ESTRO 37 Abstract book

S1236

ESTRO 37

Were measured repeatability, reproducibility, half value layer (HVL), homogeneity, output and delivery time by an independent chronometer. To mimic reference conditions presented in IAEA TRS398, for each applicator diameter, the equivalence thickness was designed to be positioned at the exit of the Xray tube (micronode). This accessory will permit an easier way of measurement. Results security tests were conform with the IEC60601, especially concerning treatment interruption. Only results concerning 50kV (energy clinically used) are presented here. Repeatability of results were lower than 1% and reproducibility 1.5%. Applicator diameter (mm) FSD (mm) HVL (mmAl) 30 15 0.65 35 17.5 0.75 40 20 0.85 45 22.5 1 50 25 1 Table1: results obtained for the different applicators If we compare the spherical applicator to a terrestrial globe, the results of homogeneity for a 35mm and a 45mm applicator diameter are respectively : in latitude 7.3% and 6.5%; in longitude 7.4% and 4.7%. Dose rate in reference conditions (30mm diameter applicator) will be settled to 30Gy/min by adjusting mA level. Conclusion Characteristics of this new device are congruent with medical physics expectations. The characteristics of this new device make it possible to deliver intraoperative treatments in senology for a dose of 20Gy in less than 2 minutes. The anaesthesia duration is thus little increased, allowing an optimization of the occupancy time of the operating room and the mobilization of the multidisciplinary caregiver team. Mandatory quality assurance includes a priori risk analysis and risk mapping. Quality control tests of this machine are based on Canadian recommendations for quality assurance. Next development will concern applicators for rectal, skin and vaginal treatments. EP-2236 Sensitivity of dose-volume indices to organ reconstruction settings in HDR prostate brachytherapy M.C. Van der Meer 1 , P.A.N. Bosman 2 , B.R. Pieters 1 , Y. Niatsetski 3 , T. Alderliesten 1 , A. Bel 1 1 Academic Medical Center, Radiation Oncology, Amsterdam, The Netherlands 2 Centrum Wiskunde & Informatica, Life Sciences and Health, Amsterdam, The Netherlands 3 Elekta, Physics and Advanced Development, Veenendaal, The Netherlands Purpose or Objective Dose-volume indices (DVIs) play a decisive role in the evaluation of HDR prostate brachytherapy (BT) treatment plans. Calculation of DVIs is normally based on 2D delineated contours; reconstructed 3D organs may thereby vary depending on reconstruction assumptions. Therefore, the quality of a treatment plan expressed by DVIs may vary with different reconstruction settings. We investigated the sensitivity of DVIs to different organ reconstruction settings, compared to standard settings in a commercially available BT treatment planning system (TPS). Material and Methods Data of 26 prostate cancer patients consecutively treated with HDR BT were included. Contours were delineated on

axial MRI scans (slice thickness: 3.3mm). The studied settings in obtaining a 3D organ from 2D contours were: • In- or excluding the urethra: it can be considered part of the prostate or not. • Smooth or stepwise interpolation: contours drawn on each slice can fill the volume spanned by the slice, or interpolation can be used between contours for a smoother organ surface.

•

Full or partial top/bottom slice thickness: here, contours can span only the half-slice-thickness towards the other contours, or the full volume spanned by the slice.

An in-house stand-alone software was developed and validated with Oncentra Brachy (OB) TPS (Elekta), which we use clinically. The impact of different settings was compared to a baseline following the default settings in OB, where the urethra is included in the prostate, interpolation is used, and the last contours span half-a- slice thickness. Studied DVIs are given in the table. The DVI calculation used a number of sample points (256.000 per organ for targets, 2.560 per cm 3 for OARs) resulting in a small variation in the DVIs; in the baseline setting, averaged over all patients, the 95% confidence interval was at most ±0.2% for targets and ±0.5% for OARs.

Results DVIs of rectum and urethra were relatively insensitive to the reconstruction settings (range of [-0.1, 1.3]%). The D 90 and V 150 of the prostate increased when the urethra was excluded from the prostate, with medians of 1.0% respectively 1.2%. DVIs of the bladder and seminal vesicles changed up to 2.5% without interpolation, and up to 8.3% respectively 9.5% when the full slice thickness was used for the first and last contour, resulting in a decrease for the vesicles and an increase for the bladder (see figure). For the vesicles, this can be explained by the small average volume of 4.08cm 3 . For the bladder, the region of interest of the most irradiated 1cm 3 or 2cm 3 was often near the most caudally delineated contour.