RT Journal Article SR Electronic T1 Micro vs Clinical Scale Dosimetry of Alpha Particle Targeted Radionuclide Therapy – Comparison of Absorbed Dose and DNA Damage JF Journal of Nuclear Medicine JO J Nucl Med FD Society of Nuclear Medicine SP 242227 OP 242227 VO 65 IS supplement 2 A1 Ghaseminejad, Shahin A1 De Sarno, Danny A1 Duong, Thanh-Tai A1 Opara, Chidera A1 Bertolet, Alejandro A1 Lee, Ting-Yim YR 2024 UL http://jnm.snmjournals.org/content/65/supplement_2/242227.abstract AB 242227 Introduction: Targeted radionuclide therapy (TRT) has demonstrated notable success in treating neuroendocrine and prostate cancer. The current choice of radionuclide, 177Lu, primarily emits beta particles. Alpha emitters like 225Ac are expected to be more effective due to their higher linear energy transfer (LET) and extremely short range (~80 μm), allowing for more precise targeting of cancer cells while minimizing damage to surrounding healthy tissue. This study investigated how microscale 225Ac dosimetry and its DNA breaks can inform clinical dosimetry based on centimeter-scale SPECT imagingMethods: Three 3D cell models (A, B and C) were constructed with 2mm voxel resolution. Models had cubic cells and nuclei with specifications as follows: model A had 6mm nuclei in 10mmcells; models B and C had 4mm and 8mm nuclei respectively in 20mm cells. Each model had 4 versions in which, cells were placed at different center-to-center distances which decreased from version 1 to 4 corresponding to cell fractions of 0.1, 0.22, 0.34 and 0.56 for model A, and 0.1, 0.27, 0.43, 0.77 for models B and C. Number of cells was increased as we decreased cells distances to keep the model volumes roughly the same for all models/versions. Activity was uniformly distributed in the cytoplasm simulating radiopharmaceuticals that have been internalized into the cell. Model dimensions were chosen to ensure that a central cluster of 3×3×3 cells experienced the full cross-fire effect from the neighboring cells based on the 80mm range of alpha particles from the 225Ac chain. Each model had the same total time-integrated activity (TIA) and its 3D distribution was convolved with the 2mm-scale 225Ac chain dose-point kernel (DPK) to determine the micro-scale 3D dose distribution. The DPK was generated with TOPAS, a Geant4 wrapper, using Geant4-DNA physics lists option 2. The heterogeneous micro-scale TIA was then converted into a uniform distribution with the same total TIA and convolved with the 225Ac DPK to simulate the calculated absorbed dose distribution from SPECT-measured TIA according to the MIRD-S value formalism. After acquiring the absorbed dose to each cell for all models/versions, TOPAS n-Bio () was used to calculate the number of single-strand breaks (SSB) and double-strand breaks (DSB) in the DNA for each model/versions, and also for the uniform SPECT clinical dose using Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm.Results: Figure1 shows dose profiles normalized to the same total TIA through the central cluster of 3 cells. Normalized TIA corresponds to administrating 100 kBq/kg of 225Ac at the Standardized Uptake Value (SUV) of 5.0 g/mL. As expected, the micro-scale absorbed dose was highly heterogeneous. For model A, peak and trough ratios decreased from 1.5 to 1.1 and cell doses from 18.5 to 11.9 Gy as cell fraction increased. These ratios decreased from 2.8 to 1.2 for model B and from 2.58 to 1.16 for model C and cell doses from 8.4 to 3.6 and 8.0 to 3.6 Gy respectively. Figure2 shows DNA breaks for each model with 225Ac sources randomly distributed inside the cytoplasm to deliver the same cell dose as the convolution dose for each model. In model A, mean SSB decreased from 160 to 103 and 18 to 13 for mean DSB as cell fraction increased. Model B showed a decrease in mean SSB from 295 to 129 and mean DSB from 43 to 19. Mean SSB and DSB also decreased from 303 to 137 and 39 to 18 respectively for Model C. The clinical uniform dose obtained from SPECT however showed lower mean SSBs (105) and mean DSBs (14).Conclusions: Microscale models reveal that clinical dosimetry, reliant on TIA distribution measured through SPECT, may significantly underestimate cellular doses and also number of DNA breaks by several folds. The extent of this underestimation is tied to a crucial factor—the cellular fraction. This finding is of relevance to TRT because as cycles of treatment progress, the cellular fraction would decrease if the treatment is effective.