Hypoxia-related and immune phenotype-related fusion model for non-invasive prognostication of hepatocellular carcinoma treated by TACE: a multicentre study - PubMed
3 hours ago
- #Prognostication Model
- #TACE
- #Hepatocellular Carcinoma
- Study aims to develop a multimodal prognostication model for hepatocellular carcinoma (HCC) patients treated with transarterial chemoembolisation (TACE).
- Enrolled 1448 HCC patients across multiple cohorts, including a TACE cohort, biomarker subset, single-cell RNA sequencing cohort, and TCGA HCC cohort.
- Utilized pre-treatment contrast-enhanced CT images to construct deep learning and conventional radiomic models.
- Compared early-fusion and late-fusion models (LFMs), integrating the better-performing LFM with clinical variables to form a clinical-radiologic model (CRM).
- CRM effectively stratified patients' survival across independent cohorts and provided more granular risk stratification than existing models.
- Multi-omic analyses revealed activation of myelocytomatosis oncogene, enhanced epithelial-mesenchymal transition, upregulated glycolysis, and hypoxia pathway activation in high-score groups.
- Single-cell transcriptomic data confirmed high hypoxia scores in all cell types of high-risk patients and reduced cytotoxic activity in T cells.
- CRM model can non-invasively predict the prognosis of HCC patients treated with TACE therapy.