer has spread to the ureters or kidneys. The levels of alanine aminotransferase (ALT), aspartate transaminase (AST) and alkaline phosphatase can be measured to assess liver function. Higher-than-normal levels could indicate that the cancer has spread to the liver. Tumor antigen 125 (CA 125) can be measured. A higher-than-normal level could indicate advanced or metastatic uterine cancer. Finally, magnetic resonance imaging (MRI) uses magnetic forces and radio waves to produce cross-sectional images of the body's organs, tissues, bones and blood vessels. A computer assembles the images into 3-dimensional snapshots. MRI is used to determine the extent of cancer invasion in the muscular layer of the uterine wall (called the myometrium). It can also help doctors determine whether the cancer has spread to other organs or reappeared after treatment [ 68 – 70 ] . Therapeutic options are offered depending on the classifications previously presented. In fact, patients with type 1 cancer with a FIGO grade of 1 A G1/G2 are offered a hysterectomy, adnexectomy or salpingectomy. If the patient is FIGO 1B but still type 1, in addition to the previous treatment, the patient can undergo lymphonodectomy and brachytherapy. When there is a high or intermediate prognosis, i.e., type I FIGO IA group 3 or FIGO IA/B G1/G2 with lymphovascular invasion, clinicians can consider radiotherapy. Finally, for the other types, such as type I FIGO IB G3, FIGO superior at 1 or type II cancer, cytoreductive surgery and chemotherapy are considered [ 64 ]. Other therapies, such as immunotherapies and targeted therapies, will be presented further. The molecular classification is currently used to determine the treatment that will be used. In fact, patients with POLE are classified as having a good prognosis and are offered surgery with or without radiotherapy. Patients with type 1 or 2 MMR defects will be tested for surgery and immunotherapy with or without radiotherapy. A woman with high-grade endometrial cancer (EC) with either a p53 abnormality or wild-type (WT) p53 will undergo surgery and radiotherapy, with or without chemotherapy. The same approach applies to women who are at high risk of metastases. Finally, if all the markers are negative or at low levels, only surgery is performed [ 65 ]. While considering the classifications that have been previously named, it is possible to classify the survival rates in that manner [ 64 ]: Type I, FigoIA, G1/G2: 93.4% survival at 5 years. Type I, FIGO IB, G1/G2: 86.3% survival at 5 years. Type I FIGO IA G3 and type I FIGO IA/B G1/G2 with LVSI: 82% survival at 5 years. Type I, FIGO IB, G3, type II, and FIGO > I: < 74% survival at 5 years. Quality of life (QOL) after an EC is important for clinicians to consider in order to manage the healthcare professionals involved with the patient and for biologists to understand the impact of treatment on the patient's life. Many factors, such as socioeconomic, psychological, and physical factors, are involved. The effect of each factor on tumor grade also needs to be considered. Socioeconomic effects: The cost of any cancer therapy is high in some countries, and the time and cost of travel between home and hospital must be considered. Some studies have shown that low income and lower education are associated with poor QOL. Some patients may look for less expensive treatment options that are less efficient. Owing to the high cost, some patients may decide to discontinue their treatment [ 71 ]. Finally, while very few studies have demonstrated this in the case of EC, socioeconomic factors play a significant role, similar to the disparities observed in breast cancer outcomes, for example. The difficulty in accessing early screening emphasizes that low-income patients are less likely to survive cancer [ 72 , 73 ]. Physical effects: Individuals with obesity and EC seem to have a worse quality of life, with more fatigue, urological symptoms, and pain in the back and pelvic regions [ 74 ]. Moreover, the most important symptom that appears is menopausal symptoms; even if EC generally occurs after menopause, patients who undergo a premenopausal hysterectomy have difficulties with menopausal symptoms afterwards [ 75 ]. Chemoradiation has been reported to cause lower physical and quality of life scores [ 71 ]. Psychological effects: Many studies have shown that patients with EC have depressive and anxiety symptoms, which are linked to lower overall survival, even in some patients 4.5 years after the end of treatment and those with early-stage EC [ 71 ]. If we look at psychological and physical effects, in terms of sexuality, patients with an EC had less sexual interest than did the general population. After surgery, 55.9% never returned to being sexually active, not because of physical sexual dysfunction but because of a lack of interest [ 74 ]. It is essential to consider these studies and provide the necessary support for these individuals. Psychosocial education should be an integral part of EC treatment, as the emotional impact of the diagnosis is significant. Furthermore, discussions about life after EC show that most women like psychological support as well as lifestyle coaching. In fact, they appreciate services such as nutritional guidance to prevent overweight or obesity, advice on the increased risk of cardiovascular disease, and support with physical activity to maintain good health [ 76 – 81 ]. There are many different types of therapies that are used to treat EC depending on the stage, type of mutation and genetic characteristics, as listed previously. Surgery and radiation therapy are often used in the early stages of cancer. Chemotherapy, hormonotherapy, targeted therapy and immunotherapy are treatment options that depend on the condition of the patient [ 74 , 82 ]. Importantly, younger patients usually want to have children in the future. Unfortunately, chemotherapies are highly harmful to the body. For this reason, other therapies can be prioritized. Hormones can be administered to patients with early-stage, low-grade EC. It can be given to patients with recurrent, low-grade and slow-grade cancer. They can be used as monotherapies or in combination with other treatments. Patients are often given progestins such as megestrol (progestogenic activity, antigonadotropic effects) and medroxyprogesterone (progestin). Other options include aromatase inhibitors, such as letrozole (antiestrogen, which prevents aromatase from producing estrogen); estrogen receptor modulators, such as tamoxifen (selective, which requires estrogen receptors on the surface of cancer cells, competitive inhibition); and gonadotropin-releasing hormone agonists, such as goserelin (which stimulates the production of sex hormones and disrupts the feedback system, ultimately resulting in downregulation) [ 83 , 84 ]. Chemotherapy is an important treatment option. There are platinum (which causes crosslinking in DNA), anthracycline (which intercalates between the nitrogenous base pairs of DNA and inhibits the activity of topoisomerase II) and taxane-based chemotherapy drugs (inhibits microtubule function). The most common chemotherapy drugs for EC include cisplatin, carboplatin, paclitaxel (mitotic spindle poison, which blocks the mechanism of mitosis) and doxorubicin (anthracycline, which intercalates between two DNA base pairs, preferably between two G‒C bases). They can be used alone or in combination [ 85 ]. Unfortunately, chemotherapies are not precise and can also target healthy cells that die because of their generic mechanisms of action. As mentioned above, immunotherapy can be used alone or in combination with chemotherapy [ 86 ]. The combination of immunotherapy drugs with other treatments has been shown to be beneficial for patients with MMR or MSI-H subtypes. Ongoing studies regarding the outcome of this type of combined therapy are in progress for those with advanced as well as metastatic EC [ 86 ]. For example, Jemperli , also known as Dostarlimab ((PD-1)–blocking monoclonal antibody), is an immunotherapy drug that was recently approved (2021–2023). It can be used alone or in combination with carboplatin or paclitaxel. Dostarlimab is indicated for patients with recurrent mismatch repair deficiency and high microsatellite instability tumors. Clinical trials have shown that patients with primary advanced or recurrent EC with MMRD/MSI-H that has been treated with chemotherapy and Dostarlimab have increased progression-free survival [ 87 ]. Another drug that targets immune checkpoint inhibitors in patients with advanced EC is Keytruda , also known as pembrolizumab (another PD-1-blocking monoclonal antibody). This treatment is usually given after another type of therapy for patients with MMR, MSI-H or high mutational burden (TMB-H) tumors. It can also be used with targeted therapies, such as lenvatinib (brand name: Lenvima , kinase inhibitor against VEGFR1/2/3), for MMR-proficient patients [ 87 ]. Unfortunately, as is the case in many instances, the beneficial effects are largely attributable to the overall condition of the patients. A recent study involving nearly 5,200 patients demonstrated that Dostarlimab improved survival outcomes, whereas pembrolizumab had a more moderate response rate. On the other hand, the two studies were similar in terms of induced deaths and treatment discontinuations [ 88 ]. Finally, targeted therapy drugs such as lenvatinib and larotrectinib (brand name: Vitrakvi , an inhibitor of tropomyosin kinase TrkA/B/C), which are kinase inhibitors, and everolimus, an mTOR inhibitor, are interesting future prospects for EC treatment. Many clinical trials are ongoing regarding the combination of targeted therapy with other types of treatments. For example, in patients with recurrent EC, the combination of everolimus and letrozole has been studied during a phase II trial by the GOG Foundation Study and has shown beneficial results. Combinations of targeted therapy with immunotherapy are also possible. For example, pembrolizumab can be combined with lenvatinib for patients whose previous treatments did not work and for those with non-MSI-H, non-MMR-d and advanced EC [ 89 , 90 ]. Targeted therapies are promising new treatments, but several challenges are emerging: tumor heterogeneity, resistance that develops as treatment progresses, and the limitations of biomarkers (for example, different AKT isoforms, which may not all respond to the same inhibitors). Additionally, toxicity profiles, similar to those seen with other therapies, can also arise [ 91 – 96 ]. In summary, the names of the studies and each target are as follows: (Tables 3 , 4 , 5 , 6 , 7 ). Table 3 Clinical studies of patients with EC receiving immunotherapy Target Names of the studies Scientific reasoning PD-1 RUBY (part 1 and 2), NRG-GY018, KEYNOTE-158, GARENT, PODIUM-101 PD-1 is a protein located on T lymphocytes, whose role is to negatively regulate the immune system: when a PDL1 protein binds to the PD1 receptor, the T lymphocyte is deactivated Some tumor cells possess PDL1 ligand. They bind to PD1 receptors on T lymphocytes, preventing them from activating and destroying tumor cells. They can therefore continue to grow PD-L1 AtTEND, DUO-E, MITO-END3 Cancer cells are thus no longer recognized by the immune system and can continue to grow. Immunotherapy prevents the binding between PD1 and PD-L1, thus stimulating the immune system to work on these cells [ 89 ] Table 4 Clinical study of ECs with no specific profile Target Names of the studies Scientific reasoning Progestogen GOG81, GOG119, GOG153 Progesterone key hormone in counter balancing estrogen in EC [ 28 ] Aromatase and mtTOR NCT01068249 , NCT02730923 /VICTORIA The PI3K pathway is generally dysregulated in EC on the first hand. On the other hand, hormonal manipulation leads to a patient's answer but there is resistance coming from PI3K pathway dysregulated, targeting mTOR + hormones may overcome this resistance [ 97 ] Aromatase and CDK4/6 NGSO-PALEO/ENGOT-EN3NCT02657928, NCT03675893 ER + EC are generally altered in the PI3K pathway, these cancers are frequently with a high ER transcriptional activity and upregulation of cdk4 and 6. They are key regulators in hormonal therapy resistance [ 97 , 98 ] DKK-1 NCT03395080 DKK1 modulates Wnt signaling and gynecologic endometrioid tumors have a high prevalence of β-catenin mutations [ 98 ] PORCN NCT02521844 PORC is required for the posttranslational modification of Wnt [ 99 ] Table 5 Clinical study of ECs with growth factor receptors Target Names of the studies Scientific reasoning VEGFR GOGO209, EORTC, GOG229E, NCT00462826 , AMG386, GOG229J VEGF is one of the most important stimulators of angiogenesis in EC [ 100 ] FGFR KEYNOTE775, LEAP001, NCT00888173 , NCT01225887 , NCT01379534 , NCT01111461 , KEYNOTE 146 Mutations in the FGFR pathways have been identified in endometrium cancer [ 101 ] HER 2 NCT01367002 , NCCH1615/STATICE, DESTINY-PanTumor02, NCT042053630, NCT04235101 , NCT05150691 Prolonged survival when patient have HER-2 + in serous EC [ 102 ] Table 6 Clinical study of EC with signaling pathways Target Names of the studies Scientific reasoning mTOR NCT00739830 , AGO-GYN8, NCT02725268 The PI3K/AKT/mTOR pathway is the most altered pathway in EC [ 102 ] PI3K SAR245408 XL147 PI3K/mtor MAGGIE KRAS Pathway NCT01935934 , code break101 KRAS mutations associated with type I estrogen-related EC. KRAS mutations occur at the early stages of the EC pathway, KRAS mutations are present in 6–16% of endometrial hyperplasia [ 90 ] HRD UTOLA Target DNA repair in EC. Parp-inhibitor maintenance for advanced and metastatic EC [ 52 ] ARID1A ATARI, NCT05523440 ARID1A mutations have been reported in gynecological cancers, including clear cell carcinoma. Studies show that ARID1A mutant cells display sensitivity to ATR inhibition [ 103 – 105 ] PTEN ENDOLA PTEN has been identified as a marker for premalignant endometrial hyperplasia, target it can stop the switch between hyperplasia and cancer [ 106 ] Table 7 Characterization of the human cell lines most frequently used to model endometrial cancer Cell line Type of EC Age of patient Characterization Other Ishikawa Endometrial adenocarcinoma, Grade 1 39 years old Receptors ER and PR presents, Type I, Hypermutated (MSI) molecular group ER and PR disap- pear after long term culture and the cells transform themselves into undifferentiated cells 80% of the genome is from East Asian North KLE Endometrial adenocarcinoma, but cells are coming from the colon's metastasis Grade 3 Between 64 and 68 years old depending on sources Receptors ER and PR presents, Poorly differentiated, KRAS wild type, Low microsatellite instability, P53 WT molecular group Mostly type II due to mutation, age of the patient 60% of the genome is from North of Europe and 30% is from South Europe HEC-1-A Endometrial adenocarcinoma, Grade 2 71 years old Receptors ER and PR absents, Diploid, Heterozygous KRAS, Homozygous TP53, P53 WT molecular group Most likely, type II in function of the mutation, 91% of the genome is from East Asian North HEC-1-B Endometrial adenocarcinoma, Grade 2 71 years old Receptors ER and PR mostly absents (not in all cells), Tetraploid, Heterozygous KRAS, Homozygous TP53, P53 WT molecular group Child of the subtype HER-1-A, Most likely, type II in function of the mutations, 92% of the genome is from East Asian North Clinical studies of patients with EC receiving immunotherapy PD-1 is a protein located on T lymphocytes, whose role is to negatively regulate the immune system: when a PDL1 protein binds to the PD1 receptor, the T lymphocyte is deactivated Some tumor cells possess PDL1 ligand. They bind to PD1 receptors on T lymphocytes, preventing them from activating and destroying tumor cells. They can therefore continue to grow Cancer cells are thus no longer recognized by the immune system and can continue to grow. Immunotherapy prevents the binding between PD1 and PD-L1, thus stimulating the immune system to work on these cells [ 89 ] Clinical study of ECs with no specific profile Progesterone key hormone in counter balancing estrogen in EC [ 28 ] The PI3K pathway is generally dysregulated in EC on the first hand. On the other hand, hormonal manipulation leads to a patient's answer but there is resistance coming from PI3K pathway dysregulated, targeting mTOR + hormones may overcome this resistance [ 97 ] ER + EC are generally altered in the PI3K pathway, these cancers are frequently with a high ER transcriptional activity and upregulation of cdk4 and 6. They are key regulators in hormonal therapy resistance [ 97 , 98 ] DKK1 modulates Wnt signaling and gynecologic endometrioid tumors have a high prevalence of β-catenin mutations [ 98 ] PORC is required for the posttranslational modification of Wnt [ 99 ] Clinical study of ECs with growth factor receptors VEGF is one of the most important stimulators of angiogenesis in EC [ 100 ] Mutations in the FGFR pathways have been identified in endometrium cancer [ 101 ] Prolonged survival when patient have HER-2 + in serous EC [ 102 ] Clinical study of EC with signaling pathways The PI3K/AKT/mTOR pathway is the most altered pathway in EC [ 102 ] KRAS mutations associated with type I estrogen-related EC. KRAS mutations occur at the early stages of the EC pathway, KRAS mutations are present in 6–16% of endometrial hyperplasia [ 90 ] Target DNA repair in EC. Parp-inhibitor maintenance for advanced and metastatic EC [ 52 ] ARID1A mutations have been reported in gynecological cancers, including clear cell carcinoma. Studies show that ARID1A mutant cells display sensitivity to ATR inhibition [ 103 – 105 ] PTEN has been identified as a marker for premalignant endometrial hyperplasia, target it can stop the switch between hyperplasia and cancer [ 106 ] Characterization of the human cell lines most frequently used to model endometrial cancer Endometrial adenocarcinoma, Grade 1 Receptors ER and PR presents, Type I, Hypermutated (MSI) molecular group ER and PR disap- pear after long term culture and the cells transform themselves into undifferentiated cells 80% of the genome is from East Asian North Endometrial adenocarcinoma, but cells are coming from the colon's metastasis Grade 3 Receptors ER and PR presents, Poorly differentiated, KRAS wild type, Low microsatellite instability, P53 WT molecular group Mostly type II due to mutation, age of the patient 60% of the genome is from North of Europe and 30% is from South Europe Endometrial adenocarcinoma, Grade 2 Receptors ER and PR absents, Diploid, Heterozygous KRAS, Homozygous TP53, P53 WT molecular group Most likely, type II in function of the mutation, 91% of the genome is from East Asian North Endometrial adenocarcinoma, Grade 2 Receptors ER and PR mostly absents (not in all cells), Tetraploid, Heterozygous KRAS, Homozygous TP53, P53 WT molecular group Child of the subtype HER-1-A, Most likely, type II in function of the mutations, 92% of the genome is from East Asian North EC represents a major clinical challenge because of its increasing incidence and the diversity of its clinical and biological forms. Recent advances offer new prospects for diagnosis and treatment, but further efforts are needed to improve the prognosis and quality of life of patients. When we search for "EC" in the ATCC database, 25 results appear, with Ishikawa, HEC-1-A, HEC-1-B, and KLE being the most commonly used cell lines in research [ 106 ]. However, the use of cell lines presents challenges, particularly with respect to reproducibility [ 107 – 109 ]. Variations in cell density, culture media, and experimental techniques across studies can lead to inconsistent results, especially in cancer biology and drug development. Over time, cell lines undergo genetic, epigenetic, and morphological changes, further distancing their behavior from that of primary tissues or cells, which limits their clinical applicability. Additionally, cell line contamination is a significant concern that applies to all cell lines, including ECs. For example, residues of MCF-7 or HeLa cells have been detected in the ECC-1 cell line, raising questions about the reliability of certain experimental models. Below, we briefly describe the four major EC cell lines and their profiles, with all the information sourced from [ 109 ] . On the basis of current molecular classification, we were able to link cell lines to their group. However, we would like to add that some of the cell lines have high mutation rates and are sometimes very different from the primary cells taken from the patient, so the validation of molecular features prior to experimental use is recommended. Several cell lines are used in research, each with different genetic characteristics typical of EC, such as Ishikawas with PTEN loss or KLE with PIK3CA mutations. These findings allow researchers to study tumor behavior in vitro, as well as the response of tumors to various treatments. However, even though in vitro models are becoming increasingly relevant (e.g., organoids) and help reduce the use of animals, in vivo models remain essential in the field of medical research. The use of more complex and complete models not only enables cells to be grown in 3D but also, in the case of orthotopic models, enables interactions with the tumor microenvironment within the organ of interest to be taken into account, as well as the organism's metabolism, which cannot be accomplished in vitro [ 110 , 111 ]. Mice are among the most widely used animals in research, since they have several advantages over other in vivo models: they are small, easy to handle and take up very little space, they multiply rapidly, and their rapid metabolism enables faster tumor development than that of humans, enabling faster results. Therefore, various models are commonly used in EC research [ 112 ]. Over time, these models have become increasingly complex. Some spontaneously develop ECs, such as Han: Wistar rats, Donryu rats, DA/Han rats or BDII/Han rats [ 112 ]. Other models are chemically induced with estrogen injections, such as indole-3-carbinol, or chemical molecules that induce breaks in single- and double-stranded DNA, such as N-ethyl-N'-nitro-N- nitrosoguanidine [ 113 ]. Transgenic models present mutations characteristic of EC, such as mutations in PTEN [ 114 ], p53 [ 115 ], PPAP (POLE et POLD1) [ 116 ], and Mig6 [ 112 , 117 ]. Currently, some models are more relevant to ECs since they have a greater resemblance with human patients. These models involve the grafting of tumor cells from murine or human cell lines or directly from patients, either subcutaneously or orthotopically. This approach gives researchers control over the tumor, allowing them to select mutations of the cells relevant to their studies. For example, established cell lines can be used as xenograft models or genetically modified to overexpress or underexpress specific genes, allowing the study of tumor behavior and treatment responses [ 118 ]. Similarly, patient-derived xenografts (PDXs) involve the use of cells from patients to better reflect the genetic heterogeneity observed in clinical settings [ 119 ]. Finally, patient-derived organoid xenograft (PDOX) models involve generating organoids from patient biopsy cells and grafting them into the uterus of an in vivo model, further enhancing their clinical relevance (Table  7 ) [ 120 ]. In conclusion, these models allow researchers to study the involvement of genes in the development of EC, as well as to test different molecules and their impact on tumors in a much more representative way than they do in vitro [ 118 ]. Thus, it is also possible to study chemical resistance or the impact of diet on the evolution of EC, as Ke Shen et al. did by studying the effect of a high-fat diet on PDOX model mice, which confirmed that these models can consider the impact of an individual's metabolism and tumor microenvironment [ 121 ]. Overall, these models closely replicate patient characteristics, preserving biomarkers, genetic profiles, and tissue structures observed in clinical settings [ 122 ]. However, the immune system has not been studied thoroughly, and in vivo models are ideal for studying its mechanisms. Thus, in future studies, the various stages of development and maturation of immune cells can be studied within the tumor (immune infiltration, cytotoxicity, response to treatment, etc.) as well as in the various organs in which the cells develop and mature (Table  8 ). Table 8 Summary of the strengths and limitations of the PDX/PDOX models Strengths Limitations PDX/PDOX Cells grown in 3D (Takes into account intercellular interaction) Can consider the impact of an individual’s metabolism and tumor microenvironment. [ 111 , 121 ] In the future, these models may be useful for studying the immune system Preserve patient characteristics (biomarkers, genetic profiles) observed in clinical settings [ 122 ] Good representation of the genetic heterogeneity observed in clinical settings. [ 119 ] Clinical relevance [ 120 ] Enables study of tumor evolution and drug efficacy [ 118 ] Creating a mouse model is expensive [ 115 ] Cell culture takes time (PDOX) Variable grafting success rate (10 to 90%) [ 123 ] Models can take a long time to set up (6 months to 2 years) [ 123 ] Sometimes, the time to generate enough PDOX to test drugs is longer than the patient's lifespan [ 123 ] It is challenging to recover all of the patient data [ 123 ] Murin microenvironment is different from humans (e.g., cancer-associated fibroblasts, endothelial cells) [ 123 ] Potential selection of subclonal tumor cells during the transplantation phase [ 123 ] Particularly in the case of orthotopic grafting, organs are sometimes difficult to access due to their location or size (pancreas, mouse, kidneys) Summary of the strengths and limitations of the PDX/PDOX models Cells grown in 3D (Takes into account intercellular interaction) Can consider the impact of an individual’s metabolism and tumor microenvironment. [ 111 , 121 ] In the future, these models may be useful for studying the immune system Preserve patient characteristics (biomarkers, genetic profiles) observed in clinical settings [ 122 ] Good representation of the genetic heterogeneity observed in clinical settings. [ 119 ] Clinical relevance [ 120 ] Enables study of tumor evolution and drug efficacy [ 118 ] Creating a mouse model is expensive [ 115 ] Cell culture takes time (PDOX) Variable grafting success rate (10 to 90%) [ 123 ] Models can take a long time to set up (6 months to 2 years) [ 123 ] Sometimes, the time to generate enough PDOX to test drugs is longer than the patient's lifespan [ 123 ] It is challenging to recover all of the patient data [ 123 ] Murin microenvironment is different from humans (e.g., cancer-associated fibroblasts, endothelial cells) [ 123 ] Potential selection of subclonal tumor cells during the transplantation phase [ 123 ] Particularly in the case of orthotopic grafting, organs are sometimes difficult to access due to their location or size (pancreas, mouse, kidneys) The PI3K/AKT/mTOR signaling pathway is one of the most frequently dysregulated pathways in cancer and plays a critical role in cell growth, survival, proliferation, and metabolism [ 124 , 125 ]. It is activated by external signals such as growth factors and hormones, which bind to receptor tyrosine kinases (RTKs) on the cell surface, initiating a cascade of intracellular events (Fig.  3 ) [ 126 ]. Dysregulation of this axis, through mutations or other mechanisms, leads to increased tumorigenic potential by promoting cell cycle progression, evasion of apoptosis, and metabolic reprogramming. Fig. 3 Major inhibitors and the most commonly mutated components of the PI3K/AKT/mTOR signaling pathway. The diagram illustrates the PI3K/AKT/mTOR signaling pathway and the mode of action of the major inhibitors. The most commonly mutated components of this pathway are PTEN, PIK3CA, and AKT Major inhibitors and the most commonly mutated components of the PI3K/AKT/mTOR signaling pathway. The diagram illustrates the PI3K/AKT/mTOR signaling pathway and the mode of action of the major inhibitors. The most commonly mutated components of this pathway are PTEN, PIK3CA, and AKT EC has the highest prevalence of mutations in the PI3K/AKT/mTOR pathway, with alterations found in more than 80% of cases [ 8 , 127 ]. While the involvement of this pathway in tumorigenesis is well established, its exceptionally high mutation rate in EC highlights a unique vulnerability that could be exploited for therapeutic interventions. PI3K is a heterodimer consisting of the regulatory subunit PIK3R1 and the catalytic subunit PIK3CA [ 128 ]. Upon activation by RTKs or G-protein coupled receptors (GPCRs), PI3K phosphorylates phosphatidylinositol-4,5-bisphosphate (PIP2) into phosphatidylinositol-3,4,5-trisphosphate (PIP3). This lipid product recruits downstream effectors such as AKT to the plasma membrane [ 126 – 131 ]. In EC, PIK3CA mutations, including hotspot mutations such as H1047R, increase PI3K activity, leading to persistent pathway activation. Notably, the H1047R mutation selectively activates AKT2 over AKT1, driving processes such as EMT, metastasis, and chemoresistance [ 124 ]. PTEN, a tumor suppressor, antagonizes PI3K by dephosphorylating PIP3 back to PIP2, thereby suppressing AKT activation [ 132 ]. Loss-of-function mutations or deletions in this gene occur in up to 60% of ECs, leading to unchecked PI3K signaling [ 133 ]. Its inactivation is especially prevalent in early-stage endometrial tumors, highlighting its role in the initiation of tumorigenesis [ 134 ]. AKT is a serine/threonine kinase that acts as a central mediator in the PI3K/AKT/mTOR signaling pathway. Activation of the pathway results in the recruitment of AKT to the plasma membrane through its pleckstrin homology (PH) domain, where it is phosphorylated by PDK1 and mTORC2, where it becomes fully active [ 134 – 137 ]. AKT then phosphorylates a variety of downstream substrates involved in cell survival, growth, metabolism, and apoptosis. Some major substrates of this kinase include the cell cycle regulator FOXO, the NFkB regulator IKK, and the p53 inhibitor MDM2 [ 138 , 139 ]. Through these interactions, AKT plays pivotal roles in promoting cell proliferation, survival, and resistance to apoptosis, all of which are key processes in tumorigenesis. There are three isoforms of AKT—AKT1, AKT2, and AKT3—which share a conserved structural framework but differ in tissue distribution, subcellular localization, and specific functions. All three isoforms contain a PH domain, a central kinase domain, and a regulatory hydrophobic motif, with variations in these domains contributing to isoform-specific functions [ 140 ]. Differences in the PH domain, for example, affect their affinity for PIP3 and other membrane lipids, which influences their localization and activity [ 92 ]. Chemoresistance is one of the major challenges in cancer therapy, and AKT isoforms are central to this phenomenon [ 141 ]. AKT1 is ubiquitously expressed and primarily regulates cell growth, survival, and proliferation. In EC, dysregulation of AKT1 contributes to increased tumor proliferation and evasion of apoptosis. While AKT1 plays a key role in promoting tumor growth, its involvement in chemoresistance is less pronounced than that of other isoforms [ 142 , 143 ]. On the other hand, AKT2 is strongly associated with metastasis, EMT, and chemoresistance. It is critical for survival following chemotherapy, particularly in response to cisplatin, as its activation induces the upregulation of antiapoptotic proteins such as Bcl-2, reducing cell death and facilitating drug resistance [ 141 , 143 ]. The role of AKT2 is further underscored by its selective activation in the presence of the PIK3CA H1047R mutation, which enhances tumor invasiveness and resistance to treatment [ 124 ]. Additionally, AKT2 mediates the activation of the NFκB pathway, which promotes angiogenesis and tumor invasiveness, potentially contributing to chemoresistance in EC [ 144 , 145 ]. AKT3 remains comparatively less characterized in EC but has been implicated in metabolic regulation and drug resistance. While its role in EC remains to be fully elucidated, studies have shown that downregulation of AKT3 sensitizes cells to doxorubicin but does not significantly affect cisplatin sensitivity [ 141 ]. In ovarian cancer, reducing AKT3 expression in tumor cells results in reduced metastasis but does not fully mimic the effects of AKT1 inhibition, which decreases tumor growth and angiogenesis [ 146 ]. In triple-negative breast cancer, AKT3 downregulation promotes tumor cell migration and metastasis via the upregulation of S100A4 [ 147 ], suggesting a role in metastatic potential across various cancers. These findings highlight that the function of AKT3 in cancer progression may be context dependent, but its role in EC still requires further investigation to understand its full impact on tumor biology and therapy response. Although AKT mutations are rare (∼2%) in EC [ 143 ], the functional divergence of AKT isoforms emphasizes their distinct contributions to tumorigenesis and chemoresistance. These isoform-specific roles suggest that targeted therapies aimed at selectively modulating AKT isoforms could offer promising strategies for overcoming resistance and improving treatment outcomes in patients with EC. mTOR is a serine/threonine kinase that acts as a central regulator of cellular metabolism, growth, proliferation, and survival upstream and downstream of AKT. It functions as part of two distinct complexes: mTORC1 and mTORC2 [ 148 ]. mTORC1 integrates signals from growth factors, nutrient availability, and energy status to regulate processes such as protein synthesis, autophagy, and lipid metabolism. This process is mediated through key downstream effectors, such as S6 kinase (S6K) and 4E-BP1, which promote protein translation and cell growth [ 143 ]. On the other hand, mTORC2 is involved primarily in cytoskeletal organization and cell survival. It phosphorylates and activates AKT at Ser473, completing its activation and enabling its full functionality [ 149 ]. Aberrant activation of mTOR signaling, often driven by mutations in PIK3CA or loss of PTEN , is a hallmark of EC [ 127 ]. Persistent mTOR activation promotes tumor growth, angiogenesis, and resistance to apoptosis, contributing to the aggressive behavior of endometrial tumors [ 142 ]. Treatments targeting the PI3K/AKT/mTOR pathway in endometrial cancer (EC) represent a promising therapeutic avenue, as this pathway is frequently dysregulated in this malignancy. Several inhibitors targeting different nodes of the pathway are currently under investigation or are in clinical use. PI3K inhibitors include alpelisib [ 150 , 151 ], which specifically targets PIK3CA mutations and has shown clinical activity in advanced gynecological cancers, and copanlisib [ 152 ], which is being evaluated in phase II trials for PIK3CA-mutated endometrial carcinoma. AKT inhibitors, such as ipatasertib and capivasertib, are in clinical trials, with evidence suggesting their potential to improve outcomes when combined with other therapies, including antiangiogenic agents and immune checkpoint inhibitors [ 153 , 154 ]. mTOR inhibitors include everolimus, temsirolimus, ridaforolimus, AZD-8055, and vistusertib. Everolimus and temsirolimus are more advanced and have been tested in combination with hormonal therapies in recurrent or metastatic cases, whereas ridaforolimus and vistusertib are still being evaluated in earlier trials [ 155 – 157 ]. EC and its treatments show considerable diversity. Over time, many classifications have been developed, both from a clinical perspective (stage, grade, and histology) and a more theoretical biological standpoint (immortalized cell lines and molecular classification). Over the years, categorizations have become more refined, allowing for improved patient care. Treatments have evolved through a variety of approaches, including hormonal therapies, chemotherapies, targeted inhibitors (e.g., mTOR and PI3K inhibitors), and immunotherapies. The pace of translation from preclinical studies to clinical practice has occasionally led to discrepancies between laboratory efficacy and real-world clinical outcomes. EC is heterogeneous, with some subtypes sharing molecular features with colorectal cancer, such as microsatellite instability (MSI), POLE mutations, and defective mismatch repair, whereas others resemble high-grade serous ovarian carcinoma in terms of p53 mutations and genomic instability. The fact that EC is linked to the female reproductive system and regulated by hormones further complicates the tumor environment, making treatment challenging. Acquired therapeutic resistance, or even disease progression despite treatment, is a possible outcome, and preserving a woman's fertility can also pose a significant challenge. Another important issue, as previously discussed, is the difficulty of diagnosing patients at an early stage. Unfortunately, gender-related barriers sometimes complicate this process. In some countries, access to care is politically unfeasible, whereas in others, follow-up is hindered by a lack of financial or time resources [ 71 – 81 ]. This leads to delays in diagnosis, and diagnosing women with advanced EC makes treatment even more difficult [ 158 ]. If the cancer has metastasized, more complex and expensive surgeries and treatments must be considered. Although recent advances in targeted therapies and combination regimens offer encouraging prospects, significant challenges remain in optimizing efficacy, managing toxicity, and overcoming therapeutic resistance. Research is ongoing, and promising new approaches are emerging. Combination therapies represent a significant step forward, offering the potential to manage and reduce side effects in affected women. In particular, the PI3K/AKT signaling pathway appears to be a promising target given its high mutation frequency; however, emerging evidence from clinical trials indicates that while some patients benefit from these inhibitors, toxicity profiles and the development of resistance remain substantial obstacles [ 158 ]. Although our laboratory is focused on these inhibitors, their clinical efficacy remains variable depending on the therapeutic context: 56% of patients in SAR245408/XL147 had a progression-free survival of 6 months or more [ 159 ], and 38% of patients discontinued their treatment as a result of disease progression in NCT00739830 [ 160 ] and 40% in AGO-GYN8 [ 161 ]. Significant challenges remain to be addressed in this field [ 157 – 161 ]. Artificial intelligence, whether in predictive modeling, histopathological image analysis, risk stratification or direct inspection, is a path to explore that will undoubtedly lead to effective, real-world solutions for endometrial cancer. Models such as CgMLP, EndoNet and HIENet are already helping to detect and/or classify endometrial cancers more accurately than humans alone [ 162 – 164 ].

EC is the most common type of gynecological cancer in developed countries, with an increasing trend over the past 10 years [ 1 – 3 ]. This increase is partly attributed to various known risk factors for EC, such as rising obesity rates, an aging population and changes in reproductive behavior. In 2023, the highest incidence rates were reported in North America, Eastern Europe and Western Europe [ 1 ]. On the other hand, mortality rates have shown a more stable trend in recent years [ 4 , 5 ], with a slight increase in some countries. EC is generally associated with a favorable prognosis, with a 5-year relative survival rate of approximately 81% in Canada and the United States [ 2 , 6 ]. However, despite advances in treatment and earlier diagnosis, the 5-year relative survival rate has not improved significantly in recent years. Thus, current trends highlight the importance of continuing to improve prevention, early diagnosis and treatment strategies to better manage the increasing incidence and significant impact of EC worldwide. In fundamental cancer research, one protein is highly important: p53. Deletion of this protein or its inactivation by PTEN mutations plays a part in endometrioid carcinoma. This mutation is considered a “driver” protein for this type of cancer, and loss of TP53 leads to the majority of serous carcinomas [ 7 ]. However, a mutation in this first gene is present in 83% of endometrial adenocarcinomas, and its deletion or loss is associated with high levels of AKT phosphorylation [ 8 ]. Thus, the PI3K-AKT signaling axis has been identified as a promoter of EC. This pathway is well known for its high mutation rate in ECs [ 9 – 11 ], as well as its involvement in proliferation and the cell cycle. New AKT inhibitors are emerging and are being used clinically as single or dual therapies. Our lab is studying the role of AKT in endometrial cancer, and we have dedicated a final section to exploring the current changes and the involvement of these proteins in the progression of EC. As mentioned earlier, clinical observations in this field have remained stagnant for years. However, what progress has been made on the fundamental research side? While research in this area continues, bridging the gap between clinical practice and biological insights remains a challenge. The aim of this literature review is to evaluate the situation from both clinical and fundamental perspectives, providing researchers and clinicians with updated information on recent developments.