The first article in our new Recent Highlights Series discusses the recent article “Beyond 18F-FDG: characterization of PET/CT and PET/MR scanners for a comprehensive set of positron emitters of growing application – 18F, 11C, 89Zr, 124I, 68Ga, and 90Y” by Soderlund et al. published in the Journal of Nuclear Medicine (doi:10.2967/jnumed.115.156711). The purpose of the study was to investigate the image quality obtained on two clinical PET/CT and PET/MR scanners using a comprehensive set of isotopes: 18F, 11C, 89Zr, 124I, 68Ga, and 90Y.
As PET technology has grown over the last 30 years the number of isotopes used routinely has also increased. Although the initial clinical PET focused predominantly on 11C, 13N, and 15O, recent developments in fields such as immunotherapy, radioimmunotherapy, monoclonal antibody-based therapies, and radionuclide therapy have led to the exploitation of a variety of isotopes owing to their differing characteristics. In spite of the increasing use of these different isotopes, the authors state that few data are available on the image quality and performance achieved with non-18F imaging. Therefore, they analyzed the image quality and special resolution achieved using these isotopes on a Siemens Biograph mCT PET/CT and a Biograph mMR PET/MR scanner. Both scanners had LSO crystals couple to photo multiplier tubes (mCT) or avalanche photo diodes (mMR). The quality of the images was determined using a NEMA IEC body phantom consisting of six fillable spheres of differing diameters. The largest spheres were filled with water (cold lesions) and the others were filled with radioactivity (hot lesions). The full experimental methods can be found in the manuscript.
The key results reported by the authors are as follows. Fewer counts were acquired for 90Y and there was a higher proportion of Bremsstrahlung. Therefore, the quality of the reconstructed data was poorer that that obtained with the other isotopes. Except for 90Y (and to a lesser extent 124I) there were few differences in the contrast recovery performance of the different isotopes. For the smaller hot spheres, the contrast recovery depended generally on the positron range of the isotope: 18F had the best performance whereas 124I and 68Ga had the poorest performance. The results obtained with cold spheres were similar. When FWHM was compared across isotopes, the positron range had little effect on the measured special resolution, except that the axial spatial resolution at 1–10 cm was significantly higher for 124I and 68Ga using the mMR.
When using isotopes for PET imaging the image resolution and isotope performance are critical. As the technology continues to advance beyond using 18F it is increasingly important to assess the performance of common PET scanners against alternative isotopes. The data in the current study revealed that different isotopes achieved different special resolution performance and image quality. Importantly, the authors concluded that all of the isotopes investigated, except for 90Y, performed in a manner similar to 18F. However, the residual bias, contrast recovery, high Bremsstrahlung and background variability were inferior for 90Y compared with the other isotopes.
It is important to consider what these observations mean for the “big picture” use of these isotopes during PET imaging clinically. The two PET systems used in the current study achieve a higher spatial resolution than the whole body PET systems used in many hospitals, at least in the mean positron ranges of the isotopes used. Previous studies with high-resolution animal PET imaging systems revealed that longer positron range isotopes such as 124I and 68Ga resulted in decreased FWHM-to-FWTM ratios. Although, this was not observed in the current study, presumably because of the reduced spatial resolution of the clinical systems used, this is not likely to be a concern for human imaging. Taken together these observations suggest that the isotopes elicited comparable imaging performance as the “gold standard” 18F. High-energy positron emitters such as 124I and 68Ga might have a broader axial spatial resolution in the magnetic field in the PET/MR scanner, which should be taken into account during clinical practice.
We would be interested to hear your thoughts on this study. Should it be repeated using different scanners? Was it necessary? Does it advance the field? Will any results change the choices you make in your practice?