S1752
Physics - Dose prediction/calculation, optimisation and applications for particle therapy planning
ESTRO 2026
technique that may reduce normal tissue toxicity through spatial fractionation of proton beams. This study investigates the influence of multislit collimator (MSC) slit width and center-to-center (CTC) distance on robust target coverage and peak-to-valley dose ratio (PVDR) in single-beam pMBRT. Material/Methods: Simulations were performed in RayStation 2024B research version using a Monte Carlo (MC) dose engine including MSCs. A homogeneous cubic water phantom with a spherical clinical target volume (CTV) of diameter 6.0 cm was modelled. Nineteen pMBRT plans were generated with slit widths of 0.04–0.20 cm and CTC distances of 0.2–0.6 cm, plus an open beam reference. Four single-beam geometries were simulated, representing different entrance-to-target depths (6–24 cm) and angles. Robust composite minmax optimization (CMRO) with 28 scenarios (5 mm setup, 3% range uncertainty) was applied. All treatment plans were assessed for target dose coverage with the constraints D98 ≥ 95% and D2 < 107%, using the voxelwise minimum and maximum dose distributions. Next to this, PVDR at beam entrance and CTV center were analyzed. Tumor control probability (TCP) was calculated for illustrative purposes. Results: A trade-off was observed between entrance PVDR and robust target coverage. High PVDR values at entrance reduced coverage, while sufficient coverage reduced PVDR, see Figures 1 and 2. CTV depth and beam geometry were key factors: larger depths and oblique angles reduced entrance PVDR (<2.7) and coverage. Intermediate depths (6–15 cm) provided the most favorable balance. TCP was comparable between pMBRT and open beams when PVDR in the target was <1.2, but decreased with heterogeneous target doses.
RobOpt improved OAR sparing compared with PTV- based planning but did not prevent target coverage degradation. Enlarging setup uncertainty to 5 mm enhanced target robustness but worsened OAR doses. Conclusion: Despite RobOpt, proton plans lost a substantial portion of their dosimetric benefit after three weeks of treatment due to anatomical evolution, whereas photon plans remained more stable. Incorporating scheduled adaptive replanning or anatomical robustness into optimization is essential to maintain the proton advantage in HN radiotherapy. References: [1] Langendijk, J. A., Hoebers, F. J., De Jong, M. A., Doornaert, P., Terhaard, C. H., Steenbakkers, R. J., ... & Schuit, E. (2021). National protocol for model-based selection for proton therapy in head and neck cancer. International journal of particle therapy, 8(1), 354-365. Keywords: protons, robust optimization, anatomical changes Impact of center-to-center distance and slit width on robust target coverage and PVDR for proton minibeam radiation therapy using RayStation Anna C Prins 1,2 , Erik Traneus 3 , Mischa S Hoogeman 2,1 , Kelvin Ng Wei Siang 1,2 1 Department of Medical Physics & Informatics, Holland Proton Therapy Center, Delft, Netherlands. 2 Department of Radiotherapy, Erasmus MC Cancer Institute, Rotterdam, Netherlands. 3 RaySearch Laboraties AB, RaySearch Laboratories AB, Stockholm, Sweden Digital Poster 3340
Purpose/Objective: Proton minibeam radiation therapy (pMBRT) is a novel
Made with FlippingBook - Share PDF online