PT  - JOURNAL ARTICLE
AU  - Li, Bolun
AU  - Ding, Song
AU  - Zou, Zhiguo
AU  - Wang, Ningchao
AU  - Chen, Yumei
AU  - Zhou, Yun
AU  - Wang, Yihan
AU  - Liu, Jianjun
AU  - Pu, Jun
TI  - <strong>Using a computationally consistent and efficient graphic analysis of 18F-PBR06 total-body dynamic PET in health volunteers</strong>
DP  - 2023 Jun 01
TA  - Journal of Nuclear Medicine
PG  - P902--P902
VI  - 64
IP  - supplement 1
4099  - http://jnm.snmjournals.org/content/64/supplement_1/P902.short
4100  - http://jnm.snmjournals.org/content/64/supplement_1/P902.full
SO  - J Nucl Med2023 Jun 01; 64
AB  - P902 Introduction: 18F-PBR06 PET was recently proposed for translocator protein 18-kDa (TSPO) imaging, but lack of knowledge of 18F-PBR06 kinetics measured by human dynamic PET. Kinetic modeling and parametric imaging of human 18F-PBR06 total-body dynamic PET is even challenging in terms of high heterogeneity in tracer kinetics and computational cost. In quantitative radioligand-receptor PET analyses, the graphical analysis including the Logan plot and the relative equilibrium based graphical plot (RE plot) are used to estimate total distribution volume (VT). Previous studies show that the Logan plot in VT estimation is robust to tracer kinetics but sensitive to the noisy levels of data. In contrast, the RE plot is computationally efficient and consistent in VT estimation, but its application is limited to the condition of relative equilibrium of tracer kinetics. The objective of this study is to evaluate the Logan and RE plots for quantification of 18F-PBR06 total-body dynamic PET in the healthy volunteers.Methods: A total of 6 healthy human volunteers were enrolled in this study. All participants underwent a 90-min dynamic 18F-PBR06 PET scan on a high sensitivity and high resolution total-body PET/CT scanner (uEXPLORER, united-Imaging Healthcare). The PET images were reconstructed into a total of 98 dynamic frames (30×2 s, 12×5 s, 6×10 s, 4×30 s, 25×60 s, 15×120 s, 6×300 s) with an image matrix of 360×360 pixel, 4 iterations, and 20 subsets by list-mode ordered subset-expectation maximization (OSEM) algorithm with time-of-flight and point-spread function (PSF). Six regions of interest (ROI) were drawn manually using PMOD, including descending aorta, kidney, spleen, liver, lung, left ventricle wall (LV_wall). The time activity curve (TAC) of the descending aorta ROI was used as input function for the Logan and RE plots. Brain PET was spatially normalized standard space and 35 brain ROIs were drawn on a MRI template in standard space to extract brain ROI TACs. The distribution volume (VT) was estimated by using RE plot and Logan plot with blood input function for each ROI with t*=20-min. The VT parametric images were also generated from both RE plot and Logan plot.Results: The spatial distribution and kinetics of 18F-PBR06 measured by total-body PET is demonstrated in Fig. a. The graphical analysis of Logan plot and RE plot at ROI TAC level from a typical study is shown in Fig. b-e. The VT estimates obtained by linear regression slope from the study are quite comparable. As shown in Fig.f and Fig.j, the VT estimates from the ROI TACs with the RE plot are close to the ones from the ROI TAC with Logan plot. High to low VTs were observed in LV_wall, spleen, kidney, liver, lung, and brain. VT in thalamus is highest in the brain ROIs. Parametric images of VT generated from a typical study using the Logan plot and RE plot were shown in Fig. h. Compared the RE plot, the noise-induced under estimation in the VT images generated by the Logan plot was observed (arrowed) in the typical study.Conclusions: In conclusion, this is a first-in-human kinetic modeling and parametric imaging of 18F-PBR06 total-body dynamic PET in healthy participants. The parameter VTs from RE plot and Logan plot kinetic modeling were comparable at ROI kinetic level. The computationally consistent and efficient graphic analysis using RE plot is suggested for VT parametric image generation.