Additional research is needed to better understand the relationship between early tau accumulation and cognitive decline in cognitively healthy (CH) older adults. The tau PET tracer 18F-MK-6240 has shown favorable imaging characteristics and is an ideal candidate to identify early tau accumulation in CH older adults. In the present study, we evaluated the associations between in vivo tau levels and retrospective 5-year cognitive change.

Using 18F-MK-6240 PET, we evaluated tau accumulation in 41 CH participants from Cognitive Reserve (CR) and Reference Ability Neural Network (RANN) studies. We investigated the relationships between regional PET signal and retrospective cognitive change, focusing in three cognitive domains well-established to change with aging: episodic memory, speed processing and reasoning. Latent change score analysis was applied to generate latent variables estimating the change in these domains from baseline to follow-up. Regarding tau data, we created meta-ROIs based on 16 AD-vulnerable subregions selected a priori. The tau SUVR of the subregions were averaged to create four meta-ROIs: 1) Total-AD ROI comprising all 16 areas; 2) Medial Temporal Lobe-ROI (MTL-ROI), 3) Lateral Temporal Lobe-ROI (LTL-ROI) and Cingulate/Parietal Lobe (C/P-ROI). The associations between regional tau levels and retrospective cognitive change over 5-years were investigated through regression models adjusted by age, sex, education, and baseline cognitive performance.

The mean age of the sample was 67.5 years (SD=5.8, range 55 to 77 years), 53.8% were male, 63.4% White, 26.1% Black, and 7.3% Latinx. Overall, the sample was highly educated (16.20 ± 2.26) and presented high IQ scores (118.97 ± 7,76) based on American National Adult Reading Test. All participants were CH and showed cognitive decline over 5 years in all cognitive domains analyzed. Most participants were classified as Braak stage 0 (82.9%), some as Braak stage 1 (14.6%) and one participant as Braak stage 2 (2.4%). The regression models showed that higher [18F]MK-6240 SUVR was associated with steeper decline in memory and speed in the Total-AD meta-ROI (memory: p = .03; speed: p = .04), and in the regional ROIs, such as LCL (memory: p = .03; speed: p = .04), C/P (memory: p = .02, speed: p = .05) and MTL (memory: p = .05, speed: p = .02), with large effect sizes (f2) (>.40) for memory and medium effect sizes for speed (.20-.25). There were no associations between [18F]MK-6240 SUVR and reasoning change.

Our finding reinforces the notion that pathological tau in areas of early accumulation influence changes in cognitive domains known to be affected in AD even in cognitively normal individuals. The novel contribution of this work is the relevance of tau accumulation beyond episodic memory, as we observed its association with speed proceeding decline. The fact reasoning decline is commonly observed in normal aging, but here not associated with tau suggest the specificity of the tau-cognition associations. Our results should be considered with caution due to the modest sample size.