Validation for retreatment strategy and survival analysis after second session of transarterial chemoembolization in patient with hepatocellular carcinoma

초록

Background & Aims: Transarterial chemoembolization (TACE) is the recommended modality for hepatocellular carcinoma (HCC) patients with intermediate stage of Barcellona Clinic Liver Cancer (BCLC) system. Assessment for retreatment with TACE (ART) score was developed to identify a group of patients who may have a poor prognosis after second TACE session. We aimed to validate ART score and, if necessary, to evaluate survival factors in our HCC patients for drawing another prognostic score. Methods: Between January 2004 and December 2009, 182 patients who were first diagnosed as HCC and treated with more than 2 sessions of TACE within ≤90 days in Korea University Guro Hospital were enrolled for validation of ART score and analysis of survival factors. The 94 patients in Soonchunhyang University Bucheon Hospital, were included for external validation of the new scoring system. We collected demographic and clinical data at baseline and 1 day before second session of TACE. Tumor response was assessed according to modified RECIST criteria. OS was defined as time from second TACE to last day of follow up or death. We performed survival analysis using Kaplan-Meier method and Cox regression. Results: There was no survival difference between ART score ≥2.5 group and ART score 0-1.5 group in our patients (median OS : 37.0 months vs. 29.0 months, p=0.236). Therefore, we performed univariate analysis to find survival factors from second session of TACE using various baseline and pre-second TACE data including three factors of ART score. According to univariate analysis, age, tumor size, tumor number, BCLC stage, AFP prior to first and second TACE session, AFP response, Child-Pugh class prior to second TACE, and serum creatinine increase > 25%, platelet count decrease > 25% were entered into a stepwise cox regression model. Age, AFP prior to second TACE session, BCLC stage, Child-Pugh class and tumor number (AABCN) were independent factors for survival (all p< 0.05). We developed a new AABCN score (Range, 0-13.5) using independent survival parameters. AABCN score ≥ 7.5 significantly predicted poor survival after second TACE (HR 3.115, 95% CI 2.0-4.9, P < 0.001). These results were confirmed in the external validation cohort (HR 1.915, 95% CI 1.1-3.3, P = 0.017). Conclusions: ART score was not a valuable method to select a group of HCC patients who are not optimal candidate for repeated TACE in our cohort. The newly developed AABCN score well predicted prognosis of HCC patients after second TACE and could be a useful alternative to guide decision for continuous TACE or not.