ESTRO 2020 Abstract Book

S1065 ESTRO 2020

Purpose or Objective The margins from the CTV to the PTV must be specific for each center, since it influences: treatment units, CT simulator, personnel and workflow. Given the difficulty in quantifying the data in our center, due to the type of image verification in these patients until 2017, the data reported in the literature were used. With this work, we wanted obtained the margins from CTV to PTV to be used in patients with breast cancer in our center. Secondary endpoint was to assess if there was any difference in margins between mastectomy (MS) versus (vs.) breast conserving surgery (BCS). Material and Methods Women breast cancer radiotherapy treatments performed in 2017/2018 were reviewed. The inclusion criteria for the analysis were: treatments in Clinac 2100-IX linear accelerator (CL2100), electronic portal imaging (EPID) within of the extended-no-action-limit correction strategy (eNAL) and inclined plane use. 283 patients with BCS and 142 patients with MS were analyzed. A total of 117 BCS patients and 117 MS patients met the criteria for the study. All treatments were in the CL2100, following of the eNAL, based on 3 pairs of orthogonal images the first 3 days, and a weekly imaging verification. We used a bone matching protocol: vertebral column to assess the VRT and LAT axes, and sternoclavicular joint for the LNG axis. When the mean shift was >5mm, the setup was modified, checking it the next two days with new imaging. The shifts of all EPIDs were collected in the three axis of space: Vertical (VRT), Longitudinal (LNG) and Lateral (LAT). The Van Herk formula was used to obtain the margins from CTV to PTV. Independent t-test and Mann-Whitney U test were used for comparing groups. Results We analyzed 3444 images from 1725 sessions in the 234 treatments reviewed. In the MS group there were 1776 images and 1668 in the BCS group. The values obtained from the shifts in VRT, LNG and LAT axes, respectively, were: - Mean systematic error: MS Group: -1.6mm, 0.9mm, -0.7mm; and BCS Group: - 1.8mm, 0.4mm, -0.2mm. - Standard deviation of the systematic error: MS Group: 1.8mm, 1.4mm, 2.1mm; BCS Group: 1.8mm, 1.4mm, 2.3mm. - Standard deviation of the random error: MS Group: 2.5mm, 2.1mm, 2.5 mm; BCS Group: 2.5mm, 2.3mm, 3.0mm. - The margins from the CTV to the PTV: MS Group: 6.6mm , 7.1mm y 6.4mm; BCS Group: 6.6mm, 7.7mm and 6.6mm. When comparing the groups it was observed that the shifts made in the LNG and LAT axes were different, being statistically significant. There were no differences in the VRT axis. LNG: MS vs. BCS = 0.545mm, 95%CI: 0.309, 0.781; p<0.0001. LAT: MS vs. BCS = -0.508mm, 95%CI: -0.829, -0.187; p=0.0019. Conclusion We have verified that the margins from CTV to PTV used to date in our center are suitable in breast cancer patients. The shifts in LNG and LAT axes are different in patients according to the surgery performed, but do not lead to a clinically significant change in the margins from CTV to PTV.

0.871-0.957). The carina was used to guide difficult matches (20%) n the mediastinum and timings were comparable with 2DkV imaging (4 minutes). Conclusion This study supports the introduction of CBCT imaging and soft tissue matching to tumour as routine clinical practice for radical non-SABR lung radiotherapy. Manual inspection of the visible tumour is highly recommended after automatic matching to ensure accurate target coverage, with the use of the carina as a guide when problematic matching occurs. Further dosimetric assessment of direct tumour matching is advised PO-1909 Interobserver variability of CBCT for prostate radiation therapy: Fiducial Markers vs CTV/PTV F.A. Lima Aires 1 , D. Saraiva 1 , F. Costa 1 , P. Peixoto 1 , A. Monteiro 1 , S. Garcia 1 , M.G. Pinto 1 1 Centro Hospitalar Universitário de São João, Radioncology, Porto, Portugal Purpose or Objective The usage of image guidance is crucial for the radiation therapy treatment (RT) of prostate cancer with VMAT. For this purpose a daily CBCT is acquired prior to the treatment and then it is then compared to a reference CT scan for deviation assessment. At our department two methods may be used to assess these deviations: using implanted fiducial markers as a reference or, when in lack of such markers, using the target volumes (CTV/PTV) boundaries. The purpose of this study is to compare both methods in order to evaluate which presents less interobserver variability. Material and Methods Forty-four patients diagnosed with prostate cancer, treated between March and September 2019, are included in this study. The sample was divided in two equally sized groups and CBCT images were analyzed in offline-review : Group A(n=22) – Calculation of deviation and the time through fiducial markers alignment; Group B(n=22) – Calculation of deviation and the time through PTV/CTV adjustment (without fiducial markers). The inter-observer agreement of CBCT review was assessed using the Bland- Altman method, modified for each alignment process. The 95% limit of agreement (LoC) was measured regarding the average deviation for each observer. The difference between group A and group B inter-observer alignment only focused on translational movements, excluding rotational movements. Results 357 images were analyzed, presenting a mean standard deviation (SD) of 0.079 to group A and 0.153 to group B; The 95% LoC of group A was ±0.14, ±0.16 and ±0.16 and ±0.18, ±0.43 e ±0.29 for group B, at left-right direction, superior-inferior and anterior-posterior, respectively. Conclusion This study shows differences of deviations between group A and group B. Achieved results suggest that The fiducial markers alignment method presents a reduction of the time and of the inter-observer variability, leading to a higher positioning reproducibility. Based on this study, the implementation of fiducial markers in all patients diagnosed with prostate cancer may represent an added value for the treatment of prostate cancer with VMAT. PO-1910 Margins from CTV to PTV in Breast Cancer Pacients: Comparison between tumorectomy and mastectomy J. Matheos Pérez 1 , M. Altabas Gonzalez 1 , B. Pérez Esteve 1 , M. Comino Muriano 1 , A. Giraldo Marín 1 , S. Micó Milla 1 , R. Granado Carrasco 1 , D. Garcia Relancio 1 , V. Reyes López 1 1 Hospital Universitario Vall d'Hebron, Radiation Oncology, Barcelona, Spain