• Qi Guo, Alessandro Colasanti, David R Owen, Mayca Onega, Aruloly Kamalakaran, Idriss Bennacef, Paul M Matthews, Eugenii A Rabiner, Federico E Turkheimer, and Roger N Gunn.
• Centre for Imaging Sciences, Imanova, London, United Kingdom.
• J. Nucl. Med. 2013 Nov 1; 54 (11): 1915-23.

UnlabelledPET is used to image active inflammatory processes by targeting the translocator protein (TSPO). In vitro, second-generation TSPO radioligands, such as PBR111, have been shown to bind to human tissue samples with either high affinity (high-affinity binders, HABs), low affinity (low-affinity binders, LABs), or an intermediate, mixed affinity (mixed-affinity binders, MABs). We previously explained these differences in affinity in human tissue via the rs6971 polymorphism in the TSPO gene and predicted that the specific signal from PET ligands in vivo would vary accordingly. In silico modeling predicted that (18)F-PBR111 would have a moderate to high specific-to-nonspecific ratio in the normal human brain. To test these predictions, we present here the analysis and modeling of (18)F-PBR111 data in healthy humans.MethodsTwenty-one subjects (9 HABs, 8 MABs, and 4 LABs), 28-62 y old, genotyped for the rs6971 polymorphism, underwent 120-min PET scans with arterial sampling after a bolus injection of (18)F-PBR111. Compartmental models and Logan graphical methods enabled estimation of the total volume of distribution (VT) in regions of interest (ROIs). To evaluate the specific signal, we developed 2 methods to estimate the nondisplaceable volume of distribution (V(ND)): the first assumed that the in vitro affinity ratio of (18)F-PBR111 in HABs relative to LABs (4-fold) is preserved in vivo; the second modeled the difference in the HAB and MAB signals in the context of an occupancy plot.ResultsA 2-tissue-compartment model described the data well, and a significant difference was found between the VT of HABs, MABs, and LABs across all ROIs examined (P < 0.05). We also found a significant correlation between VT and age for both HABs and MABs in most ROIs. The average V(ND) estimated by the 2 methods was 1.18 ± 0.35 (method I: V(ND) = 0.93, method II: V(ND) = 1.42), implying that the (18)F-PBR111 BPND was 2.78 ± 0.46 in HABs, 1.48 ± 0.28 in MABs, and 0.51 ± 0.17 in LABs and that the in vivo affinity ratio was similar to that measured in vitro.Conclusion(18)F-PBR111 shows a high specific signal in the healthy human brain in vivo. A large component of the variability in the signal across subjects is explained by genetic variation and age, indicating that (18)F-PBR111 can be used for the quantitative assessment of TSPO expression.

27 keywords...

hide…