一种新型免疫代谢预后模型对肝细胞癌生存进行分层并实现个体化治疗

The etiology of HCC is multifactorial, with pathogenesis involving immune and metabolic pathways. This study developed a prognostic model based on immune- and metabolism-related genes (IMRGs) to improve personalized HCC management.

Transcriptomic data from HCC patients were obtained from TCGA (n = 377) and GEO (n = 115). Differentially expressed IMRGs between tumor and adjacent normal tissues were identified, and patients were stratified using non-negative matrix factorization (NMF). A LASSO-Cox prognostic model was constructed and validated, with assessments of immune microenvironment, therapeutic sensitivity, and functional pathways.

HCC patients were classified into two subgroups with distinct immune microenvironment features. LASSO regression identified nine key prognostic genes from 54 consensus IMRGs, forming a robust risk model. High-risk patients had significantly worse survival, and the model outperformed existing immune/metabolic signatures (5-year AUC = 0.700 vs. 0.570-0.603) and clinical parameters (AUC = 0.720 vs. 0.470-0.712). Functional analysis revealed enhanced chromosome separation and nuclear division in high-risk patients, while low-risk patients showed elevated amino acid catabolism and fatty acid metabolism. High-risk patients also exhibited reduced Natural Killer cell activity and type II interferon responses but increased oxaliplatin sensitivity and 5-Fluorouracil resistance.

This IMRG-based prognostic model effectively predicts HCC outcomes, enabling risk stratification and personalized therapy. It demonstrates strong clinical utility, advancing precision medicine in HCC management.