Imaging Capabilities of the IRIS and IRIS XL-260 PET/CT Systems for High-Throughput Imaging: A Quantification Cross-Validation Study

The concept of imaging several subjects simultaneously is an active preclinical research topic. In this article, we assessed the imaging capabilities of two positron emission tomograph (PET)/computed tomograph (CT) systems based on similar detector technology. The IRIS system is a mice/rat imaging system arranged in two rings of eight detectors each, the more recent IRIS XL 260, dedicated to nonhuman primate (NHP) imaging has a single ring of 16 detectors. Both systems were equipped with standard animal cells (Minerve) and 3-mice adapters. Our objective is to study which of these systems could be more appropriate to perform quantitative high-throughput imaging on mice. Phantoms and [18F]FDG tumor-bearing mice acquisitions have been conducted in single- and high-throughput modes using both the IRIS (up to 3 mice) and the IRIS XL-260 PET/CT (up to 6 mice) systems. Image quality phantom results obtained in high-throughput mode show some slight degradation of the recovery coefficient for rods of 1, 2, and 3 mm in diameter compared to the single-mode results, as one would normally expect. Similarly, we observed a decrease in image uniformity between the single- and the high-throughput modes for both the IRIS PET and the IRIS XL-260 PET systems. We performed [18F]FDG tumor-bearing mice PET acquisitions with both systems. In order to estimate the quantification differences in all the studied configurations, we calculate the ratios between the %ID/g values extracted from the PET images and the ex-vivo values. In single-mode acquisitions, ratios of 0.94 ± 0.09 and 0.83 ± 0.08 were obtained for the IRIS and the IRIS XL-260 PET systems, respectively. In high-throughput mode, ratios of 0.78 ± 0.12 and 0.73 ± 0.13 were obtained for the IRIS and the IRIS XL-260 PET systems, respectively. The difference in %ID/g between static acquisitions and ex-vivo value is not statistically significant ( $p$ -value > 0.1) for both PET systems in single mode, as well as for the IRIS system in the high-throughput mode. We noted a slightly higher statistical difference between static acquisitions and ex-vivo values for the IRIS XL-260 system with a $p$ -value of 0.015. Phantoms and in-vivo studies have made it possible to highlight the capability of the two systems to perform high-throughput acquisitions. Our results suggest that the IRIS system configuration may be the most suitable when aiming for quantitative high-throughput mice imaging. The large transverse field of view of the IRIS XL-260 makes it possible to image a greater number of mice simultaneously, which may be useful in specific cases, such as studies using an expensive radiotracer and/or with a short half-life. However, the IRIS PET/CT offers a full axial coverage of the animals and a higher sensitivity.