Background and aims: Many staging systems for hepatocellular carcinoma (HCC) have been proposed but the best tool for staging of HCC remains controversial. The aim of the present study was to identify the best staging system evaluating the predictive ability for outcome for each of the seven different staging systems applied in a homogeneous group of patients who underwent percutaneous radiofrequency ablation (RFA).
Methods: We analyzed retrospectively 112 patients with HCC and cirrhosis treated with percutaneous RFA from January, 1998 to April, 2005. Response to treatment after 30 days and for long-term follow-up was evaluated with computed tomography (CT) or magnetic resonance imaging (MRI) and serum alpha-fetoprotein level (AFP). All of the 112 patients were grouped according to each one of the seven different staging systems: Okuda, TNM, BCLC, CLIP, GRETCH, CUPI, JIS.
Results: The mean follow-up time of the 112 patients submitted to RFA was 24 months (range 3-92 months) with survival rates at 1, 3, and 5 yr of 82%, 40%, and 18%, respectively. Univariate and multivariate analyses showed that factors related to survival were Child-Pugh score (P</= 0.01), serum AFP (P</= 0.01), and the response to treatment (P</= 0.01) with hazard ratios of 2.09 (95% CI 1.21-3.61), 2.79 (95% CI 1.59-4.90), and 2.76 (95% CI 1.25-6.09), respectively. The comparison of the results of the different staging systems in all of the 112 patients and in a subgroup of 96 patients with complete response to treatment showed that BCLC had the best discrimination ability, monotonicity of gradient (linear trend chi(2) 6.07, P= 0.01), and homogeneity ability (LR chi(2) test 10.00, P= 0.008).
Conclusions: The BCLC staging system shows a superior discriminatory power in our cohort of HCC patients who underwent RFA; moreover, it can give important prognostic information after complete response to treatment. Our study confirms the validity of the BCLC staging system in patients with HCC in cirrhosis.