Identification of immunogenic cell death signature genes in hepatocellular carcinoma: from single-cell transcriptomics to in vitro mechanistic validation and comprehensive prognostic modeling with hundreds of machine learning algorithms

Hepatocellular carcinoma (HCC) lacks reliable prognostic biomarkers for immunotherapy. Immunogenic cell death (ICD) represents a promising therapeutic target, but its comprehensive characterization in HCC remains unexplored. We performed multi-omics integration of single-cell RNA sequencing data fro...

Full description

Saved in:
Bibliographic Details
Published in:Frontiers in immunology Vol. 16; p. 1649618
Main Authors: Liu, Hongliang, Sun, Zhenni, Wang, Xi, Zhou, Bin, Cha, Lichao
Format: Journal Article
Language:English
Published: Switzerland Frontiers Media S.A 2025
Subjects:
Biomarkers, Tumor - genetics
Carcinoma, Hepatocellular - genetics
Carcinoma, Hepatocellular - immunology
Carcinoma, Hepatocellular - mortality
Carcinoma, Hepatocellular - pathology
Gene Expression Profiling
Gene Expression Regulation, Neoplastic
Hep G2 Cells
hepatocellular carcinoma
Humans
immunogenic cell death
Immunogenic Cell Death - genetics
Liver Neoplasms - genetics
Liver Neoplasms - immunology
Liver Neoplasms - mortality
Liver Neoplasms - pathology
Machine Learning
multi-omics integration
precision medicine
Prognosis
Single-Cell Analysis
Transcriptome
tumor microenvironment
Tumor Microenvironment - genetics
Tumor Microenvironment - immunology
multi-omics integration
precision medicine
tumor microenvironment
machine learning
hepatocellular carcinoma
immunogenic cell death
ISSN:1664-3224, 1664-3224
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Hepatocellular carcinoma (HCC) lacks reliable prognostic biomarkers for immunotherapy. Immunogenic cell death (ICD) represents a promising therapeutic target, but its comprehensive characterization in HCC remains unexplored. We performed multi-omics integration of single-cell RNA sequencing data from 7 HCC samples (GSE112271, 44,461 cells) with bulk transcriptomics from three independent cohorts (TCGA-HCC [n=371], GSE14520 [n=242], ICGC [n=445]). ICD activity was quantified using ssGSEA. We identified HCC-specific ICD-related (HCC-ICDR) genes via WGCNA and optimized a prognostic model by benchmarking machine learning algorithms. Experimental validation included functional assays using CLIC1 and NAP1L1 overexpression in HepG2 cells. The ICD-based risk score (ICDRS) demonstrated superior prognostic accuracy (C-index=0.839), validated across cohorts. Single-cell profiling revealed macrophages exhibited the highest ICD activity. High-risk patients displayed immunosuppressive microenvironments with enriched Tregs, M0 macrophages, and neutrophils, alongside hyperactivated DNA repair and MYC signaling. Low-risk patients showed anti-tumor immunity with increased CD8+ T cells and M1 macrophages. ICDRS predicted differential therapeutic vulnerabilities: low-risk patients showed enhanced sensitivity to standard immunotherapy-compatible treatments including sorafenib and doxorubicin, while high-risk patients demonstrated preferential sensitivity to EGFR-targeted therapies. Experimental validation confirmed CLIC1 and NAP1L1 significantly promoted HCC malignant behaviors. We established the comprehensive ICD-based prognostic framework for HCC, revealing novel tumor-immune interactions and therapeutic vulnerabilities. This model provides robust stratification for immunotherapy selection and advances precision medicine in HCC management. Future clinical translation includes prospective validation and development of companion diagnostics, offering potential pathways for personalized HCC treatment implementation.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1664-3224
1664-3224
DOI:10.3389/fimmu.2025.1649618