Neuroimaging markers associated with maintenance of optimal memory performance in late-life

Background: Age-related memory decline has been well-documented; however, some individuals reach their 8th - 10th decade while maintaining strong memory performance. Objective: To determine which demographic and biomarker factors differentiated top memory performers (aged 75 +, top 20% for memory) from their peers and whether top memory performance was maintained over 3 years. Methods: Clinically normal adults (n = 125, CDR = 0; age: 79.5 +/- 3.57 years) from the Harvard Aging Brain Study underwent cognitive testing and neuroimaging (amyloid PET, MRI) at baseline and 3-year follow-up. Participants were grouped into Optimal (n = 25) vs. Typical (n = 100) performers using performance on 3 challenging memory measures. Non-parametric tests were used to compare groups. Results: There were no differences in age, sex, or education between Optimal vs. Typical performers. The Optimal group performed better in Processing Speed (p = 0.016) and Executive Functioning (p < 0.001). Optimal performers had larger hippocampal volumes at baseline compared with Typical Performers (p = 0.027) but no differences in amyloid burden (p = 0.442). Twenty-three of the 25 Optimal performers had longitudinal data and16 maintained top memory performance while 7 declined. Non-Maintainers additionally declined in Executive Functioning but not Processing Speed. Longitudinally, there were no hippocampal volume differences between Maintainers and Non-Maintainers, however Non-Maintainers exhibited higher amyloid burden at baseline in contrast with Maintainers (p = 0.008). Conclusions: Excellent memory performance in late life does not guarantee protection against cognitive decline. Those who maintain an optimal memory into the 8th and 9th decades may have lower levels of AD pathology.