TY - JOUR T1 - Combined cut-off point determination of <sup>18</sup>F-AV-1451 (T807) imaging biomarker for tau burden and 2018 NIA-AA definition facilitate reclassification of Alzheimer’s disease JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 254 LP - 254 VL - 60 IS - supplement 1 AU - Ivan Ho AU - Sirong Chen AU - Yim Lung Leung AU - Man Ki Cheung AU - Chi-Lai Ho AU - Vincent Mok Y1 - 2019/05/01 UR - http://jnm.snmjournals.org/content/60/supplement_1/254.abstract N2 - 254Objectives: In 2018, the National Institute on Aging and Alzheimer’s Association (NIA-AA) updated the definition for Alzheimer’s disease (AD) to base on in vivo biomarkers of abnormal β-amyloid (Aβ) and pathologic tau deposits instead of clinical symptoms. We had previously reported a cut-off point for preclinical/clinical AD using 11C-PIB PET as a valid in vivo biomarker of Aβ plaque. We hereby, in this prospective study, evaluated the ability of 18F-AV-1451 (T807) PET, a selective tau tracer, to measure the global and regional tau burden in the brain and aimed to define a cut-off point for AD. Methods: During Feb to Nov 2018, 43 patients (M: 17, F: 26; mean age=66.4±8.5 y, range: 50-79 y) with decreased recent memory in 1-2 years, with/without cognitive impairment, but without language dysfunction, tremor or balancing problem, prior history of stoke or other neurodegenerative diseases, were recruited into this study. They all underwent 3-tracers PET/CT: 18FDG (biomarker for prediction of future cognitive decline), 11C-PIB and T807 within 1 week. 18FDG PET was assessed visually for temporal-parietal hypometabolism. Regional and global cortical to cerebellum SUV ratios (SUVRs) were calculated for both 11C-PIB and T807 PET (35 and 85 min post injection respectively). Global SUVR on 11C-PIB PET &gt;= 1.3 was defined as abnormal in Aβ burden. Based on 11C-PIB and 18FDG PET findings, patients were classified as: (1) 11C-PIB+ and 18FDG+ as probable-AD, (2) 11C-PIB+ and 18FDG- as AD pathologic change, (3) 11C-PIB- and 18FDG- as non-AD. Results: According to 11C-PIB and 18FDG PET findings, 17 patients (mean age=70±6.3 y) were diagnosed as probable AD, 4 AD pathologic change (mean age=66±12.5 y) and 22 non-AD (mean age=63.7±8.6 y). Regional T807 SUVR of probable-AD patients exhibited significantly higher tau burden than non-AD patients in lateral temporal (1.45±0.29 vs 1.09±0.06), posterior cingulate (1.40±0.28 vs 1.11±0.08), medial temporal (1.38±0.17 vs 1.10±0.10), occipital (1.37±0.29 vs 1.07±0.06), precuneus (1.35±0.30 vs 1.03±0.06), superior parietal (1.28±0.31 vs 0.97±0.06) and frontal (1.22±0.21 vs 1.00±0.07) regions (all P&lt;0.05). Global T807 binding was also significantly higher in probable-AD than non-AD patients (SUVR=1.31±0.21 vs 1.05±0.05, P&lt;0.05). ROC curve analysis defined the cut-off point of global T807 SUVR for probable-AD as 1.14, with a sensitivity of 82.4% (14/17), specificity 95.5% (21/22), AUC 0.92. Of the 4 AD patients with pathologic change (11C-PIB+ and 18FDG-), 1 patient showed high tau burden (global SUVR=1.32), thus qualified as biological AD without neuronal injury by this new NIA-AA definition. Conclusions: Based on the cut-off SUVR value we found for T807 tau burden, and the 2018 NIA-AA definition, 82.4% (14/17) of the probable-AD patients and 25% (1/4) of AD patients with pathologic change were reclassified as AD, while no change in classification was found in the non-AD patients. SUVR measurement of tau burden with T807 PET may also have potential in future research for severity assessment of the degree of cognitive impairment in AD patients. ER -