Summary: Like native DOPA, [18Fl-6-fluoro-L-3 ,4-dihydroxyphenylalanine ([18F1FDOPA) is subject to methylation and decarboxylation. To determine the rates of formation and elimination of [18F1FDOPA metabolites, plasma from human subjects undergoing positron emis sion tomographic (PET) studies was analyzed by high performance liquid chromatography (HPLC). In addition to the principal metabolite O-methyl-[18F1FDOPA (OMe [18F1FDOPA), two decarboxylated metabolites were de tected in plasma from carbidopa pretreated subjects. The concentrations of each metabolite during 90 min following tracer injection could be described as a function of the concentration of [18F1FDOPA, and two rate constants; ko, the rate of formation, and k_ I ' the rate of clearance. Plasma metabolite time series generated from total plasma activity curves and measured rate constants wereThe labeled amino acid 6-[lsP]fluoro-L-3,4-dihydroxyphenylalanine (esp]PDOPA) is employed as a tracer of the enzyme aromatic amino acid de carboxylase (AAAD) in positron emission tomogra phy (PET) studies of the human brain (Garnett et aI. , 1983; CaIne et aI. , 1985; Gjedde et aI., 1991; Kuwabara et aI. , 1992
668in close agreement with the actual concentrations deter mined by HPLC fractionation. Population means for ko (0.011 ± 0.002 min-I ) and L I (0.010 ± 0. 003 min-I ) were used to generate "simulated" plasma curves. The measured and generated plasma curves were used as in puts for estimation of partition and decarboxylation co efficients of [18F1FDOPA in brain. The use of generated input functions from normal population means of transfer coefficients did not introduce a systematic error into the estimate of the enzyme activity. However, the high vari ability of these estimates in patients precludes the use of this technique as an alterative to individual HPLC mea surements.ery. In carbidopa-pretreated subjects, circulating eSp]PDOPA is rapidly metabolized by catechol O-methyltransferase (COMT) to O-Methyl [lsp]PDOPA (OMe-esP]PDOPA), which pene trates the blood-brain barrier more readily then the parent compound (Reith et aI. , 1990). Kinetic mod els for the estimation of cerebral AAAD activity therefore require the fractionation of plasma in or der to determine the relative amounts of esp] PDOPA and its principal metabolite (Gjedde et aI. , 1991; Huang et aI. , 1991). However, the pharma cokinetics of eSp]PDOPA with respect to minor plasma metabolites has not previously been de scribed.Time series for the plasma concentrations of eSp]PDOPA and its metabolites can be fitted to lin ear differential equations described the respective coefficients of formation (ko) and clearance (k _ I) of a metabolite (Gjedde et aI. , 1991). The present study consists in a retrospective reanalysis of 28 PET studies, in which the coefficients were deter-