Abstract
1577
Objectives: Validation and commissioning of dosimetry software, such as Monte Carlo (MC) radiation transport codes used for patient-specific dose estimation, is critical prior to their use in clinical decision making. However, direct experimental validation is generally not performed for low/medium-energy beta emitters due to the challenges of measuring energy deposited by short-range particles. Exploiting ready access to 3D printing we designed and tested a reproducible geometry for radiochromic film (RF) based absorbed dose measurements of radionuclides that have a beta range sufficiently penetrating to ~200 microns in water. Measurements were used to validate absorbed dose estimates from our previously in-house developed Dose Planning Method (DPM) MC code dedicated to internal dosimetry as well as the general purpose MCNP6 code.
Methods: Assayed 177Lu and 131I liquid sources (75 - 450 MBq) were injected into pairs of 3D printed half-cylinders sealed with 12.7µm - 25.4µm thick Kapton Tape. Phantoms were printed on a Form2 Labs SLA 3D printer with clear resin and fits in a standard urine specimen container. Activity concentration was estimated to deliver 2-5 Gy to the sensitive layer of the RF. A 5x3.5cm strip of GafChromic EBT3 RF was secured between the 2 half-cylinders. After a 20-26 hour exposure, films were retrieved and wipe tested for contamination. Absorbed dose to the RF was measured using triple channel dosimetry and a calibration generated via 6MV beam. Profiles were analyzed across the central 1cm2 area of the RF for validation. Quantitative SPECT/CT of the cylinders using an xSPECT-Quant system was also performed as an additional measurement of the activity. The experiment with 177Lu was repeated 5 times to assess reproducibility, while that with 131I was performed only once due to radiation safety logistics. Voxelized MC absorbed dose estimates were generated by DPM and MCNP and compared against the RF measurements. MC calculations were scaled based on total number of decays. For MC modeling, the thickness of base and sensitive layer of RF were verified on adjacent RF samples that were delaminated and measured with calibrated micrometer. The Kapton tape thickness was verified in a similar manner, inclusive of the silicone adhesive layer. Material density and compositions for MC were taken from Pacific Northwest National Laboratory publications. The beta spectrums as well as the x-ray and gamma ray yields were provided by Beta Shape, Brookhaven National Lab, and ICRP07.
Results: The agreement between the dose estimates from the RF measurements and DPM was 4%, 8%, 4%, -5%, and -6% for each 177Lu experiment respectively and 2% for 131I. 177Lu experiments measured with 25.4um thick Kapton tape averaged 5% less dose to the RF and experiments measured with 12.7um Kapton tape averaged 5% more dose to the RF than predicted. Dose estimates from the RF measurements and MCNP was within 12%. Currently, our single 131I experiment showed 2% less dose to the RF than predicted by DPM. MC absorbed dose from betas was estimated to be greater than 92% and 67% of the total (betas + others) for 177Lu and 131I respectively, indicating dominant beta contribution with our design. The simulated dose estimates using total activity in the samples measured from the dose calibrator vs. xSPECT agreed to within 2% for 177Lu.
Conclusions: With agreement within 8% for 177Lu and 2% for 131I between measurement and simulation we demonstrate direct validation of absorbed dose calculations with DPM for medium to low energy beta emitters. The reproducible results with a simple 3D printed phantom designed for liquid sources demonstrate that this is a reliable and practical setup for experimentally validating dose calculation algorithms including beta emitters not investigated in this study. Research Support: R01CA240706 awarded by NCI and R01EB022075 awarded by NIBIB, National Institute of Health.
