Early detection of hepatocellular carcinoma (HCC) is crucial for curative treatment, yet current screening strategies for high-risk liver cirrhosis (LC) patients lack sufficient sensitivity. This study evaluates plasma cell-free DNA(cfDNA) concentration and fragmentomics as biomarkers to improve HCC diagnosis and prognosis. Plasma samples from 39 HCC and 46 LC patients were analyzed for cfDNA concentration and fragment patterns. A multivariate logistic regression model (CMAC), integrating cfDNA concentration, mononucleosome proportion (%MN), alpha-fetoprotein (AFP), and c-reactive protein (CRP), was developed and validated using Leave-One-Out Cross-Validation and bootstrapping. HCC patients exhibited significantly higher cfDNA concentrations (p < 0.0001) and longer fragment lengths (p < 0.05) compared to LC patients. The CMAC model demonstrated superior diagnostic performance (AUROC = 0.946) compared to AFP alone (AUROC = 0.777, p < 0.001). Notably, in early-stage HCC, the CMAC model remained highly accurate (AUROC = 0.941), whereas AFP failed to reach statistical significance. Higher CMAC scores were significantly associated with advanced BCLC stages (p = 0.009), lymphovascular invasion (p = 0.0063) and reduced overall survival (p = 0.0037). Integration of cfDNA analysis with established clinical markers in the CMAC model shows promise as a complementary tool for the early detection of HCC in LC patients. Validation in larger, multicenter cohorts will be necessary to confirm these findings and their clinical applicability.
Keywords: cancer biomarker; circulating free DNA; cirrhosis; fragmentomics; hepatocarcinoma; liquid biopsy; prognostic and predictive markers.