Technical notes Quantitative comparison between the commercial software STRATOS ® by Philips and a homemade software for voxel-dosimetry in radiopeptide therapy E. Grassi a, * , F. Fioroni a , V. Ferri a , E. Mezzenga a , M.A. Sarti a , T. Paulus c , N. Lanconelli d , A. Filice b , A. Versari b , M. Iori a a Medical Physics Dept., IRCCS e S. Maria Nuova Hospital, Viale Risorgimento 80, I-42123 Reggio Emilia, Italy b Nuclear Medicine Dept., IRCCS e S. Maria Nuova Hospital, Viale Risorgimento 80, I-42123 Reggio Emilia, Italy c Philips Technologie GmbH Innovative Technologies, Pauwelsstr.17, 52074 Aachen, Germany d Dept. of Physics, University of Bologna, I-40126 Bologna, Italy article info Article history: Received 10 December 2013 Received in revised form 2 October 2014 Accepted 5 October 2014 Available online 25 October 2014 Keywords: Dosimetry Voxel Dose distribution DVH Radionuclide therapy abstract Background: Targeted radionuclide therapy is a rapidly growing modality. A few commercial treatment planning systems are entering the market. However, some in-house systems are currently developed for a more flexible and customized dosimetry calculation at voxel-level. For this purpose, we developed a novel software, VoxelMed, and performed a comparison with the software STRATOS. Methods: The validation of both of them was undertaken using radioactive phantoms with different volume inserts. A cohort of 10 patients was also studied after a therapeutic administration of 177 Lu- labelled radiopeptides. The activity, number of disintegrations, absorbed dose and dose-volume histo- gram (DVH) were calculated for the phantoms and the kidneys in patients, which were the main critical organs at risk in this study. Results: In phantoms the absorbed doses computed with VoxelMed and STRATOS agree within 5%. In patients at the voxel-level the absorbed dose to kidneys (VoxelMed: mean 0.66 Gy/GBq) showed a limited difference of 5%, but with a remarkable range (40%, þ60%) between the two software packages. Voxel-dosimetry allows to estimate the dose non-homogeneities in volumes, which may be evaluated through DVHs. Conclusion: This study demonstrates that a fully 3D voxel-dosimetry with multiple SPECT images is feasible by using home-made or commercial software package and absorbed dose results obtained are similar. The main difference between the studied tools was observed in the activity integration method (effective vs physical half-time to time activity curve tail). We believe that an effective half-time inte- gration method produces a more accurate approximation of clinical uptake and resultant dosimetry. © 2014 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved. Introduction In the last decade, peptide receptor radionuclide therapy (PRRT) with somatostatin analogues has been increasingly used for the treatment of metastasized neuroendocrine tumours. On the other hand patient-specific dosimetry can provide information to assess both organ-specific and tumour absorbed doses, and to avoid healthy organ toxicity [1]. An accurate dosimetry is so necessary to understand the radiobiological considerations that affect treatment response [2]. Different dosimetric methods can be applied. They are generally based on the MIRD (Medical Internal Radiation Dose) indications. The MIRD scheme [3] may be employed through either hybrid (3D SPECT plus serial planar) or fully 3D SPECT-CT-based dosimetry. Organ and equivalent dose values may be then calculated with MIRDose or the OLINDA/EXM software (Vanderbilt University, Nashville, USA). MIRD no. 23 [4] describes a voxel-level approach, in which the absorbed dose is computed at voxel-level, accounting for the non- uniformity at a maximum level of detail accessible in vivo. This produces dose-volume-histograms (DVHs) and radiobiological pa- rameters of great interest for radionuclide therapy. The voxel-based methods are not widely used in clinical practice and only two commercial software packages are available: * Corresponding author. Tel.: þ39 0522 296937; fax: þ39 0522 296392. E-mail address: elisa.grassi@asmn.re.it (E. Grassi). Contents lists available at ScienceDirect Physica Medica journal homepage: http://www.physicamedica.com http://dx.doi.org/10.1016/j.ejmp.2014.10.002 1120-1797/© 2014 Associazione Italiana di Fisica Medica. Published by Elsevier Ltd. All rights reserved. Physica Medica 31 (2015) 72e79