Author: Carl D. Elliston, Lawrence Koutcher, Michael J. Price, Adam C. Riegel, Michael B Sisti, Tony J.C. Wang, Andy (Yuanguang) Xu 👨🔬
Affiliation: Columbia University Irving Medical Center 🌍
Purpose: An inverse treatment planning software was recently introduced to Gamma Knife radiosurgery. The purpose of this study is to compare the plan quality of the Gamma Knife ”Lightning” with that of the Linac-based SRS planning software.
Methods: Twelve patients with 38 brain mets were imaged with both MR and CT. GTVs and OARs were contoured in Eclipse and imported into GammaPlan. Treatment plans for all targets were generated using the one isocenter per target approach on a Gamma Knife Icon and a Varian Truebeam with 6Xfff beams. All plans were prescribed to 20 Gy and normalized to 100% PTV (defined as 2 mm margin from GTV) coverage. The conformality index, the 12Gy isodose volume and the estimated beam-on time were compared between each treatment plan pair. The cumulative doses to the brain, the brainstem, the optical structures and the cochleae for each patient were also compared.
Results: The average conformality index for the 38 treatment targets is 1.09 for the Truebeam plans and 1.15 for the Gamma Knife plans. The ratio of the 12 Gy isodose volume from Truebeam plan to that from the Gamma Knife plan is about 1 for tumor larger than 4cc but increases with decreasing tumor volume, with a maximum value of 2.6 for a 0.33cc tumor. The doses to the brain structure and the OARs from the Truebeam plans are in general larger than those from the Gamma Knife plans. The maximum and minimum ratio of the estimated beam-on times between the 2 treatment modalities are 23.5 and 2.97 respectively.
Conclusion: With improved target coverage optimization tool from the ”Lightning” inverse planning software, the conformality index of the Gamma Knife plans are comparable to those of the Linac plans. Gamma Knife treatment gives less dose to OARs and normal tissues, especially for small metastases.