Impact of morphological changes on small animal dosimetry in nuclear medicine

Molecular radiotherapy consists in targeting radioactive atoms as close as possible to tumours in order to concentrate energy deposition (and cytotoxic effects) at the level of cancer cells. Preclinical experiments including absorbed dose calculation allow the adjustment of injected radiopharmaceutical activity in order to optimise the therapeutic window. Even though Monte-Carlo modelling combined with hybrid anatomical models allows the accurate and realistic description of energy deposits, inter-specimen morphologic variations can have a marked impact on computed absorbed dose. The work presented in this manuscript investigates the dosimetric impact of using such preclinical models in nuclear medicine. A characterisation of MCNPX and GATE was first performed, based on the generation of reference dosimetric data (Dose Point Kernel and Voxel Dose Kernel). Organ S-value calculations for two realistic digital rodent whole-body models (Moby/Roby) showed that mass ratio scaling for two geometric models obtained from the same specimen led to significantly different dosimetric results, raising the issue of the reproducibility of generated dosimetric results. Besides, dosimetric calculation based on one specimen that can be used as a reference for the entire experimental batch may not allow full and accurate dosimetric assessment. Our conclusion is that even though it may be possible to use reference models to compare dosimetric results for diagnostic radiopharmaceutical experiments, animal-specific anatomy should be considered when trying to characterise ionizing radiation effects on biologic targets of interest for MRT. (author)