412. TARGETING MICROGLIAL ACTIVATION IN SCHIZOPHRENIA: A LONGITUDINAL [18F]DPA-714 PET IMAGING STUDY INVESTIGATING THE EFFECTS OF A 3-MONTH TREATMENT WITH NATALIZUMAB

Abstract Background Microglial activation is implicated in the pathophysiology of schizophrenia. The monoclonal antibody natalizumab antagonizes α4β1-integrins, and is known to suppress activation of microglia in neuroinflammatory disorders. Aims & Objectives To pharmacologically reduce microglial activation, we investigated the effects of a 3-month treatment with natalizumab on [18F]DPA-714 distribution volume ratio (DVR) and symptom measures in patients with first-episode psychosis (FEP). We hypothesized that natalizumab would decrease [18F]DPA-714 DVR at follow-up, and this change would correlate with symptom improvement. Method A baseline case-control comparison of patients with FEP and healthy volunteers was performed, as well as a nested longitudinal study where patients were assessed following a 3-month treatment with natalizumab 300mg or placebo. All patients were symptomatic despite receiving dopamine blockers (median chlorpromazine equivalent dose 287mg/day). Baseline and follow-up brain imaging was carried out using [18F]DPA-714, a positron emission tomography (PET) radiotracer for the translocator protein (TSPO), which is upregulated in activated microglia. Participants with the low affinity binder TSPO genotype were excluded. Total gray matter (GM), frontal lobe GM, and temporal lobe GM were defined as a priori regions of interest (ROI). We used DVR as our outcome measure to quantify TSPO binding, while using a supervised clustering approach to define a pseudo-reference region. For the case-control comparison, we used ANCOVA with age and TSPO genotype included as covariates. Paired t-tests were used to analyze changes in DVR at follow-up. We used repeated-measures ANOVA to test if effects of natalizumab on symptom measures differed from placebo. Spearman’s correlation was used for correlations. All analyses were two-tailed with p<0.05 considered statistically significant. Results 62 patients with FEP (median duration of illness 22 months, median PANSS total score 55) and 41 healthy volunteers received baseline PET imaging. Demographics including age, sex, and TSPO genotype were not statistically different between groups. 47 patients completed follow-up PET imaging after receiving natalizumab (n=31) or placebo (n=16). At baseline, DVR was higher in patients relative to controls in total GM (η2=0.043, p=0.038) and temporal lobe GM (η2=0.057, p=0.016), but not in frontal lobe GM (p=0.406). Contrary to our expectation, there was no significant change in DVR across these ROIs following natalizumab treatment (all p>0.05). The magnitude of improvement in PANSS total scores was similar between natalizumab and placebo-treated groups, with a significant effect of time (p<0.001) but not a time*group interaction (p=0.768). There was no significant correlation between percent change in PANSS total scores and change in DVR from baseline to follow-up in all three ROIs (all p>0.05). Discussion & Conclusions To our knowledge, this is the largest PET TSPO study to date in FEP, and the first study to test the effects of natalizumab in patients with schizophrenia. Our findings indicate higher TSPO levels in temporal and total GM in patients with FEP compared to healthy controls, with a small to medium effect size. Natalizumab treatment did not normalize this signal or improve symptom measures relative to placebo, indicating this is unlikely to be an efficacious therapeutic approach for schizophrenia. ClinicalTrials.gov: NCT03093064.