ESTRO 2024 - Abstract Book

S4919

Physics - Quality assurance and auditing

ESTRO 2024

ionisation chamber at the centre. Each patient was measured five times in each accelerator on separate days and by different operators to account for various uncertainties, including device setup variations, inter-operator variability, MLC positioning, and beam output fluctuations. For each measurement, we calculated the gamma passing rates (global 3D gamma, absolute dose differences, 10% of maximum dose threshold) for two distinct criteria: 3% - 2 mm, as recommended by the AAPM TG-218, and a higher sensitivity criterion of 2% - 2 mm. Additionally, we corrected the ionisation chamber dose for daily output variations. We compared the PSQA results between the two accelerators for each treatment type. We determined the mean value and dispersion (σ) of the recorded parameters in each patient and accelerator, and employed paired t-tests to assess the significance of the differences (p<0.05). The dispersion values were used to evaluate the repeatability and reproducibility of the PSQA process, while the mean values served as the basis for comparing the PSQA results.

Results:

Figure 1 shows the repeatability and reproducibility of PSQA measurements in each accelerator for the considered locations. The red rectangles indicate the statistically significant differences. The variability depended on the treatment type but remained within 2% for all cases. The patterns were similar across both accelerators and only the gamma passing rates for the prostate and the ionisation chamber dose for lung SBRT differed. However, these discrepancies were below 1%, so the measurements can be considered equally repeatable and reproducible in the two accelerators.

Figure 2 shows the PSQA results with the statistically significant differences highlighted by red rectangles. Gamma 3%-2 mm passing rates were comparable and well above the 95% tolerance level recommended by the AAPM TG-218 for all locations except brain HSRT. Still, they were within the 90% action level that we consider for these treatments since the spatial resolution of the ArcCHECK device is low for these field sizes. The stricter gamma 2%-2 mm passing rates differed for head and neck (0.9%), prostate (1.9%), and brain HSRT (1.7%). However, the clinical impact should be minimal as these differences remained below 2%, and the passing rates were systematically above the acceptance criteria. Similarly, the ionisation chamber doses also presented some differences, but none exceeded 1.5% for any patient and all results were within the 5% tolerance.