Abstract
BACKGROUND: Conversion therapy with hepatic arterial infusion chemotherapy (HAIC) combined with bevacizumab and sintilimab has shown promise for unresectable hepatocellular carcinoma (uHCC). However, predictors of postoperative recurrence remain unclear. METHODS: We retrospectively analyzed 112 HCC patients treated with HAIC + bevacizumab + sintilimab followed by surgical resection. Patients were stratified into recurrence (n = 30) and non-recurrence (n = 82) groups. Demographics, laboratory values, and tumor measurements were collected before and after conversion therapy. Recurrence-free survival (RFS) was estimated by Kaplan-Meier analysis. Restricted cubic spline (RCS) logistic regression was used to identify thresholds for AFP decline and tumor size decline associated with 1-year recurrence. Multivariable logistic regression was used to determine independent predictors of recurrence. RESULTS: During conversion therapy, the non-recurrence group exhibited greater tumor shrinkage (5.67 ± 3.06 cm vs. 8.77 ± 3.92 cm; p<0.001), lower ALT (p=0.017), higher AST (p=0.008), and lower bilirubin (p=0.006). The median RFS was 22.2 months (95% CI: 18.3-28.0); the 1- and 2-year RFS rates were 71.7% and 46.9%, respectively. The RCS model showed that an AFP decline greater than 25% and tumor size reduction significantly lowered the risk of 1-year recurrence, but reductions in tumor size beyond 60% did not confer additional benefits in reducing recurrence risk. In multivariate analysis, tumor size decline ratio (OR=0.002; 95% CI: 0.000-0.117; p=0.002) and AFP decline ratio (OR=0.240; 95% CI: 0.067-0.862; p=0.029) during conversion therapy independently predicted a lower recurrence risk. Elevated post-therapy bilirubin level remained an adverse predictor (OR=1.020; 95%CI: 1.000-1.030; p=0.039). Adverse events were predominantly grade 1-2, and grade 3-4 adverse events were manageable and well-controlled. CONCLUSIONS: Decline ratios of tumor size and AFP during HAIC + bevacizumab + sintilimab conversion therapy were robust and independent predictors of 1-year postoperative recurrence in HCC. Monitoring of these dynamic biomarkers may guide optimal surgical timing and follow-up strategies.