Relationship between brain AD biomarkers and episodic memory performance in healthy aging
Introduction
Episodic memory changes are the initial cognitive deficits observed in typical Alzheimer’s disease (AD) and its prodromal stage (McKhann et al. 2011). Subtle memory impairment can also be seen when patients are still in the preclinical stages (Schindler et al. 2017). Consequently, it is essential to distinguish preclinical manifestations from normal episodic memory difficulties that accompany healthy aging for a better detection of people at risk for AD (Tromp et al. 2015). In healthy aging, some aspects of episodic memory decline more than others (Nyberg et al. 2003). Typically, episodic memory decline starts to accelerate over 60 years, with decreased immediate and delayed free recall for verbal information, while recognition memory remains mostly preserved (Albert 1997). In addition, a prominent difficulty in the ability to discriminate similar memory traces is found in healthy aging (Toner et al., 2009, Holden et al., 2013).
One way to identify the nature of memory changes associated with incipient Alzheimer pathology is to characterize memory performance of healthy older individuals in relationship to the presence of AD biomarkers: β-amyloid (Aβ) deposits (Buckner, 2005, Palmqvist et al., 2017), tau neurofibrillary tangles (Braak and Del Tredici 2015) and hippocampal atrophy (Sarazin et al. 2007). Indeed, these neuropathological signs are observed decades before cognitive symptoms (Jack et al. 2010, 2013) and affect brain areas that are part of the episodic memory network (Rugg and Vilberg 2013).
Along those lines, the presence of AD biomarkers in healthy aging match memory changes at the group level in standard neuropsychological tests. Cross-sectional studies reported that, in middle-aged and older participants (mean age: 60 ± 8.4 years), high level of CSF tau/Aβ42 was associated with decreased verbal episodic memory performance in the Free and Cued Selective Reminder Test (FCSRT-free recall) and Logical Memory (Schindler et al. 2017). A positive association was also observed between hippocampal volume and performance in verbal and visual memory tasks in older adults (O’Shea et al. 2016). In addition, in vivo PET imaging studies in cognitively normal adults aged 65 years and older reported that early tau staging in the hippocampus and entorhinal cortex (Braak and Braak 1995) predicts decrease in an episodic memory composite score (Schöll et al., 2016, Maass et al., 2018). In contrast, other studies in similar populations failed to demonstrate a direct association between global cortical Aβ and episodic memory performance (Mormino et al., 2009, Mormino et al., 2011, Maass et al., 2018). Altogether, these studies suggest that the presence of regional tau deposits in the medial temporal lobe (MTL) is mainly associated with decreased episodic memory performance, an impairment also associated with hippocampal atrophy.
Additionally, there is growing interest for a specific episodic memory process called behavioral pattern separation which more specifically recruits the hippocampus (Marks et al. 2017). Behavioral pattern separation refers to the capacity to discriminate between similar memory events (Yassa and Stark 2011) and is typically assessed with the Mnemonic Similarity Task (Stark et al. 2019). In healthy aging, the ability to identify similar lures in an object recognition memory task is positively associated to the volume of the hippocampus, specifically the combined dentate gyrus (DG) and CA3 subfields, and the subiculum (Stark and Stark 2017). Moreover, the association between lure discrimination and hippocampal volume was mediated by the presence of tau in the medial temporal lobe as assessed with tau PET imaging (Marks et al. 2017). Impaired lure discrimination was also observed in patients with Mild Cognitive Impairment (MCI) (Stark et al. 2013). Decreased performance of object pattern separation and recognition memory are respectively associated with decreased DG and CA3 volumes in a group of older adults (over 60 y.) including 18% of MCI (Dillon et al. 2017).
Interestingly, a recent study combining quantitative MRI for hippocampal volume and Aβ in cerebrospinal fluid (CSF) reported that hippocampal atrophy is the main predictive factor for memory decline in older adults (mean age 77 ± 3.5 years), and CSF Aβ levels in a slightly younger population (mean age: 69 ± 2.1 years) (Svenningsson et al. 2019). These results suggest that relationships between memory performance and AD biomarkers are sensitive to age. Intriguingly, previous cross-sectional and longitudinal studies including middle-aged participants (before 60 y.) failed to report associations between hippocampal atrophy and decreased verbal and visual episodic memory performance (Charlton et al., 2010, Paul et al., 2011, Gorbach et al., 2017). Similar results were observed for Aβ deposits (Okonkwo et al., 2014, Mielke et al., 2016), suggesting that sensitivity of AD-related biomarkers to memory performance is lesser during middle, compared to older, age (≥60 y.).
In this context, the aim of the present study was to assess the relationship between several AD biomarkers (Aβ, tau/neuroinflammation, hippocampal atrophy) and several episodic memory processes (free recall, recognition memory, behavioral pattern separation), and to determine whether these associations already exist in a group comprising healthy late middle-aged participants, rather than only in an older population (above 60 y.), as suggested in the literature.
Section snippets
Participants
Sixty healthy French-speaking community dwelling participants in their late middle-age (50–69 y, Table 1) were enrolled in a multimodal cross-sectional study on the relationships between AD biomarkers, sleep-related processes and cognitive aging (COFITAGE). Participants were recruited in senior facilities, by advertising in local newspapers and by word of mouth. No participant reported recent history of psychiatric and neurological disorders, or were taking medication likely to affect the
Relationship between AD biomarkers and memory performance in late middle-aged healthy participants (50–69 y.)
Demographic characteristics, memory performance and AD biomarker values are presented in Table 1.
The GLMM analysis showed no significant association between AD biomarkers (tau/neuroinflammation, Aβ and hippocampal atrophy) and memory performance at the FCSRT, Logical Memory and MST in the whole group of participants (Table 2). The observation of 95% confidence intervals indicated that the possibility that there is no difference between groups should not be ruled out for low statistical power (
Discussion
The main purpose of the present study was to evaluate the associations between the amount of AD-related biomarkers and episodic memory performance in healthy late middle-aged participants and older individuals. To this end, participants completed episodic memory tasks, volumetric MRI for hippocampal atrophy and two PET-scans for tau NFT/neuroinflammation quantification within Braak stage I/II regions (Braak and Braak 1991) and Aβ burden in early Thal stages I/II regions. Contrary to previous
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
We thank C. Hagelstein, E. Lambot, P. Villar González, P. Ghaemmaghami, S. Laloux, A. Claes, B. Herbillon, B. Lauricella, P. Hawotte, C. Le Goff, E. Cavalier, M. Blanpain, L. Pietquin, X. Pépin, M. Cerasuolo, G. Boraita-Amador, E. Tezel, D. Marzoli, L. Veithen, P. Cardone, M. André, C. Mouraux for their help in different steps of the research project.
Author contributions
E.S., P.M., G.V., C.B. and F.C. designed the experiment. L.R., J.N., M.V.E, D.C., G.B., M.B., E.S., P.M., G.V., C.B., and F.C. helped in data acquisition, analysis, and interpretation. M.A.B, E.S., and P.M. provided administrative, technical, or material support. L.R., C.B., and F.C. wrote the manuscript. L.R., J.N., M.V.E, D.C., G.B., M.B., M.A.B., E.S., P.M., G.V., C.B. and F.C. contributed to manuscript revising.
Funding
This work was supported by Fonds National de la Recherche Scientifique (FRS-FNRS, FRSM 3.4516.11, MEMODYN 30446199, Belgium), the Wallonia-Brussels Federation (grant for Concerted Research Actions – SLEEPDEM), University of Liège, Fondation Simone et Pierre Clerdent, European Regional Development Fund (ERDF, Radiomed Project). J.N. was supported by University of Liège. M.V.E., C.S., C.B., G.V., F.C. were supported by the FNRS-Belgium. [18F]Flutemetamol doses were provided and cost covered by GE
References (85)
A fast diffeomorphic image registration algorithm
Neuroimage
(2007)- et al.
Unified segmentation
Neuroimage
(2005) - et al.
Staging of alzheimer’s disease-related neurofibrillary changes
Neurobiology of Aging
(1995) - et al.
Aging and Memory in Humans
- et al.
The relationship between episodic long-term memory and white matter integrity in normal aging
Neuropsychologia
(2010) - et al.
Beta-amyloid and cognitive decline in late middle age: Findings from the Wisconsin Registry for Alzheimer’s Prevention study
Alzheimer’s Dement.
(2016) - et al.
Longitudinal association between hippocampus atrophy and episodic-memory decline
Neurobiology of Aging
(2017) - et al.
Tracking pathophysiological processes in Alzheimer’s disease: An updated hypothetical model of dynamic biomarkers
Lancet Neurology
(2013) - et al.
Hypothetical model of dynamic biomarkers of the Alzheimer’s pathological cascade
Lancet Neurology
(2010) - et al.
Amyloid burden and neural function in people at risk for Alzheimer’s Disease
Neurobiology of Aging
(2014)