Radiomics predicts poorly differentiated hepatocellular carcinoma and uncovers ribosomal-immune dysregulation mechanism

TL;DR

This study uses radiomics to noninvasively predict liver cancer grades and identifies a link between imaging patterns and tumor biology, including ribosomal and immune system dysregulation.

Contribution

A novel radiogenomic framework is introduced for noninvasive HCC grading and uncovering ribosomal-immune dysregulation mechanisms.

Findings

A multi-VOI random forest model achieved high accuracy in predicting HCC grades (AUC 0.959 internally, 0.860 externally).

Radiogenomic analysis revealed associations between imaging features and ribosomal dysregulation and immune exhaustion.

A prognostic signature derived from radscore independently predicted patient survival in TCGA-HCC data.

Abstract

Hepatocellular carcinoma (HCC) shows marked spatial heterogeneity, limiting biopsy-based Edmondson-Steiner (ES) grading. We developed a multicenter radiogenomic framework to noninvasively predict ES grade and explore underlying molecular mechanisms. Arterial-phase DCE-MRI from 295 patients and The Cancer Imaging Archive (TCIA) cases were analyzed using three tumor regions (body, edge, and out). An integrated volume-of-interest (VOI) random forest (RF) model was trained with selected features and externally validated. Radiogenomic analysis correlated radscore with TCIA transcriptomic profiles using weighted gene co-expression network analysis (WGCNA). The model achieved high discrimination (area under the curve [AUC] 0.959 internally; 0.860 externally). Radscore-associated modules revealed ribosomal dysregulation and immune exhaustion. A derived prognostic signature stratified and The Cancer Genome Atlas (TCGA) patients into distinct risk groups and independently predicted survival (hazard ratio [HR] 3.95, p < 0.0001; C index 0.643). This integrated radiogenomic approach enables noninvasive ES grading and provides insight into biologically relevant tumor heterogeneity.

•Integrated multi-VOI radiomics enables noninvasive Edmondson-Steiner grading in HCC•An externally validated random forest model achieves robust multicenter performance•Radiogenomic analysis links imaging heterogeneity to ribosomal dysregulation and immunity•A radscore-derived signature independently predicts survival in TCGA-HCC patients

Integrated multi-VOI radiomics enables noninvasive Edmondson-Steiner grading in HCC

An externally validated random forest model achieves robust multicenter performance

Radiogenomic analysis links imaging heterogeneity to ribosomal dysregulation and immunity

A radscore-derived signature independently predicts survival in TCGA-HCC patients

Cancer; molecular interaction; molecular network

Figures