Hepatic resection for "BCLC stage A" hepatocellular carcinoma. The prognostic role of alpha-fetoprotein

BACKGROUND: Our aim was to assess the capability of Barcelona Clinic Liver Cancer (BCLC) staging system in allocating stage A patients to hepatic resection (HR) and the effect on survival. METHODS: We enrolled 132 patients with hepatocellular carcinoma (HCC) amenable to HR. All patients underwent ultrasound (US)-guided anatomical resection (≤2 segments) and then postoperative results were evaluated. RESULTS: Results showed 95% of patients were Child A, 49% in BCLC A1, 21% in A2, 6% in A3, and 24% in A4. No 30-day mortality occurred. Overall survival got worse from A1 to A4 (P = 0.0271), while no differences were found in Childs A patients with or without portal hypertension (P = 0.1674). Multivariate analysis (Cox model) shows that only AFP (<20 ng/ml) was an independent predictor of survival: If the AFP is incorporated in BCLC staging system (all A1 and A2 patients with abnormal AFP levels were included in A3 subgroup), 5-year survival rate including normal AFP for A1 was 57% and for A2 was 65%, whereas the survival rates impaired in the worst candidates (5-year survival rate including AFP abnormal for A3 and A4 was 36%; P = 0.002). So, introducing AFP in BCLC classification it is possible to simplify the algorithm in only 2 classes, well-separated in survival curves (class 1 [AFP-]: 60%; class 2 [AFP+]: 37%; P = 0.0001). CONCLUSION: Our experience stressed the high value of BCLC system in staging of patients with HCC, but underlined that in selected patients (normal AFP) even A2 group may benefit from HR with a good survival.