ESTRO 38 Abstract book

S896 ESTRO 38

The primary sites were 17 (71%) gastrointestinal including seven (41%) pancreatic, genitourinary (5, 21%), or unknown primary sites (2, 8%). The median Ki67 was 90% for all patients. Five patients were diagnosed with limited disease and operated, one received neoadjuvant treatment and is still free of disease 10 years later, while one patient received adjuvant treatment and had a DFS of 4.7 months. Nineteen patients were diagnosed with extensive disease, 14 of them received Platinum based chemotherapy as first line of treatment, mostly (13, 93%) with Etoposide, same treatment protocol as in SCLC. 6 (31.6%) and 3 (15.7%) patients received second or more lines of systemic treatment respectively. Of the 10 patients we have an imaging test after first course of treatment we can determine that 3 (30%) had a partial response and 4 (40%) had a stable disease. For the total of patients with extensive disease the median survival was 8.2 months. While no patient received prophylactic cranial irradiation, only one patient (4%), presenting with pancreatic SCC developed brain metastasis, compared to the 40-50% known in SCLC. Conclusion We report herein a relatively large number of patients. Our data demonstrates a relative efficacy for the classic platinum based treatment. While treatment of EPSCC is often extrapolated from SCLC, the rarity of brain metastasis in this cohort suggest that PCI should not be incorporated in the treatment protocol of these patients. EP-1666 ARENA: Improving training in target volume delineation for radiotherapy E. Evans 1 , C. Piazzese 2 , E. Spezi 2 , J. Staffurth 1 , S. Gwynne 3 1 Velindre Cancer Centre, Clinical Oncology, Cardiff, United Kingdom ; 2 Cardiff University, School of Engineering, Cardiff, United Kingdom ; 3 South West Wales Cancer Centre, Clinical Oncology, Swansea, United Kingdom Purpose or Objective Inter-observer variation radiotherapy (RT) Target Volume Delineation (TVD) may be considered the ‘weakest link’ in the RT planning process. Accurate outlining is imperative to ensure patients receive optimal outcomes with minimal toxicity. In response to this the ARENA project has been launched by clinical oncologists in South Wales to facilitate a standardised approach to TVD training. The project will develop tumour site-specific TVD instructional modules with corresponding outlining modules, offering test cases for trainees to outline, upload and receive semi- automated feedback on through software developed by Cardiff University. Experience has been drawn from the team’s involvement with National Radiotherapy Trials Quality Assurance (RTTQA) group, providing feedback for submitted pre-trial RT outlining. Material and Methods To ascertain TVD training needs of clinical oncology trainees, 406 UK clinical oncology trainees were surveyed regarding TVD training quality and preferential format for TVD modules. Using the survey results, a TVD instructional module and corresponding series of cases of varying difficulty with semi-automated feedback have been developed using oesophageal cancer as a pilot site. 6 UK UGI oncologists outlined 6 oesophagus cases to develop a reference volume for each case. Software was developed in MATLAB to provide semi-automated on the cases outlined by trainees on their training centre’s planning software. Feedback developed included trainee outline against a reference volume, over /under contoured regions, minimum and maximum acceptable volumes, a conformity score and a ‘red flag’ when volumes contoured inappropriately include normal tissue.The latter module with semi-automated feedback was piloted locally with 3 clinical oncology trainees.

last glass of water and CBCT was noted. The bladder protocol was followed from the planning day and then on daily basis before treatment. The CBCT data set was fused to the planning CT and was used to characterize the bladder each day. Bladder contouring was done on all planning CTs and treatment CBCT images by the same oncologist. The volumetric changes of the bladder were measured to analyze the inter-fractional filling variation. Total 274 CBCT was contoured in 26 patients and were compared with the 26 planning CT to see the average bladder filling variation and standard deviation in transverse, anteroposterior and longitudinal diameter. Results The mean bladder volumes in-room 183.07 cc (range, 62– 475cc) and SD were 90.43cc. Similarly, the mean time was 40.12 minutes with SD of 2.15 min and p-value 0.059 which was non- significant. Bladder wall motion of these patients was analyzed in transverse, anteroposterior and longitudinal diameter shown in Table-1.

Table 1- Mean± SD of bladder wall movements

Conclusion This study finds a possible answer to a very pertinent question on bladder wall motion and bladder volume variations during radiation therapy. Image-guided radiotherapy (IGRT) with verification of the organ position before the daily treatment after following bladder filling protocol has allowed for lessening of inter-fraction bladder wall motion and showed considerable benefits in terms of margin reduction on lateral side as there is less displacement on transverse diameter and more liberal margins should anteroposterior in anteroposterior dimension and longitudinal dimension. EP-1665 Extra-pulmonary Neuroendocrine Carcinoma. Rarity of Brain Metastases L. FERRO 1 , I. Golomb 2 , H. Ligumsky 3 , O. Gutfeld 1 , I. Wolf 3 , V. Soyfer 1 1 Sourasky Medical Center, Radiation oncology- Oncology Division, Tel Aviv, Israel ; 2 Tel Aviv University, Sackler school of medicine- Tel Aviv University, Tel Aviv, Israel ; 3 Sourasky Medical Center, Oncology Division, Tel Aviv, Israel Purpose or Objective Extra-pulmonary small cell carcinoma (EPSCC) is a rare clinical entity and while its management is usually extrapolated from small cell lung cancer (SCLC), this approach is controversial and is not evidence-based. Specifically, the current clinical practice in the management of SCLC involves prophylactic cranial irradiation (PCI). The PCI proved the decrease in incidence of brain metastases both in limited and extensive disease. We aimed to analyze the clinical characteristics of a relatively large cohort of patients treated at single tertiary center. Material and Methods A search of the entire database of the Center, consisting of over 25,000 patients, yielded 24 patients diagnosed with EPSCC or high-grade neuroendocrine carcinoma between 2015 and 2018. Clinical characteristics were obtained from the medical files. The overall median survival, response rate, and the incidence of local and distant progression ware calculated. Results