Super-Resolution CEMRI Radiomics with Peritumoral Expansion Predicts Macrotrabecular-Massive HCC and Immune Microenvironment

A radiomics model using super-resolution CEMRI with intratumoral+5mm peritumoral features predicted macrotrabecular-massive HCC (AUC 0.892 train, 0.812 test). Adding Child-Pugh and PIVKA-II in a nomogram improved AUC to 0.906/0.853. High-risk group showed angiogenic, EMT, and im…

• In 209 HCC patients (88 MTM-HCC), a radiomics model combining intratumoral and 5 mm peritumoral features from super-resolution portal venous phase CEMRI achieved AUC of 0.892 (training) and 0.812 (test).
• A nomogram integrating the radiomics model with Child-Pugh class/score and PIVKA-II further improved AUC to 0.906 (training) and 0.853 (test).
• Transcriptomic analysis of 32 TCGA HCC cases stratified by radiomics risk showed enrichment in angiogenesis, epithelial-mesenchymal transition, and inflammatory pathways, with increased resting mast cell infiltration suggesting immunosuppression.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.