Characterization of gamma camera performance at high holmium-166activity for imaging protocol optimization

Objectives: Holmium-166 (¹⁶⁶Ho) is used in therapeutic radioembolization, ideally followed by SPECT imaging for dosimetry purposes, since it also emits low-energy (81 keV) gamma photons. This study assesses the SPECT image quality at high ¹⁶⁶Ho photon count rate. This may affect image quality, dosimetric accuracy and necessary time intervals.

Methods: The liver compartment, including two tumors, of an anthropomorphic phantom was filled with ¹⁶⁶Ho-chloride. Multiple SPECT/CT (Siemens Symbia T16, medium-energy low-penetration collimators) were acquired at a range of different activities up to 2.7 GBq, with a tumors to background activity concentration ratio of 10. Images were reconstructed using a commercially available protocol incorporating attenuation and scatter correction (3D OSEM) and an academic Monte Carlo algorithm (UMCS). To calculate the sensitivity of the gamma camera, the ratio between the number of photons per second recovered in tumor spheres and in a sphere delineated in the background was compared with the theoretical tumors to background activity concentration ratio. Image quality of reconstructions was blindly assessed on a 1 to 5 scale by a nuclear medicine physician (experience’s years > 5).

Results: The acquisition demonstrated paralyzable behavior, with a linear response between counting rate and activity up to approximately 600 MBq. At 1.3 GBq, the system presented a 20% count loss. Recovery coefficients (RCs) had a fairly constant trend over the different activities in both tumors. Sensitivity steadily decreased with increasing activity in both tumors and liver compartments. Based on the qualitative assessment, all clinical and UMCS reconstructions were considered to be of high quality (all images scored either 4 or 5).

Conclusions: It is possible to obtain visually satisfying SPECT reconstructions at high ¹⁶⁶Ho count rate. The recovery coefficients are not affected by the increase in activity. However, the higher the activity the more count loss and hence the underestimation of the reconstructed dose. Based on these results, we suggest a maximum activity of approximately 1 GBq to perform an appropriate ¹⁶⁶Ho SPECT scan. Research Support: Nederlandse Organisatie voor Wetenschappelijk Onderzoek