This chapter should be cited as follows:
Moro F, Culcasi C, et al., Glob Libr Women's Med
ISSN: 1756-2228; DOI 10.3843/GLOWM.419603

The Continuous Textbook of Women’s Medicine Series – Gynecology Module

Volume 10

Ultrasound in gynecology

Volume Editors: Professor Antonia Testa, Agostino Gemelli University Hospital, Rome, Italy
Professor Simona Fragomeni, Agostino Gemelli University Hospital, Rome, Italy

Chapter

Endometrial Pathology

First published: April 2025

Study Assessment Option

By answering four multiple-choice questions (randomly selected) after studying this chapter, readers can qualify for Continuing Professional Development points plus a Study Completion Certificate from GLOWM.
See end of chapter for details.

INTRODUCTION

Endometrial pathology includes a wide spectrum of benign and malignant conditions that can significantly impact a woman's reproductive and overall health. Common benign abnormalities, such as endometrial polyps, hyperplasia and synechiae, are often associated with abnormal uterine bleeding or infertility.^1,2,3 On the other hand, endometrial cancer is the most prevalent gynecological malignancy in developed countries, with the incidence rising due to lifestyle and hormonal factors.⁴ An early and accurate diagnosis is crucial, with transvaginal ultrasound examination serving as the first-line imaging modality for endometrial assessment.

This chapter explores ultrasound-based diagnostic criteria, including International Endometrial Tumor Analysis (IETA) terminology, and advanced sonographic techniques for enhanced lesion characterization.⁵

EXAMINATION TECHNIQUE

Transvaginal ultrasound examination is the primary imaging technique for evaluating the uterus, endometrium and related pathologies. It should be performed with an empty bladder, ideally during the early proliferative phase in premenopausal women, and between 5 and 10 days after the last progestin intake in postmenopausal patients undergoing cyclic hormone replacement therapy. In postmenopausal patients not receiving therapy or on continuous hormone replacement therapy, it can be performed at any time. When transvaginal ultrasound is not feasible, such as in cases of virginity, vaginismus or secondary vaginal stenosis, the transrectal approach is recommended; meanwhile, a transabdominal scan with a 90° angle of insonation may be necessary in cases of large fibroids or a globally enlarged uterus. Evaluation of the uterus should begin by identifying the bladder and cervix, followed by assessing its position and obtaining measurements.

For a comprehensive evaluation of the entire endometrium, sagittal scans should first be performed by sweeping the probe across the entire uterine cavity, from the right to the left tubal angle and in the transverse (or oblique) plane from the cervix to the fundus.

After obtaining a comprehensive view of the uterus, the image is adjusted to concentrate solely on the uterine corpus. In this context, real zoom is a valuable tool, as it allows real-time magnification of a specific area without loss of resolution, unlike conventional zoom, which may degrade image quality. This function enables a more detailed analysis of endometrial morphology, enhancing the visualization of subtle features such as endometrial echogenicity, irregularities or small lesions. Real zoom is particularly useful for refining the diagnosis of endometrial pathology, including hyperplasia, polyps or neoplasms, and for optimizing patient management. In addition to grayscale examination, color or power Doppler should always be performed to assess vascularization of the endometrium or any endometrial pathology.

IETA terms and measurement

Standardized terminology to describe the endometrium and endometrial pathology has been established by the IETA group.⁵ According to this terminology, for an accurate assessment of endometrial thickness, it should be measured in the longitudinal plane and reported in mm. In the presence of intracavitary fluid, endometrial thickness is determined by summing the measurements of both endometrial layers, excluding the fluid from the calculation (Figure 1). When intracavitary endometrial pathology is present, the total endometrial thickness, including the lesion, should be recorded. However, if an intracavitary myoma is clearly visualized, it should be excluded from the measurement. The endometrial lesion should be measured in three orthogonal planes: anteroposterior (thickness) and craniocaudal (length) in the sagittal view, and laterolateral (width) in the transverse view. Lesion volume is then estimated using the ellipsoid approximation formula: (d1 × d2 × d3)/2.

The main parameters included in the IETA consensus are echogenicity of the endometrium or endometrial lesion, endometrial outline, endometrial midline, intracavitary fluid, endometrial–myometrial junction, color score and vascular pattern.

Echogenicity

Echogenicity is the property of a tissue to reflect sound waves, which are then detected and displayed as echoes on an ultrasound image. Endometrial echogenicity is classified as uniform or non-uniform.

Uniform includes a three-layer pattern, typical of the periovulatory phase and characterized by a central echogenic line, a surrounding hypoechoic functional layer and an outer echogenic basal layer, or homogeneous echogenicity, which can be hypoechogenic, isoechogenic or hyperechogenic relative to the myometrium (Figure 2A−D).

Non-uniform includes heterogeneous echogenicity, asymmetrical appearance and presence of cystic areas (Figure 2E,F).

Endometrial outline

The endometrial outline or margin of an intracavitary lesion should be defined as smooth or irregular. The presence of a ‘bright edge’, which refers to the echogenic interface created between an intracavitary lesion and the endometrium, should also be reported (Figure 3).

Endometrial midline

The endometrial midline is defined as linear, non-linear (wavy), irregular or not defined (Figure 4).

Intracavitary fluid

Intracavitary fluid is described as anechogenic or as having low-level echogenicity, ground-glass appearance or mixed echogenicity.

Endometrial–myometrial junction

The endometrial–myometrial junction is described as regular, irregular, interrupted or not defined (Figure 5).

Color score

Color and power Doppler assessment should encompass both the endometrium and adjacent myometrium. Doppler imaging should be performed with a pulse repetition frequency (PRF) set between 0.3 and 0.9 kHz, adjusting it and the gain to ensure clear visualization of vessels without artifact-related blooming. The vascularity within the endometrium can be evaluated using the International Ovarian Tumor Analysis (IOTA) color score, which has been previously applied to ovarian masses (Figure 6):⁶ no flow (score 1), minimal flow (score 2), moderate flow (score 3) and abundant flow (score 4).

Vascular pattern

Evaluation of endometrial vascularization involves assessing the presence or absence of prominent vessels and identifying specific vascular patterns. These patterns can be categorized as single ‘dominant’ vessel (without or with branching), multiple vessels with focal origin (two or more vessels appear to share a common stem) or with multifocal origin (vessels originate from multiple points at the myometrial–endometrial junction), scattered vessels meaning dispersed color signals within the endometrium, without a visible origin at the myometrial–endometrial junction, and, circular flow (Figure 7).

BENIGN CONDITIONS

Benign endometrial pathology comprises various conditions, including endometrial polyps and hyperplasia and uterine synechiae. Additionally, although intracavitary leiomyomas originate from the myometrium rather than the endometrium, they should be considered in differential diagnoses. The clinical manifestations of these conditions vary according to the patient's hormonal status, particularly in relation to menopause. The most common symptoms include abnormal uterine bleeding, such as intermenstrual spotting, menometrorrhagia, and unexpected postmenopausal bleeding, as well as fertility issues. However, in many cases, diagnosis is incidental, occurring during routine check-ups without evident symptoms.

Endometrial polyps

Endometrial polyps are benign focal lesions originating from the endometrial mucosa, composed of a stromal core covered by cylindrical epithelium, which contains endometrial glands, stroma and blood vessels. Polyps can be single or multiple, ranging in size from a few mm to several cm, and their morphology may be sessile with large or small implantation base or pedunculated.^7,8 Most endometrial polyps originate from the fundal region and extend toward the internal os. In some cases, they may protrude through the external cervical os and become visible in the vaginal canal. The precise etiology of endometrial polyps remains uncertain; however, their development is likely associated with estrogenic stimulation. Established risk factors include advanced age, hypertension, obesity and tamoxifen use, with an estimated prevalence in the latter group ranging from 30% to 60%.^7,8

Polyps are frequently observed, with a prevalence between 11% and 24%, in women of reproductive age. However, their occurrence tends to rise over time, being more common in postmenopausal women. In this group, up to 50% of those experiencing abnormal uterine bleeding are diagnosed with endometrial polyps. These growths can be asymptomatic. The most common symptoms include abnormal uterine bleeding (68% of cases, including postmenopausal bleeding), and less commonly, infertility (with a prevalence between 6% and 32%). Malignant transformation is uncommon, occurring in approximately 0.5%–3% of cases, with a higher incidence in postmenopausal women.⁹ The risk factors for malignant transformation are presence of abnormal uterine bleeding, age over 60 years old, postmenopausal status, hypertension, diabetes mellitus, tamoxifen use and obesity.^7,8

Two-dimensional transvaginal ultrasound represents the primary imaging modality for the detection of endometrial polyps, offering a sensitivity ranging from 19% to 96% and a specificity between 53% and 100%.¹⁰ However, hysteroscopy with guided biopsy remains the gold standard for diagnosis, as it allows for direct visualization of the endometrial cavity and the simultaneous removal of polyps, providing both diagnostic and therapeutic benefits.

Ultrasound characteristics

In women with endometrial polyps with abnormal uterine bleeding, the endometrial thickness varies between 8 mm and 14 mm.¹¹

Typically, an endometrial polyp appears as a localized endocavitary lesion with homogeneous hyperechogenicity (Figure 8) or as non-uniform with no cysts or regular cysts, the latter being more frequent in postmenopausal women and corresponding histologically to dilated glands containing proteinaceous material. Moreover, tamoxifen use is frequently associated with (sub) endometrial cysts on ultrasound and cystically dilated endometrial glands on histological examination, often occurring in the context of polyps, endometrial cystic atrophy or adenomyosis.^12,13

Endometrial polyps are categorized as either sessile or pedunculated based on the proportion between their basal diameter at the endometrial interface (a) and their maximum transverse diameter (b). A polyp is classified as pedunculated when the a/b ratio is less than 1, whereas a ratio equal to or greater than 1 indicates a sessile morphology.¹⁴ A polyp’s outline is usually regular, and thin hyperechoic echoes, known as the ‘bright-edge’ sign, may appear along the margins, oriented perpendicularly to the ultrasound beam, detected more often before than after menopause (55% vs 37%).¹¹ This feature is recognized as a reliable sonographic marker for polyp detection, particularly in women with a secretory endometrium.¹⁵

The polyp may focally interrupt the endometrial midline in 64% of cases, especially in postmenopausal patients, while the myometrial–endometrial junction appears regular in 77% of cases.¹¹ The use of color or power Doppler enhances the diagnostic accuracy of two-dimensional transvaginal ultrasound in evaluating endometrial polyps. Typically, color Doppler identifies a single vascular structure, known as the ‘feeding vessel’ (sensitivity 89%, specificity 87%) (Figure 9), running through the connective axis of the polyp. This pattern contrasts with the multiple vessel pattern more commonly observed in atypical hyperplasia and malignant endometrial lesions.

In approximately 69% of cases, a single vessel, with or without branching, is detected, often corresponding to a color Doppler score of 2 or 3.¹¹ Polyps can also be multiple, each with their own vessel, making them distinguishable from blood clots, which may appear within the endometrial cavity but do not exhibit a vascular pedicle or blood flow on color/power Doppler.¹¹

Endometrial polyps with well-defined margins, presence of the bright-edge sign and small intralesional cysts are generally indicative of benignity. In contrast, large polyps, with increased vascularization, non-uniform echogenicity, irregular lesion surface within a fluid-distended uterine cavity, and abnormal uterine bleeding, raise the suspicion for potential malignancy.¹⁶

Uterine synechiae

Intrauterine adhesions, or synechiae, refer to the formation of scar tissue within the uterine cavity. These adhesions can vary in severity, ranging from partial involvement, affecting 30% to 60% of the endometrial surface, to complete obliteration of the cavity in the most severe cases. When accompanied by symptoms such as hypomenorrhea/amenorrhea or reduced fertility, the condition is classified as Asherman’s syndrome.¹⁷ The main cause is trauma to the uterine cavity by diagnostic curettage, spontaneous miscarriage, abortion, a hysteroscopic procedure (e.g. myomectomy), inflammation or infection (e.g. endometritis), or they can form after childbirth.¹⁸

Ultrasound characteristics

Intrauterine adhesions can sometimes mimic endometrial polyps, making differential diagnosis essential for accurate evaluation (Table 1). Ultrasonographic features of uterine synechiae include isoechoic bands of varying thickness that interrupt the continuity of the endometrium (Figure 10). The endometrium may appear either thin or thick, depending on the extent of intrauterine adhesions. In some cases, anechoic areas filled with fluid may be seen, surrounded by fibrous adhesions. When subendometrial fibrosis coexists, the endometrium may appear focally hyperechoic and may produce acoustic shadows. On two-dimensional saline infusion sonography (SIS), synechiae appear as avascular echogenic bands crossing the uterine cavity and, in severe cases, cavity distension may be difficult or impossible if the fibrosis is extensive.⁵ Three-dimensional SIS can be useful for evaluating the extent of intrauterine adhesions in the coronal plane and determining the volume of the uterine cavity, which is often reduced in such cases.

Endometrial hyperplasia

Endometrial hyperplasia is a condition characterized by abnormal endometrial growth, exceeding the normal proliferative phase and leading to an increased gland-to-stroma ratio. According to the World Health Organization (WHO) classification, endometrial hyperplasia is divided into two main types: non-atypical endometrial hyperplasia and atypical endometrial hyperplasia, also known as endometrioid intraepithelial neoplasia (EIN).¹⁹ Non-atypical endometrial hyperplasia is marked by an overgrowth of endometrial glands with irregular sizes and shapes but without significant cytological atypia. The risk of progression to endometrial carcinoma in this case is relatively low, estimated at 1–3%.²⁰ Atypical endometrial hyperplasia (EIN), on the other hand, is characterized by abnormal proliferation of endometrial glandular cells with atypical features, including enlarged nuclei, irregular cell shapes and loss of normal glandular architecture. Long-term studies indicate a significantly increased risk of malignant transformation, with estimates ranging from 20% to 50%, influenced by factors such as histopathological characteristics and genetic mutations.²¹

The risk factors for endometrial hyperplasia closely overlap with those for endometrioid adenocarcinoma of the endometrium. The most significant risk factor is prolonged estrogen dominance due to an imbalance between estrogen and progesterone. This condition commonly occurs in women with anovulatory cycles, obesity or polycystic ovary syndrome (PCOS), those using tamoxifen, or in nulliparous women. Genetic predisposition also plays a role, with Lynch syndrome being a recognized risk factor, while Cowden syndrome, though less common, is also associated with an increased risk of endometrial abnormalities.²¹

The most common and often earliest symptom of endometrial hyperplasia is abnormal uterine bleeding.²² A recent study applied the terminology of the IETA group to describe endometrial pathology in premenopausal and postmenopausal women with abnormal uterine bleeding. The study included 2216 women with a final histological diagnosis, among whom 118 had simple hyperplasia (5.3%) and 18 (0.9%) had atypical hyperplasia, with no significant differences in prevalence between premenopausal and postmenopausal women.¹¹

Ultrasound characteristics

There is no standardized cut-off for endometrial thickness to enhance the diagnostic accuracy in differentiating between normal endometrium, endometrial hyperplasia and malignant pathology, as the literature presents conflicting and inconsistent findings.

In endometrial hyperplasia without atypia, the endometrial thickness is 9–17 mm. The endometrium frequently appears with non-uniform echogenicity without cysts or is uniformly hyperechogenic. The midline is often indistinct particularly in postmenopausal patients. The endometrial–myometrial junction is generally regular, and the color score is usually absent or minimal. When vascularization has been identified on color Doppler, the prevailing vessel morphology included multiple vessels with a multifocal origin or a scattered pattern (Figure 11) (Table 2).¹¹

In atypical hyperplasia, endometrial thickness is 8–18 mm. The most frequently observed echogenicity is non-uniform without cysts. The endometrial midline is often indistinct and the endometrial–myometrial junction appears regular in 65% of cases. When color Doppler signals are present, the predominant vascular pattern consists of multiple vessels with a multifocal origin or a scattered distribution. A color score of 2 (minimal) or 3 (moderate) is reported in 53% of cases (Figure 12) (Table 2).¹¹

ENDOMETRIAL CANCER

Epidemiology

Endometrial cancer is the sixth most common cancer in women with approximately 420 368 cases diagnosed worldwide in 2022. The estimated number of new cases of endometrial carcinoma in Europe in 2022 was 124 874 with 30 272 deaths.^23,24 More than 90% of endometrial cancer cases occur in women over the age of 50, with a median age at diagnosis of 63 years.²⁵ However, 4% of women diagnosed with endometrial cancer are under 40 years old, many of whom still wish to preserve their fertility.²⁶ These trends are partially attributed to lifestyle factors such as increasing rates of obesity and declining fertility, both of which are established risk factors for endometrial cancer.²⁷

Risk factors

Endometrial cancer is influenced by several well-established risk factors. Obesity (body mass index > 30) significantly increases the risk, with an estimated three- to four-fold elevation. Other contributing factors include chronic hyperestrogenism, whether endogenous or exogenous, PCOS, tamoxifen therapy, hormone-secreting neoplasms such as granulosa cell tumors, anovulatory cycles and diabetes mellitus. In addition to a specific genetic predisposition to endometrial carcinoma, some cases involve a familial predisposition to developing malignant tumors in multiple organs. This is seen in Lynch syndrome type II, where the risk of developing cancer is 40–80% for colorectal cancer, 40–60% for endometrial cancer and 10–12% for ovarian cancer.²⁸

Clinical presentation

Endometrial cancer is a highly curable malignancy, especially when diagnosed at an early stage. The most common symptom, abnormal uterine bleeding, occurs in 75–90% of cases and frequently facilitates early detection. According to the FIGO annual report, the 5-year overall survival rate for endometrial cancer is approximately 80%. However, histopathological features of the disease significantly influence both patient prognosis and the recommended therapeutic approach.²⁹ Sometimes it is detected incidentally through abnormal cervical cytology, with findings such as atypical glandular cells or adenocarcinoma which may originate from either the endocervix or the endometrium. In other cases, it is discovered incidentally during imaging studies such us ultrasound, computed tomography (CT) or magnetic resonance imaging (MRI) performed for unrelated indications or as pathological finding following hysterectomy or abdominopelvic surgery. The time interval between symptoms onset and diagnosis is generally short, with over 40% of cases diagnosed within 3 months. However, certain cases, particularly those with serous histology, may remain asymptomatic. In fact, serous histology is often associated with endometrial atrophy, which can delay diagnosis and negatively impact prognosis.³⁰

Histotypes

Endometrioid carcinoma

Endometrioid carcinoma is the most common histotype (75–80%), typically diagnosed at early stages and associated with a favorable prognosis. Variants include carcinoma with squamous differentiation and carcinoma with mucinous differentiation. It is estrogen-related and commonly presents as Grade (G) 1 or 2 (low grade) or, less frequently, as G3 (high grade).³¹

Serous carcinoma

Serous carcinoma is the second most common histotype (10%). It often shows overexpression of Her2, which can be targeted therapeutically. This histotype is more aggressive and is often associated with aberrant expressions of p53 (i.e. either absent or overexpressed). It should be suspected in women who are approximately 10 years older than those with endometrioid adenocarcinoma or who have a history of pelvic radiation, prolonged tamoxifen therapy or breast cancer. It presents with myometrial and vascular invasion in up to 75% of cases, and it often presents at Stage III or IV, with metastasis to pelvic and para-aortic lymph nodes. The prognosis is poorer than for the endometrioid form.^30,31

Clear cell carcinoma

Clear cell carcinoma is a rare histotype (< 5% of cases). It typically presents in older women and is usually negative for estrogen receptors. The prognosis is poor, often presenting at an advanced stage.³²

Mixed carcinoma

Mixed carcinoma is characterized by the presence of at least two distinct histological components: endometrioid associated with serous or clear cells. The non-endometrioid carcinoma component must represent at least 10% of the tumor to be classified as mixed.³⁰

Undifferentiated carcinoma

Undifferentiated carcinoma lacks glandular or squamous differentiation but expresses epithelial markers, such as cytokeratin. It forms solid masses of undifferentiated cells and may coexist with endometrioid adenocarcinoma or appear as the sole tumor component. A rarer subtype, small-cell undifferentiated carcinoma, resembles small-cell carcinomas of other organs and shows neuroendocrine differentiation, with markers like chromogranin and synaptophysin. Another type, dedifferentiated carcinoma, often linked to Lynch syndrome, combines a well-differentiated adenocarcinoma (G1 or G2) with undifferentiated carcinoma. Variants may include myxoid stroma or rhabdoid cells.³³

Carcinosarcoma

Carcinosarcoma (malignant mixed Müllerian tumor) is a rare and aggressive biphasic tumor composed of a high-grade sarcomatous component, which can be homologous (e.g. endometrial stromal sarcoma or leiomyosarcoma) or heterologous (e.g. chondrosarcoma or rhabdomyosarcoma), and a high-grade carcinomatous component. The carcinoma often exhibits features of high-grade endometrioid, serous, clear cell or undifferentiated carcinoma.³⁴

Mucinous adenocarcinoma

Mucinous adenocarcinoma should be distinguished from primary endocervical adenocarcinoma (which has a worse prognosis) by its lesser mucin content.²⁹

Squamous carcinoma

Squamous carcinoma should be distinguished from squamous-differentiated adenocarcinoma due to the absence of glandular differentiation and from squamous cell carcinoma of the cervix that has spread to the endometrium.²⁹

G3 endometrioid carcinomas, serous carcinomas and clear cell carcinomas are classified as high-grade carcinomas.²⁹

Molecular classification

In 2020, the WHO classification of endometrial cancer integrated the molecular classification introduced by The Cancer Genome Atlas (TCGA) in 2013.³⁵ This paradigm-shifting framework has transformed the understanding and management of endometrial cancer by defining four distinct molecular subgroups based on genomic and molecular characteristics. These subgroups provide critical prognostic insights and have become instrumental in guiding personalized treatment strategies.³⁶

POLE-ultramutated

The POLE-ultramutated (POLEmut) subgroup is characterized by an exceptionally high mutational burden due to somatic or germline mutations in the POLE gene, which encodes the catalytic subunit of DNA polymerase epsilon. These tumors are frequently associated with mutations in the PI3K/PTEN signaling pathway and are mostly seen in Grade-2 endometrioid carcinomas. POLEmut tumors have an excellent prognosis, with low rates of recurrence and high survival rates, often requiring minimal therapeutic intervention. Due to their favorable outcomes, they represent candidates for treatment de-escalation.

Mismatch repair-deficient

The mismatch repair-deficient (MMRd) subgroup is associated with defective DNA mismatch repair, leading to microsatellite instability. This subgroup commonly involves mutations in MLH1, MSH2, MSH6 and PMS2, which can be either somatic or related to Lynch syndrome. These tumors are frequently observed in G2 or G3 endometrioid carcinomas and occasionally in serous carcinomas. They are often accompanied by a prominent lymphocytic infiltrate. MMRd tumors have an intermediate prognosis, with a higher likelihood of local recurrence than distant metastasis. Patients in this subgroup may benefit from immunotherapy, such as immune checkpoint inhibitors targeting PD-1 or PD-L1.

P53-abnormal

P53-abnormal (p53abn) subgroup is defined by alterations in the TP53 gene, resulting in abnormal p53 protein expression. These tumors are often associated with chromosomal instability and are predominantly observed in serous carcinomas, although they can also occur in high-grade endometrioid carcinomas. The prognosis for p53abn tumors is poor, with high rates of recurrence and progression. These cases require aggressive treatment, including chemotherapy and radiotherapy, to mitigate the high risk of relapse.

No specific molecular profile

The no specific molecular profile (NSMP) subgroup lacks the defining molecular characteristics of the other three subgroups. Common mutations in this group include PIK3CA and KRAS. These tumors are typically associated with G2 or G3 endometrioid carcinomas. Prognosis for NSMP tumors is intermediate, with significant heterogeneity in clinical outcomes. Treatment strategies for this subgroup are primarily guided by traditional clinicopathological features rather than molecular data.

In 3–5% of cases, endometrial carcinomas present with multiple concurrent mutations. These cases are classified as ‘multiple classifiers’. In cases in which multiple classifiers include POLEmut or MMRd along with a secondary p53mut alteration, current scientific evidence suggests that these tumors should not be classified as p53mut, as they retain a favorable prognosis similar to that of tumors with POLEmut or MMRd. Therefore, patients harboring both POLEmut and p53mut should be classified as POLEmut, while those with MMRd and p53mut should be categorized as MMRd.³⁷

Staging

The 2023 FIGO staging system (Table 3)²⁹ arises from the need to integrate the old FIGO 2009 staging with the new ESGO/ESTRO/ESP 2021 molecular classification. The staging integrates: histotype and grade of differentiation (‘non-aggressive histological types’: low-grade endometrioid tumors vs ‘aggressive histological types’: high-grade endometrioid tumors, serous carcinoma, clear cell carcinoma, mixed carcinoma, undifferentiated carcinoma, carcinosarcoma, mesonephric-like carcinoma, gastrointestinal-type mucinous carcinoma), invasion of lymphovascular spaces (‘negative’ if no vessels are involved, ‘focal’ if fewer than five vessels are involved, ‘substantial’ if at least five vessels are involved) and molecular classification.

This new classification means that the detection of POLE mutations in Stage II leads to downstaging to Stage I, while the detection of p53 mutations results in upstaging to Stage II. The stratification of endometrial cancer in these four categories allows identification of the most appropriate surgical treatment for each patient and subsequent adjuvant therapy when necessary.²⁹

Risk stratification

The European Society of Gynaecological Oncology (ESGO), the European Society for Medical Oncology (ESMO) and the European Society of Pathology (ESP) guidelines for the management of patients with endometrial carcinoma have introduced a refined risk stratification system that integrates both histopathological and molecular features.³⁸ This classification allows for a more precise prognostic assessment and guides treatment decisions by categorizing patients into five distinct risk groups (Table 4). By incorporating molecular markers such as p53 status, mismatch repair (MMR) deficiency and POLE mutations, this approach enables a more personalized management strategy, optimizing therapeutic interventions while minimizing overtreatment.

Diagnosis

Pathological assessment of an endometrial biopsy is essential for determining the histopathologic tumor type and grade, which are crucial for appropriate therapy planning and prognostic evaluation.³⁸ Hysteroscopy is considered the gold standard for assessing the uterine cavity due to its high accuracy and feasibility, with a sensitivity of 86.4% and a specificity of 99.2% for diagnosing endometrial cancer.³⁹ According to ESGO/ESTRO/ESP, the preoperative workup should include a detailed family history, a comprehensive clinical evaluation, including a pelvic examination, and advanced imaging techniques such as expert transvaginal or transrectal ultrasonography or pelvic MRI, which are essential for evaluating deep myometrial and cervical stromal invasion. Preoperative CT or positron emission tomography-CT (PET-CT) are clinically useful for detecting metastatic disease in patients with advanced endometrial cancer.⁴⁰

Standard surgical procedures

The standard treatment for early-stage endometrial cancer consists of total hysterectomy with bilateral salpingo-oophorectomy, without vaginal-cuff resection. Infracolic omentectomy is recommended as part of the staging procedure for patients with clinical Stage-I serous endometrial carcinoma, carcinosarcoma and undifferentiated carcinoma. However, it may be omitted in cases of Stage-I clear cell and endometrioid carcinoma.

Lymph node staging using the sentinel lymph node (SLN) technique can be considered for patients with low- to intermediate-risk disease but may be omitted when there is no evidence of myometrial invasion. In contrast, for patients with intermediate- to high-risk disease, systematic lymphadenectomy is recommended.³⁸ However, for appropriate management of endometrial cancer, please refer to the 2021 ESGO/ESTRO/ESP evidence-based guidelines.³⁸

Ultrasound assessment

Endometrial thickness is a crucial parameter in assessing oncologic risk in postmenopausal patients with abnormal uterine bleeding. A cut-off value of 4 mm is commonly used to identify high-risk patients. When endometrial thickness exceeds this threshold, further evaluation through histological assessment is recommended to rule out endometrial pathology.⁴¹

Recognizing grayscale and power Doppler ultrasound features is important for predicting the phenotype of low- and high-risk endometrial cancer. By assessing both morphological and vascular characteristics, ultrasound provides valuable insights into tumor aggressiveness, aiding in risk stratification and preoperative assessment. Integrating these findings into the diagnostic workflow offers a non-invasive yet highly informative method for guiding treatment decisions. Recognizing distinct ultrasound patterns enables tailored surgical and therapeutic approaches, ultimately enhancing patient management and improving outcomes.⁴²

Low-risk endometrial cancers

Ultrasound characteristics of low-risk endometrial cancers (FIGO Stage IAG1–G2) typically present as hyperechoic lesions and are generally small in size. The regularity of the endometrial–myometrial junction is one of the distinguishing features, remaining well-defined and intact. Additionally, these tumors tend to have a greater amount of tumor-free myometrium, further contributing to their lower risk profile (Figure 13).

From a vascular perspective, low-risk tumors demonstrate reduced vascularization, which can manifest in various forms. Power Doppler assessment often reveals a single vessel with branching or multiple vessels originating from a focal point. The overall color Doppler signal intensity in these cases tends to be low to moderate, reflecting limited perfusion and reduced tumor angiogenesis.⁴³

High-risk endometrial cancer

On ultrasound, high-risk endometrial cancers (FIGO stage IAG3 or non-endometrioid histotype and/or FIGO Stage ≥ IB), compared with low-grade cancer, show typically greater tumor size (endometrial thickness and volume) and non-uniform echogenicity. A key morphological hallmark of these tumors is the irregularity of the endometrial–myometrial junction, indicating more aggressive infiltration into surrounding tissues (Figure 13).

Vascular characteristics further distinguish high-risk tumors from their low-risk counterparts. These tumors display increased vascularization, with multiple vessels originating from multifocal points. Power Doppler imaging frequently reveals a higher color Doppler score, indicative of moderate-to-high perfusion. This enhanced vascular network suggests a more aggressive angiogenic profile, which correlates with increased tumor aggressiveness and a poorer prognosis.⁴³

Transvaginal ultrasound is the primary modality for assessing myometrial invasion, cervical stromal infiltration and potential ovarian involvement, which are key prognostic factors influencing staging and surgical planning.

Transabdominal ultrasound should always complement transvaginal ultrasound, offering a broader assessment of extrauterine disease. The preoperative ultrasound evaluation of endometrial cancer should include a systematic assessment of lymph node metastasis and peritoneal involvement, which are crucial for accurate staging and treatment planning.

Myometrial invasion

Myometrial invasion is one of the most important independent prognostic factors. It has been demonstrated that the diagnostic performance of transvaginal ultrasound, when performed by an expert, is comparable to that of MRI.^44,45 Various ultrasonographic approaches have been proposed to evaluate myometrial invasion, encompassing both subjective assessment and objective measurement techniques, such as Gordon's method and Karlsson's method.⁴⁶

Subjective assessment refers to the ultrasound examiner's personal impression, based on the comparison between the width of the myometrium and the depth of myometrial invasion.^47,48 This method has been shown to be superior to objective assessment, with a sensitivity ranging from 56% to 100% and a specificity between 65% and 90% (Table 5).

According to Gordon’s method, depth of infiltration can be measured as the ratio of the distance between the maximum tumor depth (B) and the total myometrial thickness (A), with B/A > 50% indicating deep myometrial infiltration (Figure 14).⁶⁷ This method has been shown to have a sensitivity ranging from 68% to 100% and a specificity between 65% and 91% (Table 6).

 According to Karlsson's method, depth of myometrial infiltration can be measured as the ratio between the maximum anteroposterior (AP) diameter of the endometrial lesion (B) and the uterine AP diameter (A), with B/A > 50% indicating deep myometrial infiltration (Figure 15).⁸¹ This approach has a sensitivity and specificity comparable to those of subjective evaluation, a finding that was confirmed in a prospective multicenter cohort study.⁵¹ Moreover, a minimal tumor-free margin of less than 7 mm is a useful predictor of deep myometrial infiltration. This measurement is obtained in the plane in which the shortest distance from the tumor to the serosa is observed.⁴⁸ This method has been shown to have a sensitivity ranging from 56% to 87% and a specificity between 64% and 88% (Table 7).

Stromal infiltration

Cervical involvement in endometrial cancer occurs in about 15% of cases. Sensitivity and specificity of transvaginal ultrasound in assessing cervical involvement are 69% and 93%, respectively, while for MRI the equivalent values are 69% and 91%.⁴⁵

Subjective assessment of ultrasound images is the most effective method for predicting cervical stromal invasion in endometrial cancer, particularly in patients with G1 or G2 tumors.⁴⁸ This effectiveness is related to the dynamic nature of transvaginal examination, which involves applying pressure with the probe and observing the movement of the tumor tissue in relation to the cervical mucosa. Moreover, in cases of stromal infiltration, a loss of clear demarcation between the endometrial lesion and the cervical stroma may be observed, further supporting the diagnosis.

Objective evaluation of cervical stromal invasion relies on measuring the distance between the external cervical os and the lower tumor margin, with a validated cut-off value of ≤ 20.5 mm (Figure 16).⁵¹

Assessment of pelvic and abdominal metastases

The preoperative ultrasound evaluation of endometrial cancer should include a systematic assessment of the ovaries, lymph nodes and abdomen to detect potential pelvic or abdominal metastases. Ultrasound features that help differentiate synchronous primary cancers of the endometrium and ovary from ovarian metastases of endometrial cancer have been reported.⁸⁶ These two conditions exhibit distinct ultrasound patterns. In cases of synchronous primary cancers, the endometrial tumor is usually small, with no myometrial infiltration and minimal or no vascularization. Ovarian masses are typically unilateral, with a multilocular-solid or solid morphology. Conversely, in cases of ovarian metastases from endometrial cancer, the endometrial tumor is generally larger, shows myometrial infiltration, and exhibits high vascularization with a multiple-vessel pattern. Ovarian masses are predominantly bilateral and solid (Figure 17).⁸⁶

Lymph node metastasis in endometrial cancer is a strong prognostic factor for survival. Deep myometrial invasion (≥ 50%) is strongly associated with lymph node metastasis (Figure 18). Studies in the literature on sentinel lymph nodes have shown a more frequent involvement of external iliac lymph nodes, including obturator nodes, in both endometrioid and non-endometrioid tumors.⁸⁷

For a standardized measurement and description of lymph nodes, we recommend using VITA terminology.⁸⁸ Fischerova et al. have recently published a methodological consensus on the ultrasound assessment of lymph nodes for staging gynecological cancer, aiming to illustrate the ultrasound techniques necessary for performing an accurate preoperative workup.⁸⁹

Role of ultrasound in selecting patients for fertility sparing

Fertility-sparing treatment in patients with atypical endometrial hyperplasia or endometrial carcinoma is a viable option for young women with reproductive potential, provided they meet the selected criteria established by the ESGO/ESHRE/ESGE Guidelines.⁹⁰ This approach can be considered for patients with G1, Stage IA endometrioid endometrial carcinoma without myometrial invasion or additional risk factors. Evidence supporting fertility preservation in G2 endometrioid endometrial carcinoma or in cases with minimal myometrial infiltration remains limited; therefore, treatment decisions should be individualized. Preoperative assessment of myometrial invasion in these patients is conducted using transvaginal ultrasound performed by an expert or MRI.⁹⁰

Ultrasound examination also plays a crucial role in follow-up. After conservative treatment, which includes hysteroscopic tumor resection followed by oral progestins and/or a levonorgestrel-releasing intrauterine device, an ultrasound scan is recommended every 3 months. Endometrial histological assessment should be performed every 3–6 months via hysteroscopy, depending on imaging results.⁹⁰

ARTIFICIAL INTELLIGENCE AND RADIOMICS

In recent years, several authors have explored the role of artificial intelligence (AI) models and radiomics applied to ultrasound imaging in endometrial cancer.⁹¹ These studies have primarily focused on using AI to differentiate between benign and malignant histology, predict myometrial invasion and distinguish between high- and low-risk endometrial cancers. In the latest study, two machine learning models were developed: one based solely on clinical and ultrasound variables, which demonstrated strong performance with an area under the curve (AUC) of 0.90, sensitivity of 80% and specificity of 84% and another relying exclusively on radiomics features, achieving an AUC of 0.80, sensitivity of 59% and specificity of 86%.⁹²

Radiomics, which combines medical imaging with advanced computational analysis, enables the extraction of a vast array of quantitative features from ultrasound images. When applied to ultrasound, it enhances the assessment of morphological characteristics by capturing intricate spatial patterns, texture variations and pixel-level heterogeneity that may be imperceptible to the human eye.

These high-dimensional data could contribute to the development of predictive models that refine risk stratification, improve tumor phenotyping and support personalized management strategies. By integrating radiomics with ultrasound, a traditionally operator-dependent technique, this approach may provide a more objective and reproducible diagnostic tool, offering deeper insights into tumor biology and aggressiveness.

PRACTICE RECOMMENDATIONS

Online Study Assessment Option
All readers who are qualified doctors or allied medical professionals can automatically receive 2 Continuing Professional Development points plus a Study Completion Certificate from GLOWM for successfully answering four multiple-choice questions (randomly selected) based on the study of this chapter. Medical students can receive the Study Completion Certificate only.

 

(To find out more about the Continuing Professional Development awards programme CLICK HERE)

 

I wish to proceed with the Study Assessment for this chapter