Published online May 15, 2024. doi: 10.4251/wjgo.v16.i5.1808
Peer-review started: November 10, 2023
First decision: January 30, 2024
Revised: February 2, 2024
Accepted: March 12, 2024
Article in press: March 12, 2024
Published online: May 15, 2024
Processing time: 181 Days and 10.3 Hours
Hepatocellular carcinoma (HCC) is one of the most common malignant tumors, and HCC patients have a poor prognosis. Vessels encapsulating tumor clusters (VETC) are a vascular pattern associated with a novel metastasis mechanism and have been proven to be an independent poor prognostic factor for early HCC patients.
It seems that no one has focused on predicting the VETC pattern of small HCC (sHCC; ≤ 3 cm) patients in multicenter studies.
To construct a nomogram that combines preoperative clinical parameters and image features to predict patterns of VETC and evaluate the prognosis of sHCC patients.
We collected patients with VETC and HCC status from three hospitals. Data from one hospital were used as a training set to train the prediction model, while those from the other two hospitals were used as test and validation sets, respectively. Univariate and multivariate logistic regression analyses were used to screen the independent predictive factors associated with VETC, and these factors were included to construct a model for predicting the pattern of VETC in sHCC patients. The performance of the model was evaluated using area under curve (AUC), decision curve analysis (DCA), and calibration curve. Kaplan-Meier survival analysis was performed to confirm whether the VETC status predicted by the model was associated with early recurrence in sHCC patients.
The independent predictive factors that we identified include alpha-fetoprotein_lg10, carbohydrate antigen 199 (CA199), irregular shape, non-smooth margin, and arterial peritumoral enhancement. The model for predicting VETC status, which incorporates these factors, showed good results under the evaluation of AUC, DCA, and calibration curves in the three sets. Finally, Kaplan-Meier survival analysis confirmed that the VETC pattern was associated with early recurrence in sHCC patients.
The nomogram constructed by incorporating preoperative clinical parameters and imaging features has undergone extensive validation across multiple patient centers, demonstrating strong predictive performance and having good significance for predicting postoperative recurrence.
Improving the predictive performance of VETC status in preoperative prediction of sHCC patients also requires the combination of radiomics and artificial intelligence, in order to better provide assistance for clinical treatment decision-making.