ESTRO 37 Abstract book

ESTRO 37

S525

two field configurations: A single posterior field (1f-IMPT) and two oblique posterior fields separated by 30 degrees (2f-IMPT). To ensure coverage of the CTV for IMRT, a PTV margin (5mm AP, 5mm LR, 8mm CC) was added. For IMPT, three strategies were pursued. Robust optimization (RO) (3mm isocenter shifts and 3% density uncertainty) of the CTV (IMPT RO ), RO of the CTV and full coverage of the PTV (IMPT PTVRO ) and no RO, but full coverage of the PTV (IMPT PTV ). For all patients, a surveillance CT-scan (sCT) was acquired at F10 and used for recalculation of both IMRT and IMPT plans. V95% CTV and dose to OARs were compared for all plans between planning CT(pCT) and sCT.

Results Dose to OARs was lower for IMPT plans than IMRT plans (Fig 1). Higher skin doses were seen for IMPT plans compared to IMRT. V30Gy Skin was in median 22.3cm 3 (0- 204) for 1f-IMPT PTVROB , decreasing to 0 cm 3 (0-136) for 2f- IMPT PTVRO . Thirteen pts experienced no anatomical changes and maintained V95% CTV >99.5% on sCT for both modalities. For the remaining 13 pts, anatomical changes leading to dose deterioration were seen. Comparing the 6 IMPT strategies, we found no difference in robustness between using one or two posterior fields, but an increase in robustness using IMPT PTVRO (4 pts had V95% CTV <99%) compared to IMPT RO (10 pts had V95% CTV <99%). Comparing IMPT PTVRO with IMRT revealed different results for 7 pts experiencing target deformation (group A) and 6 pts experiencing changes in diapraghm position combined with target deformations (group B). For group A, IMRT (V95% CTV >98.3% for all pts) was slightly more robust than IMPT PTVRO (V95% CTV >97.9% for all pts). For group B, more severe underdosage was seen. The worst case had a large decrease in V95% CTV coverage to 65.3% by IMRT (Fig 2). The IMPT PTVROB was not affected by the diaphragm movement due to the posterior-anterior field entrance, but underdosage was still observed due to target deformation (V95% CTV =91%). If we require V95% CTV >99%, none of the IMRT plans in group B fulfil the requirement, while all but two IMPT PTVRO pass.

Conclusion Our novel patient set-up for SBRT gynecological boost appears to be stable and accurate. Preliminary results show that re-calculation of actual delivered doses using non-rigid deformation is achievable. Larger number of CBCT images and more patients will be used for further analysis.

Poster: Physics track: Adaptive radiotherapy for inter- fraction motion management

PO-0958 Anatomical changes in oesophageal cancer patients: Posterior beam IMPT is more robust than IMRT D.S. Moeller 1 , M. Nordsmark 2 , T.B. Nyeng 1 , M. Alber 3 , L. Hoffmann 1 1 Aarhus University Hospital, Department of Medical Physics, Aarhus C, Denmark 2 Aarhus University Hospital, Department of Oncology, Aarhus, Denmark 3 Heidelberg University Hospital, Department of Radiation Oncology, Heidelberg, Germany Purpose or Objective Anatomical changes are seen in oesophageal cancer patients during the RT course. Delivery of proton therapy is sensitive to anatomical changes. We investigate the robustness towards anatomical changes using IMPT instead of IMRT. Material and Methods 26 patients with central or distal oesophageal cancer were included. 22 patients were treated according to the pre-operative CROSS regimen (41.4Gy/23 fractions(F)), while 4 patients were treated with definitive ChRT (50Gy/27F). The median CTV volume was 289 cm 3 (119- 692). All patients were treated with 5-8 IMRT fields. IMPT plans were generated retrospectively in Eclipse 13.7 for