sign in sign up
<<
Home , Dopamine

PET Evaluation of the System of the

Nora D. Volkow, Joanna S. Fowler, S. John Gatley, Jean Logan, Gene-Jack Wang, Yu-Shin Ding and Stephen Dewey Medical and Chemistry Departments, Brookhaven National Laboratory, Upton, New York and Department of Psychiatry, SUNY-Stony Brook, Stony Brook, New York

in three neuronal groups: the retrobulbar, the Dopamine plays a pivotal role in the regulation and control of (SN) and the (VTA). Dopamine movement, and cognition. It also is closely linked to from the SN project predominantly to the dorsal and reward, and . Abnormalities in brain dopa mine are associated with many neurological and psychiatric disor are mainly concerned with initiation and execution of move ders including Parkinson's disease, and substance ments (9). Those from the VTA project predominantly to limbic abuse. This close association between dopamine and neurological and limbic-connected regions including , and psychiatric diseases and wfth substance abuse make ft an orbital and cingulate cortices, and and important topic in research in the neurosciences and an important are involved with reinforcement, motivation, mood and thought molecular target in drug development. PET enables the direct organization (1, 10—13). Dopamine neurons from the retrobul measurement of components of the dopamine system in the living bar area project to the where they regulate . It relies on radiotracers which label dopamine recep tors, dopamine transporters, precursors of dopamine or com secretion from the pituitary (14). Dopamine cells in pounds which have specificity for the which degrade the SN and in the VTA and their major projections are shown dopamine. Additionally, by using tracers that provide information on in Figure 1. regional brain or blood flow as well as neurochemically Dopamine is synthesized in the dopamine neurons where it is specific pharmacological interventions, PET can be used to assess stored within vesicles which protect it from oxidation by the functional consequences of changes in brain dopamine activity. (MAO). Dopamine is released into the PET dopamine measurements have been used to investigate the synapse in response to an and interacts with normal human brain and its involvement in psychiatric and neuro logical diseases. It has also been used in psychopharmacological postsynaptic dopamine receptors. The concentration of dopa research to investigate dopamine drugs used in the treatment of mine in the synapse is regulated primarily by its by the Parkinson's disease and of schizophrenia as well as to investigate dopamine transporters, to maintain low (nanomolar) steady the effects of drugs of abuse on the dopamine system. Since state concentrations (15). Dopamine is also removed by oxida various functional and parameters can be studied in tion by MAO A in neurons and by MAO B in glia which the same subject, PET enables investigation of the functional integ surround the nerve terminals and by rity of the dopamine system in the human brain and investigation of 0- (COMT). Brain dopamine release is the interactions of dopamine with other . Through the parallel development of new radiotracers, kinetic models and regulated by as well as by other neuroanatomically better instruments, PET technology is enabling investigation of distinct neurotransmitters through interactions with the dopa increasingly more complex aspects of the human brain dopamine mine (16). system. This paper summarizes the different tracers and experimen PET is an imaging method used to track the regional tel strategies developed to evaluate the various elements of the distribution and kinetics of chemical compounds labeled with dopamine system in the human brain with PET and their applications short-lived positron-emitting isotopes in the living body (1 7). It to clinical research. was the first technology that enabled direct measurement of KeyWords:imaging;neurotransmitters;;psychiatriccomponents of the dopamine system in the living human brain. illnesses; neurological illnesses. Dopamine was labeled with I‘C25 yr ago for the purpose of J NucIMed1996;37:1242—1256 imaging metabolism in peripheral organs such as the adrenals and the (18). Because dopamine does not Thedopaminesystemisinvolvedintheregulationofbraincross the blood-brain barrier, however, imaging studies of regions that subserve motor, cognitive and motivational behav dopamine in the living brain have been indirect, relying on the iors (1—3).Disruptions of dopamine function have been impli development of radiotracers to label dopamine receptors, dopa cated in neurological (4) and psychiatric illnesses including mine transporters, precursors of dopamine or chemical com substance abuse (5 ), as well as on some of the deficits pounds which have specificity for the enzymes which degrade associated with aging ofthe human brain (6). This has made the synaptic dopamine (Fig. 2). Table 1 gives the brain concentra dopamine system an important topic in research in the neuro tion and activity of the dopamine-related elements which have sciences and neuroimaging as well as an important molecular been measured with PET. Additionally, through the use of target for drug development (4, 7,8). tracers that provide information on regional brain activity (i.e., Dopamine cells reside predominantly in the mesencephalon brain glucose metabolism and cerebral blood flow) and of appropriate pharmacological interventions, it has been possible Rec@vod Sept. 12, 1995; revision accepted Oct. 18, 1995. For correspondence or reprints contact: Nora D. volkow, MD, Medic@Department, to assess the functional consequences of changes in brain Bldg. 490, Brookhaven National Laboratory, Upton, NY 11973. dopamine activity.

1242 THE JOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 ‘July 1996 TABLE I Binding Site Concentrations and Enzymatic Activities in the Human Striatum*

ConcentrationActivityBinding site or activity (pmole/g)(pmole/g/sec)

ICA PreF Receptors Dl (82) 50 D2 (238) 20 Transporters

: OFC.. Dopaminetransporter(239) 400 650t Enzymes Tyrosinehydroxylase(240) 25 AromaticL-aminoacid decarboxylase 200

Monoamineoxidase A (240,241) 200 4@Q* Monoamineox@aseB (240) 200 Catechol-O-methyftransferase(242,243) 40@ FIGURE 1. Simplifieddiagramfor the dopaminergicprojectionsof the substantia nigra(SN)and the ventraltegmental area çVTA).CG = Cingulate Gyrus, PreF = PrefrOntalCortex, OFC = OrbitofrontalCortex, NAc = ‘ValuescitedarefairlyrepresentaThieofvalues reported inotherstudies of NucleusAccumbens, CA = Caudate, PUT = Putamen, AM = Amygdala, postmortem .However,considerablevariationisfound inthe literature, Hipp = Hippocampus. associated withdifferencesin methodologiesand in indMdualsubjects. A detailed review of invitro studins isfar beyond the scope ofthis article. Since This article summarizes the different tracers and experimen enzymaticactivitiesare usuallymeasured withsaturatingsubstrate concen tal strategies developed to evaluate the various elements of the trations,the cited values probab@@do not reflectmetabolicfluxesinvn,o. t@ermined in rat striatum. dopamine system in the human brain with PET and their 1Cornbined MAO A and MAO B activities assayed using as applications in clinical research. Although SPECT methodol substrate. ogy can also be used to measure some of the same components @Determinedin human . (19), this chapter will not explicitly discuss this methodology.

DOPAMINE RECEPTORS dopamine receptors which are grouped into two major families: Dopamine receptors are present both pre- and postsynapti those which stimulate adenyl cyclase (Dl, D5) and those which cally at dopaminergic synapses (Fig. 2). At the postsynaptic inhibit adenyl cyclase (D2, D3, D4) (21 ). The concentrations as side they function in cell-to-cell communication, and at the well as the locations of these receptors in human brain differ presynaptic side they modulate the release and synthesis of and the most ubiquitous are the Dl (50 pmole/g) and D2 dopamine (16). The dopamine receptors have been the main receptors (20 pmole/g). The highest concentration for Dl and therapeutic targets in the treatment of psychotic symptoms in D2 receptors occurs in striatum (22 ), where D 1 receptors schizophrenics and of extrapyramidal motor symptoms in appear to be predominantly expressed by striatal output neurons Parkinson's disease patients (20). There are five subtypes of projecting to the SN; and D2 receptors appear to reside mainly in output neurons projecting to the (23). Extrastriatal regions have much lower densities (0.3—4 pmole/g) of D2 (22,24) and Dl receptors (25). D3 receptors exist in low concentrations (1 pmole/g) in the shell of the nucleus accumbens and in the (26). The concentration of D4 receptors is also very low (2. 1 pmole/g tissue) and they are localized in several limbic and cortical regions with relatively lower levels in striatum (27,28). D5 receptors are located in limbic areas (29). PET tracers to measure D2 and Dl receptors have been developed; however there are currently no specific PET ligands to differentially evaluate D3, D4 and D5 receptors.

100 _—!@.@____.[ i.—i: 1_:@::—.-_ 200 The D2 receptors were the first to be imaged with PET @ D2 receptors f \ Dl receptors (30—33).Several D2 antagonists have been labeled for Ic0MT use with PET (34). These ligands differ with respect to their I Poit@ceIl affinity for the D2 receptors, their specificity and their kinetics (Table 2). More recent studies have started to focus on labeling FiGURE2@Simpkfieddiagramof a typicalstriataldopaminesynapse.In (34). The most widely used D2 practice, electron microscopic studies show great structural heterogeneity. There are approximately1011dopamineterminalsper gram of stilatum,out PET ligands are the dopamine receptor antagonists [I ‘C]raclo ofatotalnumberofabout1012terminalspergram.Theiraveragevolumeis pride (33) and [‘1C] or [18FJ-labeled N-methylspiropendol that of a sphere of radius 0.55 @,and there are about 5000 terminalsper (NMS) (30,31 ). is a with moderate affinity stviatalcell.Thespontaneous flnngrate ofdopaminecellsis about 3 Hz.The for D2 receptors (Kd = 1000—2000 pM) (35). It has a high @ volumeof each synapticcleft is about 1017 I, total dopamine cleft selectivity in terms of affinity for receptors for other neuro volumeisabout 1 @.dper gram.Theaverage distance between cleftsisabout 2 @.For convenience, Dl and D2 receptors are shown on the same transmitters but also binds to D3 receptors. NMS has a higher postsynaptic cell, whereas they are almost certainly localized on different affinity for D2 receptors than raclopride (Kd = 50—300pM) but cells. it also binds to 5HT2 as well as D4 receptors (27). Because the

PET DOPAMINETRACERS•Volkow et al. 1243 TABLE 2 Radioligandsfor Dopamine Receptors I'1CINNC 112 data data invitro max Time max (mm)DlSCH0.3392023390(244,245)NNC7560.17530-60(244,43)NNC1128-12(34)D2Raclopride1.1420-4075(246,247)NMS0.0512>50010>60(60,168,247)IBZM0.451.66(78,248-250)Epidepride0.06>10>20020120(249)RadioligandKi (nm)HumST/CBan (mm)RodenST/CBt

FiGURE3. Brainimagesobtained I with the Dl receptorradioligand r1c]NNC112(50)inahealthycon II trol.Theimagecorrespondsto the averageactivityobtainedbetween9 and68mm(courtesyof Halldinand collaborators).

The Dl receptors have also been measured with PET though they have been much less investigated. Several Dl receptor antagonists have been labeled with positron emitters and in dude [“C]SCH23390 (47), [“C]SCH39166 (48), [“C]NNC 687, [1‘C}NNC756 (49) and NNC 112 (50) (Table 2). A problem for most of these radioligands is that they also have considerable affinity for 5HT2 receptors (49). [1‘CISCH23390 has been the most widely used (25). High uptake of these dissociation of NMS at the 5HT2 receptors is significantly radioligands occurs in striatum which is the area with the faster than at the D2 receptors, tracer kinetic modeling differ highest Dl receptor density (25). The concentration of Dl entiates these two components of NMS binding (36). Since the receptors in cortex is relatively low but is higher than that for ratio of D2/(D3 + D4) receptor binding is quite high in most D2 receptors (25) and has been measured with PET. Ofthe PET brain regions (37,38) the PET measurements from these two Dl radiotracers the one that fives the highest cortical to tracers predominantly reflect binding to D2 receptors. Also their cerebellar ratios (1 .8—2.2)is [‘C]NNC 112 which makes it binding, at least for NMS, appears to be predominantly to probably the best of the Dl radiotracers for extrastriatal postsynaptic receptors (39,40). For most PET studies D2 imaging (Fig. 3) (34). The dopamine Dl receptor SKF receptor measurements have been limited to the striatum where 75670 was labeled with ‘‘Cand proposed as a ligand that may neurons containing D2 receptors are predominantly GABAergic be potentially useful for measuring high-affinity Dl receptor or cholinergic. sites (51). The use of PET to measure D2 receptors in areas other than Radioligands for Dl and D2 receptors have been used to striatum has been limited by the relatively low concentration of investigate their involvement in aging, in psychiatric and D2 receptors in extrastriatal regions as well as the small neurological illnesses and to assess receptor occupancy by volumes of some of the areas (41 ). Although D2 receptors in drugs. Ofthe brain changes associated with aging, extrastriatal areas have been imaged with PET and SPECT those in the dopamine system are among the most conspicuous (41—44), the limited sensitivity of these instruments do not and are probably responsible for some of the motor and allow accurate quantification. The development of D2 ligands behavioral changes in the elderly. PET studies evaluating the with higher affinity for D2 receptors and of PET instruments effects of age on D2 receptors have consistently reported a with a higher sensitivity and spatial resolution may enable significant decline in D2 receptors with age. The estimates for improved quantitation of D2 receptors in extrastriatal areas. D2 losses obtained from PET studies range between 4—8% per Some ligands with very high affinity for D2 receptors have decade of life (52—55) and are slightly higher than those recently been developed that may facilitate these measure reported for postmortem studies which range between 2—5%per ments. Imaging of the human brain with one of these high decade (56). Dopamine Dl receptors in striatum and frontal affinity ligands [‘‘C]FLB457(Kd = 20 pM), showed signifi cortex have also been found to decrease with age (57). cant accumulation in , substantia nigra, colliculus and Increased brain dopamine activity is an important contributor temporal cortex (41,45). to the symptomatology observed in schizophrenic patients (8). Initially the development of D2 ligands was targeted towards PET studies in schizophrenic patients measuring D2 receptors tracers with very high affinity for D2 receptors. However, tracer have yielded inconsistent results with some studies reporting kinetic modeling of these tracers is limited by tracer delivery elevations, others no changes and others documenting eleva which makes them sensitive to cerebral blood flow (CBF) in tions in some patients but not in others (41 ). Several reasons addition to receptor concentration (46). This is particularly have been given for these discrepancies related to differences in problematic for areas of high receptor concentration since choice of tracers and subject populations (58), including the binding equilibrium is not reached within the time period when suggestion that the elevations observed with NMS could be due the measurements can be made. However, these tracers may be to increases in D4 receptors in schizophrenic patients (27). useful to quantify D2 receptors in areas with relatively low D2 Studies in schizophrenic patients have also been done to assess concentrations as discussed for [‘‘CJFLB457. On the other the relationship between receptor occupancy by antipsychotic hand, recent interest in the development of tracers with lower drugs and therapeutic efficacy. For typical neuroleptics it was affinities has been stimulated by the feasibility of using them to estimated that 70—80%ofthe D2 receptors need to be occupied measure relative changes in synaptic dopamine concentration. for therapeutic efficacy and that higher occupancies are asso

1244 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 •July 1996 TABLE 3 Radioligandsfor DopamineTransporters

dataTimeBmax/TimePeak dataRodent I@inHuman ST@maxbKd@CST/maxbRadioligand (nm)(nCVcc/mCi) ST/CBS (mm)(mVg)PIdST/CB@(mm) (500')12+>12006.T25010@120WIN35428(10,93)12150(60')4+>1205?460dtMPRTI-55(90,251)1.5130

.625215Nomifensine(84)[email protected]—300.9°10°——Cocalne(86,209,253,254)100851.75—70.641.55(96,252)401002.330-401

°Ratstha@ membranes. @‘Timeof maximumuptake in striatum. @CaIcUIatedusingtheratioofdistributionvolume(DV)instriatumtothatinreferencetissue,obtainedusinggraphical(209)forcocaineanddtMP;Bmax/Kd from two-compartment model for ;calculatedas k@,/k@from three-compartmentmodel for WiN 35,428and R11-55. dAt time of maximal striatal binding; not corrected for plasma protein binding. °Metabolitecorrection based on plasma/erythrocytepartitioning. @Maximumvalueobserved;at endof measurementperiodif fOllOWedby +. @Representatjveradioactivityconcentrationin striatum (nCi/cc/mCiinjected)at peak or at indicatedtime after injection. @ from MBq (g tissue)-1/MBq (9 body wt)-l. 1@ermedV3, or equilibriumpartition coeffident (251).

ciated with side effects (80—90%)(59—60). In contrast, for Studies with Parkinson's disease patients have shown that atypical neuroleptics, such as , the average D2 occu while there may be initial increasesduring early stagesof the pancy was significantly lower and ranged between 20—67% disease, for the most part D2 receptor concentrations do not (61 ). Levels of D2 receptor occupancy by typical neuroleptics differ from those of age-matched controls (74—77). Because were linearly relatedto drugplasmaconcentrations(62,63). For L-DOPA requires the presence of dopamine receptors in order Dl receptors, levels of occupation by typical neuroleptics to exert a therapeutic effect, imaging with D2 receptors has ranged between 16—44%and occupation by atypical neurolep been used to predict responsiveness in patients with movement tics ranged between 36—59%(59). These findings may suggest disorders (78). In patients with Huntington's chorea, where the that D2 receptors are a main target for typical but not for pathology is localized in striatal neurons, significant reductions atypical neuroleptics. Though for the atypical neuroleptics, in striatal D2 (79—81) and Dl (82) receptors have been blockade of Dl receptors is larger than that of D2 receptors, documented. Furthermore, because Huntington's disease affects therapeutic efficacy may also involve blockade ofD4 and 5HT2 predominantly medium-sized spiny neurons in striatum, where receptors (64). Studies done to determine if the lack of Dl receptors are located, it has been suggested that Dl ligands therapeutic response seen in nonresponding schizophrenic pa may be more sensitive than D2 ligands in detecting degenera tients was a result of inadequate D2 receptor blockade by the tion in this disease (82). There is also some evidence of antipsychotic drug showed that there were no differences in reductions in Dl receptors in the frontal cortex of patients with receptor blockade by neuroleptics in responders and non Huntington's chorea (82). responders patients (65—66). However, the two groups of patients differed in their baseline D2 receptor measures sug DOPAMINETRANSPORTERS gesting that there may be underlying biological differences that Interest in the (DAT) has been stimu may predict responsiveness to antipsychotic agents (62). These lated, in part, by the fact that it constitutes the main target site fmdings also question the common practice of increasing for the reinforcing properties of (83). Additionally, antipsychotic doses beyond the therapeutic window in nonre because transporters are localized on the presynaptic terminal sponding patients. they serve as markers of dopamine neurons. Several radioli The dopamine systemhas been linked with addictive disor gands have been developed for their suitability as PET and ders. Imaging studies have shown significant reductions in D2 SPECT probes of the DAT. These include [‘‘C]nomifensine receptor availability in cocaine abusers which persist after (84—85), [“C]cocaine(86), [18F] GBR 13119 (87,88), [‘8F] cocaine withdrawal and which were correlated with self ratings GBR 12909 (89), ç―C]and [‘23I]RTI-55(90, 91), [“C]WIN of dysphoria (67,68). Abnormalities in D2 receptor measures 35428 (92—94), [‘C] (95) and [‘‘C]d-threo have also been documented in alcoholics (69). No drug abuse methylphenidate (96). These radioligands differ with respect to studies have been done with PET Dl radioligands. their affinities for the DAT, their speciflc-to-nonspecificbind In psychotic depressed patients increases in striatal D2 ing ratios and their specificity for the DAT as well as their receptors were recently reported. Because the relation was kinetics. Table 3 summarizes this information for DAT ligands between D2 and and not mood this finding was which have been examined in reasonable detail in human interpreted as most likely reflecting the psychotic state rather subjects. The cocaine analogs WIN 35,428 (also known as than the mood abnormality (70). Studies with the Dl receptor CFT) and RTI-55 (alsoknown as a-CIT), have affinities for the ligand [‘‘C]SCH23390 in depressed patients showed reduc DAT approximately 10 and 100 higher, respectively, than tions in the binding potential in frontal cortex but no changes in those of cocaine and S-(+)-nomifensine and d-threo-methyl striatum (71 ). These findings provide further evidence that the phenidate have affinities intermediate between those of cocaine dopamine system may be involved in affective disorders and WIN 35,428. Their times to maximum uptake vary by a (72,73). factor of over 200, that is between about 6 mm for cocaine, and

PET DOPAMINETRACERS•Volkow et al. 1245 over 1200 mm for the high-affinity cocaine analog RTI-55. The initial uptake in brain for the five radioligands in Table 3 is high Nomt@d Pivldnson's and corresponds to approximately 7—10%of the injected dose. Carbon-i 1-nomifensine was the first DAT ligand developed for PET (97). Its uptake in brain is relatively fast and peak DopatrineTransporters concentrations are achieved approximately 20 mm after admin (I―CId-threo-MP) istration which allows for proper modeling and quantification (84). A potential disadvantage of this ligand is that it binds more tightly to the transporter than to the DAT. Carbon-i 1-cocaine exhibits faster kinetics than [1 ‘C]nomifensine, with a half-life in striatum of approximately 20 mm. Though in vitro studies show binding of cocaine to Dopanine02 Receptors as well as norepinephrine transporters, as a PET ( tlIq radopride) ligand it shows specificity ofbinding to DAT (98). A limitation for this tracer is that its specific-to-nonspecific binding ratio is relatively low and that its fast kinetics limit the statistical quality of PET images. The specific binding of [l ‘C]methyl phenidate and of [‘‘C]d-threomethylphenidate in the brain is mainly due to DAT. Its clearance is slower than that of [I ‘C]cocaine which is an advantage for kinetic analysis in that it is fast enough to allow for proper quantification but is slow l@ucoseMetabolism enough to permit appropriate counting statistics. A disadvan (‘@DG) ta@e of racemic [l ‘CJmethylphenidate when compared with [I ‘C]d-threo methylphenidate is that the 1 enantiomer in the racemic mixture contributes to nonspecific binding and con founds its quantification. An advantage for [‘‘C]d-threometh FIGURE4@Brainimages froma mumpletmcer study comparinga patient ylphenidate and for [I 1C}methylphenidate is that its pharmacol withParkinson'sdisease and a control.The imagesforD2receptors and for ogy in is well studied so that unlabeled dopamine transporters correspond to the distribution volumes of r1C]rack pride and of C1C]d-threomethyiphenidate,respectively.Metabolicimages methylphenidate can be administered to humans to assess were obtained v@th18FD0. nonspecific binding. RTI-55 has highly desirable properties which are optimized down-regulate in response to treatment and hence may be less in the ‘23I-labeledcompound used with SPECT (i.e., very high sensitive to the confounding effects of (100). The affinity for the DAT and a high specific-to-nonspecific binding disadvantage is a lack of specificity for the different mono ratio) (90). The SPECT measurements with [1231]RTI-55 are amine transporters. made when the radioligand distribution corresponds to an It is to be noted that most studies in humans using DAT equilibrium situation, which is thought to be the case after about radioligands have shown age-related decreases in binding 18 hr. However, PET measurements with [l 1C]RTI-55 (half (85, 101, 102) which are consistent with the reduction in DAT life = 20 mm) must be conducted very far from equilibrium and in dopamine cells documented in postmortem studies (91 ). A danger under these conditions is that tissues with some (103, 104). In patient populations these radioligands have been threshold concentration of binding sites may trap essentially used to assess dopamine cell degeneration in subjects with every molecule of radiotracer which is delivered. This is Parkinson disease who show marked reductions in DAT when expected to diminish the sensitivity of radiotracer binding to compared with healthy age-matched controls (Fig. 4) small decreases in binding site concentration. Furthermore, (85,90,91,94). Studies in cocaine abusers have shown that radiotracer binding will become more sensitive to alterations in while there are increases in DAT shortly after withdrawal (105) tissue perfusion, which proportionately alters delivery of radio there are decreases or no changes with protracted withdrawal ligand. Carbon-i i-WIN 35,428 also does not achieve a maxi (106). A preliminary study done in violent alcoholics showed mum value of striatal binding within the time constraints of significant elevations of DAT when compared with nonalco PET experiments; its striatal uptake is still rising after more holic subjects (107). In contrast, nonviolent alcoholics had than 5 half-lives for I‘Chave elapsed (1 10 mm) (93). The lower DAT levels than controls (107). extent to which this may compromise its ability to detect small decreases in DAT concentration, or its usefulness in circum DOPAMINE SYNThESIS stances of altered CBF, is presently unclear. PET studies are Dopamine synthesis occurs within the dopamine neuron as required to determine under what circumstances one DAT shown in Figure 5. is transported via amino ligand may be more accurate than another. For example, under carriers in the blood-brain barrier and cell membranes. Once in conditions of decreased DAT availability and/or to examine the intracellular space it is hydroxylated to L-3,4-dihydroxy regions with low DAT densities, ligands with very high (L-DOPA) by (TH, E. C. affinities such as [l ‘C]W[N35428 may be desirable. However, 1.14.16.2).L-DOPA is then decarboxylatedby aromaticL for quantification in patients with decreased CBF, li@ands with decarboxylase (AADC, E. C. 4. 1.i .28) (108) to a relatively lower affinity for the DAT such as [‘CJd-threo form dopamine. methylphenidate may be more appropriate. TH is the rate-limiting in the synthesis of brain Though most of the radioligands developed for monoamine dopamine. PET radiotracers which would enable an assessment transporters have been targeted to the cellular transporters, of TH activity are still under development (109,110). The first [1 ‘C} has been proposed as a PET ligand to PET tracer developed for studies of dopamine metabolism in measure the vesicular (99). The advan the human brain was ‘8F-labeledfluoro L-DOPA (111,112). tage for this type of ligand is that it does not appear to up- or Like L-DOPA, 6-['8F]fluoro-L-DOPA crosses the blood-brain

1246 THEJOURNALOFNUCLEARMEDICINEVol. 37 •No. 7 •July 1996 radioactivity. Recently this tracer has been applied in the study DOPAMINE SYNTHESIS of dopamine metabolism in patients with unipolar depression (120). The accumulation of 18F after the injection of 6-[18F]fluoro L-DOPA has been used as a measure of the integrity of the nigrostriatal dopamine system in Parkinson's disease (7) and a recent postmortem study has correlated fluoro-DOPA uptake constants with nigrostriatal neuronal density (121 ). The extent Tyrosine L-DOPA Do@ to which ‘8Faccumulation is sensitive to smaller age-related losses in dopamine neurons is controversial with some investi gators finding an age-related loss in dopamine neurons is controversial with some investigators finding an age-associated decline and others finding no change in PET studies of normal subjects (122—124). Studies in the postmortem human brain suggest that AADC may be up-regulated in dopamine neurons that are spared during aging and that PET measures of 6-[18F]fluoro-L-DOPA uptake may overestimate the number of dopamine nerve terminals during normal aging (125). In patients with Parkinson's disease, a correlation has been found between 6-['8F]fluoro-L-DOPA uptake in putamen and motor deficits and between 6-['8F]fluoro-L-DOPA uptake in caudate and memory (delayed recall test) (126). Though most FIGURE5. Pathways forthe synthesis and metabolismof dopamine inthe studies done with 6-['8F]fluoro-L-DOPA measure AADC ac CNS.Th, tyroelnehydroxylase(EC1.14.16.2);MDC, aromaticamino acid tivity in striatum, concentrations in the amygdala, mesenceph decarboxylase (EC 4.1.1.28); MAO,monoamine oxidase (EC i.4.3.4);AD, alon, hippocampus and thalamus are higher than those in the aldehydedehydrogenase (EC1.2.1.3);COMT,catechol-O-methyltransferase (116). (EC2.1.1.6). Other tracers have been developed in an attempt to improve quantitation and to reduce the spectrum of labeled from 6-[18F]fluoro-L-DOPA. The best characterized is barrier resulting in an accumulation of ‘8Fin striatum. This [‘8F]fluoro-m-tyrosine which is missing the hydroxyl group on striatal accumulation appears to be unidirectional over the first carbon-4 of the aromatic ring (127, 128). This molecule is a 90 mm and probably reflects the synthesis of fluorodopamine substrate for AADC but is not a substrate for COMT. This and subsequent storage within vesicles (1 13). However, be tracer has been investigated in monkeys and showed low cause L-DOPA (and ) and its metabolites are peripheral metabolism as predicted (118). Initial studies have extensively metabolized by MAO and by COMT, labeled been carried out in humans and it has been proposed that the use metabolites, especially 3-O-methyl-6-[18F]fluoroDOPA also of labeled fluoro-m-tyrosine will be a considerable advance contribute to striatal ‘5Factivity as visualized by PET (114). over the use of labeled fluoro-DOPA for the measurement of This complicates the estimation of indices which reflect the rate AADC activity (129). Another tracer, [‘8F}fluoro-@-fluoro of dopamine synthesis. To minimize the extensive peripheral methylene-m-tyrosine is also being investigated (130). metabolism of fluoroDOPA by AADC, which limits tracer availability in brain, peripheral AADC inhibitors have been DOPAMINE METABOUSM used to spare [‘8F]fluoroDOPA (115). Unfortunately, AADC Monoamine oxidase (MAO, EC i.4.3.4) formally oxidizes blockade enhances COMT-catalyzed formation of peripheral such as dopamine to the corresponding aldehyde (131). [‘8F@3-O-methylfluoroDOPA which enters the brain and con The enzyme exists in two subtypes, MAO A and MAO B (132). tributes to the nonspecific signal. Though both forms can oxidize each of the biogenic amines, It has been suggested that the accumulation of 6-['8F]fluoro MAO A preferentially oxidizes 5-hydroxytryptamine whereas L-DOPA metabolites in the striatum parallels AADC activity MAO B oxidizes benzylamine. Dopamine is a substrate for both and that this reflects the brain's capacitj for dopamine synthesis subtypes of MAO (133), and both subtypes have similar (116). A recent PET study using 6-[ 8F]fluoro-L-DOPA em affinities for dopamine (134). There are selective inhibitors for ployed a kinetic model that took into account labeled metabo each form of the enzyme, the best known of which are lites to calculate the relative rates of conversion of clorgyline for MAO A (135) and L-deprenyl for MAO B (136). 6-['8F}fluoro-L-DOPA to 6-['8F]fluorodopamine. The highest In the human brain MAO B predominates (B:A = 4: 1) (137). values for AADC activity were found in caudate and putamen Since MAO B is largely compartmentalized in glial cells, it has which are consistent with in vitro measurements. been suggested that MAO B inhibition spares extraneuronal The extent to which 6-['8F]fluoro-L-DOPA parallels the dopamine (138). Many studies support a link between increases behavior of L-DOPA has been addressed by comparing in MAO B (but not MAO A) and aging and neurodegenerative 6-['8F]fluoro-L-DOPA and either [j3-' ‘C]L-DOPA, [13-'‘C}6- disease (139). fluoro-L-DOPA or [3H]L-DOPA (11 7,118). The behavior of By far the major medical interest in MAO stems from the 6-['8F]fluoro-L-DOPA qualitatively parallels that of L-DOPA neuropsychological effects of MAO inhibitor drugs on the but its uptake is higher and its metabolism by COMT is faster. human brain. Indeed, MAO inhibitors were among the first The brain kinetics of L-[f3-1‘C]DOPAhave been measured in drugs. The fact that MAO inhibition spares brain human volunteers with PET (119). The use of [@3-'‘C}L-DOPA dopamine led to the use of the MAO B inhibitor L-deprenyl as allows PET studies to be carried out at tracer conditions even an adjunct with L-DOPA in the treatment of Parkinson's though its use was reported to be limited by large statistical disease (140). More recently it has been shown that L-deprenyl variation due to the limited time frame for quantitation of low slows the natural progression of Parkinson's disease (141).

PET DOPAMINETRACERS•Volkow et al. 1247 MonoamineOxidaseB ([1t]L-deprenyl-D2) [11CjRaclopride 4 It Baseline 4 I Methyiphenidat. FIGURE 6. Influx constant images of human brain MAO B using r1cj H-deprenyl-d2.Notk@ethe highvalues inthe subcorticalregionsand lower values incortex inagreement withpostmortemassays. FIGURE7. Breln images of the distributionvolume for C1C]raclor,ñde obtained with no pharmacologicalinterventionand after administrationof methyiphenidate. Though there is controversy concerning the mechanism(s) responsible for the therapeutic actions ofL-deprenyl (142), this mine and the concentration of this compound in the brain is finding has stimulated interest in the development ofnew MAO taken as an index ofdopamine release (151 ). Although the main inhibitor drugs. function of COMT was described in the late 1950s (152), the PET studies of MAO have measured its regional distribution, significance of this enzyme has been less studied, in part, its inhibition and its synthesis rate. Two general approaches to because of the lack of suitable in vivo inhibitors. The recent mapping MAO with PET in vivo have been developed, one uses development of selective and potent COMT inhibitors (150) the labeled suicide inactivators, L-deprenyl and clorgyline, to has presented the opportunity to investigate the distribution of label the two forms of the enzyme (143) and the other uses COMT in the living body with PET and appropriately labeled labeled dimethylphenethylamine to produce ‘‘C-labeled di COMT inhibitors. In this regard Ro 41-0960 (3,4-dihydroxy-5- methylamine, a MAO B-generated which is intra nitro-2'-fluorobenzophenone), a potent and selective fluorine cellularly trapped (144). The most frequently used approach is containing COMT inhibitor which has been reported to cross the measurement of MAO B with [‘‘C]L-deprenyland [‘‘C]L the blood-brain barrier, has been labeled with ‘8Fin order to deprenyl-D2 (a deuterium-substituted analog of L-deprenyl study brain COMT with PET (153). Though PET studies in the with improved sensitivity (145)). baboon using ‘8F-labeledRo 41-0960 demonstrated a negligi The general approach to the in vivo labeling of MAO B with ble uptake in the brain, high uptake was observed in the kidneys [1 ‘C]L-deprenyl is based on the principle of suicide inhibition and in other organs known to have high COMT activity in which the radiotracer becomes covalently attached to the demonstrating the potential of PET to investigate peripheral, if enzyme as a result of oxidation and the formation of a very not central, COMT activity (154). reactive intermediate. The distribution of ‘‘Cin brain after the injection of [‘‘C]L-deprenylclosely parallels the distribution of DOPAMINERGIC RESPONSIVITY AND MAO B as determined in the postmortem human brain (Fig. 6). DOPAMINERELEASE Carbon-i i-L-deprenyl has been used to determine the duration The competition of endogenous dopamine with D2 receptor of MAO B inhibition by pharmacological doses of L-deprenyl. radioligands presents the opportunity to measure changes in Since L-deprenyl is a suicide inactivator of MAO B, enzyme synaptic dopamine in vivo by observing the degree ofchange in recovery after withdrawal from the drug requires the synthesis the binding of a D2 receptor radiotracer by a pharmacological ofa new enzyme. PET measurements showed that the half-time challenge or other perturbation (155). The in vivo striatal for recovery of brain MAO B averaged about 40 days (146). binding of [3H}raclopride has been shown to be sensitive to Though L-deprenyl is given to patients at a dose of 10 mg/day, pharmacologically-induced changes in endogenous dopamine it is obvious from the PET study that complete MAO B (35, 156). This property in turn has been used to assess changes inhibition can be obtained at a fraction of this dose. A similar in endogenous synaptic dopamine concentration with PET with investigation was carried out to study the duration of MAO B labeled raclopride and with labeled NMS (157—162). Compar inhibition by the new reversible, MAO B inhibitor drug Ro 19 isons are made between the binding of the radiotracer at 6327 (147, 148) which is being proposed for the treatment of baseline and after administering a drug that changes dopamine Parkinson's disease. This study showed total recovery of the concentration (i.e., methylphenidate, ) (Fig. 7). enzyme after 36 hr of Ro 19 6327 discontinuation. Studies in humans have shown that methylphenidate signifi Catechol-O-methyltransferase (COMT; EC 2.1.i.6) is one of cantly reduced [‘‘C]raclopridebinding in the brain. The reduc the two major enzymes which metabolize dopamine (149). tions varied among subjects and decreased with age (160). COMT catalyzes the transfer of a methyl group from S Because [‘‘C]raclopridebinding is highly reproducible (163— adenosyl-L-methionine to the phenolic group of a substrate that 164) these reductions most likely reflect changes in synaptic must have a catechol . It is distributed throughout the dopamine. The effects of amphetamine on dopamine release in body and brain and is an important molecular target in the the human brain have also been measured with SPECT and the development of drugs to treat Parkinson's disease (150). The D2 radioligand [‘231]IBZM(165). product of dopamine metabolism by COMT is 3-methoxytyra Investigating relative changes in dopamine concentration in

1248 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 •July 1996 response to a drug provides a tool with which to evaluate the (1 78) and the system (1 79) to modulate striatal dopa relationship between drug-induced changes in dopamine con mine release. In humans the interactions between dopamine and centration and its behavioral effects (160,165). It may also have been investigated in healthy controls (1 77). enable the detection of dopamine dysfunction with a higher Striatal dopamine D2/acetylcholine interactions have also been sensitivity than when studying patients during baseline condi studied in primates using [‘8F](—)-4-N-ethyl-fluoroacetamido tions. Limitations of this strategy are: the inability to measure benzovesamicol as a radioligand marking cholinergic activity baseline dopamine concentration; the limited spatial resolution (180). of PET (measures are of large volumes of tissue which may be The possibility that abnormal interactions between cortical heterogeneous in their responses to drugs); the inability to glutaminergic neurons and the striatal dopamine system may differentiate whether changes in binding reflect the magnitude underlie schizophrenia was also recently investigated with PET of the rise in dopamine or the ratio of high- versus low-affinity where it was demonstrated that DOPA decarboxylase activity is D2 receptors (dopamine predominantly competes with the elevated in patients with psychosis as measured with radiotracers for the binding to high-affinity D2 receptors); and 6-['8F]fluoro-DOPA (181 ). This observation was interpreted as the relatively low sensitivity ofthe technique which requires the supporting the hypothesis that in schizophrenia, psychosis is the use of pharmacological challenges that raise dopamine levels result of insufficient cortical (glutamatergic) stimulation of severalfold. Also, because the pharmacological challenge can nigrostriatal terminals leading to a low baseline concentration affect tracer delivery (166), studies are required to assess its of extracellular dopamine in the striatum and a corresponding effects on tracer kinetics and on the model parameter. In the increase in the activity of the enzymes involved in dopamine case of[' ‘C}raclopride,the model parameter used (BmaxlKd) is synthesis (182). insensitive to changes in CBF (167). To our knowledge, the sensitivity of these strategies to nonpharmacological, i.e., be REGIONAL BRAIN GLUCOSE METABOUSM AND CBF havior-related (168) changes in dopamine has not been dem Brain metabolism is tightly coupled with brain function and onstrated. This methodology differs from the measures obtained it can therefore be used to assess the regional functional with (169) in that it simultaneously measures all changes associated with manipulation of the dopamine system. brain regions, it measures predominantly synaptic rather than Local brain metabolic rates can be measured using FDG (183). extracellular dopamine concentration, and it is minimally inva Though a number of energy-requiring processes contribute to sive permitting its use in awake human subjects. the basal rate ofglucose utilization, it is neuronal activity which Recent studies in our laboratory using electrically-stimulated is the major contributor to glucose utilization (184—186). Under brain slices have suggested that the binding of Di radioligands physiological conditions CBF is tightly coupled to glucose may be more sensitive to changes in synaptic dopamine than to metabolism so that it can also be used as an index of brain D2 radioligands. To our knowledge this has not yet been function. However, care must be taken when measuring CBF as evaluated in PET experiments (170). a functional tracer in experiments that use pharmacological challenges since some of the psychoactive drugs have vasoac INTERACtiONS OF DOPAMINE WITh OThER tive properties (166). In humans, manipulations of the dopa NEUROTRANSMITrERS mine system have been made using drugs that raise dopamine Dopamine does not function alone and behaviors that involve concentration and with drugs that block D2 receptors. For the dopamine are also regulated by other neurotransmitters (16). most part, drugs that increase dopamine concentrations (i.e., The dopamine system interacts with several neurotransmitters cocaine and amphetamine) when given acutely have been both in the mesencephalic area and in projection regions. Close shown to produce a widespread decrease in brain glucose interactions between dopamine and norepinephrine, serotonin, metabolism (187, 188). Acute administration of D2 receptor gamma-amino butyric acid (GABA), glutamate, acetylcholine, antagonists was shown to have a minimal effect on regional and CCK among others, have been documented (16). brain glucose metabolism in schizophrenic patients (189). In These interactions regulate dopamine function and are of contrast, acute administration in normal subjects relevance for understanding brain pathology as well as for the decreased metabolism in frontal and limbic cortices, in thala therapeutic drug development. For example, interactions of mus and in (190). Though not always consis dopamine with glutamate have been associated with schizo tent, most studies evaluating the effects of chronic treatment phrenia (1 71 ) and Parkinson's disease (23); those with seroto with D2 receptor antagonists have shown increases in striatal nm with schizophrenia (1 72) and those with the opiate system metabolism and relative decreases in frontal metabolism (191— with substance abuse (1 73). Understanding how the dopamine 198). The inability to differentiate changes in metabolism that system is modulated by other neurotransmitters is leading to the result from direct changes in dopamine levels from those which development of candidate drugs which exploit these interac reflect secondary adaptation processes limits the interpretation tions, e.g., glutamate antagonists for Parkinson's disease (1 74) of these findings. and serotonin antagonists in schizophrenia (1 75). The feasibility of studying interactions with MULtiPLE TRACERS STUDIES PET was first demonstrated in the baboon brain for the The relatively low dosimetry from positron-labeled tracers interaction between dopamine and acetylcholine (1 76). The permits injection of volunteers with more than one PET methodology takes advantage of the sensitivity of receptor radiotracer. Multiple tracer studies that measure glucose metab specific radioligands to competition with endogenous neuro olism and/or CBF in conjunction with specific dopamine transmitters in an analogous way to the strategy described to tracers, enable one to assess the functional significance of these measure dopamine responsivity. The difference is that instead dopamine elements. Such studies have been done to investigate of changing dopamine concentration with a drug that directly the relation between brain glucose metabolism and dopamine interacts with dopamine neurons, it uses drugs which affect D2 receptors in cocaine abusers. A significant correlation was specific neurotransmitters that are known to modulate dopa reported between dopamine D2 receptors and glucose metabo mine (158). Using this strategy PET has been used to evaluate lism in , cingulate gyrus and superior frontal the ability of acetylcholine (1 76,177), GABA (157), serotonin cortex (68). Lower values for D2 receptor concentration are

PET DOPAMINETRACERs•Volkow et al. 1249 associated with lower metabolism in these brain regions. In (33,46,202—204) but the goal is to separate the effects of Parkinson's disease studies that evaluated in the same subject transport from that of receptor binding. The parameter of 6-['8F]fluoroDOPA and brain glucose metabolism have also interest is k3 a Bmax or k3/k4 a Bmax/Kd where Bmax is the reported an association between the decreases in [‘8F]DOPA total receptor concentration and Kd is the equilibrium dis uptake in striatum and decreases in metabolic activity in frontal sociation constant; Bmax/Kd is sometimes referred to as the cortical areas (199,200). Multiple tracer studies also enable binding potential (202). General aspects of parameter esti investigation of the relationship between the different elements mation and modeling techniques used in PET have been of the dopamine system and may aid the evaluation of patients. reviewed (211—214). An example of such a study is shown in Figure 4 from a For reversible ligands, the distribution volume (DV) can Parkinson's disease patient who was evaluated with three be determined directly. The DV is the ratio of tissue radioli tracers: [I ‘C]d-threomethylphenidate for DAT; [‘‘C]raclopride gand concentration to plasma radioligand concentration at for D2 receptors; and FDG for regional brain glucose metabo equilibrium. lism. While this patient showed marked reductions in DAT and Alternatives to the methods that require explicit model in frontal metabolic activity; there were no differences in parameter determination are the model independent methods dopamine D2 receptors. The DAT/D2 ratio was therefore which are much easier to apply. In the case of irreversibly markedly reduced when compared with that ofcontrols. The use binding ligands, the graphical technique (Patlak plot) (207,208) of multiple tracers is slowly starting to emerge as a powerful has been extensively used. In this method a plot of tool to investigate the relations between functional and neuro fT @ chemical parameters in the normal and the diseased brain. ROI(T) I Cp(T) versus Cp(t)/Cp(T) Jo QUANTITATIVE MEASURES OF DOPAMINE PARAMETERSWITHPET becomes linear after some time with a slope Ki, referred to as the influx constant. Cp(T) is the plasma radioactivity of the The models used to describe the uptake and binding of PET labeled ligand and ROI(T) is the tissue radioactivity measured ligands are generally simple two- or three-compartment models by the PET camera at time T. K1determines the steady-state rate with at most four kinetic parameters even though the physio oftransfer ofthe ligand into the irreversible compartment which logical processes underlying the PET image are complex and is given by [email protected]@,a function of the number of free receptors involve, among other factors, delivery via capillary network, or binding sites, is also a function oftracer delivery (K,). In the diffusion to the binding site, competition with endogenous case of a very high rate of trapping (controlled by k3) K. K1 neurotransmitter, binding to plasma proteins and association and therefore contains little information about free receptor and dissociation with specific binding sites (of possibly more concentration, a situation referred to as flow limited. By than one kind). The models used are simply of necessity since reducing the rate of trapping, the sensitivity of K@to variations there are only two measurements: plasma radioactivity due to in free receptor concentration can be improved. This was done the labeled ligand and the radioactivity measured by the PET with deuterium-substituted L-deprenyl ([I ‘C]L-deprenyl-D2) camera which is the total radioactivity due to all processes for which the trapping rate was reduced over that of L-deprenyl occurring within the voxel. As a result only a limited number of in regions of high MAO B (145). Under these conditions Ki is model parameters can be uniquely determined (201 ). Three much less sensitive to changes in delivery and more sensitive to compartment models for D2 dopamine receptors have been changes in enzyme/receptor concentration. The condition of proposed (33,46,202—206). Alternatively there are model-inde being flow limited is a consequence of the rapid trapping and pendent methods for the analysis of PET data which do not irreversible nature of the binding and limits the ability to depend upon a particular model but only require that linear determine reliable measures of free receptors whether expressed first-order kinetics are applicable to the movement of the in terms of k3 of the three-compartment model or K. The slope radiotracer (207—210).Both approaches provide some measure of a plot of radioactivity in the receptor region to that in a which is a function of the free receptor concentration. nonreceptor region versus time as a measure of the number of A radioligand generally can be classified as either reversible free receptors has also been used (46,60). This method can be when there is uptake and loss of the ligand from tissue during derived from the Patlak analysis for irreversibly binding ligands the course of the experiment or irreversible, in which case the (46,215). The steady state (when this ratio becomes constant). ligand is taken up by receptors/enzyme and remains bound for Since the DV is a steady-state quantity, it is independent of the duration of the experiment. The three-compartment model blood flow (167,216). In terms of model parameters the DV is which has been applied to both types of radiotracers consists of given by three to four parameters. Two parameters control transport between plasma and tissue, K1 is the plasma to tissue influx K1/ @ constant and k2 is the tissue-to-plasma efflux constant. There DV = — (@i+ Bmax' @/Kd'1 Eq. 1 are one or two receptor parameters, k3, which is proportional to k2 the number of free receptors or binding sites and k4, the where ligand-receptor dissociation constant (in the case of irreversible — ligands k4 = 0) (33,46,203,206. ) The transport constants, as Bmax'1 _ (Bmax1 L L') 1 Eq. 2 @ they are generally used, are functions ofblood flow, permeabil Kd'1 Kd1 @l+NS ity, plasma protein binding (K1) and nonspecific binding (k2). In some models blood flow and nonspecific binding are consid Bmax, is the receptor concentration of receptor type i and Kd@ ered separately (202). Values for the model parameters can be is the equilibrium dissociation constant, NS refers to the ratio of determined by fitting the model to PET data or by relying on a kinetic constants for nonspecific binding (assumed to be suffi pseudo equilibrium to estimate receptor number from PET data ciently rapid that it is always in a steady state (202)), L is the directly (33,204). There are many different approaches and concentration of endogenous neurotransmitter and L' is the assumptions used in the determination of the model parameters concentration of receptors occupied with unlabeled ligand or even within the framework of the three-compartment model drug. The free receptor concentration is related to the DV. By

1250 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 •July 1996 taking the DV ratio of receptor region (ROl) to a nonreceptor inverse agonists stabilize R favoring the inactive state and region (NR), the dependence upon plasma protein binding antagonists do not perturb the equilibrium (225,226). The which is contained in K, is eliminated. It is assumed that the presence of constitutively active receptors was recently docu ratio oftransport constants is close to the same for both regions. mented for dopamine D5 receptors (227). Though the physio logical significance of these constitutive dopamine receptors is DV(ROI) E 3 not understood there is evidence from other receptor systems @ DV(NR) = @Bmax'1/Kd'1. q. that abnormal expression of constitutive receptors can lead to pathology as is the case for mutations in the rhodopsin receptor There are several methods for determining the DV. One of which lead to retinitis pigmentosa (228), and for mutations of them involves an infusion protocol to maintain a constant the thyrotropin-stimulating hormone receptor which are associ plasma level and the DV is directly determined from the ratio of ated with the presence of oncogenic lesions (229). Access to radioactivity in tissue to that in plasma (21 7). However, for experimental strategies with PET that would enable discrimi most PET studies the ligand is introduced as a bolus injection nation of R* from R would allow the assessment of the which requires a modeling technique to determine the DV. functional state of the receptors in the normal and the diseased Under these conditions the DV can be determined using a brain. The possibility that there is increased expression of two-compartment two-parameter model that does not explicitly constitutive receptors in diseases with overactivity of the contain receptor parameters (210). The DV can also be deter dopamine system, as has been postulated for schizophrenia, mined from the plot of merits investigation. The PET strategy described to measure (T CT dopaminergic responsivity after a pharmacological challenge @ I ROI(t)dt/ROI(T)vs Cp(t) could perhaps be useful to assess the functional state of Jo Jo receptors since it reflects not only changes in intrasynaptic which becomes linear after some time with a slope equal to the dopamine but also the ratio of R* to R for dopamine binds DV (209). predominantly to R*. Unfortunately this strategy is confounded The determination of absolute values for the receptor con by baseline measures, since in a state with increased expression centration Bmax has been reported for NMS (218) and of R@, dopamine would also be more effective in blocking raclopride (219), although with conflicting results. Such exper binding of the receptor radioligand. Another strategy which has iments require a loading dose of unlabeled ligand or drug to been proposed to measure the functional state of the receptors reduce the number of available receptors. However, for an with PET is to differentially evaluate binding of dopamine ‘8F-labeled congener of raclopride, the three-compartment receptor agonists from binding of dopamine antagonists (230). @ model failed to give consistent results for the ligand-receptor _& S_ dissociation (k4) under conditions of saturation compared to the The binding of dopamine to its receptors results in an high specific activity nonsaturation experiments (220). The intracellular response transmitted by second messenger mole authors postulate that the geometry of the synapse and the cules such as cAMP which in turn regulate enzymatic processes physical arrangement of receptors on the membrane are respon within the cell such as (16). Changes in second sible for the discrepancy. In spite of this failure the three messenger systems will therefore affect the functional effects of compartment model has proved to be useful in assessing the receptors on the cells expressing them and may be involved changes in receptor availability even though it may lack the in disease, i.e., affective disorders (231 ) and drug addiction complexity to reproduce the observed behavior of ligand (232). The pharmacological properties of certain therapeutic binding under some circumstances. agents, such as , appear to be in part mediated by their ability to change these second messenger systems (233). FUTURE DIRECTiONS Development of radiotracers for second messenger systems Radiotracer Design and Development would be an enormous advance because there are theoretical Advances made in radiotracer design and synthesis have reasons to expect that brain diseases and therapeutic interven played a major role in shaping the PET field as we know it tions will express themselves more clearly at the level of today. These advances have made it possible to create the second- and third-messenger systems. radiotracers required to investigate new biomolecular targets Fluorine-18-labeled 1,2-diacylglycerols are currently being and to improve the quantitative nature of PET measurements. investigated as PET ligands to image second-messenger sys Development of more selective radioligands for dopamine tems (234). New tracers targeted to specific elements of the receptor subtypes and radiotracers with a range of kinetic second-messenger systems may allow investigation of their properties to maximize information from specific experimental involvement in brain diseases. paradigms are important immediate goals. Also, new radiotrac Molecular Gene&s ers which have the sensitivity to probe small changes in Antisense nucleotides are finding applications in many fields synaptic dopamine, as might be expected to arise from behav and have the potential to map mRNA for various molecular ioral stimulation, and/or to assess baseline dopamine concen targets including elements of the dopamine system such as tration would also be an important advance in our understand receptors and transporters. The antisense nucleotide comple ing of the functional activity of the dopamine system. ments the coding strand of DNA and selectively binds to a Functional State of Receptors strand of specific mRNA. The possibility of using this tech The documentation of constitutively active receptors that nique in nuclear medicine has been investigated primarily to couple to G proteins without the need of an agonist (221—224) permit the evaluation of oncogenes amplified in cancer cells has led to the renewed consideration of the two-state receptor (235). Using PET to measure regional expression of mRNA in model for G-protein coupled receptors (225). The two-state brain would be predicted to be limited by difficulties in receptor model describes an equilibrium between active (R*) preparation of the required labeled oligonucleotides and deliv and inactive receptors (R) which is perturbed by the presence of ery of nucleotides into the brain as well as the relatively low receptor ligands; agonists stabilize R* favoring the active state, rate of trapping and diffusion in tissue. Dopamine plays an

PET DOPAMINETRACERs•Volkow et al. 1251 important role in regulating the functions of various peripheral ACKNOWLEDGMENT organs such as the heart and kidneys. The extent to which This work was supported by Department of Energy grant imaging of these peripheral organs with positron-labeled anti DE-ACO2-76CH00016 and National Institute of Drug Abuse sense probes is at all feasible is currently difficult to predict. grants ROIDAO6891 and RO1DA06278.

Higher Spatial Resolution REFERENCES Most studies measuring dopamine receptors and dopamine I. Goldman-RakicP.Circuitryof theprefrontalcortex.In: PlumF,ed.Handbookof transporters are done in the caudate and the putamen. However, psychology. Bethesda, MD: American Physiological Society; 1987:373—417. these striatal regions are known to be functionally as well as 2. LeMoalM, SimonH. Mesocorticolimbicdopaminenetwork:functionalandregula tory roles. Physiol Rev 1991;71:l55—234. neurochemically heterogeneous. It is anticipated that future 3. Koob OF, Bloom FE. Cellular and molecular mechanisms of drug dependence. PET instruments with higher spatial resolution and sensitivity Science 1988;242:715—723. 4. Kopin Ii. The pharmacology of Parkinson's disease therapy: an update. Ann Rev will enable measurements in specific striatal subregions, such as Pharmacol Toxicol 1993 ;32:467—495. the nucleus accumbens. 5. Creese I, Burt DR. Snyder 5H. Dopamine receptor binding predicts clinical and pharmacological potencies of antischizophrenic drugs. Science 1976;192:481—483. Integration with Functional MRI Studies 6. McGeer PL, McGeer EG, Suzuki iS. Aging and extrapyramidal function. Arch Neurol 1977;34:33—35. PET has benefited from advances in MRI predominantly by 7. Weeks PA, Brooks DJ. Positron emission tomography and central neurotransmitter the use of co-registration procedures that improve regional systems in movement disorders. Fundam Clin Pharmacol 1994;8:503—517. quantitation of PET images. Also, since it is difficult to locate 8. Volkow ND, Fowler J5. Neuropsychiatric disorders: investigation of schizophrenia and substance abuse. Semin Nuci Med 1992;22:254—267. some brain in PET images due to variability in their 9. Di Chiara 0, Morelli M, Acquas E, Carboni E. Functions of dopamine in the location and size between patients, the individual subject's MRI extrapyramidal and liinbic systems. Arzneim-Forsch Drug Rev 1992;42:231—237. can be used to identify them. Newer strategies are being 10. Willner P, Muscat R, Papp M, Sampson D. In: Willner P. Scheel-Kruger J, eds. The mesolimbicdopaminesystem:frommotivationto action. New York, NY: JohnWiley developed for fusion imaging in which the data from both andSons;1992:387—400. image modalities are fused to generate an image that has the 11. Glowinski J, Tassin JP, Thierry AM. The mesocortical-prefrontal dopaminergic neurons. TINS 1984:418—451. spatial resolution of MRI with the biochemical information 12. Selemon LD, Goldman Rackic PS. Longitudinal topography and interdigitations of from PET. The development of functional MRI that enables corticostriatal projections in the rhesus monkey. J Neurosci 1984;5:776—794. assessment of functional brain activation with a higher spatial 13. stuss DT, Benson DF. Thefrontallobes. New York, NY: Raven Press; 1986:12—38. 14. McGeer PL, Eccles JC, McGeer EG. and pharmacology of dopamine. and temporal resolution than that of PET offers the opportunity Molecular neurobiologyof the mammalian brain. New York, NY: Plenum Press; of incorporating both technologies for investigating the tempo 1987:294—308. ral dynamics of the functional brain responses to acute changes 15. Gartis PA, Ciolkowski EL, Pastore P, Wightman RM. Efflux of dopamine from the synaptic cleft in the nucleus accumbens ofthe rat brain. JNeurosci 1994;14:6084-6093. in dopamine (i.e., temporal relationship between the receptor 16. Jackson DM, Westlind-Danielsson A. Dopamine receptors: molecular biology, occupancy by an agonist and/or antagonist and its functional biochemistry and behavioral aspects. Pharmacol Ther 1994;64:291-369. 17. Fowler J5, WoIfAP. New directions in positron emission tomography. In: Bristol JA, consequences). ed. Annual reports in medicinal chemistry. 1989;24:277—286. 18. Christman DR, Hoyte RM, Wolf AP. Organic radiopharmaceuticals labeled with Other Clinical Applications isotopes of short half-life. I: dopamine-hydrochloride-1-' ‘C.J Nucl Med 1970;1 1: Though D2 and DAT radioligands are being used for the 474—478. 19. Chumpradit S. Kung MP, Billings J, Kung HF. Fluorinated and iodinated dopamine evaluation of patients with movement disorders and D2 radio agents: D2 imaging agents for PET and SPECT. J Med Chem 1993;36:221—228. ligands have been shown to be useful for evaluating receptor 20. Seeman P, Niznik H. Dopamine receptors and transporters in Parkinson's disease and blockade by antipsychotic agents, their clinical utility has not schizophrenia. FASEB J I990;4:2737—2744. 21. 5ibley DR. Monsma FJ. Molecular biology ofdopamine receptors. TrendsPharmacol been fully evaluated. For example, recent studies have docu Sci 1992;13:61—69. mented a role of dopamine in malignant neuronal cell prolifer 22. Kessler RM, Whetsell WO, Ansasi Ms. et al. Identification ofextrastriatal D2 receptors in postmortem human brain with [Wljepidepride. Brain Re.s 1993;609:237-243. ation and have shown significant concentrations ofDl receptors 23. starr Ms. Glutamate/dopamine D1ID2 balance in the and its relevance in human brain meningiomas (236). Similarly, the presence of to Parkinson's disease. Synapse 1995;19:264—293. dopamine receptors, which are expressed on certain pituitary 24. Camps M, Cofles R, Gueye B, Probst A, Palacios JM. Dopamine receptors in human brain autoradiographic distribution of D2 sites. Neuroscience 1989;28:275—290. adenomas and which may be related to the cellular origins of 25. Hall H, Sedvall 0, Magnusson 0, Kopp J, Halldin C, Farde L. Distribution ofDI- and the neoplastic cells, has been detected by the dopamine D2 D2-dopamine receptors, and dopamine and its metabolites in human brain. Neuro , ‘‘CNMS (237). Thus, some of the dopa 1994; 11:245—256. 26. Murray AM, Ryoo H, Joyce iN. vis@1i@ation of dopamine D3-like receptors in mine receptor radioligands may be useful in the evaluation of human brain with [‘251]epidepride.Eur J Pharmacol 1992;227:443—445. abnormal cell proliferation in the human brain. 27. Seeman P, Guan H, Van Tol H. Dopamine D4 receptors elevated in schizophrenia. Nature 1993;365:441—445. 28. Strange PG. Dopamine D4 receptors: curiouser and curiouser. Trends Pharmacol Sci CONCLUSION 1994;15:3 17—319. PET, in conjunction with appropriate radiotracers, is being 29. Meador-Woodruff JH, Mansour A, et al. Distribution of D5-dopamine receptor mRNA in rat brain. Neurosc Leti l992;145:209—212. used to assess the dopamine system in the normal and the 30. Wagner HN, Burns HD, Dannals RF, et al. Imaging DA receptors in the human brain diseased human brain. As a result, information which could by PET.Sciencel983;221:1264—1266. 31. Ameft CD, Wolf AP, Shiue CY, et al. Improved delineation of human dopamine only be previously investigated in animals or in postmortem receptors using [‘8F]-N-methylspiroperidol and PET. J Nuci Med 1986;27: I878— human brains is accessible in living human subjects. This has 1882. enabled initial investigations of the relation between changes in 32. Maziere B, Loc'h C, Hantraye P. et al. 76Br-bromospiroperidol: a new tool for quantitative in vivo imaging of neuroleptic receptors. Life Sci 1984;35:1349—1356. the dopamine system in the human brain and its functional and 33. Farde L, Hall H, Ehrin E, Sedvall 0. Quantitative analyses ofD2-dopamine receptor clinical consequences. At the same time these studies have been binding in the living human brain by positron emission tomography. Science able to assess the effects of drugs in the dopamine system. l986;231:258—260. 34. Halldin C. Dopamine receptor radioligands. Med Chem Res 1994;5:l27—149. Future advances in radiotracer development and in instrumen 35. Seaman P. Guan C, Niznik HB. Endogenous dopaminc lowers the dopamine D2 tation will enable performance of more selective measurements receptor density as measured by 3H raclopride: implications for positron emission tomography of the human brain. Synapse 1989;3:96—97. with respect to the parameters investigated as well as the brain 36. Wang G-J, Volkow ND, Logan J, et al. Evaluation ofage-related changes in serotonin regions assessed. Such studies will increase our understanding 5-HT2 and dopamine D2 receptor availability in healthy human subjects. Life Sci of the mechanisms by which the dopamine system regulates 1995;56:249—253. 37. VanTol HM, Bunzow JR. Guan C, et al. Cloning of the gene for a human dopamine motor, cognitive and motivational behaviors in the human D4 receptor with high affinity for the antipsychotic clozapine. Nature 1991;350:610— brain. 614.

1252 THE JOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 •July 1996 38. Huntley OW, Morrison JH, Prikhozhan A, 5ealton SC. Localization of multiple disorders measured by positron emission tomography. Psychopharmacology 1992; dopamine receptor subtype mRNAs in human and monkey motor cortex and striatum. 106:14—18. Mo! BrainRes 1992;15:181—188. 72. Caldecott-Hazzard S. Morgan DO, Dc Leon-Jones F, et al. Clinical and biochemical 39. Filloux F, Dawson TM, Wamsley 1K. Localization ofnigrostriatal dopamine receptor aspects of depressive disorders. II. Transmitter/receptor theories. Synapse 1991;9: subtypes and adenylate cyclase. Brain Res Bull 1988;20:447—459. 251—301. 40. JoyceThi,MarshallJF. Quantitativeautoradiographyof dopamineD2 sitesin rat 73. BrownAS, Gershon5. Dopamineand depression.J Neural Transm[Gen Sect] caudate-putamen: localization to intrinsic neurons and not to neocortical afferents. 1993;91 :75—109. Neurosci 1987;20:773—794. 74. AquiloniusSM. WhathasPETtoldusaboutParkinson'sdisease?ActaNeurolScand 41 . Farde L, Nordstrom A-L, Karisson P. Halldin C, Sedvall 0. Positron emission 1991;136:37—39. tomography. Studies on dopamine receptors in schizophrenia. Clin Neuropharmacol 75. Rinne UK, Laihinen A, Rinne JO, et al. Positron emission tomography demonstrates 1995;l8:512 1—S129. dopamine D2 receptor supersensitivity in the striatum of patients with early 42. Wang 0-i, Volkow ND, Fowler iS, et al. Comparison of two PET radioligands for Parkinson's disease. Mov Disord l990;5:55—59. imaging extrastriatal dopamine receptors in human brain. Synapse 1993;15:246—249. 76. Hassan MM, Thakar I. Dopamine receptors in Parkinson's disease. Prog Neuropsy 43. Yousef K, Volkow ND, Schlyer D, et al. Inhibition of extrasttiatal dopamine chopharmacol Bio! Psychiatry 1988; 12:173—182. receptors by haloperidol. Synapse 1995; 19:14—17. 77. Rinne JO, Laihinen A, Lannberg P, Maramaki P, Rinne UK. A postmortem study on 44. Kessler RM, Whetsell WO, Ansari MS. et al. Identification ofextrastriatal dopamine striatal dopamine receptors in Parkinson's disease. Brain Res 1991;556: I 17—122. D2 receptors in postmortem human brain with [‘251]epidepride.Brain Res 1993;609: 78. SchwartzJ,TatschK, Arnold0, Ctal. ‘23I--5PECTpredictsdopami 237—243. nergic responsiveness in patients with de novo . Neurology 1992;41: 45. Halldin C, Farde L, Hogberg T, et al. Carbon-I 1-FLB 457: a radioligand for 556—561. extrastriatal D2 dopamine receptors. J Nuc! Med 1995;36:1275—128l. 79. Leenders K, Frackowiak R, Quinn N. Brain energy metabolism and dopaminergic 46. Wong DF, Gjedde A, Wagner HN. Quantification of neuroreceptors in the living functions in Huntington's disease measured in vivo using positron emission tomog human brain. I. Irreversible binding of ligands. J Cereb Blood Flow Metab raphy. Mov Disord 1986;1:69—77. 1986;6: 136—146. 80. Hagglund J, Aquilonius S-M, Eckernas S-A, et al. Dopamine receptor properties in 47. Farde L, Halidin C, 5tone-Elander 5, Sedvall 0. PET analysis of human dopámine Parkinson's disease and Huntington's chores evaluated by positron emission tomog receptor subtypes using ‘‘C-sCH23390 and ‘‘Craclopride. Psychopharmacology raphy using ‘‘C-N-methylspiperone.Ada Neurol Scand l987;75:87—94. 1987;92:278—284. 81. Brandt I, Foistein 5, Wong D, et al. D2 receptors in Huntington's disease: positron 48. Halldin C, Farde L, Barnett A, Sedvall 0. Synthesis of carbon-l I-labeled SCH emission tomography findings and clinical correlates. J Neuropsychiatry C/in 39166, a new selective dopamine Dl receptor ligand, and preliminary PET investi Neurosci 1990;2:20—27. gations. App! Radiat Isot 199 1;42:45 1—455. 82. Sedvall 0, Karlsson P. Lundin A, et al. Dopamine Dl receptor number—a sensitive 49. Karlsson P, Farde L, Halldin C, et al. PET examination of [‘‘C]NNC687 and PET marker for early brain degeneration in Huntington's disease. Eur Arch [I ‘C]NNC 756 as new radioligands for the DI-dopamine receptor. Psychopharma Psychiatry C/in Neurosci l994;243:249—255. cology1993;113:149—156. 83. Ritz MC, Lamb RI, GoldbergSR. KuharMi. Cocainereceptorson dopamine 50. HalldinC, FogedC, LochC, et al. [‘‘C]NNCI 12,a selectivePETradioligandfor transporters are related to self administration of cocaine. Science 1987;237:1219— examination of extrastriatal dopamine Dl receptors. J Nuci Med 1994;35:l22P. 1223. 51. Da5ilva iN, Wilson AA, 5eeman P. Houle S. Synthesis of [‘‘C]SKF75670 as a 84. Salmon E, Brooks Di, Leenders KL, et al. A two-compartment description and potential dopamine Dl receptor agonist imaging agent for PET. J Labeled Compd kinetic procedure for measuring [I ‘C]nomifensineuptake using positron emission Radiopharm 1994;35:460—461. tomography. J Cereb Blood Flow Metab I990; 10:307—317. 52. Wong DF, Wagner HN ir, Dannals RF, et al. Effects of age on dopamine and 85. TedroffI, AquiloniusS-M, HartvigP.etal. Monoaminereuptakesitesinthehuman @ serotonin receptors measured by positron emission tomography in the living brain. brain evaluated in vivo by means of ‘Cnomifensine and positron emission Science I986;226: I393—1396. tomography: the effect of age and Parkinson's disease. Acta Neurol Scand 1988;77: 53. Baron JC, Maziere B, Loch C, et al. Loss of striatal [76Br]bromospiperone binding 92—101. sites demonstrated by positron emission tomography in progressive supranuclear 86. Fowler iS, Volkow ND, WolfAP, et al. Mapping cocaine binding sites in human and palsy. J Cereb Blood Flow Metab 1986;6:131—136. baboon brain in vivo. Synapse I989;4:371—377. 54. Antonini A, Leenders KL, Reist H, et al. Effect of age on D2 dopamine receptors in 87. Kilbourn MR. In vivo binding of [“F]GBR 131 19 to the brain dopamine uptake normal human brain measured by positron emission tomography and ‘‘Craclopride. system. Life Sci 1988;42:1347—1351. Arch Neurol 1993;50:474—480. 88.KilbournMR. HakaMS, MulhollandOK, JewettDM, Kuhl D. Synthesisof 55. Rinne JO, Hietala J, Ruotsalainen U, et al. Decrease in human striatal dopamine D2 radiolabeled inhibitors of presynaptic monoamine uptake systems: [‘‘F]GBR13119 receptor density with age: a PET study with ‘‘Craclopride. J Cereb Blood Flow (DA), [‘‘C](NE) and [‘‘C](5-HT). J Lab Cmpd Radiopharm Metab 1993;13:310—314. 1989;26:4l2—414. 56. 5eeman P, Bzowe NH, Gum HG, et al. Human brain dopamine receptors in children 89. KoeppeRA, KilbournMR. FreyKA, et al. Imagingandkineticmodelingof andagingadults.Synapse1987;l:399—404. [‘8FJGBR12909, a dopamine uptake inhibitor. J NucI Med 1990;31 :720. 57. Suhara T, Fukuda H, lnoue 0, et al. Age-related changes in human Dl dopamine 90. Innis RB, Seibyl JP, Scanley BE, et al. Single-photon emission computed tomo receptors measured by positron emission tomography. Psychopharmacology 1991; graphic imaging demonstrates loss of striatal dopamine transporters in Parkinson's 103:41—45. disease. Proc Nat! Acad Sci USA I993;90: 11965—11969. 58. Andreasen NC, Carson R, Diksic M, et al. Workshop on schizophrenia, PET and the 91. Laihinen AO, Rinne JO, Nagren KA, et al. PET studies on brain monoamine dopamine D2 receptors in the human neostriatum. Schiz Bull 1988;14:471—484. transporters with carbon-I 1-f3-CIT in Parkinson's disease. J Nuc! Med l995;36: 59. FardeL, NordstromA-L, WieselFA, etal.Positronemissiontomographyanalysisof 1263—1267. central DI and D2 receptor occupancy in patients treated with classical neuroleptics 92. Madras BK, Spealman RD. Fahey MA, et al. Cocaine receptors labeled by and clozapine. Relation to extrapyramidal side effects. Arch Gen Psychiatry 1992; 2f3-carbomethoxy-3@3-(4-fluorophenyI)tropane. Mo! Pharm I989;36:5 I8—524. 49:538—544. 93. Wong DF, Yung B, Dannals RF, et al. In vivo imaging of baboon and human 60. 5mith M, WolfAP, Brodie ID, et al. Serial ‘8F-N-methylspiroperidol PET studies to dopamine transporters by positron emission tomography using (‘‘C]WIN35,428. measure changes in antipsychotic drug D2 receptor occupancy in schizophrenic Synapse1993;15:130—142. patients. Biol Psychiatry 1988;23:653—663. 94. Frost ii, Rosier AJ, Reich SO, et al. Positron emission tomographic imaging of the 61. Farde L, Nordstrom A, Nyberg 5, Halldin C, 5edvall 0. Dl, D2 and 5HT2 occupancy dopamine transporter with ‘‘C-WIN 35,428 reveals marked declines in mild in clozapine treated patients. J Clin Psychiatry 1994;55(suppl B):67—69. Parkinson's disease. Am J Neuro! 1993;34:423—431. 62. Wolkin A, Brodie JD, Rotrosen I, et al. Dopamine receptor occupancy and plasma 95. Ding Y-S, Fowler IS, Volkow ND, et al. Pharmacokinetics and in vivo specificity of haloperidol levels. Arch Gen Psychiatry 1989;46:482—483. [I ‘C]d-threo-methylphenidate for the presynaptic dopaminergic neuron. Synapse 63. Cambon H, Baron IC, Boulenger JP. In vivo assay for neuroleptic receptor binding 1994;18:152—160. in the striatum positron tomography in humans. BrJ Psychiatry 1987;15l:824—830. 96. Volkow ND, Ding YS, Fowler IS, et al. [‘‘C]d-threo-methylphenidate: a new PET 64. Carlsson A. Towards a new understanding of dopamine receptors. Clin Neurophar ligand for the dopamine transporter. II. Studies in the human brain. J Nuc! Med macol 1995;18:S6—S13. 1995;36:2162—2168. 65. Wolkin A, Barouche F, Wolf AP, et al. Dopamine blockade and clinical response: 97. Aquilonius 5-M, Bergstrom K, Eckernas 5A, et al. In vivo evaluation of striatal @ evidence for two biological subgroups of schizophrenia. Am J Psychiatiy 1989;46: dopamine reuptake sites using ‘C-nomifensine and positron emission tomography. 905—908. Acta Neurol Scand 1987;76:283—287. 66. Coppens W, 51off Ci, Paans AMJ, et al. High central D2 dopamine receptor 98. Volkow ND, Fowler iS, Logan I, Ct al. Comparison of [‘‘Cicocainebinding at occupancy as assessed with positron emission tomography in medicated but therapy sub-pharmacological and pharmacological doses: a PET study. J Nuc! Med I995;36: resistant schizophrenic patients. Biol Psychiatry 1991;29:629—634. 1289—1297. 67. Volkow ND, Fowler IS, Wolf AP, et al. Effects of chronic cocaine abuse on 99. Kilboum MR. DaSilva IN, Frey KA, Koeppe RA, KuhI DE. In vivo imaging of postsynaptic dopamine receptors. Am J Psychiatry 1990;l47:719—724. vesicular monoamine transporters in human brain using [‘‘C]tetrabenazineand 68. Volkow ND, Fowler 15, Wang 0-i, et al. DecreaseddopamineD2 receptor positron emission tomography. J Neurochem 1993;60:23 15—2318. availability is associated with reduced frontal metabolism in cocaine abusers. Synapse 100. Kilbourn MR. Shades of grey: radiopharmaceutical chemistry in the 1990s and 1993;14:169—177. beyond. NucI Med Biol l992;19:603—606. 69. Hietala I, West C, Syvalahti E, et al. Striatal D2 dopamine receptor binding 101. Volkow ND, Fowler 15, Wang 0-i, et al. Decreased dopamine transporters with age characteristics in vivo in patients with dependence. Psychopharmacology in healthy human subjects. Ann Neurology l994;36:237—239. 1994;1 16:285—290. 102. Van Dick CH, Seibyl iP, Malison RT, et al. Age-related decline in dopamine 70. PearlsonGD, Wong DF, Tune LE, et al. In vivo dopaminereceptordensityin transporter binding in human striatum with [l-l23J@3-CIT SPECT. J Nucl Med psychotic and nonpsychotic patients with . Arch Gen Psychiatry l995;36:1175—1181. 1995;52:471—477. 103. Zelnik N, Angel I, Paul SM, Kleinman JE. Decreased density of human striatal 71. Suhara 1, Nakayama K, Inoue 0, et al. Dl dopamine receptor binding in mood dopamine uptake sites with age. Eur J Pharmacol l986;126:175—176.

PET DOPAMINETRACERS•Volkow et al. 1253 104. DeKeyserI, EbingerG, Vauquelin0. Age-relatedchangesin thehumannigrostriatal 139. Strolin, Benedetti M, Dostert P. Monoamine oxidase: from physiology and patho dopaminergic system. Ann Neurol 1990;27:157—l6l. physiology to the design and clinical application of reversible inhibitors. Adv Drug 105. Malison RT, Wallace EA, Best 5, et al. SPECT imaging ofdopamine transporters in Research1992;23:67—125. cocaine dependent and healthy control subjects with [‘23I]@3CIT.Soc Neurosci Abstr 140. Birkmayer W, Knoll J, Youdim MBH, et al. Increased life expectancy resulting from 1994;20:1625. addition of L-deprenyl to L-DOPA treatment in Parkinson's disease: a Iongterm 106. Volkow ND, Wang 0-i, Fowler IS, et al. Cocaine binding is decreased in the brain study.JNeural Transm 1985;64:l13—127. of detoxified cocaine abusers. J Neuropsychopharmaco! 1996;14:159—168. 141. Tetrud JW, Langston 1W.The effect ofdeprenyl (selegeline) on the natural history of 107. Tiihonen I, Kuikka I, Bergstrom K, et al. Altered striatal dopamine reuptake site Parkinson's disease. Science l989;245:5l9—522. densities in habitually violent and nonviolent alcoholics. Nature Medicine 1995;l: 142. Olanow CW, Calne D. Does selegeline monotherapy in Parkinson's disease act by 654—657. symptomatic or protective mechanisms? Neurology l99l;41 (suppl 4):l3—26. 108.ZhuM-Y,JuorioAV. AromaticL-aminoaciddecarboxylase:biologicalcharacter 143. Fowler IS, MacGregor RR, Wolf AP, et al. Mapping human brain monoamine ization and functional role. Gen Pharmaco! 1995;26:681—696. oxidase A and B with ‘‘C-suicideinactivators and positron emission tomography. I09. Cumming P, Venkatachalam 1K, Rajagopal 5, Diksic M, Ojedde A. Brain uptake of Science 1987;235:48 1—485. a-['4C]methyl-para-tyrosine in the rat. Synapse l994;l7:125—l28. 144. Shinotoh H, Inoue 0, Suzuki K, et al. Kinetics of [‘‘C]N,N-dimethylphenethylamine I 10. Deiesus 01, Murali D, Kitchen R, et al. Evaluation of 3['8F]fluoro-a-fluoromethyl in mice and humans: potential for measurement ofbrain MAO B activity. J NucI Med p-tyrosine, a potential PET tracer for tyrosine hydroxylase activity. Nuc/ Med Bio! 1987;28:I006—lOl 1. 1994;2l:663—667. 145. Fowler 15, Wang 0-i, Logan J, et al. Selectivereductionof radiotracertrappingby I 11. Garnett ES, Firnau 0, Nabmias C. Dopamine visualized in the basal ganglia of living deuterium substitution: comparison of [‘‘C]L-deprenyland [‘‘C]L-deprenyl-D2 for man. Nature 1983;305:137—l38. MAO B mapping. JNuclMed 1995;36:1255—1262. I 12. Garnett Es, Fimau 0, Chan PKH, Sood 5, Belbeck LW. [“F]Fluoro-dopa,an analog 146. Fowler J5, Volkow ND, Logan I, et al. Slow recovery ofhuman brain MAO B after of dopa, and its use in direct external measurements of storage, degradation and L-Deprenyl (selegeline) withdrawal. Synapse I994; 18:86—93. turnover of intracerebral dopamine. Proc Nat Acad Sci USA 1978;75:464—467. 147. Bench Ci, Price OW, Lammertsma AA, et al. Measurement of human cerebral 113. Hoshi H, Kuwabara H, Leger 0, Cumming P, Guftman M, Ojedde A. 6-[―F]Fluoro (MAO B) activity with positron emission tomography (PET): L-dopa metabolism in living human brain: a comparison ofsix analytical methods. J a dose ranging study with the reversible inhibitor Ro 19 6327. EurJ Clin Pharmacol Cereb B!ood Flow Metab 1993; 13:57—69. 1991;1O:l69—l73. I 14. Dhawan V, Ishikawa T, Chaley T, et al. Combined FDOPA and 3OMFD PET studies 148. Fowler 15, Volkow ND, Logan I, et al. MAO B inhibition therapy in Parkinson's in Parkinson's disease: modeling issues. J Nuc! Med l996;in press. disease: the degree and reversibility of human brain MAO B inhibition by RO 115. Melega WP, Hoffman JM, Luxen A, et al. The effects ofcarbidopa on the metabolism 196327.Neurology1993;42:1984—1992. of 6-[―F]fluoro-L-dopain rats, monkeys and humans. Life Sci l990;47:l49—l57. 149. Guldberg HC, Marsden CA. Catechol-O-methyltransferase: pharmacological aspects 116. Gjedde A, Reith I, Dyve 5, et al. Dopa decarboxylase activity in the living human and physiological role. Pharmacol Rev 1975;27:135—206. brain. Proc Nat Acad Sci USA 1991;88:2721—2725. 150. Mannisto TP, Kaakkola S. New selective COMT inhibitors: useful adjuncts for 117. Hartvig P, Lindner KJ, TedroffJ, et al. Regional brain kinetics of6-fluoro-(@-' ‘C) Parkinson's disease? TIPS 1989;I0:34—36. L-dopa and (a-' ‘C)-L-dopafollowing COMT inhibition. A study in vivo using 151. Wood PL, Altar CA. Dopamine release in vivo from nigrostriatal mesolimbic and positron emission tomography. J Neural Transm 1992;87:15—22. mesocortical neurons: utility of 3-methoxytyramine measurements. Pharmacol Rev 118. Melega WP, Luxen A, Perlmutter MM, et al. Comparative in vivo metabolism of 1988;40:l63—187. 6-[―F]fluoro-L-dopa and [3H]L-dopa in rats. Biochem Pharm 1990;39:1853—l860. 152. Axelrod I. The 0-methylation ofepinephrine and other in vitro and in vivo. 119. Hartvig P, Agren H, Reibring L, et al. Brain kinetics of L-[@-'‘C]DOPAin humans Sciencel957;126:1657—l660. studied by positron emission tomography. J Neural Transm 199l;86:25—41. 153. Mannisto PT, Kaakkola 5, Nissinen E, et al. Potent, selective and orally active 120. Agren H, Reibring L, Hartvig P. Ctal. Monoamine metabolism in human prefrontal inhibitors of catechol-O-methyltransferase. L@feSci l988;32: 1465—1471. cortex and basal ganglia. PET studies using [a-' ‘C]L-5-hydroxytryptophan and 154.DingYS,GatleySI, FowlerIS, et al. Mappingcatechol-O-methyltransferase(in [p-I ‘C]L-DOPA in healthy volunteers and patients with unipolar depression. Dc vivo—initial studies with [“F]Ro41-0960. Life Sci 1995;58:195—208. pressionl993;l:7l—8l. 155. Friedman AM, Deiesus 01, Revenaugh I, et al. Measurements in vivo of parameters 121. Snow BJ, Tooyama I, McGeer EG, et al. Human positron emission tomographic ofthe dopamine system. Ann Neurol l984;15:566—576. [“F]fiuorodopastudies correlate with dopamine cell counts and levels. Ann Neurol 156. Ross SB, Jackson DM. Kinetic properties ofthe accumulation of3H raclopride in the 1993;34:324—330. mouse in vivo. Arch Pharmaco! l989;340:6—12. 122. Eidelberg D, Takikawa 5, Dhawan V, Ctal. Striatal “F-DOPAuptake: absence of an 157. Dewey SL, Smith OW, Logan J, et al. GABAergic inhibition of endogeneous aging effect. J Cereb Blood Flow Metab l993;l3:88l—888. dopamine release measured in vivo with ‘‘C-racloprideand positron emission 123. Martin WRW, Palmer MR. Patlak CS, Calne DB. Nigro-striatal function in man tomography. J Neurosci 1992;12:3773—3780. studied with positron emission tomography. Ann Neurol 1989;26:535—542. 158. Dewey SL, Smith 05, Logan I, et al. Striatal binding of the PET ligand ‘‘C- 124. Sawle GV, Colebatch JO, Shah A, et al. Striatal function in normal aging: raclopride is altered by drugs that modify synaptic dopamine levels. Synapse implications for Parkinson's disease. Ann Neurol l990;28:799—804. 1993;13:350—356. 125. Kish 51, Zhong XH, Hornykiewicz 0, et al. Striatal 3,4-dihydroxyphenylalanine 159. Hume 5P, Myers R, Bloomfield PM, et al. Quantitation ofcarbon-l I-labeled raclopnde decarboxylase in aging: disparity between postmortem and positron emission tomog in rat striatum using positron emission tomography. Synapse l992;12:47—54. raphy studies. Ann Neurol 1995;38:260—264. 160. Volkow ND, Wang G-J, Fowler IS, et al. Imaging endogenous dopamine competition 126. HolthoffVA, Kessler I, Pietrzyk U, et al. Motor and cognitive deficits in Parkinson's with [“C]raclopridein the human brain. Synapse 1994;l6:255—262. disease are related to striatal dopamine uptake [Abstract]. J Cereb Blood F/ow Metab 161. Dewey SL, Logan I, Wolf AP, et al. Amphetamine induced decreases in [“F] 1993;13:S253. N-methylspiroperidol binding in baboon brain using positron emission tomography 127. Deiesus OT, Mukhergee J. Radiobrominated m-tyrosine analogs as potential CNS (PET). Synapse 1991;7:324—327. L-dopaPET tracers.BiochemBiophysRetCommun 1988;l50:l027—lO3l. 162. Logan I, Dewey SL, WoIfAP, et al. Effects ofendogenous dopamine on measures of 128. Melega WP, Perlmutter MM, Luxen A, et al. 4-['8F]Fluoro-L-m-tyrosine: An [“F]N-methylspiroperidol binding in the basal ganglia: comparison of simulations L-3,4-dihydroxyphenylalanine analog for probing presynaptic dopaminergic function and experimental results from PET studies in baboons. Synapse 1991;9:195—207. with positron emission tomography. J Neurochem 1989;53:31 1—314. 163. Nordstrom AL, Farde L, Pauli 5, et al. PET analysis of central [‘‘C]raclopride 129. Nahmias C, Wahl L, Chirakal R, et al. A probe for intracerebral aromatic amino-acid binding in healthy young adults and schizophrenic patients, reliability and age effects. decarboxylase activity: distribution and kinetics of [“F]6-fluoro-L-m-tyrosine in the Human Psychopharmacol 1992;7: I57—165. human brain. Mov Disord l995;lO:298—304. 164. Volkow ND, FowlerlS, Wang 0-i, et al. Reproducibility ofrepeated measures of' ‘C 130. DeiesusOT, Holden JE, EndresC, Ctal. Visualization of dopaminenerve terminals raclopride binding in the human brain. J Nuc! Med l993;34:609—613. by positron tomography using [“F)[email protected] Rev 165. Laruelle M, Abi-Dargham A, van Dick C, et al. SPECT imaging ofstriatal dopamine 1992;597:l5l—154. release after amphetamine challenge. J Nuc! Med 1995;36:l 182—1190. 131. Berry MD, Juorio AV, Paterson IA. The functional role of monoamine oxidases A 166. Wang 0-i, Volkow ND, Fowler iS, et al. Methylphenidate decreases regional and B in the mammalian central . Prog Neurobiol l994;42:375—391. cerebral blood flow in normal human subjects. L@feSci l994;54:1433—1446. 132. Breakefield XO, Chen Z-Y, Tivol E, Shalish C. Molecular genetics and inheritance 167. Logan J, Volkow ND, Fowler IS, et al. Effects of blood flow on [‘‘C]raclopride of human monoamine oxidases A and B. Neuro! Dis Ther 1994;21:95—l12. binding in the brain: model simulations and kinetic analysis of PET data. J Cereb 133. Youdim MBH, RiedererP. Dopamine metabolismand neurotransmissionin primate Blood Flow Metab 1994;14:995—l0l0. brain in relationship to and B inhibition. J Neural Transm 168. Inoue 0, Tsukada H, Kobayashi K, et al. Swim stress alters in vivo binding of l993;9l:18l—195. [3H]N-methylspiperone. Neuropharmacol 1991 ;30: 1101—1106. 134. O'Carroll M, Fowler Ci, Phillips JP, et a!. The deamination of dopamine by human 169. Deterding LI, Dix K, Burka LT, Tomer KB. On-line coupling ofin vivo microdialysis brain monoamine oxidase. Specificity for the two enzyme forms in seven brain with tandem mass spectrometry. Anal Chem 1992;64:2636-264l. regions. Arch Pharmacol 1983;322: 198—202. 170. Gifford AN, Gatley 51, Ashby CR. Endogeneously released dopamine inhibits the 135. Johnston JP. Some observations upon a new inhibitor ofmonoamine oxidase in brain binding of dopaminergic PET and SPECT uganda in superfused rat striatal slices. tissue. Biochem Pharmacol 1968;l7:1285—l297. Synapse1990;22:232—238. 136. Knoll I, Magyar K. Some puzzling effects of monoamine oxidase inhibitors. Adv 171. Riederer P. Lange KW, Kornhuber I, et al. Glutamatergic-dopaminergic balance in Biochem Psychopharmaco/ l972;5:393—408. the brain: its implications in motor disorders and schizophrenia. Arzneim-Forsch 137. Oreland L, Arai Y, Stenstrom A, Fowler Ci. Monoamine oxidase activity and 1992;42:265—268. localization in the brain and the activity in relation to psychiatric disorders. Mod Prob 172. Meltzer H. Clinical studies on the mechanism of action of clozapine: the dopamine Pharmacopsychiafty 1983; 19:246—254. serotonin hypothesis of schizophrenia. Psychopharmacol 1989;99:518—527. 138. RiedererP. KonradiC, SchayV. et al. LocalizationofMAO A andMAO B in human 173. Schaefer GJ. Opiate antagonists and rewarding brain stimulation. Neurosci Behav Rev brain: a step in understanding the therapeutic action of L-deprenyl. In: Yahr MD, 1988;l2:l—17. Bergman Ki, eds. Advances in Neurology. New York, NY: Raven Press, 1986:111— 174. Girault IA, Halpain 5, Greengard P. Excitatory amino acid antagonists and 118. son's disease. Trends Neurosci 1990;13:325—326.

1254 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 •July 1996 175. Stockmeier CA, DiCarlo Ii, Zhang Y, et al. Characterization oftypical and atypical from multiple-time uptake data. Generalization. J Cereb Blood Flow Metab 1985;5: antipsychotic drugs based on in vivo occupancy of serotonin-2 and dopamine-2 584—590. receptors. J Pharmacol Exp Ther l993;266: 1374—1384. 209. Logan I, Fowler 15, Volkow ND, et al. Graphical analysisof reversibleradioligand 176. Dewey 5L, Brodie ID, Fowler 15, et al. Positron emission tomography studies of binding from time-activity measurements applied to [N-' ‘C-methyl]-(—)-cocaine dopaminergic/cholinergic interactions in the baboon brain. Synapse l990;6:32l—327. PET studies in human subjects. J Cereb Blood Flow Metab l990;lO:740—747. 177. Dewey SL, Smith OS, Logan I, et al. Effects of central cholinergic blockade on 210. Koeppe RA, Holthoff VA, Frey KA, et al. Compartmentalanalysisof [‘‘C]fluma striatal dopamine release measured with positron emission tomography in normal zenil kinetics for the estimation ofligand transport rate and receptor distribution using human subjects. Proc NatI Acad Sci USA l993;90:l 1816—11820. positron emission tomography. J Cereb Blood Flow Metab 1991 ;l 1:735—744. 178. Dewey SL, Smith OS, Logan I, et al. modulation of striatal dopamine 21 1. Carson RE. Parameter estimation in positron emission tomography. In: Phelps M, measured with positron emission tomography and in vivo microdialysis. J Neurosci Mazziotta I, Schelbert H, eds. Positron emission tomography and autoradiography: 1995; 15:821—829. principles and applications for the brain and heart. New York, NY: Raven Press; 179. Smith 05, Dewey SL, Logan I, et al. Opiate modulation of striatal dopamine release I986:347—390. measured with positron emission tomography and ‘‘C-raclopride[Abstract]. Soc 212. Huang S-C, Barrio IR, Phelps ME. Neuroreceptor assay with positron emission NeurosciAbstrl993;19:128.9. tomography: equilibrium versus dynamic approaches. J Cereb Blood Flow Metab 180. Ingvar M, Stone-Elander 5, Rogers 0, et al. Striatal D2/acetylcholine interactions: I986;6:5I5—521. PET studies of the vesamicol receptor. NeuroReport l993;4:l311—13l4. 213. Zeeberg BR, Gibson RE, Reba RC. Accuracy of in vivo neuroreceptor quantification 181. Reith I, Benkelfat C, Sherwin A, et al. Elevated dopa decarboxylase activity in living by PET and review of steady-state, transient, double injection and equilibrium brain of patients with psychosis. Proc Nail Acad Sci USA 1994;9l :11651—11654. models. IEEE Trans Med Imaging l988;7:203—2l5. I82. Grace AA. Phasic versus tonic dopamine release and the modulation ofthe dopamine 214. Delforge I, Syrota A, Mazoyer BM. Experimental design optimization: theory and system responsivity. Neurosci 199l;4l :1—24. application to estimation of receptor model parameters using dynamic positron 183. Reivich M, KuhI D, Wolf AP, et al. The [“F]fluorodeoxyglucose method for the emission tomography. Phys Med Biol 1989;34(4):419—435. measurement oflocal cerebral glucose utilization in man. Circ Res l979;44:l27—137. 215. Logan I, Schlyer DI, Wolf AP, et al. D-2 receptor occupancy and 184.MatsM, FinkDI, GainerH.Activity-dependentenergymetabolismin ratposterior plasma concentration. Biol Psychiatry (Letter) 1990;28:1068—l070. pituitary primarily reflects sodium pump activity. J Neurochem l980;34:2l3—215. 216. Holthoff VA, Koeppe RA, Frey KA, et al. Differentiation of radioligand delivery and 185. Sokoloff L, Reivich M, Kennedy C, et al. The [‘4C]deoxyglucose method for binding in the brain: validation of a two-compartment model for [‘‘C]flumazenil.J measurement of local cerebral glucose utilization. J Neurochem l977;28:897—916. Cereb Blood Flow Metab 1991 ;I 1:745—752. 186. Pomno U, Crane AM. Metabolic mapping of the effects of drugs of abuse with the 217. Carson RE, Channing MA, Blasberg RO, Ct al. Comparison of bolus and infusion 2-['4C]deoxyglucose method. Modern methods in pharmacology, volume 6, testing methods for receptor quantitation: application to [“F]cyclofoxy and positron emis and evaluation ofdrugs ofabuse. New York, NY: Wiley-Liss; 1990:147—164. sion tomography. J Cereb Blood Flow Metab l993;l3:24—42. 187. Wolkin A, Angrist B, Wolf A, et al. Effects of amphetamine on local cerebral 218. Wong DF, Wagner HN Ir, Tune LE, et al. Positronemission tomography reveals metabolism in normal and schizophrenic subjects as determined by positron emission elevated D2-dopamine receptors in drug-naive schizophrenics. Science 1986:234: tomography. Psychopharmacol I987;92:24 1—246. 1558—1563. 188. London ED, Cascella NO, Wong DF, et al. Cocaine-inducedreductionof glucose 219. Farde L, Wiesel F-A, Stone-Elander 5, Ctal. D2-dopamine receptors in neuroleptic utilization in human brain. A study using positron emission tomography and naive schizophrenic patients. Arch Gen Psychiatry 1990;47:213—219. [fluorine-l8]-fluorodeoxyglucose. Arch General Psychiatry 1990;47:567—574. 220. Votaw iR, Kessler RM, Dc Paulis T. Failure of the three-compartmentmodel to 189.VolkowND,BrodieID, WolfAP,etal.Brainmetabolismin schizophrenicsbefore describe the pharmacokinetics in brain of a high affinity substituted benzamide. and after acute neuroleptic administration. J Neurol Neurosurg Psychiatry 1986;49: Synapse 1993;l5:177—190. I199—1202. 221. Samana P. Cottechia 5, Costa T, Lefkowitz Ri. A mutation-induced activated state of 190.BartlettEl, BrodieID, SimkowitzP. Ctal. Effectsof haloperidolchallengeon the @32-adrenergicreceptor. J Biol Chem l993;268:4625—4636. regional cerebral glucose utilization in normal human subjects. Am J Psychiatry 222. Costa T, Ogino Y, Munson P1, Ct al. Drug efficacy at guanine nucleotide-binding 1994;l51:681—686. regulatory protein-linked receptors: thermodynamic interpretation of negative antag 191. Wolkin A, iaeger I, Brodie ID, et aI. Persistence ofcerebral metabolic abnormalities onism and of receptor activity in the absence of ligand. Mo! Pharmacol l992;41: in chronic schizophrenia as determined by positron emission tomography. Am 549—560. J Psychiatry l985;142:564—571. 223. Black 1W, Shankley NP. Inverse agonists exposed. Nature 1995;374:214—215. 192. DeLisi LE, Holcomb HH, Cohen RM, Ct a]. Positron emission tomography in 224. Bond RA, Leff P, iohnson TD, et al. Physiologicaleffects of inverse agonistsin schizophrenic patients with and without neuroleptic medication. J Cereb B!ood Flow transgenic mice with myocardial overexpression of the f32-adrenoceptor. Nature Metab I985;5:20l—206. 1995;374:272—276. 193. BuchsbaumMS. Wu IC, DeLisi LE, et al. Positronemissiontomographystudiesof 225. Leff P. The two-state model of receptor activation. Trends Pharmacol Sci 1994; 15: basal ganglia and somatosensory cortex neuroleptic drug effects. Biol Psychiatry 408—409. 1987;22:479—494. 226. Milligan 0, Bond RA, Lee M. Inverse agonists: pharmacological or 194.SzechtmanH,NahmiasC,GarnettES,etal.Effectofneurolepticsonalteredcerebral potential therapeutic strategy? Trends Pharmaco! Sci 1995;16: 10—13. glucose metabolism in schizophrenia. Arch Gen Psychiatry l988;45:523—532. 227. Tiberi M, Caron MC. Constitutive activity of dopamine Dl receptor subtypes. C/in 195. Wik 0, Wiesel F-A, Sjorgen I, et al. Effects of sulpinde and chlorpromazineon Neuropharmaco! 1995;18:S43—548. regional cerebral glucose metabolism in schizophrenic patients as determined by 228. Robinson PR, Cohen GB, Zhukovsky EA, Oprian DD. Constitutively active mutants positron emission tomography. Psychopharmaco! 1989;97:309—318. of rhodopsin. Neuron 1992;9:719—725. 196. Buchsbaum MS. Potkin SO, Siegel BV Ir, et al. Striatal metabolic rate and clinical 229. Parma I, Duprez L, Van Sande I, et al. Somatic mutations in the thyrotropin receptor response to neuroleptics in schizophrenia. Arch Gen Psychiatry l992;49:966—974. gene cause hyperfunctioning adenomas. Nature 1993:365:649—651. 197. Bartlett El, Wolkin A, Brodie ID, Ctal. Importance ofpharmacologic control in PET 230. Yang Z-Y, Mukherjee I. Development of agonists as radiotracers for the assessment studies: effects of thiothixene and haloperidol on cerebral glucose utilization in of the functional state of the dopamine D2 receptors [Abstract]. J Nuc! Med chronic schizophrenia. Psychiatry Res 199 1;40: 115—124. 1995;36(suppl): l52P. 198. Holcomb HH, Cascella NO, Thaker OK, et al. Functional sites of neuroleptic drug 231.ManjiHK,Chen0, SimonH,Ctal.Guaninenucleotide-bindingproteinsin bipolar action in the human brain: PETIFDG studies with and without haloperidol. Am J affective disorder. Arch Gen Psychiatry l995;52:l35—144. PsychiatryI996;l53:41—49. 232. Nestler El. Molecular neurobiology of drug addiction. Neuropsychopharmacol 199. Eidelberg D, Moeller JR. Ishikawa T, et al. Early differential diagnosis of Parkinson's 1994;!1:77—88. disease with “F@@fluorodeoxyglucoseand positron emission tomography. Neurology 233.ManjiHK,PotterWZ,LenoxRH.Signaltransductionpathways.Moleculartargets l995;45: 1995—2004. for lithium's actions. Arch Gen Psychiatry 1995;52:531—543. 200. Eidelberg D, Moeller JR. Dhawan V. et al. The metabolic anatomy of Parkinson's 234. Takahashi T, Ido T, Ootake A, et al. [“F]Labeledl,2-diacylglycerols; a new tracer @ disease: complementary and ‘8F-fluorodopapositron emis for the imaging of [Abstract]. J Labeled Compds Radiop sion tomography studies. Mov Disord l990;5:203—2l3. harm 1994;35:5l7—5l9. 201. Swan IAA, Korf I. In vivo dopaminereceptorassessmentfor clinical studiesusing 235. Dewanjee MK, Ohafounpour AK, BootheT, et al. Gallium-67 labeling of antisense positron emission tomography. Biochem Pharmacol l987;36:2241—2250. deoxyoligonucleotide (GAASDON) probes and uptake in leukemic cells (P388) 202. Mintun MA, Raichle ME, Kilbourn MR. et al. A quantitative model for the in vivo [Abstract]. J Labe!ed Compd Radiopharm 1994;35:356—357. assessment of drug binding sites with positron emission tomography. Ann Neurol 236. Schrell UM, Nomikos P. Fahlbusch R. Presence of dopamine Dl receptors and 1984;l5:217—227. absence of dopamine D2 receptors in human cerebral meningioma tissue. J Neuro 203.WongDF,OjeddeA,WagnerHN,etal.Quantificationofneuroreceptorsintheliving surg l992;77:288—294. human brain. II. Inhibition studies of receptor density and affinity. J Cereb Blood 237. Muhr C, Bergstrom M, Lundberg P0, Ct al. Dopamine receptorsin pituitary Flow Metab 1986;6:147—153. adenomas: PET visualization with ‘‘C-N-methylspiperone. J Comp Assist Tomog 204. Farde L, Eriksson L, Blomquist 0, Halldin C. Kinetic analysis ofcentral [‘‘C]raclo 1986;l0:l75—180. pride binding to D2-dopamine receptors studied by PET—a comparison to the 238. Luabeya MK, Maloteaux IM, Laduron PM. Regional and cortical laminar distribu equilibrium analysis. J Cereb Blood Flow Metab l989;9:696—708. tions of serotonin 52, benzodiazepine, muscarinic and dopamine D2 receptors in 205. PerlmutterIS, Larson KB, Raichle ME, Ctal. Strategiesfor in vivo measurementof human brain. J Neurochem 1984;43:1068—1071. receptor binding using positron emission tomography. J Cereb B!ood Flow Metab 239.McElvainIS, Schenk10. A multisubstratemechanismof striataldopamineuptake 1986;6: 154—169. and its inhibition by cocaine. Biochem Pharmacol l992;43:2 189—2199. 206. Logan I, Wolf AP, Shiue C-Y, Fowler IS. Kinetic modeling of a receptor-ligand 240.MackayAVP,DaviesP.DewarCM,YatesCM.Regionaldistributionof enzymes binding applied to positron emission tomographic studies with neuroleptic tracers. associated with by monoamines, acetylcholine and GABA in the J Neurochem l987;48:73—83. human brain. J Neurochem 1978;30:827—839. 207. Patlak CS, Blasberg RG, Fenstennacher ID. Graphical evaluation of blood-to-brain 241. O'CarroIl AM, Anderson MC, Tobbia I, et al. Determinationof the absolute transferconstants from multiple-time uptake data.JCerebBloodFlowMetab l983;3:l—7. concentrations ofmonoamine oxidase A and B in human tissues. Biochem Pharmacol 208. Patlak CS, Blasberg RO. Graphical evaluation of blood-to-brain transfer constants 1989;38:901—905.

PET DOPAMINETRACERS•Volkow et al. 1255 242.HuhMM,FriedhoffAJ.Multiplemolecularformsof catechol-O-methyltransferase.249.KesslerRM,AnsariMS.SchmidtDE,et al. High-affinitydopamineD2receptor J Biol Chem 1979;254:299—308. radioligands. 2. [‘23ljepidepride,a potent and specific radioligand for the character 243. KastnerA, AngladeP. BounaixC, et al. lmmunohistochemicalstudyof catechol-O ization ofstriatal and extrastriatal dopamine D2 receptors. Life Sci 1991;49:6l7—628. methyltransferase in the human . Neurosci 1994;62:449—457. 250. Pellevoisin C, Chalon 5, Zouakia A, et al. Comparison oftwo radioiodinated ligands 244. AndersenPH,GronvaldFC, HohlwegR, et al. NNC-l 12,NNC-687andNNC-756: ofdopamine D2 receptors in animal models: iodobenzamide and iodocthylspiperone. new, selective and highly potent dopamine DI-receptor antagonists. Eur J Pharmaco! Life Sci 1993;52:185l—1860. I992;2I9:45—52. 251. LaruelleM, WallaceE, SeibylIP, et al. Graphical,kineticandequilibriumanalyses 245. Inoue0, KobayashiK, SakiyamaY, SuzukiT. Theeffectof benzodiazepineson the of in vivo [‘23I]f3-CITbinding to dopamine transporters in healthy human subjects. binding of[3H]SCH 23390 in vivo. Neuropharmacol 1992;31:ll5—12l. J Cereb Blood Flow Metab 1994; 14:982—994. 246. Farde L, Ehrin E, Enksson L, et al. 5ubstituted benzamides as ligands for visualization of dopamine receptor binding in the human brain by positron emission 252. Gatley SI, Ding V-S. Volkow ND, Ctal. Binding ofd-threo-[' ‘C]methylphenidateto tomography. Proc Nat! Acad Sci USA 1985:82:3863—3867. the dopamine transporter in vivo: insensitivity to synaptic dopamine. Eur J Pharma 247. Inoue0, KobayashiK, TsukadaH, et al. Differencesin in vivo receptorbinding col 1995;281:14l—l49. between [3H]N-methylspiperone and [3H]raclopride in -treated mouse brain. 253. Volkow ND, Fowler IS, Wolf AP, et al. Distribution and kinetics of carbon-I I- iNeural Transm l991;85:l—lO. cocaine in the human body measured with PET. J NucI Med 1992;33:52l—525. 248.KungHF,Pan5, KungMP,etal.In vitroandin vivoevaluationof [‘231]IBZM:a 254. Scheffel U, Boja 1W, Kuhar Mi. Cocaine receptors: in vivo labeling with 3H- potential CNS D2-dopamine receptor imaging agent. J Nuc! Med 1989;30:88—92. (—)cocaine, 3H-WLN 35,065-2 and 3H-WIN 35,428. Synapse 1989;4:390—392.

Ethical Clinical Practice of Functional Brain Imaging

Society of Nuclear Medicine Brain Imaging Council Reston, Virginia

patterns are variable and not easily interpreted as being causally The development and evolutionof functionalbrain imagingtechnol related to a particular disease entity (e.g., mild head , ogies and their broad application to a wide range of neurological and AIDS, chronic fatigue syndrome, toxic exposures, foreign-body psychiatric disorders have led to their scientifically sound use in specific clinical situations. In addition, there is a growing diversity of reactions, autoimmune disorders, substance abuse, violence and empirical new applicatkxis where there is ktde prevkxis research or others) (30,31 ). Furthermore, while specific PET and SPECT clinical experience. Therefore, a committee of the Brwn Imaging defects appear useful to confirm a certain disease diagnosis or Council of the Society of Nuclear Medicine was formed to address to support the localization of a particular clinical finding theneedforspecificguidelinesregardingscaninterpretationand (32—35),there is only limited evidence that specific diseases or reporting. This committee considered the wide range of current and neurological, psychiatric or behavioral deficits can be predicted potential uses of PET and SPECT, including its growing role in from specific scan patterns (36,37). While these types of studies forensics. A set of basic guidelines for the reporting and interpreta remain extremely important for identifying previously unrecog tion of brain imaging studies applicable to all clinical situations, nized brain abnormalities and potential disease mechanisms in including forensics, was formulated. These gukielines were corn a variety of neuropsychiatric illness, their utility in the manage posed in a manner sensitive to the need for standards that are scientifically defensible now, and which willcontinue to be valid as ment of individual patients is still far from clear. the field evolves. It is the intent of the committee and this summary As SPECT and PET become more widely available, enthu document to positively influence the clinical use of brain SPECT and siasm tempered by a cautious attitude seems appropriate regard PET by offering guidance concerning the elements essential to a ing their use in brain disorders as a whole, given clear evidence complete and useful clinical report, defining standards to differenti that functional patterns are highly dependent on a large number ate well-established clinical applications from research uses and of technical, analytical and physiological variables. The pur providing a framework in which to consider the appropriateness of pose ofthis document is to provide recommendations and basic functional brain imaging used in the forensic arena. guidelines for brain SPECT and PET acquisition, interpretation Key Words PET; SPECT; clinical applications; forensics; gukielines and reporting, with particular attention to recognized and J NucIMed1996;37:1256-1259 generally accepted clinical indications, and to urge caution regarding applications in unstudied behavioral disorders. TheuseofSPECTandPETinthemanagementofpatients with stroke, epilepsy, brain tumors and , and in some BASIC METhODOLOGICAL ISSUES cases, movement disorders and moderate-to-severe head trauma The Society of Nuclear Medicine has recently published is now well recognized (1—9).Scan abnormalities have also technical guidelines on brain SPECT acquisition and image been identified in patients with certain psychiatric diagnoses, reconstruction (38,39). However, even with adherence to these including depression, obsessive compulsive and panic disor recommendations, the quality of the study will vary from ders, schizophrenia and substance abuse (10—19), but consis institution to institution and is dependent on several factors, tent patterns for these disorders have not been confirmed. including instrumentation, collimation, filters, behavioral state Sensitivity and specificity are often unknown and in many ofthe subject during tracer uptake, timing ofthe scan relative to cases, group patterns may actually be too subtle to detect in tracer injection, scan duration, patient movement, attenuation, individual patients. reconstruction and analytical methods, as well as quality con More elusive or less well-characterized behavioral syn trol. dromes have also been studied (20—29). In many cases, the In general, disease patterns are established using specific instruments, physiological measurements and methods of anal ysis. However, for some situations, the imaging technology Received Mar. 12, 1996; accepted Mar. 20, 1996. ForcorrespondenceorreprintscontactJOannaWIISOn,CouncilCoordinator,Society available at a particular site may not be sufficient for diagnostic ofNuclearMedicine,1850SamuelMorseDr.,Reston,VA22090. purposes. Accordingly, the degree to which subsequent studies

1256 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 July 1996