Intensity-modulated arc therapy using the gate Monte Carlo simulation platform in a grid environment

Abstract

Introduction The GATE Monte-Carlo (MC) simulation platform ( www.opengatecollaboration.org ) based on the GEANT4 toolkit has come into widespread use for simulating Positron Emission Tomography, Single Photon Emission Computed Tomography, Computing Tomography (CT) imaging devices and radiation therapy. In this work, we explore the simulation of an intensity-modulated arc therapy (IMAT) treatment using GATE 6.2. To reach this objective, we first performed a complete modeling of the Varian's Novalis Tx linear accelerator (15MV photon mode) using the multileaf collimator (MLC). Then, the simulation of the geometry was validated in water before experiencing a head-and-neck cancer treatment. Material and methods The first stage is to model the Novalis Tx head with the GATE MC platform using manufacturer's data. We simulated this linac in 15 MV mode and compared depth doses, in-plane and cross-plane profiles with reference measurements (performed with ionization chambers (Semiflex 31013, PTW and LA48 Linear Array, PTW) and PTW60012 Diode E) in water for several field sizes ranging from 1 × 1 to 10 × 10 cm 2 . Target and flattening filter geometries and initial electron beamare adjusted in order to fit simulation calculations with dosemeasurements. In a second stage, the MLC geometry implementation enabled the simulation of clinical IMAT treatment plans compared with the iPlan treatment planning software (BrainLab) calculations. In order to reduce the computing time of GATE calculations, the simulations are executed on a distributed infrastructure using the GateLab ( http://vip.creatis.insa-lyon.fr/ ) platform. Results Weshowa good agreement between measurements and simulated depth dose profiles in liquid water. GATE and iPlan normalized treatment plans were compared using agammaindex of ±2%/mm.The ongoing work concerning treatment plan comparisons demonstrates limits of the iPlan Pencil Beam algorithm close to heterogeneities. Using the distributed infrastructure, space phase file was produced in 14 h for 500 million primary photons generated reaching a gain of a factor 50 comparing to calculations performed on a single CPU. Conclusions This study demonstrates that GATE offers efficient tools to make possible IMAT applications. In a subsequent work a microdosimetry study should be performed for a complete validation.