KEY POINTS
- This prospective single-center R-IDEAL stage IIb clinical evaluation from University Hospital Tübingen, Germany, included 45 patients with hepatocellular carcinoma (HCC) and 51 treated lesions between August 2019 and September 2024. Patients were treated within the prospective MRL-01 (NCT04172753) or MOMENTUM (NCT04075305) studies. Median age was 72 years, and baseline liver function was classified as Child-Pugh A in 33 patients (73%), B in nine (20%), and C in three (7%). More than half the cohort (24/45, 53%) had received previous liver-directed treatment, and 13 patients had targets close to the stomach or bowel.
- Treatment was delivered using a 1.5-T MR-Linac with daily MRI guidance and respiratory motion monitoring. Simulation included both 4DCT and MRI, while target delineation incorporated additional diagnostic imaging when necessary. Individualized 3–6 mm PTV margins were applied based on expected positional variability. Importantly, all fractions used the Adapt-to-Position (ATP) workflow, involving virtual couch shifts rather than daily target recontouring and full plan reoptimization. During beam delivery, 5-Hz cine MRI enabled monitoring of target position and potential interruption when coverage was insufficient.
- Prescription doses were individualized according to tumor characteristics, functional liver reserve, and adjacent OAR exposure. Median prescribed dose was 35 Gy (IQR 35–40 Gy), with 40/45 patients (88.9%) receiving five fractions. The most common regimens were 35 Gy in five fractions (17 patients), 50 Gy in five fractions (ten patients), and 40 Gy in five fractions (seven patients). Median GTV volume was 16.4 cm³, with median GTV D98 of 38.3 Gy and corresponding BED10 D98 of 67.6 Gy (IQR 61.9–102.0). Mean dose to uninvolved liver was 7.2 Gy, with median liver V10Gy of 24.0%.
- Local control was high during the first two years after treatment. At a median follow-up of 28 months (95% CI 21–35), estimated local control was 94% at 12 months (95% CI 86–100%) and 90% at 24 months (95% CI 79–100%). These results were achieved despite individualized prescriptions and the frequent use of regimens with BED10 below 100 Gy. However, the study did not include a conventional image-guided SBRT comparison group, and the limited number of local failures prevented reliable assessment of dose-response relationships.
- Overall survival remained relatively favorable, but progression outside the irradiated lesion was common. Estimated 12- and 24-month overall survival was 84% and 70%, respectively, while progression-free survival declined from 54.7% at 12 months to 29% at 24 months. Twenty-four patients (53%) developed intrahepatic progression outside the treated lesion, and three subsequently underwent repeat SBRT for new intrahepatic disease. These outcomes highlight the difference between durable treatment-site control and the broader risk of recurrence within the remaining liver.
- MR-guided SBRT was also used as a bridge to liver transplantation. Among 14 patients treated with bridging intent, eight (57%) subsequently underwent transplantation during the reported follow-up. The technique offered noninvasive treatment without catheter placement or thermal ablation, but transplantation rates depended on eligibility, waiting-list status, disease progression, and organ availability. Without a comparator group, the study cannot establish whether MR-guided SBRT improves transplantation probability or prevents wait-list dropout relative to TACE, conventional SBRT, or thermal ablation.
- No acute CTCAE grade ≥3 adverse events were documented, supporting the observed short-term tolerability of this approach in a cohort with variable baseline liver function. Nevertheless, low-grade toxicities were not systematically recorded, and patient-reported outcomes will be reported separately. All patients received ATP rather than Adapt-to-Shape treatment, so the study cannot quantify the additional benefit of daily anatomical recontouring or dose reoptimization. The single-center population, heterogeneous prescriptions, lack of a comparison group, and limited longer-term follow-up prevent conclusions about comparative effectiveness or late hepatic toxicity.
CLINICAL TAKEAWAY
MR-guided SBRT achieved encouraging two-year local control with no reported severe acute toxicity in a prospective cohort of patients with HCC, including individuals previously treated with liver-directed therapies and those awaiting transplantation. These findings support MR-guided SBRT as a feasible local treatment approach when surgery or other ablative interventions are unsuitable, but do not demonstrate superiority over conventional SBRT or other liver-directed modalities. Because all treatments used Adapt-to-Position, the study should not be interpreted as evidence for the clinical benefit of full daily adaptive replanning.