TY - JOUR T1 - Experimental validation of Monte Carlo-generated beta absorbed doses for 3D voxelwise dosimetry JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 533 LP - 533 VL - 61 IS - supplement 1 AU - Ashok Tiwari AU - Stephen Graves AU - Sarah Strand AU - John Sunderland Y1 - 2020/05/01 UR - http://jnm.snmjournals.org/content/61/supplement_1/533.abstract N2 - 533Objectives: The critical need for accurate determination of dose from beta particles in both normal tissue and tumors in the setting of targeted radiopharmaceutical therapy has motivated the development of image-based voxelwise dosimetric techniques that, at their core, depend upon Monte Carlo simulation of energy deposition. The fundamental weakness of this approach is that, to date, Monte Carlo-based dose calculations have never been empirically validated in a controlled laboratory setting due to the fundamental difficulty of measuring spatial dose deposition for therapeutic beta emitters (i.e. 90Y and 177Lu) with a very high spatial resolution. The goal of this work was to develop and test an experimental methodology for high-resolution beta dosimetry and to compare results from this method with Monte Carlo simulations performed in the same geometry. Methods: 90Y and 177Lu line sources (plastic capillary tube of length 13 cm, 0.8 mm outer diameter, 0.4 mm inner diameter) were inserted longitudinally through blocks of low-density polyethylene and cortical bone and lung tissue-equivalent plastic (CIRS, Norfolk, VA). Radiochromic film EBT3 was laser-cut to provide a 0.8 mm diameter hole to accommodate orthogonal line-sources of radioactivity with the same outer diameter. The film was sandwiched tightly between the rectangular slabs with identical 0.8mm holes to achieve scatter equilibrium in the plane of the film. Final activity densities in the line sources were (0.34 ± 0.015) MBq/cm and (0.35 ± 0.014) MBq/cm for 90Y and 177Lu, respectively. Film exposures were conducted for durations ranging from 10 minutes to 38 hours to achieve exposures within the linear range of the EBT3 film (0.1-10 Gy). All experimental geometries were precisely simulated within the GATE Monte Carlo toolkit, which has previously been used to produce dose point kernels for 90Y and 177Lu. Radiochromic film calibration was achieved by irradiation with 6 MV bremsstrahlung x-rays from a calibrated linear accelerator (Siemens Oncor), in accordance with literature recommendations [1]. Results: Experimentally measured doses agreed closely with Monte Carlo calculations. For 90Y and 177Lu absorbed doses, the mean local percentage difference in the dose distribution compared to Monte Carlo simulation was ~6.0%. Discrepancies were highest close to the line source, due to the delamination of the film during the laser-cutting process. 1D gamma analysis [2] was performed using the 10%/1 mm gamma criterion to compare the simulated and measured dose distributions. The percentage of measurement points meeting the gamma criterion, averaged over all tests, was 93.5%. Conclusions: We report the first high-resolution experimental validation of Monte Carlo-simulated dose deposition from beta particles. Excellent agreement was observed between the experimental beta absorbed doses compared with GATE Monte Carlo simulations for 90Y and 177Lu, suggesting that dosimetry based on beta dose kernels derived from GATE Monte Carlo simulations are sufficiently accurate for clinical use. Future work should expand these methods to other therapeutic radionuclides and measurement geometries. ER -