ESTRO 2022 - Abstract Book

S1478

Abstract book

ESTRO 2022

in the online adaptive workflow. The number of MUs of the online fraction plan was compared to the pre-treatment plan and previous fractions of the patient prior to online treatment delivery. The number of MUs and other plan characteristics were collected for 23 clinical patients for both the pre-treatment as well as the online adapted treatment plans. Pre- treatment QA measurements of the plans were performed using an Octavius 1500MR array (PTW, Freiburg). Results The average number of MUs in the pre-plan was 1485 ± 132 (1 SD). Figure 1 shows the number of monitor units of the pre- treatment plan as well as the treatment fractions for all 23 patients. No statistically significant difference was found between the average number of MUs over all adaptive plans of all patients (1501 ± 130 MU) and the pre-plan average. Inter- fraction variation of the number of MUs was reduced compared to the inter-patient variation (average SD of 62 MUs). For 14 of the 23 patients, this variation was below 3% of the total number of MUs compared to the pre-treatment plan. A larger inter-fraction variation correlated with larger variations in patient anatomy (bladder and rectal filling). No correlation was found between the number of MUs and the number of segments in the treatment plan. Pre-treatment QA measurements of the treatment plans had a pass rate above 98.5% (median 100%), and an average mean gamma of 0.397 (2%/2mm local gamma criterion). No correlation was found between the mean gamma and the number of MUs for the plan.

Conclusion An efficient and effective QA procedure was developed for online adaptive MR-guided treatment of prostate cancer patients. The MU range check can be used in combination with a strict class solution and integrated QA checks of the online adaptive workflow. The inter-patient variation of MUs is limited, and the intrapatient variation between different fractions is further reduced. An MU range check can thus be used for detecting outliers as an alternative to a secondary dose calculation.

PO-1679 Clinical evaluation of a novel system for real time monitoring of deep inspiration breath hold

M. Kusters 1 , E. van Bronkhorst-van denBos 1 , L. Bieamans-VanKastel 1 , P. Westhoff 1 , H. Meijer 1 , J. Barnhoorn 2 , M. Luesink 2

1 Radboud university medical center, Radiation Oncology, Nijmegen, The Netherlands; 2 Cablon Medical, Radiation Oncology, Leusden, The Netherlands Purpose or Objective It is recommended that left breast radiotherapy is performed with deep inspiration breath hold (DIBH), since this results in a significant reduction in radiation dose to the heart and left anterior descending coronary artery. In this study a newly developed system for real-time monitoring of DIBHs has been tested and evaluated during treatment of 27 patients during breast cancer radiation treatment. Materials and Methods Patients with left sided breast cancer are guided to perform a voluntary breath hold using verbal instructions and a device touching the abdomen providing sensible feedback. The number of breath holds that are necessary depends on the delivery technique and can vary between 4 and 10. The system “CNERGY Breath Hold” (Cablon Medical, Leusden, The Netherlands) consists of a single full HD real-time depth camera, installed at the ceiling of the treatment room that measures the respiratory motion of the chest by tracking the surrounding area of the fixed marked tattoo point on the patient’s xiphoid process. During setup this tattoo point is aligned in DIBH at the isocentre of the linac (Elekta, Crawley, UK), where the baseline of the DIBH is defined as 0 mm. When the couch is moved for treatment this translation is also translated in the monitoring system to keep track at the same point on the patient. The system allows RTTs in the control room to monitor the breath holds during treatment. The system will interrupt the treatment automatically when the DIBHs are not performed well and are outside the gating window of -5 to 5 mm. The DIBH data was analysed for reproducibility and stability. For reproducibility the standard deviation of the mean of each DIBH level was calculated. For stability all breath holds were fitted by first order polynomials, the slopes were multiplied by their breath hold lengths to find a range and all these ranges were averaged.