A Prediction Model for Preoperative Risk Assessment in Endometrial Cancer Utilizing Clinical and Molecular Variables

Abstract

The utility of comprehensive surgical staging in patients with low risk disease has been questioned. Thus, a reliable means of determining risk would be quite useful. The aim of our study was to create the best performing prediction model to classify endometrioid endometrial cancer (EEC) patients into low or high risk using a combination of molecular and clinical-pathological variables. We then validated these models with publicly available datasets. Analyses between low and high risk EEC were performed using clinical and pathological data, gene and miRNA expression data, gene copy number variation and somatic mutation data. Variables were selected to be included in the prediction model of risk using cross-validation analysis; prediction models were then constructed using these variables. Model performance was assessed by area under the curve (AUC). Prediction models were validated using appropriate datasets in The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. A prediction model with only clinical variables performed at 88%. Integrating clinical and molecular data improved prediction performance up to 97%. The best prediction models included clinical, miRNA expression and/or somatic mutation data, and stratified pre-operative risk in EEC patients. Integrating molecular and clinical data improved the performance of prediction models to over 95%, resulting in potentially useful clinical tests.

Differential DNA methylation in high-grade serous ovarian cancer (HGSOC) is associated with tumor behavior

“…4000 genome sequencing platform employing 150 bp paired-end sequencing by synthesis chemistry was used. All library preparation and sequencing procedures were performed in the Genome Core Facility of the University of Iowa Institute of Human Genetics 27 .…”

Population Substructure Has Implications in Validating Next-Generation Cancer Genomics Studies with TCGA

“…Rather, our aim was to assess the genetic background of patients with endometrial and ovarian cancer that we diagnose and treat, and compare them with the genetic background of patients with the same cancer types in TCGA. The motivation of this study stemmed from the need to validate a prediction model of clinical outcomes that integrated clinical, pathological, and diverse molecular data: gene and miRNA expression, gene copy number and somatic mutations [ 2 , 3 ]. The validation of these prediction models in TCGA datasets was not ideal, so we wanted to investigate possible reasons for this discordance.…”

“…Although TCGA and other publicly available datasets have made these validations much simpler, the limitations of doing so have yet to be fully addressed. For example, in our previous studies designed to predict clinical outcomes by integrating clinical, pathological and molecular features of patients with cancer, we found that the best prediction models, developed using our internal patient cohort (University of Iowa), performed 10–20 percentage points worse in TCGA datasets [ 2 , 3 ]. Based on the characteristics of our population (northern European origin), it was obvious that our patients’ backgrounds seemed to be more homogeneous than the population from which TCGA derived their datasets.…”

Bacterial, Archaea, and Viral Transcripts (BAVT) Expression in Gynecological Cancers and Correlation with Regulatory Regions of the Genome

“…Of the 193 patients identified in the original HGSC panel, we were able to obtain 112 tumor tissues with sufficient RNA yield and quality for analysis [ 16 ]. Similarly, of the 126 patients identified in the original endometrial endometrioid cancer panel, we were able to obtain 62 primary tumor tissues with sufficient RNA yield and quality for analysis [ 17 ]. From the 20 original normal fallopian tube samples, 12 had sufficient RNA yield and quality for analysis [ 16 ] ( Figure 1 ).…”