Radiomics and Radiogenomics Models to Predict Molecular Integrated Risk Classes and Prognostic Factors in Endometrial Cancer.

Radiomics and Radiogenomics Models to Predict Molecular Integrated Risk Classes and Prognostic Factors in Endometrial Cancer ID: ROMANTIC STUDY

The aim is to develop radiogenomics models to stratify patients into three main risk categories (Favorable, Intermediate, and Unfavorable) according to the ProMisE model (9) and use these models to predict the most prognostically relevant EC histopathological features (i.e. FIGO stage, degree of tumor differentiation, histotype, LVSI status, myometrial and cervical invasion, lymph node metastases).

These models would support clinicians in personalizing surgical and adjuvant treatment choice among the options considered by the international guidelines.

Inclusion Criteria:

• Pathologically confirmed diagnosis of primary endometrial cancer (endometrioid, clear cell, serous, mixed, any grade)
• FIGO stage IA-IB
• Formalin-fixed, paraffin-embedded (FFPE) tissue at the diagnosis available - Availability of preoperative MRI scans in dicom (.dcm) format
• Availability of preoperative US images in dicom (.dcm) format
• Availability of preoperative CT-scan images in dicom (.dcm) format (optional)
• Available clinical information (e.g. baseline information, surgery, adjuvant therapy, median follow up period 24 months)

Exclusion Criteria:

• Metastatic cancer to the uterus (not primary EC)
• Uterine sarcoma
• Conservative surgery
• FIGO stage > II
• Formalin-fixed, paraffin-embedded (FFPE) tissue at the diagnosis not available
• Patients without available MRI, US or CT-scan images on digital media
• Clinical information not available or incomplete
• Any other malignancy in the previous 5 years or synchronous
• Patients aged under 18 years