ESTRO 2025 - Abstract Book

S2808

Physics - Dose prediction, optimisation and applications of photon and electron planning

ESTRO 2025

CC001 protocol and 30Gy in 10 fraction prescription. Replanning was performed on a TrueBeam machine (Millennium MLC) using four plan configurations: (P1) 2-Arc VMAT using clinical RapidPlan model; (P2) 2-Arc VMAT using user-defined objectives template (UDOT); (P3) 1-Arc VMATp using dynamic collimator rotation, 6 STAMPs with patient-specific gantry angles, and UDOT; and (P4) 1-Arc VMATp using dynamic collimator rotation, 6 STAMPs using templated STAMP angles, and UDOT. The UDOT was designed to maximize generalized Equivalent Uniform Dose (gEUD) objectives. The templated STAMP locations were developed based on preliminary testing to optimize angle selection using control point maps. All 40 plans were normalized to V95%=100% coverage and analyzed using NRG CCOO1 dose metrics were reported. Results: All four plan types demonstrated acceptable plan quality, with quality criteria passing rates of 77% for VMAT, 90% for VMAT using UDOT, 99% for VMATp + patient-specific STAMPs + UDOT, and 100% for VMATp + template STAMPs + UDOT. The failed quality metrics per protocol were still within acceptable variation. VMATp maintained similar target dose coverage as traditional VMAT while reducing the planning time by 40% and allowing single arc delivery. P4 achieved a higher D98% of 28.00Gy compared to 27.11Gy in the P1 (p<0.05). Hippocampal D100% was significantly lower in P4 (6.53 Gy) compared to P1 (7.43 Gy, p<0.05), while hippocampal D0.03cc was comparable between P4 and P1 (12.72 Gy vs. 12.68 Gy). For other critical structures (Optic Chiasm, Optic Nerves, and Eyes), P4 showed statistically significant dose reductions compared to P1-2 plans. The templated STAMPs (P4) produced results slightly better or comparable to the patient-specific STAMPs (P3).