Objective Ultrastaging is accurate in detecting nodal metastases, but increases costs and may not be necessary in certain low-risk subgroups. In this study we examined the risk of nodal involvement detected by sentinel lymph node (SLN) biopsy in a large population of apparent early-stage endometrial cancer and stratified by histopathologic characteristics. Furthermore, we aimed to identify a subgroup in which ultrastaging may be omitted. Methods We retrospectively included patients who underwent SLN (with bilateral mapping and no empty nodal packets on final pathology) ± systematic lymphadenectomy for apparent early-stage endometrial cancer at two referral cancer centers. Lymph node status was determined by SLN only, regardless of non-SLN findings. The incidence of macrometastasis, micrometastasis, and isolated tumor cells (ITC) was measured in the overall population and after stratification by histotype (endometrioid vs serous), myometrial invasion (none, <50%, ≥50%), and grade (G1, G2, G3). Results Bilateral SLN mapping was accomplished in 1570 patients: 1359 endometrioid and 211 non-endometrioid, of which 117 were serous. The incidence of macrometastasis, micrometastasis, and ITC was 3.8%, 3.4%, and 4.8%, respectively. In patients with endometrioid histology (n=1359) there were 2.9% macrometastases, 3.2% micrometastases, and 5.3% ITC. No macro/micrometastases and only one ITC were found in a subset of 274 patients with low-grade (G1-G2) endometrioid endometrial cancer without myometrial invasion (all <1%). The incidence of micro/macrometastasis was higher, 2.8%, in 708 patients with low-grade endometrioid endometrial cancer invading <50% of the myometrium. In patients with serous histology (n=117), the incidence of macrometastases, micrometastasis, and ITC was 11.1%, 6.0%, and 1.7%, respectively. For serous carcinoma without myometrial invasion (n=36), two patients had micrometastases for an incidence of 5.6%. Conclusions Ultrastaging may be safely omitted in patients with low-grade endometrioid endometrial cancer without myometrial invasion. No other subgroups with a risk of nodal metastasis of less than 1% have been identified.

Objectives Transvaginal ultrasound is typically the initial diagnostic approach in patients with postmenopausal bleeding for detecting endometrial atypical hyperplasia/cancer. Although transvaginal ultrasound demonstrates notable sensitivity, its specificity remains limited. The objective of this study was to enhance the diagnostic accuracy of transvaginal ultrasound through the integration of artificial intelligence. By using transvaginal ultrasound images, we aimed to develop an artificial intelligence based automated segmentation model and an artificial intelligence based classifier model. Methods Patients with postmenopausal bleeding undergoing transvaginal ultrasound and endometrial sampling at Mayo Clinic between 2016 and 2021 were retrospectively included. Manual segmentation of images was performed by four physicians (readers). Patients were classified into cohort A (atypical hyperplasia/cancer) and cohort B (benign) based on the pathologic report of endometrial sampling. A fully automated segmentation model was developed, and the performance of the model in correctly identifying the endometrium was compared with physician made segmentation using similarity metrics. To develop the classifier model, radiomic features were calculated from the manually segmented regions-of-interest. These features were used to train a wide range of machine learning based classifiers. The top performing machine learning classifier was evaluated using a threefold approach, and diagnostic accuracy was assessed through the F1 score and area under the receiver operating characteristic curve (AUC-ROC). Results 302 patients were included. Automated segmentation–reader agreement was 0.79±0.21 using the Dice coefficient. For the classification task, 92 radiomic features related to pixel texture/shape/intensity were found to be significantly different between cohort A and B. The threefold evaluation of the top performing classifier model showed an AUC-ROC of 0.90 (range 0.88–0.92) on the validation set and 0.88 (range 0.86–0.91) on the hold-out test set. Sensitivity and specificity were 0.87 (range 0.77–0.94) and 0.86 (range 0.81–0.94), respectively. Conclusions We trained an artificial intelligence based algorithm to differentiate endometrial atypical hyperplasia/cancer from benign conditions on transvaginal ultrasound images in a population of patients with postmenopausal bleeding.

Endometrial cancer (EC) is the most diagnosed gynecologic malignancy, and its incidence and mortality are increasing. The prognosis is highly dependent on the disease spread. Surgical staging includes retroperitoneal evaluation to detect potential lymph node metastases. In recent years, systematic lymphadenectomy has been replaced by sentinel lymph node (SLN) biopsy and ultrastaging, allowing for the detection of macrometastases, micrometastases, and isolated tumor cells (ITCs). Micrometastases and ITCs have been grouped as low-volume metastases (LVM). The reported prevalence of LVM in studies enrolling more than one thousand patients with apparent early-stage EC ranges from 1.9% to 10.2%. Different rates of LVM are observed when patients are stratified according to disease characteristics and their risk of recurrence. Patients with EC at low risk for recurrence have low rates of LVM, while intermediate- and high-risk patients have a higher likelihood of being diagnosed with nodal metastases, including LVM. Macro- and micrometastases increase the risk of recurrence and cause upstaging, while the clinical significance of ITCs is still uncertain. A recent meta-analysis found that patients with LVM have a higher relative risk of recurrence [1.34 (95% CI: 1.07–1.67)], regardless of adjuvant treatment. In a retrospective study on patients with low-risk EC and no adjuvant treatment, those with ITCs had worse recurrence-free survival compared to node-negative patients (85.1%; CI 95% 73.8–98.2 versus 90.2%; CI 95% 84.9–95.8). However, a difference was no longer observed after the exclusion of cases with lymphovascular space invasion. There is no consensus on adjuvant treatment in ITC patients at otherwise low risk, and their recurrence rate is low. Multi-institutional, prospective studies are warranted to evaluate the clinical significance of ITCs in low-risk patients. Further stratification of patients, considering histopathological and molecular features of the disease, may clarify the role of LVM and especially ITCs in specific contexts.