Room: ePoster Forums
Purpose: To investigate the impact of Volumetric Modulated Arc Therapy (VMAT) delivery errors on 2D and 3D percentage gamma passing rate (%GPs) and dosimetry for nasopharyngeal carcinoma (NPC), and evaluate the relationship among the 2D %GPs, whole-body 3D %GPs, and dosimetry differences.
Methods: Machine output, gantry rotation angle, collimator position, and multi-leaf collimator (MLC) position errors were deliberately simulated for all control points of ten VMAT plans which were treated, and the dose distributions were recalculated. The 3D and 2D %GPs with four different criteria and DVH-based dosimetry differences were obtained by comparing the simulated and original plans in specific regions of interest (ROI), the differences were compared by using a Wilcoxon test, and among the relationships was performed by using Pearson's r correlation analysis.
Results: For MU, gantry angle, collimator position, and MLC leaf shift errors upto 2%, 2Â°, 2mm and 2mm, respectively, the maximum dose difference in all ROIs were less than 2.12%, both the 2D and whole-body 3D %GPs with 3%/2mm criterion were above 90%, and there was no statistically significant correlation among the 2D %GPs, whole-body 3D %GPs, and dosimetry differences. However, the MLC blank open and close errors had a significant effect on dosimetry differences, 3D %GPs and 2D %GPs with the criterion stricter than 3%/3mm (P<0.05), and the Pearson correlation analysis showed a strong correlation (r > 0.8 and P < 0.001) among the 2D %GP, the whole-body 3D %GP, and dosimetry differences.
Conclusion: Compared with MU, gantry angle, collimator position and MLC shift errors, MLC blank close and open errors have a high significantly effect on %GPs and dosimetry for NPC VMAT. In addition, it is not enough to only use 2D %GP or 3D %GP with the 3%/2mm criterion to detect VMAT delivery errors, stricter 2%/2mm criterion should be needed.
Funding Support, Disclosures, and Conflict of Interest: This research work was supported by National Natural Science Foundation of China (Grant No. 81472807).