Initial evaluation of [18F]GE-180 PET imaging in relapsing-remitting multiple sclerosis patients.

Introduction: Multiple sclerosis (MS) is characterised by demyelination, but it is known from pathology studies that inflammation, evidenced by infiltrating immune cells and activated microglia in normal appearing white matter (NAWM), is also present at different disease stages. The effects of inflammation on MS progression are not yet understood; positron emission tomography (PET) offers the opportunity to monitor these effects and thus aid patient stratification and guide therapy selection, adding to the information already available from MRI techniques.The 18 kDa translocator protein (TSPO) is upregulated in activated microglia, which are present in inflammatory lesions and NAWM in MS. [11C]-(R)-PK11195 is a TSPO-PET ligand which is known to show localisation and extent of inflammation in many brain diseases, including MS1, but suffers from a poor signal to noise ratio (SNR) and short (20 minutes) half-life. [18F]GE-180 is a novel TSPO-PET ligand which has shown improved SNR and lower non-specific binding than [11C]-(R)-PK11195 in pre-clinical models2. Here, we present initial results from a phase 1 study in relapsing-remitting MS (rrMS) and healthy control subjects (HVs). Methods: HVs and rrMS subjects were assessed for binding status. One non-binder in each group was included. Subjects underwent T1 pre and post gadolinium enhanced MRI scans and T2 scans additionally. 90 minute dynamic [18F]GE-180 PET scans were performed typically within 1 month. PET and T1 pre-contrast MR images were coregistered in the software package PMOD, and the Hammers atlas3 was applied to acquire standardised uptake values (SUVs) for regions of interest (ROIs). Selected areas of T2 hyperintensity were defined manually to give MS lesion SUVs. A cortical composite grey matter (GM) region was chosen as reference for SUVRs. Results: No Gd-enhancing lesions were identifiable from T1 postcontrast MRIs. With [18F]GE-180, whole WM uptake (SUVRGM) was slightly higher in rrMS than HVs, while a lesion-rich periventricular (PV) area yielded SUVRs which were approximately 10% higher in rrMS than HVs. PET-defined lesions in rrMS showed an elevation of 20 - 40% compared to HVs.Conclusion: Preliminary results suggest that [18F]GE-180 is able to identify areas of inflammation in PET images which are non-Gd enhancingin T1. 1 Rissanen, E. et al. 2014, J. Nucl. Med., 55: 1-6 2 Boutin, H et al. 2015, Eur. J. Nucl. Med. Mol. Imaging, 42: 503-511 3 Hammers, A. et al. 2002, Hum. Brain Mapp., 15: 165-174.