RT Journal Article
SR Electronic
T1 Test–Retest Reproducibility for the Tau PET Imaging Agent Flortaucipir F 18
JF Journal of Nuclear Medicine
JO J Nucl Med
FD Society of Nuclear Medicine
SP 937
OP 943
DO 10.2967/jnumed.117.200691
VO 59
IS 6
A1 Michael D. Devous, Sr
A1 Abhinay D. Joshi
A1 Michael Navitsky
A1 Sudeepti Southekal
A1 Michael J. Pontecorvo
A1 Haiqing Shen
A1 Ming Lu
A1 William R. Shankle
A1 John P. Seibyl
A1 Ken Marek
A1 Mark A. Mintun
YR 2018
UL http://jnm.snmjournals.org/content/59/6/937.abstract
AB Alzheimer disease (AD) is characterized by β-amyloid (Aβ) plaques and tau neurofibrillary tangles. There are several PET imaging biomarkers for Aβ including 11C-PiB and 18F-florbetapir. Recently, PET tracers for tau neurofibrillary tangles have become available and have shown utility in detection and monitoring of neurofibrillary pathology over time. Flortaucipir F 18 is one such tracer. Initial clinical studies indicated greater tau binding in AD and mild cognitive impairment patients than in controls in a pattern consistent with tau pathology observed at autopsy. However, little is known about the reproducibility of such findings. To our knowledge, this study reports the first data regarding test–retest reproducibility of flortaucipir F 18 PET. Methods: Twenty-one subjects who completed the study (5 healthy controls, 6 mild cognitive impairment, and 10 AD) received 370 MBq of flortaucipir F 18 and were imaged for 20 min beginning 80 min after injection and again at 110 min after injection. Follow-up (retest) imaging occurred between 48 h and 4 wk after initial imaging. Images were spatially normalized to Montreal Neurological Institute template space. SUVRs were calculated using AAL (Automated Anatomical Labeling atlas) volumes of interest (VOIs) for parietal, temporal, occipital, anterior, and posterior hippocampal, parahippocampal, and fusiform regions, as well as a posterior neocortical VOI composed of average values from parietal, temporal, and occipital areas. Further, a VOI derived by discriminant analysis that maximally separated diagnostic groups (multiblock barycentric discriminant analysis [MUBADA]) was used. All VOIs were referenced to a subsection of cerebellar gray matter (cere-crus) as well as a parametrically derived white matter–based reference region (parametric estimate of reference signal intensity [PERSI]). t test, correlation analyses, and intraclass correlation coefficient were used to explore test–retest performance. Results: Test–retest analyses demonstrated low variability in flortaucipir F 18 SUVR. The SD of mean percentage change between test and retest using the PERSI reference region was 2.22% for a large posterior neocortical VOI, 1.84% for MUBADA, 1.46% for frontal, 1.98% for temporal, 2.28% for parietal, and 3.27% for occipital VOIs. Further, significant correlations (R2 > 0.85; P < 0.001) were observed for all regions, and intraclass correlation coefficient values (test–retest consistency) were greater than 0.92 for all regions. Conclusion: Significant test–retest reproducibility for flortaucipir F 18 was found across neocortical and mesial temporal lobe structures. These preliminary data suggest that flortaucipir F 18 tau imaging could be used to examine changes in tau burden over time.