Evaluation of the LI-RADS radiation treatment response assessment v2024 in comparison with the modified RECIST.

Abstract

Conventional radiologic response criteria inadequately address imaging features and delayed responses after radiation-based therapies, limiting clinical decisions. This study evaluated treatment response to radiation-based locoregional therapies using modified RECIST (mRECIST), LI-RADS treatment response algorithm (LR-TRA) v2017, and radiation LR-TRA v2024, assessed changes in response categories during follow-up, and evaluated their association with overall survival (OS). This single-center retrospective study included consecutive patients with hepatocellular carcinoma treated with transarterial radioembolization or external beam radiation therapy between 2011 and 2023. Two abdominal radiologists independently assessed the first radiologic response 3 months post-treatment using the mRECIST, LR-TRA v2017, and radiation LR-TRA v2024. The changes in the mRECIST and LR-TR categories during follow-up were evaluated. The Cox proportional hazard model was used to assess the association between response categories and OS. A total of 115 patients were included (66.6 ± 12.9 years; 73.9% [85/115] men). Complete response (CR), partial response, stable disease, and progressive disease (PD) based on the mRECIST were 36.5% (44/115), 26.1% (30/115), 17.4% (20/115), and 20% (26/115), respectively. In LR-TRA v2017, the nonviable, equivocal, and viable rates were 48.7% (56/115), 13% (15/115), and 38.3% (44/115), respectively. In LR-TRA v2024, the nonviable, nonprogressing, and viable rates were 47% (54/115), 51.3% (59/115), and 1.7% (2/115), respectively. During follow-up, the proportion of LR-TR nonprogressing decreased, while that of LR-TR nonviable increased over time. Achieving nonviable disease during follow-up was strongly associated with improved OS (p < 0.001). Radiation LR-TRA v2024 effectively reflected a typical delayed response pattern after radiation-based therapy. Question Conventional radiologic response criteria do not fully address the imaging features and delayed response pattern after radiation-based therapies, limiting clinical decision making. Findings The new LI-RADS radiation treatment response assessment v2024 (LR-TRA v2024) effectively reflects unique post-treatment imaging features and delayed response to radiation-based treatment, facilitating management and communication. Clinical relevance The new radiation LR-TRA v2024 captured the evolving nature of post-radiation imaging features by incorporating the concept of temporal change and LR-TR nonprogressing disease to allow for a watch-and-wait protocol.