Room: Exhibit Hall | Forum 3
Purpose: To assess the dosimetric impact of three different clinical techniques for stereotactic radiosurgery (SRS) treatment of multiple brain metastases in order to identify the most optimal approach.
Methods: Five patients with 6-9 metastases were each planned for SRS treatment with Leksell GammaKnife, BrainLab multi-aperture dynamic conformal arcs (Elements), and Varian volumetric arc therapy (VMAT). Prescriptions ranged from 15-24Gy in a single fraction, dependent upon target volume size. Each case was planned to be clinically feasible in terms of achieving maximum conformality while satisfying all organ-at-risk dose limits. Quantitative comparison parameters included RTOG conformity index (CIRTOG), Paddick conformity index (CIPaddick), mean dose to the brain (Brain_Dmean) and volume of the brain receiving at least 12Gy (V12Gy). GammaKnife plans served as the reference dataset to compare Elements and VMAT against. Statistical analysis was performed via the Wilcoxon signed rank test for paired datasets to identify any significant differences (p<0.05) between the extracted parameters.
Results: The target volumes across the total of 35 studied metastases ranged from 0.12 - 8.3cmÂ³. The total target volume per patient ranged from 9.4 - 23.7 cmÂ³. The median CIRTOG for GammaKnife, Elements, and VMAT was 1.52, 1.36, and 1.25, and the median CIPaddick was 0.60, 0.71, and 0.79, respectively. The median Brain_Dmean for GammaKnife, Elements, and VMAT was 324.8cGy, 370cGy, and 450.2cGy; the median V5Gy was 179.5cmÂ³, 292cmÂ³, and 323.1cmÂ³; and the median V12Gy was 39.4cmÂ³, 50.6cmÂ³, and 36.1cmÂ³, respectively. Elements and VMAT demonstrated statistically significant improvements in CI compared to GammaKnife, meanwhile neither V12Gy, V5Gy, nor Brain_Dmean exhibited significant differences across plan comparisons.
Conclusion: Based on this preliminary analysis, VMAT plans resulted in conformity indices closest to 1 and in general demonstrated greater low dose spread. However, the differences in brain dose were not statistically significant, which is likely due to the limited sample size.