Advances in CT and MRI for Yttrium-90 Radioembolization of Hepatocellular Carcinoma.

RATIONALE AND OBJECTIVES: Transarterial radioembolization (TARE) is increasingly used for patients with hepatocellular carcinoma (HCC) across Barcelona Clinic Liver Cancer stages. It provides local disease control and downstaging or bridging to definitive therapy in selected cases. Imaging plays a central role in patient selection, dosimetry planning, and post-treatment assessment. This review aims to summarize current advances on Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) biomarkers for optimizing TARE clinical management and prognostic evaluation. MATERIALS AND METHODS: This narrative review was performed based on recent evidence regarding CT and MRI biomarkers used to predict TARE eligibility (eg, lung shunt fraction [LSF] prediction, tumour-to-non tumour ratio [TNR], functional reserve), quantify treatment response Response Evaluation Criteria in Solid Tumors (RECIST) 1.1, modified RECIST, and Liver Imaging Reporting and Data System (LI-RADS) radiation Treatment Response Algorithm [TRA]), and predict survival (conventional features, body composition, radiomics). RESULTS: Pre- and intraprocedural CT/MRI, including dynamic contrast-enhanced MRI and cone-beam CT (CBCT) with perfusion/iodine mapping, can approximate microsphere distribution and support personalized dosimetry. For response assessment, enhancement-based criteria outperform purely size-based metrics for therapies inducing necrosis. LI-RADS TRA v2024 introduces radiation-specific guidance and a "non-progressing" category that can underpin watch-and-wait strategies in appropriate contexts. Radiomics and machine learning show promising accuracy for predicting early response and overall survival (OS). However, they require standardization and prospective validation. CT/MRI-derived body composition metrics (eg, sarcopenia, adipose tissue density, fat-free muscle area) are independently associated with outcomes and are practical to extract from routine scans. CONCLUSION: Standardization and multi-center validation are essential to translate these imaging biomarkers into clinical practice.