Test-retest reproducibility of extrastriatal dopamine D2 receptor imaging with [123I]epidepride SPECT in humans

This study evaluated the test-retest reproducibility of D2 receptor quantification in the thalamus and temporal cortex using [123I]epidepride SPECT.

Ten healthy volunteers (4 men, 6 women; age range, 19-46 y) underwent 2 SPECT studies (interval, 2-26 d) using a bolus-plus-constant-infusion paradigm (bolus-to-infusion ratio = 6 h; infusion time = 9 h). Plasma clearance (in liters per hour) and free fraction (f1) of the parent tracer were measured. Radioactivity (in becquerels per gram) in the thalamus, temporal cortex, and cerebellum were normalized to the infusion rate (in becquerels per hour). Normalized striatal radioactivity was also measured to assess reproducibility in regions with a high density of receptors and better counting statistics. The outcome measures obtained were V3 (receptor density [Bmax]/equilibrium dissociation constant [KD]), V3' (f1 x Bmax/KD), and RT (specific-to-nondisplaceable tissue ratio).

Test-retest variability and reliability (intraclass correlation coefficient) were 10.8% and 0.88, respectively, for plasma clearance and 15.3% and 0.77, respectively, for f1. The test-retest variability of brain-specific (target minus nondisplaceable) radioactivity was higher in the thalamus and temporal cortex than in the striatum, although reliability was comparable. Among the outcome measures, V3' showed better test-retest variability and reliability in the thalamus (13.3% and 0.75, respectively) and temporal cortex (13.4% and 0.86, respectively).

Brain radioactivity was the main source of variability for quantification of extrastriatal D2 receptors with [123I]epidepride. The reproducibility of outcome measures in extrastriatal regions was good. However, because receptor density was lower in extrastriatal regions than in the striatum, the counting statistics in these regions were low and reproducibility was affected by the higher test-retest variability of brain-specific radioactivity. Compared with V3 and V3', RT showed less test-retest variability in the thalamus and temporal cortex but lower reliability. Moreover, measurement of RT may be affected by the presence of potential lipophilic metabolites entering the brain.