ESTRO 2022 - Abstract Book

S95

Abstract book

ESTRO 2022

Department of Oncology, Pamplona, Spain; 21 Gold Coast University Hospital, ICON Cancer Centre, Gold Coast, Australia; 22 Institut Claudius Regaud—Institut Universitaire du Cancer de Toulouse, Department of Medical Physics, Toulouse, France; 23 Hospital Sant Joan de Reus, Department of Medical Physics, Reus, Spain Purpose or Objective Careful modelling of the MLC is crucial for IMRT/VMAT. It has been shown that TPS modelling errors were present in most centres failing end-to-end dosimetry audit [1] and a large centre-to-centre variability in MLC TPS parameters has been reported [2]. The goals of this study are to evaluate the feasibility of using a common set of tests to evaluate the MLC models implemented in TPSs and a common characterization of MLCs across vendors. Materials and Methods 24 centers worldwide were invited to participate in this pilot study. DICOM plans or scripts for the tests were supplied. The set of tests included reference, transmission, sweeping gap (SG) and asynchronous sweeping gap (aSG) fields for the 10 and 20 mm gaps. SG fields are sensitive to the transmission and rounded-leaf-end of the MLC [3,4], while aSG are sensitve to the tongue and groove (TG) width and the leaf tip-TG region [5,6]. In the aSG tests the gap is constant but adjacent leaves are shifted different distances ‘s’ to expose the leaf sides, causing a dose reduction as ‘s’ increases. All centers used a Farmer-type ionization chamber with its long axis perpendicular to leaf movement with an isocentric set-up in water or solid water (SSD 90 cm, depth 10 cm) with a 6 MV photon beam. The DICOM set included images of a water phantom and an evaluation structure mimicking the ion chamber. Centers reported the average dose per field to the evaluation structure calculated by their TPS. Both measured and calculated doses were normalized to the reference field dose for comparison within the same MLC type.

Results A summary of MLC and TPSs evaluated is given in Fig1. Normalized doses for the aSG fields for 2 representative MLC types are shown in Fig2. For each type: • Measured normalized dose curves were nearly identical. The standard deviation with respect to the mean were at most 1% (Agility), 1% (HD120) 1.2% (Millennium) and 0.4% (Halcyon). The slope and shape of the dose reduction caused by the TG depended on the type. These results reflect that these tests are sensitive to the dosimetric details of each type. • Calculated dose curves for TPSs showed large differences with respect to measured doses for all MLCs except for the Halcyon: 8% (Agility), 8% (HD120) and 3.5% (Millennium).