Abstract Book

S1243

ESTRO 37

PLN’s. Pre-treatment imaging seems to have some predictive value for excessive displacement during treatment. EP-2374 Importance of immobilization during lung SBRT treatments E. Heras Vila 1 , C. Montiel Pacheco 1 , A. Lorente Pau 1 , T. Borda Zambrana 1 , D. Garcia Relancio 1 , A. Giraldo Marin 1 1 Hospital Universitari Vall d´Hebron, oncology- radiotherapy, Barcelona, Spain Purpose or Objective The IGRT protocol used in our center for SBRT lung, includes the evaluation of the intrafraction movement through the realization of a CBCT after each fraction.We wanted to analyze the data obtained from each CBCT, to assess whether our type of immobilization was enough or not. Material and Methods We evaluated retrospectively the patients treated in our centre with SBRT for early stage NSCLC, between February 2015 - August 2017. All patients were immobilized with the Stradivarius SBRT device. We used a custom vacuum mattress for the positioning and accessories system (equipped with abdominal comp- ression), but always looking for the the well-being of the patient. We stabilize the position with a body vacuum mattress having the arms over head, taking into account the patient's elbows opening, to avoid collisions with the machine and guarantee the comfort of the patient. To prevent movement and rotation, legs and feet are semi- covered by the cushion. The abdominal compressor will be located under the sternum, limiting the patient's breathing without becoming painful. We recognize the importance during the simulation, to get comfort patient but at the same time a position with enough effecti- veness and reproducibility. More details in figure 1.

(mm)

AP LR CC

Mean systematic error

-0.6 -0.2 -1

SD of Systematic setup error 1

0.6 0.8

SD of Random setup error

1.2 1.2 1.2

Conclusion After implemented IGRT protocol, some actions of improvement have been carried out: CBCT daily (pre and post) to correct the variability of the position inter and intrafraction; mattress height in the pelvis area is asymmetrical to prevent deformation of the mattress during positioning (always on the opposite side of the lesion). According to the results the lateral axis have the best control trough the treatment and we need to focus a little more in vertical and longitudinal axis. Wherever is important to know that currently we are working with a planning tumor volume margin of 5 mm, therefore the majority of our patients were correctly covered during the treatment. Consequently our immobilization is adequate. Through our results we can conclude that an effective procedure to simulation should reduce the probability of errors during positioning and treatment, making less important the respiratory movement, and increasing the comfort and stability of the patient. EP-2375 Time dependence of intrafraction motion in SRS with mask fixation for linac and Gamma Knife ICON J. Stam 1 , A. Licup 1 , P. Remeijer 1 , A. Van Mourik 1 1 Netherlands Cancer Institute, Department of Radiation Oncology, Amsterdam, The Netherlands Purpose or Objective Patients treated with stereotactic radiosurgery (SRS) on the Gamma Knife ICON can be immobilized using a thermoplastic mask, similar to linac based SRS. However, the treatment time on the Gamma Knife is typically expected to be longer than on the linac, which might lead to an increase in intrafraction motion (IFM). Patients treated with multi-isocentric fractionated SRS on 2-3 lesions on a linac spend a comparable amount of time in the mask. In this study, we evaluated the IFM of these linac patients as a bench mark for the IFM that can be expected in ICON patients. Material and Methods 85 patients with two lesions and 26 patients with three lesions, treated with single fraction (18-24 Gy) linac based SRS within the same session were evaluated. The patients were immobilized in a hybrid mask. Each lesion had its own isocentre and for each lesion a Cone Beam CT (CBCT) to correct for the setup error, a verification CBCT and a post-treatment CBCT was performed. A rigid registration was performed on the bony anatomy. If the residual translation was > 2mm on the verification CBCT, the correction-verification procedure was repeated. If rotations were > 3° the patient was repositioned. The residual errors of the post CBCT and verification CBCT of each lesion were subtracted to calculate the IFM vector length per lesion. The sum of the IFM of all lesions was calculated to give the total cumulative IFM per patient. The mean and standard deviation (SD) was then calculated over all patients. The time between the first CBCT and the last post CBCT was defined as session time. The correlation between the session time and cumulative IFM was determined.

Through a checklist we avoid mistakes and capture all the information about the immobilization, in order to improve the daily reproducibility of the treatment, minimizing the error inter and intrafraction. Results We analyzed a total of 40 patients, aged between 55-89y and 164 images were acquired from post-treatment CBCT. The majority of the techniques used were VMAT but some patients received IMRT or 3D. The mean time for each fraction was 23 min (Range 9-51). Movements range reported in every axis were: vertical (AP) -4 to 5mm, longitudinal (CC) -6 to 4mm and in lateral(LR) -3 to 3mm. Other results are displayed in table 1.