Identification of Regorafenib Prognostic Index (REP Index) via Recursive Partitioning Analysis in Patients with Advanced Hepatocellular Carcinoma Receiving Systemic Treatment: A Real-World Multi-Institutional Experience

Background: The results of the pivotal RESORCE trial led to the approval of the tyrosine kinase inhibitor regorafenib as second-line treatment in advanced hepatocellular carcinoma (HCC) after sorafenib failure. Data about prognostic factors in a second-line HCC setting are scarce. Objective: The aim of the present study was to investigate prognostic factors in a cohort of patients with advanced HCC treated with regorafenib after progressing on sorafenib. Methods: We retrieved the data of 259 patients affected by advanced HCC treated with regorafenib as second-line treatment from four different Italian institutions and one South Korean institution and performed a recursive partitioning analysis to build a score system. Results: At the first-step univariate analysis for overall survival (OS), alkaline phosphatase (ALP) was the most significant parameter and was chosen as the first node in our tree model. In the subpopulation of patients presenting with ALP ≤122 U/L (n=155) at baseline, the most statistically significant split was by progression-free survival (PFS) on previous sorafenib treatment, between patients with a PFS ≥ 6 months (n = 59) and patients with a PFS < 6 months (n = 96). In the subpopulation of patients with ALP ≤ 122 U/L and PFS to sorafenib ≥ 6 months, the final split was determined between patients with hepatitis B virus (HBV)-related liver disease (n = 22) and patients with no HBV-related liver disease (n = 37). In the subpopulation of patients presenting ALP >122 U/L (n = 104) at baseline, the most statistically significant split was by aspartate aminotransferase (AST) value, between patients with AST ≤ 56 U/L (n = 48) and patients with AST > 56 U/L (n = 56). We built the Regorafenib Prognostic Index (REP index) stratifying the population into “low-risk,” “medium-risk,” and “high-risk” groups. The difference in median OS between the three risk groups was statistically significant, being 20.8 months (95% confidence interval [CI] 10.0–46.3) in the “low-risk” group, 8.4 months (95% CI 7.2–1435.8) in the “medium-risk” group, and 5.5 months (95% CI 3.5–13.2) in the “high risk” group. The median PFS was 7.7 months (95% CI 3.7–19.3), 2.5 months (95% CI 2.1–28.8), and 2.4 months (95% CI 1.6–9.1) for the “low-risk,” “medium-risk,” and “high-risk” groups, respectively. Conclusion: The REP index is an independent prognostic factor for OS and PFS in patients with advanced HCC treated with regorafenib.