Do estrogen, progesterone, P53 and Ki67 receptor ratios determined from curettage materials in endometrioid-type endometrial carcinoma predict lymph node metastasis?

A Machine Learning Approach to Build and Evaluate a Molecular Prognostic Model for Endometrial Cancer Based on Tumour Microenvironment

Endometrial cancer (EC) incidence and the associated tumour burden have increased globally. To build a molecular expression prognostic model based on the tumour microenvironment to guide personalised treatment using a machine learning approach. Two datasets were reviewed, including a training cohort (n = 698) and a testing cohort (n = 151). All patients underwent hysterectomy ± adnexectomy ± lymph nodes dissection between December 2014 and June 2020 at the PLA General Hospital First Medical Center and received necessary and regular follow-up. We developed novel models using R software to predict factors that affect survival, such as progression-free survival and overall survival. Then, the model was optimised by evaluating the prediction efficiency in multiple dimensions. Eight hundred and forty-nine patients with EC were included in the study. Survival-related influences on EC patients were identified by univariate analysis and cox regression equations. In addition, a nomogram was visualised in conjunction with demographic characteristics and the above meaningful clinicopathological variables. Ultimately, through a comprehensive assessment, a random forest model (RF16) was developed for complementing the findings of the molecular classification of EC. The RF16 not only specifically characterises tumour molecules, but also enhances the generalizability of the model by replacing gene sequencing with immunohistochemistry. This study showed that the machine learning model (RF16) is low-cost, efficient, and clinically valuable in guiding treatment for EC patients.

Early recurrence after surgery in FIGO 2023 stage I-III endometrial cancer: characteristics and risk factors

Background

Understanding the risk factors for early recurrence is crucial for improving endometrial cancer (EC) patient outcomes.

Methods

We conducted a retrospective analysis of clinicopathological data from 473 patients diagnosed with EC at the First Affiliated Hospital of Chongqing Medical University between October 2013 and May 2019. We evaluated factors influencing early recurrence(defined as occurring within 12 months after treatment) based on 2023 International Federation of Gynecology and Obstetrics (FIGO) staging system.

Results

Among the 473 patients, 284 (60.1%) were diagnosed with stage I, 117 (24.7%) with stage II, and 72 (15.2%) with stage III. A total of 343 patients (72.5%) had non-aggressive EC, while 130 patients (27.5%) had aggressive EC. Our findings identified higher FIGO stage, lymphovascular space invasion, estrogen receptor negativity, and abnormal P53 expression as significant independent risk factors for early recurrence. Of the 473 patients, 83 (17.6%) experienced recurrence, with 44 patients (53.0%) relapsing within 12 months post-treatment. Patients with early recurrence had significantly worse prognoses compared to those with late recurrence or no recurrence(P < 0.001).

Conclusion

The identification of these risk factors is essential for developing individualized treatment plans and postoperative management strategies. Our study highlights the need for targeted therapies and intensified follow-up for high-risk patients to improve outcomes in endometrial cancer.

Clinicopathological characteristics of endometrial carcinoma with different molecular subtypes and their correlation with lymph node metastasis

Endometrial carcinoma (EC) is one of the three major malignancies of the female reproductive organs. With intense research of tumor molecular mechanisms and development of precision medicine in recent years, the traditional pathomorphological classification fails to meet the needs of clinical diagnosis and treatment for EC. This study aims to analyze the correlation of different Proactive Molecular Risk Classifier for Endometrial Cancer molecular subtypes with lymph node metastasis (LNM) and other clinical features in EC. 120 treatment-naive EC patients with surgery were enrolled in this study. The molecular subtypes of these patients were classified as follows by Proactive Molecular Risk Classifier for Endometrial Cancer (ProMisE) molecular subtyping: mismatch repair deficiency (MMRd) in 22 cases (18.33%), polymerase epsilon exonuclease domain mutation (POLE EDM) in 2 cases (1.67%), p53 wild-type (p53-wt) in 64 cases (53.33%), and p53 abnormal (p53-abn) in 32 cases (26.67%). The clinicopathological features of 120 patients were retrospectively analyzed. Statistical significance was identified among the four molecular subtypes in terms of histological classification, International Federation of Gynecology and Obstetrics (FIGO) staging, pathological grading, and LNM. Among the enrolled cases, 26 had LNM and 94 had no lymph node involvement. According to the multivariate Logistic regression analysis, p53 wt (P=0.008, OR=0.078, 95% CI: 0.012-0.510) was a protective factor for LNM in EC patients, while poorly differentiated histology (P=0.001, OR=15.137, 95% CI: 3.013-76.044) was a risk factor. ProMisE classification system, being more objective and reproducible, can provide an important reference for preoperative decision-making. The patients with p53 wt by ProMisE had a low risk of LNM in preoperative diagnostic curettage specimens, while there was a higher risk of LNM among the patients with poorly differentiated EC.

Strengthening the Combinational Immunotherapy from Modulating the Tumor Inflammatory Environment via Hypoxia-Responsive Nanogels

Despite the success of immuno-oncology in clinical settings, the therapeutic efficacy is lower than the expectation due to the immunosuppressive inflammatory tumor microenvironment (TME) and the lack of functional lymphocytes caused by exhaustion. To enhance the efficacy of immuno-oncotherapy, a synergistic strategy should be used that can effectively improve the inflammatory TME and increase the tumor infiltration of cytotoxic T lymphocytes (CTLs). Herein, a TME hypoxia-responsive nanogel (NG) is developed to enhance the delivery and penetration of diacerein and (-)-epigallocatechin gallate (EGCG) in tumors. After systemic administration, diacerein effectively improves the tumor immunosuppressive condition through a reduction of MDSCs and Tregs in TME, and induces tumor cell apoptosis via the inhibition of IL-6/STAT3 signal pathway, realizing a strong antitumor effect. Additionally, EGCG can effectively inhibit the expression of PD-L1, restoring the tumor-killing function of CTLs. The infiltration of CTLs increases at the tumor site with activation of systemic immunity after the combination of TIM3 blockade therapy, ultimately resulting in a strong antitumor immune response. This study provides valuable insights for future research on eliciting effective antitumor immunity by suppressing adverse tumor inflammation. The feasible strategy proposed in this work may solve the urgent clinical concerns of the dissatisfactory checkpoint-based immuno-oncotherapy.

Estrogen receptor and programmed death ligand-1 expression in type 1 endometrial cancer and its associated clinicopathological characteristics

BACKGROUND

This study aimed to determine the association of estrogen receptor (ER) and programmed death ligand-1 (PD-L1) expression with the clinicopathological characteristics of type 1 endometrial cancer.

MATERIALS AND METHODS

A total of 85 patients with type 1 endometrial cancer who underwent surgery at the Dr. Soetomo Hospital, Surabaya, Indonesia were retrospectively studied. Data about the age, menopausal status, body mass index, disease stage, cell differentiation, angiolymphatic invasion, myometrial invasion, and adjuvant therapy of the patients were collected from medical records. Immunohistochemistry with ER and PD-L1 antibodies was performed on all samples. The association between ER and PD-L1 expression and clinicopathological characteristics was statistically analyzed.

RESULTS

The positivity rates of ER and PD-L1 in type 1 endometrial cancer were 68.2 % and 78.5 %, respectively. ER positivity was significantly correlated with body mass index (BMI) ≥25, premenopausal status, early stage of disease, <1/2 myometrial invasion, negative nodal metastasis, and lack of adjuvant therapy. It was also associated with age <55 years, low-grade cells, and angiolymphatic invasion, but the correlation was not significant. Meanwhile, PD-L1 positivity was significantly correlated with BMI <25, menopausal status, advanced stage of disease, high-grade cells, angiolymphatic invasion, and adjuvant therapy. It was also associated with age ≥55 years and nodal metastasis, but the correlation was not significant.

CONCLUSION

ER and PDL-1 positivity is associated with the clinicopathological characteristics of type 1 endometrial cancer.

Development and validation of a nomogram involving immunohistochemical markers for prediction of recurrence in early low-risk endometrial cancer

Background

The purpose of this study was to construct a nomogram based on classical parameters and immunohistochemical markers to predict the recurrence of early low-risk endometrial cancer patients.

Methods

A total of 998 patients with early low-risk endometrial cancer who underwent primary surgical treatment were enrolled (668 in the training cohort, 330 in the validation cohort). Prognostic factors identified by univariate and multivariate analysis in the training cohort were used to construct the nomogram. Prediction performance of the nomogram was evaluated using the calibration curve, concordance index (C-index), and the time-dependent receiver operating characteristic curve. The cumulative incidence curve was used to describe the prognosis of patients in high-risk and low-risk groups divided by the optimal risk threshold of the model.

Results

In the training cohort, grade ( P = 0.040), estrogen receptor ( P < 0.001), progesterone receptor ( P = 0.001), P53 ( P = 0.004), and Ki67 ( P = 0.002) were identified as independent risk factors of recurrence of early low-risk endometrial cancer, and were used to establish the nomogram. The calibration curve showed that the fitting degree of the model was good. The C-indexes of training and validation cohorts were 0.862 and 0. 827, respectively. Based on the optimal risk threshold of the nomogram, patients were split into a high-risk group and a low-risk group. The cumulative incidence curves showed that the prognosis of the high-risk group was far worse than that of the low-risk group ( P < 0.001).

Conclusion

This nomogram, with a combination of classical parameters and immunohistochemical markers, can effectively predict recurrence in early low-risk endometrial cancer patients.

A Model to Identify Candidates for Lymph Node Dissection Among Patients With High-Risk Endometrial Endometrioid Carcinoma According to Mayo Criteria

Background

The Mayo criteria are the most widely accepted algorithm for predicting the risk of lymph node metastasis in endometrial endometrioid carcinoma (EEC). However, the clinical value of these criteria in high-risk patients is limited and inconclusive.

Methods

A total of 240 patients with EEC meeting the Mayo high-risk criteria between January 1, 2015, and December 31, 2018 were included in our study. We retrospectively collected the laboratory reports, basic clinical information, clinicopathological and immunohistochemistry (IHC) findings, and the sequences of molecular pathological markers of these patients. A nomogram for predicting the likelihood of positive lymph node status was established based on these parameters.

Results

Among the 240 patients, 17 were diagnosed with lymph node metastasis. The univariable analyses identified myometrial invasion >50%, aberrant p53 expression, microsatellite instable (MSI), and cancer antigen 125 (CA125) ≥35 U/ml as potential risk factors for lymph node metastasis. The multivariable analyses showed that aberrant p53 expression, MSI, and CA125 ≥35 U/ml were independent predictors of lymph node metastasis. The area under the curve (AUC) for the nomogram was 0.870, as compared to 0.665 for the Mayo criteria.

Conclusions

Our novel prediction model effectively identifies patients at high risk for lymphatic metastasis. This model is a promising strategy for personalized surgery in patients with high risk according to the Mayo criteria.

Analysis of immunohistochemical characteristics and recurrence after complete remission with fertility preservation treatment in patients with endometrial carcinoma and endometrial atypical hyperplasia

OBJECTIVE

To investigate the relationship between immunohistochemical characteristics and recurrence after complete remission (CR) with fertility preservation treatment in patients with endometrial cancer (EC) and endometrial atypical hyperplasia (AH).

METHODS

The clinical data and immunohistochemical results of 53 patients with EC and 68 patients with AH admitted to Peking University People's Hospital from January 2010 to January 2021 were retrospectively analyzed. Patients were divided into two groups according to whether recurrence after complete remission (CR): group 1: recurrence after CR; group 2: no recurrence after CR, for statistical analysis.

RESULTS

(1) The expression rate of ER in group 1 was lower than that in group 2, (P < 0.05). The expression rate of Ki-67 in group 1 was significantly higher than that in group 2, (P < 0.01). The expression rates of PR, P16, P53, and PTEN were not significantly different between the two groups (P > 0.05); (2) combination index ER/ Ki-67 row ROC curve analysis, there was a significant difference (P < 0.01), the best cut-off value was 3.55, sensitivity 0.730, specificity 1.000, Youden index 0.730. The 3-year RFS of high rate patients was 100%, and that of low rate patients was 42.3%, P < 0.01.

CONCLUSIONS

The expression rate of Ki-67 is of great significance in predicting the recurrence of EC after fertility preservation therapy. The best cut-off value of combination index ER/ Ki-67 (3.55) was better than a single immunohistochemical marker in predicting recurrence of EC after fertility preservation treatment.

Combining Clinicopathological Parameters and Molecular Indicators to Predict Lymph Node Metastasis in Endometrioid Type Endometrial Adenocarcinoma

Background

Lymph node metastasis (LNM) is a critical unfavorable prognostic factor in endometrial cancer (EC). At present, models involving molecular indicators that accurately predict LNM are still uncommon. We addressed this gap by developing nomograms to individualize the risk of LNM in EC and to identify a low-risk group for LNM.

Methods

In all, 776 patients who underwent comprehensive surgical staging with pelvic lymphadenectomy at the First Affiliated Hospital of Chongqing Medical University were divided into a training cohort (used for building the model) and a validation cohort (used for validating the model) according to a predefined ratio of 7:3. Logistics regression analysis was used in the training cohort to screen out predictors related to LNM, after which a nomogram was developed to predict LNM in patients with EC. A calibration curve and consistency index (C-index) were used to estimate the performance of the model. A receiver operating characteristic (ROC) curve and Youden index were used to determine the optimal threshold of the risk probability of LNM predicted by the model proposed in this study. Then, the prediction performance of different models and their discrimination abilities for identifying low-risk patients were compared.

Result

LNM occurred in 87 and 42 patients in the training and validation cohorts, respectively. Multivariate logistic regression analysis showed that histological grade (P=0.022), myometrial invasion (P=0.002), lymphovascular space invasion (LVSI) (P=0.001), serum CA125 (P=0.008), Ki67 (P=0.012), estrogen receptor (ER) (0.009), and P53 (P=0.003) were associated with LNM; a nomogram was then successfully established on this basis. The internal and external calibration curves showed that the model fits well, and the C-index showed that the prediction accuracy of the model proposed in this study was better than that of the other models (the C-index of the training and validation cohorts was 0.90 and 0.91, respectively). The optimal threshold of the risk probability of LNM predicted by the model was 0.18. Based on this threshold, the model showed good discrimination for identifying low-risk patients.

Conclusion

Combining molecular indicators based on classical clinical parameters can predict LNM of patients with EC more accurately. The nomogram proposed in this study showed good discrimination for identifying low-risk patients with LNM.

A Nomogram Model Involving Immunohistochemical Markers for Predicting the Recurrence of Stage I-II Endometrial Cancer

Background

The purpose of this study was to establish a nomogram combining classical parameters and immunohistochemical markers to predict the recurrence of patients with stage I-II endometrial cancer (EC).

Methods

419 patients with stage I-II endometrial cancer who received primary surgical treatment at the First Affiliated Hospital of Chongqing Medical University were involved in this study as a training cohort. Univariate and multivariate Cox regression analysis of screening prognostic factors were performed in the training cohort to develop a nomogram model, which was further validated in 248 patients (validation cohort) from the Second Affiliated Hospital of Chongqing Medical University. The calibration curve was used for internal and external verification of the model, and the C-index was used for comparison among different models.

Results

There were 51 recurrent cases in the training cohort while 31 cases in the validation cohort. Univariate analysis showed that age, histological type, histological grade, myometrial invasion, cervical stromal invasion, postoperative adjuvant treatment, and four immunohistochemical makers (Ki67, estrogen receptor, progesterone receptor, P53) were the related factors for recurrence of EC. Multivariate analysis demonstrated that histological type (P = 0.029), myometrial invasion (P = 0.003), cervical stromal invasion (P = 0.001), Ki67 (P < 0.001), ER (P = 0.009) and P53 expression (P = 0.041) were statistically correlated with recurrence of EC. Recurrence-free survival was better predicted by the proposed nomogram with a C-index of 0.832 (95% CI, 0.752–0.912) in the training cohort, and the validation set confirmed the finding with a C-index of 0.861 (95% CI, 0.755–0.967).

Conclusion

The nomogram model combining classical parameters and immunohistochemical markers can better predict the recurrence in patients with FIGO stage I-II EC.

Researches on the Correlation Between Estrogen and Progesterone Receptors Expression and Disease-Free Survival of Endometrial Cancer

Objective

In this study, 345 patients with endometrial carcinoma (EC) were selected to investigate the correlation between ER/PR status and the EC disease-free survival (DFS) rate.

Methods

The intensity and proportion of tumor cell expression of estrogen receptors and progesterone receptors (ER/PR) status of 345 postoperative tumor specimens in ECs were independently assessed semi-quantitatively by two pathologists using immunohistochemistry, the summed score ranged from 0 to 8 points was worked out by adding proportion score and intensity score based on the breast cancer hormone receptor immunohistochemical Allred scoring system. The association between DFS in ECs and ER/PR expression (intensity, proportion and summed score) was assessed using Cox regression analysis. Gene expression data were obtained from The Cancer Genome Atlas research network (TCGA).

Results

According to inclusion criteria, 201 type I and 144 type II EC patients were enrolled in this study. In the univariate analysis of type I endometrial carcinoma, the intensity, proportion and summed score of ER/PR status were significantly correlated with DFS. After adjusting for factors known to significantly impact survival, the influence of ER/PR status on DFS is generally decreased but the correlation is still significant. In the univariate analysis of type II endometrial carcinoma, the intensity, proportion and summed score of ER/PR status were significantly correlated with DFS. After adjusting for factors known to significantly impact survival, the influence of ER status on DFS is generally decreased, but the correlation is still significant, the effect of PR expression on DFS is not statistically significant.

Conclusion

Higher ER/PR expression status was associated with better DFS in patients with type I endometrial cancer after adjusting for known factors that significantly affect survival. In patients with type II endometrial cancer, patients with positive ER expression were significantly associated with better DFS. However, the effect of PR expression on DFS was not statistically significant.

The optimal cut-off value of immunohistochemical parameter P53 for predicting recurrence of endometrial cancer

OBJECTIVE

To explore the optimal cut-off value of immunohistochemical parameter P53 for predicting the recurrence of Stage I-III endometrial cancer.

METHODS

A total of 473 patients who were treated between October 2013 and May 2018 were retrospectively studied. Receiver operating characteristic (ROC) curves and the Youden index were used to calculate the optimal cut-off value of P53. Cox regression analysis was used to detect the association between the threshold of P53 and recurrence of endometrial cancer. Recurrence-free survival (RFS) and overall survival (OS) were exhibited by Kaplan-Meier curve.

RESULTS

The study showed that 67% was the optimal cut-off value of P53 to predict the recurrence of endometrial cancer. P53 above 67% was an independent predictor for relapse of endometrial cancer (p < 0.001). The 3-year RFS was 89.7% in the low-value group and 66.6% in the high-value group (p < 0.001), while the 3-year OS was 93.9% and 76.4%, respectively (p < 0.001). Furthermore, the 3-year RFS of patients who did not receive adjuvant chemotherapy or radiotherapy was 95.7% and 78.2% between the two groups (p < 0.001).

CONCLUSION

The optimal cut-off value of immunohistochemical parameter P53 for predicting recurrence was confirmed as 67% and a P53 index above 67% was an independent prognostic factor.

Prognostic Value of Ki67 in Patients with Stage 1-2 Endometrial Cancer: Validation of the Cut-off Value of Ki67 as a Predictive Factor

Objective

The purpose of this study was to find a cut-off value of the immunohistochemical parameter Ki67 for stage I-II endometrial cancer.

Materials and Methods

The clinicopathological data of 318 patients with stages I-II endometrial cancer who received primary surgical treatment were retrospectively analyzed. A cut-off value of Ki67 for predicting recurrence of endometrial cancer was determined by using the receiver operating characteristic curve and the Youden index. The Cox regression was performed to screen factors associated with recurrence of endometrial cancer. Based on the cut-off value of Ki67, the patients were divided into two groups, and the differences of clinicopathological parameters between the two groups were compared.

Results

The receiver operating characteristic curve showed that the optimal cut-off value of Ki67 for predicting recurrence of patients with stages I-II endometrial cancer was 38%. The multivariate Cox regression analysis demonstrated that the histotypes (P=0.012), myometrial invasion (P=0.014), cervical stromal invasion (P=0.001), Ki67 (P=0.002), estrogen receptor (ER) (P=0.045) and P53 (P=0.032) were significant prognostic predictors for recurrence of endometrial cancer. The recurrence-free survival and the disease-specific survival of patients in the high-Ki67 group (Ki67 ≥38%) were much lower than those in the low-Ki67 group (Ki67 <38%) (P=0.000, P=0.001, respectively). Among the 118 patients with early low-risk endometrial cancer who did not receive adjuvant treatment after surgery, the recurrence-free survival of patients in the high-Ki67 group was also lower than those in the low-Ki67 group (P=0.000).

Conclusion

The Ki67 was demonstrated to be a useful prognostic factor in patients with stages I-II endometrial cancer, and the Ki67 labeling index 38.0% was optimal cut-off value for predicting recurrence.

Predicting Recurrence in Endometrial Cancer Based on a Combination of Classical Parameters and Immunohistochemical Markers

Objective

The aim of this study was to establish a nomogram to predict the recurrence of endometrial cancer (EC) by immunohistochemical markers and clinicopathological parameters and to evaluate the discriminative power of this model.

Methods

The data of 473 patients with stages I–III endometrial cancer who had received primary surgical treatment between October 2013 and May 2018 were randomly split into two sets: a training cohort and a validation cohort at a predefined ratio of 7:3. Univariate and multivariate Cox regression analysis of screening prognostic factors were performed in the training cohort (n=332) to develop a nomogram model for EC-recurrence prediction, which was further evaluated in the validation cohort (n=141).

Results

Univariate analysis found that FIGO stage, histological type, histological grade, myometrial invasion, cervical stromal invasion, postoperative adjuvant treatment, and four immunohistochemical markers (Ki67, ER, PR, and p53) were associated with recurrence in EC. Multivariate analysis showed that FIGO stage, histological type, ER, and p53 were superior parameters to generate the nomogram model for recurrence prediction in EC. Recurrence-free survival was better predicted by the proposed nomogram, with a C-index value of 0.79 (95% CI 0.66–0.92) in the validation cohort.

Conclusion

This nomogram model involving immunohistochemical markers can better predict recurrence in FIGO stages I–III EC.

The combined ratio of estrogen, progesterone, Ki-67, and P53 to predict the recurrence of endometrial cancer

BACKGROUND AND OBJECTIVES

We aimed to explore the capacity of the combined ratio of biomarkers to predict the recurrence of Stage I-III endometrial cancer (EC).

METHODS

A total of 473 patients were enrolled after screening. The cut-off value of the ratio was calculated by the receiver operating characteristic curve (ROC). The univariate and multivariate Cox regression analysis was used to assess the correlation between the combined ratio and the recurrence of EC. The differences of clinicopathological parameters between the two groups divided based on the threshold were compared.

RESULT

The ROC curve showed that 0.92 was the optimal cut-off value of the ratio ([ER + PR]/[P53 + Ki67]). The multivariate analysis demonstrated that only International Federation of Gynecology and Obstetrics stage (p = .031) and the combined ratio (p = .004) were independent risk factors of recurrence. The 3-year recurrence-free survival (RFS) and overall survival of patients in the low-ratio group were 54.1% and 66.8%, respectively; while in the high-ratio group were 94.9% and 97.9%, respectively (p < .001). The 3-year RFS of 194 patients, who did not receive the adjuvant therapy, was 54.7% and 97.2% between two groups (p < .001).

CONCLUSIONS

The optimal cut-off value (0.92) of the combined ratio was demonstrated to be better to predict the recurrence of EC than a single immunohistochemical marker.

Establishment and evaluation of a risk-scoring system for lymph node metastasis in early-stage endometrial carcinoma: Achieving preoperative risk stratification

AIM

To establish a risk-scoring system for lymph node metastasis (LNM) of early-stage endometrial carcinoma (EC), and to stratify the preoperative risk of LNM.

METHODS

We retrospectively analyzed the clinical data of 507 patients diagnosed with the early-stage EC (i.e., confined to the uterine corpus). We determined the risk factors for LNM by logistic regression analysis; then constructed a simple logistic scoring system, and an additive scoring system based on the regression coefficient (β), and odds ratio, of each variable, respectively.

RESULTS

The overall rate of LNM was 9.1% (46/507). Multivariate analysis showed that preoperative serum cancer antigen 125 (CA125) ≥35 U/mL, histopathology of grade 3 and/or type II, depth of myometrial invasion ≥1/2 and positive immunostaining for Ki-67 ≥50%, were independent risk factors for LNM (P < 0.05). The simple logistic and additive scoring systems exhibited good predictive ability (area under the curve [AUC] >0.8). Based on the additive scoring system, the risk of LNM in patients with early-stage EC was classified into three groups: a low-risk group (total score: <5), an intermediate-risk group (total score: 5-10) and a high-risk group (total score: >10). The incidence of LNM differed significantly across these three groups (P < 0.05).

CONCLUSION

The risk-scoring system constructed in this study can effectively predict the risk of LNM in patients with early-stage EC, achieve preoperative risk stratification and provide a reference guideline for the use of lymphadenectomy.

Radiologists with MRI-based radiomics aids to predict the pelvic lymph node metastasis in endometrial cancer: a multicenter study

OBJECTIVE

To construct a MRI radiomics model and help radiologists to improve the assessments of pelvic lymph node metastasis (PLNM) in endometrial cancer (EC) preoperatively.

METHODS

During January 2014 and May 2019, 622 EC patients (age 56.6 ± 8.8 years; range 27-85 years) from five different centers (A to E) were divided into training set, validation set 1 (351 cases from center A), and validation set 2 (271 cases from centers B-E). The radiomics features were extracted basing on T2WI, DWI, ADC, and CE-T1WI images, and most related radiomics features were selected using the random forest classifier to build a radiomics model. The ROC curve was used to evaluate the performance of training set and validation sets, radiologists based on MRI findings alone, and with the aid of the radiomics model. The clinical decisive curve (CDC), net reclassification index (NRI), and total integrated discrimination index (IDI) were used to assess the clinical benefit of using the radiomics model.

RESULTS

The AUC values were 0.935 for the training set, 0.909 and 0.885 for validation sets 1 and 2, 0.623 and 0.643 for the radiologists 1 and 2 alone, and 0.814 and 0.842 for the radiomics-aided radiologists 1 and 2, respectively. The AUC, CDC, NRI, and IDI showed higher diagnostic performance and clinical net benefits for the radiomics-aided radiologists than for the radiologists alone.

CONCLUSIONS

The MRI-based radiomics model could be used to assess the status of pelvic lymph node and help radiologists improve their performance in predicting PLNM in EC.

KEY POINTS

• A total of 358 radiomics features were extracted. The 37 most important features were selected using the random forest classifier. • The reclassification measures of discrimination confirmed that the radiomics-aided radiologists performed better than the radiologists alone, with an NRI of 1.26 and an IDI of 0.21 for radiologist 1 and an NRI of 1.37 and an IDI of 0.24 for radiologist 2.