Long-term survival of patients undergoing liver resection for very large hepatocellular carcinomas.

BACKGROUND: This study aimed to assess long-term survival after liver resection for huge hepatocellular carcinoma (HCC).

METHODS: Patients with stage I-III HCC who underwent hepatectomy from 2002 to 2010 were identified retrospectively from prospective national databases and followed until December 2012. Patients were assigned into four groups according to tumour size: less than 3·0 cm (small), 3·0-4·9 cm (medium), 5·0-10·0 cm (large) and over 10·0 cm (huge). The primary endpoint was overall survival. The Kaplan-Meier method and Cox proportional hazards model were used for survival analysis.

RESULTS: A total of 11 079 patients with HCC (mean(s.d.) age 59·7 (12·0) years) were eligible for this study. Median follow-up was 72·5 months. Patients with huge HCC had the worst prognosis; overall survival rates for patients with small, medium, large and huge HCC were 72·0, 62·1, 50·8 and 35·0 per cent respectively at 5 years, and 52·6, 41·8, 35·8 and less than 20·0 per cent at 10 years (P < 0·001). Multivariable analysis showed that tumour size affected long-term survival (hazard ratio (HR) 1·31, 1·55 and 2·38 for medium, large and huge HCC respectively versus small HCC). Prognostic factors for huge HCC were surgical margin larger than 0·2 cm (HR 0·70; P = 0·025), poor differentiation (HR 1·34; P = 0·004), multiple tumours (HR 1·64; P < 0·001), vascular invasion (HR 1·52; P = 0·008), cirrhosis (HR 1·37; P = 0·013) and the use of nucleoside analogues (HR 0·69; P = 0·004).

CONCLUSION: Huge HCCs have a worse prognosis than smaller HCCs after liver resection. A wide resection margin and antiviral therapy with nucleoside analogues may be associated with favourable long-term survival.