Test-Retest performance of [18F]MK-6240 tau burden and relative delivery indices in cognitively normal older subjects using PET/MRI

Abstract

Background: Accurate interpretation of quantitative PET outcomes hinges on understanding the test-retest variability (T-RT). Previous studies of the tau-PET ligand [¹⁸F]MK-6240 reported adequate T-RT performance of tau burden estimates over a short-term 21-day and over a longer-term 6-month T-RT period, primarily involving Alzheimer’s disease (AD) and cognitively normal (CN) subjects, respectively. However, several T-RT characteristics have not yet been reported, particularly in older CN (oCN) subjects. Here, we investigate the short-term T-RT performance of dynamic [18F]MK-6240 outcomes in a group largely consisting of oCN. We report T-RT for uptake in potential reference regions, for extracerebral off-target signal, and for estimates of tau burden and relative delivery indices in tau-bearing target regions. Eight participants (7 oCN, 1 AD) underwent baseline dynamic [¹⁸F]MK-6240 PET/MRI (Biograph mMR) and a retest follow-up PET/MRI scan within approximately 3 weeks. T-RT was evaluated using absolute percentage differences and interclass correlation coefficients (ICC) in three groups of regions: 1) potential reference regions using standardized-uptake-values 90-110 minutes post-injection (SUV_90-110min); 2) target regions using SUV ratios (SUVR_90-110min), distribution volume ratios (DVR), and relative delivery (R₁); and 3) extracerebral region using SUVR_90-110min. A voxel-based partial volume correction (PVC) was applied. T-RT was evaluated with and without PVC. Results: In oCN subjects, the SUV_90-110min T-RT in the evaluated reference regions ranged from 6-11% (ICC > 0.9); target region T-RT was similar for SUVR_90-110min (4-9%, ICC: 0.62-0.97), DVR (3-6%, ICC: 0.66-0.92), and R₁ (3-11%, ICC: 0.77-0.92). PVC had minimal impact on reference region SUV_90-110min T-RT, but increased target region T-RT variability (SUVR_90-110min: 10-26%; DVR: 6-15%; R₁: 4-14%). Extracerebral SUVR_90-110min exhibited higher T-RT variability (~12%, ICC: 0.85) than other target regions (average 6%) and increased to ~15% after PVC. Conclusion: Our findings are consistent with previous reports and provide further evidence of acceptable [¹⁸F]MK-6240 T-RT in low-signal oCN subjects. Our results suggest [¹⁸F]MK-6240 is suitable for detecting early tau deposition and longitudinal changes over time, and further support the viability of [¹⁸F]MK-6240 R₁ to evaluate longitudinal changes in perfusion. However, the extracerebral signal exhibited higher T-RT variability than other target and reference regions and may affect their signal.