ESTRO 2020 Abstract Book

S874 ESTRO 2020

due to residual rotational error. The shifts larger than 5 mm was not noted in Y axis direction. 15 of 22 patients had no treatment fractions with the shifts larger than 5 mm in Z axis direction. Conclusion The presented method allows a simple assessment of the effects of rotational movements. The obtained translations of the point A approximate the shifts of CTV points resulting from roll and pitch rotations. For significant number of patients, the pitch rotational errors caused the shifts larger than 5 mm in superior-inferior and anterior-posterior direction. Setup variations caused by rotations cannot be ignored for prostate patients treated with whole-pelvic radiation therapy. The impact of rotational errors should be taken into consideration in calculations of the CTV-PTV margin. Replacing the point A coordinates with the coordinates of selected CTV points in the mentioned method allows to calculate the shifts of these points as well as the CTV-PTV margin. It is recommended that margin increases with distance from the isocenter in order to take rotational errors into account. PO-1606 Measurement of dose distribution of cardio- synchronous brain motion in microbeam radiation therapy M. Petasecca 1 , M. Duncan 1 , M. Donzelli 2 , P. Pellicioli 3 , E. Brauer-Krisch 4 , J. Davis 1 , A. Rosenfeld 1 , M.L.F. Lerch 1 1 University of Wollongong, Centre for Medical Radiation Physics, Wollongong, Australia ; 2 The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust, Joint Department of Physics, London, United Kingdom ; 3 Swansea University- Singleton Park, Medical School, Swansea, United Kingdom ; 4 European Synchrotron Radiation Facility, Biomedical beamline ID17, Grenoble, France Purpose or Objective Microbeam radiation therapy (MRT) is an emerging radiation oncology modality ideal for treating inoperable brain tumours. MRT employs quasi-parallel beams of low energy x-rays produced from modern synchrotrons. The peak to valley dose ratio (PVDR) is of critical importance to the efficacy of MRT. The underlying radiobiological advantage of MRT relies on high peak dose (50um FWHM) for tumour control and low valley dose for healthy tissue sparing (400um spacing between peaks). The brain is known to exhibit cardio synchronous pulsating motion of the order of 100-200 mm, which is comparable to the microbeam width and spacing. This can result in dose blurring through overlap of microbeams and a reduced PVDR which will compromise the radiobiological effectiveness. This work presents the first experimental measurement of the effect of brain motion on MRT dose distribution. We present microbeam profiles measured under simulated brain motion conditions using a silicon strip detector and Gafchromic film.These are compared to a dataset calculated using GEANT4 Monte Carlo toolkit. Material and Methods The Centre for Medical Radiation Physics has developed a real time dosimetry system for MRT based on a silicon strip detector (SSD). Measurements were performed at the ID17 beamline at the European Synchrotron Radiation Facility (Grenoble – France). The detector was mounted in a 15cm x 15cm x 15cm solid water phantom and the brain motion was realised by a stepper motor. Profiles at different scanning speed, with/out motion, were reconstructed by stepping the detector horizontally through the width of the radiation field while the dose is delivered by shifting the target vertically in front of the beam apertures.Gafchromic HD-V2 film was used under the same conditions. PDVR and FWHM was calculated for each of the profiles and for each detector dataset. Comparison to Montecarlo simulations were performed using the Geant4

toolkit.

Results When brain motion is applied there was a decrease in total dose delivered (to the specific point of measurement) making the delivered dose not the same as the planned dose. Fig.1a shows the distortion introduced in the relative dose map of the microbeams. Profiles reconstructed using the SSD detector (Fig.1b) show quantitatively the distortion in Gy. Low target scanning speed affect the dose distortion the most with a reduction of the PVDR up to 40% as shown in Fig.2a. This is confirmed by the simulated data as shown in Fig.2b.

Conclusion The SSD was able to reconstruct dose profiles under motion conditions and predict similar effects on FWHM and PVDR variations as per the simulations. Low scanning speed of the target leads to 40% decrease of the PVDR. Higher scan speeds mitigate the dose distortion effects with no effect on the displacement of the microbeams. This has implications for MRT treatment planning when multiple ports are used. The technique developed can be used to assess the effect of motion in MRT for different motion patterns associated to different organs. PO-1607 Development of a Robot Assisted Ultrasound Based Radiation Therapy (USgRT) P.K. Seitz 1 , A. Schwahofer 1 , B. Baumann 2 , R. Bendl 2 1 German Cancer Research Center, Department of Medical Physics in Radiation Oncology, Heidelberg, Germany ; 2 Heilbronn University, Medical Informatics, Heilbronn, Germany Purpose or Objective One of the central problems of radiotherapy is the position monitoring of tumors during irradiation. Therefore, imaging methods are searched in order to detect the tumor position during the irradiation. One approach to this is ultrasound, which requires a constant pressure on the patient surface. The aim of this work is to develop a robot application for respiratory and motion compensation with constant contact pressure. This is to be integrated into a medical image processing program for the registration of ultrasound images with planning data. Material and Methods The lightweight robot (Kuka lbr iwa 7 R800) was integrated into the image processing program Medical Interaction Toolkit (MITK) [1] via a newly developed interface. The detection of the transducer position was observed via an optical tracking system (Spectra, Polaris). Experimental setup is displayed in Fig. 1.