Low-dose 90Y PET/CT imaging optimized for lesion detectability and quantitative accuracy: a phantom study to assess the feasibility of pretherapy imaging to plan the therapeutic dose.

Objective: The overall aim of this work is to optimize the reconstruction parameters for low-dose yttrium-90 (Y) PET/CT imaging, and to determine Y minimum detectable activity, in an endeavor to investigate the feasibility of performing low-dose Y imaging in-vivo to plan the therapeutic dose in radioembolization.

Materials And Methods: This study was carried out using a Siemens Biograph 6 True Point PET/CT scanner. A Jaszczak phantom containing five hot syringes was imaged serially over 15 days. For 128 reconstruction parameters/algorithms, detectability performance and quantitative accuracy were evaluated using the contrast-to-noise ratio and the recovery coefficient, respectively.

Results: For activity concentrations greater than 2.5 MBq/ml, the linearity of the scanner was confirmed while the corresponding relative error was below 10%. Reconstructions with smaller numbers of iterations and smoother filters led to higher detectability performance, irrespective of the activity concentration and lesion size. In this study, the minimum detectable activity was found to be 3.28±10% MBq/ml using the optimized reconstruction parameters. Although the recovered activities were generally underestimated, for lesions with activity concentration greater than 4 MBq/ml, the amount of underestimation is limited to -15% for optimized reconstructions.

Conclusion: Y PET/CT imaging, even with a low activity concentration, is feasible for depicting the distribution of Y implanted microspheres using optimized reconstruction parameters. As such, in-vivo PET/CT imaging of low-dose Y in the pretherapeutic stage may be feasible and fruitful to optimally plan the therapeutic activity delivered to patients undergoing radioembolization.