The primary aim of this study was to demonstrate that small-animal SPECT/CT imaging is capable of measuring DAT density. We validated a DAT-binding imaging protocol for the mouse brain, using 123I-β-CIT. To our knowledge, the present study is the first to demonstrate the 3D kinetics of 123I-β-CIT in mice, using SPECT/CT imaging.
The SPECT/CT used was equipped with four collimators and apertures, each with nine 1-mm pinholes, resulting in high counting rates. The resolution for the camera was in the submillimeter range; Jaszczak phantom imaging showed that we can reliably image the mouse striatum.
From the template, we obtained a striatal volume of approximately 22.3 mm3 with a threshold of 50% of the maximum intensity of the striatum. Similar striatal volumes have been reported in equivalent strains
. At the same threshold, we obtained a striatal volume of 17.2 mm3 ± 20%. In Figure
3, the striatal volume is shown as a function of thresholds in individuals and the template. The striatal volume increases until it becomes stable at approximately 50% to 60% of the maximum activity. The phenomenon was found in both cases (Figure
3). For threshold values higher than 60%, the volume spreads outside the striatal volumes, into other parts of the brain. The higher striatal volume in the template is due to individual variation, coregistration, and averaging. This comparison indicates that the template represents data in healthy mice well and that no coregistration or technical problems occurred. Further, we compared the templates created from the first and second imaging of the test-retest study and found no significant difference. This result demonstrates that our method has good repeatability. In future works, we will address individual variation in striatal structures and minimize the processing steps needed using the template.
In the present study, we report the kinetics of the 123I-β-CIT tracer in the striatum and cerebellum. As indicated earlier, the quantitative validity of the BP method (striatal to background brain region ratio) depends on the achievement of steady levels of activity in these regions
. As shown in Figure
4b, 3 h postinjection, there is parallel washout in cerebellum and striatal volumes. Thus, BP analysis can be considered valid 4 h post-123I-β-CIT injection in the mouse brain. These results agree with the earlier autoradiography works
[7–9]. However, our results show moderate increase in BP, which is evident at the 7-h imaging point. This might be due to the individual kinetics of mice and also because of low statistics in the cerebellum volume. To resolve this, further kinetic studies should be performed at late imaging points.
We used BP to perform semiquantitative analysis of dopaminergic neurotransmission post-123I-β-CIT injection. BP increased over the time period followed, similar to previous rat studies with SPECT
[11, 12] and autoradiography work with both rats
 and mice
[7–9]. In healthy human subjects, the evolvement of BP is similar, differing only by time scale
[23, 24], due to the slower metabolism of humans compared with mice. In the present work, 4 h postinjection was a suitable time point for use in the test-retest study, which is equivalent to 20 to 24 h in human patients
In the test-retest study, one of the five animals showed a clearly larger variation among BPs, which probably resulted from unsuccessful injection at either one of the time points. The variability between test and retest was clearly better than that in healthy human subjects
[25, 26], which shows that our method has good test-retest reproducibility. Such improvement was expected, due to the better sensitivity and resolution of the small-animal SPECT instrument compared with clinical systems.
Successful validation of the imaging protocol and establishment of the template will help to reduce the number of control mice needed for the experiments and will further reduce the overall costs and time needed. The template created here will be further used as a reference to differentiate between normal and genetically modified mouse DAT densities. Also, we are planning to extend the template to include serotonin transporters. In present and future works, the template will offer a reference for coregistration and will take into account the variance between individuals in the VOI analysis. Furthermore, correct and repeatable delineation of VOIs is needed to gain enough statistical power because the targets are submillimeter in size. Herein, we focused on certain VOIs, but the template may also be used in voxel-by-voxel analysis.