A New Prognostic Model Based on Albumin–Bilirubin Grade for Hepatocellular Carcinoma Beyond the Milan Criteria

The survival of patients with advanced hepatocellular carcinoma (HCC) is highly variable due to heterogeneous tumoral characteristics. We proposed and validated an albumin–bilirubin (ALBI)-based model for HCC beyond Milan criteria, the ALBI-HOME, for these patients.

A total of 2186 patients were enrolled and randomly assigned to the derivation cohort (n = 1093) and validation cohort (n = 1093). Multivariate Cox proportional hazards model was used to determine significant prognostic factors in the derivation cohort. The performance of ALBI-HOME was evaluated in the validation cohort.

In the Cox model, six factors were identified as independent predictors of poor survival: ALBI grade 2 [hazard ratio (HR) 1.848, 95% confidence incidence (CI) 1.556–2.195, p < 0.001], ALBI grade 3 (HR 3.266, 95% CI 2.531–4.215, p < 0.001), serum AFP ≥ 100 ng/ml (HR 1.482, 95% CI 1.279–1.717, p < 0.001), total tumor volume ≥ 250 cm³ (HR 1.503, 95% CI 1.294–1.746, p < 0.001), ascites (HR 1.400, 95% CI 1.187–1.561, p < 0.001), performance status 0–1 (HR 1.756, 95% CI 1.485–2.076 p < 0.001), and vascular invasion or metastasis (HR 2.110, 95% CI 1.809–2.0, p < 0.001). The ALBI-HOME is based on these six parameters, and the score ranges from 0 to 7. This model was associated with the best prognostic ability among different HCC staging systems to predict survival in patients beyond Milan criteria; its ability remained consistently stable in different treatment subgroups and viral etiologies.

The proposed ALBI-HOME is a simple and feasible predictive model for HCC beyond Milan criteria. It demonstrates superior prognostic performance among the currently used staging systems and may help identify at-risk patients to undergo more aggressive treatments.