Development of a Dosimetric Validation Framework for Clinical Radiopharmaceutical Therapy Treatment Planning Systems
Abstract
Purpose
Accurate patient-specific dosimetry is essential for safety and efficacy in radiopharmaceutical therapy (RPT). Yet standardized approaches for commissioning dosimetry workflows are not well established. This work presents a framework for dosimetric validation of an RPT treatment planning system using organ S-values in a computational reference phantom.
Methods
Raw CT and region of interest volumes of ICRP 110 adult male computational phantom were converted into a DICOM CT and RT structure set. DICOM activity volumes were generated using ROI masks, assigning uniform unit activity to a single source organ and all other voxels zero. Time-integrated activity was calculated and absorbed dose computed using the Monte Carlo dose engine in the Torch® (Voximetry) RPT dosimetry software. Organ S-values were calculated as absorbed dose to a target organ per unit time-integrated activity in a source organ (mGy/MBq-s). Torch S-values were compared to OpenDose reference values for 7 source organs, 55 non-source organs, across Lu-177, I-131, Y-90 (beta-emitters) and Ra-223, Ac-225 (alpha-emitters).
Results
Torch source organ self-dose S-values demonstrated good agreement with OpenDose across most organs and radionuclides. For lungs and soft-tissue organs, agreement was within 5.2% with most values within 1%. Tracheal wall and thoracic spine spongiosa self-dose S-values were within 3.7% for all radionuclides except Ac-225 and Ra-223, for which differences of 10% were observed. Largest differences (17.1%) occurred for cortical bone self-dose. Reasonable agreement within ~20% was seen for most non-source organ S-values across all organs and radionuclides except Ac-225, which was ~30% smaller. Larger relative differences in non-source organs corresponded to negligible absolute dose differences, as they were orders of magnitude lower than source organ dose.
Conclusion
This study presents a framework for validating RPT dosimetry across multiple radionuclides using a full-body computational phantom and reference S-values, enabling standardized commissioning and quality assurance for clinical RPT treatment planning systems.