Parametric mapping of TSPO expression in a longitudinal study of the mouse brain: an image derived input function approach

Objectives: he translocator protein (TSPO) is of great interest as a biomarker for the study of diseases that have a component of neuroinflammation, such as epilepsy, multiple sclerosis, Alzheimer’s disease and Parkinson’s disease. TSPO PET imaging can follow neuroinflammation associated with neurodenenerative disorders using [¹⁸F]-DPA-714, but there are quantification challenges due to the small size of the brain structures in the mouse causing partial volume effects, to the difficulty of arterial sampling in mice, especially in longitudinal studies, and to the absence of a reference region for some pathologies and disease models. The aim of this study was to investigate the use of an image derived input function (IDIF) extracted using a factor analysis (FA) for the estimation of binding parameters and to use the IDIF for the creation of a parametric map of binding parameters across the brain. The model was induced by injection of KA into the right dorsal hippocampus of adult male C57/Bl6 mice (n=4). A dynamic [¹⁸F]DPA-714 PET/CT (60 min) was performed at 2 time points: (i) baseline scan (BL) before and (ii) one month after KA injection (KA1m). One mouse was also scanned at six months post KA injection to show the parametric map and spread of binding (KA6m). FA (PIXIES software, http://www.apteryx.fr/) was applied to all images using 4 factors to extract the IDIF and used in the Logan plot to estimate the total volume of distribution (V_T). The V_T was compared to the estimated binding potential (BP_ND) from the Logan Reference model using the averaged striatum as a reference region which was previously shown to have low TSPO expression compared to other regions in this model. The regional mean and standard deviation were calculated for BP_ND and V_T over all animals for the BL and KA1m scans. The average % coefficient of variation (%CoV=100[asterisk]SD/mean) was calculated for the BP_ND and the V_T for each region over all animals. The extracted IDIFs had a peak that was highly correlated with the injected dose (r²=0.9), and there were strong regional correlations between the IDIF-based Logan V_T estimates and the Logan reference BP_ND estimates for baseline (r²=0.99) and KA1m datasets (r²=0.98). The average %CoV over all regions for the V_T parameter was 15.4% for BL and 15.2 for KA1m, whereas using the reference region, the BP_ND had a %CoV of 163.5% at BL and 141.1% at KA1m. The IDIF was used in a voxel wise Logan plot, and three examples of a V_T map are shown in Figure 1 (the same mouse scanned at BL, KA1m and KA6m). In conclusion, our results suggest that we can quantify TSPO PET in the mouse brain when there is no reference region available by extracting an IDIF using FA. There was a strong correlation between the average regional parameter estimates (V_T vs BP_ND) and using the IDIF yielded much more stable estimates between experiments compared to using a reference region. This is likely due to the variable levels of radiotracer present in brain tissue, even in healthy animals. The IDIF is applicable for use in parametric mapping of TSPO expression. The IDIF method for parameter estimation will be useful for longitudinal studies of neuroinflammation especially in pathologies where the pattern of neuroinflammation is unknown. Figure 1: Parametric maps of TSPO binding in the mouse brain at three points post KA induction. The lesion is clear in the right hippocampus (inversed) at 1m post KA injection and the spread of TSPO expression is clear after 6 months.

• Download figure
• Open in new tab
• Download powerpoint