Dynamic Sequencing Optimization for Synchrocyclotron-Based Proton Arc Therapy
Abstract
Purpose
Spot-scanning proton arc (SPArc) therapy combines the physical advantages of proton therapy with continuous arc delivery. However, most current SPArc planning approaches assume static beam delivery and do not account for the synchrocyclotron-based proton therapy systems with rotating gantries. This mismatch between nominal planning assumptions and actual treatment delivery may result in dose deviations, particularly for highly precise treatments such as stereotactic radiosurgery (SRS).
Methods
A five-step sequencing optimization framework was developed to bridge static SPArc planning and dynamic delivery on a synchrocyclotron-based system. The framework consists of: Machine-specific modeling of delivery times, spot disassembling, embedding delivery timing information into control points, spot-weight tuning, and reconstruction of optimized energy-layer sequences. Five patients with multiple brain metastases SRS cases were retrospectively analyzed. Delivery accuracy, dosimetric plan quality, and treatment efficiency were evaluated using reconstructed dose distributions derived from virtual machine logfiles.
Results
The new algorithm significantly improved delivery accuracy while maintaining plan quality and efficiency. For the total gross tumor volume, the mean absolute D98 deviation between planned and reconstructed doses decreased from 77.4 ± 66.2 cGyE (4.6 ± 3.7%) in static SPArc plans to 9.6 ± 7.2 cGyE (0.5 ± 0.3%) after optimization. For the worst-case metastasis, D98 deviation decreased from 184.4 ± 127.8 cGyE (9.7 ± 5.6%) to 19.0 ± 13.3 cGyE (1.0 ± 0.6%), while D2 deviation decreased from 148.4 ± 103.5 cGyE (6.8 ± 4.2%) to 13.2 ± 10.2 cGyE (0.6 ± 0.4%). Target coverage and normal brain dose metrics were not statistically different between plan types (p > 0.05). Total dynamic delivery times differed by less than 1 second.
Conclusion
This sequencing optimization framework effectively mitigates inconsistencies between static planning and dynamic delivery for synchrocyclotron-based proton arc therapy, substantially improving delivery accuracy without compromising dosimetric quality or treatment efficiency.