ESTRO 2020 Abstract book

S962 ESTRO 2020

Stockholm, Sweden) tabletop. Patient immobilization was carried out with individual thermoplastic masks. Translational (x,y,z) and rotational(yaw, pitch, roll) corrections were identified using on-board kV imaging. Corrections, both translational and rotational, were carried out before treating each fraction. Offsets from registration processes can be analyzed in terms of systematic and random components. Overall population mean set-up error, population systematic error and population random error components were derived from the acquired data. Results It has been shown that rotational errors of 2° have clinical relevance at patients with non-spherical, elongated treatment volumes. Rotational deviation with such magnitude induces about 1 mm translational deviation in a point 3 cm from correction point. Translational overall population mean error values were: -0.03 cm, -0.04 cm and -0.04 cm in Superior-Inferior (SI), Left-Right (LR) and Anterior-Posterior (AP) directions respectively. Rotational overall population mean error values were: 0.0°, 0.07°and 0.18° in yaw, pitch and roll directions respectively. Overall population mean setup errors tending to zero indicates, there are no underlying common errors in this group of patients. Systematic error component values were: 0.14 cm, 0.15 cm, 0.14 cm in x, y, z and 0.77°, 0.82° and 0.74° in yaw, pitch, roll directions. Random error component values were: 0.2 cm, 0.2 cm, 0.17 cm and 0.84°, 0.65° and 0.73°. Conclusion Translational systematic and random error components are often incorporated into the treatment margins, but rotational displacements of concave IMRT target volumes can be a source of error in treatment delivery as well and can have dosimetric impact on the disease site. The HexaPod™ 6DoF treatment couch enables us to correct rotational errors occurring during patient setup. PO-1660 Feasibility of PTV margin reduction with online adaptive MR-guided radiotherapy for rectal cancer. H. Eijkelenkamp 1 , L. Van Zijp 1 , R. Tijssen 1 , M. Intven 1 1 UMC Utrecht, Radiotherapy, Utrecht, The Netherlands Purpose or Objective The shape and volume of the clinical target volume (CTV) for neoadjuvant radiotherapy in rectal cancer is influenced by bladder and rectum dynamics. Large planning target volume (PTV) margins are needed to overcome these daily variations. This may cause radiation-induced toxicity. By using online recontouring and replanning at each fraction, MR-guided radiotherapy has the potential to significantly reduce the required PTV margins, and herewith reduce toxicity. The aim of this study was to assess the feasibility of reduced PTV margins using a workflow with daily online recontouring and replanning on a 1.5T MR-linac for patients with rectal cancer. Material and Methods Six consecutive patients (3 female, 3 male) with intermediate risk rectal cancer were treated with 5 fractions of 5Gy on the 1.5T MR-linac (Unity, Elekta AB, Stockholm Sweden). Both on the pre-treatment 3D T2w MRI and the post-treatment 3D T2w MRI mesorectal and elective lymph node region CTVs were delineated according to the Dutch national consensus delineation protocol for rectal cancer. The PTV expansion needed to compensate for intrafraction motion was analyzed by creating PTV volumes by applying systematic volume expansions on the CTVs on the pre-treatment MRI. The CTV was expanded anisotropically to create different PTVs according to the margin recipe in table 1. The largest margin (PTV A) was the PTV margin used clinically, the medium (PTV B) and small (PTV C) margins were created for this margin reduction analysis. Mesorectal margins were slightly different compared to elective lymph node

another according to the tracking method of the Cyberknife. The dose of the three plans and the target volumes (GTV and PTV) were automatically rigidly transferred to the fraction CT. Organs at Risk (OAR) contours from the initial planning CT were transferred to the fraction CT scan by deformable image registration (DIR). For each plan, the relevant DVH parameters such as PTV coverage and dose to bowel (V35) were presented in a pass/fail table. On the basis of a protocol decision tree , the RTT selected the best plan of the day and the patient was treated with the chosen plan. The RTT’s received training in order to perform the whole work flow and to make plan choice. Fig 1:The CyberKnife with in-room CT scan

Results Due to anatomical changes visible on the fraction CT scan, 2 patients needed re-planning: one due to severe constraint violation for both duodenum and stomach, abandoning the plan-of-the-day (POTD) strategy. The other due to inter-fraction movement of the GTV outside the PTV. Plan A was most frequently selected from the library in 68% of the fractions, followed by plan C in 26%. Plan B was chosen in 6%. Plan A was selected in only 5 fractions out of 58 due to a bad correspondence between the contours generated by DIR on the fraction CT and the visible anatomy (fallback option). The time between fraction CT scan and treatment was within 15 minutes for with automated organ at risk contouring and a plan-of-the-day strategy to improve OARs sparing in SBRT of lymphadenopathies located in the abdominopelvic region is feasible. Treatment plan selection can be handed over to RTT’s after training. PO-1659 Importance of rotational setup errors in H&N patients treated with IMRT M. Simon 1,2 , E. Csiki 1 , J. Papp 1,2 , Á. Kovács 1,2 1 University of Debrecen, Clinic of Oncoradiology, Debrecen, Hungary ; 2 University of Pécs, Faculty of Health Sciences- Doctoral School of Health Sciences, Pécs, Hungary Purpose or Objective In order to mitigate the toxicity of Intensity Modulated Radiotherapy (IMRT) the reduction of treatment planning margins and the use Image Guided Radiotherapy (IGRT) is necessary. In most IGRT systems only the 3 translational components are corrected, the rotational components often ignored. The aim of this study was to determine the rotational set-up error components of our head and neck Between April 2016 and October 2018 a total of 194 patients with head and neck tumors received some form of IMRT treatment to 360 regions with six degree of freedom (6DoF) positional correction using HexaPod™ (Elekta, all patients. Conclusion Online-adaptive stereotactic radiotherapy (H&N) patient cohort. Material and Methods