TSPO imaging using the novel PET ligand [ 18 F]GE-180: quantification approaches in patients with multiple sclerosis

Abstract

Background

PET ligands targeting the translocator protein (TSPO) represent promising tools to visualise neuroinflammation. Here, we analysed parameters obtained in dynamic and static PET images using the novel TSPO ligand [¹⁸F]GE-180 in patients with relapsing remitting multiple sclerosis (RRMS) and an approach for semi-quantitative assessment of this disease in clinical routine.

Seventeen dynamic [¹⁸F]GE-180 PET scans of RRMS patients were evaluated (90 min). A pseudo-reference region (PRR) was defined after identification of the least disease-affected brain area by voxel-based comparison with six healthy controls (HC) and upon exclusion of voxels suspected of being affected in static 60–90 min p.i. images. Standardised uptake value ratios (SUVR) obtained from static images normalised to PRR were correlated to the distribution volume ratios (DVR) derived from dynamic data with Logan reference tissue model.

Results

Group comparison with HC revealed white matter and thalamus as most affected regions. Fewest differences were found in grey matter, and normalisation to frontal cortex (FC) yielded the greatest reduction in variability of healthy grey and white matter. Hence, FC corrected for affected voxels was chosen as PRR, leading to time-activity curves of FC which were congruent to HC data (SUV_60–90 0.37, U test P = 0.42). SUVR showed a very strong correlation with DVR (Pearson ρ > 0.9). Focal MS lesions exhibited a high SUVR (range, 1.3–3.2).

Conclusions

This comparison with parameters from dynamic data suggests that SUVR normalised to corrected frontal cortex as PRR is suitable for the quantification of [¹⁸F]GE-180 uptake in lesions and different brain regions of RRMS patients. This efficient diagnostic protocol based on static [¹⁸F]GE-180 PET scans acquired 60–90 min p.i. allows the semi-quantitative assessment of neuroinflammation in RRMS patients in clinical routine.

http://ift.tt/2mBsCnL