Identification of endometrial cancers and atypical hyperplasia: Development and validation of a simplified system for ultrasound scoring of endometrial pattern

Postmenopausal endometrial non-benign lesion risk classification through a clinical parameter-based machine learning model

OBJECTIVE

This study aimed to develop and evaluate a machine learning model utilizing non-invasive clinical parameters for the classification of endometrial non-benign lesions, specifically atypical hyperplasia (AH) and endometrioid carcinoma (EC), in postmenopausal women.

METHODS

Our study collected clinical parameters from a cohort of 999 patients with postmenopausal endometrial lesions and conducted preprocessing to identify 57 relevant characteristics from these irregular clinical data. To predict the presence of postmenopausal endometrial non-benign lesions, including atypical hyperplasia and endometrial cancer, we employed various models such as eXtreme Gradient Boosting (XGBoost), Random Forest (RF), Logistic Regression (LR), Support Vector Machine (SVM), Back Propagation Neural Network (BPNN), as well as two ensemble models. Additionally, a test set was performed on an independent dataset consisting of 152 patients. The performance evaluation of all models was based on metrics including the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, precision, and F1 score.

RESULTS

The RF model demonstrated superior recognition capabilities for patients with non-benign lesions compared to other models. In the test set, it attained a sensitivity of 88.1% and an AUC of 0.93, surpassing all alternative models evaluated in this study. Furthermore, we have integrated this model into our hospital's Clinical Decision Support System (CDSS) and implemented it within the outpatient electronic medical record system to continuously validate and optimize its performance.

CONCLUSIONS

We have trained a model and deployed a system with high discriminatory power that may provide a novel approach to identify patients at higher risk of postmenopausal endometrial non-benign lesions who may benefit from more tailored screening and clinical intervention.

IETA Ultrasonic Features Combined with GI-RADS Classification System and Tumor Biomarkers for Surveillance of Endometrial Carcinoma: An Innovative Study

Objectives: We were the first to combine IETA ultrasonic features with GI-RADS and tumor biomarkers for the surveillance of endometrial carcinoma. The aim was to evaluate the efficacy of single IETA ultrasonography GI-RADS classification and combined tumor biomarkers in differentiating benign and malignant lesions in the uterine cavity and endometrium. Methods: A total of 497 patients with intrauterine and endometrial lesions who had been treated surgically between January 2017 and December 2021 were enrolled; all of them had undergone ultrasound examinations before surgery. We analyzed the correlation between the terms of ultrasonic signs of the uterine cavity and endometrial lesions defined by the expert consensus of IETA and the benign and malignant lesions and then classified these ultrasonic signs by GI-RADS. In addition, the tumor biomarkers CA125, CA15-3, CA19-9 and HE4 were combined by adjusting the classification. The results of the comprehensive analysis were compared with pathological results to analyze their diagnostic efficacy. Results: (1) The statistic analysis confirmed that there were seven independent predictors of malignant lesions, including thickened endometrium (premenopause ≥ 18.5 mm, postmenopause ≥ 15.5 mm), non-uniform endometrial echogenicity (heterogeneous with irregular cysts), endometrial midline appearance (not defined), the endometrial-myometrial junction (interrupted or not defined), intracavitary fluid (ground glass or "mixed" echogenicity), color score (3~4 points) and vascular pattern (focal origin multiple vessels or multifocal origin multiple vessels). (2) In traditional ultrasound GI-RADS (U-T-GI-RADS), if category 4a was taken as the cut-off value of benign and malignant, the diagnostic sensitivity, specificity, PPV, NPV and diagnostic accuracy were 97.2%, 65.2%, 44.0%, 98.8% and 72.2%, respectively, and the area under the ROC curve (AUC) was 0.812. If 4b was taken as the cut-off value, the diagnostic sensitivity, specificity, PPV, NPV diagnostic accuracy and AUC were 88.1%, 92.0%, 75.6%, 96.5% and 91.2%, 0.900, respectively. The diagnostic sensitivity, specificity, PPV, NPV diagnostic accuracy and AUC were 75.2%, 98.5%, 93.2%, 93.4%, 93.4% and 0.868, respectively, when taking category 5 as the cutoff point. In modified ultrasound GI-RADS (U-M-GI-RADS), if 4a was taken as the cut-off value, The diagnostic efficacy was the same as U-T-GI-RADS. If 4b was taken as the cut-off value, the diagnostic sensitivity, specificity, PPV, NPV, diagnostic accuracy and AUC were 88.1%, 92.3%, 76.2%, 96.5%, 91.3% and 0.902, respectively. If 4c was taken as the cutoff point, the diagnostic sensitivity, specificity, PPV, NPV diagnostic accuracy and AUC were 75.2%, 98.7%, 94.3%, 93.4%, 93.6% and 0.870, respectively. The diagnostic sensitivity, specificity, PPV, NPV diagnostic accuracy and AUC were 66.1%, 99.7%, 98.6%, 91.3%, 92.4% and 0.829, respectively, if taking category 5 as the cutoff point. (3) In the comprehensive diagnostic method of U-T-GI-RADS combined tumor biomarkers results, the AUC of class 4a, 4b and 5 as the cutoff value was 0.877, 0.888 and 0.738, respectively. The AUC of class 4a, 4b, 4c and 5 as the cutoff value in the comprehensive diagnostic method of U-M-GI-RADS combined tumor biomarkers results was 0.877, 0.888, 0.851 and 0.725, respectively. There was no significant difference in diagnostic efficiency between the two comprehensive diagnostic methods. Conclusions: In this study, no matter which diagnostic method was used, the best cutoff value for predicting malignant EC was ≥GI-RADS 4b. The GI-RADS classification had good performance in discriminating EC. The tumor biomarkers, CA125, CA19-9, CA15-3 and HE4, could improve the diagnostic efficacy for preoperative endometrial carcinoma assessment.

The Sonographic Appearance of Endometrial Intraepithelial Neoplasia

OBJECTIVES

To describe the sonographic findings of endometrial intraepithelial neoplasia (EIN), a precursor of endometrial cancer.

METHODS

Cases were found by word search of pathology database 1/2013 to 6/2019. One hundred and seventy-eight patients with ultrasound <1 year prior to biopsy were included. Medical records were searched for patient data. Two radiologists blindly classified images. Differences of opinion were decided by clinical report. Univariate and multivariate analyses were performed.

RESULTS

Median time between ultrasound and first sampling procedure was 49 days. Median age was 55 (range 28-85) years. Endometrial thickness ranged from 2 to 90 mm. Mean endometrial thickness was 13 ± 6 mm in the noncancer group and 16 ± 11 mm in the cancer group (P = .02). The endometrium was almost always heterogeneous 175/178 (98%). Cysts were almost always multiple (89/109, 82%) and >1 mm (72/109, 66%). Masses were most often >5 mm (56/105, 55%) and ill-defined (41/105, 39%). Vascularity was present in 93/178 examinations (52%) and always associated with cysts and/or mass. There were 92 cancers, 25 with invasion (including 4 with tumor extension into adenomyosis). In 47 cases, the endometrial-myometrial interface was graded as ill-defined, 39 of whom had hysterectomy. There was macroscopic cancer in 11, microscopic cancer in 4, and invasive carcinoma in 12 patients (P for invasive cancer versus other outcomes = .02). Depth of invasion was 5- >95%, with 6 cancers >50%. Multivariate analysis showed thickness, polyps, and type of bleeding as the best set of independent variables for cancer (area under the receiver operating characteristic (ROC) curve [AUC] = .75). Replacing type of bleeding with age or menopausal status had AUC of .73 and .74, respectively.

CONCLUSIONS

EIN has a variety of sonographic appearances with thickened endometrium with cysts and masses being common. Ill-definition of the endometrial-myometrial interface is a poor prognostic finding when seen in the absence of adenomyosis.

Color Doppler Imaging of Vascular Abnormalities of the Uterus

ABSTRACT

Many uterine abnormalities present clinically with bleeding encompassing a broad spectrum of patients from postmenopausal spotting to life-threatening hemorrhage. Color and spectral Doppler imaging of the pelvis is often the first crucial investigation used to quickly establish the correct etiology of the uterine bleeding and guide clinical decision making and patient management.

The Risk of Endometrial Malignancy and Other Endometrial Pathology in Women with Abnormal Uterine Bleeding: An Ultrasound-Based Model Development Study by the IETA Group

OBJECTIVES

The aim of this study was to develop a model that can discriminate between different etiologies of abnormal uterine bleeding.

DESIGN

The International Endometrial Tumor Analysis 1 study is a multicenter observational diagnostic study in 18 bleeding clinics in 9 countries. Consecutive women with abnormal vaginal bleeding presenting for ultrasound examination (n = 2,417) were recruited. The histology was obtained from endometrial sampling, D&C, hysteroscopic resection, hysterectomy, or ultrasound follow-up for >1 year.

METHODS

A model was developed using multinomial regression based on age, body mass index, and ultrasound predictors to distinguish between: (1) endometrial atrophy, (2) endometrial polyp or intracavitary myoma, (3) endometrial malignancy or atypical hyperplasia, (4) proliferative/secretory changes, endometritis, or hyperplasia without atypia and validated using leave-center-out cross-validation and bootstrapping. The main outcomes are the model's ability to discriminate between the four outcomes and the calibration of risk estimates.

RESULTS

The median age in 2,417 women was 50 (interquartile range 43-57). 414 (17%) women had endometrial atrophy; 996 (41%) had a polyp or myoma; 155 (6%) had an endometrial malignancy or atypical hyperplasia; and 852 (35%) had proliferative/secretory changes, endometritis, or hyperplasia without atypia. The model distinguished well between malignant and benign histology (c-statistic 0.88 95% CI: 0.85-0.91) and between all benign histologies. The probabilities for each of the four outcomes were over- or underestimated depending on the centers.

LIMITATIONS

Not all patients had a diagnosis based on histology. The model over- or underestimated the risk for certain outcomes in some centers, indicating local recalibration is advisable.

CONCLUSIONS

The proposed model reliably distinguishes between four histological outcomes. This is the first model to discriminate between several outcomes and is the only model applicable when menopausal status is uncertain. The model could be useful for patient management and counseling, and aid in the interpretation of ultrasound findings. Future research is needed to externally validate and locally recalibrate the model.

The Value of Shear Wave Elastography in Predicting the Risk of Endometrial Cancer and Atypical Endometrial Hyperplasia

OBJECTIVES

To evaluate shear wave elastography (SWE) technology diagnosis value of endometrial cancer (EC) and atypical endometrial hyperplasia (AEH), and to establish predictive logistic regression models for the diagnosis of EC and AEH.

METHODS

Clinical information collection, transvaginal conventional ultrasonography, and SWE check were performed on 122 patients, who were perimenopausal or postmenopausal vaginal bleeding with ≥4.5 mm thick endometrium. The maximal (Emax) and mean (Emean) of Young's modulus for the endometrium were obtained. Using pathology as the gold standard, ROC curves were plotted to evaluate Young's modulus on the diagnostic effectiveness of EC and AEH. Single-factor analysis and bivariate logistic regression methods were applied to assess the clinical variables, transuaginal conventional ultrasonography variables, and Young's modulus on the identification of EC and AEH.

RESULTS

Out of 122 cases of endometrial lesions, 85 cases were benign lesions, and the remaining 37 cases were EC and AEH. The Emax and Emean for the benign group were 29.80 ± 11.40 and 17.96 ± 8.05 kPa, respectively. The Emax and Emean values for EC and AEH group were 59.49 ± 16.95 and 38.46 ± 17.10 kPa, respectively. Emax and Emean for both groups were statistically significant, with p <.001. In the logistical regression analysis, endometrial thickness, Color score, and Young's modulus were identified as independent risk factors for EC and AEH.

CONCLUSIONS

SWE technology plays an important role in the diagnosis of EC and AEH, and the diagnostic effectiveness would be higher when combined with conventional ultrasonography.

Combination IETA Ultrasonographic Characteristics Simple Scoring Method With Tumor Biomarkers Effectively Improves the Differentiation Ability of Benign and Malignant Lesions in Endometrium and Uterine Cavity

Objectives

To evaluate International Endometrial Tumor Analysis (IETA) ultrasonographic characteristics simple scoring method and tumor biomarkers for the diagnosis of uterine cavity and endometrial lesions.

Methods

We classified and scored the normalized description of IETA ultrasonic characteristics, according to IETA expert consensus literature, previous IETA-related research articles, and the previous research experience of this project group. We conducted a retrospective analysis of the ultrasound images of 594 patients enrolled from January 2017 to June 2020, scored them item by item, and finally calculated the total score of each case. Meanwhile, we combined the results of seven tumor biomarkers. The objective was to evaluate the sensitivity, specificity, coincidence rate, and the area under receiver operating characteristic (ROC) curve of IETA ultrasonographic characteristics simple scoring method and tumor biomarkers for benign and malignant uterine cavity or endometrial lesions. The diagnostic efficiency between the combined method and the single method was compared.

Results

A total of 594 cases were confirmed by postoperative pathology or surgery records, including 475 benign lesions and 119 malignant lesions. In the simple ultrasound scoring method, the average score of benign lesions was 3.879 ± 1.279 and that of malignant lesions was 9.676 ± 4.491. If ≥6.5 points was taken as the cutoff value for the judgment of malignant lesions, the sensitivity, specificity, coincidence rate, and the area under receiver operating characteristic (ROC) curve (AUC) were 76.5%, 96.0%, 92.1%, and 0.935, respectively. The difference in tumor antigen 19-9 (CA19-9) and human epididymal protein 4 (HE4) between benign and malignant lesions was statistically significant (all p ≤ 0.01). The other five tumor biomarkers (CA125, CA15-3, SCC-Ag, AFP, and CEA) showed no statistically significant difference in benign and malignant lesions. If the value of CA19-9 ≥13.96 U/ml was taken as cutoff value, the sensitivity, specificity, and coincidence rate of the diagnosis of endometrial benign and malignant lesions were 54.8%, 74.7%, and 70.7%, respectively, and the AUC was 0.620. If the value of HE4 ≥ 39.075 pmol/L was taken as cutoff point, the sensitivity, specificity, coincidence rate, and AUC were 77.4%, 67.9%, 69.8%, and 0.796, respectively. The sensitivity was increased to 97.6% and the AUC was 0.939 when IETA ultrasound characteristics simple scoring method combined CA19-9 and HE4 in parallel test.

Conclusions

In IETA ultrasound characteristics simple scoring method, with ≥6.5 points as the cutoff value, it could quickly and accurately assess the benign and malignant in uterine cavity and endometrial lesions, with high diagnostic value. The diagnostic efficacy of seven tumor biomarkers was all mediocre. Combining with these two methods, the comprehensive diagnosis could improve sensitivity and accuracy and reduce the risk of missed diagnosis.

Quality of life and sexuality of postmenopausal women with intrauterine pathologies: a recommended three-step multidisciplinary approach focusing on the role of hysteroscopy

Intrauterine pathologies are common in postmenopausal women and clinicians must identify signs and symptoms accurately to provide the adequate diagnosis and treatment. The quality of life (QoL) and sexuality of women are important outcomes to be considered to provide adequate clinical management of the postmenopausal patient with gynecologic pathologies. The aim of this paper is to propose a simple and replicable three-step multidisciplinary approach to evaluate the psychological outcomes of postmenopausal women with intrauterine pathologies, focusing on the role of hysteroscopy. In particular, the article describes three evaluation steps of those psychological outcomes corresponding to three fundamental moments of the patient's diagnostic and therapeutic path: the initial symptoms, diagnosis, and treatment. In our viewpoint, the standard use of such a protocol might considerably improve the QoL of postmenopausal patients undergoing hysteroscopic procedures due to intrauterine pathologies.

A new risk-scoring system to predict Xp11.2 translocation renal cell carcinoma in adults

Objective

The objective was to derive and validate a practical scoring system for preoperative diagnosis of Xp11.2 translocation renal cell carcinoma (RCC) in adults.

Methods

Epidemiology, symptomatology, and imaging methods were correlated between patients with common RCC and those with Xp11.2 translocation RCC using a derivation study (N = 6352) and a validation study (N = 127). Univariate analysis of risk factors was performed to derive a scoring system to predict the occurrence of Xp11.2 translocation RCC in adults. The Hosmer–Lemeshow goodness-of-fit test and receiver operating characteristic (ROC) curve were used to validate the scoring system.

Results

Based on odd ratios, three low-risk factors (sex, gross haematuria, and intratumoural calcification) and three high-risk factors (age, unenhanced computed tomography density, and enhancement pattern) were given weighted scores of 1 and 2, respectively. Patients who scored 3 to 5 points underwent an additional magnetic resonance imaging examination. The final scoring system had a sensitivity of 81.0% and a specificity of 98.0%.

Conclusion

We established a practical scoring system for the preoperative diagnosis of Xp11.2 translocation RCC in adults, which can be optimised through further clinical findings in the future.

Diagnostic value of endometrial volume and flow parameters under 3D ultrasound acquisition in combination with serum CA125 in endometrial lesions

OBJECTIVE

This study aims to discuss the differential diagnosis value of endometrial volume and flow parameters in combination with serum carbohydrate antigen 125 (CA125) in endometrial benign and malignant lesions.

MATERIALS AND METHODS

The data of 250 patients with endometrial lesions were retrospectively analyzed. Carbohydrate antigen 125 (CA125) was determined before the operation. The morphology, hemodynamics, volume and flow parameters of the endometrium were measured by transvaginal three-dimensional-power Doppler angiography (3D-PDA). The endometrial volume (EV), 3D-PDA vascular index (VI), flow index (FI) and vascularization flow index (VFI) were calculated using the virtual organ computer-aided analysis software (VOCAL).

RESULTS

According to the pathological results, 202 patients (80.8%) had benign endometrial lesions and 48 patients (19.2%) had endometrial cancer (EC). The endometrium of EC patients was thicker (15.64 ± 7.26 mm vs. 9.24 ± 5.06 mm, P < 0.001), the endometrial volume was larger (9.23 ± 4.08 ml vs. 2.26 ± 3.42 ml, P < 0.001), and the flow parameters VI, FI and VFI were higher, when compared to those of benign lesions (P < 0.001). The area under the receiver operating characteristic curve (AUROCC) of VI receptors was 0.86, while the AUC of endometrial thickness (ET) was only 0.66. Therefore, the best variable for distinguishing benign and malignant endometrial lesions was VI. The level of CA125 in the EC group significantly increased (40.57 ± 17.45 vs. 17.87 ± 7.64, P < 0.001), and the level of CA125 increased (P < 0.05) with the increase in clinical grade, degree of tumor differentiation, and pelvic lymph node metastasis (P < 0.05). However, the difference in myometrial invasion was not statistically significant (P > 0.05).

CONCLUSION

Transvaginal 3D-PDA can clearly show the morphological and hemodynamic characteristics of endometrial lesions, and assist in the detection of EC in combination with serum CA125. This may have important clinical application value.

Risk Assessment of Endometrial Hyperplasia or Endometrial Cancer with Simplified Ultrasound-Based Scoring Systems

BACKGROUND

Abnormal uterine bleeding (AUB) represents a common diagnostic challenge, as it might be related to both benign and malignant conditions. Endometrial cancer may not be detected with blind uterine cavity sampling by dilatation and curettage or suction devices. Several scoring systems using different ultrasound image characteristics were recently proposed to estimate the risk of endometrial cancer (EC) in women with AUB.

AIM

The aim of the present study was to externally validate the predictive value of the recently proposed scoring systems including the Risk of Endometrial Cancer scoring model (REC) for EC risk stratification.

MATERIAL AND METHODS

It was a retrospective cohort study of women with postmenopausal bleeding. From June 2012 to June 2020 we studied a group of 394 women who underwent standard transvaginal ultrasound examination followed by power Doppler intrauterine vascularity assessment. Selected ultrasound features of endometrial lesions were assessed in each patient.

RESULTS

The median age was 60.3 years (range ± 10.7). The median body mass index (BMI) was 30.4 (range ± 6.0). Histological examination revealed 158 cases of endometrial hyperplasia (EH) and 236 cases of EC. Of the studied ultrasound endometrial features, the highest areas under the curve (AUCs) were found for endometrial thickness (ET) (AUC = 0.76; 95% CI: 0.71-0.81) and for interrupted endomyometrial junction (AUC = 0.70, 95% CI: 0.65-0.75). Selected scoring systems presented moderate to good predictive performance in differentiating EC and EH. The highest AUC was found for REC model (AUC = 0.75, 95% CI: 0.70-0.79) and for the basic model that included ET, Doppler score and interrupted endometrial junction (AUC = 0.77, 95% CI: 0.73-0.82). REC model was more accurate than other scoring systems and selected single features for differentiating benign hyperplasia from EC at early stages, regardless of menopausal status.

CONCLUSIONS

New scoring systems, including the REC model may be used in women with AUB for more efficient differentiation between benign and malignant conditions.

Multiparametric transvaginal ultrasound in the diagnosis of endometrial cancer in post-menopausal bleeding: diagnostic performance of a transvaginal algorithm and reproducibility amongst less experienced observers

OBJECTIVE

(a) To comparatively evaluate the performance of grayscale ultrasound features, power Doppler (PD) blood flow characteristics, and gel infusion sonography (GIS) in diagnosing endometrial cancer during real-time examination, (b) to compare the performance of real-time diagnosis of endometrial cancer by experienced observers with offline analysis by blinded observers using similar sonographic criteria during review of cine loop clips.

METHODS

152 females with post-menopausal bleeding (PMB) had ET ≥ 4 mm at first-line ultrasound were included. Two experienced radiologists evaluated endometrial patterns at real-time evaluation (grayscale ultrasound, PD, and GIS), then examinations were stored as video clips for later evaluation by two less-experienced radiologists. The reference standard was hysteroscopy (HY) and/or hysterectomy with the histopathological examination. The area under (AUC) the receiver operating characteristic (ROC) curve was calculated to assess the diagnostic performance for the prediction of endometrial cancer.

RESULTS

Among 152 females with ET ≥ 4 mm at first line TVUS, 88 (57.9%) patients had endometrial cancer on final pathologic analysis. Real-time ultrasound criteria (ET ≥ 5 mm with the presence of irregular branching endometrial blood vessels or multiple vessels crossing EM or areas with densely packed color-splash vessels with non-intact or interrupted EMJ at the grayscale ultrasound and/or GIS) correctly diagnosed 95% of endometrial cancers with 92% diagnostic efficiency.There is comparable accuracy of real-time evaluation (96%) and offline analysis (92%) after the exclusion of poor quality videos from the analysis. The diagnostic criteria showed good to an excellent agreement between real-time ultrasound and offline analysis.

CONCLUSION

When real-time ultrasound is performed with good technique, utilizing multiple parameters, it is possible to diagnose endometrial cancer with a high degree of accuracy and reproducibility.

ADVANCES IN KNOWLEDGE

when real-time ultrasound is performed with good technique, utilizing multiple parameters, it is possible to diagnose endometrial cancer with a high degree of accuracy and reproducibility.

Clinical application of red cell distribution width, mean platelet volume, and cancer antigen 125 detection in endometrial cancer

Background

Red cell distribution width (RDW) and mean platelet volume (MPV) are considered to be associated with tumors. We investigated the diagnostic value of RDW, MPV, and cancer antigen 125 (CA125), alone or in combination, in the diagnosis of endometrial cancer and endometrial hyperplasia.

Methods

This study included 144 patients with endometrial cancer (stage I: 32; II: 42; III: 48; and IV: 22), 104 patients with endometrial hyperplasia, and 80 healthy control subjects. The whole blood cell parameters were analyzed by a Mindray Blood Cell Analyzer (CAL8000), whereas CA125 was analyzed using an Architect i2000 Analyzer (Abbott).

Results

Significant differences in RDW, MPV, and CA125 level were observed in the endometrial cancer, endometrial hyperplasia, and control groups (P < .05). Red cell distribution width was positively correlated (r = .735) whereas MPV was negatively correlated with (r = −.736) endometrial cancer staging. The area under the receiver operating characteristic curve of the combined diagnosis of endometrial cancer based on RDW, MPV, and CA125 was 0.924 (95% CI: 0.881‐0.955). The sensitivity and specificity of the combined diagnosis were larger than those of the independent detections involving RDW, MPV, and CA125.

Conclusions

The combination of RDW, MPV, and CA125 can improve the differential diagnosis of endometrial cancer and endometrial hyperplasia.