ESTRO 2024 - Abstract Book

S4928

Physics - Quality assurance and auditing

ESTRO 2024

nominal dose distribution and recalculated dose distribution from delivery log-files, summarized for all patient cases and all fractions. The blue boxplots indicate the pass rates for individual fields, the orange boxplots represent the pass rates for fields combined into the treatment plan, and the red boxplots denote pass rates for fractions combined into the treatment series.

2. Systematicity of deviations: Spot and couch position deviations for the first fraction delivery and all subsequent fractions show high Pearson correlation coefficients (Fig. 2).

Figure 2 (Left) Pearson correlation coefficient between spot position deviations at fraction one and all subsequent fractions summarized for all patients. (Right). Pearson correlation coefficient between the couch positions for all treatment fields for each patient at the first fraction and their positions in the following 10 fractions, for two spatial directions.

Conclusion:

In our analysis, which covered a wide range of indications treated with proton therapy, we identified two main sources of delivery uncertainty: errors related to couch and spot positions. These uncertainties have a more pronounced effect on single-field treatments and are somewhat mitigated when treatment plans include multiple fields. Importantly, the effect on dose distribution remains consistent regardless of the planning technique chosen, whether SFO or IMPT, as well as the geometry and location of the target volume.

Interestingly, this effect is systematically propagated throughout the treatment course, highlighting the potential for systematic error management.

This work is performed as part of the RAPTOR project, funded by the EU’s Horizon 2020 MSCA, Grant Agreement No. 955956.

Keywords: Proton therapy, log-files, dose recalculation