Albumin−bilirubin (ALBI) grade is an objective and reproducible model for evaluating overall survival (OS) in patients with hepatocellular carcinoma (HCC). However, the original ALBI grade was established for patients with Child−Pugh classes A−C. HCC patients with Child−Pugh class C or poor performance status (Barcelona Clinic Liver Cancer (BCLC) stage D) usually receive hospice care. Thus, optimized cutoffs for the ALBI grade for stratifying OS in HCC patients receiving anticancer therapy are pertinent for accurate prognostication. This study retrospectively enrolled 2116 patients with BCLC stages A−C HCC after the exclusion of those ineligible for receiving anticancer therapy. The modified ALBI (mALBI) grades were: an ALBI score ≤−3.02 for mALBI grade 1, an ALBI score >−3.02 to ≤−2.08 for mALBI grade 2, and an ALBI score >−2.08 for mALBI grade 3. The original ALBI and mALBI grades were independent predictors of OS in all the enrolled patients and those receiving transarterial chemoembolization. In patients receiving curative therapy (radiofrequency ablation and surgical resection), the mALBI grade (grade 2 vs. 1 and grade 3 vs. 2) was an independent predictor of OS. Original ALBI grade 2 vs. 1 was an independent predictor of OS but not ALBI grade 3 vs. 2. The mALBI model can differentiate between patients with early, intermediate, or advanced HCC who received anticancer therapy into three prognostic groups. External validation of the proposed mALBI grade is warranted.
Keywords: albumin–bilirubin (ALBI) grade; curative therapy; hepatocellular carcinoma; modified ALBI grade; survival.