ESTRO 2025 - Abstract Book

S3621

Physics - Quality assurance and auditing

ESTRO 2025

size/shape, a lattice configuration with 1.5cm sphere diameter spaced 6cm apart, or a higher-density lattice with 1.0cm sphere diameter and 4cm spacing was applied. For each lattice configuration, two plans were generated with peak-dose/valley-dose prescriptions of 66.7Gy/20Gy and 66.7Gy/10Gy, increasing peak-to-valley-dose-ratio (PVDR) from ~350% to ~700%. A total of 20 plans were created and delivered using 6MV-beams on a Varian TrueBeam Linac using a Millennium120-MLC. Dose delivery accuracy was verified using SNC-ArcCHECK and ScandiDos-Delta4 3D-diode arrays, with passing rates analyzed under various evaluation criteria. Additionally, the in-house Brems Monte-Carlo dose calculation system 1 was commissioned as a QA tool for recalculation of patient-delivered doses from machine Trajectory-Log-Files (TLF). For Brems commissioning purposes, TLF-based calculated doses on the Delta4 phantom were compared with measured doses and planned doses from RayStation. An independent end-to end test was performed using an IROC Head&Neck phantom, equipped with 8 TLDs and two film planes and planned with 66.7Gy/20Gy dose prescription. Results: Median (10 th –90 th percentiles) passing rates for the 66.7Gy/10Gy vs 66.7Gy/20Gy prescriptions were: (1) For γ (2%,2mm) evaluation criteria, ArcCHECK achieved 97.6%(94.3%-99.2%) vs 98.3% (94.4%-99.5%) while Delta4 results averaging 2.3% and 0.5% higher than ArcCHECK; (2) For γ (3%,2mm) evaluation criteria, ArcCHECK achieved 99.3%(97.4%-99.9%) vs 99.5%(98.1%-100.0%). ArcCHECK γ (3%,2mm) passing rates were above 95% in 9/10 patients for the 66.7Gy/10Gy and in all patients for the 66.7Gy/20Gy prescription (Table 1). The 8 TLDs embedded in the IROC phantom met the ±7% criteria while the sagittal/axial films passed the γ (7%,4mm) criteria. TLF-based calculated dose to the Delta4 was in good agreement with planned dose and with Delta4 diode measurements, with an absolute dose difference of < 0.7 Gy (Figure1b).