PO-0774: PET-guided simultaneous integrated boost of lung tumors: Alanine/EPR dosimetry in an anthropomorphic phantom

J.G. Svestad, I.S. Knudtsen, E.P.S. Sande, B.L. Rekstad, J. Rødal, W. van Elmpt, M. Öllers, E.O. Hole, E. Malinen
2013 Radiotherapy and Oncology  
Conclusions: TG119 structures and plans were found to be easily adaptable to Delta4 phantom CT enabling a rigorous evaluation procedure for IMRT plans verification. Preliminary results on VMAT plans showed that TG119 could be a practical commissioning procedure for modulated arc therapy too. Delta4 allows a fast on-line 3D dose distributions analysis substituting more traditional dosimeters suggested in TG119, but this solution is more expensive than the one proposed inside this report and it
more » ... his report and it is not available in every department for intercomparison purposes. Purpose/Objective: Quality control (QC) of volumetric modulated arc therapy (VMAT) delivery represents a significant additional workload during monthly accelerator checks. This study therefore aims to implement VMAT QC in as few arcs as possible using portal imaging. Synchronisation between gantry position, multileaf collimator (MLC) leaves and dose rate is the focus of the study, since this is the aspect which is covered least by already-implemented tests of intensity modulated radiation therapy (IMRT) delivery. Materials and Methods: An Agility accelerator with iViewGT portal imager (Elekta AB, Stockholm, Sweden) was used for this study. A steel bar of diameter 12 mm was accurately positioned in the G-T direction, 80 mm laterally from the isocentre (Figure 1a ). An arc prescription was designed which irradiated the bar with a 16 mm x 220 mm field during a complete 360° arc, so as to cast a shadow of the bar onto the portal imager. This resulted in a sinusoidal sweep of the field and shadow across the portal imager and back. The prescribed monitor units (MU) were chosen so as to give a uniform intensity during this sweep. A total of 640 MU were delivered with control point spacing 10°. The method was tested by simulating an MLC leaf position error of 2 mm at one control point, a gantry position error of 9° at one control point, and a dose error of 20 MU (3% of overall dose) at one control point. The portal images were viewed as integrated images. Coefficient of variation of mid-leaf profiles was used to evaluate the magnitude of features appearing in the portal images due to simulated errors. Results: The test requires a counter-clockwise arc and a clockwise arc to fully evaluate the VMAT performance of all MLC leaves. In the absence of simulated errors, the integrated images show uniformity. With simulated delivery errors, irregular patterns appear in the integrated portal images (Figures 1b -1d) . The increase in coefficient of variation relative to no delivery error is 42% due to a 2 mm MLC leaf position error at one control point, 108% due to a 9° gantry position error at one control point and 239% due to a 20 MU dose error at one control point. The method is more sensitive to errors at gantry angle 90°/270° than at 0°/180° due to the geometry of the test. Conclusions: This method provides fast and effective VMAT QC suitable for inclusion in a monthly accelerator QC programme. The test is able to detect errors in the delivery of individual control points, with the possibility of using movie images to further investigate suspicious image features. We are grateful for engineering assistance and to Elekta AB for their collaboration on VMAT and Agility. PO-0774 PET-guided simultaneous integrated boost of lung tumors: Alanine/EPR dosimetry in an anthropomorphic phantom
doi:10.1016/s0167-8140(15)33080-2 fatcat:qvdejdj2inhw3ncxbpk6rdje3a