Pharmacokinetic assessment of the tau radiotracer [18F]MK6240

Objectives: [¹⁸F]MK6240 was developed for PET imaging of tau protein aggregates which are implicated in Alzheimer’s disease and other neurological disorders. The goal of this work was to provide an initial characterization of [¹⁸F]MK6240 pharmacokinetics in control and mild cognitive impairment (MCI) subjects using metabolite corrected arterial input functions.

Methods: Four subjects (3 control, 1 MCI) underwent [¹⁸F]MK6240 PET scans. Dynamic PET data were measured for 135 min after bolus injection of [¹⁸F]MK6240. Samples drawn from a radial artery catheter were processed to measure whole blood and plasma concentrations. Radiometabolite analysis was performed on selected plasma samples (drawn 3, 5, 8, 15, 30, 60, and 90 min post-injection) using column-switching radio-HPLC with eluent fractions measured in a well-counter. Dynamic PET images were warped to a standardized space using the subject’s T1-weighted structural MRI to allow delineation of regions of interest for extraction of time activity curves. Total volume of distribution (V_T) was estimated using the Logan graphical method and various one- (1T) and two- (2T) tissue compartment models.

Results: Whole blood:plasma ratio stabilized to 0.67 by 20 minutes (range: 0.58-0.8). Percent parent in plasma (%PP) followed a single exponential (T_1/2: 5.8±0.4 min) and primarily polar metabolites were observed on HPLC. At 90 minutes, %PP ranged from 0-10% across individuals. Metabolite corrected plasma activity fit a single exponential after 5 minutes (T_1/2: 4.8±0.4 min). [¹⁸F]MK6240 in gray matter peaked quickly (3-5 SUV at ~5 minutes). The preferred compartmental model was 2Tv where ‘v’ denotes that the vascular fraction was included as a parameter. Regional V_T (ml×cm^-3) in gray matter of controls ranged from 4.6-8.7 and 3.5-6.2 in white matter, and the cerebellum ranged from 5.8-8.6. In the MCI subject, VT in gray matter was 34, 16 in white matter, and 9.6 in the cerebellum. Highest regional V_T was found in the precuneus of the MCI subject with a V_T of 68. Microparameters were relatively stable where regional average of K₁ ranged 0.23-0.34 ml/min/cm³ among subjects. Regionally averaged k₃ (min^-1) was 0.016±0.003 (S1), 0.0145±0.002 (S2) 0.009±0.002 (S3) and 0.037±0.02 (S4[MCI]); k₄ (min^-1) by subject was 0.008±0.003, 0.008±0.003, 0.012±0.005, and 0.004±0.002. Logan plots did not reach linearity within the 135 minute scanning period. In the four subjects examined here, no evidence of skull uptake was observed.

Conclusions: [¹⁸F]MK6240 exhibited rapid clearance from plasma with little to no radioactivity in the parent form at 90 minutes. 2Tv was the preferred compartmental model, yielding good fits and stable microparameters across regions and subjects. Logan blood-based graphical methods did not linearize within the 135-minute scanning period. Future [¹⁸F]MK6240 scans with arterial sampling are scheduled and will provide a more thorough assessment of [¹⁸F]MK6240 kinetics.