ESTRO 2025 - Abstract Book

S3254

Physics - Intra-fraction motion management and real-time adaptive radiotherapy

ESTRO 2025

Material/Methods: The study included 20 consecutive patients with locally advanced pancreatic cancer, treated in five fractions on a 1.5 T MRI-Linac. 3D target motion data were extracted from the cinematic MR images acquired during treatment using an in-house tracking algorithm [2]. Adaptive treatment plans followed a consensus protocol for treating pancreatic tumours in 5x10 Gy using only a 2 mm PTV margin [3]. Treatment sessions were simulated using various gating and drift windows, defined as an isotropic 2 to 5 mm expansion of the GTV (see Figure 1). The delivered dose was accumulated for each simulated session to monitor the target coverage.

Results: The treatment delivery efficiency, number of baseline shift re-plans, and changes in target coverage (GTV D98% and mean dose) are shown as a function of the gating and drift window size for each of the 100 adaptive fractions in Figure 2. As expected, the delivery efficiency increases and target dose coverage decreases with increasing gating window size. Reducing the drift window size removes the worst target coverage reduction outliers but can lead to excessive usage of baseline shift re-planning. Using 4 mm for both the gating and drift window size provided a compromise, maintaining reasonable target coverage compared to smaller window sizes and achieving over 75% delivery efficiency in 95% of the fractions and a maximum of one baseline shift re-plan.