Related Experiment Video
Updated: Jan 11, 2026
Author Spotlight: Investigating Immune Cell Dynamics in the Tumor Microenvironment — Challenges and Innovations in Cancer Prognosis
Published on: April 12, 2024
Preoperative contrast-enhanced CT for predicting the macrotrabecular-massive subtype and its prognostic significance
Pinxiong Li1,2, Chuanyan Yang2, Liping Pan2,3
1School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China.
Background:
The macrotrabecular-massive subtype of hepatocellular carcinoma (MTM-HCC) is a special histological type of hepatocellular carcinoma (HCC), characterized by aggressive behavior and poor prognosis. Identification of MTM-HCC during pretreatment assessments may carry significant prognostic and therapeutic implications. This study aimed to develop a diagnostic prediction model for MTM-HCC using preoperative contrast-enhanced computed tomography (CECT) features and machine learning (ML) algorithms.
Methods:
Patients diagnosed with HCC who underwent liver resection surgery and preoperative CECT between October 2018 and December 2021 were retrospectively enrolled. These patients were subsequently categorized into MTM-HCC and non-MTM-HCC groups. Independent predictors of the MTM subtype were identified through multivariable logistic regression analyses. Clinical and CECT features were then selected to construct predictive models using ML algorithms. The performance of each model was evaluated using the area under the receiver operating characteristic (ROC) curve (AUC). Furthermore, Kaplan-Meier survival curves were employed to evaluate the association with early recurrence.
Results:
The final study cohort comprised 122 patients diagnosed with HCC (mean age: 53±10 years; 102 males), of which 46 tumors (37.7%) were classified as MTM-HCC. Multivariable analyses identified substantial necrosis as an independent predictor of MTM-HCC [odds ratio (OR) =2.23, 95% confidence interval (CI): 1.04-5.05; P=0.04]. The clinicoradiological histogram model, which integrated clinical imaging features and whole-tumor computed tomography (CT) histogram features, demonstrated AUC values of 0.917 (95% CI: 0.862-0.973) and 0.853 (95% CI: 0.728-0.977) in the training and test sets, respectively. Additionally, Kaplan-Meier survival curves indicated that substantial necrosis (P<0.01) was significantly associated with early recurrence.
Conclusions:
Substantial necrosis on CECT was an independent predictor of MTM-HCC and was associated with early recurrence of HCC. Meanwhile, the clinicoradiological histogram model based on preoperative CECT was shown to be useful in predicting MTM-HCC.
