Efficacy of targeted hepatocellular carcinoma therapy using modified response evaluation criteria in solid tumors and alpha-fetoprotein: A single center retrospective exploratory study

Backgrounds/Aims: To evaluate the combination of modified response evaluation criteria in solid tumors (mRECIST) with alpha-fetoprotein (AFP) in patients with intermediate-advanced hepatocellular carcinoma receiving combination therapy with a tyrosine kinase inhibitor (TKI) and a programmed cell death protein-1 (PD-1) inhibitor, focusing on stable disease (SD) patients with low AFP levels.Methods: We analyzed 79 patients who received TKI plus PD-1 inhibitor combination therapy from 2018 to 2021.Based on AFP at 8 weeks, patients were divided into Group A (AFP ≥200 ng/mL, n = 26) and Group B (AFP <200 ng/mL, n = 53).Kaplan-Meier and Cox models assessed overall survival (OS) and progression-free survival (PFS).Results: Only Barcelona Clinic Liver Cancer stage C independently predicted OS (hazard ratio [HR] 7.193, p < 0.001).For PFS, independent predictors included tumor diameter (HR 1.009, p = 0.025), mRECIST response (global likelihood-ratio p = 0.070), and AFP group (HR 0.474, p = 0.049).Virus type was excluded due to collinearity with antiviral therapy.Among SD patients, low AFP was associated with longer PFS than high AFP (24.0 vs. 3.0 months, p < 0.001).In Group B, low-AFP SD showed numerically longer PFS than partial response (24.0 vs. 15.0 months; HR 0.389; p = 0.116).Conclusions: This exploratory analysis suggests that the prognostic value of AFP is largely mediated by tumor burden.While mRE-CIST response predicted PFS but not OS, low-AFP SD patients exhibited favorable outcomes in both PFS and OS.These findings indicate that integrating mRECIST with AFP may enhance response stratification, though prospective validation is necessary.