Association of Depressive Symptoms and Cognition in Older Adults Without Dementia Across Different Biomarker Profiles

Abstract

Background:

Depression is a late-life risk factor for cognitive decline. Evidence suggests an association between Alzheimer’s disease (AD) associated pathologic changes and depressive symptoms.

Objective:

To investigate the influence of AT(N) biomarker profile (amyloid-β [A], p-tau [T], and neurodegeneration [N]) and gender on cross-sectional associations between subclinical depressive symptoms and cognitive function among older adults without dementia.

Methods:

Participants included 868 individuals without dementia from the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Depressive symptoms were measured using the Geriatric Depression Scale (GDS). ADNI neuropsychological composite scores assessed memory and executive function (EF). PET, cerebrospinal fluid, and MRI modalities classified the study sample into biomarker profiles: normal biomarkers (A–T–N–), AD continuum (A+T±N±), and suspect non-AD pathology (SNAP; A–T±N–or A–T–N±). Multivariate regression models were used to investigate associations between GDS and cognitive domains.

Results:

GDS was negatively associated with memory (β= –0.156, p < 0.001) and EF (β= –0.147, p < 0.001) in the whole sample. When classified by biomarker profile, GDS was negatively associated with memory and EF in AD continuum (memory: β= –0.174, p < 0.001; EF: β= –0.129 p = 0.003) and SNAP (memory: β= –0.172, p = 0.005; EF: β= –0.197, p = 0.001) subgroups. When stratified by sex, GDS was negatively associated with memory (β= –0.227, p < 0.001) and EF (β= –0.205, p < 0.001) in men only.

Conclusion:

The association between subclinical depressive symptoms and cognitive function is highly influenced by the AT(N) biomarker profile.

Keywords

Alzheimer’s disease amyloid biomarker cognition depressive symptoms neurodegeneration tau

INTRODUCTION

Depressive symptoms are prevalent among older adults [1]. Prior research indicates that subclinical and clinical depressive symptoms are associated with cognitive decline, the onset of mild cognitive impairment (MCI), and the development of Alzheimer’s disease (AD) and related dementias [2 –6]. However, the relationship between depressive symptoms and cognition in the presence of AD pathology remains unclear [7].

Historically, AD has been diagnosed primarily through clinical diagnostic criteria [8]. However recent progress in in vivo assessment of neuroimaging and biofluids provide positive evidence to support neuropathological criteria for AD pathology [9 –12]. AD occurs through an accumulation of neuropathologic changes, and therefore the 2018 NIA-AA guideline suggests defining AD and its diagnosis through neuropathologic indicators, using in vivo biomarkers, neuroimaging, and autopsy, as opposed to clinical diagnosis alone [13].

Previous work has shown that depressive symptoms in cognitively normal older adults may be associated with the presence of amyloid-β (Aβ) (A), tau (T), and neurodegeneration (N), the hallmarks of AD pathology often referred to as the AT(N) system [13 –16]. Based on the AT(N) system, populations can be categorized into three main subtypes: normal biomarkers, AD continuum, and suspected non-AD pathology (SNAP). Reports linking depression, AT(N) classification, and cognition are limited, with some studies referring to biomarker analysis in the study of AD, without looking at AT(N) biomarkers specifically [15, 16], while other studies have assessed the validity of AT(N) biomarkers in predicting clinical outcomes, without also assessing the role depression may play in cognitive decline [17, 18]. Therefore, the relationship between depressive symptoms and cognitive function in preclinical AD pathologic changes has not been clearly established. Additionally, given the well-documented sex differences in depressive symptom severity in older adults [19, 20] and also known sex differences in performance on tests of different cognitive domains [21], we investigated whether sex differences impact the relationship between depressive symptoms and cognition in memory and executive function (EF) domains.

For this purpose, we used data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), a longitudinal cohort of older adults with a focus on the collection of biomarker data across the AD spectrum to address the following questions: 1) In participants without dementia, are depressive symptoms associated with cognitive performance (memory and EF)? 2) In participants without dementia, does the association between depressive symptoms and cognition differ by AT(N) status (Normal AT(N) biomarkers, AD continuum, SNAP)? 3) In participants without dementia, does the association between depression and cognition differ by clinical diagnostic category (cognitively normal and amnestic MCI)? 4) In participants without dementia, is the relationship between depression and cognitive function influenced by gender? Does the impact of gender differ among stratification by biomarker and clinical diagnosis? 5) In participants without dementia with abnormal levels of Aβ, does the amount of Aβ present moderate the effect of depressive symptoms on cognition? We hypothesized that in individuals without dementia, depressive symptoms would be associated with cognitive performance in memory and EF domains. Furthermore, we hypothesized that these associations would be present in individuals with pathologic evidence of disease (i.e., individuals with in vivo evidence of AT(N) pathology) but not in those with normal biomarkers (i.e., individuals with no AT(N) pathology in brain).

2 METHODS

Study design and participants

Data used for this study was derived from the ADNI-1, ADNI-GO, and ADNI-2 studies. ADNI is an ongoing cohort of older adults, which was launched in 2003 as a public–private partnership with the primary goal of testing whether serial magnetic resonance imaging (MRI), positron emission tomography (PET), other biological markers, and clinical and neuropsychological assessments can be combined to measure the progression of MCI and mild dementia. More information about the ADNI study can be found at https://adni.loni.usc.edu. ADNI data collection was approved by the Institutional Review Boards of all participating institutions. Informed written consent was obtained from all participants at each site.

Eligible participants for this cross-sectional study included ADNI participants who were cognitively normal (CN) or were diagnosed with amnestic MCI (aMCI) at baseline visit. Patients diagnosed with AD at baseline were excluded. Eligible patients also had complete biomarker data for cerebrospinal fluid (CSF) Aβ₄₂ and p-tau, and structural MRI. All eligible participants had also completed a Geriatric Depression Scale (GDS) test in the same visit (Fig. 1).

Fig. 1

Flow chart of study participants.

Fig. 2

Scatter plots showing the relationship between GDS and Cognitive Domains across different biomarker subgroups. A–C) Association between depressive symptoms (GDS) and memory (z-score) by biomarker: A) Normal biomarkers; B) AD continuum; C) SNAP; D–F) Association between depressive symptoms (GDS) and EF (z-score) by biomarker: D) Normal biomarkers; E) AD continuum; F) SNAP.

CN participants had Mini-Mental State Examination (MMSE) scores of 24 or higher and a Clinical Dementia Rating (CDR) score of 0. All MCI participants were diagnosed as having aMCI; diagnostic criteria included a MMSE score between 24 and 30 (inclusive), a memory complaint, objective memory loss measured by education-adjusted scores on the Wechsler Memory Scale Logical Memory II, a CDR of 0.5, absence of significant impairment in other cognitive domains, essentially preserved activities of daily living, and absence of dementia. For more information on ADNI protocols see https://adni.loni.usc.edu/methods/documents/.

Study measures

Neuropsychological assessment

Primary cognitive outcomes for this study were ADNI memory and EF composite scores. Methods for the development of these composite scores have been outlined in detail in previous articles [22, 23]. Composite scores are analyzed in Z-score units.

The 15-item GDS was used to evaluate the presence of depressive symptoms [24]. GDS scores range from 0 to 15; however, individuals who met criteria for clinical depression (GDS >6) at the screening visit were excluded from ADNI study. Therefore, participants included in this study all had GDS scores of ≤6 at the time of enrollment.

CSF biomarkers

CSF Aβ₁₄₂, tau, and p-tau were measured at the ADNI Biomarker Core Laboratory at the University of Pennsylvania using the multiplex xMAP Luminex platform (Luminex Corp, Austin, TX) with Innogenetics (INNO-BIA AlzBio3; Ghent, Belgium; for research use only reagents) immunoassay kit-based reagents [9]. All CSF biomarker assays were performed in duplicate and averaged.

Structural MRI

All participants completed structural MRIs. MRI data were automatically processed using the FreeSurfer software package (available at https://surfer.nmr.mgh.harvard.edu/) by the Schuff and Tosun laboratory at the University of California-San Francisco as part of the ADNI shared dataset. FreeSurfer methods for identifying and calculation of regional brain volume are previously described in detail [25].

AT(N) profiling

A β: Aβ abnormal (A+) and normal (A–) groups were determined by applying a cutoff value of 192 pg/ml on the CSF Aβ₁₄₂ [9].

p-tau: p-tau abnormal (T+) or normal (T–) was determined by a cutoff value of 23 pg/ml for CSF p-tau level [9].

Neurodegeneration: Neurodegeneration present (N+) or absent (N–) was defined using hippocampal volume as a proxy for neurodegeneration. Abnormal N was defined as hippocampal volume adjusted for total intracranial volume (HVa) of less than 6,723 mm³ [9, 26].

These biomarkers were used to classify participants into three AT(N) biomarker categories: normal biomarkers (A–T–N–), AD continuum (A+T±N±), and suspected non-AD pathology (SNAP; A–T±N+ or A–T+N±).

Statistical analysis

All statistical analyses were performed using SPSS version 25. First, multivariate regression models were used to investigate the cross-sectional associations between GDS scores and the cognitive domains of memory and EF in the entire sample, while controlling for age, sex, and education. Subsequently, we stratified the sample based on clinical diagnosis (CN versus aMCI) and biomarker (normal biomarkers, AD continuum, and SNAP) and performed regression analysis of the stratified samples. Additionally, each regression analysis was repeated after further stratifying the sample by sex. Models were also run to explore moderation effects, if any, of Aβ on memory and EF by measuring the association between GDS and cognitive domains among participants without dementia with pathologic levels of Aβ, where Aβ was also a continuous covariate. To account for type 1 error due to running multiple comparisons, significance level was established as p-value <0.01.

RESULTS

Demographics and sample characteristics

Table 1 summarizes the sample characteristics. Participants had a mean age of 72.7 (SD = 6.85) years, 92.7% were white, and 46.3% were women. Average composite memory scores were 0.548 (SD = 0.755). Men had a mean memory composite score of 0.375 (SD = 0.706) and women had a mean composite memory score of 0.748 (SD = 0.763). Mean EF scores were 0.465 (SD = 0.894). Men had a mean composite EF score of 0.399 (SD = 0.897) and women had a mean EF score of 0.541 (SD = 0.886). Mean GDS score for the entire sample was 1.38 (SD = 1.39). Men had mean GDS score of 1.35 (SD = 1.37) and women had a mean GDS score of 1.42 (SD = 1.42).

Table 1

Baseline characteristics of 868 participants by diagnosis and AT(N) biomarker profiles.

Variable	Total Sample	CN	aMCI	Normal Biomarker	AD Continuum	SNAP	Male	Female
N (%)	868 (100%)	338 (38.9%)	530 (61.1%)	161 (18.5%)	482 (55.5%)	225 (25.9%)	466 (53.7%)	402 (46.3%)
Age, y, mean (SD)	72.7 (6.85)	73.76 (5.78)	72.09 (7.39)	70.59 (6.25)	73.50 (6.68)	72.64 (7.31)	73.50 (6.74)	71.85 (6.89)
Female, %	46.3%	52.7%	42.3%	53.4%	44.8%	44.4%	–	–
Education, y, mean (SD)	16.13 (2.79)	16.38 (2.61)	15.98 (2.78)	16.25 (2.64)	15.99 (2.76)	16.36 (2.68)	16.63 (2.66)	15.56 (2.67)
White, %	92.7%	90.5%	94.2%	89.4%	94.6%	91.1%	91.8%	93.6%
GDS, mean (SD)	1.38 (1.39)	0.85 (1.12)	1.72 (1.45)	1.37 (1.52)	1.43 (1.36)	1.28 (1.36)	1.35 (1.37)	1.42 (1.42)
Memory composite score, mean (SD)	0.548 (0.755)	1.051 (0.570)	0.227 (0.681)	0.930 (0.609)	0.310 (0.726)	0.784 (0.727)	0.375 (0.706)	0.748 (0.763)
Executive function composite score, mean (SD)	0.465 (0.894)	0.798 (0.815)	0.252 (0.878)	0.821 (0.779)	0.244 (0.876)	0.683 (0.877)	0.399 (0.897)	0.541 (0.886)

Normal biomarkers = A–T–N–; AD pathologic change = A+ T±N±; and SNAP = A–T±N+ or A–T+N±. GDS, Geriatric Depression Scale; CN, cognitively normal; aMCI, amnestic mild cognitive impairment; SNAP, suspect none-AD pathology.

Association between GDS and cognitive domains in the entire sample

Multivariate regression analysis in the entire sample showed that higher GDS was associated with lower memory and EF performance (memory: β= –0.156, p < 0.001, CI: –0.118, –0.051; EF: β= –0.147, p < 0.001, CI: –0.135, –0.055). When the sample was stratified by sex, GDS was negatively associated with memory and EF in men (memory: β= –0.227, p < 0.001, CI: –0.161, –0.073; EF: β= –0.205, p < 0.001, CI: –0.189, –0.079). However, no significant associations between GDS and cognition were found among women (memory: β= –0.079, p = 0.116, CI: –0.095, 0.011; EF: β= –0.078, p = 0.107, CI: –0.108, 0.011) (Table 2).

Table 2

Multivariate regression analysis: Association between GDS and cognitive domains memory and EF in the entire sample and by sex

	Memory				Executive Function
	β	t	p	95% CI	β	t	p	95% CI
All^* (N = 868)	–0.156	–4.925	<0.001	–0.118, –0.051	–0.147	–4.642	<0.001	–0.135, –0.055
Women^** (N = 402)	–0.079	–1.574	0.116	–0.095, 0.011	–0.078	–1.616	0.107	–0.108, 0.011
Men^** (N = 466)	–0.227	–5.266	<0.001	–0.161, –0.073	–0.205	–4.784	<0.001	–0.189, –0.079

^*Models adjusted for age, sex, education. ^**Models adjusted for age and education. Standardized β coefficient is used throughout.

Association between GDS and cognitive domains among clinical subgroups

Next, the sample was divided by clinical diagnoses into CN and aMCI subgroups and multivariate regression analysis was repeated for the subgroups. There were no significant associations between GDS and memory or EF in either the CN or aMCI subgroups: [CN (Memory: β= 0.004, p = 0.934, CI: –0.047, 0.051; EF: β= 0.006, p = 0.900, CI: –0.068, 0.077); aMCI (memory: β= –0.007, p = 0.871, CI: –0.041, 0.035; EF: β= –0.084, p = 0.038, CI: –0.100, –0.003). This remained true when the sample was further subdivided by sex: Women [CN (memory: β= 0.073, p = 0.303, CI: –0.027, 0.085; EF: β= 0.091, p = 0.200, CI: –0.032, 0.151); aMCI (memory: β= 0.071, p = 0.286, CI: –0.030, 0.102; EF: β= –0.070, p = 0.266, CI: –0.120, 0.033). Men [CN (memory: β= –0.059, p = 0.423, CI: –0.127, 0.054; EF: β= –0.097, p = 0.165, CI: –0.203, 0.035); aMCI (memory: β= –0.074, p = 0.177, CI: –0.077, 0.014; EF: β= –0.106, p = 0.052, CI: –0.127, 0.001) (Table 3).

Table 3

Multivariate regression analysis: Associations between GDS and cognitive domains by diagnosis and sex

	Diagnosis	Memory				Executive Function
		β	t	p	95% CI	β	t	p	95% CI
All^*	CN (N = 338)	0.004	0.083	0.934	–0.047, 0.051	0.006	0.126	0.900	–0.068, 0.077
	aMCI (N = 530)	–0.007	–0.162	0.871	–0.041, 0.035	–0.084	–2.077	0.038	–0.100, –0.003
Women^**	CN (N = 178)	0.073	1.034	0.303	–0.027, 0.085	0.091	1.285	0.200	–0.032, 0.151
	aMCI (N = 224)	0.071	1.070	0.286	–0.030, 0.102	–0.070	–1.116	0.266	–0.120, 0.033
Men^**	CN (N = 160)	–0.059	–0.803	0.423	–0.127, 0.054	–0.097	–1.395	0.165	–0.203, 0.035
	aMCI (N = 306)	–0.074	–1.355	0.177	–0.077, 0.014	–0.106	–1.951	0.052	–0.127, 0.001

^*Models adjusted for age, sex, education. ^**Models adjusted for age and education. GDS, Geriatric Depression Scale; CN, cognitively normal; aMCI, amnestic mild cognitive impairment. Standardized β coefficient is used throughout.

Association between GDS and cognitive domains among biomarker subgroups

When the sample was stratified by AT(N) biomarker subgroups, regression models showed that higher GDS was negatively associated with memory and EF performance in both AD continuum (memory: β= –0.174, p < 0.001, CI: –0.138, –0.048; EF: β= –0.129, p = 0.003, CI: –0.138,–0.028) and SNAP (memory: β= –0.172, p = 0.005; CI:–0.156, –0.028; EF: β= –0.197, p = 0.001, CI: –0.204, –0.051) subgroups. However, there was no significant association between GDS and cognitive domains in the normal biomarker subgroup (memory: β= –0.045, p = 0.556, CI: –0.079, 0.043; EF: β= –0.079, p = 0.312, CI: –0.119, 0.038). Additionally, when biomarker subgroups were further stratified by sex, men showed significantly worsening memory as GDS increased in AD continuum and SNAP subgroups (AD continuum: β= –0.223, p < 0.001, CI: –0.163, –0.052; SNAP: β= –0.218, p = 0.010, CI: –0.190, –0.026). There was no significant association between GDS and memory in women (Normal: β= 0.085, p = 0.444, CI: –0.046, 0.104; AD continuum: β= –0.120, p = 0.079, CI: –0.142, 0.008; SNAP: β= –0.133, p = 0.180, CI: –0.178, 0.034). Additionally, in men, lower EF was associated with higher GDS scores in the AD continuum subgroup (AD continuum: β= –0.203, p<0.001, CI: –0.205, –0.058), while amongst women, worse performance on EF tests was associated with higher GDS scores in the SNAP subgroups (β= –0.259, p = 0.005, CI: –0.280, –0.051) (Table 4).

Table 4

Multivariate regression analysis: Association between GDS and cognitive domains among different AT(N) categories overall and by sex

	Biomarker	Memory				Executive Function
		β	t	p	95% CI	β	t	p	95% CI
All^*	Normal biomarkers (N = 161)	–0.045	–0.589	0.556	–0.079, 0.043	–0.079	–1.014	0.312	–0.119, 0.038
	AD continuum (N = 482)	–0.174	–4.050	<0.001	–0.138, –0.048	–0.129	–2.959	0.003	–0.138, –0.028
	SNAP (N = 225)	–0.172	–2.828	0.005	–0.156, –0.028	–0.197	–3.270	0.001	–0.204, –0.051
Women^**	Normal biomarker (N = 86)	0.085	0.769	0.444	–0.046, 0.104	–0.003	–0.028	0.978	–0.105, 0.103
	AD continuum (N = 216)	–0.120	–1.768	0.079	–0.142, 0.008	–0.030	–0.453	0.651	–0.104, 0.065
	SNAP (N = 100)	–0.133	–1.352	0.180	–0.178, 0.034	–0.259	–2.878	0.005	–0.280, –0.051
Men^**	Normal biomarker (N = 75)	–0.235	–2.041	0.045	–0.208, –0.002	–0.278	–2.441	0.017	–0.261, –0.026
	AD continuum (N = 266)	–0.223	–3.826	<0.001	–0.163, –0.052	–0.203	–3.531	<0.001	–0.205, –0.058
	SNAP (N = 125)	–0.218	–2.602	0.010	–0.190, –0.026	–0.149	–1.814	0.072	–0.203, 0.009

^*Models are adjusted for age, sex, and education. ^**Models adjusted for age and education. GDS, Geriatric Depression Scale; SNAP, suspect non-AD pathology. Standardized β coefficient is used throughout.

Effects of Aβ at suprathreshold levels on the relationship between GDS and cognitive domains

Finally, we investigated whether Aβ levels in individuals with abnormal Aβ values (A+) affect the association between GDS and cognitive domains (memory and EF). This was done to assess if higher levels of amyloid moderate the effect of GDS on cognitive domains in people with abnormal amounts of amyloid present. Aβ levels were added to the regression model as a continuous measure, only among individuals who had abnormal Aβ levels (A+). Results showed that in this A+ subsample, the association between GDS and memory (β= –0.156, p < 0.001, CI: –0.127, –0.040) or EF (β= –0.121, p = 0.005, CI: –0.133, –0.023) remained significant and was only modestly affected by the addition of Aβ as a covariate (Table 5). This suggests that after adjusting for Aβ among A+ individuals, the associations between GDS and cognitive domains remain significant.

Table 5

Multivariate regression analysis: Effect of Aβ on GDS and memory among A+ participants

	Memory				Executive Function
N = 482	β	t	p	95% CI	β	t	p	95% CI
Age	–0.028	–0.664	0.507	–0.127, –0.040	–0.257	–5.763	<0.001	–0.045, –0.022
Sex	–0.260	–6.074	<0.001	–0.503, –0.257	–0.056	–1.254	0.210	–0.254, 0.056
Education, y	0.188	4.461	<0.001	0.028, 0.071	0.114	2.581	0.010	0.009, 0.064
GDS	–0.156	–3.770	<0.001	–0.127, –0.040	–0.121	–2.80	0.005	–0.133, –0.023
Aβ	0.253	6.084	<0.001	0.005, 0.010	0.105	2.426	0.016	0.001, 0.007

Models adjusted for age, sex, and education. GDS, Geriatric Depression Scale; Aβ, amyloid β. Standardized β coefficient is used throughout.

DISCUSSION

Our results indicate that, at cross-section, the presence of subclinical depressive symptoms was inversely associated with cognition in memory and EF domains. This association was significant among individuals with AD continuum and SNAP biomarkers, but not in individuals with normal AT(N) biomarkers. Additionally, significant associations between depressive symptoms and cognitive function were detected in men.

The association between depressive symptoms and cognition in our study is in line with previous studies suggesting a relationship between depressive symptoms and cognitive decline [3 , 27]. A 2021 study by Xu et al. found that in individuals without dementia, minimal depressive symptoms were associated with overall decreased cognition as well as greater amyloid burden. Xu et al. further argued that amyloid deposition plays a mediating role in the relationship between depressive symptoms and cognition. As in our study, Xu et al. investigated the association of depressive symptoms in older adults without dementia [28]. A meta-analysis examining cognitive impairment and severity of depression has reported significant correlations between depression severity and cognitive performance in the domains of episodic memory, executive function, and processing speed [29]. Other studies indicate that depressive symptoms, even at subclinical levels, are associated with longitudinal cognitive decline and incident AD or incident dementia [3]. However, these findings have not always been consistent, and some studies failed to show relationships between depressed mood and cognition at baseline or longitudinally [30, 31]. Study findings may differ due to variation in the methods for the evaluation of depression and threshold scores in the diagnosis of clinical depression. These differences may lead to underestimation of the effect of mild or moderate depressive symptoms in people who do not meet criteria for diagnosis of clinical depression [3]. Our results indicate that depressive symptoms are negatively associated with performance in both memory and EF domains in older adults without dementia. When we stratified our sample by CN and aMCI groups, these associations were no longer significant. This is likely because stratification by clinical diagnosis leads to both a smaller sample size and narrower range of cognitive scores that result in decreased power to detect weak associations.

There are several hypotheses that link cognitive impairment, AD pathology, and depressive symptoms. The glucocorticoid cascade hypothesis suggests that depressive symptoms may elevate endogenous glucocorticoids, increasing AD pathology, which in turn can lead to cognitive decline [32]. Prior studies have shown that glucocorticoids increase Aβ and tau pathology in the brain and may consequently damage brain regions, especially the hippocampus, leading to more accelerated decline in memory or other related cognitive functions over time [14 , 34]. In addition, inflammatory factors may have significant involvement in the development of cognitive decline [35]. Studies have shown elevated inflammatory biomarkers in both depression and AD [36] and have also suggested an interaction between depression and Aβ through inflammatory-mediated pathways [37, 38].

In recent years, studies have reported conflicting results regarding associations between in vivo AD pathology and depressive symptoms. One study reported no difference in cortical Aβ between matched participants with high and low depressive symptoms in a sample of MCI and AD participants from ADNI [39]. Another ADNI study reported that A+ MCI subjects with depressive symptoms have an elevated Aβ load, especially in the frontotemporal and insular cortices, as compared to cognitively matched non-depressed individuals [40]. Another study using the ADNI MCI sample showed that MCI individuals with subclinical depressive symptoms had diminished CSF Aβ₁₄₂ levels but did not differ from MCI individuals without depressive symptoms in CSF tau levels [41]. Xu et al., in their previously mentioned study, has shown that depressive symptoms are associated with cognitive impairment, using outcomes of memory and EF. This study also examined the associations of minimal depressive symptoms and CSF biomarkers, and suggests that amyloid burden mediates the relationship between depressive symptoms and cognition [28]. In contrast to this, a 2021 study by Mackin et al. found that late life depression was associated with decreased amyloid burden. Furthermore, Mackin et al. showed that the relationship between late life depression and cognitive decline are not related to amyloid deposition [42]. While our study does analyze whether there is a moderation effect of amyloid burden on the association between GDS and cognitive domains, we do not address mediation. Additionally, the main goal of our study is to analyze the role of biomarker stratification on the associations between GDS and memory and EF. Our data suggest that amyloid burden may play a role in moderating the relationship between depressive symptoms and cognitive function. However, differences among study findings indicate that more research is needed to understand the role of amyloid in the relationship between depression and cognitive function.

Our results indicate that depressive symptoms are associated with both memory and EF domains in individuals with AD continuum pathology (A+T±N±) and SNAP (A–T±N+ or A–T+N±), while there was no association between depressive symptoms and cognition among individuals with normal biomarkers (A–T–N–). Furthermore, our results indicate that among A+ individuals, higher Aβ levels modestly affected the association between GDS and cognition. There are two distinctions between our study and previous studies. First, to our knowledge, this study is the first to evaluate the relationship between depressive symptoms and cognition using the AT(N) biomarker system classification and, therefore, our biomarker-defined groups are slightly different from other studies. Second, our classification is only based on biomarkers such that all participants without dementia are included in our analyses, resulting in a wider range of cognitive scores for participants. This increases our ability to detect associations.

Biological sex differences in depression and cognitive function are frequently reported in older adults [2 , 44]. Prevalence of depression, anxiety, stress, and other mood disorders is higher in women than men [45]. These differences can lead to varying impacts of depression on cognition in men and women, respectively. Here, we showed that in older adults without dementia, higher levels of depressive symptoms are associated with lower cognition in men. However, this association was not significant in women. A similar outcome was observed when the sample was stratified by biomarker. The neurobiological mechanisms underlying these differences among men and women is poorly understood [46]. Bioavailability of estrogen is thought to affect brain pathology in late life, which can lead to depression or cognitive dysfunction [45 –47]. Evidence suggests that testosterone levels modulate serotonergic transmission in brain which in turn affects both cognition and mood [47]. As testosterone gradually declines, associated changes in serotonin transmission may be one of the pathophysiologic mechanisms through which cognition declines and depressive symptoms increase in men.

Our findings should be viewed in light of certain limitations. Due to the cross-sectional nature of this study, we cannot make any conclusions about the causative relationship between subclinical depression and cognitive outcomes. While we used a large sample size of older adults without dementia, stratification based on sex, biomarkers, or clinical diagnosis decreases the power to detect associations in subsamples. Additionally, this study uses established biomarker thresholds and clinical categories to draw relationships between cognitive function and depressive symptoms. While this serves a functional purpose, it overlooks the transitory, progressive nature of AD pathological changes and may poorly categorize participants on the border of thresholds. Furthermore, ADNI is not a population-based study and there are strict inclusion and exclusion criteria for selection of subjects. Therefore, results of this study might not be generalizable to community-based samples.

In conclusion, this study contributes to the body of research suggesting a biological link between depressive symptoms and cognitive decline in older adults. Our findings indicate the importance of monitoring clinical and subclinical depression in older adults, especially in subgroups with biomarker evidence of AD pathologic change. Depression is a preventable and remediable risk factor for AD and as such has huge potential for being targeted with a variety of pharmacological and non-pharmacological interventions. This could, in turn, decrease the adverse effects of depression on cognitive function.

Footnotes

ACKNOWLEDGMENTS

Data collection and sharing for ADNI project was funded by the Alzheimer’s Disease Neuroimaging Initiative (ADNI) (National Institutes of Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National Institute of Biomedical Imaging and Bioengineering, and through generous contributions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company; CereSpir, Inc.; Cogstate; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.; Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier; Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of Health Research is providing funds to support ADNI clinical sites in Canada. Private sector contributions are facilitated by the Foundation for the National Institutes of Health (). The grantee organization is the Northern California Institute for Research and Education, and the study is coordinated by the Alzheimer’s Therapeutic Research Institute at the University of Southern California. ADNI data are disseminated by the Laboratory for Neuro Imaging at the University of Southern California.

This work was supported by grants from the National Institute of Health (Ezzati: NIA K23 AG063993; Lipton: NIA P01 AG03949), the Alzheimer’s Association (Ezzati, 2019-AACSF-641329), the Cure Alzheimer’s Fund (Ezzati, Lipton), the Leonard and Sylvia Marx Foundation (Lipton), and the Dr. Louis Glazer and Jacob N. Glazer Endowment Fund (Rubin-Norowitz).

Authors’ disclosures available online ().

DATA STATEMENT

Data in this article were previously presented as an abstract and poster in a scientific meeting at the American Academy of Neurology from April 17-22, 2021.

References

Beekman

ATF

, Geerlings

, Deeg

DJH

, Smit

, Schoevers

, de Beurs

, Braam

, Penninx

, van Tilburg

(2002) The natural history of late-life depression: A 6-year prospective study in the community. Arch Gen Psychiatry 59, 605–611.

Sundermann

, Katz

, Lipton

(2017) Sex differences in the relationship between depressive symptoms and risk of amnestic mild cognitive impairment. Am J Geriatr Psychiatry 25, 13–22.

Ezzati

, Katz

, Derby

, Zimmerman

, Lipton

(2019) Depressive symptoms predict incident dementia in a community sample of older adults: Results from the Einstein Aging Study. J Geriatr Psychiatry Neurol. doi: 10.1177/0891988718824036.

Lipnicki

, Makkar

, Crawford

, Thalamuthu

, Kochan

, Lima-Costa

, Castro-Costa

, Ferri

, Brayne

, Stephan

, Llibre-Rodriguez

, Llibre-Guerra

, Valhuerdi-Cepero

, Lipton

, Katz

, Derby

, Ritchie

, Ancelin

, Carrière

, Scarmeas

, Yannakoulia

, Hadjigeorgiou

, Lam

, Chan

, Fung

, Guaita

, Vaccaro

, Davin

, Kim

, Han

, Suh

, Riedel-Heller

, Roehr

, Pabst

, van Boxtel

, Köhler

, Deckers

, Ganguli

, Jacobsen

, Hughes

, Anstey

, Cherbuin

, Haan

, Aiello

, Dang

, Kumagai

, Chen

, Narazaki

, Ng

, Gao

, Nyunt

MSZ

, Scazufca

, Brodaty

, Numbers

, Trollor

, Meguro

, Yamaguchi

, Ishii

, Lobo

, Lopez-Anton

, Santabárbara

, Leung

, Lo

, Popovic

, Sachdev

(2019) Determinants of cognitive performance and decline in 20 diverse ethno-regional groups: A COSMIC collaboration cohort study. PLoS Med 16, e1002853.

, Shu

, Wu

, Chen

, Hu

, Zhang

, Yan

, Feng

(2020) Neuropsychiatric symptoms as prognostic makers for the elderly with mild cognitive impairment: A meta-analysis. J Affect Disord 271, 185–192.

Santabarbara

, Sevil-Perez

, Olaya

, Gracia-Garcia

, Lopez-Anton

(2019) Clinically relevant late-life depression as risk factor of dementia: A systematic review and meta-analysis of prospective cohort studies. Rev Neurol 68, 493–502.

Morris

, Clark

, Vissel

(2014) Inconsistencies and controversies surrounding the amyloid hypothesis of Alzheimer’s disease. Acta Neuropathol Commun 2, 135.

Jack

Jr , Albert

, Knopman

, McKhann

, Sperling

, Carrillo

, Thies

, Phelps

(2011) Introduction to the recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 257–262.

Shaw

, Vanderstichele

, Knapik-Czajka

, Clark

, Aisen

, Petersen

, Blennow

, Soares

, Simon

, Lewczuk

, Dean

, Siemers

, Potter

, Lee

, Trojanowski

(2009) Cerebrospinal fluid biomarker signature in Alzheimer’s Disease Neuroimaging Initiative subjects. Ann Neurol 65, 403–413.

10.

Vemuri

, Wiste

, Weigand

, Shaw

, Trojanowski

, Weiner

, Knopman

, Petersen

, Jack

Jr; . Alzheimer’s Disease Neuroimaging Initiative (2009) MRI and CSF biomarkers in normal, MCI, and AD subjects. Neurology 73, 287–293.

11.

Villemagne

, Burnham

, Bourgeat

, Brown

, Ellis

, Salvado

, Szoeke

, Macaulay

, Martins

, Maruff

, Ames

, Rowe

, Masters

(2013) Amyloid β deposition, neurodegeneration, and cognitive decline in sporadic Alzheimer’s disease: A prospective cohort study. Lancet Neurol 12, 357–367.

12.

Wiltfang

, Lewczuk

, Riederer

, Grünblatt

, Hock

, Scheltens

, Hampel

, Vanderstichele

, Iqbal

, Galasko

, Lannfelt

, Otto

, Esselmann

, Henkel

, Kornhuber

, Blennow

(2005) Consensus paper of the WFSBP Task force on biological markers of dementia: The role of CSF and blood analysis in the early and differential diagnosis of dementia. World J Biol Psychiatry 6, 69–84.

13.

Jack

CR Jr

, Bennett

, Blennow

, Carrillo

, Dunn

, Haeberlein

, Holtzman

, Jagust

, Jessen

, Karlawish

, Liu

, Molinuevo

, Montine

, Phelps

, Rankin

, Rowe

, Scheltens

, Siemers

, Snyder

, Sperling

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

14.

Ezzati

, Zimmerman

, Katz

, Lipton

(2013) Hippocampal correlates of depression in healthy elderly adults. Hippocampus 23, 1137–1142.

15.

Gatchel

, Rabin

, Buckley

, Locascio

, Quiroz

, Yang

, Vannini

, Amariglio

, Rentz

, Properzi

, Donovan

, Blacker

, Johnson

, Sperling

, Marshall

, Harvard Aging Brain Study (2019) Longitudinal association of depression symptoms with cognition and cortical amyloid among community-dwelling older adults. JAMA Netw Open 2, e198964.

16.

Zhang

, Wei

, Shen

, Ma

, Chen

, Dong

, Tan

, Yu

; Alzheimer’s Disease Neuroimaging Initiative (2020) Associations of subsyndromal symptomatic depression with cognitive decline and brain atrophy in elderly individuals without dementia: A longitudinal study. J Affect Disord 274, 262–268.

17.

, Li

, Suckling

, Feng

, Pan

, Wang

, Song

, Zhu

, Li

, Wang

, Tan

, Dong

, Tan

, Mok

, Aisen

, Weiner

; Alzheimer’s Disease Neuroimaging Initiative (2019) Frequency and longitudinal clinical outcomes of Alzheimer’s AT(N) biomarker profiles: A longitudinal study. Alzheimers Dement 15, 1208–1217.

18.

Bertens

, Tijms

, Vermunt

, Prins

, Scheltens

, Visser

(2017) The effect of diagnostic criteria on outcome measures in preclinical and prodromal Alzheimer’s disease: Implications for trial design. Alzheimers Dement (N Y) 3, 513–523.

19.

Young

, Fogg

, Scheftner

, Keller

, Fawcett

(1990) Sex differences in the lifetime prevalence of depression: Does varying the diagnostic criteria reduce the female/male ratio? . J Affect Disord 18, 187–192.

20.

Parker

, Fletcher

, Paterson

, Anderson

, Hong

(2014) Gender differences in depression severity and symptoms across depressive sub-types. J Affect Disord 167, 351–357.

21.

Laws

, Irvine

, Gale

(2018) Sex differences in Alzheimer’s disease. Curr Opin Psychiatry 31, 133–139.

22.

Crane

, Carle

, Gibbons

, Insel

, Mackin

, Gross

, Jones

, Mukherjee

, Curtis

, Harvey

, Weiner

, Mungas

, Alzheimer’s Disease Neuroimaging Initiative (2012) Development and assessment of a composite score for memory in the Alzheimer’s Disease Neuroimaging Initiative (ADNI). Brain Imaging Behav 6, 502–516.

23.

Gibbons

, Carle

, Mackin

, Harvey

, Mukherjee

, Insel

, Curtis

, Mungas

, Crane

, Alzheimer’s Disease Neuroimaging Initiative (2012) A composite score for executive functioning, validated in Alzheimer’s Disease Neuroimaging Initiative (ADNI) participants with baseline mild cognitive impairment. Brain Imaging Behav 6, 517–527.

24.

Yesavage

, Sheikh

(1986) 9/Geriatric Depression Scale (GDS) recent evidence and development of a shorter violence. Clin Gerontol 5, 165–173.

25.

Fischl

(2012) FreeSurfer. Neuroimage 62, 774–781.

26.

Mormino

, Betensky

, Hedden

, Schultz

, Amariglio

, Rentz

, Johnson

, Sperling

(2014) Synergistic effect of β-amyloid and neurodegeneration on cognitive decline in clinically normal individuals. JAMA Neurol 71, 1379–1385.

27.

Wang

, Shen

, Ma

, Ou

, Dong

, Tan

, Yu

; Alzheimer’s Disease Neuroimaging Initiative (2021) Associations of the rate of change in geriatric depression scale with amyloid and cerebral glucose metabolism in cognitively normal older adults: A longitudinal study. J Affect Disord 280, 77–84.

28.

, Feng

, Shen

, Bi

, Ma

, Li

, Dong

, Tan

, Yu

(2021) Amyloid pathologies modulate the associations of minimal depressive symptoms with cognitive impairments in older adults without dementia. Biol Psychiatry 89, 766–775.

29.

McDermott

, Ebmeier

(2009) A meta-analysis of depression severity and cognitive function. J Affect Disord 119, 1–8.

30.

Becker

, Chang

, Lopez

, Dew

, Sweet

, Barnes

, Yaffe

, Young

, Kuller

, Reynolds

3rd (2009) Depressed mood is not a risk factor for incident dementia in a community-based cohort. Am J Geriatr Psychiatry 17, 653–663.

31.

Henderson

, Korten

, Jacomb

, Mackinnon

, Jorm

, Christensen

, Rodgers

(1997) The course of depression in the elderly: A longitudinal community-based study in Australia. Psychol Med 27, 119–129.

32.

Butters

, Young

, Lopez

, Aizenstein

, Mulsant

, Reynolds

, 3rd DeKosky

, Becker

(2008) Pathways linking late-life depression to persistent cognitive impairment and dementia. Dialogues Clin Neurosci 10, 345–357.

33.

Green

, Billings

, Roozendaal

, McGaugh

, LaFerla

(2006) Glucocorticoids increase Amyloid-β and tau pathology in a mouse model of Alzheimer’s disease. J Neurosci 26, 9047.

34.

Sierksma

, van den Hove

, Steinbusch

, Prickaerts

(2010) Major depression, cognitive dysfunction and Alzheimer’s disease: Is there a link? Eur J Pharmacol 626, 72–82.

35.

Jordanova

, Stewart

, Davies

, Sherwood

, Prince

(2007) Markers of inflammation and cognitive decline in an African-Caribbean population. Int J Geriatr Psychiatry 22, 966–973.

36.

Maes

, Yirmyia

, Noraberg

, Brene

, Hibbeln

, Perini

, Kubera

, Bob

, Lerer

, Maj

(2009) The inflammatory & neurodegenerative (I&ND) hypothesis of depression: Leads for future research and new drug developments in depression. Metab Brain Dis 24, 27–53.

37.

Kang

, Cirrito

, Dong

, Csernansky

, Holtzman

(2007) Acute stress increases interstitial fluid amyloid-beta via corticotropin-releasing factor and neuronal activity. Proc Natl Acad Sci U S A 104, 10673–10678.

38.

Guo

, Zheng

, Justice

(2012) Central CRF system perturbation in an Alzheimer’s disease knockin mouse model. Neurobiol Aging 33, 2678–2691.

39.

Chung

, Plitman

, Nakajima

, Chakravarty

, Caravaggio

, Gerretsen

, Iwata

, Graff-Guerrero

(2016) Cortical Amyloid β deposition and current depressive symptoms in Alzheimer disease and mild cognitive impairment. J Geriatr Psychiatry Neurol 29, 149–159.

40.

Brendel

, Pogarell

, Xiong

, Delker

, Bartenstein

, Rominger

; Alzheimer’s Disease Neuroimaging Initative (2015) Depressive symptoms accelerate cognitive decline in amyloid-positive MCI patients. Eur J Nucl Med Mol Imaging 42, 716–724.

41.

Gonzales

, Insel

, Nelson

, Tosun

, Schöll

, Mattsson

, Sacuiu

, Bickford

, Weiner

, Mackin

; Alzheimer’s Disease Neuroimaging Initiative (2018) Chronic depressive symptomatology and CSF amyloid beta and tau levels in mild cognitive impairment. Int J Geriatr Psychiatry 33, 1305–1311.

42.

Mackin

, Insel

, Landau

, Bickford

, Morin

, Rhodes

, Tosun

, Rosen

, Butters

, Aisen

, Raman

, Saykin

, Toga

, Jack

Jr , Koeppe

, Weiner

, Nelson

; Alzheimer’s Disease Neuroimaging Initiative and the ADNI Depression Project (2021) Late-life depression is associated with reduced cortical amyloid burden: Findings from the Alzheimer’s Disease Neuroimaging Initiative Depression Project. Biol Psychiatry 89, 757–765.

43.

Fuhrer

, Dufouil

, Dartigues

, Study

FTP

(2003) Exploring sex differences in the relationship between depressive symptoms and dementia incidence: Prospective results from the PAQUID study. J Am Geriatr Soc 51, 1055–1063.

44.

Koran

MEI

, Wagener

, Hohman

; Alzheimer’s Neuroimaging Initiative (2017) Sex differences in the association between AD biomarkers and cognitive decline. Brain Imaging Behav 11, 205–213.

45.

Streiner

, Cairney

, Veldhuizen

(2006) The epidemiology of psychological problems in the elderly. Can J Psychiatry 51, 185–191.

46.

Janicki

, Schupf

(2010) Hormonal influences on cognition and risk for Alzheimer’s disease. Curr Neurol Neurosci Rep 10, 359–366.

47.

Gonzalez

, Farabollini

, Albonetti

, Wilson

(1994) Interactions between 5-hydroxytryptamine (5-HT) and testosterone in the control of sexual and nonsexual behaviour in male and female rats. Pharmacol Biochem Behav 47, 591–601.