Addition of tumor multiplicity improves the prognostic performance of the hepatoma arterial-embolization prognostic score

Background & Aims: The hepatoma arterial-embolization prognostic (HAP) score predicts survival outcome in patients with hepatocellular carcinoma (HCC) treated with trans-arterial chemoembolization (TACE). We validated the HAP score in Korean subjects with HCC and investigated whether its prognostic performance is improved with additional parameters. Methods: A total of 280 patients with HCC treated with TACE between 2003 and 2009 were included. Validation and modification of HAP score were performed based on multivariate Cox regression models. Results: The median age of the study population (211 men, 69 women) was 60 years. Viral etiology of HCC accounted for 80.4% (n = 181 for hepatitis B, 44 for hepatitis C). The median overall survival (OS) was 40.5 months. On multivariate analysis, together with the original components of the HAP score (serum albumin <3.6 g/dl, total bilirubin >0.9 mg/dl, alpha-foetoprotein >400 ng/ml, and tumor size >7 cm), tumor number >= 2 was selected as an independent unfavorable prognostic factor for OS (hazard ratio 2.3; P < 0.001). Accordingly, a modified HAP-II (mHAP-II) score was established by adding tumor number >= 2. Although both HAP and mHAP-II scores discriminated the four different risk groups (log-rank test, all P < 0.001), the mHAP-II score performed significantly better than the HAP score, as per the areas under receiver-operating curves predicting OS at 3 years (0.717 vs. 0.658) and 5 years (0.728 vs. 0.645), respectively (all P < 0.05). Conclusions: Although the HAP score predicted OS for Korean subjects with HCC undergoing TACE, the addition of tumor number significantly improved the prognostic performance. The mHAP-II score can be used for accurate prognostication and selection of optimal candidates for TACE.