ESTRO 37 Abstract book

S1194

ESTRO 37

EP-2157 A simple way to improve quality of treatment for locally advanced lung cancer M. Yaneva 1 , I. Neycheva 1 1 Complex Oncological Center, Radiotherapy, Plovdiv, Bulgaria Purpose or Objective The main goal was to evaluate two different VMAT approaches – pure VMAT and hybrid VMAT, for their efficacy of quality of treatment for locally advanced lung cancer. Material and Methods The study was based on comparative analysis for 15 clinical cases previously treated for locally advanced lung cancer with or without hilar/mediastinal limph nodes. The mean PTV volume was 323.7cc±289.8cc. For each case treatment planning was performed in two ways: using double arc-VMAT and using hybrid VMAT by adding 3D-CRT component. All plans were clinically acceptable where 95% of prescribed dose covered 95% or more of PTV. All of them were calculated by one physicist to avoid inter-observer variations. Hybrid plans were made as simple as possible: monoisocentric, coplanar, distributing delivered dose between two static fields (6/18 MV) and two complete arcs. The ratio between static and VMAT-contribution was 65% and 35% of prescribed dose, respectively. Conformity index (CI) and homogeneity index (HI) were calculated for each case and each aproach to assess quality of treatment together with clinicaly relevant doses for lung and other OARs. Homogeneity index was defined as ratio between difference of near Dmax (D2%) and near Dmin (D98%) to Dmedian (D50%). Also, we analized the difference between absolute Dmax and near Dmax (D2%) called “dose tail”. Conformity index was defined as ratio between absolute volume TV received 95% of prescribed dose to PTV received 95% of prescribed dose. MDL and V20 were estimated for healthy lung (whole lung minus PTV). Paired sample t-test and describtive analizes were used for statistical purposes. Results The basic finding was better PTV coverage for Hybrid RapidArc with better homogeneity index (p<0.001). The mean HI for pure VMAT was 0.12±0.02 vs. 0.08±0.02 for hybrid VMAT. In hybrid plans we found lower “dose tail” with mean value of 1.36±0.38 Gy. The differences for CI and MDL were not statistically significant. The mean CI was 1.23±0.20 and 1.29±0.23 for VMAT and hybrid approach, respectively. The doses were lower in a hybrid approach for V20 (p = 0.01) and CL-5 (p = 0.002). In clinical manner the mean difference of 11.55%, (SD: 11.57%) was remarkable result for sparing contralateral lung and demonstrates benefit of hybrid technique together with better dose homogeneity of PTV. The time needed to prepare a hybrid plan was at least three times shorter and in most cases could be realized without additional structures for PTV density heterogeneity. Because of small VMAT contribution, (only 35%), inaccuracies in optimizer, which does not model lateral electron transport, had very low influence for final dose calculation and could be neglected. Conclusion In our opinion the hybrid RapidArc techique could be a simple way to improve quality of treatment in localy advanced lung cases. EP-2158 Feasibility of Mobius 3D as an independent MU checker for the adaptive work flow on the MR-Linac A. Pollitt 1 , G. Budgell 1 , A. Pooler 1 , J. Wood 1 , R. Chuter 1 , A. McWilliam 1 1 Christie Hospital NHS, Radiotherapy, Manchester, United Kingdom

Purpose or Objective The Elekta Unity MR Linac will provide on-treatment MR imaging allowing for daily online adaptive radiotherapy. This may range from a small adaptation of the reference plan for geometric shifts or to a full re-optimisation. Physical verification is impossible with the patient on the bed, therefore software based verification becomes critically important. However, the presence of the magnetic field causes perturbations of the dose to superficial tissues and at interfaces which cannot be modelled fully without Monte Carlo calculations. This study aims to establish the magnitude of the difference between the Monte Carlo dose calculation performed by the Monaco Treatment Planning System (TPS) and a collapsed cone calculation (without magnetic field) performed by Mobius 3D. Material and Methods Twenty-nine patients from four tumour sites (Head and Neck (7), Sinus (10), Sarcoma (3) and Lung (9)) were planned using Monaco (v5.19.07, Elekta AB Stockholm, Sweden) using an Agility beam model and step and shoot IMRT to match the available beam model in Mobius 3D. All of the plans met the departmental constraints for Target and OAR doses. Plans were optimised with the 1.5 T magnetic field present and recalculated without it. The DICOM for both plans was exported to Mobius 3D which performed an independent dose calculation. A comparison was performed of the 3%/3mm gamma- analysis pass rate and the doses for the overall plan and to the reportable targets and OARs. Results The average change to the gamma-analysis pass rate, between with and without magnetic field, was -6.0 ± 2.2% (1SD). Table 1 shows the results for the gamma- analysis pass rate and dose change for the target volumes and OARs for each site. The mean change in dose for all of the target and OAR volumes was less than 1% between the cases with and without the 1.5T magnetic field. This is shown for the PTV dose in Figure 1. The maximum dose change for target structures was less than 1% and for OAR structures was less than 3%, within the typical tolerance for an MU check.