The model for end-stage liver disease based cancer staging systems are better prognostic models for hepatocellular carcinoma: a prospective sequential survey.

OBJECTIVE: Liver cirrhosis is a major component in staging for hepatocellular carcinoma (HCC). The model for end-stage liver disease (MELD) is a better prognostic predictor for cirrhotic patients compared to the Child-Turcotte-Pugh (CTP) system, which is a parameter in Cancer of Liver Italian Program (CLIP), Barcelona Clinic Liver Cancer (BCLC), and Japan Integrated Scoring (JIS) system.

AIM: To investigate if the MELD-based models have a better prognostic ability.

METHODS: In the MELD-based model, the CTP class was replaced with MELD score at cutoffs of <10, 10-14, and >14. The modified systems were prospectively compared with the original counterpart in 430 consecutive HCC patients. Using 6-month mortality as the end point, the area under receiver operating characteristic curve (AUC) between the original and modified system was compared sequentially on a 3-monthly basis.

RESULTS: At the final inspection, the modified CLIP and JIS system had a significantly higher AUC compared to the original system (0.92 vs 0.893 for CLIP, P < 0.018; 0.88 vs 0.842 for JIS, P= 0.002), but there was no significant difference for the BCLC system (0.848 vs 0.841, P= 0.561). Survival analysis showed modified CLIP and JIS, and to a lesser extent, modified BCLC system, had a higher homogeneity likelihood ratio and discriminatory ability linear trend, and a lower Akaike information criterion in the Cox multivariate model, indicating a better discriminatory ability for different stage categories.

CONCLUSIONS: The MELD-based CLIP and JIS staging systems have an improved predictive ability compared to the original system and are feasible models for HCC staging in the MELD era.