Validation of the Eclipse(TM) Treatment Planning System for Synchrotron Microbeam Radiotherapy

Sammanfattning: This work presents the first data from a calculation engine for synchrotron Microbeam Radiation Therapy (MRT) implemented in the EclipseTM Treatment Planning System from Varian Medical Services. The calculation engine constitutes a simple pencil beam algorithm estimating ‘peak’ dose and of Monte Carlo simulated Peak Valley Dose Ratios (PVDRs) that are used to obtain ‘valley’ dose. The aim of this work is to benchmark data calculated using Eclipse against measured and Monte Carlo simulated data in order to enable future clinical applications of the calculation engine. Percentage Depth Dose (PDD) curves and field profiles in a homogeneous water phantom as well as PDD curves and point dose values in inhomogeneous, tissue equivalent phantoms were calculated. These data were compared against data measured using a PTW 31014 Pin-Point ionization chamber and GafChromic HD-V2 film as well as against Monte Carlo data simulated using the GEANT4 toolkit. The data show that Eclipse performs very well in calculating ‘peak’ dose in a homogeneous water phantom and that it calculates ‘valley’ dose with some accuracy, although an underestimation of valley dose for smaller field sizes and an overestimation for larger field sizes is discovered. This is found to depend on the inability of Eclipse to account for the field size dependence of the peak to valley dose ratio (PVDR). Field profiles are also calculated with reasonable accuracy, although due to the simplistic nature of the algorithm used to calculate dose it does not completely calculate dose deposition from a synchrotron radiation X-ray beam. This simplification results in errors in the field profiles. When it comes to simulating dose in inhomogeneous mediums, Eclipse is struggling. The Eclipse calculated ‘peak’ dose distributions in phantoms containing lung equivalent tissue do not agree with Monte Carlo simulations or with what you would theoretically expect. Furthermore, for the ‘valley’ dose, an error in the algorithm causes the deposited dose to inexplicably drop to zero in some regions of the phantom for some phantom configurations. Although there are limitations in the calculation engine in Eclipse and although there is still work to be done, the data generated in this work are overall encouraging and indicate that with further improvements there are possibilities and potential for this calculation engine to be implemented in the clinic in the future.