Serum-biomarker-based population screening model for hepatocellular carcinoma

• DOI: 10.1016/j.isci.2025.111981
• Publication Date: 2025-02-08T23:22:38Z
• Authors: Wenmin Liao; Wenbin Lin; Zhonglian He; Chenyang Feng; Yuying Liu; Zixian Wang; Ruizhi Wang; Meifang He; Shuqin Dai; Ying Sun; Wei Wei; Peisong Chen; Chaofeng Li
• Keywords: Public health; Science; Q; Cancer

Hepatocellular carcinoma (HCC) early identification is crucial for improving patient outcomes. Current screening methods are often complex and costly. This study developed a simplified, cost-effective HCC model using serum marker data. A diverse population from two Chinese hospitals was recruited, including cancer patients, hospital healthy individuals. two-stage created: LASSO logistic regression preliminary screening, followed by incorporating alpha-fetoprotein (AFP). The model's performance evaluated in multiple cohorts. Across five populations, the showed strong with AUC-ROC ranging 0.868 to 0.907, accuracy between 87.43% 96.96%, sensitivity over 75% specificity above 90%. Compared solely AFP models, second-stage improved risk estimates significantly higher AUC (0.930 vs. 0.827) net reclassification improvement (NRI) up 56.2%. offers practical, cost-efficient tool detection, addressing significant public health need.