OA087] IVD methodology in apbi brachytherapy

Abstract

Purpose To establish a methodology for the comparison of the dose measured by dosimeters used in in vivo dosimetry (IVD) of 192Ir brachytherapy (BT) treatments with the dose planned by treatment planning system (TPS) with two different algorithms. Methods IVD was performed on 9 accelerated partial breast irradiation (APBI) patients. The dosimeter used was a micromosfet placed on the skin surface in 3 positions marked with radiopaque markers: a position near the tumor site, and two other positions located at the beginning and end of the treatment tubes. For the calculation of the dose planned by TPS, and taking into account a possible positional error of the dosimeter due to the markers artifact in the planning CT image, we created a volume of interest (VOI) expanded by 2 mm around these radiopaque markers, calculated minimum and maximum doses in that VOI, and compared the dosimeter data within that dose range. Treatment plans were performed using the traditional TG43 method and a commercially model-based dose calculation algorithm (MBDCA) following TG186 recommendations. Results Regarding the planned dose range calculated with TG43, 4 of the total 27 measurements performed fell outside this range. The minimum and maximum deviation found was 2 and 16%. Regarding the TG186 planned dose range, 3 of the 27 measurements fell outside, with a minimum and maximum deviation of 4 and 23%. Conclusions The dosimeter deviation from the planned dose may have been influenced by a variation of its position relative to the radiopaque marker, an important factor in regions of high dose gradients, and by the fact that TPS may not include heterogeneities in its dose calculation, an important factor in the tissue-air barrier. Nevertheless, it happened in a maximum of 4 out of 27 treatments, thus highlighting the good performance of these dosimeters in BT IVD within our methodology of using a planned dose range. As future work, we intend to observe the influence of a bolus material on the skin in the dosimeter behavior and acquire a deeper knowledge of this new algorithm and its applications in a more accurate dose calculation.