ESTRO 36 Abstract Book

S412 ESTRO 36 2017 _______________________________________________________________________________________________

Magnetic Resonance Image (MRI) has the potential to increase the accuracy and effectiveness of proton therapy. Previous studies on that topic demonstrated that corrections in dose calculation algorithms are strictly required to account for the dosimetric effects induced by external magnetic fields. So far, a real dose calculation possibility including a trajectory corrected approach was missing. In this study, we developed a pencil beam algorithm (PBA) for dose calculation of a proton beam in magnetic fields. Material and Methods MC simulations using the GATE 7.1 toolkit were performed to generate first benchmarking data and subsequent validation data for the PBA. The PBA was based on the theory of fluence weighted elemental kernels. A novel and non-symmetric exponential tailed Gauss fitting function was used to describe the lateral energy deposition profiles in water. Nuclear corrections, multiple scattering and charged particle drifting were accounted by means of a look-up table (LUT) approach. Longitudinal dose depositions were estimated from the LUT and corrected using a water-equivalent depth scaling. In a first step proton beams in the clinical required energy range 60 – 250 MeV with transverse external magnetic fields ranging from 0 – 3T were analyzed in a 40x40x40 cm 3 water phantom. Next validation simulations were performed for different phantom configurations, e.g. using a simple water box or slab-like geometries with inhomogeneities of different materials and volumes. Percentage depth dose curves (PDD) and two-dimensional dose distributions were calculated to assess the performance of the PBA. Results For PDD in water discrepancies between the PBA and MC of less than 1.5% were observed for all the depth values before the Bragg-Peak (see Figure 1). An increasing value of up to 6% was found in the distal energy falloff region, where dose values represents around 1% of the maximum dose deposition. In all cases, maximum range deviations of the results were less than 0.2 mm. Deviations between two dimensional dose maps obtained with PBA and GATE remained below 1% for almost all the proton beam trajectory, reaching a maximum value up to 4% in the Bragg-Peak region, see Fig. 2. As expected, agreement became worse for high energy protons and high intensity magnetic fields.

Fig. 2 Relative dose difference map for a 240 MeV proton beam in water exposed to a 3T transverse field. Conclusion The proposed pencil beam algorithm for protons can accurately account for dose distortion effects induced by external magnetic fields. Corrections of dose distributions using an analytical model allows to reduce dose calculation times considerably, making the presented PBA a suitable candidate for integration in a treatment planning system. The current work demonstrates that proton MRI is feasible from a dosimetric point of view. PO-0786 Energy dependence investigation for detectors used in out-of-filed dosimetry L. Shields 1 , L. Leon-Vintro 2 , B. Mc Clean 3 1 St Luke's Hospital, Medical Physics, Dublin, Ireland 2 University College Dublin, Schoool of Physics, Dublin, Ireland 3 St. Luke's Radiation Oncology Network, Medical Physsics, Dublin, Ireland Purpose or Objective Traditionally, energy dependence of a range of detectors used in radiotherapy has been investigated mainly in the Cobalt-60 and 6-15MV photon range. However, when considering detectors for use in out-of-field dosimetry, it is more important that the energy dependence is investigated over a much lower range. This study examined (i) the mean incident energy of radiation out- of-field for a 6MV photon beam and (ii) the energy dependence of a range of clinically available detectors to the typical energies experienced out-of-field and (iii) Monte Carlo (MC) calculated and detector measured out- An Elekta Synergy Linac operating at 6MV and a water phantom at 90cm SSD was defined in BEAMnrc. Phase spaces were scored at 6 different planes in the water phantom - 0.2, 1.4 (dmax), 5, 10, 15 and 20cm. Each phase space file was analysed using the EGSnrc program package BEAMDP to extract energy spectra from each of the phase space files to examine the change in energy spectra with increasing distance from the field edge and depth in the phantom. The energy dependence of each of the detectors was examined using 70, 100, 125 and 200 kV beams on a Gulmay D3225 Orthovoltage Unit and a 6MV Elekta Synergy beam. The kV energies lied within the range of energies which were found to be dominant out-of-field in a 6MV beam. A dose of 1 Gy was delivered to each detector as determined by their respective calibration protocols, and the signal was recorded for all energies. In-plane and cross-plane profiles were measured by each detector and compared to MC calculated. of-field dose profiles. Material and Methods

Fig. 1. PDD curves comparing the PB algorithm with MC simulations for proton beams in water. Relative discrepancies are shown in the top region of the graph.