Characterization of extrastriatal D2 in vivo specific binding of [¹⁸F](N-methyl)benperidol using PET

Abstract

PET imaging studies of the role of the dopamine D2 receptor family in movement and neuropsychiatric disorders are limited by the use of radioligands that have near-equal affinities for D2 and D3 receptor subtypes and are susceptible to competition with endogenous dopamine. By contrast, the radioligand [¹⁸F]N-methylbenperidol ([¹⁸F]NMB) has high selectivity and affinity for the D2 receptor subtype (D2R) and is not sensitive to endogenous dopamine. Although [¹⁸F]NMB has high binding levels in striatum, its utility for measuring D2R in extrastriatal regions is unknown. A composite MR-PET image was constructed across 14 healthy adult participants representing average NMB uptake 60 to 120 min after [¹⁸F]NMB injection. Regional peak radioactivity was identified using a peak-finding algorithm. FreeSurfer and manual tracing identified a priori regions of interest (ROI) on each individual's MR image and tissue activity curves were extracted from coregistered PET images. [¹⁸F]NMB binding potentials (BP(ND) s) were calculated using the Logan graphical method with cerebellum as reference region. In eight unique participants, extrastriatal BP(ND) estimates were compared between Logan graphical methods and a three-compartment kinetic tracer model. Radioactivity and BP(ND) levels were highest in striatum, lower in extrastriatal subcortical regions, and lowest in cortical regions relative to cerebellum. Age negatively correlated with striatal BP(ND) s. BP(ND) estimates for extrastriatal ROIs were highly correlated across kinetic and graphical methods. Our findings indicate that PET with [¹⁸F]NMB measures specific binding in extrastriatal regions, making it a viable radioligand to study extrastriatal D2R levels in healthy and diseased states.

Figure 1

Visualization of target regions for BP_ND analysis. Target ROIs (A-E) for one representative individual are overlaid on an MRI in axial and coronal views. The cerebellar reference region is depicted in F. To control for partial volume effects, ROIs were eroded when possible. Light blue coloring indicates original, uneroded regions as defined by FreeSurfer or manual tracing. Dark blue coloring indicates eroded regions. BP_ND's were obtained from eroded areas in all ROIs except for nucleus accumbens, hypothalamus, midbrain, and cortical regions. Sup Temp Sulcus, banks of superior temporal sulcus. rAnt Cing Cortex, rostral anterior cingulate cortex. Trans Temp Gyrus, transverse temporal gyrus. mOrbitofrontal Cortex, medial orbitofrontal cortex.

Figure 2

Main results of the radioactivity peak analysis (n = 14). A peak analysis identified brain regions with high radioactivity counts in PET images including (A) striatal areas, (B) thalamus and hippocampus, (C) midbrain and (D) temporal cortex. Note the decrease in intensity threshold necessary to identify peaks (dark red, arrows) in extrastriatal regions (color bar on right side of each PET image). Mean PET counts in the cerebellar reference region was 1.

Figure 3

Radioactivity levels by region over scan time. NMB activity over time of the PET scan is shown in one representative subject for (A) striatal ROIs, (B) subcortical extrastriatal ROIs and (C) cortical ROIs relative to the cerebellar reference region. Sup Temp Sulcus, banks of superior temporal sulcus. rAnt Cing Cortex, rostral anterior cingulate cortex. Trans Temp Gyrus, transverse temporal gyrus. mOrbitofrontal Cortex, medial orbitofrontal cortex.

Figure 4

Analyses of ROI BP_ND's as estimated by the Logan graphical method with cerebellum as a reference region. BP_ND's (mean ± S.D., n = 14) differed between and within (A) striatal,(B) extrastriatal subcortical and (C) cortical regions. Note the approximately 10-fold difference in scale between striatal and extrastriatal binding regions. NAc, nucleus accumbens. Sup Temp Sulcus, banks of superior temporal sulcus. rAnt Cing Cortex, rostral anterior cingulate cortex. Trans Temp Gyrus, transverse temporal gyrus. mOrbitofrontal Cortex, medial orbitofrontal cortex. *, ***, p < 0.05, p < 0.001 relative to other striatal regions; ^###, p < 0.001 relative to hippocampus; ^†, ^††, p < 0.05, p < 0.01 relative to midbrain; ^{$, $$$,} p < 0.05, p < 0.001 relative to transverse temporal gyrus; ^{@, @@@}, p < 0.05, p < 0.001 relative to medial orbitofrontal cortex.

Figure 5

Relationship between age and BP_ND. Age was negatively correlated with D2R BP_ND's in striatal regions including putamen (r₁₄ = -0.81, p < 0.001), caudate (r₁₄ = -0.89, p < 0.001) and nucleus accumbens (r₁₄ = -0.63, p < 0.05).