ESTRO 2024 - Abstract Book

S4658

Physics - Optimisation, algorithms and applications for ion beam treatment planning

ESTR0 2024

5MU/3MeV. Aside from minimum MU and layer-spacing, these plans had identical optimization objectives and robustness settings as the nominal plans.

All RapidScan plans were compared in silico to the nominal plan in terms of target coverage, OAR sparing, field specific doses, and robustness. Delivery times for the nominal and RapidScan plans were also compared using the predicted times calculated by the prototype. For 5 of the patients, verification plans were created for the nominal and the fastest (estimated) RapidScan plan. These two sets of plans, per patient, were delivered to a planar ion-chamber patient-specific QA array on a Varian ProBeam Compact machine. Delivery time was measured for individual fields and plan quality was assessed using a gamma analysis. The patient cohort contained SFO and MFO plans with per-fraction dosages of 2-7Gy and target volumes of 83-2299cc.

Results:

In silico, all treatment plans provided similar and clinically acceptable dose distributions. Minor variations in plan hotspots and point dose maximums to some OARs were observed when changing minimum MU’s and layer-spacing, but most dose-volume constraints varied by only ±1% Table 1 presents plan statistics and calculated treatment time changes compared to the nominal plan. Increasing minimum MUs led to the most significant treatment time reduction: 28±12% and 32±16% for plans with 3MU and 5MU minimums, respectively. Adjusting energy layer spacing to 1MeV increased delivery time by 84±24%, while changing to 5MeV decreased it by 17±6%.

For the 5 patients selected for verification measurements, the fastest calculated treatment times were for the 5MU/3MeV plans. The measured delivery time of these fields are shown in Fig. 1 and were approximately 50% quicker than the nominal fields (40.6±10.4 vs. 81.9±24.7 seconds). The measured gamma indices at 3%/3mm were 99.3±1.7% for nominal and 99.1±1.8% for RapidScan plans. At 2%/2mm, these were 96.4±3.1% and 96.0±3.5%, respectively.