TY - JOUR T1 - Simplified Non-Invasive Tracer Kinetic Analysis for <sup>18</sup>F-Florbetaben PET Using a Dual Time-Window Acquisition Protocol JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1640 LP - 1640 VL - 59 IS - supplement 1 AU - Henryk Barthel AU - Santi Bullich AU - Norman Koglin AU - Georg Becker AU - Aleksandar Jovalekic AU - Susan De Santi AU - Andrew Stephens AU - Osama Sabri Y1 - 2018/05/01 UR - http://jnm.snmjournals.org/content/59/supplement_1/1640.abstract N2 - 1640Objectives: Accurate quantitation of amyloid-beta (Aß) plaque load using positron emission tomography (PET) is important for monitoring Aß accumulation in therapeutic trials of Aß-modifying treatments for Alzheimer’s disease (AD). A standardized uptake value ratios (SUVR) approach is currently used in most clinical research settings, but this approach might be biased by radiotracer clearance or cerebral blood flow (CBF) changes over time. The objectives of this work were: (1) to validate a non-invasive kinetic modeling method for 18F-florbetaben PET using a simplified dynamic acquisition protocol and (2) to assess the influence of CBF and radiotracer clearance changes on SUVRs and non-invasive kinetic modeling data. Methods: Data from twenty subjects (10 probable AD dementia, 10 age-matched healthy volunteers) scanned dynamically for 140 min were used. Arterial samples were collected during scanning and corrected for metabolites. The binding potential (BPND) from non-invasive tracer kinetic methods (simplified reference tissue model (SRTM), and multilinear reference tissue model (MRTM2)) and SUVR were compared for different shortened or interrupted acquisition approaches. Simulations were carried out to assess the effect of regional CBF affecting either the target or reference region and radiotracer clearance changes. For the MRTM2, the effect on BPND estimates of different initial equilibration times (t[asterisk]) ranging from 10 to 70 min were analyzed. Results: A 0-30 and 120-140 min dual time-window acquisition protocol provided the best compromise between patient comfort and quantification accuracy. Excellent agreement was found between BPND obtained using full and two time-window (2TW) acquisition protocols (BPND,2TW= 0.01+ 1.00·BPND,FULL, R2=0.97 (MRTM2); BPND,2TW= 0.05+ 0.92·BPND,FULL, R2=0.93 (SRTM)). Simulations showed a limited impact of CBF and radiotracer clearance changes on MRTM2 parameters and SUVRs. The optimal t[asterisk] that provided lowest bias on BPND estimates for the MRTM2 was 70 min. Conclusions. This study demonstrates accurate non-invasive kinetic modeling of 18F-florbetaben PET data using a simplified dual time-window dynamic acquisition protocol. The influence of CBF and radiotracer clearance changes on the PET outcome measures of Aß plaque load was small. Thus, for most clinical research applications, the SUVR approach is sufficient. In cases in which maximum quantification accuracy is needed a dynamic dual time-window scan protocol may be beneficial. ER -