ESTRO 2021 Abstract Book

S100

ESTRO 2021

radiation therapy.

PH-0164 Abdominal compression; development of a non-gated pancreas MRIgRT workflow S. Alexander 1 , R. Lawes 2 , G. Adair Smith 2 , H. Barnes 2 , I. Hanson 3 , T. Herbert 2 , R. Huddart 4 , C. Lacey 2 , H. McNair 5 , A. Mitchell 3 , S. Nill 3 , C. Ockwell 2 , U. Oelfke 3 , H. Taylor 5 , A. Wetscherek 6 , K. Aitken 7 , A. Hunt 2 1 The Royal Marsden NHS Foundation Trust, Radiotherapy, Sutton, United Kingdom; 2 Royal Marsden NHS Foundation Trust, Radiotherapy, Sutton, United Kingdom; 3 Royal Marsden NHS Foundation Trust and Institute of Cancer Research, Joint Department of Physics, Sutton, United Kingdom; 4 Royal Marsden NHS Foundation Trust and Institute of Cancer Research, Urology Unit, Sutton, United Kingdom; 5 Royal Marsden NHS Foundation Trust and Institute of Cancer Research, Radiotherapy, Sutton, United Kingdom; 6 Royal Marsden NHS Foundation Trust and Institute of Cancer Research, The Joint Department of Physics, Sutton, United Kingdom; 7 Royal Marsden NHS Foundation Trust and Institute of Cancer Research, Gastrointestinal Unit, Sutton, United Kingdom Purpose or Objective For patients with locally advanced pancreatic cancer dose escalated radiotherapy suggests a survival advantage 1 . Safe dose escalation is complicated by abdominal motion and the pancreas’ proximity to dose sensitive organs at risk. Gated MRIgRT has provided a solution for this 2 , but motion management strategies are not currently supported on the Unity MR-linac (MRL) (Elekta, Stockholm, Sweden). The purpose of this study was to examine the use of abdominal compression (AC) as a solution to enable non-gated MRIgRT. Table 1: AC criteria for use

Materials and Methods Two AC devices were identified through discussion with UK radiotherapy equipment distributers and the workflow development multi-disciplinary team. Five non-patient volunteers were recruited to the PRIMER trial (NCT02973828) with the aim of determining optimal MRL immobilisation. Volunteers had 4 scanning sessions. During each session, MRI sequences (T1w, DIXON, 3D Vane and coronal cine) with and without AC were acquired; 1/4 sessions used an in-house adapted large elastic AC belt and 3/4 sessions used a commercially available system (ZiFix TM , Qfix, USA). Diaphragmatic motion in the superior/inferior direction was assessed as a surrogate for respiratory associated abdominal motion and measured on coronal cine images, with and without AC. Normal tissue visibility (pancreas, liver, duodenum, stomach and bowel) on T1w, Dixon and 3D Vane was graded using a 4-part Likert scale. Three MRL radiographers independently registered 10 image pairs, 5 with and 5 without AC, from which inter-observer registration variability was measured. Participant experience questionnaires were used to assess session tolerability. In an additional session Respiratory Gating for Scanners (RGSC)(Varian, Palo Alto, USA), an infrared motion tracking system, was used to capture volunteer respiratory motion with and without AC. To detect an accurate respiratory trace and enable 4DCT acquisition abdominal motion (rise and fall) has to be ≥ 4 mm. Results Both AC devices were MRI safe and fitted within the magnet bore. Mean free breathing diaphragmatic motion was 11.64 mm, both AC devices reduced this motion below 10 mm with ZiFix TM having the greatest effect, reducing mean diaphragmatic motion to 6.76 mm. Image quality and registration accuracy were not adversely impacted by AC. No scanning sessions were terminated due to volunteer discomfort. 4/5 volunteers had a RGSC motion tracking session, for 2 volunteers the ZiFix TM AC device reduced abdominal motion so much that

an accurate trace could not be detected. Table 2: Results related to AC criteria