ESTRO 36 Abstract Book

S433 ESTRO 36 _______________________________________________________________________________________________

Purpose or Objective Current clinical portal dosimetry from Varian evaluates dose using calibrated units (CU). This work assesses the quality of the first Varian implementation of the MAASTRO algorithm for pre-treatment absolute portal dosimetry (in Gy) of 6X-FF fields. Material and Methods To achieve the proposed goal, a comparison was made between the gamma analysis results obtained using both Varian’s clinical portal dosimetry (PDIP v10.0.28) and the MAASTRO algorithm [1] implementation made available for the authors in the Portal Dosimetry (PD) application accessible through the Varian Citrix Research Environment (CRE). For this study, 10 breast IMRT breast plans and 10 VMAT prostate plans were chosen from the patients’ database. In total, 71 IMRT fields and 40 VMAT arcs were compiled for analysis. Each plan was recalculated with gantry zero on a water-equivalent slab phantom, for later comparison of absolute dose at 5cm depth. Verification plans were created for irradiation with 6X-FF beams at the Varian Edge LINAC in order to measure the doses at the Electronic Portal Imaging Device (EPID) level. For each field/arc, the measured doses and the calculated doses were compared by gamma analysis in CU for PDIP and in absolute dose values (Gy) for the PD system on the CRE. [1] Nijsten SM et al, 'A global calibration model for a-Si EPIDs used for transit dosimetry”, Med. Phys. 34(10): 3872-84, 2007 Results Table 1 presents the summary of the gamma analysis results obtained in the comparison between the measured dose at the EPID and the calculated dose using the MAASTRO algorithm implementation and PDIP. The results show that the analyzed IMRT plans using the MAASTRO algorithm obtained, on average, a higher gamma pass rate, lower mean gamma values and lower dose differences than while using PDIP. The same is observed for VMAT plans. Figure 1 shows the graphical comparison between the gamma passing rate obtained using the MAASTRO algorithm and PDIP, where the black circles represent the comparison of the gamma passing rates for IMRT plans (averaged over all beams) and the open triangles represent the comparison of the gamma passing rates for VMAT plans (averaged over all arcs). One can see that the gamma pass rate obtained using the MAASTRO algorithm is consistently higher than the one obtained using PDIP.

Figure 1. 3D printed patient-specific head phantom filled with dosimetric gel during the treatment planning process. Results Results from mono-energetic irradiation of the cubic phantoms showed proton range agreement to the TPS within 1 mm for 90 MeV and 115 MeV, supporting the SPR gel characterization accuracy. Dose-response linearity was confirmed for the delivered dose range, except at the Bragg peak position where a LET dependence was revealed. Gamma index and relative dose distribution profiles showed good agreement between TPS and gel, as shown in in Figure 2.

Figure 2. (A) Slice of the 3D SFUD dose distribution converted from a T2 relaxation map obtained from MR scanning the irradiated 3D printed head phantom filled with polymer gel. The PTV is indicated in white. (B) Gel (RTsafe) and TPS dose profiles along the path marked in red in (A). (C) 3%/2mm gamma index along the profile. Conclusion In this work we have shown that patient-specific 3D polymer gel dosimetry is applicable to PT using PBS. Further characterization and correction of the LET dependence and comparison to MC dose calculations will PO-0810 Absolute dose pre-treatment Portal Dosimetry using the Varian MAASTRO implementation A. Taborda 1 , J. Stroom 1 , C. Baltes 2 , A. Seabra 1 , K. Dikaiou 2 , C. Greco 1 1 Champalimaud Centre for the Unknown, Clinical Department, Lisboa, Portugal 2 Varian Medical Systems, Varian Medical Systems Imaging Laboratory, Baden-Dättwil, Switzerland be carried out and presented. Acknowledgements: DFG-MAP

Table

1