ESTRO 37 Abstract book

S776

ESTRO 37

prior to treatment. The fiducial marker was percutaneously placed, under local anesthesia, near the tumor by a hepatobiliary surgeon. The GA was 0.28 mm in diameter and 10 mm in length and had a shape that could be bent to fold the marker. MRI was performed 1 week after the implantation of the GA. Planning contrast- enhanced four-dimensional (4D) computed tomography (CT) scans and MRI images were used to determine gross tumor volume. The prescription radiation doses were designed to deliver the prescription dose to cover 95% of the planning target volume (PTV) in stereotactic body radiation therapy (SBRT) and to deliver the mean dose of the prescription dose for PTV in conventional radiotherapy using IMRT. IMRT was administered using volumetric modulated arc therapy (VMAT). A phantom study was performed in order to evaluate if the variety of the placed forms affects the precision of the verification. The GA was placed in three different forms (linear, folded, and tadpole-like forms) in gelatin. 4D-CT and 4D cone-beam CT (CB-CT) were performed in order to acquire the localization images.Then, automatic verification was performed between average images of the reference images and those of the localization images. Each 3D vector setup error was calculated from the acquired setup errors using the following formula; d3D=√dx2+dy2+dz2. Results All procedures for placement of GA were successfully performed with no complications. The GA showed various forms and was well detected in both CT and MRI and in CB-CT images in all fractions of the radiotherapy course. In the phantom study, no significant differences were observed in the automatic validated setup errors among the three different forms of the GA. Conclusion Our findings show that an iron-containing marker, the GA, is useful in image registration including CB-CT and MRI and that the tadpole-like shape can be recommended for liver radiotherapy. The present findings suggest that the GA will indeed be useful in clinical practice. EP-1426 ChemoRadiotherapy for Oesophageal cancer followed by Surgery Study (CROSS): Single Center Experience A. Tozzi 1 , C. Iftode 1 , T. Comito 1 , C. Franzese 1 , F. De Rose 1 , D. Franceschini 1 , G.R. D'Agostino 1 , G.A. Carta 1 , G. Reggiori 1 , M. Scorsetti 1 1 Istituto Clinico Humanitas, Radiotherapy and Radiosurgery, Rozzano Milan, Italy Purpose or Objective Multimodal approach is a standard treatment for locally advanced oesophageal and junctional cancer (OC). CROSS is one of the referral regimens for the treatment of this disease and it has been adopted as our center’s main regimen since 2012. The purpose of our study was to evaluate the role of volumetric-modulated arc therapy (VMAT) in the neoadjuvant chemoradiotherapy management of advanced medium and distal oesophageal cancer in terms of toxicity and response to treatment. Material and Methods From November 2012 to June 2017 consecutive fit patients (ECOG-PS) with resectable, locally advanced (T3-4,N0 and AnyT,N+ M0) OC were enrolled to receive neoadjuvant CROSS regimen. All underwent FDG-PET scanning before and after induction chemoradiotherapy. Pathologic uptake was observed in all patients. Patients showing no progression over neoadjuvant treatment were evaluated for oesophagectomy. Results The analysis included 66 patients, with a median age of 65 years (range of 40 - 81) and a generally low ECOG-PS (0-2). We had no cases of pulmonary or cardiologic toxicity > G2-G3. Sixteen (24.24%) patients had disphagia G2. Ten patients (15.15 %) did not receive a complete

CROSS regimen. Fifteen patients (22.7 %) did not undergo surgery due to progressive disease (9 pts), unfitness (3 pts) and death occurring during neoadjuvant treatment (3: 2 myocardial infarction and 1 aortic dissection). Fifty- one patients (82.2 %) underwent oesophagectomy. Two experienced radiologists and nuclear medicine physicians specialized in gastro-oesophageal cancer reviewed both pre- and post neoadjuvant treatment CT images and PET/CT images, respectively. The sensitivity of CT and PET/CT were 94% and 95% respectively versus final histology. Partial or complete response was observed in more than 90% of the cases (radiological/metabolic) and was confirmed after surgical intervention (67% partial or complete and 27% stable response). Tumor down-staging was recorded in 67% of patients and nodal down-staging in 50%. Complete pathological response was recorded in 15 cases (22.7 %). Median follow-up was 21.9 months. Conclusion VMAT was applied in the context of neoadjuvant chemoradiotherapy for the treatment of medium and distal oesophageal carcinoma with satisfactory results in terms of tolerance and toxicity. EP-1427 A comparative study of the normal oesophageal wall thickness based on 3DCT, 4DCT and CBCT S.C.H.&.I. Chaoyue Hu 1 , S.C.H.&.I. Jianbin Li 1 , S.C.H.&.I. Jinzhi Wang 1 , S.C.H.&.I. Qian Shao 1 , S.C.H.&.I. Wei Wang 1 , S.C.H.&.I. Yanluan Guo 1 , S.C.H.&.I. MIn Xu 1 , S.C.H.&.I. Wenwu li 1 , S.C.H.&.I. Yong Huang 1 1 Shandong cancer hospital affiliated to Shandong Universi ty, Department of Radiology- Shandong Cancer Hospital & Institute, Jinan, China Purpose or Objective The purpose of this study was to compare the normal oesophageal wall thickness based on three-dimensional computed tomography (3DCT), four-dimensional CT (4DCT) and cone beam CT (CBCT) to provide a reference for the delineation of oesophageal tumours. Material and Methods Contrast-enhanced 3DCT, 4DCT and CBCT scans were acquired from 33 patients with lung cancer or metastatic lung cancer. The outer oesophageal wall was manually contoured on each 3DCT, 4DCT MIP ( the maximum intensity projection of 4DCT), 4DCT 50 (the end expiration phase of 4DCT) and the CBCT data sets. The average thoracic and intra- abdominal oesophageal wall thicknesses were measured (defined as R 3DCT , R 50 , R MIP , and R CBCT ). Results For the thoracic and the intra-abdominal segments, there were no significant differences between R 3DCT and R 50 , but there were significant differences between R 3DCT and R MIP , and R 3DCT and R CBCT (p=0.000–0.013). For the lower and intra-abdominal oesophagus, there were no significant differences between R CBCT and R MIP (p=0.170, p=0.130). However, for the upper and middle oesophagus, R CBCT were larger than R MIP (p=0.014, p=0.006). There were no significant difference between upper and middle segments on 3DCT, 4DCT (4DCT 50 and 4DCT MIP ) and CBCT images. Intra- abdominal oesophageal wall thickness was greater than that of the thoracic oesophagus. There were no differences between the upper and lower, and middle and lower oesophagus (p=0.053, p=0.377) on CBCT images. Conclusion Oesophageal wall thickness of the same segment differed according to imaging modality. Uniform criterion to delineate the gross target volume could not be adopted. The thickness of the upper and middle segments on the same CT image differed no remarkably.