ESTRO 2023 - Abstract Book

S256

Saturday 13 May

ESTRO 2023

The logistic regression model accuracy was 90%. Conclusion

The study suggests that for the POSNOC trial, the position of the superior border of the tangential fields, relative to BOC, can be used as a time efficient QA tool for defining high tangents. This simple approach can be used for future trials as part of the radiotherapy QA programme. The distance from HH can be used to predict amount of AX within the field which may suggest unintentional treatment of the axillary lymph nodes. Further work may still be required to investigate the robustness of these metrics and to ensure their reliability for future radiotherapy trials QA. PD-0322 Automated VMAT gating for left-sided breast DIBH: few breathholds and small position variations S. Huijskens 1 , P. Granton 1 , K. Fremeijer 1 , C. van Wanrooij 1 , K. Offereins-van Harten 1 , S. Schouwenaars-van den Beemd 1 , B. Heijmen 1 , J. Penninkhof 1 1 Erasmus MC Cancer Institute, University Medical Center Rotterdam, Department of Radiotherapy, Rotterdam, The Netherlands Purpose or Objective Evaluate breathhold (BH) numbers and durations, and position variations in left-sided breast DIBH treatment with automated VMAT gating steered with surface guidance imaging. Materials and Methods All fractions of 15 included patients treated with 15x2.67Gy (n=11) or 20x2.67Gy (n=4) were analyzed. The applied VMAT plans were generated while ensuring a total estimated delivery time <200sec. Automated gating was performed with surface imaging (AlignRT, VisionRT Ltd.) in combination with the Response Module (Elekta AB, Stockholm, Sweden). If the surface region of interest moved in/out the gating tolerance (+/- 3 mm, 3̊), the beam was automatically switched on/off. Patients received BH training a few days prior to planning CT acquisition to hold their breath as long as possible in a comfortable way. Visual feedback was provided to all patients. Depending on the patient’s preference, audio coaching was also performed. During all fractions, the surface guidance system collected real time delta information with respect to the reference surface of the day, for all three orthogonal directions. We analyzed the performance of this optically steered automated VMAT gating for numbers of BHs needed to complete dose delivery, BH durations, and position variations in the three directions. Results Evaluation of the in total 245 treatment fractions showed that only a median of 4 BHs per fraction (5-95% range: 2-6) were needed to complete VMAT dose delivery per fraction. The overall median BH duration with beam-on was 47 sec (patient range: 29-74 sec). In some fractions, a shorter BH was sufficient to complete total dose delivery, clarifying the large range of BH durations for some patients (Figure 1A). Over all patients, the median intra-BH 5-95% spans of motion were 1.0 mm, 1.1 mm, and 1.0 mm for vertical, longitudinal and lateral directions, respectively (Figure 1B). The span of motion range was largest for the longitudinal direction, which could be explained by some relaxation/contraction of the shoulders during BH. BH reproducibility was in most cases < 1.0 mm in each direction (Figure 1C). No differences in BH durations were found for patients who preferred to perform the BH procedure without audio-coaching (n=4, highlighted in bold in Figure 1A).