External validation of a simplified BCLC staging system for early hepatocellular carcinoma

BACKGROUND AND AIMS: The aim was to externally validate the capability of a simplified Barcelona Clinic Liver Cancer (s-BCLC) staging system in allocating patients to hepatic resection (HR) and the effect on survival: S-BCLC was defined by only 2 groups: AA included BCLC A1 + A2 classes with alpha-fetoprotein (AFP) ≤ 20 ng/ml and AB included A1 + A2 with AFP > 20 ng/ml plus A3 + A4 subgroups. METHODS: This study compared a training group (TG) with hepatocellular carcinoma (HCC) submitted to hepatic resection (HR) in Milan with another group of patients, the validation group (VG) in Creteil. All patients underwent ultrasound-guided anatomical resection (<3 segments). RESULTS: Overall survival got worse from A1 to A4 (p = 0.0271) in TG (n = 132), as well as in VG (n = 100) (p = 0.0044) with a more important overlapping of each curves. According s-BCLC classification, the survival curves of TG (p = 0.0001) and VG (p = 0.0250) showed a definitive separation in two different staging groups. The s-BCLC provided the best predictive accuracy and it also presented the highest separability index and C-statistics in both TG and VG. On the other hand, in the evaluation of discriminatory ability for death, measured by ROC curve areas, the s-BCLC system gave better results than the others. CONCLUSION: This experience stressed the high value of BCLC system in staging of HCC, but the s-BCLC system seems to be more useful for therapeutic decision making.