ESTRO 2024 - Abstract Book

S4740

Physics - Quality assurance and auditing

ESTRO 2024

The aim of this study is to assess the results of the pre-treatment verifications performed for prostate SBRT plans adapted in an online fashion.

Material/Methods:

Prostate SBRT treatments are planned in our department using an online adaptive method developed and validated by our group (Pract Radiat Oncol. 2022 Mar-Apr;12(2):e144-e152). The first 50 prostate patients treated with the PACE B schedule (5 × 7.25 Gy) were included, i.e., 250 adapted SBRT plans were collected in this study. These were computed on the daily CBCT scan using the Acuros XB v. 16.1 algorithm of the Varian Eclipse treatment planning system (TPS). Adapted plans consisted of a single VMAT with 6 MV FFF energy from a Varian TrueBeam linac. Pre-treatment verification of the adapted plan created in each treatment session was performed using the Mobius 3D v. 3.1 secondary dose calculation program (M3D). Acuros and M3D reported dose-to-medium. Tuning of the dynamic leaf gap correction (DLGC) parameter of M3D was performed using a suite of 16 test plans for typical sites (prostate, lung, head and neck, brain) treated using the 6 MV FFF VMAT technique. Comparisons of the MD3-calculated point doses vs. measurements in a homogeneous Mobius Verification Phantom (MVP) were used to select the optimal DLGC value in M3D. For each online adapted plan, a global 3D gamma comparison between the Eclipse 3D dose and the M3D dose in the patient CBCT was performed. Gamma passing rates (GPRs) for the whole external patient contour (Body) and the PTV were recorded. In addition, the target mean dose (Dmean) and target coverage (D90%) differences between the Eclipse and Mobius doses were analyzed.

Results:

Results:

1. DLGC of -1 mm was found from the set of 16 test plans. Average percent dose difference (±SD) between the MD3 doses and measurements in the MVP was -0.1% ± 2.0% (range: -4.6% to 3.7%). The derived confidence limit was 4.1% (mean + 1.9× SD). 2. From the results obtained in 1), we decided to use ±5% as the dose criterion for the gamma index analysis for comparing clinical plans computed by the Eclipse vs. MD3 calculations. 3. For the 250 online adapted plans, 5%/2 mm GPR values (mean ± SD) of 100% ± 0% and 99.7% ± 1.3% (range: 86.5% to 100%) were obtained for the Body and PTV structures, respectively (Fig 1). Only a plan reported GPR < 90% for the PTV. 4. For the 250 online adapted plans, Dmean and D90% values (mean ± SD) of -1.5 ± 1.3% (range: -4.8% to 2.5%) and -3.2% ± 1.2% (range: -6.0% to 3.7%) were obtained (Fig 2). Only a plan reported GPR < 90% for the PTV structure. Twelve plans (4.8% of plans) reported D90% < -5%.