ESTRO 2021 Abstract Book

S1364

ESTRO 2021

Conclusion Bead TLDs can be arranged so that they are detectable on standard radiotherapy CT protocols. Further work on the feasibility study will investigate the use of the bead TLDs for in vivo dosimetry. With the use of fiducial markers alongside higher density beads, an identical set of bead TLDs can be repositioned on the phantom for treatment delivery. The dose to each bead TLD can then be assigned to its position on the CT scan allowing for an accurate comparison between the planned and delivered doses. PO-1643 End-2-End testing of PBS proton therapy workflow with the PTW RUBY phantom J. Wulff 1 , J. Horn 2 , F. Kugel 1 , D. Poppinga 2 , B. Koska 1 , C. Bäumer 1 , B. Timmermann 1 1 University Hospital Essen, Westdeutsches Protonentherapiezentrum Essen, Essen, Germany; 2 Physikalisch- Technische Werkstätten, R&D, Freiburg, Germany Purpose or Objective The complexity of pencil-beam scanning proton therapy requires a thorough, yet efficient quality assurance program. The underlying single components and subsystems of the delivery system are tested periodically, while an end-to-end test allows demonstrating data integrity and proper function with all dependencies in the treatment workflow. The PTW RUBY modular phantom was applied for such an end-to-end test with pencil- beam scanning in IBA ProteusPlus system. Materials and Methods A computed tomography was acquired in a Philips BigBore for the RUBY phantom with different QA inserts. The “Linac QA” insert with bone-substitutes in different orientations was used to test the patient positioning workflow with the IBA AdaptInsight position verification system (3D/2D kV imaging) as well as with the VisionRT surface guided alignment. The phantom was initially set-up on the treatment couch by means of lasers with the defined rotational and translational shifts. The “System QA” insert with different tissue substitutes served for planning in the RayStation planning system (9B SP1). Treatment plans of different complexity were created, ranging from single-field box-shaped dose distributions to non-coplanar field configurations with robust IMPT optimization (3 mm/ 3.5%) for a spherical targets from 1-10 cm diameter. Dose was calculated with the Monte-Carlo algorithm in a 1 mm dose grid (0.3% stat. unc.). Dose measurements were based on a PTW PinPoint 3D chamber + Unidos Tango electrometer in the center of the phantom. Treatment plans were created to measure dose within the target, as well as in structures mimicking an organ- at risk (OAR). Results The manual position correction of a deliberately mis-alinged RUBY phantom (“Linac QA” insert + tilting base) by clinical staff and the automatic correction determined within AdaptInsight was <1 mm and <0.2°. The treatment plans were irradiated with the “System QA” insert after translation correction, which agreed <1 mm with known offsets. The determined position correction by the AlignRT system was within 1.5 mm. The general agreement between plan and measurement for the target dose was well within 2% and 4% for the OAR structure. Depending on the local gradient, the single field doses were within 2.5% within the target structure and 6% for the OAR region. The latter occurred due to close vicinity of the chamber locations to dose gradients. A small, but systematic offset in calculated range can be attributed to the fact that most of the phantom is made of polystyrol, which is not optimally reflected in the clinical Hounsfield-Unit calibration of the TPS used. Conclusion The RUBY phantom was used for a systematic end-2-end test at a proton therapy facility. The modular design allows to efficiently test the complete treatment workflow. The determined results for patient positioning and the dose calculation accuracy are within the expected limits and are reflected in the clinical protocols.

PO-1644 Dose calculation validation of AAA and Acuros XB algorithms in presence of titanium spinal