Related Experiment Video
Updated: Jul 1, 2025
06:54
MR Molecular Imaging of Prostate Cancer with a Small Molecular CLT1 Peptide Targeted Contrast Agent
Published on: September 3, 2013
11.2K
Radiomic Analysis of Contrast-Enhanced CT Predicts Glypican 3-Positive Hepatocellular Carcinoma
Shifang Sun1,2, Shungen Xiao3, Zhen Jiang1
1School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, China.
Current Medical Imaging
|March 11, 2024
Summary
This study developed a contrast-enhanced CT radiomics model to predict Glypican 3 (GPC3) expression in Hepatocellular Carcinoma (HCC). The combined model, incorporating radiomics score and Alpha-Fetoprotein (AFP) levels, effectively predicts GPC3-positive HCC preoperatively.
Area of Science:
- Radiology
- Oncology
- Medical Imaging
Background:
- Glypican 3 (GPC3) expression in Hepatocellular Carcinoma (HCC) indicates a poorer prognosis and is an emerging immunotherapeutic target.
- Accurate preoperative diagnosis of GPC3-positive HCC is crucial for guiding systemic therapy.
- Dynamic contrast-enhanced CT is a widely used imaging modality for HCC diagnosis.
Purpose of the Study:
- To develop and validate a radiomics model using contrast-enhanced CT to predict GPC3 expression in HCC.
- To evaluate the predictive performance of the radiomics model compared to clinical factors.
Main Methods:
- Retrospective analysis of 141 pathologically confirmed HCC patients.
- Extraction of radiomics features from the Artery Phase (AP) of contrast-enhanced CT scans.
- Construction of a radiomics score (Rad-score) using LASSO regularization and a combined model with clinical risk factors (serum Alpha-Fetoprotein - AFP).
- Assessment of predictive performance using Receiver Operating Characteristic (ROC) curve analysis and Decision Curve Analysis (DCA).
Main Results:
- A 5-feature AP radiomics model was constructed. The AP radiomics model outperformed the AFP model in training cohorts but not validation cohorts.
- The combined model, integrating AP Rad-score and AFP levels, demonstrated improved predictive performance (AUC 0.867-0.895) compared to the AFP model (AUC 0.651-0.718) in both cohorts.
- Decision Curve Analysis indicated that the combined model provided greater net benefit than the AP radiomics model at probability thresholds above 60%.
Conclusions:
- A combined model incorporating AP Rad-score and serum AFP levels, derived from contrast-enhanced CT, can effectively predict GPC3-positive expression in HCC preoperatively.
- This approach aids in identifying patients who may benefit from GPC3-targeted immunotherapies.
- Radiomics analysis of contrast-enhanced CT offers a valuable non-invasive tool for preoperative assessment of HCC biomarkers.
