ESTRO 2020 Abstract Book

S1064 ESTRO 2020

corrections (rotation was not allowed). After the treatment, the PT-CBCT was acquired and registered to the planning CT. The tumour intrafraction movement in the three principal axes (left-right, LR; cranio-caudal, CC and antero- posterior, AP) was calculated as the difference between PT-CBCT and BT-CBCT registrations. Group Mean (GM), systematic (∑) and random (σ) errors were obtained from these data. To study the intrafraction patient movement, we registered again both CBCTs to the planning CT using the closest vertebrae to the tumour as the matching structure instead of tumor and made the same analysis. Results A total of 384 BT-CBCT and PT-CBCT were analyzed. The tumour intrafraction analysis resulted in a group mean error of LR = -0.2 mm (range -7 to 3 mm); CC = -0.9 mm (range -8 to 4 mm) and AP = -0.5 mm (range -5 to 7 mm). The systematic and random setup errors (σ and ∑) are displayed in Table 1. Intrafraction tumor motion exceeded 3 and 5 mm margins in 5% and 2% of the fractions, respectively, remaining the lateral axis the one with the smallest errors. In the retrospective evaluation carried out with vertebrae matching, the mean systematic error was in LR = 0.0 mm (range -4 to 5 mm); CC = 0.3 mm (range -11 to 14 mm) and AP = 0.5 mm (range -9 to 12 mm).

Tomography (CBCT) imaging for the first three fractions and then every five fractions (group A). 38 patients were positioned using Catalyst HD-C-RAD® in addition to tattoos and verified with CBCT according to aforementioned schedule. (group B).A total of 272 and 268 shifts along the three axes for group A and B, were respectively analyzed for the assessment of patient setup accuracy. The shift distribution along the three axes was compared to the two groups. Results An improved setup accuracy was observed for group B compared to group A. The standard deviation of the shift distribution decreases of 41%, 33% and 50% for the lateral, vertical and longitudinal direction, respectively, when SGRT positioning procedure is used. The histograms of the 3D displacement vector for group A and group B are shown in figure 1.

Conclusion SGRT allows an improvement of setup accuracy for the treatment of deep targets in the thoracic region.

PO-1908 Comparing soft tissue and bone to verify radiotherapy treatment position for lung cancer patients A. Sweeney 1 , D. Finn 1 1 Edinburgh Cancer Centre, Western General Hospitel, Edinburgh, United Kingdom

Conclusion We have evaluated the tumour and patient intrafraction movements along a SBRT treatment. Both are not negligible and depend on many factors. Using the tumour for correcting the patient setup is a safe way to ensure that it is inside the treatment area. Tumour movement resulted to be smaller than patient movement and could be correlated. The implementation of a CBCT after fraction increases the total treatment time, but we do consider it necessary for a qualified treatment. PO-1907 The effectiveness of SGRT for patient set-up in thoracic deep target radiotherapy S. Leva 1 , G. Libonati 1 , M. Tettamanti 1 , M. Casiraghi 2 , S. Presilla 2 , M.C. Valli 1 , A. Richetti 1 1 Ente Ospedaliero Cantonale, Radiotherapy, Bellinzona, Switzerland ; 2 Ente Ospedaliero Cantonale, Medical Physics Division- Istituto Imaging della Svizzera Italiana, Bellinzona, Switzerland Purpose or Objective Surface Guided Radiotherapy (SGRT) has been proven to improve set-up accuracy for breast radiotherapy. However, few data on the use of SGRT in the treatment of deep tumors are available up to now. The aim of this study is to analyze the benefit of SGRT with Catalyst HD® C-RAD for the positioning of patients treated for thoracic deep tumors. Material and Methods 80 patients treated with VMAT-RapidArc® for mediastinal and esophageal cancer were retrospectively evaluated. 42 patients were positioned using three tattoos and subsequently verified using Cone Beam Computed

Purpose or Objective Introduction

As the complexity of radiotherapy treatments increase, more on treatment imaging is being used in routine clinical practice. Studies vary in which anatomical landmark to match to in lung cancer treatment, to ensure treatment accuracy. Prior to transitioning from bone matching using two-dimensional kilovoltage (kV) imaging to 3D soft tissue matching using Cone Beam CTs (CBCTs), we retrospectively assessed matching to bone, carina and the tumour to determine the optimum landmark for image- matching purposes. Material and Methods Five radiographers conducted automatic and manual matches to bone, carina and tumour in 88 CBCTs of 20 patients using CBCT. For each of the 2600 matches, couch shifts were recorded in the anterior/posterior, left/right and superior/inferior directions. Tumour coverage was graded using target volume margins. The level of agreement between automatic and manual matches and the percentage of set-up errors out of tolerance (5mm) were calculated. CBCT feasibility was assessed by examining inter-observer reliability, reporting difficult matches and comparing timings of CBCTs with kV images. Results There was a significant improvement in target coverage when matching to tumour, instead of bone or carina (P<0.001). However, Bland-Altman analysis demonstrated tumour matching had the lowest levels of automatic and manual agreement. Tumour matching detected the highest proportion of set-up errors (26.1%), then carina (19.05%) and bone (18.41%). All methods demonstrated good or excellent inter-observer reliability (intraclass correlation