Integrating GPC3 with Other Biomarkers to Improve the Diagnosis of Early-Stage Liver Cancer

Serum Glypican-3 (GPC3) levels in HCC patients are significantly higher than those in healthy individuals or patients with non-malignant liver diseases, making it a diagnostic marker for HCC. However, its diagnostic capability remains controversial due to its low sensitivity. The common marker AFP has limitations in terms of sensitivity and specificity, particularly in early-stage HCC. We sought to combine GPC3 detection with multi-biomarker panels to enhance sensitivity and specificity in early-stage HBV-, HCV-, and ALD-related liver cancer diagnosis. We applied receiver operating characteristic (ROC) analysis, which is used to evaluate the diagnostic performance of different biomarker tests, to develop comprehensive multi-biomarker panels that include GPC3, along with other biomarkers such as gender, age, AFP, AFP-L3%, and DCP, for assessment in the selected patients. We also applied univariate and multivariate logistic regression analysis to generate a specific diagnostic model for early HBV-induced HCC detection. We found that GPC3 levels in serum were significantly higher in HCC patients compared to CLD patients. We performed univariate and multivariate logistic regression analysis on the relevant indicators of early HCC to establish a new GDATA model for diagnosing early HCC. The new model included five indicators of early HCC: GPC3, DCP, AFP-L3%, TBIL and age. The diagnostic efficacy was better than that of GPC3, AFP, DCP and AFP-L3 alone. The diagnostic accuracy of the GDATA model for early HCC was significantly higher than that of the GALAD model or single indicators alone. The GDATA model thus provides a new promising diagnostic strategy for early HCC detection.