Head‐to‐head comparison of [18F]MK‐6240 and [18F]flortaucipir (AV‐1451) in autosomal dominant Alzheimer disease

Abstract

AbstractBackgroundThe accumulation of misfolded tau proteins in the brain is an early pathological indicator of Alzheimer disease (AD). The introduction of [18F]flortaucipir (AV‐1451), has enabled early investigations of tau spatial dispersion in vivo but has prominent off‐target binding in the choroid plexus and basal ganglia. More recent tau tracers, such as [18F]MK‐6240, may improve the ability to detect pathology, but direct comparisons between these imaging agents are limited. We provide preliminary data from a direct head‐to‐head comparison of AV‐1451 and MK‐6240 in a sample of asymptomatic and symptomatic autosomal dominant AD (ADAD) participants and healthy controls.MethodTau PET imaging using AV‐1451 and MK‐6240 was acquired on consecutive days for 14 adult Dominantly Inherited Alzheimer Network (DIAN) participants. [11C]PiB‐PET amyloid and T1‐weighted head MR scans were also collected. MR images were segmented into 40 bilateral regions of interest (ROIs) using FreeSurfer‐5.3. ROI standardized uptake value ratios (SUVrs) were calculated for the 80‐100 minute and 90‐110 minute post‐injection time windows for AV‐1451 and MK‐6240, respectively, with cerebellar gray as the reference region. Partial volume correction (PVC) was applied using a geometric transfer matrix approach. For analysis, bilateral SUVrs were correlated between the tau tracers to assess their spatial and quantitative agreement.ResultExemplar voxel‐wise PET images are shown for ADAD participants in Figure 1 and healthy controls in Figure 2. Visually, MK‐6240 appears to capture many of the high binding regions of AV‐1451. MK‐6240 also displayed diffuse signal along pial/leptomeningeal surfaces in 42% of participants, regardless of amyloid status. This may relate to arterial, venous, or glymphatic uptake. Quantitatively, regional SUVrs between the two tracers for tauopathy regions, such as inferior temporal and precuneus, were strongly correlated before and after PVC (see Figure 3). PVC reduced correlation strengths in areas thought to have minimal true tau uptake. The tracers were highly correlated amongst amyloid(+, n=2) but less so amongst amyloid(‐, n=12; see Figure 4).ConclusionMK‐6240 and AV‐1451 display a relatively strong quantitative and spatial relationship amongst those with AD. Future studies should further investigate off‐target leptomeningial binding in MK‐6240 and its influence in quantitative tau PET studies.