A mitochondrial function-based prognostic model for hepatocellular carcinoma uncovers MRPS15 as a key driver of tumor...

Hepatocellular carcinoma (HCC) is highly heterogeneous with unpredictable outcomes. Mitochondrial dysfunction drives HCC progression, but its prognostic implications remain unclear. This study aimed to develop a mitochondrial-related prognostic model for HCC.

We analyzed transcriptomic data from 370 HCC patients from The Cancer Genome Atlas (TCGA) using consensus clustering of mitochondrial pathway genes. A mitochondrial-associated scoring model (MASM) was developed via least absolute shrinkage and selection operator (LASSO) and Cox regression, validated across independent cohorts [GSE116174 and the International Cancer Genome Consortium Liver Cancer-RIKEN, Japan (ICGC-LIRI)]. Functional roles of MRPS15, a key gene in the model, were investigated through in vitro experiments and subcutaneous tumor formation experiment in nude mice.

Consensus clustering identified three mitochondrial subtypes with distinct metabolic profiles: subtype A exhibited elevated oxidative phosphorylation (OXPHOS) and preserved metabolic homeostasis; subtype B showed adaptive metabolic reprogramming; subtype C displayed selective activation of biosynthetic pathways despite widespread mitochondrial dysfunction. The MASM prognostic signature, comprising eight mitochondrial genes, achieved predictive accuracy with area under the curve (AUC) values of 0.805, 0.751, and 0.742 for 1-, 3-, and 5-year survival, respectively, and showed favorable trends compared with clinical staging. MRPS15 was validated as a critical regulator of HCC cell proliferation, invasion, migration, and apoptosis, primarily through sustaining OXPHOS and redox homeostasis.

This study establishes a mitochondrial-based classification and eight-gene MASM model for HCC, which stratifies patients into three subtypes with distinct metabolic, genomic, and immune profiles. The model outperforms traditional staging in survival prediction. MRPS15 is validated as a key oncogenic driver sustaining OXPHOS. These findings provide a framework for personalized therapy and warrant prospective clinical validation.