Multi-parameter gene expression profiling of peripheral blood for early detection of hepatocellular carcinoma

Area of Science:

• Biomarkers and Diagnostics
• Genomics and Molecular Biology
• Hepatocellular Carcinoma Research

Background:

• A previous study established a nine-gene expression detection system using the GeXP platform.
• Hepatocellular carcinoma (HCC) diagnosis requires accurate and non-invasive methods.

Purpose of the Study:

• To analyze gene expression data using various multi-parameter methods.
• To develop a diagnostic model for classifying HCC patients and healthy individuals.
• To evaluate the diagnostic performance of a nine-gene expression signature in peripheral blood.

Main Methods:

• Utilized logistic regression, discriminant analysis, classification tree analysis, and artificial neural networks.
• A diagnostic model was constructed using data from 103 early HCC patients and 54 healthy controls.
• Model performance was validated on 52 HCC patients and 34 healthy individuals, assessing AUC, sensitivity, and specificity.

Main Results:

• The artificial neural network model incorporating all nine genes demonstrated the highest diagnostic value.
• Achieved an Area Under the Curve (AUC) of 0.943, with 98% sensitivity and 85% specificity in initial model building.
• Validation results showed a sensitivity of 96% and specificity of 86% for the nine-gene model.

Conclusions:

• Multi-parameter analysis of gene expression significantly enhances diagnostic value compared to single-factor approaches.
• The developed nine-gene expression signature shows promise as a future clinical diagnostic tool for HCC.
• Gene expression profiling in peripheral blood offers a potential non-invasive strategy for early HCC detection.