Improving delivery time for breath-hold proton therapy via combination of advanced spot positioning algorithms and ridge filter

Pencil beam scanning (PBS) is particularly susceptible to respiration-induced tumor motions, which can result in unnecessary dose exposure of healthy tissues or inadequate dosing of the target. Various motion mitigation techniques, such as rescanning, gating, and breath-hold, are used to address this issue. However, breath-hold's widespread use in proton therapy, particularly for scanned delivery, is limited due to the need for rapid treatment delivery. In this study, we aimed at assessing the feasibility of delivering an entire field dose within one single breath-hold in PSI's Gantry 2 by examining the advantages and limitations of combining a novel designed Ridge filter (RF) with various spot placement techniques to reduce the treatment delivery time, trying to achieve delivery times per field (in case of small tumors) or per patch (in case of large tumors) of approximately less than 20 seconds. To achieve our goal, treatment plans were created on a cohort of 12 breath-hold cases using three different configurations with three different spot placement algorithms. Configurations performances were then compared in terms of plans quality, clinical acceptability and delivery time reduction for each patient. Results indicated that the configuration that showed the best performance was the one that combined the RF with a spot placement algorithm based on a fixed grid, which met all medical prescriptions and robustness criteria. Furthermore, it was demonstrated that the use of the RF with different spots positioning algorithms significantly reduced the dead time during dose delivery compared to the clinical setups currently used at PSI's proton therapy center. Finally, in terms of breath-hold feasibility, it was determined that, out of the clinically acceptable plans we created, the delivery of an entire field dose within a single breath-hold was feasible for small tumors, while for extended tumor volumes performing one breath-hold per patch was a feasible solution.