[18F]fallypride binding in the mouse brain: test-retest and effects of registration

The quantification of in vivo receptor kinetics with PET tracer experiments is an intricate and challenging problem especially for small animals such as rats and mice. A test-retest scan is usually set up in order to confirm an observed experimental effect or to examine the reliability of the experiment design. Inadequate processing of the image data may also mask small effects. The purpose of this study was to investigate the effect of image registration on [18F]fallypride binding potentials calculated from PET mouse test-retest data. Sub-optimal registration affected the quantification of in vivo receptor kinetics with [18F]fallypride. The absence of anatomical information in the [18F]fallypride image and the lack of a homogeneous tracer distribution, even during the earlier minutes of the scan, lead to erroneous automatic registration. A FDG scan after each [18F]fallypride test could improve registration as FDG provides a more homogeneous brain image. Variability in the data could also result from stress induced by anaesthesia or the experimental environment.