Associations Between Cognitive Complaints,Memory Performance,Mood,and Amyloid-β Accumulation in Healthy Amyloid Negative Late-Midlife Individuals

Abstract

Background:

Cognitive complaints are gaining more attention as they may represent an early marker of increased risk for AD in individuals without objective decline at standard neuropsychological examination.

Objective:

Our aim was to assess whether cognitive complaints in late middle-aged individuals not seeking medical help are related to objective cognitive outcomes known as early markers for AD risk, concomitant affective state, and amyloid-β (Aβ) burden.

Methods:

Eighty-seven community-based cognitively normal individuals aged 50–69 years underwent neuropsychological assessment for global cognition, using Preclinical Alzheimer’s Cognitive Composite 5 (PACC5) score, and a more specific episodic memory measure. Affective state was based on self-assessment questionnaires for depression and anxiety. Aβ PET burden was assessed via [¹⁸F]Flutemetamol (N = 84) and [¹⁸F]Florbetapir (N = 3) uptake. Cognitive complaints were evaluated using Cognitive Difficulties Scale.

Results:

Higher cognitive complaints were significantly associated with lower episodic memory performance and worse affective state. Moreover, higher level of cognitive complaints was related to higher (but still sub-clinical) global Aβ accumulation (at uncorrected significance level). Importantly, all three aspects remained significant when taken together in the same statistical model, indicating that they explained distinct parts of variance.

Conclusion:

In healthy Aβ negative late middle-aged individuals, a higher degree of cognitive complaints is associated with lower episodic memory efficiency, more anxiety and depression, as well as, potentially, with higher Aβ burden, suggesting that complaints might signal subtle decline. Future studies should untangle how cognitive complaints in healthy aging populations are related to longitudinal changes in objective cognition and AD biomarker correlates.

Keywords

Affective disorders amyloid cognition cognitive decline episodic memory middle-age

INTRODUCTION

Preclinical Alzheimer’s disease (AD) is characterized by normal cognition on standard neuropsychological tests and the presence of AD-related biomarkers referring to pathological protein accumulation in the brain, i.e., amyloid-β (Aβ) plaques and tau neurofibrillary tangles, irrespective of the status of neurodegeneration [1 –3]. However, measurement of AD biomarkers using cerebrospinal fluid (CSF) assessment or positron emission tomography (PET) is invasive, demanding, and costly. In addition, the abnormal value for these biomarkers in cognitively normal individuals does not predict AD symptom outbreak but rather represents a marker of increased risk for developing AD [4]. Therefore, the identification of other less invasive markers, which could help to detect individuals with a higher risk for AD-related pathologies in the general population without initial invasive assessment, is of particular importance.

The concept of subjective cognitive decline (SCD) has been developed in that framework [5 –7]. SCD might be described as a broad spectrum of cognitive complaints reported by individuals without objective decline at neuropsychological examination [5]. Although not specific, SCD is associated with an increased risk for subsequently developing AD, and may also be considered as a preclinical AD stage preceding mild cognitive impairment (MCI) [7]. According to its definition, the presence of SCD is not linked to impaired objective cognition. Nevertheless, in most cases, cognitive scores match patients’ complaints, such that higher level of SCD is related to lower (but still normal) performance on global cognition [8, 9], episodic memory or executive tasks [10].

A series of self and informant based questionnaires have been developed to quantify cognitive difficulties experienced by individuals in the general population or memory clinics [7 , 11–16]. Unfortunately, there is no consensus about which questionnaire and cut-off differentiate best SCD individuals from those without SCD. The SCD-Initiative [17] recommended the use of two main criteria for SCD diagnosis in clinical settings: 1) self-experience of persistent decline in cognition compared to previous normal status, a condition which is not related to an acute event; 2) normal performance on standardized neuropsychological tests, adjusted for sex, age, and education. The latter criterion, i.e., lack of objective cognitive decline, is the reason why SCD is not considered as a clinical category in the Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5). In contrast, medical conditions characterized by objective cognitive decline, such as MCI and Alzheimer’s disease correspond to mild and major neurocognitive disorder respectively (mild NCD or major NCD) in DSM-5 classification [18]. It is also important to note, that one diagnostic criterion of mild NCD includes “concern of the individual, a knowledgeable informant, or the clinician” for “evidence of modest cognitive decline from a previous level of performance in one or more cognitive domains” [19]. Therefore, in terms of DSM-5, SCD could sometimes, but not always, be a precursor to mild NCD. Furthermore, active medical help seeking is considered as an additional risk factor of future objective cognitive decline [17, 20]. Only a fraction of individuals with cognitive concerns seek for medical help, however [21].

Age-related brain changes associated with AD, including deposits of Aβ and tau proteins, begin as early as in late middle-age (i.e., ∼ 50 y) [22 –25]. Similarly, several cognitive functions such as processing speed, episodic memory and executive functions begin to decline already in midlife [26 –31]. Previous studies focusing on SCD included patients that were usually 70 years or older [32 –38], whereas middle-aged cohorts remain underrepresented [39, 40], and only few studies took Apolipoprotein E (APOE) polymorphism, the most important AD-related genetic risk factor, into account [34, 39]. Consequently, participants included in these studies may represent a group with a higher AD risk [17]. Furthermore, the link between SCD and affective state is not well understood and whether affective issues can be observed beyond objective cognitive difficulties is not established [41 –43]. Therefore, assessing whether cognitive complaints in late midlife are related to AD-related features both at the cognitive and brain level is of prime importance to further establish SCD as an early marker for AD risk, that could provide a large time window for prevention strategies [44, 45].

Consequently, the present study aimed to explore whether cognitive complaints in late middle-aged individuals (50–69 years) not seeking medical help are related to objective cognitive outcomes known as early markers for AD risk, or if other variables associated to increased risk for AD (depression and anxiety status, Aβ PET burden) explain the subjective feeling of cognitive decline. Objective cognitive measures, affective state and continuous Aβ PET values were considered jointly to assess whether they differentially explain cognitive complaints, including APOE genotype as a covariate. Given the good health status and relatively young age of our target population, we expected that Aβ burden would remain low (or sub-clinical). Our hypothesis was that higher level of cognitive complaints would be related to lower (but still within normal range) scores for global cognition and episodic memory, worse subclinical affective state, and higher global Aβ burden.

MATERIALS AND METHODS

Experimental design

Participants were enrolled in a multi-modal study designed to identify biomarkers and lifestyle factors associated with normal cognitive aging in the context of preclinical dementia (the Cognitive Fitness in Aging study, COFITAGE, with trial registration number EudraCT 2016-001436-35; see [46]). In that context, we assessed objective cognition through neuropsychological examination, cognitive complaints, and affective state via self-report questionnaires, and we also assessed Aβ via PET imaging. All participants also underwent quantitative multi-parametric magnetic resonance imaging (MRI) acquisitions for subsequent pre-processing of PET data.

Participants

Participants were healthy late middle-aged (50 to 69 years old) community-dwelling French speaking men and women (Table 1; N = 87; 60 women [69%]) not actively seeking for help in memory clinics. They were recruited in social events for seniors, by advertising in local newspapers and by word of mouth. The study was presented as an exploration of the biological and psychological mechanisms implicated in sleep, cognition, and aging. No participants reported any recent history of neurological or psychiatric disease or were taking medication likely to affect the central nervous system. All had normal or corrected-to-normal vision and hearing. Other exclusion criteria were sleep disorders, assessed during an in-lab night of sleep under polysomnography, body mass index < 18 and > 31 kg/m², smoking, psychoactive drug consumption, excessive consumption of caffeine (> 4 cups/day) or alcohol (> 14 units/week), diabetes, and shift-work. Participants with clinical levels of depression or anxiety as assessed by the Beck Depression Inventory [47] and by the Beck Anxiety Inventory [48], respectively, and/or with ongoing pharmacological treatment were excluded. A few participants reached mild depression and/or anxiety levels according to Beck scales but this was not considered as clinical. Participants with treated (> 6 months) hypertension and hypothyroidism were included. All participants showed normal performance on the Mattis Dementia Rating Scale [49] (i.e., score > 130/144). The procedures involving experiments on human subjects were in accordance with guidelines on human experimentation and were approved by the local Ethics Committee of the Faculty of Medicine (University of Liege). All participants gave their signed informed consent prior to the experiment and received a financial compensation.

Table 1

Descriptive statistics of demographical data, subjective factors, objective cognition, Aβ, and APOE (N = 87)

	Mean	SD	Min	Max
Demographical data
Age, y	59.16	5.22	50.00	69.00
Sex, female, n (%)				60 (69%)
Ethnic status, Caucasian, n (%)				87 (100%)
Education, y	15.10	3.07	9.00	25.00
Educational level:
Primary School, n (%)				0.0 (0%)
Secondary School, n (%)				23 (26.4%)
Bachelor’s degree, n (%)				33 (37.9%)
Master’s degree, n (%)				25 (28.7%)
PhD or higher, n (%)				6 (6.9%)
Subjective factors
Subjective cognitive complaints (CDS)	28.77	19.42	0.0	95.00
Depression (BDI)	5.21	4.53	0.0	17.00
Anxiety (BAI)	2.98	3.34	0.0	17.00
Objective cognition
PACC5 score	0.0	2.93	–7.92	8.21
FCSRT, Free and Total recall	81.09	6.63	66.00	92.00
Logical Memory Task, delayed recall	12.26	3.79	3.00	22.00
DSST, total 2 min score	73.10	12.66	39.00	99.00
Semantic Fluency Test, 1 min score	20.41	4.21	13.00	37.00
Mattis Dementia Rating Scale	142.48	1.92	134.00	144.00
Aβ accumulation
Global Mask, CL	–5.15	5.65	–15.85	7.36
PCC and Precuneus Mask, CL	–0.10	7.89	–23.93	20.72
APOE
APOE ɛ4 carriers, n (%)				16 (18.4%)
APOE ɛ4 non-carriers, n (%)				71 (81.6%)

SD, standard deviation; CDS, Cognitive Difficulties Scale; BDI, Beck Depression Inventory; BAI, Beck Anxiety Inventory; FCSRT, Free and Cued Selective Reminding Test; DSST, Digit Symbol Substitution Test; Aβ, amyloid-β; CL, Centiloid units; PCC, posterior cingulate cortex; APOE, Apolipoprotein E. See Methods section for the references of the tests.

Neuropsychological examination

Neuropsychological examination consisted of a battery of cognitive tasks assessing short-term and episodic memory, executive, and attentional functions. Original Preclinical Alzheimer’s Cognitive Composite 5 (PACC5) score [50, 51] was computed as the sum of z-scores of the following cognitive measures: Free and Total Recall in the Free and Cued Selective Reminding Test (FCSRT) [52], Delayed Recall in the Logical Memory Test [53], Total score in the Digit Symbol Substitution Test [54], scores in the Verbal Fluency Test for the categories of Animals, Fruits and Vegetables (1 min each), and Mini-Mental State Examination [55]. Here we introduced three changes to the initial PACC5: Mini-Mental State Examination was replaced by the score from the Mattis Dementia Rating Scale, the Verbal Fluency Test score was calculated for the animal category only, and a more recent version of the Digit Symbol Substitution Test having a larger range of scores was used [56].

The episodic memory score was assessed through the FCSRT [52], which, as a list learning test, was proposed by the European Prevention of Alzheimer’s Dementia (EPAD) program as a sensitive measure for early cognitive decline [57]. The raw score of episodic memory, i.e., the sum of free recall and total recall, was converted to z-scores for our analyses.

Neuropsychological evaluation was performed during 2 sessions taking approximately 75 min each. As our study was designed for several research objectives, additional neuropsychological tasks were included in the assessment, which are not mentioned here.

Cognitive complaint scores

These were obtained with the Cognitive Difficulties Scale (CDS) [11]. In this questionnaire, participants had to evaluate their cognitive difficulties during the last 3 weeks. There were 39 statements (e.g., “I have trouble recalling frequently used phone numbers”), where individuals had to indicate their cognitive difficulties on a 5-point Likert scale between 0 (“never”) and 4 (“very often”), with higher value meaning more cognitive complaints. This questionnaire was already shown to be a sensitive measure of cognitive complaints in studies seeking associations with Aβ burden [58, 59].

Affective state

Affective state was based on two questionnaires collected after one of the neuropsychological evaluation sessions, the Beck Depression Inventory (BDI) [47], and the Beck Anxiety Inventory (BAI) [48]. We computed a composite score for affective state based on the average of z-scores of total scores for each questionnaire, and final score was itself converted to z-scores. Higher value means worse affective state (i.e., more depression and more anxiety).

Quantitative multi-parametric MRI acquisition

Quantitative multi-parametric MRI acquisition was performed on a 3-Tesla MR scanner (Siemens MAGNETOM Prisma, Siemens Healthineers, Erlangen, Germany). Quantitative maps were obtained by combining images using different parameters sensitive to distinct tissue properties. Multi-parameter mapping was based on multi-echo 3D fast low angle shot at 1 mm isotropic resolution [60]. This included three datasets with T1, proton density (PD), and magnetization transfer (MT)–weighted contrasts imposed by the choice of the flip angle (FA = 6° for PD & MT, 21° for T1) and the application of an additional off-resonance Gaussian-shaped RF pulse for the MT-weighted acquisition. Fieldmap data for the correction of B1 transmit and receiver fields, and B0 inhomogeneity were also acquired.

Quantitative multi-parametric MRI processing

MRI multi-parameter maps were processed with the hMRI toolbox [61] (http://hmri.info) and SPM12 (Welcome Trust Centre for Neuroimaging, London, UK) to obtain notably a quantitative MT map as well as segmented tissue maps (grey matter, white matter, CSF), using the “unified segmentation” approach. [62]. Inter-subject alignment into the standard MNI space was performed with diffeomorphic anatomical registration using exponentiated lie algebra (DARTEL), providing subject specific deformation flow-fields.

PET acquisition and processing

Aβ PET imaging was performed with radiotracers [¹⁸F]Flutemetamol for 97 participants and with [¹⁸F]Florbetapir for 3 participants on an ECAT EXACT HR+ scanner (Siemens, Erlangen, Germany). For all radiotracers, participants received a single dose of the respective radio-ligands in an antecubital vein (target dose 185±10%MBq). Aβ-PET image acquisitions started 85 min after injection, and 4 frames of 5 min were obtained. All PET images were reconstructed using filtered back-projection algorithm including corrections for measured attenuation (10 min transmission scan using three retrac table 68Ge line sources), dead time, random events, and scatter using standard software (Siemens ECAT –HR+ V7.1, Siemens/CTI, Knoxville, TN, USA).

Before proceeding with further PET processing, a PET average image was created using all frames. Averaged PET images were first manually reoriented then automatically co-registered to the structural MT map in subject space. Then, deformation flow-fields derived by DARTEL from the MT maps were applied to averaged co-registered PET images [63]. Standardized uptake value ratio (SUVR) was calculated using the whole cerebellum as the reference region for Aβ-PET [64]. Volumes of interest (VOIs) were determined using the masks provided by automated anatomical labelling (AAL) atlas [65]. As Aβ-PET images were obtained using two radiotracers, their SUVR values were scaled to Centiloid units for common scale [64 , 66–68]. Finally, we applied two masks for the estimations of Aβ uptake. One mask considered global brain Aβ burden and included lateral and medial cortices in frontal, parietal, temporal, occipital brain regions, and basal ganglia together with anterior and posterior cingulate cortex. Another, more specific mask, considered brain regions which show early and rapid Aβ increase in Aβ negative individuals [69], and included posterior cingulate cortex (PCC) and precuneus. Given the relatively young age and good health of participants, they were all Aβ negative (Table 1).

APOE genotyping

Genotyping was performed based on blood sample DNA extraction. Common single nucleotide polymorphisms (SNPs) were assessed using Infinium OmniExpress-24 BeadChip (Illumina, San Diego, CA, USA) based on human genome build hg19 (GRCh37). Genotype imputation was performed using the “Sanger Imputation Server” (https://imputation.sanger.ac.uk/) by choosing Haplotype Reference Consortium (release 1.1) (HRC) [70] as reference panel and the pre-phasing algorithm Eagle2 [71]. APOE ɛ variants were determined by rs7412 and rs429358 SNPs. Participants were classified into ɛ4 carriers (heterozygous and homozygous) and non-carriers, based on their APOE status.

Statistical analyses

All statistical analyses were performed with SAS 9.4 for Windows (SAS Institute, Cary, NC, USA). Generalized linear mixed models (GLMM; PROC GLIMMIX) were applied to compute all statistics. Dependent variable distributions were estimated using allfitdist function in MATLAB (developed by Mike Sheppard, part of the MvCAT package [72]) and set accordingly for each GLMM. Collinearity diagnosis was performed on all predictors using Tolerance (TOL) and Variance Inflation Factors (VIF) as criteria via PROC REG function on SAS. Degrees of freedom (DF) were estimated using Kenward-Roger’s correction. Subject (intercept) effect was included as a random factor. Since we computed 5 separate models (prior to the composite model), we defined p-value≤0.01 as significant following the Bonferroni correction for multiple comparisons.

All models included cognitive complaints as the dependent variable, and controlled for sex, age, education, and APOE status. GLMM first evaluated the association with cognitive performance in two different models, one with PACC5 values, and another with the episodic memory score from the PACC5. Then, models sought to evaluate the link between cognitive complaints and subclinical affective state. Next GLMMs evaluated the association between subjective cognition and AD biomarker (Aβ) in two separate models, one with Aβ burden in global brain mask, whereas another one with the more specific mask targeting regions related to early Aβ pathophysiology. A final GLMM sought to evaluate if the significant predictors of cognitive complaints that were identified in each previous step (episodic memory, affective state, and Aβ) remained significant when assessed simultaneously in one model. Semi-partial R² ( $R_{sp}^{2}$ ) was reported for each significant effect as described previously [73].

RESULTS

Descriptive statistics about demographic data, cognitive complaints, affective status, cognitive performance, Aβ and APOE status are presented in Table 1.

Links between cognitive complaints and cognitive performance

We tested whether CDS score was linked to PACC5 and episodic memory measure in two separate models. The first GLMM showed no significant association between CDS score and PACC5 (Table 2A). The second model revealed a significant negative association between CDS score and episodic memory measure, indicating that a higher level of cognitive complaints is linked to worse performance on the FCSRT task, as well as an uncorrected positive association between education and CDS in the second model, such that higher education may be associated with higher cognitive concerns (Table 2B; Fig. 1A, 1D). Sex, age, and APOE allele (i.e., ɛ4 carriers versus non-carriers) were not associated with cognitive complaints in either model.

Table 2

Statistical outcome of the GLMM seeking for associations between cognitive complaint level (dependent variable), and: A) PACC5; B) episodic memory measure

Model A	Estimate±SE	F value (df)	p	CI
Sex^*	–0.13±0.21	0.38 (1,81)	0.542	[–0.54, 0.29]
Age	–0.03±0.09	0.13 (1,81)	0.720	[–0.21, 0.15]
Education	0.18±0.10	3.34 (1,81)	0.071	[–0.02, 0.37]
APOE	–0.06±0.23	0.06 (1,81)	0.801	[–0.52, 0.40]
PACC5	–0.15±0.10	2.42 (1,81)	0.124	[–0.35, 0.04]
Model B	Estimate±SE	F value (df)	p	CI
Sex^*	–0.18±0.20	0.80 (1,81)	0.373	[–0.56, 0.21]
Age	–0.02±0.09	0.04 (1,81)	0.851	[–0.19, 0.16]
Education	0.18±0.09	4.11 (1,81)	0.046 ( $R_{sp}^{2}$ = 0.05)	[0.003, 0.36]
APOE	–0.10±0.22	0.22 (1,81)	0.644	[–0.55, 0.34]
Episodic memory (FCSRT)	–0.28±0.09	9.67 (1,81)	0.003 ( $R_{sp}^{2}$ = 0.11)	[–0.46, –0.10]

SE, standard error; df, degrees of freedom; CI, confidence interval of 95%; APOE, Apolipoprotein E; FCSRT, Free and Cued Selective Reminding Test. Grey background represents control variables included in all GLMM. Significant associations at uncorrected level are in italic, while those meeting criteria of correction for multiple tests are in bold. ^*Sex: 1 = male; 2 = female.

Fig. 1

Scatter plots visualizing the associations between cognitive complaints (higher score refers to worse cognitive difficulties) and A) Memory performance (higher is better); B) Affective state (higher is worse); C) Aβ accumulation in global brain mask (higher means worse); D) Education (higher is better). Simple regressions were used for visual display only, and not as a substitute for the full GLMM statistics.

Association between level of cognitive complaints and affective state

GLMM showed that the level of cognitive complaints was positively associated with affective state, at uncorrected p-value level but, though close to it, did not reach corrected level. (Table 3; Fig. 1B). In contrast, sex, age, education and APOE genotype were not associated with cognitive complaints.

Table 3

Statistical outcome of the GLMM seeking for associations between cognitive complaint level (dependent variable) and affective state

	Estimate±SE	F value (df)	p	CI
Sex^*	0.04±0.19	0.05 (1,81)	0.825	[–0.34, 0.43]
Age	–0.01±0.09	0.02 (1,81)	0.876	[–0.19, 0.16]
Education	0.14±0.09	2.54 (1,81)	0.115	[–0.04, 0.32]
APOE	–0.04±0.23	0.04 (1,81)	0.844	[–0.50, 0.41]
Affective state¹	0.22±0.09	6.26 (1,81)	0.014 ( $R_{sp}^{2}$ = 0.07)	[0.05, 0.40]

SE, standard error; df, degrees of freedom; CI, confidence interval of 95%; Rsp², Semi-partial R²; APOE, Apolipoprotein E. Grey background represents control variables included in all GLMM. Significant associations at uncorrected level are in italic, while those meeting criteria of correction for multiple tests are in bold. ^*Sex: 1 = male; 2 = female. ¹Higher value means worse affective state.

Relationship between level of cognitive complaints and Aβ burden

Then, we tested whether CDS score was associated to global Aβ burden (i.e., using a brain mask covering the entire brain, except the cerebellum) and to early amyloid pathophysiology regions (i.e., posterior cingulate cortex and precuneus, see [69]). The first GLMM yielded a positive association between CDS score and global Aβ accumulation (Table 4A; Fig. 1C), but at uncorrected p value level only. In contrast, the second GLMM did not show any significant association between CDS score and Aβ over early AD pathophysiology regions (Table 4B). Again, sex, age, education and APOE genotype were not associated with cognitive complaints.

Table 4

Statistical outcome of the GLMM seeking for associations between cognitive complaint level (dependent variable) and: A) Aβ global mask; B) Aβ specific mask

Model A	Estimate±SE	F value (df)	p	CI
Sex^*	0.04±0.20	0.03 (1,81)	0.853	[–0.36, 0.43]
Age	–0.04±0.09	0.19 (1,81)	0.665	[–0.22, 0.14]
Education	0.16±0.09	3.07 (1,81)	0.083	[–0.02, 0.34]
APOE	0.04±0.24	0.04 (1,81)	0.849	[–0.42, 0.51]
Aβ (global)	0.19±0.09	4.02 (1,81)	0.048 ( $R_{sp}^{2}$ = 0.05)	[0.001, 0.38]
Model B	Estimate±SE	F value (df)	p	CI
Sex^*	–0.01±0.20	0.0 (1,81)	0.977	[–0.41, 0.40]
Age	–0.03±0.09	0.14 (1,81)	0.708	[–0.22, 0.15]
Education	0.14±0.10	2.10 (1,81)	0.152	[–0.05, 0.33]
APOE	–0.04±0.24	0.03 (1,81)	0.859	[–0.52, 0.43]
Aβ (specific)	0.03±0.10	0.11 (1,81)	0.737	[–0.16, 0.23]

SE, standard error; df, degrees of freedom; CI, confidence interval of 95%; Rsp², Semi-partial R²; APOE, Apolipoprotein E; Aβ, amyloid-β. Grey background represents control variables included in all GLMM. Significant associations at uncorrected level are in italic, while those meeting criteria of correction for multiple tests are in bold. ^*Sex: 1 = male; 2 = female.

Simultaneous associations between level of cognitive complaints and episodic memory, affective state, and Aβ burden

In our final step, we considered jointly the single parameters significantly related to CDS score at uncorrected (or corrected in case of FCSRT score) p values in the previous models. This final GLMM revealed that episodic memory performance, affective state, and global Aβ remained associated at similar significance level with subjective cognitive score (Table 5; Fig. 1). Moreover, education was significantly positively linked to CDS, indicating that higher education may be associated with higher cognitive concerns (Table 5; Fig. 1D). As for the previous models, sex, age, and APOE genotype were not associated with cognitive complaints.

Table 5

Statistical outcome of the GLMM seeking for associations between cognitive complaint level (dependent variable), episodic memory, affective state, and Aβ global mask

	Estimate±SE	F value (df)	p	CI
Sex^*	–0.05±0.19	0.06 (1,79)	0.812	[–0.43, 0.34]
Age	–0.003±0.08	0.0 (1,79)	0.971	[–0.17, 0.16]
Education	0.21±0.09	6.07 (1,79)	0.016 ( $R_{sp}^{2}$ = 0.07)	[0.04, 0.39]
APOE	0.0004±0.22	0.0 (1,79)	0.999	[–0.44, 0.44]
Episodic memory (FCSRT)	–0.24±0.09	7.59 (1,79)	0.007 ( $R_{sp}^{2}$ = 0.09)	[–0.42, –0.07]
Affective state¹	0.20±0.08	5.50 (1,79)	0.022 ( $R_{sp}^{2}$ = 0.06)	[0.03, 0.37]
Aβ (global)	0.18±0.09	4.10 (1,79)	0.046 ( $R_{sp}^{2}$ = 0.05)	[0.003, 0.35]

SE, standard error; df, degrees of freedom; CI, confidence interval of 95%; Rsp², Semi-partial R²; APOE, Apolipoprotein E; FCSRT, Free and Cued Selective Reminding Test; Aβ, amyloid-β. Grey background represents control variables included in all GLMM. ^*Sex: 1 = male; 2 = female. ¹Higher value means worse affective state.

DISCUSSION

SCD is a recent concept that might represent a preclinical stage within AD pathophysiology indicative of an increased risk for developing AD [7]. Its links with various cognitive or physiological aspects have been assessed multiple times already, but the simultaneous association with several behavioral and brain aspects remains scarce, especially in individuals not seeking for clinical help. The main objective of this study was therefore to assess how subjective feeling of one’s own cognitive abilities is related to objective cognitive outcomes known as early markers of AD (global cognitive outcome using PACC5 measure, and verbal episodic memory performance for a list learning task), concomitant subclinical affective state, and in vivo Aβ PET global brain uptake (taken as a continuous value) in cognitively normal and amyloid-β negative late middle-aged individuals, controlling for sex, age, education and APOE status.

To evaluate cognitive complaints we used the Cognitive Difficulties Scale [11], which meets current consensus as the optimal method to assess one’s perception of cognitive status [6 , 59]. Our analyses revealed that a higher level of cognitive complaints was related to a worse score on the episodic memory task, but not to PACC5 outcome. We further report that cognitive complaints may also be linked to worse subjective affective state, while they were also potentially associated with a higher Aβ brain accumulation when considered over the entire brain, but not when focusing only on the brain regions considered to be first affected by Aβ accumulation over the course of AD pathophysiology. Since these correlates of cognitive complaints remained similarly associated with CDS when included together in a single statistical model, we further suggest that episodic memory, together with a potential contribution of affective state and whole-brain Aβ burden, explain at least partially distinct parts of variance in cognitive complaints.

It is important to note that the definition of SCD posits that cognitive complaints are not necessarily limited to the memory domain, but may also include other cognitive domains [5, 74]. Previous cross-sectional studies have also reported that, while remaining clinically normal, a higher SCD level is related to lower performance on global cognition [8 , 76], and more specific cognitive functions, including verbal episodic memory and executive functions [10]. Our results show that, in late midlife, cognitive complaints in cognitively normal individuals are associated with episodic memory functioning, but not with a global measure of cognition known to be sensitive to the earliest AD-related cognitive changes. We may therefore extend previous findings by showing that if cognitive complaints do not reach a status leading to help seeking, which would allow formal clinical SCD diagnosis, we presume that episodic memory performance may be the main indicator one uses to judge his/her cognitive status.

This link must have underlying neural correlates, including changes in brain structural and functional integrity. Atrophy in the regions of medial temporal lobe (MTL) is often viewed as the primary target to account for alterations in episodic memory. For example, Perrotin and colleagues have reported that higher level of self-reported cognitive difficulties among seniors was associated with hippocampal atrophy, at least in a group seeking medical help [58]. The decline in verbal episodic memory, measured using the same FCSRT task as in our study, was also related to MTL atrophy in healthy seniors (mean age 72.7 years) in a longitudinal study [77]. Episodic memory functioning is not only limited to MTL integrity, but also includes medial and lateral prefrontal cortex, precuneus, posterior cingulate cortex [78], regions which were included in our global brain Aβ PET estimation.

Our analyses revealed that cognitive complaints are weakly related to a higher global Aβ burden (at uncorrected significance level), but not in the specific mask related to early Aβ pathophysiology. Whereas previously some negative findings were reported [79], most of the previous studies reported that a higher level of SCD is related to a higher Aβ PET accumulation [32–34 , 80–83]. Most of previous studies have included, however, older participants (around 70 years) from both the general community and memory clinic, thus concentrating on SCD individuals with increased risk for AD [17]. Here, we add therefore that cognitive complaints in healthy midlife (mean age 59 years) community-based individuals, not seeking for medical help, could also correspond to a higher global Aβ concentration, even when the values of the latter biomarker remain in sub-threshold level for amyloid positivity. Our attempt to investigate the regional specificity of the association lead to a negative result when including the brain regions first affected by Aβ plaque progression. It seems therefore that cognitive complaints may reflect somehow the diffused global low level of Aβ rather the initial step of Aβ brain pathology. Furthermore, our results related to Aβ may be also affected by PET processing method. We used two radiotracers to estimate Aβ burden in global and specific brain masks. In order to standardize uptake values of two different radiotracers, we have validated standardized Centiloid pipeline according to our particular data [64 , 67] and computed scaling equations allowing to convert SUVR of two radiotracers into common Centiloid units. However, this standardized pipeline has one possible limitation, as it does not include partial volume correction (PVC) as a PET pre-processing step, which is often used to correct for spill-in effects from adjacent tissue classes [84]. Therefore, in Supplementary Material, we have repeated our statistical analyses with Aβ data using PVC, but only with one radiotracer (Flutemetamol), as inclusion of this pre-processing step does not allow to apply Centiloid scaling equations to convert the SUVR values of two radiotracers into Centiloid units Our statistical models in Supplementary Material showed that Aβ was no longer a significant predictor of cognitive complaints. These results may be explained by a smaller sample size than the main sample, or by the fact that PVC can sometimes increase sensitivity to inaccuracy in image registration and segmentation [85]. Thus, these last results do not allow us to imply firm conclusions about the exact role of Aβ burden in the context of cognitive complaints. Yet, further investigations are needed to confirm our relatively weak association and establish whether the regional distribution of Aβ matters in its association with cognitive complaints using voxel-based analyses.

Surprisingly and contrary to our hypothesis, we found no associations between PACC5 score and cognitive complaints. It might be that PACC5, while considered as a sensitive measure, is limited by one of its components, the Mini-Mental State Examination score, which in our study was replaced by the Mattis Dementia Rating Scale. Both scores are designed as short dementia screening tests, but display a ceiling effect in cognitively healthy individuals [86], and it was even suggested that PACC score should be computed without such a measure [87]. In addition, our study did not assess the associations of cognitive complaints with other specific cognitive domains (e.g., processing speed, attention, executive functions, language, visuospatial abilities), thus, we cannot dismiss the possibility of other cognitive correlates of cognitive complaints.

On top of the link between memory performance and Aβ burden, and not surprisingly, we found that cognitive complaints may be linked (at relatively low but still uncorrected p value level) to worse affective state in a healthy study sample that do not meet criteria for a clinical affective condition. Previous studies have also reported that SCD individuals show higher levels of depressive symptoms [8 , 42] and anxiety [43 , 88] than healthy controls. According to current views, worse affective state may reflect reaction to cognitive complaints [42], or may have a common or a separate underlying cause [17], such as hypothalamic-pituitary-adrenal axis hyperactivity [89].

While it was not our main goal, we found that higher education is potentially related to more cognitive complaints. These results are in line with previous studies reporting that better education is linked to SCD [8 , 91]. Interestingly, it was also shown that when better educated individuals reached objective cognitive decline, they reported more self-perceived difficulties, whereas lower educated seniors did not show concomitant cognitive complaints when reaching objective cognitive decline [92]. It may be that individuals with a higher educational status have also a better self-perception of their cognitive status. Alternatively, it may also be that high cognitive efficiency in better educated individuals is more important to them due to professional or more personal reasons. The concept of cognitive reserve also suggests that in the presence of brain pathology, persons with lower education would reach clinically significant cognitive impairment faster than individuals with better education, who may be able to compensate for longer time but can sometimes express higher level of cognitive complaints [93]. For example, one study demonstrated that the link between SCD and amyloid level becomes stronger with greater educational attainment [94].

We neither observed an effect of APOE ɛ4 status for the level of cognitive complaints. A systematic review [95] reported APOE ɛ4 allele frequency is comparable between SCD individuals and healthy controls (see, however, [96]). Nonetheless, it seems that APOE ɛ4 carrier status in SCD individuals increases the probability for clinical progression to MCI [97]. These results come from studies which mostly studied clinically defined SCD individuals, and not cognitive complaint in a healthy population, as it was in our case.

While previous data indicate that age is associated to the level of memory complaints [33], and that the age for onset of SCD [17] can even be more predictive of dementia risk [34, 98], we did not observe an association between age and level of cognitive complaints. However, the age range was between 50 and 69 years in our study, thus limiting the possible age effect for cognitive complaints.

From a more clinical viewpoint, it seems important to emphasize that the symptom of subjective cognitive complaints as risk factor for AD is rather unspecific and should be associated to other minimally invasive biomarkers to identify high-risk individuals in primary care setting. For example, MRI-based lower hippocampal volume was related to poorer memory performance in a group of elderly individuals with subjective memory complaints [99]. Blood-based biomarkers are also gaining attention, as it was reported that high folate and low bilirubin levels in SCD individuals were related to cognitive decline at 3-year follow-up [100], while decreased total antioxidant and increased serum homocysteine levels were related to more subjective memory complaints [101]. Moreover, tau protein burden is also a main AD-related biomarkers [4]. Interestingly, blood-based tau biomarker (plasma tau phosphorylated at threonine 181; P-tau 181) was recently suggested as an early AD biomarker, even preceding PET-tau [102]. Furthermore, the presence of cardiovascular disease and depressive symptoms history may also indicate about possible progression or reverting in cognitive decline [103].

We stress that our study bears some limitations, the first one being its cross-sectional nature which controls less efficiently for different environmental effects or life experiences, and, therefore, does not allow to approach causal interpretation as longitudinal approaches would do. For instance, the increase in Aβ over time may be more important for cognitive decline than baseline amyloid level [104]. Another possible limitation in our study is selection bias due to the requirements of the larger project in which our volunteers were enrolled. Our participants underwent quite extensive screening for health and affective status, consumption of psychoactive substances (e.g., alcohol, drugs, caffeine), and their educational level was high. This type of selection bias is however also present in SCD studies, sometimes including highly educated volunteers from academic memory clinic or community [33, 80]. In addition, we did find weak behavioral and Aβ correlates of cognitive complaints despite this rather excellent health status, preventing any bias from potential comorbidities or unspecific heath conditions. Finally, our study did not include the evaluation of worry associated with subjective cognitive complaints. In fact, SCD accompanied with worry can greatly increase dementia risk, and was therefore included in SCD plus category [7, 17]. As we evaluated healthy volunteers from general population and many of them had quite low level of cognitive complaints (see Table 1), we did not ask a question about possible worry over their cognitive status. Future studies assessing similar population may particularly consider inclusion of the question about worry, as it might better characterize cognitive complaints.

To sum up, our results provide new insights about the associations between cognitive complaints, objective cognition, and AD biomarker status. We demonstrate that as early as in late midlife, cognitive complaints may be considered as an important marker for higher sub-threshold amyloid accumulation. Moreover, we show that, beyond brain integrity, perception of cognitive status has some actual ground which relies mostly on the episodic memory and affective domains. Future studies are needed to untangle how cognitive complaints in community-based late middle-aged population is related to longitudinal changes in objective cognition and AD biomarker correlates.

Footnotes

ACKNOWLEDGMENTS

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, and C. Mouraux for their help in different steps of the research project.

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 –grant 17/21-09 SLEEPDEM), University of Liège, Fondation Simone et Pierre Clerdent, European Regional Development Fund (ERDF, Radiomed Project), the SAO-FRA (grant 15018). 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. VM is supported by a European Research Council starting grant (CS; ERC-StG 757763). []Flutemetamol doses were provided and cost covered by GE Healthcare Ltd (Little Chalfont, UK) as part of an investigator sponsored study (ISS290) agreement. This agreement had no influence on the protocol and results of the study reported here. None of funding sources had an involvement in study design; in the collection, analysis, interpretation of data; in the writing of the report; in the decision to submit the article for publication.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Dubois

, Hampel

, Feldman

, Scheltens

, Aisen

, Andrieu

, Bakardjian

, Benali

, Bertram

, Blennow

, Broich

, Cavedo

, Crutch

, Dartigues

, Duyckaerts

, Epelbaum

, Frisoni

, Gauthier

, Genthon

, Gouw

, Habert

, Holtzman

, Kivipelto

, Lista

, Molinuevo

, O’Bryant

, Rabinovici

, Rowe

, Salloway

, Schneider

, Sperling

, Teichmann

, Carrillo

, Cummings

, Jack

(2016) Preclinical Alzheimer’s disease: Definition, natural history, and diagnostic criteria. Alzheimers Dement 12, 292–323.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

, Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

Sperling

, Aisen

, Beckett

, Bennett

, Craft

, Fagan

, Iwatsubo

, Jack

, Kaye

, Montine

, Park

, Reiman

, Rowe

, Siemers

, Stern

, Yaffe

, Carrillo

, Thies

, Morrison-Bogorad

, Wagster

, Phelps

(2011) Toward defining the preclinical stages of Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 280–292.

Jack

, Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

, Elliott

, Masliah

, Ryan

, Silverberg

(2018) NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimers Dement 14, 535–562.

Jessen

, Amariglio

, Van Boxtel

, Breteler

, Ceccaldi

, Chételat

, Dubois

, Dufouil

, Ellis

, Van Der Flier

, Glodzik

, Van Harten

, De Leon

, McHugh

, Mielke

, Molinuevo

, Mosconi

, Osorio

, Perrotin

, Petersen

, Rabin

, Rami

, Reisberg

, Rentz

, Sachdev

, De La Sayette

, Saykin

, Scheltens

, Shulman

, Slavin

, Sperling

, Stewart

, Uspenskaya

, Vellas

, Visser

, Wagner

(2014) A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer’s disease. Alzheimers Dement 10, 844–852.

Rabin

, Smart

, Crane

, Amariglio

, Berman

, Boada

, Buckley

, Chételat

, Dubois

, Ellis

, Gifford

, Jefferson

, Jessen

, Katz

, Lipton

, Luck

, Maruff

, Mielke

, Molinuevo

, Naeem

, Perrotin

, Petersen

, Rami

, Reisberg

, Rentz

, Riedel-Heller

, Risacher

, Rodriguez

, Sachdev

, Saykin

, Slavin

, Snitz

, Sperling

, Tandetnik

, van der Flier

, Wagner

, Wolfsgruber

, Sikkes

SAM

(2015) Subjective cognitive decline in older adults: An overview of self-report measures used across 19 international research studies. J Alzheimers Dis 48(Suppl 1), S63–86.

Jessen

, Wiese

, Bachmann

, Eifflaender-Gorfer

(2010) Prediction of dementia by subjective memory impairment. Arch Gen Psychiatry 67, 414–422.

Garcia-Ptacek

, Cavallin

, Kåreholt

, Kramberger

, Winblad

, Jelic

, Eriksdotter

(2014) Subjective cognitive impairment subjects in our clinical practice. Dement Geriatr Cogn Dis Extra 4, 419–30.

Jacinto

, Brucki

, Porto

, Martins

, Nitrini

(2014) Subjective memory complaints in the elderly: A sign of cognitive impairment? Clinics (Sao Paulo) 69, 194–197.

10.

Langlois

, Belleville

(2014) Subjective cognitive complaint in healthy older adults: Identification of major domains and relation to objective performance. Aging Neuropsychol Cogn 21, 257–282.

11.

McNair

, Kahn

(1984) Self-assessment of cognitive deficits. In Assessment in geriatric psychopharmacology, Crook T, Ferris S, Bartus R, eds. Mark Powley, New Canaan.

12.

Farias

, Mungas

, Reed

, Cahn-Weiner

, Jagust

, Baynes

, Decarli

(2008) The measurement of everyday cognition (ECog): Scale development and psychometric properties. Neuropsychology 22, 531–44.

13.

Rattanabannakit

, Risacher

, Gao

, Lane

, Brown

, McDonald

, Unverzagt

, Apostolova

, Saykin

, Farlow

(2016) The cognitive change index as a measure of self and informant perception of cognitive decline: Relation to neuropsychological tests. J Alzheimers Dis 51, 1145–1155.

14.

Rami

, Mollica

, Garcfa-Sanchez

, Saldafia

, Sanchez

, Sala

, Valls-Pedret

, Castellvi

, Olives

, Molinuevo

(2014) The subjective cognitive decline questionnaire (SCD-Q): A validation study. J Alzheimers Dis 41, 453–466.

15.

Miebach

, Wolfsgruber

, Polcher

, Peters

, Menne

, Luther

, Incesoy

, Priller

, Spruth

, Altenstein

, Buerger

, Catak

, Janowitz

, Perneczky

, Utecht

, Laske

, Buchmann

, Schneider

, Fliessbach

, Kalbhen

, Heneka

, Brosseron

, Spottke

, Roy

, Teipel

, Kilimann

, Wiltfang

, Bartels

, Düzel

, Dobisch

, Metzger

, Meiberth

, Ramirez

, Jessen

, Wagner

(2019) Which features of subjective cognitive decline are related to amyloid pathology? Findings from the DELCODE study. Alzheimers Res Ther 11, 66.

16.

Ferreira

, Machado

, Molina

, Nieto

, Correia

, Westman

, Barroso

(2017) Cognitive variability during middle-age: Possible association with neurodegeneration and cognitive reserve. Front Aging Neurosci 9, 188.

17.

Jessen

, Amariglio

, Buckley

, van der Flier

, Han

, Molinuevo

, Rabin

, Rentz

, Rodriguez-Gomez

, Saykin

, Sikkes

SAM

, Smart

, Wolfsgruber

, Wagner

(2020) The characterisation of subjective cognitive decline. Lancet Neurol 19, 271–278.

18.

Sachdev

, Blacker

, Blazer

, Ganguli

, Jeste

D V.

, Paulsen

, Petersen

(2014) Classifying neurocognitive disorders: The DSM-5 approach. Nat Rev Neurol 10, 634–642.

19.

American Psychiatric Association (2013) Diagnostic and Statistical Manual of Mental Disorders. American Psychiatric Association, Washington, DC.

20.

Slot

RER

, Sikkes

SAM

, Berkhof

, Brodaty

, Buckley

, Cavedo

, Dardiotis

, Guillo-Benarous

, Hampel

, Kochan

, Lista

, Luck

, Maruff

, Molinuevo

, Kornhuber

, Reisberg

, Riedel-Heller

, Risacher

, Roehr

, Sachdev

, Scarmeas

, Scheltens

, Shulman

, Saykin

, Verfaillie

SCJ

, Visser

, Vos

SJB

, Wagner

, Wolfsgruber

, Jessen

, Boada

, de Deyn

, Jones

, Frisoni

, Spiru

, Nobili

, Freund-Levi

, Soininen

, Verhey

, Wallin Å

, Touchon

, Rikkert

, Rigaud

, Bullock

, Tsolaki

, Vellas

, Wilcock

, Froelich

, Bakardjian

, Benali

, Bertin

, Bonheur

, Boukadida

, Boukerrou

, Chiesa

, Colliot

, Dubois

, Epelbaum

, Gagliardi

, Genthon

, Habert

, Houot

, Kas

, Lamari

, Levy

, Metzinger

, Mochel

, Nyasse

, Poisson

, Potier

, Revillon

, Santos

, Andrade

, Sole

, Surtee

, Thiebaud de Schotten

, Vergallo

, Younsi

, van der Flier

(2019) Subjective cognitive decline and rates of incident Alzheimer’s disease and non-Alzheimer’s disease dementia. Alzheimers Dement 15, 465–476.

21.

Waldorff

, Rishoj

, Waldemar

(2008) If you don’t ask (about memory), they probably won’t tell. J Fam Pract 57, 41–44.

22.

Ritchie

, Ritchie

, Jaffe

, Skoog

, Scarmeas

(2015) Is late-onset Alzheimer’s disease really a disease of midlife? Alzheimers Dement (N Y) 1, 122–130.

23.

Jack

, Holtzman

(2013) Biomarker modeling of Alzheimer’s disease. Neuron 80, 1347–1358.

24.

Beason-Held

, Goh

, An

, Kraut

, O’Brien

, Ferrucci

, Resnick

(2013) Changes in brain function occur years before the onset of cognitive impairment. J Neurosci 33, 18008–18014.

25.

Coupé

, Manjón

, Lanuza

, Catheline

(2019) Lifespan changes of the human brain in Alzheimer’s disease. Sci Rep 9, 3998.

26.

Eckert (2010) Age-related changes in processing speed: Unique contributions of cerebellar and prefrontal cortex. Front Hum Neurosci 4, 10.

27.

Fjell

, Sneve

, Grydeland

, Storsve

, Walhovd

(2017) The disconnected brain and executive function decline in aging. Cereb Cortex 27, 2303–2317.

28.

Harada

, Natelson Love

, Triebel

(2013) Normal cognitive aging. Clin Geriatr Med 29, 737–752.

29.

Hoogendam

, Hofman

, Van Der Geest

, Van Der Lugt

, Ikram

(2014) Patterns of cognitive function in aging: The Rotterdam Study. Eur J Epidemiol 29, 133–140.

30.

Olaya

, Bobak

, Haro

, Demakakos

(2017) Trajectories of verbal episodic memory in middle-aged and older adults: Evidence from the English Longitudinal Study of Ageing. J Am Geriatr Soc 65, 1274–1281.

31.

Rönnlund

, Nyberg

, Bäckman

, Nilsson

L-G

(2005) Stability, growth, and decline in adult life span development of declarative memory: Cross-sectional and longitudinal data from a population-based study. Psychol Aging 20, 3–18.

32.

Buckley

, Hanseeuw

, Schultz

, Vannini

, Aghjayan

, Properzi

, Jackson

, Mormino

, Rentz

, Sperling

, Johnson

, Amariglio

(2017) Region-specific association of subjective cognitive decline with tauopathy independent of global β-amyloid burden. JAMA Neurol 74, 1455–1463.

33.

Amariglio

, Mormino

, Pietras

, Marshall

, Vannini

, Johnson

, Sperling

, Rentz

(2015) Subjective cognitive concerns, amyloid-β, and neurodegeneration in clinically normal elderly. Neurology 85, 56–62.

34.

Zwan

, Villemagne

, Doré

, Buckley

, Bourgeat

, Veljanoski

, Salvado

, Williams

, Margison

, Rembach

, Macaulay

, Martins

, Ames

, Van Der Flier

, Ellis

, Scheltens

, Masters

, Rowe

(2016) Subjective memory complaints in APOE ɛ4 carriers are associated with high amyloid-β burden. J Alzheimers Dis 49, 1115–1122.

35.

Teipel

, Cavedo

, Weschke

, Grothe

, Rojkova

, Fontaine

, Dauphinot

, Gonzalez-Escamilla

, Potier

, Bertin

, Habert

, Dubois

, Hampel

, Christelle

, Hugo

, Laurie

, Federica

, Enrica

, Patrizia

, Stanley

, Stephane

, Geoffroy

, Remy

, Pailine

, Aurelie

, Marcel

, Simone

, Christiane

, Francis

, Catherine

, Stephie

, Marie

, Katrine

, Perrine

, Katia

, Antonio

, Valérie

, Marine

, Caroline

, Bruno

, Harald

, Hovagim

, Habib

, Olivier

, Marie-Odile

, Foudil

, Fanny

, Marie-Claude

, de Schotten Michel

(2017) Cortical amyloid accumulation is associated with alterations of structural integrity in older people with subjective memory complaints. Neurobiol Aging 57, 143–152.

36.

Shokouhi

, Conley

, Baker

, Albert

, Kang

, Gwirtsman

, Newhouse

(2019) The relationship between domain-specific subjective cognitive decline and Alzheimer’s pathology in normal elderly adults. Neurobiol Aging 81, 22–29.

37.

Mielke

, Wiste

, Stephen Weigand

, David Knopman

, Lowe

, Roberts

, Yonas Geda

, Swenson-Dravis

, Bradley Boeve

, Senjem

, Prashanthi Vemuri

, Petersen

, Jack

(2012) Indicators of amyloid burden in a population-based study of cognitively normal elderly. Neurology 79, 1570–1577.

38.

Toledo

, Bjerke

, Chen

, Rozycki

, Jack

Jr , Weiner

, Arnold

, Reiman

, Davatzikos

, Shaw

, Trojanowski

; Alzheimer’s Disease Neuroimaging Initiative (2015) Memory, executive, and multidomain subtle cognitive impairment: Clinical and biomarker findings. Neurology 85, 144–53.

39.

Mosconi

, De Santi

, Brys

, Tsui

, Pirraglia

, Glodzik-Sobanska

, Rich

, Switalski

, Mehta

, Pratico

, Zinkowski

, Blennow

, de Leon

(2008) Hypometabolism and altered cerebrospinal fluid markers in normal apolipoprotein E E4 carriers with subjective memory complaints. Biol Psychiatry 63, 609–618.

40.

Grambaite

, Hessen

, Auning

, Aarsland

, Selnes

, Fladby

(2013) Correlates of subjective and mild cognitive impairment: Depressive symptoms and CSF biomarkers. Dement Geriatr Cogn Dis Extra 3, 291–300.

41.

Weisenbach

, Boore

, Kales

(2012) Depression and cognitive impairment in older adults. Curr Psychiatry Rep 14, 280–288.

42.

Hill

, Mogle

, Wion

, Munoz

, DePasquale

, Yevchak

, Parisi

(2016) Subjective cognitive impairment and affective symptoms: A systematic review. Gerontologist 56, e109–e127.

43.

Sun

, Wang

, Dong

, Lu

, Scheininger

, Ewers

, Jessen

, Zuo

, Han

(2019) Anxiety correlates with cortical surface area in subjective cognitive decline: APOE ɛ4 carriers versus APOE ɛ4 non-carriers. Alzheimers Res Ther 11, 50.

44.

Livingston

, Sommerlad

, Orgeta

, Costafreda

, Huntley

, Ames

, Ballard

, Banerjee

, Burns

, Cohen-Mansfield

, Cooper

, Fox

, Gitlin

, Howard

, Kales

, Larson

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2017) Dementia prevention, intervention, and care. Lancet 390, 2673–2734.

45.

Hurt

, Burns

, Brown

, Barrowclough

(2012) Why don’t older adults with subjective memory complaints seek help? Int J Geriatr Psychiatry 27, 394–400.

46.

Narbutas

, Van Egroo

, Chylinski

, González

, Jimenez

, Besson

, Ghaemmaghami

, Hammad

, Muto

, Schmidt

, Luxen

, Salmon

, Maquet

, Bastin

, Vandewalle

, Collette

(2019) Cognitive efficiency in late midlife is linked to lifestyle characteristics and allostatic load. Aging (Albany NY) 11, 7169–7186.

47.

Beck

, Ward

, Mendelson

, Mock

, Erbaugh

(1961) An inventory for measuring depression. Arch Gen Psychiatry 4, 561–571.

48.

Beck

, Epstein

, Brown

, Steer

(1988) An inventory for measuring clinical anxiety: Psychometric properties. J Consult Clin Psychol 56, 893–897.

49.

Mattis (1976) Mental status examination for organic mental syndrome in the elderly patients. In Geriatrics Psychiatry: A Handbook for Psychiatrists and Primary Care Physicians, Bellak L, Karasu T, New York, pp. 77–121.

50.

Donohue

, Sperling

, Salmon

, Rentz

, Raman

, Thomas

, Weiner

, Aisen

(2014) The preclinical Alzheimer cognitive composite: Measuring amyloid-related decline. JAMA Neurol 71, 961–970.

51.

Papp

, Rentz

, Orlovsky

, Sperling

, Mormino

(2017) Optimizing the preclinical Alzheimer’s cognitive composite with semantic processing: The PACC5. Alzheimers Dement (N Y) 3, 668–677.

52.

Grober

, Hall

, Lipton

, Alan

, Resnick

, Kawas

(2009) NIH Public Access. J Int Neuropsychol Soc 14, 266–278.

53.

Wechsler

, Stone

(1987) Wechsler Memory Scale revised, Psychological Corporation, San Antonio.

54.

Wechsler

(1981) WAIS-R Manual:Wechsler Adult Intelligence Scale-Revised, Psychological Corporation, San Antonio.

55.

Folstein

, Folstein

, McHugh

(1975) Mini-mental state. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

56.

Wechsler

(1997) The WAIS III - WMS III Technical Manual, Psychological Corporation, San Antonio.

57.

Ritchie

, Ropacki

, Albala

, Harrison

, Kaye

, Kramer

, Randolph

, Ritchie

(2017) Recommended cognitive outcomes in preclinical Alzheimer’s disease: Consensus statement from the European Prevention of Alzheimer’s Dementia project. Alzheimers Dement 13, 186–195.

58.

Perrotin

, La Joie

, de La Sayette

, Barré

, Mézenge

, Mutlu

, Guilloteau

, Egret

, Eustache

, Chételat

(2017) Subjective cognitive decline in cognitively normal elders from the community or from a memory clinic: Differential affective and imaging correlates. Alzheimers Dement 13, 550–560.

59.

La Joie

, Perrotin

, Egret

, Pasquier

, Tomadesso

, Mézenge

, Desgranges

, de La Sayette

, Chételat

(2016) Qualitative and quantitative assessment of self-reported cognitive difficulties in nondemented elders: Association with medical help seeking, cognitive deficits, and β-amyloid imaging. Alzheimers Dement (Amst) 5, 23–34.

60.

Weiskopf

, Helms

(2008) Multi-parameter mapping of the human brain at 1mm resolution in less than 20 minutes. Proc Int Soc Magn Reson Med 16, 2241.

61.

Tabelow

, Balteau

, Ashburner

, Callaghan

, Draganski

, Helms

, Kherif

, Leutritz

, Lutti

, Phillips

, Reimer

, Ruthotto

, Seif

, Weiskopf

, Ziegler

, Mohammadi

(2019) hMRI - A toolbox for quantitative MRI in neuroscience and clinical research. Neuroimage 194, 191–210.

62.

Ashburner

, Friston

(2005) Unified segmentation. Neuroimage 26, 839–851.

63.

Ashburner

(2007) A fast diffeomorphic image registration algorithm. Neuroimage 38, 95–113.

64.

Klunk

, Koeppe

, Price

, Benzinger

, Devous

, Jagust

, Johnson

, Mathis

, Minhas

, Pontecorvo

, Rowe

, Skovronsky

, Mintun

(2015) The Centiloid project: Standardizing quantitative amyloid plaque estimation by PET. Alzheimers Dement 11, 1–15.e4.

65.

Rolls

, Joliot

, Tzourio-Mazoyer

(2015) Implementation of a new parcellation of the orbitofrontal cortex in the automated anatomical labeling atlas. Neuroimage 122, 1–5.

66.

Navitsky

, Joshi

, Kennedy

, Klunk

, Rowe

, Wong

, Pontecorvo

, Mintun

, Devous

(2018) Standardization of amyloid quantitation with florbetapir standardized uptake value ratios to the Centiloid scale. Alzheimers Dement 14, 1565–1571.

67.

Battle

, Pillay

, Lowe

, Knopman

, Kemp

, Rowe

, Doré

, Villemagne

, Buckley

(2018) Centiloid scaling for quantification of brain amyloid with [18 F]flutemetamol using multiple processing methods. EJNMMI Res 8, 107.

68.

Narbutas

, Chylinski

, Egroo Van

, Bahri

, Koshmanova

, Besson

, Muto

, Schmidt

, Luxen

, Balteau

, Phillips

, Maquet

, Salmon

, Vandewalle

, Bastin

, Collette

, Snow

(2021) Positive effect of cognitive reserve on episodic memory, executive and attentional functions taking into account amyloid-beta, tau, and apolipoprotein E status.. Front Aging Neurosci 13, 666181.

69.

Insel

, Mormino

, Aisen

, Thompson

, Donohue

(2020) Neuroanatomical spread of amyloid β and tau in Alzheimer’s disease: Implications for primary prevention. Brain Commun 2, fcaa007.

70.

Auton

, Abecasis

, Altshuler

, Durbin

, Bentley

, Chakravarti

, Clark

, Donnelly

, Eichler

, Flicek

, et al. (2015) A global reference for human genetic variation. Nature 526, 68–74.

71.

Loh

, Danecek

, Palamara

, Fuchsberger

, Reshef

, Finucane

, Schoenherr

, Forer

, McCarthy

, Abecasis

, Durbin

, Price

(2016) Reference-based phasing using the Haplotype Reference Consortium panel. Nat Genet 48, 1443–1448.

72.

Sadegh

, Ragno

, AghaKouchak

(2017) Multivariate Copula Analysis Toolbox (MvCAT): Describing dependence and underlying uncertainty using a Bayesian framework. Water Resour Res 53, 5166–5183.

73.

Jaeger

, Edwards

, Das

, Sen

(2017) An ² statistic for fixed effects in the generalized linear mixed model. J Appl Stat 44, 1086–1105.

74.

Dauphinot

, Bouteloup

, Mangin

, Vellas

, Pasquier

, Blanc

, Hanon

, Gabelle

, Annweiler

, David

, Planche

, Godefroy

, Rivasseau-Jonveaux

, Chupin

, Fischer

, Chêne

, Dufouil

, Krolak-Salmon

(2020) Subjective cognitive and non-cognitive complaints and brain MRI biomarkers in the MEMENTO cohort. Alzheimers Dement (Amst) 12, e12051.

75.

de Jager

, Schrijnemaekers

A-CMC

, Honey

TEM

, Budge

(2009) Detection of MCI in the clinic: Evaluation of the sensitivity and specificity of a computerised test battery, the Hopkins Verbal Learning Test and the MMSE. Age Ageing 38, 455–460.

76.

Calabria

, Manenti

, Rosini

, Zanetti

, Miniussi

, Cotelli

(2011) Objective and subjective memory impairment in elderly adults: A revised version of the Everyday Memory Questionnaire. Aging Clin Exp Res 23, 67–73.

77.

Pelletier

, Bernard

, Dilharreguy

, Helmer

, Le Goff

, Chanraud

, Dartigues

, Allard

, Amieva

, Gwénaëlle

(2017) Patterns of brain atrophy associated with episodic memory and semantic fluency decline in aging. Aging (Albany NY) 9, 1470.

78.

Jeong

, Chung

, Kim

(2015) Episodic memory in aspects of large-scale brain networks. Front Hum Neurosci 9, 454.

79.

Hollands

, Lim

, Buckley

, Pietrzak

, Snyder

, Ames

, Ellis

, Harrington

, Lautenschlager

, Martins

, Masters

, Villemagne

, Rowe

, Maruff

(2015) Amyloid-β related memory decline is not associated with subjective or informant rated cognitive impairment in healthy adults. J Alzheimers Dis 43, 677–686.

80.

Snitz

, Lopez

, McDade

, Becker

, Cohen

, Price

, Mathis

, Klunk

(2015) Amyloid-β imaging in older adults presenting to a memory clinic with subjective cognitive decline: A pilot study. J Alzheimers Dis 48, S151–S159.

81.

Amariglio

, Becker

, Carmasin

(2012) Subjective cognitive complaints and amyloid burden in cognitively normal older individuals. Neuropsychologia 50, 2880–2886.

82.

Perrotin

(2012) Subjective cognition and amyloid deposition imaging. Arch Neurol 69, 223.

83.

Snitz

, Weissfeld

, Cohen

, Lopez

, Nebes

, Aizenstein

, McDade

, Price

, Mathis

, Klunk

(2015) Subjective cognitive complaints, personality and brain amyloid-beta in cognitively normal older adults. Am J Geriatr Psychiatry 23, 985–993.

84.

Gonzalez-Escamilla

, Lange

, Teipel

, Buchert

, Grothe

(2017) PETPVE12: An SPM toolbox for partial volume effects correction in brain PET - Application to amyloid imaging with AV45-PET. Neuroimage 147, 669–677.

85.

, Blazey

, Snyder

, Raichle

, Marcus

, Ances

, Bateman

, Cairns

, Aldea

, Cash

, Christensen

, Friedrichsen

, Hornbeck

, Farrar

, Owen

, Mayeux

, Brickman

, Klunk

, Price

, Thompson

, Ghetti

, Saykin

, Sperling

, Johnson

, Schofield

, Buckles

, Morris

, Benzinger

TLS

(2015) Partial volume correction in quantitative amyloid imaging. Neuroimage 107, 55–64.

86.

Spencer

, Wendell

, Giggey

, Katzel

, Lefkowitz

, Siegel

, Waldstein

(2013) Psychometric limitations of the mini-mental state examination among nondemented older adults: An evaluation of neurocognitive and magnetic resonance imaging correlates. Exp Aging Res 39, 382–397.

87.

Lim

, Snyder

, Pietrzak

, Ukiqi

, Villemagne

, Ames

, Salvado

, Bourgeat

, Martins

, Masters

, Rowe

, Maruff

(2016) Sensitivity of composite scores to amyloid burden in preclinical Alzheimer’s disease: Introducing the Z-scores of Attention, Verbal fluency, and Episodic memory for Nondemented older adults composite score. Alzheimers Dement (Amst) 2, 19–26.

88.

Chen

, Yang

, Du

, Hu

, Han

(2019) Clinical characteristics in subjective cognitive decline with and without worry: Baseline investigation of the SILCODE Study. J Alzheimers Dis 72, 443–454.

89.

Iob

, Kirschbaum

, Steptoe

(2020) Persistent depressive symptoms, HPA-axis hyperactivity, and inflammation: The role of cognitive-affective and somatic symptoms. Mol Psychiatry 25, 1130–1140.

90.

Hertzog

, Hülür

, Gerstorf

, Pearman

(2018) Is subjective memory change in old age based on accurate monitoring of age-related memory change? Evidence from two longitudinal studies. Psychol Aging 33, 273–287.

91.

van Oijen

, de Jong

, Hofman

, Koudstaal

, Breteler

MMB

(2007) Subjective memory complaints, education, and risk of Alzheimer’s disease. Alzheimers Dement 3, 92–97.

92.

Amieva

, Mokri

, Le Goff

, Meillon

, Jacqmin-Gadda

, Foubert-Samier

, Orgogozo

, Stern

, Dartigues

(2014) Compensatory mechanisms in higher-educated subjects with Alzheimer’s disease: A study of 20 years of cognitive decline. Brain 137, 1167–1175.

93.

Stern

, Arenaza-Urquijo

, Bartrés-Faz

, Belleville

, Cantilon

, Chetelat

, Ewers

, Franzmeier

, Kempermann

, Kremen

, Okonkwo

, Scarmeas

, Soldan

, Udeh-Momoh

, Valenzuela

, Vemuri

, Vuoksimaa

, Arenaza Urquiljo

, Bartrés-Faz

, Belleville

, Cantillon

, Chetelat

, Clouston

SAP

, Estanga

, Ewers

, Franzmeier

, Gold

, Habeck

, Jones

, Kempermann

, Kochhann

, Kremen

, Lim

, Martínez-Lage

, Morbelli

, Okonkwo

, Ossenkoppele

, Pettigrew

, Rosen

, Scarmeas

, Soldan

, Song

, Udeh-Momoh

, Stern

, Valenzuela

, Van Loenhoud

, Vemuri

, Vuoksimaa

(2020) Whitepaper: Defining and investigating cognitive reserve, brain reserve, and brain maintenance. Alzheimers Dement 16, 1305–1311.

94.

Aghjayan

, Buckley

, Vannini

, Rentz

, Jackson

, Sperling

, Johnson

, Amariglio

(2017) The influence of demographic factors on subjective cognitive concerns and beta-amyloid. Int Psychogeriatr 29, 645–652.

95.

Ali

, Smart

, Gawryluk

(2018) Subjective cognitive decline and APOE ɛ4: A systematic review. J Alzheimers Dis 65, 303–320.

96.

Moreno-Grau

, Rodríguez-Gómez

, Sanabria

, Pérez-Cordón

, Sánchez-Ruiz

, Abdelnour

, Valero

, Hernández

, Rosende-Roca

, Mauleón

, Vargas

, Lafuente

, Gil

, Santos-Santos

MÁ

, Alegret

, Espinosa

, Ortega

, Guitart

, Gailhajanet

, de Rojas

, Sotolongo-Grau

, Ruiz

, Aguilera

, Papasey

, Martín

, Peleja

, Lomeña

, Campos

, Vivas

, Gómez-Chiari

, Tejero

MÁ

, Giménez

, Serrano-Ríos

, Orellana

, Tárraga

, Ruiz

, Boada

, Abdelnour

, Aguilera

, Alegret

, Berthier

, Boada

, Buendia

, Bullich

, Campos

, Cañabate

, Cuevas

, de Rojas

, Espinosa

, Gailhajenet

, Diego

, Gil

, Giménez

, Gismondi

, Gómez-Chiari

, Guitart

, Hernández-Olasagarre

, Hernández

, Ibarria

, Lafuente

, Lomeña

, Martín

, Martínez

, Mauleón

, Monté

, Moreno

, Moreno-Grau

, Núñez

, Orellana

, Ortega

, Páez

, Pancho

, Pavía

, Pelejà

, Pérez-Cordon

, Pérez-Grijalba

, Pesini

, Preckler

, Rodríguez-Gómez

, Romero

, Rosende-Roca

, Ruiz

, Sanabria

, Sánchez-Ruiz

, Santos-Santos

, Sarasa

, Sotolongo-Grau

, Tárraga

, Tejero

, Torres

, Valero

, Vargas

, Vivas

(2018) Exploring APOE genotype effects on Alzheimer’s disease risk and amyloid β burden in individuals with subjective cognitive decline: The FundacioACE Healthy Brain Initiative (FACEHBI) study baseline results. Alzheimers Dement 14, 634–643.

97.

Müller-Gerards

, Weimar

, Abramowski

, Tebrügge

, Jokisch

, Dragano

, Erbel

, Jöckel

, Moebus

, Winkler

(2019) Subjective cognitive decline, APOE ɛ4, and incident mild cognitive impairment in men and women. Alzheimers Dement (Amst) 11, 221–230.

98.

Wang

, Van Belle

, Crane

, Kukull

, Bowen

, McCormick

, Larson

(2004) Subjective memory deterioration and future dementia in people aged 65 and older. J Am Geriatr Soc 52, 2045–2051.

99.

Cantero

, Iglesias

, Van Leemput

, Atienza

(2016) Regional hippocampal atrophy and higher levels of plasma amyloid-beta are associated with subjective memory complaints in nondemented elderly subjects. J Gerontol A Biol Sci Med Sci 71, 1210–1215.

100.

de Leeuw

, van der Flier

, Tijms

, Scheltens

, Mendes

, Manadas

, Bierau

, van Wijk

, van den Heuvel

EGHM

, Mohajeri

, Teunissen

, Kester

(2020) Specific nutritional biomarker profiles in mild cognitive impairment and subjective cognitive decline are associated with clinical progression: The NUDAD Project. J Am Med Dir Assoc 21, 1513.e1–1513.e17.

101.

Palomar-Bonet

, Atienza

, Cantero

(2020) Blood total antioxidant status is associated with cortical glucose uptake and factors related to accelerated aging. Brain Struct Funct 225, 841–851.

102.

Janelidze

, Berron

, Smith

, Strandberg

, Proctor

, Dage

, Stomrud

, Palmqvist

, Mattsson-Carlgren

, Hansson

(2021) Associations of plasma phospho-Tau217 levels with tau positron emission tomography in early Alzheimer disease. JAMA Neurol 78, 149–156.

103.

Welstead

, Luciano

, Muniz-Terrera

, Saunders

, Mullin

, Russ

(2021) Predictors of mild cognitive impairment stability, progression, or reversion in the Lothian Birth Cohort 1936. J Alzheimers Dis 80, 225–232.

104.

Hanseeuw

, Betensky

, Jacobs

HIL

, Schultz

, Sepulcre

, Becker

, Cosio

DMO

, Farrell

, Quiroz

, Mormino

, Buckley

, Papp

, Amariglio

, Dewachter

, Ivanoiu

, Huijbers

, Hedden

, Marshall

, Chhatwal

, Rentz

, Sperling

, Johnson

(2019) Association of amyloid and tau with cognition in preclinical Alzheimer disease: A longitudinal study. JAMA Neurol 76, 915–924.