BetaEntity Annotation Prototype
← Back to institutions

Annotated abstract

IDDF2026-ABS-0014 Why current serum biomarkers fail in very early hepatocellular carcinoma: evidence of a stage-specific diagnostic gap between BCLC 0 and A

gutjnl · 2026-06-26 · canonical JSON source

21 visible annotations · policy: published · automated confidence ≥ 75.00%

Document resource

Background Early detection of hepatocellular carcinoma (HCC) at the very early stage (BCLC 0) is critical for curative treatment. Serum biomarkers including alpha-fetoprotein (AFP), AFP-L3, des-gamma-carboxyprothrombin (DCP), and the GALAD score are widely used for HCC surveillance; however, their ability to discriminate between very early and early-stage HCC remains unclear. This study aimed to evaluate whether current serum biomarkers can reliably distinguish BCLC 0 from BCLC A and to explore the underlying reasons for diagnostic failure at the very early stage.Methods This cross-sectional diagnostic study included consecutive patients with HCC and matched controls with chronic liver disease recruited at a tertiary referral center between March and October 2025. Serum AFP, AFP-L3, and DCP levels were measured, and GALAD scores were calculated. Diagnostic performance was assessed using receiver operating characteristic (ROC) curve analysis. Subgroup analyses focused on early-stage HCC (BCLC 0–A), with specific evaluation of biomarker discrimination between BCLC 0 and BCLC A.Results A total of 166 participants (83 HCC, 83 controls) were included. For overall HCC detection, DCP and the GALAD score demonstrated excellent diagnostic performance (both AUC 0.957), significantly outperforming AFP and AFP-L3 (p < 0.05). In the early-stage subgroup (BCLC 0–A, n = 23), DCP was the only biomarker capable of differentiating BCLC 0 from BCLC A (AUC 0.767; 95% CI 0.559–0.974; p = 0.039). AFP, AFP-L3, and GALAD showed no significant discriminatory ability between these stages.Conclusions At the BCLC 0 stage, the limited performance of serum biomarkers does not reflect inappropriate test selection, but rather indicates that tumor biology has not yet generated sufficiently strong serum signals to support very early detection in clinical practice.