Background The Barcelona Clinic Liver Cancer (BCLC) staging system is the most widely used staging system for hepatocellular carcinoma (HCC). However, the classifications of early (BCLC-A) and intermediate (BCLC-B) stage HCC remain controversial. Aim To refine the staging of BCLC-A and -B. Methods A total of 986 patients with HCC undergoing liver resection from two institutions formed the training cohort, and 694 from another institution were the validation cohort. Time-dependent receiver operating characteristic (ROC) curve analysis was performed to evaluate the performance of tumour size in predicting overall survival (OS), and determined the optimised cut-off. Discriminatory performance was evaluated using Harrell's concordance index (C-index). Results Patients with multiple tumours exceeding Milan criteria but within up-to-seven criteria had similar OS and disease-free survival (DFS) to those with multiple tumours meeting Milan criteria, and were assigned to the modified BCLC-A stage. The area under the ROC curve of tumour size for predicting OS was 0.778, and the diameter of 7 cm was the optimal cut-off to identify patients with single tumours who had higher OS than BCLC-B stage patients. Due to the similar OS, patients with single HCCs >7 cm were assigned to the modified BCLC-B stage. The C-indexes of the modified BCLC classification for OS and DFS were higher compared to the original version. The findings were supported by the validation cohort. Conclusions The modified staging of BCLC-A and -B, based on single tumour >7 cm and multiple tumours beyond up-to-seven criteria, could be more accurate to predict the prognosis of HCC patients. Liver resection could benefit patients with resectable multifocal HCCs beyond the Milan criteria.
