CT-Based Radiomics Features Combined with AFP for Predicting Vessels Encapsulating Tumor Clusters and Prognosis of...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Hepatocellular carcinoma (HCC) poses significant challenges in preoperative risk stratification.
• Accurate prediction of vessels encapsulating tumor clusters (VETC) is crucial for HCC patient management.
• Current methods for VETC assessment may be limited in non-invasive predictive capabilities.

Purpose of the Study:

• To develop and validate a computed tomography (CT)-based radiomics nomogram for the preoperative prediction of VETC in HCC.
• To assess the nomogram's ability to predict patient prognosis, specifically disease-free survival (DFS).
• To establish a reliable, non-invasive tool for stratifying HCC patients based on VETC status and prognosis.

Main Methods:

• A retrospective study of 231 HCC patients with preoperative CT scans.
• Extraction and selection of radiomics features using Least Absolute Shrinkage and Selection Operator (LASSO) regression.
• Development and validation of a nomogram integrating radiomics features and alpha-fetoprotein (AFP), evaluated using AUC, calibration curves, and decision curve analysis (DCA).

Main Results:

• The CT-based radiomics nomogram, incorporating four radiomics features and AFP, demonstrated strong predictive performance (AUC 0.782 training, 0.755 validation).
• Excellent agreement between predicted and observed outcomes was confirmed by calibration curves.
• Decision curve analysis confirmed significant clinical utility, and VETC+ patients showed significantly worse DFS.

Conclusions:

• A CT-based radiomics nomogram integrating radiomics features and AFP is a reliable tool for preoperative VETC prediction in HCC.
• This nomogram aids in stratifying HCC patients for prognosis, offering a non-invasive approach.
• The findings support the use of radiomics for improved HCC management and risk assessment.