ESTRO 37 Abstract book

ESTRO 37

S397

United Kingdom 2 Velindre Cancer Centre, Physics, Cardiff, United Kingdom 3 The Christie, Clinical Oncology, Manchester, United Kingdom 4 Gray Institute of Oncology, Clinical Oncology, Oxford, United Kingdom 5 South West Wales Cancer Centre, Clinical Oncology, Swansea, United Kingdom Purpose or Objective RTTQA is required to ensure adherence to the outlining and planning aspects of the trial protocol as variations can lead to poorer outcomes. NeoSCOPE, a UK phase II randomized control trial of two neoadjuvant chemoradiotherapy regimens in oesophageal cancer, undertook a comprehensive RTTQA programme with both pre-accrual and prospective on-trial QA of outlining and planning of every patient recruited into the trial. Material and Methods A detailed radiotherapy protocol was provided to 19 centres who were required to submit and pass two pre- accrual benchmark cases for outlining and planning, with detailed feedback provided. The outlining cases were compared against a consensus reference volume. On trial, prospective individual case reviews (ICR) of outlining and planning were undertaken for all patients. Real time review (feedback within 3 working days) was performed on the first 20 patients recruited and the first case submitted from each participating centre. Subsequent cases were subject to timely retrospective review, with the aim of feeding back in ≤2 weeks of commencing RT. Reviews were undertaken by a clinical oncologist and physicist within the trial QA group, against a pre-determined consensus of acceptable and unacceptable deviations. Unacceptable variation required re-submission. Results Of 55 pre-accrual cases submitted, 8 (15%) had unacceptable variations and 17(31%) had acceptable variations. The most common unacceptable variation was GTV length (7/8 cases) but following centre explanations for major deviations including differing CT slice thickness, their understanding of the protocol was confirmed and only 4 (7%) required re- submission. 83 ICR cases were undertaken, 39 (47%) of which were real-time and 44 (53%) timely-retrospective. 10% of the cases required resubmission due to unacceptable variations, the most common being errors in delineation of GTV. Of the on-trial plan reviews, 5% of cases required re- submission, most commonly due to PTV undercoverage. 6 of the 8 outlining cases requiring resubmission were either the 1st or 2nd patient recruited by a participating centre. 74% of the real-time reviews were returned within 3 working days and 100% of timely retrospective reviews were returned by the 3 rd fraction. There were no delays to patients’ treatment as a result of the QA process. Conclusion The NeoSCOPE comprehensive RTTQA programme is one of the 1st trials to perform QA review on all patients within a trial, without resulting in a treatment delay. Despite a pre-accrual programme the incidence of unacceptable outlining variation in the on-trial group was 10%, supporting the need for continuous review both prior to and for the duration of the trial. PO-0770 concurrent chemoradiotherapy using photon and proton for locally advanced pancreatic cancer Y. Hiroshima 1 , N. Fukumitsu 1 , T. Ishida 1 , M. Nakamura 1 , S. Shimizu 1 , Y. Sekino 1 , D. Miyauchi 1 , T. Iizumi 1 , K. Miura 1 , R. Kanuma 1 , K. Tanaka 1 , T. Saito 1 , D. Takizawa 1 , H. Numajiri 1 , K. Onishi 1 , M. Mizumoto 1 , T. Aihara 1 , H. Ishikawa 1 , T. Okumura 1 , K. Tsuboi 1 , H. Sakurai 1

landmarks to show a visual impression. Association between clinicopathological factors and LNs at high risk was further investigated. Results Based on the recurrent lymph nodes (LNs) contoured, we showed high-risk regions for relapse and drew a density distribution map of 16 LNs stations on CT images. The most commonly involved recurrent lymph nodes were No.16b (paraaortic LNs between the lower border of the left renal vein and the aortic bifurcation, 51.2%) and No.16a(paraaortic LNs between the diaphragmatic aortic hiatus and the lower border of the left renal vein, 39.5%). No.13, No.12, No.9 and No.14 nodes were involved in 36.4%, 33.3%, 28.7%,and 27.9%, respectively.Other regions including No.1-6 and No.10 were at low risk. Meanwhile, it is noteworthy that 72%(83/116)of recurrent No.16b LNs located at the upper half of 16b1 region. An analysis in subgroup showed a higher trend of No.12 and 13 involvements in distal third of the stomach although no significant difference was seen. Moreover, failure of LNs around the abdominal aorta regardless of 16a or 16b nodes appears more frequently in patients with N2 or N3 disease compared with stage N0-1(P=0.031,P=0.019) and pathologic N stage was the only independent risk factor of No.16b1 relapse.

Conclusion Our mapping provided a new suggestion for the vessel- guided delineation of region lymph nodes when defining the CTV in patients after standard D2 resection. LNs around the abdominal aorta and its main branches, and regions around hepatic hilar and head of pancreas were considered to be the high risk regions. Meanwhile, radiation field could be shrunk by excluding parigastric LNs, splenic hilar LNs and nodes below the middle of LRV and IMA (lower half of 16b1 and 16b2) to reduce the adverse reaction. PO-0769 NeoSCOPE RTTQA: pre-accrual and on-trial review of all patients in a UK oesophageal RT trial E. Evans 1 , G. Jones 2 , T. Rackley 1 , R. Maggs 2 , G. Radhakrishna 3 , S. Mukherjee 4 , M. Hawkins 4 , T. Crosby 1 , S. Gwynne 5 1 Velindre Cancer Centre, Clinical Oncology, Cardiff,