ESTRO 2023 - Abstract Book

S255

Saturday 13 May

ESTRO 2023

optimization. The plan is reviewed and approved by the physician. When the patient is brought to the treatment room, a kilovoltage cone-beam CT image is acquired. Targets and organs-at-risk from the original plan are auto-segmented on the new image and edited as necessary. The plan is reoptimized to account for changes in body positioning and target volume and then delivered. Results We successfully treated 10 patients requiring palliative care with 800 cGy in a single fraction using this workflow. Disease sites included: sternum, rib, spine, sacrum, iliac crest, and a retroperitoneal mass. Diagnostic scans for initial planning were on average from 47 days (range: 11-157 days) prior to treatment. Default Ethos beam arrangements were used for 6/10 cases, the remaining targets benefited from custom beam arrangements (e.g. partial arcs entering anteriorly for the sternum). The auto-segmented GTV was edited for 8/10 cases (volumes changed 13.3cc±26.6cc). In all cases, the adapted plan was selected for treatment. All adapted plans were normalized for PTV D95%=100%, PTV D0.03cc ranged from 104- 113%. If the initial plan had been delivered without adaptation, PTV D95% would have decreased to 65.1%±33.9%. Average total time for the on-couch adaptive process was 11.0±3.4 minutes. Average total in-room times were 19.1±2.9 minutes. Average total in-clinic time for patients (from consult check-in to end of treatment) was 3.2 hours. Conclusion Prior to the implementation of this workflow, palliative patients required separate appointments over multiple days for consult, simulation CT, and treatment. This new workflow reduces on-site time to a few hours and frees clinical resources. Adapting the dose ensures excellent target coverage despite changes in target volume between the diagnostic scan and treatment. PD-0321 Using the axillary vein & artery to validate metrics for high tangents within the POSNOC trial R. Butt 1 , R. Patel 1 , C. Brooks 1 , G. Jackson 1 , D. Dodwell 2 , E. Miles 1 , A. Goyal 3 1 National Radiotherapy Trials Quality Assurance Group, Mount Vernon Cancer Centre , Radiotherapy Physics, London, United Kingdom; 2 Oxford University Hospital, Oncology, Oxford, United Kingdom; 3 Royal Derby Hospital, Oncoplastic Breast Surgery , London, United Kingdom Purpose or Objective The purpose of this study was to validate the use of simple anatomical bony landmarks to define high tangents in breast radiotherapy for the UK-ANZ POSNOC (ISRCTN54765244) randomised trial of axillary treatment in women with early stage breast cancer. A comprehensive radiotherapy quality assurance (QA) programme was implemented for the trial. As part of this QA process the patients randomised to the ‘no axillary treatment’ arm were monitored for high tangent use. Materials and Methods Anatomical bony landmarks used for this study are the base of clavicle (BOC) and head of humerus (HH). Simple descriptive metrics; cranial, caudal and proximal positioning of the superior tangent border in relation to the BOC were recorded in addition to a measurement from the inferior most extent of the HH to the superior aspect of the tangents. The ACOSOG Z0011 trial defined high tangents as superior tangent border ≤ 2 cm from humeral head. The axillary vein & artery (AX) were outlined according to the 2016 ESTRO consensus guideline on target volume delineation for elective radiation therapy for early stage breast cancer (figure 1). A binomial logistic regression analysis was used to measure the reliability of the results seen and robustness for the metrics used when accounting for amount of AX within the field.

Results BOC metric, HH measurements and AX outlining was completed for 119 patients.

Use of high tangents was confirmed if any of the tangential field overlapped with the AX volume. The cranial (p<0.003) & caudal (p<0.010) metrics were good predictors for defining high tangents, however proximal (p<0.214) was not significant. The results also suggest that distance from HH is a good predictor (figure 2) for the amount of AX within the field p<0.001.