3449 - MR-Guided Adaptive Radiotherapy for Primary Liver Cancer: Impact of Online Isotoxic Prescriptions near Organs-at-Risk

02:30pm - 03:45pm PT

Hall F

Screen: 14

POSTER

Presenter(s)

Haruo Inokuchi, MD, PhD - osaka metropolitan university, Osaka-city, Osaka

H. Inokuchi, N. Mukumoto, N. Mukumoto, M. Sakagami, M. Yamagishi, H. Itoyama, Y. Hata, N. Hamaura, K. Hayashi, R. Ogino, and K. Shibuya; Department of Radiation Oncology, Graduate School of Medicine, Osaka Metropolitan University, Osaka, Japan

Purpose/Objective(s): MR-Linac is increasingly used for ablative dose delivery via MR-guided adaptive radiotherapy (MRgART), enhancing accuracy and dose coverage. ART shows promise in improving outcomes for primary liver cancer near hepatobiliary and luminal gastrointestinal organs-at-risk (OARs) through on-table adaptation, motion management, and online isotoxic prescriptions with MR imaging guidance. This study reports 12-month local control and toxicity outcomes in patients with primary liver cancer, categorized by anatomical tumor location, using a 1.5T MR-Linac for daily adaptation.

Materials/Methods: Between May 2022 and March 2024, 35 patients with primary liver cancer received 48.0–52.5 Gy in 5 fractions (BED₁₀ = 95–107 Gy₁₀) to the internal target volume. A novel compression belt workflow on the 1.5T MR-Linac system was used. A balanced turbo field echo (btFFE) 2D cine motion scan assessed tumor motion at each fraction for internal target volume adjustments. Online isotoxic prescriptions were adapted daily based on tumor proximity to OARs. Patients were categorized into two groups: those requiring online isotoxic adaptive therapy (aOARs group: n=23) and those who did not (nOARs group: n=12). Coverage trade-off was assessed using the coverage compromise index (CCI), calculated as the minimum dose received by the hottest 99% of the PTV (D99) divided by the prescription dose. Patient demographics, dose fractionation, toxicity, and 12-month clinical responses were analyzed based on RECIST 1.1 criteria.

Results: The median follow-up was 18 months (range: 10–31 months). Among 35 patients, 30 (85%) had hepatocellular carcinoma (HCC) and 5 (15%) had cholangiocellular carcinoma (CCC). Target volumes were within 2 cm to luminal structures or central biliary trees in several cases, including the esophagus (n=4), common bile duct (n=8), stomach (n=6), heart (n=3), bowel (n=10), and duodenum (n=6). The mean CCI (95% CI) was significantly lower in the aOARs group at 0.90 (0.88–1.02) versus 0.98 (0.93–1.12) in the nOARs group (p=0.04). However, mean IGTV dose (%) remained comparable at 101.5±2.28 and 106.2±1.89, respectively (p=0.19). The 12-month local control rates were 92.0% and 85.5%, respectively, in the aOARs group vs. the nOARs group. No grade 3+ toxicities were observed. Local control and progression-free survival did not significantly differ between groups.

Conclusion: MR-Linac ART provided promising local control and overall survival with no severe toxicity in primary liver cancer patients, even when tumors were near hepatobiliary and luminal gastrointestinal OARs. Online individualized isotoxic prescriptions, adjusted based on tumor location, may minimize toxicity while maintaining effective ITV dose coverage. Longer follow-up is needed for long-term outcome assessment.