Non-Invasive Assessment of Complete Regression in Endometrial Cancer Patients Undergoing Fertility Preservation Using...

Area of Science:

• Oncology
• Radiology
• Genomics

Background:

• Fertility-preserving treatment (FPT) is crucial for young women with atypical endometrial hyperplasia (AEH) or early endometrial cancer (EC).
• Current methods for assessing treatment response (complete regression) are invasive.
• There is a need for non-invasive predictive tools for FPT outcomes.

Purpose of the Study:

• To develop and validate a non-invasive tool using radiomics and molecular profiling to predict complete regression (CR) in AEH/EC patients undergoing FPT.
• To identify imaging biomarkers associated with treatment outcomes.

Main Methods:

• Retrospective analysis of 146 patients with AEH or early EC receiving FPT.
• Extraction of radiomic features from MRI scans.
• Machine learning to build a radiomics signature and a radiomics-clinical nomogram.
• Validation in independent cohorts.
• Single-cell and bulk RNA sequencing for molecular profiling.

Main Results:

• The radiomics-clinical nomogram achieved high predictive performance (AUC 0.963 training, 0.986 validation).
• Patients with high radiomics scores had significantly lower CR rates.
• Molecular profiling revealed immune alterations (reduced CD8+ T-cells, increased M2 macrophages) and metabolic pathway upregulation in tumors with poor response.

Conclusions:

• A non-invasive radiomics-based approach can accurately predict treatment outcomes in AEH/EC patients undergoing FPT.
• This tool can aid in personalizing FPT strategies.
• Radiomics signatures correlate with tumor microenvironment characteristics, including immune suppression.