Measurement of striatal and thalamic dopamine D2 receptor binding with 11C-raclopride

11C-Raclopride is a widely used positron emission tomography (PET) tracer for measurement of striatal D2 dopamine receptor binding characteristics. Recently, 11C-raclopride has also been used for quantification of thalamic D2 receptor binding. We studied reproducibility and validity issues on the thalamic D2 binding measurements using healthy volunteer test-retest data and simulated data. Eight healthy male volunteers received 11C-raclopride as a bolus injection in a standard test-retest design using 3-dimensional PET. The displacement of thalamic 11C-raclopride binding by the D2 receptor antagonist haloperidol was studied in two female schizophrenic patients. With regards to reproducibility and reliability, thalamic 11C-raclopride binding could be described with a simplified reference tissue model resulting in binding potentials (BPs) between 0.38 and 0.66. In comparison, the model failed to describe 11C-raclopride binding consistently in temporal cortex due to low specific signal. Measurement of thalamic 11C-raclopride BP was reproducible with a test-retest variability of 7.6+/-6.2% and reliable with an intraclass correlation coefficient (ICC) of 0.87. Comparable ICCs were observed in caudate and putamen (0.84-0.96). With regard to validity, subchronic low dose haloperidol treatment reduced specific 11C-raclopride binding equally in putamen and thalamus but a higher dose induced clearly higher D2 receptor occupancy in putamen than in thalamus. Noise simulations indicated that this can partly be explained by an over-estimation of thalamic D2 receptor BP in noisy conditions (low signal, high occupancy). The D2 receptor BP in putamen was clearly more resistant to noise. We conclude that the reproducibility and reliability of thalamic 11C-raclopride BP is good and equal to, or only slightly less than, those observed in caudate or putamen. However, the signal-to-noise ratio for quantification may become too low especially in receptor occupancy-type studies, leading to an artefactual underestimation of measured D2 receptor occupancy.