Pharmacokinetic Evaluation of the Tau PET Radiotracer 18F-T807 (18F-AV-1451) in Human Subjects

Dustin Wooten, Nicolas J. Guehl, Eline E. Verwer, Timothy M. Shoup, Daniel Yokell, Nevena Zubcevik, Neil Vasdev, Ross Zafonte, Keith A. Johnson, Georges El Fakhri, Marc D. NormandinJournal of Nuclear Medicine 2016 · DOI

Abstract

<sup>18</sup>F-T807 is a PET radiotracer developed for imaging tau protein aggregates, which are implicated in neurologic disorders including Alzheimer disease and traumatic brain injury (TBI). The current study characterizes <sup>18</sup>F-T807 pharmacokinetics in human subjects using dynamic PET imaging and metabolite-corrected arterial input functions. Nine subjects (4 controls, 3 with TBI history, 2 with mild cognitive impairment) underwent dynamic PET imaging for up to 120 min with arterial blood sampling. Total volume of distribution ($V_T$) was estimated using compartmental modeling and graphical analysis. Reference region-based methods, including Logan distribution volume ratio (DVR) and static SUV ratio (SUVR) using the cerebellum as a reference tissue, were explored. Results showed <sup>18</sup>F-T807 has properties suitable for tau quantification, with simplified approaches correlating highly ($R^2 = 0.97$) with gold-standard compartmental modeling.

Introduction

Tau protein aggregates are a hallmark of neurodegenerative diseases like Alzheimer’s and are also implicated in Traumatic Brain Injury (TBI). To study these diseases in living patients, researchers use Positron Emission Tomography (PET) with radiotracers that bind to tau. However, for a tracer to be clinically useful, its pharmacokinetics—how it enters the brain, metabolizes in the blood, and binds to its target—must be thoroughly understood. While the "gold standard" for quantification involves invasive arterial blood sampling, this study investigates whether <sup>18</sup>F-T807 can be reliably quantified using simpler, non-invasive imaging ratios.

Methods

The researchers performed 120-minute dynamic PET scans on 9 subjects (4 healthy controls, 3 with TBI, and 2 with Mild Cognitive Impairment [MCI]). Key technical approaches included:

  • Arterial Input Function: Continuous arterial blood sampling and HPLC analysis were used to measure the concentration of the unmetabolized tracer in the plasma.
  • Compartmental Modeling: They compared 1-tissue and 2-tissue models (1T, 2T, 2Tv) to determine the best fit for tracer kinetics.
  • Graphical Analysis: Logan and Multilinear Analysis 1 (MA1) were used to estimate $V_T$.
  • Reference Region Validation: The cerebellum was tested as a reference region to calculate DVR and SUVR, potentially eliminating the need for arterial lines.

Results and Discussion

Brain Kinetics and Penetration

<sup>18</sup>F-T807 showed high brain penetration, with gray matter uptake peaking quickly (SUV > 2 within ~5 minutes). The tracer demonstrated clear differentiation between clinical groups, with $V_T$ significantly higher in MCI subjects compared to controls in expected regions like the temporal and parietal cortices.

Regional time-activity curves (TACs) for subjects S3, S5, and S9, and cerebellum TACs across all subjects.

As shown in Figure 1, the cerebellum exhibited the lowest variability and fastest washout, confirming its suitability as a reference region for this tracer in the studied populations.

Validation of Simplified Quantification

The study's most significant finding was the extremely high correlation between complex, invasive modeling and simplified imaging metrics. The 2-tissue model with blood volume correction (2Tv) provided the best fit for the data, but simplified metrics like Logan DVR (at $t^* = 55$ min) and static SUVR (measured 80–100 minutes post-injection) were nearly identical in their results.

Validation of reference region methods showing high correlation between simplified metrics (Logan-DVR, SUVR) and the invasive 2Tv model.

As illustrated in Figure 6, the correlation between the simplified SUVR$_{80-100}$ and the gold-standard 2Tv model reached an $R^2$ of 0.96 ($P < 0.0001$). This indicates that a 20-minute static scan can provide a highly accurate measurement of tau burden without the need for arterial sampling or 2-hour dynamic imaging.

Conclusion

This study provides the "green light" for using <sup>18</sup>F-T807 in clinical research using simplified protocols. While compartmental modeling remains the gold standard for initial tracer characterization, the high accuracy of the cerebellar reference tissue approach (specifically SUVR from 80–100 min) makes <sup>18</sup>F-T807 a practical and powerful tool for large-scale studies of tau pathology in Alzheimer disease and TBI.

Summary updated

Abstract

<sup>18</sup>F-T807 is a PET radiotracer developed for imaging tau protein aggregates, which are implicated in neurologic disorders including Alzheimer disease and traumatic brain injury (TBI). The current study characterizes <sup>18</sup>F-T807 pharmacokinetics in human subjects using dynamic PET imaging and metabolite-corrected arterial input functions. <b>Methods:</b> Nine subjects (4 controls, 3 with a history of TBI, 2 with mild cognitive impairment due to suspected Alzheimer disease) underwent dynamic PET imaging for up to 120 min after bolus injection of <sup>18</sup>F-T807 with arterial blood sampling. Total volume of distribution (<i>V</i><sub>T</sub>) was estimated using compartmental modeling (1- and 2-tissue configurations) and graphical analysis techniques (Logan and multilinear analysis 1 [MA1] regression methods). Reference region-based methods of quantification were explored including Logan distribution volume ratio (DVR) and static SUV ratio (SUVR) using the cerebellum as a reference tissue. <b>Results:</b> The percentage of unmetabolized <sup>18</sup>F-T807 in plasma followed a single exponential with a half-life of 17.0 ± 4.2 min. Metabolite-corrected plasma radioactivity concentration fit a biexponential (half-lives, 18.1 ± 5.8 and 2.4 ± 0.5 min). <sup>18</sup>F-T807 in gray matter peaked quickly (SUV > 2 at ∼5 min). Compartmental modeling resulted in good fits, and the 2-tissue model with estimated blood volume correction (2Tv) performed best, particularly in regions with elevated binding. <i>V</i><sub>T</sub> was greater in mild cognitive impairment subjects than controls in the occipital, parietal, and temporal cortices as well as the posterior cingulate gyrus, precuneus, and mesial temporal cortex. High focal uptake was found in the posterior corpus callosum of a TBI subject. Plots from Logan and MA1 graphical methods became linear by 30 min, yielding regional estimates of <i>V</i><sub>T</sub> in excellent agreement with compartmental analysis and providing high-quality parametric maps when applied in voxelwise fashion. Reference region-based approaches including Logan DVR (<i>t</i>* = 55 min) and SUVR (80- to 100-min interval) were highly correlated with DVR estimated using 2Tv (<i>R</i><sup>2</sup> = 0.97, <i>P</i> < 0.0001). <b>Conclusion:</b><sup>18</sup>F-T807 showed rapid clearance from plasma and properties suitable for tau quantification with PET. Furthermore, simplified approaches using DVR (<i>t</i>* = 55 min) and static SUVR (80-100 min) with cerebellar reference tissue were found to correlate highly with compartmental modeling outcomes.
BibTeX
@article{wooten2016pharmacokinetic,
  author = {Dustin Wooten and Nicolas J. Guehl and Eline E. Verwer and Timothy M. Shoup and Daniel Yokell and Nevena Zubcevik and Neil Vasdev and Ross Zafonte and Keith A. Johnson and Georges El Fakhri and Marc D. Normandin},
  title = {{Pharmacokinetic Evaluation of the Tau PET Radiotracer <sup>18</sup>F-T807 (<sup>18</sup>F-AV-1451) in Human Subjects}},
  journal = {Journal of Nuclear Medicine},
  year = {2016},
  doi = {10.2967/jnumed.115.170910},
}
Figure