142. DoseFX: A software for dosimetry of organs and lesions in molecular radiotherapy

Abstract Purpose The primary objective was to validate the software DoseFX developed by Comecer for organ level dosimetry. Residence times ( τ ) and absorbed doses (D) to organs were compared to the quantities obtained with OLINDA1.1 and VoxelMed2.0 (voxel/organ-level dosimetry software). Methods and materials SPECT/CT sequential dosimetry scans acquired in a PRRT trial were considered. Every patient was scanned 5 times after a therapeutic administration of 177Lu-DOTATOC. Two different clinical studies were performed. Study (A): the same activity values for the organs were used for the calculation of τ and D with DoseFX and OLINDA1.1 in a sample of 25 patients. Study (B): the same activity values for the organs were used with DoseFX and VoxelMed_Organ in a sample of 20 patients, while VoxelMed_Voxel works directly on images. For this study we did not calculate τ with OLINDA1.1, but we used τ calculated with VoxelMed_Organ. The Wilcoxon test was used (p-values in plots). Results The mean percentage differences (PD) between DoseFX and OLINDA1.1 for the study (A) were ( τ ;D): −1.1;−1.3 liver, 0.1;0.1 kidney and −1.3;−1.3 spleen. For the study (B) mean PD between DoseFX and OLINDA1.1 were ( τ ;D): 5.4;4.2 liver, 0.0;2.7 kidney and 4.8;10.0 spleen; between DoseFX and VoxelMed2.0 were ( τ ;D): −0.7;9.8 liver, 0.6;13.8 kidney and 1.4;17.9 spleen. Data distribution in box plots. The study (A) showed a good agreement between DoseFX and OLINDA1.1 both for τ and for D, with mean PD smaller than 1%. In study (B) larger PDs were observed between DoseFX and VoxelMed, making them significantly different. This is most likely due to the different calculation modality of τ and D. Conclusion The main conclusion is that DoseFX is a tool for organ dosimetry as safe as OLINDA1.1 in PRRT. Particular attention should be paid to the differences in the dosimetry calculation techniques implemented in available tools.