Quantification of Human Caudate D2 Dopamine Receptor Density with Positron Emission Tomography: A Review of the Model

Abstract
The radioligand [11C] N-methylspiprone (NMSP) has been used to assess relative and absolute D2 dopamine receptor densities in the living human brain with positron emission tomography (PET). In those studies, a three-compartment model was used to analyze the NMSP accumulation in the caudate nucleus. Absolute receptor densities were determined from studies in both the presence and absence of the inhibitor haloperidol; haloperidol was used to reduce the number of available receptors. We show that the mathematical analysis of the model requires that occupancy of D2 dopamine receptors by NMSP be negligible. The published parameter values (K1, k2, k3 k4(t), and BM'max) for studies in the presence of haloperidol are shown to be inconsistent with this requirement. Potential sources of the inconsistency include a high mass dose of NMSP, as well as the need to assign values for the association and dissociation rate constants for haloperidol, and for the association rate constant for NMSP. For these reasons, the method for absolute receptor density determination may be in error. If sufficiently low masses of NMSP are injected, however, the value of BM'max might provide a measure of relative receptor density.