Dynamic ctDNA Monitoring Predicts Recurrence and Guides Therapy in Postoperative Hepatocellular Carcinoma.

BACKGROUND: Hepatocellular carcinoma (HCC) remains a global health challenge with high recurrence rates despite surgical resection. Current biomarkers lack sensitivity for early relapse detection, underscoring the need for novel strategies. Circulating tumor DNA (ctDNA) has emerged as a promising tool for minimal residual disease (MRD) monitoring, yet its prognostic and therapeutic implications in HCC remain underexplored. METHODS: This retrospective study enrolled 23 HCC patients undergoing R0 resection. Preoperative and serial postoperative plasma samples were analyzed using a 425-gene next-generation sequencing panel to detect MRD. Tissue-plasma mutation concordance, recurrence risk stratification, and therapeutic applications were evaluated. RESULTS: MRD positivity at postoperative Day 7 predicted significantly shorter disease-free survival (DFS: 3.06 vs. not reached; p = 0.0027). Dynamic monitoring revealed three recurrence risk groups: persistently MRD-positive (highest risk, HR 4.12; p < 0.001), MRD-negative-to-positive (intermediate), and continuously negative (lowest). ctDNA detected recurrence 16 weeks earlier than imaging (81.8% sensitivity, 100% specificity). High tissue-plasma mutation concordance (84.6%) confirmed ctDNA's biological relevance, with TP53 (60.9% tissue/70% plasma) and CTNNB1 (17.4%/25%) as predominant drivers. ASXL1 mutations (p = 0.025) and microvascular invasion (p < 0.05) independently predicted poor DFS. CONCLUSIONS: Dynamic ctDNA analysis improves postoperative HCC management through early recurrence prediction and precision therapy guidance. High mutation concordance validates its role in reflecting tumor heterogeneity, while MRD trajectory stratification enables personalized surveillance.