Radiomic features on contrast-enhanced images of the remnant liver predict the prognosis of hepatocellular carcinoma after partial hepatectomy

Background and aims Radiomic features extracted from preoperative contrast-enhanced computed tomography (CT) images have been shown to predict the prognosis of hepatocellular carcinoma (HCC). However, the prognostic role of radiomic features obtained from postoperative contrast-enhanced CT images of the remnant liver remains unclear. This study explored the prognostic value of radiomic features extracted from postoperative contrast-enhanced CT images in patients with HCC. Methods Robust radiomic features were obtained from postoperative contrast-enhanced CT images for 78 patients with primary HCC and used to construct a radiomics score. A clinical model and a combined model that integrated clinicopathological indicators and the radiomics score were established. The predictive performance of the model was assessed using the concordance index and net reclassification index. Results The postoperative radiomics score for the remnant liver was an independent prognostic factor for disease-free survival (DFS) and overall survival (OS). The combined model was not inferior to the clinical model in predicting DFS but was superior in predicting OS. The net reclassification index confirmed that the combined model was more accurate and efficient in predicting OS and DFS. The radiomics score for DFS was significantly associated with tumor thrombus in the portal vein and the postoperative neutrophil-lymphocyte ratio. The radiomics score for OS was associated with multiple tumors, microvascular invasion, and tumor thrombus in the portal vein. Conclusion Postoperative contrast-enhanced CT radiomic features of the remnant liver were valuable prognostic indicators that could potentially reflect tumor burden and postoperative inflammatory status and provide more information for clinical decision-making.