[18F]DPA-714 PET imaging of translocator protein TSPO (18 kDa) in the normal and excitotoxically-lesioned nonhuman primate brain

Purpose We aimed to characterize pharmacologically the TSPO- radioligand [ 18 F]DPA-714 in the brain of healthy cynomolgus monkeys and evaluate the cellular origin of its binding in a model of neurodegeneration induced by intrastriatal injection of quinolinic acid (QA). Methods [ 18 F]DPA-714 PET images were acquired before and at 2, 7, 14, 21, 49, 70, 91 days after putaminal lesioning. Blocking and displacement studies were carried out (PK11195). Different modelling approaches estimated rate constants and V T (total distribution volume) which was used to measure longitudinal changes in the lesioned putamen. Sections for immunohistochemical labelling were prepared at the same time-points to evaluate correlations between in vivo [ 18 F]DPA-714 binding and microglial/astrocytic activation. Results [ 18 F]DPA-714 showed a widespread distribution with a higher signal in the thalamus and occipital cortex and lower binding in the cerebellum. TSPO was expressed throughout the whole brain and about 73 % of [ 18 F]DPA-714 binding was specific for TSPO in vivo. The one-tissue compartment model (1-TCM) provided good and reproducible estimates of V T and rate constants, and V T values from the 1-TCM and the Logan approach were highly correlated ( r 2 = 0.85). QA lesioning induced an increase in V T , which was +17 %, +54 %, +157 % and +39 % higher than baseline on days 7, 14, 21 and 91 after QA injection, respectively. Immunohistochemistry revealed an early microglial and a delayed astrocytic activation after QA injection. [ 18 F]DPA-714 binding matched TSPO immunopositive areas and showed a stronger colocalization with CD68 microglia than with GFAP-activated astrocytes. Conclusion [ 18 F]DPA-714 binds to TSPO with high specificity in the primate brain under normal conditions and in the QA model. This tracer provides a sensitive tool for assessing neuroinflammation in the human brain.