Prognostic Value of MRI ‐Based Treatment Response Assessment Criteria in Hepatocellular Carcinoma After Stereotactic Body Radiotherapy: A Comparative Analysis of mRECIST , LI ‐ RADS TRA Version 2017, and LI ‐ RADS Radiation TRA Version 2024

BACKGROUND: Accurate imaging assessment of hepatocellular carcinoma (HCC) after stereotactic body radiotherapy (SBRT) remains challenging. Although LI-RADS CT/MRI Radiation Treatment Response Assessment (TRA v2024) was developed for this purpose, its prognostic value for survival stratification after SBRT remains to be validated. PURPOSE: To compare the prognostic performance of mid-term (3-6 months) and late-term (9-12 months) MRI-based treatment response assessed using modified RECIST (mRECIST), LI-RADS TRA v2017, and LI-RADS Radiation TRA v2024 in patients with HCC. STUDY TYPE: Retrospective cohort study. POPULATION: 125 patients with HCC underwent SBRT (111 males; mean age, 56.4 ± 10.8 years). FIELD STRENGTH/SEQUENCE: 3.0 T; axial fat-suppressed fast spin-echo T2-weighted imaging, diffusion-weighted imaging, pre-contrast gradient recalled echo T1-weighted imaging, and contrast-enhanced late arterial phase, portal venous phase, and delayed/transitional phase. ASSESSMENT: Tumor response was independently evaluated by three radiologists using three treatment assessment methods. Ancillary features (AFs) were additionally recorded to assess their incremental prognostic value. Overall survival (OS) and progression-free survival (PFS) were analyzed according to response categories. STATISTICAL TESTS: Kolmogorov-Smirnov test, Intraclass correlation coefficients, Cohen's κ, Variance inflation factor analysis, Kaplan-Meier method, Log-rank tests with Bonferroni correction, Univariate and multivariable Cox regression analyses. Statistical significance was defined as p < 0.05. RESULTS: At mid-term follow-up, LI-RADS Radiation TRA v2024 showed greater discrimination for PFS than mRECIST and LI-RADS TRA v2017, with no OS difference (p > 0.999). Across all three criteria, late-term response assessment yielded better prognostic stratification for PFS and OS than did mid-term evaluation. At the late-term follow-up, LI-RADS Radiation TRA v2024 was strongly associated with OS (HR = 15.80, 95% CI: 3.34-75.00) and PFS (HR = 6.31, 95% CI: 2.68-14.90). Adding AFs provided no further prognostic improvement. DATA CONCLUSION: LI-RADS Radiation TRA v2024 demonstrates strong prognostic stratification after SBRT, particularly at late-term MRI evaluation. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 2.