S2135
Physics - Inter-fraction motion management and daily adaptive radiotherapy
ESTRO 2026
Bertelsen 1 , Carsten Brink 1,2 , Per Pfeiffer 3,2 , Tine Schytte 3,2 , Faisal Mahmood 1,2 1 Laboratory of Radiation Physics, Department of Oncology, Odense University Hospital, Odense, Denmark. 2 Department of Clinical Research, University of Southern Denmark, Odense, Denmark. 3 Department of Oncology, Odense University Hospital, Odense, Denmark Purpose/Objective: The close proximity of luminal gastrointestinal OARs limit the use of conventional SBRT for pancreatic cancer due to the risk of severe toxicity [1]. MRI-linacs offer a superior soft-tissue contrast, enabling daily plan adaptation to accommodate anatomical changes and reduce the risk of overdosing the adjacent organs [2]. However, unlike baseline planning, adaptive planning must be fast to limit intrafraction motion, potentially compromising plan quality. The aim of this study was to compare baseline and adaptive plan quality in terms of changes in target coverage, OAR dose, and plan conformality for MRI-guided pancreatic SBRT on a 1.5 T MRI-linac. Material/Methods: The study included 152 patients prescribed 50 Gy in five fractions on a 1.5 T MRI-linac at a single institution. All completed adaptive fractions (n=737, 97%) were analyzed. Patient-specific, anisotropic GTV-PTV margins were 4-8 mm depending on respiration motion extent. Baseline planning emphasized sharp intratumoral dose gradients to optimize GTV coverage and mean dose while respecting consensus [3] near- max OAR limits (see planning goals and selected constraints in Table 1). During adaptation, targets and OAR was re-contoured and plans re-optimized. Plan quality was evaluated for target coverage, OAR constraints, and conformality based on hotspots outside the GTV and Paddick's conformity index (CI) [4]. Baseline plans were compared with the mean over each patient’s adaptive plans using Wilcoxon signed- rank tests. Results: Results are shown in Table 1. Relative to baseline, adaptive plans showed small, statistically significant decreases in target coverage as the median GTV V47.5Gy decreased by 1.3% (p<0.001) and PTV V35Gy decreased by 0.5% (p=0.001). Target coverage during adaptation was often restricted by luminal gastrointestinal OAR constraints, which were exceeded in only one patient (see Figure 1). OAR changes were generally small: The median stomach D0.5cc rose by 1.1 Gy (p=0.003). Conformality improved modestly: The median GTV CI increased by 0.1 (p<0.001), while Body-GTV D1cc decreased by 0.2 Gy (p<0.001).
Conclusion: Adaptive planning for pancreatic SBRT on a 1.5 T MRI- linac is clinically robust: Despite time-critical re- optimization, adaptive plans maintain high target coverage and OAR protection with slightly improved conformality. The statistically significant target coverage reductions were small in magnitude and predominantly driven by adherence to gastrointestinal OAR limits. These data support routine on-table adaptation as a reliable strategy for delivering 5- fraction pancreatic SBRT, and they justify continued prioritization of OAR sparing without anticipating meaningful loss of tumor coverage.
Made with FlippingBook - Share PDF online