Hepatocellular carcinoma (HCC) patients who underwent transarterial chemoembolization (TACE) have heterogeneous clinical outcomes. Accurate prognosis prediction and risk stratification are crucial for individualized treatment. We sought to develop a novel prognostic model for overall survival (OS) that incorporated contemporary clinical and laboratory factors for estimating individual prognosis.
A total of 180 HCC patients treated with TACE were used to identify the risk factors and generate prognostic models by Cox regression analyses. Model performance was evaluated by comparing it with the Tumor-Node-Metastasis (TNM) and Barcelona-Clinic Liver-Cancer (BCLC) staging systems.
A prognosis model (PI score), which consisted of neutrophil-lymphocyte ratio (NLR), γ-glutamyl transpeptidase (GGT), alpha-fetoprotein (AFP), and TNM stage, was constructed. The PI scores of each patient were calculated, and the patients were divided into subgroups based on their PI scores. The OS rate of patients in the low-risk group (PI < 0.87) was better than that of the patients in the high-risk group (PI ≥ 0.87), p < 0.001. Patients were then further divided into four stages: early stage (PI ≤ 0.49), middle stage (0.49 < PI ≤ 0.87), advanced stage (0.87 < PI ≤ 1.48), and end stage (PI > 1.48). There were statistically significant differences between the OS rates of the four groups (p < 0.001). The area under the ROC curve (AUROC) for PI score (0.746, 0.643–0.783) was higher than those of TNM (0.699, 0.620–0.763) and BCLC (0.692, 0.617–0.760).
The PI score had excellent predictive value for HCC patients undergoing TACE and was superior to TNM and BCLC.