Error Analysis for Quantification of

Abstract

To estimate receptor binding of ligand by positron emission tomography (PET) without an arterial input function, several quantitative approaches based on the use of a reference region have been proposed. We compared three approaches for quantifying extrastriatal D(2) dopamine receptors using [(11)C]FLB 457. The PET measurements were performed on seven healthy men. Binding potential (BP) of [(11)C]FLB 457 was calculated by the reference tissue model method, transient equilibrium method, and late time method. The reference tissue model describes the time-activity curve in a brain region in terms of that in the reference region, assuming that the levels of nondisplaceable radioligand binding in both regions are the same. The transient equilibrium theoretically occurs when the derivative for specific binding is zero. With the late time method, BP is calculated by integrating a late part of the time-activity curve. BP values obtained by all methods were in good agreement with those obtained by the kinetic approach, and the highest coefficient of correlation was observed in the reference tissue model method. In the simulation study, the error of BP calculated by the reference tissue model method was smallest. Moreover, the effect of the difference in the influx rate constant K(1) between the brain and the reference regions on BP was nearly avoided as theoretically predicted. We concluded that the reference tissue model method is most suitable for calculating BP of extrastriatal D(2) dopamine receptors with [(11)C]FLB 457. Copyright 2001 Academic Press.