Related Experiment Video
Updated: Jul 12, 2025
10:28
Gene Regulation and Targeted Therapy in Gastric Cancer Peritoneal Metastasis: Radiological Findings from Dual Energy CT and PET/CT
Published on: January 22, 2018
11.1K
Radiomics Analyses to Predict Histopathology in Patients with Metastatic Testicular Germ Cell Tumors before
Anna Scavuzzo1, Giovanni Pasini2,3, Elisabetta Crescio4
1Department of Uro-Oncology, Instituto Nacional de Cancerologia, Universidad Autonoma de Mexico-UNAM, Mexico City 14080, Mexico.
Journal of Imaging
|October 27, 2023
Summary
This study developed a CT radiomics model to predict non-seminomatous testicular germ cell tumor (TGCT) histopathology before surgery. The model accurately identifies tumor subtypes, potentially reducing overtreatment in young patients.
Area of Science:
- Oncology
- Radiology
- Medical Imaging
Background:
- Accurate histopathology in metastatic non-seminomatous testicular germ cell tumors (TGCT) before retroperitoneal lymph node dissection (PC-RPLND) is crucial for reducing treatment morbidity in young patients.
- Addressing survivorship concerns requires precise pre-treatment assessment of tumor characteristics.
Purpose of the Study:
- To investigate the efficacy of computed tomography (CT) radiomics models integrated with clinical predictors for personalized histopathology prediction in metastatic non-seminomatous TGCT patients.
- To enable non-invasive assessment prior to PC-RPLND, aiming to optimize treatment strategies.
Main Methods:
- A retrospective study of 122 patients with metastatic non-seminomatous TGCT.
- Extraction of quantitative features from CT images using radiomics software.
- Development and validation of machine learning models (including Support Vector Machine) with 5-fold cross-validation to predict histological subtypes.
Main Results:
- The Support Vector Machine-based radiomics model achieved a high predictive performance, with an average area under the receiver operating characteristic curve (AUC) of 0.945.
- The model demonstrated robust performance in differentiating between fibrosis/necrosis, teratoma, and viable tumor.
Conclusions:
- CT-based radiomics offers a promising non-invasive tool for predicting histopathological outcomes in metastatic non-seminomatous TGCT.
- This approach may help mitigate over- or under-treatment risks in young patients, though multi-center validation is essential for clinical implementation.
