Reproducibility of the lung anatomy under Active Breathing Coordinator control: Dosimetric consequences for scanned proton treatments.

Background and purpose: The treatment of moving targets with scanned proton beams is challenging. For motion mitigation, an Active Breathing Coordinator (ABC) can be used to assist with breath-holding. We evaluated the robustness of scanned proton therapy against anatomical uncertainties when treating lung patients during ABC controlled breath-hold. Materials and methods: Four subsequent MRIs of five healthy volunteers were acquired under ABC controlled breath-hold during two simulated treatment fractions. Deformation vector fields between these MRIs were used to deform CTs of three non-small-cell lung cancer patients. Per patient eight cases with different tumour sizes and locations were modeled. Intensity-modulated proton plans were created and split into sub-plans of 20 seconds duration (assumed breath-hold duration), and recalculated on the deformed CTs. Results: Dosimetric evaluation showed <2% target coverage loss (V95% [%]) for 19/24 cases. Simulated tumours in the caudal regions showed a loss of V95% up to 6.1%. Organs at risk doses differed little compared to the planned doses (V5Gy [%] <1% for the heart and the lungs minus GTV, <1.4 Gy difference in D0.1cc [Gy]) to the spinal cord and esophagus). Conclusions: When treating under ABC controlled breath-hold, robustly optimized IMPT plans lack robustness to caudally located lung tumours. For most other cases anatomical variations between repeated ABC breath-holds have limited dosimetric consequences.