Three-dimensional conformal radiotherapy for locally advanced hepatocellular carcinoma with portal vein tumour thrombosis: evaluating effectiveness of the model for end-stage liver disease (MELD) score compared with the Child-Pugh classification.

Abstract

The purpose of this study was to retrospectively evaluate the effectiveness of the model for end-stage liver disease (MELD) score compared with the Child-Pugh classification in patients who received three-dimensional conformal radiotherapy (3D CRT) for hepatocellular carcinoma (HCC) with portal vein tumour thrombosis (PVTT) by analyzing toxicity and prognostic factors. 56 consecutive patients who had locally advanced HCC with PVTT treated by 3D CRT between September 2007 and April 2013 were retrospectively reviewed. The median survival time of all patients was 6.4 months. Receiver-operating characteristic (ROC) analysis identified MELD score = 7.5 [area under the curve (AUC) 0.81] and Child-Pugh score = 6.5 (AUC 0.86) as the best cut-off values for predicting the incidence of complications over Common Terminology Criteria for Adverse Events grade 2. There was no significant difference in the discrimination power between the MELD score and the Child-Pugh score on comparison of the two ROC curves (p = 0.17). On multivariate analysis, age, MELD score and radiotherapy dose were significant prognostic factors for overall survival (p = 0.021, 0.038 and 0.006, respectively). In contrast, the Child-Pugh classification, tumour response, PVTT response and the number of prior interventional radiologic treatments were not significant on multivariate analysis. This study showed that the best MELD score cut-off value is 7.5 and that the MELD score is a better prognostic factor than the Child-Pugh classification in 3D CRT for HCC with PVTT. The MELD score is useful for predicting the risk of severe toxicities and the prognosis of patients treated with 3D CRT for PVTT.