1104 - Impact of Proton Therapy Delivery Systems on Circulating Blood Dose

Purpose/Objective(s): Circulating blood dose has been associated with systemic toxicities and immune-related effects in radiation therapy, yet remains poorly characterized in proton therapy, particularly with respect to machine-specific delivery dynamics. This study quantifies and compares circulating blood dose across three proton therapy delivery system for intensity-modulated proton therapy (IMPT) and proton arc therapy (PAT) using a dynamic simulation framework.

Materials/Methods: Ten patients with head-and-neck cancer were retrospectively planned using both IMPT and PAT. For each patient and technique, identical treatment plans were virtually delivered using three machine models: a synchrocyclotron (IBA Proteus®ONE, P1), a cyclotron (IBA Proteus®Plus, PPlus), and a synchrotron (HITACHI ProBeat®). A dynamic hematological dose simulation framework (HEDOS) was used to compute accumulated dose to circulating blood while accounting for realistic blood flow kinetics. Machine-specific delivery sequence models incorporating realistic timing structures of irradiation sequences were employed. Blood dose–volume histograms were generated, and dose to 90% of blood volume (D90), mean blood dose (Dmean), and dose to 10% of blood volume (D10) were extracted. Statistical comparisons between machine models were performed using the Wilcoxon signed-rank test.

Results: For IMPT, blood doses for P1 were 0.222 ± 0.122 GyE for D90, 0.245 ± 0.129 GyE for Dmean, and 0.267 ± 0.137 GyE for D10. Corresponding values for PPlus were 0.214 ± 0.121 GyE, 0.245 ± 0.129 GyE, and 0.276 ± 0.139 GyE, while HITACHI ProBeat® yielded 0.214 ± 0.120 GyE, 0.245 ± 0.130 GyE, and 0.276 ± 0.139 GyE. For PAT, blood doses with P1 were 0.176 ± 0.100 GyE for D90, 0.195 ± 0.109 GyE for Dmean, and 0.215 ± 0.117 GyE for D10. PPlus demonstrated 0.167 ± 0.097 GyE, 0.198 ± 0.110 GyE, and 0.230 ± 0.124 GyE, while HITACHI ProBeat® showed 0.172 ± 0.099 GyE, 0.196 ± 0.109 GyE, and 0.221 ± 0.119 GyE. Across both techniques, statistically significant differences between machine models were observed for D90 and D10 (p < 0.05), whereas Dmean did not differ significantly since same plan has been utilized.

Conclusion: Machine-specific delivery characteristics significantly influence circulating blood dose distributions in proton therapy. Although Dmean remained comparable, PPlus and HITACHI ProBeat® demonstrated reduced low-dose blood exposure and increased high-dose exposure relative to P1. These findings highlight the importance of incorporating realistic, machine-specific delivery timing structure into circulating blood dose modeling and may have implications for systemic toxicity mitigation.