Much enthusiasm has centered on the recently developed positron emission tomography (PET) radiotracers designed to be selective for tau deposits. Initial results are encouraging in that these compounds bind to Alzheimer's disease (AD) brain in a manner consistent with tau pathology. However, several important questions remain for all of the tau tracers including: 1) the spectrum of tau pathology to which they bind (i.e., AD vs. non-AD tau deposits); 2) the identity of off-target (non-tau) binding; and 3) the accuracy of commonly used analysis methods across the full spectrum of severity of tau pathology. The first two issues are best addressed by postmortem studies and we will address these in Project-3. The latter issue is best addressed by in vivo pharmacokinetic studies and we will do that in Project-4. The key issue addressed in Project-4 is whether the commonly employed short acquisition time (i.e., 80-100 min after bolus injection of [F-18]AV-1451) followed by analysis using tissue ratios (TR) yields values directly proportional to the number of tau binding sites across the full range of pathology. This proportionality requires a steady-state period during which a ?stable? measure of TR can be obtained. There is evidence that this may not be true for [F-18]AV-1451 - especially at the high-end of tau pathology (see preliminary data). We will address this issue by using the same detailed, fully dynamic (FD) pharmacokinetic analysis approach that we used to characterize PiB. The FD method does not rely on equilibrium between compartments (i.e., tissues or blood), but instead obtains quantitative binding measures by fitting a pharmacokinetic model (with arterial blood as input) to the tissue kinetics allowing accurate quantitation of tracer binding across the full spectrum of pathology. Thus, the FD method can serve as a standard of truthto assess the veracity of simplified methods. In Project-4, we will compare FD measures with 80-100 min bolus-TR measures across a wide range of [F-18]AV-1451 retention levels to determine if the bolus-TR method accurately reflects tau load at the high-end. We will also use the FD data to generate: 1) a simplified reference tissue model (SRTM) and 2) parameters for a bolus-plus-constant-infusion (B+I) method ? a technique to achieve stable TR when a bolus injection cannot. We will then determine whether the SRTM and B+I measures approximate FD measures better than bolus-TR. Because arterial FD methods are not routinely performed, even in research centers, we will attempt to translate the findings from our FD analysis to improved and widely-applicable methods by identifying: 1) a mathematical correction for the bolus-TR outcome measure; 2) a SRTM with improved performance relative to bolus-TR; and 3) a B+I method also with improved performance. We will study this same phenomenon in 24-month change in tracer retention. While we use [F-18]AV-1451 in Projects-1 and -2 in a way that is not critically dependent on high-end accuracy, understanding this issue of applicability across the full spectrum of pathology is necessary before the results of [F-18]AV-1451 scanning in natural history studies and drug trials can be fully understood.
Showing the most recent 10 out of 106 publications
