ESTRO 37 Abstract book

the SV tips and the SVs’ center relative to the prostate gland. Conclusion The ADT patients had significantly shorter SVs compared to the non-ADT patients, but no difference in SVmotion was observed. SV interfractional motion should thus be compensated using the same planning margins,regardless of whether ADT is used. EP-2385 Initial experience of ExacTrac X-Ray imaging for prostate patients with implanted fiducial markers. K. Crowther 1 , A. O'Neill 1,2 , C. Agnew 3 , D. McKay 1 , J. Smith 1 , A. McCrum 1 , P. Shiels 1 , S. Jain 1,2 1 Northern Ireland Cancer Centre, Radiotherapy Department, Belfast, United Kingdom 2 Centre for Cancer Research and Cell Biology, Queen's University, Belfast, United Kingdom 3 Northern Ireland Cancer Centre, Medical Physics Department, Belfast, United Kingdom Purpose or Objective To report the initial experience of Novalis ExacTrac Imaging system to verify the prostate position prior to treatment delivery using fiducial markers (FM). Material and Methods Five patients with localised prostate cancer were selected to participate in this pilot. All underwent trans- perineal placement of three, 3mm gold FM into the prostate using local anaesthesia and transrectal ultrasound. Patients attended for treatment between April 2016 - May 2017. V-MAT techniques were used with total doses from 60Gy-74Gy. Fractionation was 20-37. Patients were planned and treated with a ‘comfortably full’ bladder and empty rectum; achieved by self-administering daily micro-enemas and adhering to a bladder filling protocol. All patients had daily ExacTrac imaging, which acquires two 2D kV oblique radiographs. A point based marker match was performed. Setup corrections for translational dimensions were transferred to ARIA and all shifts applied. For treatment 1-3 and weekly thereafter post correction, pre-treatment CBCT was acquired and matched prior to treatment. In accordance with our institution’s protocol if residual CBCT set-up error was ≤3mm the correction was not applied, if >3mm all shifts were applied. Translational couch shifts and residual errors were recorded. Random (σ) and systematic (∑) set-up errors were calculated. Results 114 kV image pairs and 35 CBCTs were analysed. ∑ and σ set-up errors are presented below. ExacTrac set-up error (mm) Residual CBCT set-up error (mm) Lat Long Vert Lat Long Vert Po p ∑ err or 1.5 2.1 1.7 0.6 1.2 0.6 Po p σ err or 1.6 1.5 2.5 1.4 1.9 1.4 Overall the residual setup error following ExacTrac imaging, as determined by CBCT, were small, which demonstrates high accuracy of kV localisation when FM are present. Although for 6 fractions (4/5 patients) the CBCT residual error was >3mm requiring further correction. The advantages of this method of IGRT were evident when one patient presented with a history of Ankylosing

spondylitis and bi-lateral hip replacements. 2D kV image quality was not degraded by artefacts from hip joints or FM, enabling localisation of the prostate within PTV while CBCT image quality was degraded by artefact making visualisation of the prostate difficult. Conclusion Radiographers gained confidence in ExacTrac system reporting clear visualisation of FM. Limitations were: PTV coverage and preparation compliance could not be determined when CBCT was not acquired; cost and risk associated with FM placement. Patients with bi-lateral hip replacements receiving prostate radiotherapy will be offered FM, combined with daily ExacTrac imaging. However a larger pilot would be required to further assess the utilization for all prostate patients. EP-2386 Evaluation of image-guidance strategies of cervical and endometrial cancer A. Ryczkowski 1 , A. Jodda 1 , T. Piotrowski 1,2 1 Greater Poland Cancer Centre, Medical Physics Department, Poznan, Poland 2 Poznan University of Medical Sciences, Faculty of Health Sciences, Poznan, Poland Purpose or Objective In this study, set-up accuracy and obtained margins splitted to two regions (PTV1 - upper vagina and parametrial/paravaginal tissues; PTV2 - common, external and internal iliac lymph nodes) resulted from different image-guidance protocols used for gynecological patients were compared. Material and Methods Set-up corrections from 20 cervical and endometrial cancer patients treated on helical tomotherapy were used to simulate four types of image guidance protocols which were based on: a limited number of imaging sessions (A), reduced registration tasks during daily imaging (B), or mixed methods of imaging (C and D). Each protocol was evaluated for two referencing scenarios based on the first three fractions and first five fractions. Residual set-up error, the difference between the average set-up correction and the actual correction required, was used to evaluate the accuracy of each protocol and estimated on their basis margins.

Results The first five fractions referencing scenario provides the highest reduction of the margins for each image-guidance protocol evaluated in this study. The first type of protocol is the shortest way to the effective correction of the systematic component of set-up error. For the second type of the protocol, the control of the residual errors is better and, as a result, the reduction of the PTV1 is more significant than that obtained for the first