Glial activity load on PET (GALP) reveals persistent ‘smoldering’ inflammation in MS despite disease modifying treatment: [F-18]PBR06 study

Introduction Cortical grey (CoGM) and white matter (WM) microglial activation (MA) is involved in the pathogenesis of multiple sclerosis (MS). [F-18]PBR06 positron emission tomography (PET) targeting 18kilodalton-translocator protein (TSPO) can detect abnormal MA in MS. Aims and Objectives The goal of this study is to determine the effect of disease modifying treatment (DMT) efficacy on modulating the extent and clinical and radiological correlates of MA in MS patients. Methods Thirty [F-18]PBR06 PET scans were performed in 22 MS patients (13 RR, 9 SP, mean age 46±14 years, 15 females, median EDSS 3.5, mean T25FW 7.2±4.6s) and 8 healthy controls (HC). Individualized z-score maps of brain parenchymal MA were generated by voxel-by-voxel comparison between each subject’s PET SUVR images and a HC dataset. Logarithmically transformed ‘Glial activity load on PET’ scores (calculated as the sum of voxel-by-voxel z-scores ≥4 in CoGM and WM regions), ‘lnGALP’, were compared between MS subjects on DMT with high efficacy (HT; including rituximab, ocrelizumab, natalizumab and fingolimod, n=13) versus those on no or lower efficacy treatment (LT; including glatiramer acetate and interferons), and correlated with clinical measures and cortical thickness (measured using Freesurfer). p<0.05 was considered statistically significant. Results CoGM and WM lnGALP scores were higher in MS vs. HCs (10.0±1.5 vs. 7.5±1.5 and 9.8±1.5 vs. 6.6±2.4, both p<0.01) and were inversely correlated with cortical thickness across groups (r=-0.44 and - 0.48, both p<0.05, n=30). In HT-MS group, CoGM and WM lnGALP was significantly lower as compared to LT-MS group (9.1±1.0 vs. 11.3±1.1 and 9.1±1.3 vs. 10.8±1.4, p=0.000075 and 0.006) but remained abnormally higher than in HC group (p=0.006 and 0.02, respectively). Within HT-MS patients, CoGM lnGALP scores were higher in SP vs. RR subgroups (p=0.008), correlated positively with EDSS, T25FW, fatigue scores and serum GFAP levels (r=0.65,0.79, 0.75 and 0.67, all p<0.05), and inversely with cortical thickness (r=-0.66, p=0.01). Conclusions High-efficacy DMTs decrease, but do not normalize, CoGM and WM MA in MS patients. Such “residual” MA in CoGM is associated with clinical disability, symptom severity and cortical degeneration. Individualized mapping of TSPO-PET using [F-18]PBR06 can potentially serve as an imaging biomarker for evaluating emerging therapies targeting MA in MS patients who are worsening despite high-efficacy DMTs.