Clinical impact of dosimetric changes for volumetric modulated arc therapy in log file-based patient dose calculations

Abstract Purpose A logfile-based method cannot detect dosimetric changes due to linac component miscalibration because logfiles are insensitive to miscalibration. Herein, clinical impacts of dosimetric changes on a logfile-based method were determined. Methods and materials Five head-and-neck and five prostate plans were applied. Miscalibration-simulated logfiles were generated by inducing a linac component miscalibration into the logfile. Miscalibration magnitudes for leaf, gantry, and collimator at the general tolerancelevel were ±0.5 mm, ±1°, and ±1°, respectively, and at a tighter tolerancelevel achievable on current linac were ±0.3 mm, ±0.5°, and ±0.5°, respectively. Re-calculations were performed on patient anatomy using logfile data. Results Changes in tumor control probability/normal tissue complication probability from treatment planning system dose to re-calculated dose at the general tolerancelevel was 1.8% on planning target volume (PTV) and 2.4% on organs at risk (OARs) in both plans. These changes at the tighter tolerancelevel were improved to 1.0% on PTV and to 1.5% on OARs, with a statistically significant difference. Conclusions We determined the clinical impacts of dosimetric changes on a logfile-based method using a general tolerancelevel and a tighter tolerancelevel for linac miscalibration and found that a tighter tolerancelevel significantly improved the accuracy of the logfile-based method.