Circulating Biomarkers Predict Immunotherapeutic Response in Hepatocellular Carcinoma Using a Machine Learning...

Area of Science:

• Oncology
• Immunotherapy
• Biomarker Discovery

Background:

• Hepatocellular carcinoma (HCC) response to immune checkpoint inhibitor (ICI) therapy is limited.
• Predictive biomarkers are needed to identify patients likely to benefit from ICI treatment.
• This study focuses on developing a classifier using circulating indicators for HCC patients.

Purpose of the Study:

• To design a predictive classifier for ICI therapy response in HCC patients.
• To identify circulating biomarkers associated with survival benefits from ICI therapy.
• To construct a prognostic model for predicting patient survival post-treatment.

Main Methods:

• Retrospective study of 369 HCC patients treated with ICI therapy.
• Utilized random forest analysis to identify 11 significant serum biomarkers from 17 tested.
• Developed and refined a prognostic model incorporating circulating biomarkers and clinical features.

Main Results:

• An initial classifier using 11 biomarkers achieved an AUC of 0.724.
• Patients in the low-risk group showed a significantly better prognosis (P=0.0079).
• An enhanced model with 12 biomarkers and 5 clinical features improved AUC to 0.752, with low-risk patients having superior overall survival (P=0.026).

Conclusions:

• Circulating biomarkers can effectively predict therapeutic outcomes in HCC patients receiving ICI therapy.
• The developed classifier aids in clinical decision-making for ICI treatment selection.
• Biomarker-based prediction can optimize the use of ICI therapy in hepatocellular carcinoma.