Alpha-fetoprotein change after systemic treatments in advanced hepatocellular carcinoma: a systematic review and trial-level meta-analysis

Alpha-fetoprotein (AFP) kinetics after systemic therapy for unresectable or metastatic hepatocarcinoma has been evaluated as an early prognostic factor in many studies, but the effect size has varied largely. The current study aims to synthesize the evidence about the relationship between AFP changes and overall survival (OS). A literature search of clinical studies of metastatic hepatocarcinoma patients receiving systemic treatments was conducted in two databases, PubMed and Web of Science. After selecting studies evaluating the relationship of any AFP changes with OS, a meta-analysis was performed. Finally, the possible effect of the baseline variables on the relationship was explored by meta-regression. Thirty-three articles (3815 patients) were selected. AFP reduction was associated with longer OS (hazard ratio = 0.46, confidence interval = 0.41–0.52), with consistent heterogeneity ( Q  = 74.87; P  < 0.0001; I ² statistics = 56.8%) and publication bias. The relationship was similar for antiangiogenics and immunity checkpoint inhibitors. A meta-regression was carried out for 17 variables, and two of them (hepatitis B virus-related liver disease and Asian origin) significantly influenced the relationship of AFP changes with OS. Early AFP change after systemic treatment for metastatic hepatocarcinoma is a reliable prognostic factor and a possible intermediate endpoint for clinical trials. AFP change standardization of cutoff and timing in prospective trials is strongly recommended.