Abstract Book

S1247

ESTRO 37

in the bowel bag, with higher values for MV CBCT vs portal image (Z= -2.041, p= 0.041). Conclusion Radical treatment of prostate cancer should not be performed without daily imaging due to the prescription of high doses of radiation and the complexity of the techniques used. The daily MV CBCT image is undoubtedly more time-consuming and showed to determine a lower PTV coverage and higher doses in the bowel bag when compared with the portal image. EP-2383 Triggered imaging for hypofractionated prostate treatment on Varian TrueBeam linear accelerator J. Ter-Minasjan 1 , M. Marjamäe 1 1 North-Estonian Regional Hospital Cancer Center, Radiation Therapy, Tallinn, Estonia Purpose or Objective Dose delivery accuracy for hypofractionated radiation therapy of prostate cancer is very important in achieving the high dose in prostate whilst not exceeding the dose to organs at risk. There is a need for comprehensive imaging protocols in order to get the planned dose distribution. The aim of the poster is to present the introduction of the treatment and image guidance method for the patients who underwent hypofractionated external beam irradiation of prostate cancer. Material and Methods From September 2016 to September 2017 25 patients were treated on Varian TrueBeam linear accelerator with hypofractionated treatment using VMAT delivery technique (1-2 arcs). Patient with normal anatomy, localized prostate cancer with medium and high progression risk were selected for hypofractionated treatment. 3 fiducial gold markers (1x3 mm) were implanted in each patient for the imaging system to localize the target. The prescribed dose was 3 Gy per fraction to PTV for 20 consecutive fractions. Firstly, kV images from AP and LAT position were taken and 2D/2D match based on gold seeds and couch shift were performed. Then, Auto Beam Hold window is activated and appropriate settings are chosen: image every 60 degrees with adjusted field size to reduce the radiation dose to the patient outside the target. This results in typically 12 intrafractional images during 2 arc treatment. This is a compromise between imaging dose and accuracy achievable. The allowed movement diameter for each marker was set to 6 mm. To detect markers automatically ROI/VOI as small as possible around markers is selected. If automatic registration is not successful then the markers can be selected manually. Treatment starts and continues until markers are in allowable range. Otherwise, the treatment is paused and additional imaging is performed: marker match or 2D/2D or 2D/3D match are in option. After shifting of the couch the treatment can be resumed. Results Using described method it was possible to implement the hypofractionated prostate external beam irradiation. As a result the number of fractions (from 35–38 to 20) decreased, while the accuracy of dose delivery increased. Due to successful implementation of triggered imaging it was possible to reduce the PTV margins from 7 to 5 mm. The interruptions of the treatment due to internal movement were seen for almost all patients indicating the importance of the intrafractional imaging procedure. In most cases the fraction took on average 7 min more than the standard fraction (without intrafractional

Variation between physician registration / radiation therapist (median and maximum values, in cm)

Registration of radiation therapist after CBCT (median and maximum values, in cm)

Longitudinal axis

0,31 (0,00 – 2,59)

0,07 (0,00-0,54)

Vertical axis 0,30 (0,00 – 3,06)

0,04 (0,00 – 0,75)

Lateral axis 0,26 (0,00 – 2,10)

0,07 (0,00 – 0,78)

Conclusion IGRT improved the precision of IMRT delivery in esophageal cancers. A PTV margin of 5 mm seemed to be sufficient in most of the cases. A long term follow up is necessary. EP-2382 Impact on treatment plan of daily MV EPI versus MV CBCT imaging for prostate cancer IGRT A. Monteiro 1 , D. Rodrigues 1 , M.J. Fontes 1 , P. Varzim 1 , R. Figueira 1 , R. Lago 1 , S. Garcia 1 , T. Reis 1 , A.L. Carvalho 1 , G. Pinto 1 1 Hospital de São João, Serviço de Radioterapia, Porto, Portugal Purpose or Objective Radical radiotherapy for prostate cancer involves the administration of high doses of radiation therapy and should not be performed without a daily check image (IGRT). MV electronic portal images (EPI) or kV/MV conebeam CT (CBCT) are tipycally used as imaging modalities for prostate positioning. In the first case fiducial markers implanted in the prostate are used as a surrogate of the daily prostate position. Although the acquisition of kV/MV CBCT images is more time- consuming, this type of image is often used in our department, in detriment of portal images, due to logistics issues related to the implantation of the fiducial markers. The purpose of this study was to compare the impact on the dose distribuition when both this daily images are incorporated in the treatment plan. Material and Methods Twenty patients with radical treatment for high risk prostate cancer were selected, with a prescription of 48Gy (2Gy per day) to a PTV1 (pelvic lymph nodes + prostate) followed by a boost of 28 Gy (2Gy per day) to a PTV2 (prostate). For all selected patients, two plans were prepared: one with MV CBCT and the other with MV EPI, the imaging modalities avaliable in our department. For both cases, 38 imaging fractions were considered: two daily orthogonal EPIs, whose daily contribution is of the order of 1 cGy and was not included in the prescribed 2Gy/fraction; or one MV CBCT, whose daily contribution is approximately 10 cGy and was considered as part of the 2 Gy/fraction. The PTV was analyzed according to ICRU 83 (D50, D98, V100 and D2) and organs at risk: rectum (V60), bladder (V60), femoral heads (V50), bowel bag (V45) and bulb (D90), according to Quantec/RTOG guidelines. Statistical analysis was performed using SPSS v. 24. The comparative analysis between the two plans (MV CBCT images and EPI images) was enhanced using the Wilcoxon Signed-Rank Test. Results The comparison of the plans showed that there are significant differences in the PTV coverage (D50 and V100), with lower values for MV CBCT vs portal image (D50: Z= -2.236, p= 0.025; V100: Z= -2.023, p=0.043) and