4D-VMAT dose calculation using treatment-specific respiratory motions

The interplay between respiration-related tumour motion and modulated treatment delivery such as volumetric modulated arc therapy (VMAT) could contribute to differences between planned and delivered radiation dose distributions. The purpose of this work was to develop and experimentally validate a method to calculate the VMAT dose distribution in 4-dimensions (4D-VMAT dose calculation), accounting for tumour motion using the respiratory motion waveform recorded during each treatment fraction. The 4D-VMAT dose calculation method relies on the time sampling of machine parameters and respiratory amplitude-phase recorded during each fraction in the trajectory log file and motion waveform file, respectively. These files are recorded by a Varian TrueBeam LINAC with a respiratory motion management system. The phase information is used to sort the trajectory log file into ten discrete respiratory phases from which ten phase-specific VMAT plans are created and imported into the treatment planning system. The doses are re-calculated and summed to obtain the 4D-VMAT dose distribution. To experimentally validate the 4D-VMAT dose calculation, ten VMAT plans were created and delivered to a QUASAR respiratory motion phantom. The phantom was programmed with realistic respiratory motion waveforms from the QUASAR software database and motion amplitudes up to 15 mm were used. Each plan was delivered three times and a different waveform was used each time. Gafchromic EBT3 film was placed within the phantom to measure the delivered dose distribution. The 4D-VMAT calculated dose distributions were compared to the measured dose distributions and the agreement was assessed using gamma analysis with the 3%/3 mm dose-difference distance-to-agreement acceptance criteria and isodose and dose profile comparisons. The validation comparisons show that the 4D-VMAT dose calculation is able to reproduce fraction-specific dose distributions. There was a significant (p