Published online Nov 7, 2021. doi: 10.3748/wjg.v27.i41.7173
Peer-review started: April 2, 2021
First decision: June 24, 2021
Revised: June 26, 2021
Accepted: September 3, 2021
Article in press: September 3, 2021
Published online: November 7, 2021
Processing time: 217 Days and 13.6 Hours
Combined hepatocellular carcinoma (HCC) and cholangiocarcinoma (cHCC-CCA) arises in hepatic progenitor cells and are defined as a single nodule showing differentiation into HCC and intrahepatic cholangiocarcinoma (ICC) with 5-year postoperative overall survival (OS) rates ranging from 8% to 63%. There are different opinions in the literature on whether the prognosis of patients with cHCC-CCA is worse than that of patients with simple HCC or similar ICC.
Due to the poor prognosis of cHCC-CCA and absence of a promising way to predict prognosis of cHCC-CCA, the authors aimed to construct a radiomics nomogram for predicting postoperative survival of cHCC-CCA patients. This prognostic model may help guide treatment decisions for these patients.
The purpose of this study was to construct and validate a nomogram based on radiomics and clinical characteristics to predict the postoperative survival rate of patients with cHCC-CCA.
We collected the clinical data and computed tomography (CT) imaging data of patients with cHCC-CCA. Radiomics features were extracted from portal venous phase CT images using the least absolute shrinkage and selection operator Cox regression and random forest analysis. A nomogram integrating radiomics score and clinical factors was developed using multivariate Cox regression and each patient got a risk score. And patients were categorized as being at “high” or “low” risk based on their risk scores.
A total of five factors, which were Radiomics score, vascular invasion, anatomical resection, total bilirubin level, and satellite lesions, were independent predictors of prognosis and the nomogram was associated with OS more strongly than a model based on radiomics score or only clinical factors. Patients stratified as being at high risk showed a significantly shorter median OS than those stratified as being at low risk (6.1 vs 81.6 mo, P < 0.001).
This nomogram have potential usefulness in predicting postoperative survival of cHCC-CCA patients and may therefore help identify those more likely to benefit from it, which may facilitate clinical decision-making.
Considering the high AUC of this radiomics nomogram in predicting prognosis of cHCC-CCA, this prognostic model may help guide treatment decisions for these patients.