Evaluation of a novel PET tracer 18F-Florbetazine for Alzheimer’s disease diagnosis and amyloid deposition quantification

Introduction: ¹⁸F-Florbetazine (¹⁸F-92) is a selective PET tracer for β-amyloid (Aβ) depositions with a novel diaryl-azine scaffold to reduce lipophilicity and non-specific binding in white matter. We aimed to assess the kinetic profiles of ¹⁸F-Florbetazine in human subjects using dynamic PET.¹⁸F-Florbetazine (¹⁸F-92) is a selective PET tracer for β-amyloid (Aβ) depositions with a novel diaryl-azine scaffold to reduce lipophilicity and non-specific binding in white matter. We aimed to assess the kinetic profiles of ¹⁸F-Florbetazine in human subjects using dynamic PET.¹⁸F-Florbetazine (¹⁸F-92) is a selective PET tracer for β-amyloid (Aβ) depositions with a novel diaryl-azine scaffold to reduce lipophilicity and non-specific binding in white matter. We aimed to assess the kinetic profiles of ¹⁸F-Florbetazine in human subjects using dynamic PET.

Methods: Nine probable Alzheimer’s disease (AD) patients and six healthy controls (HCs) underwent dynamic PET examination with ¹⁸F-Florbetazine and a structural MRI scan. List-mode PET data is reconstructed in both dynamic series and 6 static frames (15 min per frame). Voxel-based simplified reference tissue mode 2 is performed on dynamic PET series with the cerebellar cortex volume of interest (VOI) as the reference tissue, resulting in non-displaceable binding potential (BPnd) maps. Standardized uptake value ratio (SUVR), BPnd, and distribution volume ratio (DVR) in 6 cerebral cortex VOIs were compared between AD and HC groups. The ratio of SUV of cerebral cortex to cerebral white matter was calculated.

Results: In AD patients, radioactivity was high in the cerebral cortex, expected to contain Aβ and low in the cerebellar cortex, whereas in HCs, radioactivity was homogenously low. The uptake in the cerebellar cortex was close between AD patients and HCs. BPnd in the cerebral cortex was significantly higher in AD patients (0.74 ± 0.23) than in HCs (0.09 ± 0.03). A cut-off value between 0.17-0.23 for BPnd or 1.17-1.23 for DVR could discriminate AD patients from HCs with 100% accuracy. Cerebral cortex SUVR of both groups from 30- min in 15 min intervals highly correlated with DVR, while correlation and slope of SUVR to DVR linear regression increase along with time (r from 0.85 to 0.93, slope from 0.66 to 1.58). Cerebral cortex SUVR from 30- min could again discriminate AD patients from HCs with 100% accuracy. In AD patients, the ratio of SUV of cerebral cortex to cerebral white matter remained over 1.3 for all time intervals: 2.06 ± 0.28, 1.57 ± 0.20, 1.36 ± 0.22, 1.30 ± 0.15, 1.31 ± 0.12, and 1.32 ± 0.14, respectively, for 0-15, 15-30, 30-45, 45-60, 60-75, and 75-90 min time frame.

Conclusions: ¹⁸F-Florbetazine satisfies the requirements as an Aβ radioligand for diagnostic use in AD. The fast clearance allows an early static scan to quantify Aβ binding with ¹⁸F-Florbetazine. High gray matter/white matter contrast aids visual assessment, especially in cases of little Aβ deposition. Weighing in kinetic profile, efficacy to differentiate AD and HC subjects, gray-to-white matter ratio, and fidelity of semi-quantitative measurement, an early static scan starting from 30- min is optimal for ¹⁸F-Florbetazine.

• Download figure
• Open in new tab
• Download powerpoint