INTRODUCTION

Several brain imaging studies have investigated D-amphetamine (D-AMPH)-induced dopamine (DA) release in the striatum (Laruelle et al, 1995; Kegeles et al, 1999; Drevets et al, 2001; Singer et al, 2002; Piccini et al, 2003; Martinez et al, 2003; Volkow et al, 1994, 2004). However, dopaminergic neurotransmission in cortex, thalamus, and limbic regions is believed to be significantly involved in psychosis, cognitive function, and psychostimulant drug abuse (Weinberger et al, 2001; Stevens, 1991; Kerwin and Murray, 1992; Goldman-Rakic, 1998; Yasuno et al, 2004; Koob and Le Moal, 2001). Most, but not all, recent brain-imaging studies suggest that extrastriatal DA D2 receptors are the site of antipsychotic drug actions (Farde et al, 1997; Pilowsky et al, 1997; Bigliani et al, 2000; Xiberas et al, 2001a, 2001b; Kessler et al, 2002; Bressan et al, 2003; Talvik et al, 2001). Extrastriatal dopaminergic neurotransmission is an important area of research in neuropsychiatric disorders (Kaasinen et al, 2001; Suhara et al, 2002).

The affinity of currently used DA D2 radioligands such as [11C]raclopride, 1.2 nM (Hall et al, 1989), and [123I]IBZM, 0.4 nM (Kung et al, 1989), permits accurate measurement of DA D2 receptor levels only in the striatum, which has the highest DA D2 receptor levels in brain (Kessler et al, 1993b; Laruelle et al, 1995; Drevets et al, 2001). With the development of high-affinity radioligands, such as [11C]FLB 457, [123I]epidepride, and [18F]fallypride, it has been possible to visualize and quantitate levels of striatal and extrastriatal DA D2/D3 receptors (Kessler et al, 1991, 1992, 1993a; Halldin et al, 1995; Mukherjee et al, 1995). The striatal uptakes of [11C]FLB 457 and [123I]epidepride are prolonged (Farde et al, 1997; Olsson et al, 1999; Fujita et al, 1999), making estimation of striatal DA D2/D3 receptor levels difficult. [18F]fallypride is a benzamide with very high-affinity for DA D2/D3 receptors, which has been used to visualize and quantify levels of striatal and extrastriatal DA D2/D3 receptors using positron emission tomography (PET) (Mukherjee et al, 1999, 2002; Rieck et al, 2004). Although [18F]fallypride has an in vitro KD for the DA D2/D3 receptors similar to that of [123I]epidepride and [11C]FLB 457, it has considerably more rapid washout from striatum, allowing estimation of DA D2/D3 receptor levels in both striatal and exrtastriatal regions (Mukherjee et al, 1997; Kessler et al, 1991; Olsson et al, 1999, 2004; Christian et al, 2004).

SPECT and PET studies of DA D2/D3 receptors performed before and after a D-AMPH challenge have been used to study striatal DA release. The decrement in striatal binding potential (b.p.) or V3″ for benzamide DA D2/D3 radioligands following D-AMPH administration in primates has been shown to be linearly related to the level of D-AMPH-induced DA release (Laruelle et al, 1997; Breier et al, 1997). In humans, PET and SPECT studies using either [11C]raclopride or [123I]IBZM have demonstrated D-AMPH (0.3 mg/kg intravenous)-induced decrements of 7.1–16.1% in striatal b.p. or V3″ (Abi-Dargham et al, 1998; Laruelle et al, 1995; Drevets et al, 2001; Martinez et al, 2003). Recently, Cardenas, using [11C]raclopride PET, reported that an oral dose of 30 mg of D-AMPH produced a decrease in striatal b.p. of 13–18% in humans (Cardenas et al, 2004). Epidepride with an affinity of 24 pM for the DA D2 receptor shows virtually no displacement after a 1 mg/kg dose of D-AMPH (Kessler et al, 1993c; al-Tikriti et al, 1994). Studies of [11C]FLB 457 with a reported affinity of 20 pM for the DA D2 receptor (Olsson et al, 1999) have produced conflicting results regarding its sensitivity to extracellular DA levels (Chou et al, 2000; Okauchi et al, 2001). Primate studies of [18F]fallypride, with an affinity of 33 pM for the DA D2 receptor and 31 pM for the DA D3, have reported D-AMPH-induced decreases in b.p. of 12–36% in the thalamus, hippocampus, amygdala, and ventral striatum following 0.3–1.0 mg/kg intravenous doses of D-AMPH (Mukherjee et al, 1997, 2002; Slifstein et al, 2004).

No PET or SPECT studies of D-AMPH-induced DA release in extrastriatal regions have been reported in humans although, as discussed above, such studies have been reported in nonhuman primates (Mukherjee et al, 1997, 2002; Slifstein et al, 2004). The aim of this study was to assess whether [18F]fallypride can be used to estimate D-AMPH-induced DA release in striatal and extrastriatal regions in humans following a 0.43 mg/kg oral dose of D-AMPH. An oral dose of 0.43 mg/kg of D-AMPH in comparison to an intravenous dose of 0.2–0.3 mg/kg produces lesser side effects such as elevated blood pressure, but similar plasma levels and displacement of [11C]raclopride in striatum (Angrist et al, 1987; Breier et al, 1997; Cardenas et al, 2004).

METHODS

In all, 14 normal subjects, six females and eight males (ages 21–32 years, mean age of 25.9 years), were recruited by advertisement. All subjects were right hand dominant and none were smokers. Exclusion criteria included a history of psychiatric or neurological condition, a history of severe concomitant or past medical illness, borderline elevated blood pressure (135/90), any psychotropic medication usage for the last 6 months, a history of substance abuse and dependence, inability to provide informed consent, an IQ less than 80, and pregnancy or lactation. After an initial assessment, the study was explained to subjects and informed consent was obtained. The consent form included a recommendation for all participants to use adequate birth control measures during the study. All female subjects capable of childbearing had pregnancy tests performed 6 h or less prior to each PET study. All subjects received a physical and neurological examination, SCID (Williams et al, 1992), blood chemistries, urine analysis and urine drug screen, EKG and MRI study. Subjects additionally completed the Sensation Seeking Scale-Form (Zuckerman et al, 1978). MRI scans were performed using a GE 1.5 T scanner with echospeed gradients. Thin-section, high-resolution, T1-weighted coronal and sagittal IR SPGR sequences were obtained (TE=3.6, TR=18.9, TI=400, slice thickness of 1.2–1.4 mm) and an axial T2-weighted sequence (TE=106, TR=5000, slice thickness of 3 mm) was obtained as well. Subjects who met the study criteria were scheduled for PET studies which were performed using a GE Discovery LS PET scanner; 3-D emission acquisitions and transmission attenuation correction scans were performed following a 5.0 mCi slow bolus injection of [18F]fallypride (specific activity greater than 3000 Ci/mmol) prior to and 3 h following a 0.43 mg/kg oral dose of D-AMPH. Serial scans were initiated simultaneously with the bolus injection of [18F]fallypride and were obtained for approximately 3.5 h. The initial scan sequence was started coincident with the start of the [18F]fallypride injection and included the following frames: 8 for 15 s, 6 for 30 s, 5 for 1 min, 2 for 2.5 min, 3 for 5 min, and 3 for 10 min. After the initial scan sequence, a 10-min transmission scan was obtained and the subject given a break. At approximately 85–90 min, a second scan sequence of two frames of 25 min each followed by a second transmission scan were obtained. The subject was then allowed a second scan break, and at approximately 165–170 min, a 40-min emission scan followed by a third transmission scan was obtained. For the post-D-AMPH study, subjects were instructed to have a light breakfast on the day of the scan, that is, a single cup of coffee, cold cereal, and/or bread, which was ingested approximately 4 h prior to D-AMPH administration. Physiological measures (blood pressure, heart rate, temperature, and respirations) were monitored throughout the study and a brief neurological examination performed.

Blood samples were collected for determination of plasma levels of D-AMPH at 1, 3, and 5 h following D-AMPH administration. The plasma samples were analyzed using a modification of the method of Campins-Falco et al, 1996. Briefly, 1.0 ml plasma samples, following addition of the internal standard B-phenylethylamine, were made basic and the amines isolated and derivatized on Waters Sep-Pak C-18 cartridge using 1,2-naphthoquinone-4-sulphate (NQS) as described by Campins-Falco. Following washing, the derivatized amines were eluted with acetonotrile : water (1 : 1), the volume reduced under vacuum, and the eluate analyzed by HPLC. The peaks corresponding to derivatized amphetamine and NQS were separated using a 7 × 53 mm Hypersil BDS ‘Rocket’ column (Alltech Assoc.) and quantitated by UV analysis at 450 nM.

Serial PET scans and thin-section T1-weighted MRI scans were coregistered to each other using a mutual information rigid body algorithm (Pluim et al, 2001). Regions of interest were delineated for the right and left caudate, putamen, ventral striatum, amygdala, substantia nigra, the medial thalami, and temporal cortex on MRI scans of the brain (see Figures 1 and 2) by a neuroradiologist experienced in PET data analysis (RMK) and automatically transferred to both the pre- and post-D-AMPH PET studies. The ventral striatum was defined according to criteria of Mawlawi et al, 2001. The substantia nigra was delineated based on landmarks from the Schaltenbrand atlas (Schaltenbrand and Wahren, 1977); the substantia nigra is located in the ventral midbrain posteromedial to the cerebral peduncles from the superior aspect of the interpeduncular fossa approximately 9 mm below the ACPC line to 16 mm below the ACPC line. These regions of interest were transferred to the coregistered PET scans and regional DA D2 receptor b.p.'s were calculated using the reference region method (Lammertsma et al, 1996). Percent displacements were calculated for each region of interest. Parametric images of DA D2/D3 receptor density and percent displacement were calculated on a pixel-by-pixel basis using the reference region method. Images were coregistered across subjects using an elastic deformation algorithm (Rohde et al, 2003). Approximately 60 min after D-AMPH administration (and at the equivalent point in time on non-drug administration days), subjects began a 75-min neuropsychological battery. The battery included measures of attention—Stroop task (Stoelting Co., 2000), information-processing speed—Digit Symbol Coding and Symbol Search (Wechsler, 1997), spatial working memory (Park et al, 1999), and affect—Positive Affect Negative Affect Scale (Watson et al, 1988). In order to minimize the practice effects, all tests were either presented using parallel forms, or utilized random order presentation of stimuli. One male subject was color-blind; although he completed all tests, he was excluded from analysis of the Stroop task that requires color processing. A second male subject had a disrupted spatial working memory trial during the baseline study and he was excluded from the working memory analysis.

Figure 1
figure 1

[18F]fallypride PET scans obtained at baseline in a normal subject at (a) the level of the thalamus showing highest uptake in medial thalamus; (b) at the level of the midbrain showing uptake in the substantia nigra in the ventral midbrain and in the colliculi dorsally in the midbrain, (c) at the level of the uncus showing uptake in the amygdala, and (d) through the inferior temporal cortex showing high uptake in the pituitary and temporal cortex.

Figure 2
figure 2

Parametric images of DA D2 receptor density coregistered with MRI images windowed to demonstrate [18F]fallypride-b.p.'s in (a) the anterior and ventral striatum on a coronal image, (b) the medial thalamus on an axial image, (c) the substantia nigra, colliculi, and temporal cortex on an axial image through the midbrain, and (d) the amygdala and temporal cortex on an axial image through the uncus of the temporal lobes.

To test the effects of D-AMPH on [18F]fallypride binding, we performed a repeated-measures ANOVA implemented in the General Linear Model module of SPSS 11.5 (SPSS Inc., IL). The model used had three within-subject factors, condition (baseline, D-AMPH), region, and laterality. Paired two-tailed t-tests and Wilcoxon signed-rank tests were performed to delineate the source of significant differences on the ANOVA. Bonferroni correction for multiple comparisons were performed. Correlations of changes in cognition, affect, and sensation seeking with parametric images of [18F]fallypride displacement were made using a Pearson product moment correlation and significance assessed using two tailed t-tests. Correction for multiple within-image comparisons was made using the method of Forman et al (1995) as implemented in the Alpha-Sim program of the AFNI analysis program. This was utilized as the image data violated the uniformity of variance assumption of SPM. After correction for multiple within-image comparisons, significant clusters were corrected for multiple behavioral tests using a Bonferroni correction.

RESULTS

Repeated-measures ANOVA using condition, region, and laterality as factors revealed significant effects of region (F=415.5, p<0.00000008) consistent with the large differences seen in regional b.p.'s (see Table 1, Figures 1 and 2), and a main effect of laterality (F=19, p<0.001) reflecting higher b.p.'s in the left temporal cortex, medial thalamus, and ventral striatum, but not caudate or putamen. There was a robust effect of condition due to a decrease in [18F]fallypride b.p.'s following D-AMPH administration (F=48.5, p=0.00001) as well as a significant condition by region interaction consistent with varying decreases across regions (F=16.99, p=0.0037). In contrast, there were no significant interactions of condition by laterality (F=0.36, p>0.1) or condition by region by laterality (F=0.56, p>0.1). Given the lack of a laterality effect, regional displacements were averaged across the right and left regions of interest.

Table 1 D-AMPH-Induced Decrements in Regional Binding Potentials (Means±SD) and Levels of Significance Corrected for Multiple Comparisons Using Paired 2-Tailed t-Tests and Wilcoxon Signed-Rank Test in Parenthesis

The greatest D-AMPH-induced displacements of [18F]fallypride were seen in the striatum and substantia nigra, while lower levels of displacement were seen in the amygdala, temporal cortex, and thalamus (see Table 1). The mean displacement in the putamen was 11.22%, while the mean displacements in the ventral striatum and caudate were 7.23 and 5.57%, respectively. The mean displacement in the substantia nigra was 6.64% similar to that seen in caudate and ventral striatum. Lower levels of displacement were seen in the amygdala—4.36%, medial thalamus—2.84%, and temporal cortex—3.67%. Using paired two-tailed t-tests with a Bonferroni correction for multiple comparisons, displacements were significant in all regions except the medial thalamus which achieved a trend level (p=0.07). The significance of regional displacements was also evaluated using the Wilcoxon signed ranks test (two-tailed) with a Bonferroni correction; displacements remained significant in the caudate (p<0.03), putamen (p=0.007), ventral striatum (p<0.014), substantia nigra (p=0.014), and temporal cortex (p<0.03), but fell to a trend level in the amygdala (p=0.06).

Correlations of changes in cognition with parametric images of DA release following D-AMPH administration showed a number of significant clusters of correlations with measures of attention and speed of cognitive processing (see Figures 3 and 4). Changes in performance on the Digit Symbol Coding, a test of attention, and speed of cognitive processing, were significantly correlated with a cluster centered in the right ventral striatum, but involving the right and left ventral striatum and basal forebrain (p<0.006, corrected for multiple-image comparisons and behavioral tests) due to significant negative correlations ranging from r=−0.53 to −0.82 (see Figure 3). A significant cluster of correlations on performance on the symbol search task, a measure of speed of cognitive processing, was observed in the left ventral putamen (p<0.05, corrected for both multiple-image comparisons and behavioral tests) due to significant negative correlations, r=−0.54 to −0.75 (see Figure 3). Significant clusters of correlations of performance on the Stroop task, a test of attention, were seen with the left temporal cortex and insula (p<0.02, corrected for multiple-image and behavioral comparisons), and with a second cluster involving the right lateral and inferior temporal cortex (p=0.01, corrected for multiple-image and behavioral comparisons) (see Figure 3); both clusters were due to significant negative correlations, r=−0.53 to −0.92. No significant clusters of correlations were seen with changes in spatial working memory.

Figure 3
figure 3

Significant clusters of correlations between D-AMPH-induced DA release and (a) digit symbol coding (ventral striatum and basal forebrain), (b) symbol search (left ventral putamen), and (c, d) stroop task (left temporal and insular cortex, right temporal cortex).

Figure 4
figure 4

Significant clusters of correlations of sensation seeking with D-AMPH-induced DA release in (a) anterior cingulate, (b) left temporal cortex, (c, d) left thalamus extending into the left hippocampus.

Sensation-seeking behavior demonstrated a number of significant clusters of correlations with D-AMPH-induced DA release (see Figure 4). These included a cluster involving the anterior cingulate with greatest extent in the pre and subgenual cingulate (p=0.04, corrected for multiple–image comparisons and behavioral tests), a cluster in the left insula and temporal cortex (p<0.006, corrected for multiple-image comparisons and behavioral tests), and a cluster in the left thalamus which extended into the left hippocampus/medial temporal cortex (p<0.006, corrected for multiple-image comparisons and behavioral tests). These clusters reflected significant negative correlations in these regions, with r's ranging from −0.53 to −0.91. No significant clusters were observed for changes in positive affect with DA release.

Plasma D-AMPH levels were 0.45±0.26 (SD) ng/ml at 1 h after administration, 0.44±0.22 at 3 h, and 0.35±0.13 at 5 h. There were no significant correlations of plasma D-AMPH levels at 1, 3, or 5 h with decrements in putamenal b.p.'s.

DISCUSSION

The results of this study demonstrate significant displacements of [18F]fallypride by DA released by oral D-AMPH in the striatum, substantia nigra, and cortex, with a trend level change in the amygdala. The magnitude of striatal displacement is similar to that reported with either [123I]IBZM or [11C]raclopride using either oral (30 mg) or intravenous doses (0.3 mg/kg) of D-AMPH. [123I]IBZM SPECT studies of D-AMPH-induced displacements report decrements in the striatum of 7–9% in normal subjects following an intravenous dose of 0.3 mg/kg (Abi-Dargham et al, 2003, 2004; Laruelle et al, 1996; Kegeles et al, 1999). [11C]raclopride PET studies of D-AMPH (0.3 mg/kg, intravenous)-induced displacements of approximately 8–16% in striatal subdivisions have reported decrements in the dorsal putamen, 4–6% in the caudate, and 15% in the ventral striatum (Martinez et al, 2003; Drevets et al, 2001). The decrements seen in the caudate and dorsal putamen in the current study are similar to those reported with [11C]raclopride studies using intravenous D-AMPH, but are considerably lower in the ventral striatum. The only previous PET study which employed a comparable oral D-AMPH dose, that is, 30 mg, to that used in the current study, reported a striatal decrement of 13% at 2 h post-D-AMPH administration, comparable to the results in the current study in which scanning commenced at 3 h post-injection. The results of the current study indicate that [18F]fallypride has a sensitivity similar to D-AMPH-released DA in comparison to [11C]raclopride in the dorsal caudate and putamen. Although these results are consistent with primate studies (Slifstein et al, 2004), they are at variance with modeling studies suggesting that the magnitude of D-AMPH induced displacement decreases with ligands having b.p. greater than 10 (Endres and Carson, 1998).

The difference in ventral striatal displacements seen between the current and previous studies of Martinez et al (2003) and Drevets et al (2001) may be related to the different routes of administration. Previous studies of oral vs intravenous methylphenidate (Volkow et al, 2004) have shown that oral compared to intravenous methylphenidate administration produces slower increases in brain methylphenidate levels, comparable peak brain levels, comparable levels of dorsal striatal DA release, but significantly less positive reinforcing effects. Positive reinforcing effects were significantly and positively correlated with dorsal striatal DA release following intravenous administration of methylphenidate, but no correlation was seen following oral administration. The ventral striatum has been shown to mediate the euphorogenic effects of intravenous D-AMPH in humans (Drevets et al, 2001). The lesser D-AMPH-induced displacement seen in the ventral striatum in the current study may reflect different regulation of dorsal vs ventral striatal DA release in response to differing rates of brain uptake of DA-releasing drugs, and may be the physiological basis for the lesser positive reinforcing effects of oral vs intravenous psychostimulant drugs. It is noteworthy that no significant correlation cluster of positive affect with DA release was seen for the group as a whole in the current study analogous to the results seen with oral methylphenidate, another psychostimulant drug which increases extracellular DA levels (Volkow et al, 2004). Additional studies are needed to confirm these results and to evaluate the mechanism by which such differential regulation occurs.

Alternative explanations for the difference in ventral striatal displacements could include differences in methods and inaccuracies in image registration. To make the results comparable to the study of Martinez et al, 2003, we used the same anatomic criteria for delineation of the ventral striatum as used in that study (Mawlawi et al, 2001). The resolution of the scanners used in all the current and previous studies of Drevets and Martinez is similar (Townsend et al, 1998). The algorithms used for coregistration in this study have been shown to have registration errors of 1.1 mm in phantom studies and 2.6 mm in previous studies of PET/MR coregistration using an older PET scanner with 6–8 mm resolution (West et al, 1997; Lavely et al, 2004); the accuracy of coregistration is demonstrated in the relatively small standard deviations, 8–9% of the means, seen in [18F]fallypride b.p.'s in small structures such as the ventral striatum and substantia nigra. An example of PET/MR coregistration is shown in Figure 2. The difference in apparent ventral striatal DA release is probably not due to methodological factors.

D-AMPH produced a mean displacement of 6.65% in the substantia nigra, which is similar to that seen in striatum. This is surprisingly high taking into consideration that extracellular DA levels at baseline and following D-AMPH administration are approximately 10% of that seen in the striatum (Gerhardt et al, 2002). In the substantia nigra, DA D2 receptors are autoreceptors which are localized on the dendrites of dopaminergic neurons (Sesack et al, 1994) and are in the high-affinity agonist state, while in the striatum they are in both the high- and low-affinity state. It has been estimated that approximately 40% of striatal DA D2 receptors are in the high-affinity agonist state (Ginovart et al, 1997; Laruelle et al, 1997; Laruelle, 2000). The b.p. of the substantia nigra is approximately 10% of that of the striatum (see Table 1). Endres has suggested that the sensitivity of benzamide radioligands to extracellular DA levels is greatest at a b.p. of 3–10, with decreasing sensitivity above this range (Endres and Carson, 1998). These factors may be responsible, at least in part, for the surprisingly high D-AMPH-induced displacement of [18F]fallypride in the substantia nigra. While the substantia nigra is a small structure whose apparent receptor levels are diminished by partial voluming, the borders of the substantia nigra are not adequately delineated on MRI studies, making implementation of a partial volume correction problematic. The value of a partial volume correction for D-AMPH-induced decrements in regional b.p.'s has been evaluated by Slifstein in nonhuman primates (Slifstein et al, 2004), who found significant spillover of striatal counts into extrastriatal regions, but no significant effect of a partial volume correction on the change in V3″ in the striatum and extrastriatal regions.

It has been previously hypothesized that D-AMPH-released synaptic DA is responsible for displacement of benzamide radioligands in striatum (Laruelle, 2000). There are a number of studies which suggest that dopaminergic neurotransmission in extrastriatal regions is largely mediated by a volume or extrasynaptic mode. In the substantia nigra and VTA, DA release appears to be largely mediated by reverse transport by the DA transporter (Falkenburger et al, 2001). The lack of synaptic specializations in the substantia nigra suggests that somatodendritic DA release does not rely on typical synaptic neurotransmission (Heeringa and Abercrombie, 1995). As DA D2 receptors are largely extrasynaptic in the substantia nigra (Sesack et al, 1994; Yung et al, 1995), it has been suggested that dopaminergic neurotransmission in the substantia nigra may be mediated by volume transmission rather than a synaptic mode (Fuxe et al, 2005; Cragg et al, 2001; Cragg and Greenfield, 1997). Previous studies of DA release and diffusion in the cortex and amygdala indicate that cortical and amygdalar dopaminergic neurotransmission is mediated largely via a volume mode (Garris and Wightman, 1994; Fuxe et al, 2005). Given these observations, the D-AMPH-induced displacements seen in the substantia nigra, amygdala, and temporal cortex are likely due, at least in part, to DA displacement of [18F]fallypride at extrasynaptic DA D2 receptors.

The magnitude of D-AMPH-induced displacements of [18F]fallypride is lower in the temporal cortex, amygdala, and medial thalamus than in the striatum, that is, 2.84–4.36 vs 5.57–11.22%. This is consistent with microdialysis studies demonstrating a 10-fold lower level of baseline extracellular DA in the cortex and amygdala and 4–5-fold lesser increases in extracellular DA levels in these regions following D-AMPH administration (Moghaddam et al, 1993; Garris and Wightman, 1994). The greater displacements observed in primate [18F]fallypride studies are likely due to the higher doses of D-AMPH administered, different routes of administration, effects of anesthesia, and species differences (Slifstein et al, 2004; Narendran et al, 2004; Tsukada et al, 1999a, 1999b; 2002).

In the current study, D-AMPH was administered approximately 180 min prior to the injection of [18F]fallypride for a number of reasons. The study of Cardenas demonstrated no diminution in displacement of [11C]raclopride from 3 to 6 h after D-AMPH administration. As the current and previous results (Angrist et al, 1987) show, the plasma levels of D-AMPH remain reasonably constant from 3 to 5 h after oral D-AMPH administration. A 3 h delay in [18F]fallypride administration following oral D-AMPH administration allows neuropsychological testing and scanning with a single dose of D-AMPH, permitting a within-subject comparison of change in cognition and affect with regional DA release. There are a number of factors which may affect apparent D-AMPH displacements, including changes in cerebral blood flow, and D-AMPH induced changes in the cerebellar distribution volume of [18F]fallypride. The effects of D-AMPH on blood flow are minimal by 3 h (Price et al, 2002); Slifstein et al, (2004) has presented simulations showing that changes in cerebral blood flow would have only minimal effects on D-AMPH-induced displacements of [18F]fallypride. It is unlikely that changes in cerebral blood flow are responsible for the apparent displacements. D-AMPH does not significantly affect the cerebellar distribution volume of [18F]fallypride in nonhuman primates. In addition, the highly significant differences in regional displacements seen in the current study suggest that a change in cerebellar distribution volume is not a major factor in the regional displacements. Further studies are needed to examine this issue.

We examined correlations of changes in neurocognitive functions, that is, attention, speed of cognitive processing, and spatial working memory, positive affect, as well as sensation-seeking behavior with D-AMPH-induced DA release using parametric image analyses. These behaviors have previously been shown to be modulated by cerebral dopaminergic neurotransmission (Rogers, 1986; Piazza et al, 1991; Hooks et al, 1994; Lawrence et al, 1998; Shah et al, 1997; Mehta et al, 1999; Marinelli and White, 2000; Drevets et al, 2001; Abi-Dargham et al, 2003). Although only a relatively small group of normal subjects was examined, significant clusters of correlations, corrected for both multiple-image and behavioral comparisons, were seen with multiple measures of attention and speed of cognitive processing, as well as with sensation seeking. These significant correlations of DA-modulated behaviors with regional DA release in the striatal and extrastriatal regions suggest that the apparent variability seen across subjects in regional DA release may reflect intersubject differences in DA release and not simply image-related noise. This is consistent with a recent PET study finding decreases in frontal and temporal DA D2 receptor availability during performance of attentional and working memory tasks (Aalto et al, 2005). These results suggest that [18F]fallypride PET studies can be used to study the relationship between regional DA release in the striatum and extrastriatal regions and neuropsychological function.

The lack of significant clusters of correlations with positive affect and spatial working memory, as well as the differences between the current study and that of Leyton et al (2002) in regard to novelty seeking, may be due to a number of factors. First, only a small group was examined which has limited power given the size and variability of the D-AMPH-induced decrements in [18F]fallypride b.p.'s. As noted above, Volkow et al, (2004) did not observe correlations of the positive reinforcing effects of methylphenidate with dorsal striatal DA release after oral administration, although such a correlation was seen with intravenous administration. The correlation coefficient observed by Abi Dargham with D-AMPH-induced DA release with positive affect in a large group of subjects, that is, 60, was approximately 0.37. In addition, the smaller D-AMPH-induced displacement in the ventral striatum compared to previous [11C]raclopride PET studies (Drevets et al, 2001; Martinez et al, 2003) may limit the ability to demonstrate a significant correlation with change in positive affect, although the smaller standard deviation in the ventral striatum in the current study (5.29 vs 10.6 and 11.8%) may moderate this limitation. The oral route of administration and the small size of the correlation with positive affect seen in the largest group reported to date may limit the ability of the current study to detect correlations with positive affect. Frontal cortical dopaminergic neurotransmission has been shown to modulate spatial working memory (Sawaguchi and Goldman-Rakic, 1991). The low b.p. of [18F]fallypride in the frontal cortex (Mukherjee et al, 2002) as well as the small number of subjects in this study may limit the ability of [18F]fallypride PET studies to detect D-AMPH-induced DA release in the frontal cortex. In regard to sensation-seeking and novelty-seeking behaviors, Leyton reported a significant positive correlation between D-AMPH-induced ventral striatal DA release, and novelty seeking as measured on the TPQ. Leyton examined eight male subjects using a lower oral dose of D-AMPH. Interestingly, no displacement of [11C]raclopride from the caudate or putamen was observed although levels of ventral striatal DA release similar to that seen in the current study were observed. While the reasons for the discrepancies between the two studies are not clear, the use of a different rating instrument, a lower dose of D-AMPH which did not appear to affect the dorsal striatum, and the use of only male subjects may be some of the factors responsible for this discrepancy. We have found gender differences in a number of correlations of behaviors with regional DA release (Riccardi et al, 2005). Correlation of region of interest data showed a correlation coefficient of 0.898, p=0.006 for sensation seeking with left ventral striatal DA release in male subjects, but a correlation coefficient of −0.713 in female subjects. This difference was significant at the p=0.005 level. When male and female subjects are grouped together, no significant correlation is seen. As we have demonstrated (Riccardi et al, 2005), sex differences need to be considered in evaluating regional DA release and its relationship to behavior.

In conclusion, the results of this study demonstrate the feasibility of using [18F]fallypride PET studies with oral D-AMPH to study DA release in extrastriatal and striatal regions, as well as the relationship of regional DA release to neuropsychological function. This method should allow study of DA release in schizophrenia, psychostimulant drug abuse, attention deficit disorder, depression, as well as other disorders.