@article {Bullichjnumed.119.236240, author = {Santiago Bullich and Olivier Barret and Cristian Constantinescu and Christine Sandiego and Andre Mueller and Mathias Berndt and Caroline Papin and Audrey Perrotin and Norman Koglin and Heiko Kroth and Andrea Pfeifer and Gilles Tamagnan and Jennifer Madonia and John P Seibyl and Kenneth Marek and Susan de Santi and Ludger M Dinkelborg and Andrew W Stephens}, title = {Evaluation of dosimetry, quantitative methods and test-retest variability of 18F-PI-2620 PET for the assessment of tau deposits in the human brain}, elocation-id = {jnumed.119.236240}, year = {2019}, doi = {10.2967/jnumed.119.236240}, publisher = {Society of Nuclear Medicine}, abstract = {18F-PI-2620 is a next generation tau positron emission tomography (PET)-tracer that has demonstrated ability to image the spatial distribution of suspected tau pathology. The objective of this study was to assess the tracer biodistribution, dosimetry and quantitative methods of 18F-PI-2620 in the human brain. Full kinetic modelling approaches to quantify tau load were investigated. Non-invasive kinetic modeling approaches and semi-quantitative methods were evaluated against the full tracer kinetics. Finally, the reproducibility of PET measurements from test and retest scans was assessed. Methods: Three healthy controls (HC) and 4 Alzheimer disease (AD) subjects underwent two dynamic PET scans including arterial sampling. Distribution volume ratio (DVR) was estimated using full tracer kinetics (2 Tissue Compartment (2TC) models, Logan Graphical Analysis (LGA)) and non-invasive kinetic models (Non-Invasive Logan Graphical Analysis (NI-LGA) and the multilinear reference tissue model (MRTM2)). Standardized uptake value ratio (SUVR) was determined at different imaging windows after injection. Correlation between DVR and SUVR, effect size (Cohen{\textquoteright}s d) and test-retest variability (TRV) were evaluated. Additionally, 6 HC subjects received one tracer administration and underwent whole-body PET for dosimetry calculation. Organ doses and the whole-body effective dose were calculated using OLINDA 2.0. Results: Strong correlation was found across different kinetic models (R2 \>0.97) and between DVR(2TC) and SUVRs between 30 to 90 min with R2\>0.95. Secular equilibrium was reached around 40 min post injection (p.i.) in most regions and subjects. The TRV and effect size for the SUVR across different regions was similar at 30-60 min (TRV=3.8\%, d=3.80), 45-75 min (TRV=4.3\%, d=3.77) and 60-90 min (TRV=4.9\%, d=3.73) and increased at later time points. Elimination was via the hepatobiliary and urinary system. The whole-body effective dose was determined to be 33.3{\textpm}2.1 μSv/MBq for an adult female and 33.1{\textpm}1.4 μSv/MBq for an adult male with a 1.5 hour urinary bladder voiding interval. Conclusion: 18F-PI-2620 exhibits fast kinetics, suitable dosimetry and low TRV. DVR measured using the 2TC model with arterial sampling correlated strongly with DVR measured by NI-LGA, MRTM2 and SUVR. SUVR can be used for 18F-PI-2620 PET quantification of tau deposits avoiding arterial blood sampling. Static 18F-PI-2620 PET scans between 45-75min p.i. provide excellent quantification accuracy, large effect size and low TRV.}, issn = {0161-5505}, URL = {https://jnm.snmjournals.org/content/early/2019/11/07/jnumed.119.236240}, eprint = {https://jnm.snmjournals.org/content/early/2019/11/07/jnumed.119.236240.full.pdf}, journal = {Journal of Nuclear Medicine} }