Predicting prognosis of endometrioid endometrial adenocarcinoma on the basis of gene expression and clinical features using Random Forest

Prognostic impact of metabolic syndrome in patients with primary endometrial cancer: a retrospective bicentric study

PURPOSE

Endometrial cancer (EC) is the most common gynaecological cancer. Its incidence has been rising over the years with ageing and increased obesity of the high-income countries' populations. Metabolic syndrome (MetS) has been suggested to be associated with EC. The aim of this study was to assess whether MetS has a significant impact on oncological outcome in patients with EC.

METHODS

This retrospective study included patients treated for EC between January 2010 and December 2020 in two referral oncological centers. Obesity, arterial hypertension (AH) and diabetes mellitus (DM) were criteria for the definition of MetS. The impact of MetS on progression free survival (PFS) and overall survival (OS) was assessed with log-rank test and Cox regression analyses.

RESULTS

Among the 415 patients with a median age of 64, 38 (9.2%) fulfilled the criteria for MetS. The median follow-up time was 43 months. Patients suffering from MetS did not show any significant differences regarding PFS (36.0 vs. 40.0 months, HR: 1.49, 95% CI 0.79-2.80 P = 0.210) and OS (38.0 vs. 43.0 months, HR: 1.66, 95% CI 0.97-2.87, P = 0.063) compared to patients without MetS. Patients with obesity alone had a significantly shorter median PFS compared to patients without obesity (34.5 vs. 44.0 months, P = 0.029). AH and DM separately had no significant impact on PFS or OS (p > 0.05).

CONCLUSION

In our analysis, MetS in patients with EC was not associated with impaired oncological outcome. However, our findings show that obesity itself is an important comorbidity associated with significantly reduced PFS.

Global endometrial DNA methylation analysis reveals insights into mQTL regulation and associated endometriosis disease risk and endometrial function

Endometriosis is a leading cause of pain and infertility affecting millions of women globally. Herein, we characterize variation in DNA methylation (DNAm) and its association with menstrual cycle phase, endometriosis, and genetic variants through analysis of genotype data and methylation in endometrial samples from 984 deeply-phenotyped participants. We estimate that 15.4% of the variation in endometriosis is captured by DNAm and identify significant differences in DNAm profiles associated with stage III/IV endometriosis, endometriosis sub-phenotypes and menstrual cycle phase, including opening of the window for embryo implantation. Menstrual cycle phase was a major source of DNAm variation suggesting cellular and hormonally-driven changes across the cycle can regulate genes and pathways responsible for endometrial physiology and function. DNAm quantitative trait locus (mQTL) analysis identified 118,185 independent cis-mQTLs including 51 associated with risk of endometriosis, highlighting candidate genes contributing to disease risk. Our work provides functional evidence for epigenetic targets contributing to endometriosis risk and pathogenesis. Data generated serve as a valuable resource for understanding tissue-specific effects of methylation on endometrial biology in health and disease.

Machine Learning for Endometrial Cancer Prediction and Prognostication

Endometrial cancer (EC) is a prevalent uterine cancer that remains a major contributor to cancer-associated morbidity and mortality. EC diagnosed at advanced stages shows a poor therapeutic response. The clinically utilized EC diagnostic approaches are costly, time-consuming, and are not readily available to all patients. The rapid growth in computational biology has enticed substantial research attention from both data scientists and oncologists, leading to the development of rapid and cost-effective computer-aided cancer surveillance systems. Machine learning (ML), a subcategory of artificial intelligence, provides opportunities for drug discovery, early cancer diagnosis, effective treatment, and choice of treatment modalities. The application of ML approaches in EC diagnosis, therapies, and prognosis may be particularly relevant. Considering the significance of customized treatment and the growing trend of using ML approaches in cancer prediction and monitoring, a critical survey of ML utility in EC may provide impetus research in EC and assist oncologists, molecular biologists, biomedical engineers, and bioinformaticians to further collaborative research in EC. In this review, an overview of EC along with risk factors and diagnostic methods is discussed, followed by a comprehensive analysis of the potential ML modalities for prevention, screening, detection, and prognosis of EC patients.

Large-Scale Screening and Machine Learning for Metal–Organic Framework Membranes to Capture CO2 from Flue Gas

To combat global warming, as an energy-saving technology, membrane separation can be applied to capture CO₂ from flue gas. Metal–organic frameworks (MOFs) with characteristics like high porosity have great potential as membrane materials for gas mixture separation. In this work, through a combination of grand canonical Monte Carlo and molecular dynamics simulations, the permeability of three gases (CO₂, N₂, and O₂) was calculated and estimated in 6013 computation–ready experimental MOF membranes (CoRE–MOFMs). Then, the relationship between structural descriptors and permeance performance, and the importance of available permeance area to permeance performance of gas molecules with smaller kinetic diameters were found by univariate analysis. Furthermore, comparing the prediction accuracy of seven classification machine learning algorithms, XGBoost was selected to analyze the order of importance of six structural descriptors to permeance performance, through which the conclusion of the univariate analysis was demonstrated one more time. Finally, seven promising CoRE-MOFMs were selected, and their structural characteristics were analyzed. This work provides explicit directions and powerful guidelines to experimenters to accelerate the research on membrane separation for the purification of flue gas.

Comprehensive Analysis of the Glycolysis-Related Gene Prognostic Signature and Immune Infiltration in Endometrial Cancer

Glucose metabolic reprogramming and immune imbalance play important roles in the progression of cancers. The purpose of this study is to develop a glycolysis-related prognostic signature for endometrial cancer (EC) and analyze its relationship with immune function. The mRNA expression profiling of the glycolysis-related genes and clinical data of EC patients were downloaded from The Cancer Genome Atlas (TCGA). We identified a glycolysis-related gene prognostic signature for predicting the prognosis of EC by using The Least Absolute Shrinkage and Selection Operator (LASSO) regression and found the patients in the high-risk group had worse survival prognosis. Multivariate Cox regression analysis showed that the gene signature was an independent prognostic factor for EC. The ROC curve confirmed the accuracy of the prognostic signature (AUC = 0.730). Then, we constructed a nomogram to predict the 1–5 years survival rate of EC patients. The association between the gene signature and immune function was analyzed based on the “ESTIMATE” and “CIBERSORT” algorithm, which showed the immune and ESTIMATE scores of patients in the high-risk group were lower, while the low immune and ESTIMATE scores were associated with a worse prognosis of patients. The imbalance of immune cells was also found in the high-risk group. Further, the protein of CDK1, a gene in the signature, was found to be closely related to prognosis of EC and inhibition of CDK1 could inhibit migration and promote apoptosis of EC cells. This study reveals a link between glycolysis-related gene signature and immunity, and provides personalized therapeutic targets for EC.

Histological Grade of Endometrioid Endometrial Cancer and Relapse Risk Can Be Predicted with Machine Learning from Gene Expression Data

The tumor grade of endometrioid endometrial cancer is used as an independent marker of prognosis and a key component in clinical decision making. It is reported that between grades 1 and 3, however, the intermediate grade 2 carries limited information; thus, patients with grade 2 tumors are at risk of both under- and overtreatment. We used RNA-sequencing data from the TCGA project and machine learning to develop a model which can correctly classify grade 1 and grade 3 samples. We used the trained model on grade 2 patients to subdivide them into low-risk and high-risk groups. With iterative retraining, we selected the most relevant 12 transcripts to build a simplified model without losing accuracy. Both models had a high AUC of 0.93. In both cases, there was a significant difference in the relapse-free survivals of the newly identified grade 2 subgroups. Both models could identify grade 2 patients that have a higher risk of relapse. Our approach overcomes the subjective components of the histological evaluation. The developed method can be automated to perform a prescreening of the samples before a final decision is made by pathologists. Our translational approach based on machine learning methods could allow for better therapeutic planning for grade 2 endometrial cancer patients.

Endometrial Cancer Molecular Characterization: The Key to Identifying High-Risk Patients and Defining Guidelines for Clinical Decision-Making?

Endometrial carcinomas (EC) are the sixth most common cancer in women worldwide and the most prevalent in the developed world. ECs have been historically sub-classified in two major groups, type I and type II, based primarily on histopathological characteristics. Notwithstanding the usefulness of such classification in the clinics, until now it failed to adequately stratify patients preoperatively into low- or high-risk groups. Pieces of evidence point to the fact that molecular features could also serve as a base for better patients' risk stratification and treatment decision-making. The Cancer Genome Atlas (TCGA), back in 2013, redefined EC into four main molecular subgroups. Despite the high hopes that welcomed the possibility to incorporate molecular features into practice, currently they have not been systematically applied in the clinics. Here, we outline how the emerging molecular patterns can be used as prognostic factors together with tumor histopathology and grade, and how they can help to identify high-risk EC subpopulations for better risk stratification and treatment strategy improvement. Considering the importance of the use of preclinical models in translational research, we also discuss how the new patient-derived models can help in identifying novel potential targets and help in treatment decisions.

Identification of a Metabolism-Related Signature for the Prediction of Survival in Endometrial Cancer Patients

Objective

Endometrial cancer (EC) is one of the most common gynecologic malignancies. The present study aims to identify a metabolism-related biosignature for EC and explore the molecular immune-related mechanisms underlying the tumorigenesis of EC.

Methods

Transcriptomics and clinical data of EC were retrieved from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Common differentially expressed metabolism-related genes were extracted and a risk signature was identified by using the least absolute shrinkage and selection operator (LASSO) regression analysis method. A nomogram integrating the prognostic model and the clinicopathological characteristics was established and validated by a cohort of clinical EC patients. Furthermore, the immune and stromal scores were observed and the infiltration of immune cells in EC cells was analyzed.

Results

Six genes, including CA3, HNMT, PHGDH, CD38, PSAT1, and GPI, were selected for the development of the risk prediction model. The Kaplan-Meier curve indicated that patients in the low-risk group had considerably better overall survival (OS) (P = 7.874e-05). Then a nomogram was constructed and could accurately predict the OS (AUC = 0.827, 0.821, 0.845 at 3-, 5-, and 7-year of OS). External validation with clinical patients showed that patients with low risk scores had a longer OS (p = 0.04). Immune/stromal scores and infiltrating density of six types of immune cells were lower in high-risk group.

Conclusions

In summary, our work provided six potential metabolism-related biomarkers as well as a nomogram for the prognosis of EC patients, and explored the underlying mechanism involved in the progression of EC.