Abstract
202
Objectives (11C)Befloxatone is a selective PET radioligand for monoamine oxidase A (MAO-A). MAO-A are involved in brain disorders, such as depression, schizophrenia, Parkinson, Alzheimer, Huntington disease, and nicotine addiction. Although (11C)Befloxatone is a promising ligand which has been used for clinical studies, its kinetic modeling has not been fully characterized.
Methods Brain PET data and arterial samples from a previous study were further analyzed. After intravenous injection of (11C)Befloxatone (274 ± 34 MBq), 15 healthy men underwent brain PET for 1 h and serial radial blood sampling to measure the arterial input function. Distribution volume (VT) was determined by compartmental (1- and 2-tissue, unconstrained and K1/k2 constrained) and noncompartmental (Logan analysis) methods.
Results After injection, the brain concentration of radioactivity showed a moderately slow wash-out after the peak. Brain radioactivity was most likely composed of parent radioligand, as no radiometabolites were detected in arterial blood. For most brain regions and in all subjects, F-test showed that the goodness-of-fit of the unconstrained 2TCM was superior to that of 1TCM. Moreover, the 2TCM had lower mean Akaike Information Criterion (-11.9) and higher mean Model Selection Criterion (4.0) scores than the 1TCM (18.0 and 3.0, respectively). The unconstrained 2TCM identified VT with average standard error (SE) across brain regions of 6.3%. The K1/k2 constrained 2TCM provided a similar goodness-of fit as the unconstrained model, but improved VT identifiability (SE: 5.0%), especially in white matter regions. Regional VT values ranged from 18 in the thalamus to 6.5 in cerebellum, with an average of 13.7 ± 2.0 mL cm-3. Logan-VT values correlated well with results from the 2TCM (average bias <10%).
Conclusions Binding to MAO-A can be reliably quantified by (11C)Befloxatone using either a constrained 2TCM or the Logan plot.