S658 3rd ESTRO Forum 2015 Conclusions: Definitive chemoradiotherapy using Carboplatin and Paclitaxel for inoperable esophageal cancer led to symptom relieve and a median overall survival of 13 months. Although acute toxicity grade ≥ 3 was common, the majority of patients were able to finish treatment. Given the toxicity profile and relatively short median survival, adequate selection of patients for this intensive treatment is required. In addition, as local recurrence rate was high, further dose escalation may be beneficial. EP-1215 Optimum time to assess complete response following radiochemotherapy in anal canal cancer patients M. Scricciolo 1 , G. Mantello 1 , F. Cucciarelli 1 , L. Vicenzi 1 , L. Fabbietti 1 , F. Fenu 1 , M. La Macchia 1 , E. Morabito 2 , S. Maggi 2 , M. Cardinali 1 1 Azienda Ospedaliero Universitaria Ospedali Riuniti, Radiation Oncology Dept., Ancona, Italy 2 Azienda Ospedaliero Universitaria Ospedali Riuniti, Physic Dept., Ancona, Italy Purpose/Objective : Response evaluation of anal canal neoplasm after Radiochemotherapy (RCT) represents a significant problem correlated to the tumour slow regression and to difficulties in response interpretation. Recent findings indicate that a complete clinical response occurs in the majority of patients within 6 months from the end of RCT, even if tumour regression may be slower. The aim of this study was to define the optimal time to assess complete response following RCT. Materials and Methods: Twenty-four patients (pts) (7 male, 17 female; median age 58 years, range 42 to 85; 15 T1-T2, 9 T3-T4, 12 N+) affected by epidermoid anal canal carcinoma, all HIV-, were treated with RCT (FUMIR) between November 2005 and July 2012. Total dose at primary tumor ranged from 45 to 65 Gy with a median of 59.4 Gy, 1.8 Gy fraction. After treatment pts were evaluated with DRE (by the same physicians GM and FC) and with serial EAUS carried out by the same operator (RG), every month in the first 4 months, and then every two months up to 1 year. Abdominal-pelvis MRI was performed at 3, 6 and 12 months. Median follow up was 34 months (range 10-80). A complete response to therapy was defined as no clinically detectable disease (or residual scar) at DRE and a reduced mass with a stable hyperechogenic images at EAUS . Results: Complete Response (CR) to therapy was seen in 21 pts (87.5%). Progressive disease occurred in 3 pts: 2 pts performed salvage surgery and 1 patient received chemotherapy for concomitant distant metastasis. Among the responders (21), only 6 ( 28.6 % ) had CR at 4 months. At 6 months (time suggested by guidelines for a final evaluation) only half patients ( 11/21) had a complete regression. All the CR were obtained within 14 months. Conclusions: In our experience, extending the evaluation time over 1 year, 10 patients with CR between 8 and 14 months avoided an unnecessary demolitive surgery. A final evaluation at 6 months to address patient to a salvage surgery could be inappropriate for late responder patients. EP-1216 Changes in preoperative 18FDG-PET/CT after neoadjuvant treatments and pathological response in rectal cancer L. Porta 1 , B. De Bari 1 , M. Cerny 2 , A. Pomoni 3 , S. Schmidt 2 , J. Prior 3 , J. Bourhis 1 , M. Ozsahin 1 1 Centre Hospitalier Universitaire Vaudois, Department of Radiation Oncology, Lausanne Vaud, Switzerland 2 Centre Hospitalier Universitaire Vaudois, Radiology Department, Lausanne Vaud, Switzerland 3 Centre Hospitalier Universitaire Vaudois, Nuclear Medicine, Lausanne Vaud, Switzerland Purpose/Objective: It is still controversial if changes in 18 FDG-PET/CT after neoadjuvant treatments for rectal cancer could predict pathological response. In this preliminary report, we analyzed changes of several 18 FDG-PET/CT parameters obtained before and after neoadjuvant treatments for locally advanced rectal cancer. Materials and Methods: Twelve locally advanced rectal cancer patients consecutively enrolled in a prospective institutional trial were analyzed. All these patients underwent whole body MRI and 18 FDG-PET/CT before and after neoadjuvant treatments for initial staging and for restaging before surgery. Nine patients received long-course chemoradiotherapy (45 Gy on pelvic nodes and mesorectum and 50 Gy on gross tumor volume and corresponding mesorectum, delivered with a simultaneous integrated boost) and 3 patients a short-course radiotherapy (25 Gy in 5 fractions delivered on pelvic nodes and mesorectum). All patients were treated with helical tomotherapy and daily image guided radiotherapy, and all of them underwent radical surgery. 18 FDG-PET/CT studies were integrated in the Velocity© software. A standardized uptake value (SUV) > 3 was used as threshold to automatically contour the rectal gross tumor volume on the pre-treatment 18 FDG-PET/CT. Then, this volume was projected on the post-treatment PET/CT, after a deformable image fusion of the pre- and post-therapy exams. We analyzed variations in terms of SUV max , SUV min , SUV mean in this area and compared them to final pathological response evaluated with the Mandard tumor regression grade (TRG). Results: Despite important variations in the studied PET/CT parameters, most of the patients presented the same TRG = 3 (7/12 patients). Moreover, we also noted that some patients presenting different trends in metabolic responses were finally classified as TRG = 3 patients. Nevertheless, linear