Effect of prescription isodose line on the brain dose for radiosurgery of brain metastasis
PO-1495
Abstract
Effect of prescription isodose line on the brain dose for radiosurgery of brain metastasis
Authors: Krayenbuehl|, Jerome(1)*[jerome.krayenbuehl@usz.ch];Mariangela|, Zamburlini(1);Stephanie|, Tanadini-Lang(1);Matthias|, Guckenberger(1);Nicolaus|, Andratschke(1);
(1)University Hospital Zürich, Department of Radiation Oncology, Zurich, Switzerland;
Show Affiliations
Hide Affiliations
Purpose or Objective
To reduce the risk of radiation necrosis during the treatment of brain metastasis (BM) with radiosurgery, steep dose gradients around the targets are desirable. Multiple factors may affect the dose gradient such as tumor size, number of BM, technique used and prescription percentage isodose line (PIL). In this study, we evaluated the impact of the prescription PIL on the dose distribution for radiosurgery of single and multiple BM for mono- and multi-isocentric treatment.
Material and Methods
Ten patients treated with radiosurgery for BM with target volumes ranging from 0.1cm3 to 5.2cm3 were re-optimized with an automated treatment planning system (Elements V2.0, BrainLab AG, Munich, Germany) using dynamic conformal arc (DCA). Plans were optimized for a TrueBeam linac (Varian Medical System, Palo Alto, CA) with high definition multi-leaf collimator. Eight to ten plans per BM were generated with the PIL prescription equally distributed between 50% and 90%. All plans were normalized such that 99.5% ± 0.5% of target volume received the prescribed dose. The plan quality was evaluated with respect to the conformity index (CI), gradient index (GI), brain volume receiving 90% of the prescribed dose (V90%), V75%, V50%, V25% as a function of the PIL prescription and BM volume.
Results
The target mean dose had a linear relationship with PIL (Dmean = -0.289Gy x PIL + 46.3 Gy, R2 = 0.97) and was independent of the target volume. The CI was lowest for PIL ranging from 86% to 75%, independent of the target volume size and the use of mono- or multi-isocentric treatment.
The brain V25% - V90% and GI was the lowest for PIL ranging from 50% to 83% for PTV volumes < 0.7cm3. Target volumes ranging from 0.7cm3 to 1.4cm3 showed the lowest brain dose for PIL ranging between 50%-60% to 83%. For target volumes > 1.4cm3, the PIL resulting in the lowest V25%-V90% and GI ranged between 83% and 60%, while a PIL < 60% lead to increased brain doses and GI.
The PIL resulting in the lowest V25%-V90% and GI for multiple BM vs. single BM was not different (p = 0.2).
Conclusion
By optimizing the PIL prescription based on BM volume, the dosimetric quality of BM radiosurgery can be improved. For small BM volumes, decreasing the PIL prescription will result in a dose escalation without increasing the brain dose. For larger BM, an optimal range of PIL exists which results in optimal achievable CI and GI. The optimal PIL prescription was similar irrespective of the number of BM.