Factors Associated with Apathy in Alzheimer’s Disease: A Cross-Sectional Analysis of the Texas Alzheimer’s Research and Care Consortium (TARCC) Study

Abstract

Background:

Apathy is among the most frequent neuropsychiatric syndromes in Alzheimer’s disease (AD).

Objective:

To determine the prevalence of apathy and the associated clinical and laboratorial parameters (focus on inflammatory biomarkers) in patients with dementia enrolled at the Texas Alzheimer’s Research and Care Consortium (TARCC) study.

Methods:

This is a cross-sectional analysis of TARCC baseline. Participants were evaluated through different clinical tools, including the Mini-Mental State Examination (MMSE) and the Lawton-Brody Instrumental Activities of Daily Life (IADL)/Physical Self-Maintenance Scale (PSMS). Apathy was defined by a positive response to the respective item in the Neuropsychiatric Inventory–Questionnaire applied to caregivers. Serum levels of 16 biomarkers were determined by HumanMap multiplex immunoassay. Comparisons between apathy versus non-apathy groups were carried out with non-parametric tests. Logistic regression and the least absolute shrinkage and selection operator (LASSO) were used to separately model apathy as a function of each biomarker, adjusted for the potential confounders.

Results:

From 1,319 patients with AD (M/F: 579/740, mean age ± SD: 75.3 ± 8.4), 373 (28.3%) exhibited apathy. When categorized according to the presence of apathy, the groups had significant differences in sex, diabetes diagnosis, and tobacco use. The apathy group also had worse cognitive performance and daily functioning than the non-apathy group as assessed, respectively, by MMSE and IADL/PSMS. Higher levels of interleukin-6, interleukin-10, and leptin were associated with higher odds of apathy.

Conclusion:

Apathy is associated with cognitive and functional status in AD. The association between apathy and peripheral inflammatory mediators deserves further investigation.

Keywords

Alzheimer’s disease apathy inflammation interleukin-6 interleukin-10

INTRODUCTION

Besides cognitive impairment, nearly all patients with Alzheimer’s disease (AD) present neuropsychiatric symptoms, also called behavioral and psychological symptoms of dementia (BPSD) [1 –3]. BPSD have been associated with negative outcomes in AD, including decreased patient and caregiver quality of life, increased risk of institutionalization, and higher healthcare costs [1 –3]. The term “BPSD” is an umbrella expression that encompasses different types of clinical problems, such as agitation, apathy, delusion, among others [1 –3].

Apathy can be defined as loss or reduction of self-initiated or environment-stimulated goal-directed behaviors alongside emotional flattening [4]. Apathy is among the most common BPSD in AD, with a 5-year prevalence over 70% in this population [5, 6]. Compared to other BPSD, such as affective symptoms (e.g., anxiety, dysphoria) that tend to fluctuate during the course of AD, the frequency and severity of apathy correlate with the progression of cognitive decline in AD [7, 8]. A growing body of evidence has also suggested that apathy is associated with the development and/or progression of neurodegenerative diseases [9, 10]. However, the matter of BPSD as predictors and/or prodromal symptoms of neurodegenerative diseases is still open to debate, and the studies diverge on which BPSD (if any) can be implicated [11, 12].

Apathy has been associated with greater functional impairment, frailty, greater caregiver burden, increased risk of institutionalization, and even higher mortality in patients with AD [13, 14]. In this context of high prevalence and clinical meaning, apathy is an important therapeutic target. Nevertheless, there are no well-established pharmacological and non-pharmacological strategies for the treatment of apathy in AD [15 –17]. Therefore, there is a great need to develop effective and translatable treatments for apathy in AD. For that, a better understanding of the pathophysiology of apathy, including potential biomarkers, is necessary [18].

Several studies have investigated neuroimaging markers of apathy in AD, especially on proxies of neurodegeneration [17, 19]. For instance, apathy was associated with neurofibrillary tangle burden in specific prefrontal cortex (PFC) areas [20], and microstructural white matter abnormalities in AD [21]. Moreover, dysfunctional brain networks involving the PFC has been implicated in AD-related apathy [22, 23]. However, the literature on blood-based biomarkers of AD-related apathy is scarce. Previous studies have focused on biomarkers of mechanisms implicated in the development of BPSD in general, mainly biomarkers of vascular risk and microvascular pathology [24, 25].

The objectives of the current study were twofold: 1) to determine the prevalence of apathy and 2) to investigate clinical and laboratorial (i.e., serum biomarkers) parameters associated with apathy in patients with AD at the baseline of the Texas Alzheimer’s Research and Care Consortium (TARCC). Given the need of identifying biomarkers associated with apathy and previous evidence showing increased peripheral levels of inflammatory/immune markers in older adults with apathy [26, 27], we specifically investigated whether apathy in AD is associated with a pro-inflammatory profile. As demographics factors (e.g., age), medical comorbidities (e.g., chronic heart disease; hypertension), and cognitive status can influence the levels of inflammatory biomarkers [24 , 29], they were carefully controlled in the analyses.

METHODS

Study design

TARCC is a longitudinal multi-site study of a large cohort of older adults, comprising non-cognitively impaired subjects, patients with mild cognitive impairment and AD recruited from outpatient clinics and the community [24 , 29]. Participants were recruited from nine Texas academic medical institutions. Each participant underwent a standardized annual examination at the respective site, which included medical evaluation, neuropsychological testing, and blood draw. Individuals with a history of major psychiatric disorders (e.g., mood and psychotic disorders) were not included in the cohort. The current study focused on cross-sectional analyses of baseline measures of individuals with AD. Institutional Review Board approval was obtained at each TARCC site, and written informed consent was obtained from all participants and/or caregivers.

Clinical assessment

The diagnosis of AD was carried out through a consensus panel and based on the criteria for probable AD according to the NINCDS-ADRDA [30].

TARC research protocol included clinical measures of global cognition (Mini-Mental State Examination, MMSE) [31], functioning (Lawton-Brody Activities of Daily Living: Physical Self-Maintenance Scale, PSMS; Instrumental Activities of Daily Living scale, IADL) [32], staging dementia severity (Clinical Dementia Rating Scale Sum of Boxes, CDR-SOB) [33], and depression (Geriatric Depression Scale, GDS) [34]. As part of the assessment, the Neuropsychiatry Inventory (NPI-Q) was also applied to family members or caregivers. The NPI-Q is a brief informant-based assessment of 12 BPSD that has been shown to be valid and reliable [35, 36]. For the purpose of this study, a positive response to the item ‘Apathy’, regardless its severity, was regarded as presence of apathy.

Clinical information obtained also included report or direct assessment of medical comorbidities (e.g., diabetes mellitus, hypertension, thyroid disease) and smoking and alcohol exposure.

Biomarker measurement

TARCC used the multiplexed immunoassay Multi-Analyte Profile (humanMAP) (https://www.rulesbasedmedicine.com, Austin, TX, USA) to measure the serum levels of different biomarkers of participants, as described elsewhere [25, 29].

Based on the literature and our prior studies in AD [25 , 37–40], we selected a panel of 16 serum-based biomarkers of microvascular pathology (vascular cell adhesion molecule 1, VCAM-1; intracellular adhesion molecule 1, ICAM-1; vascular endothelium growth factor A, VEGF-A), inflammation (interleukin-1 receptor antagonist, IL-1ra; interleukin-2, IL-2; interleukin-6; IL-6; interleukin-10, IL-10; interleukin-12 subunit p40, IL-12 p40; tumor necrosis factor alpha, TNF-α; interferon gamma, IFN-γ; interleukin-8, IL-8/CCL8; inducible protein-10, IP-10/CXCL10; C reactive protein, CRP; and serum amyloid A, SAA), and hormones (insulin and leptin) implicated in the pathophysiology of AD.

Data analysis

Descriptive statistics (i.e., central tendency, dispersion, frequency) were performed to characterize all study variables: sociodemographic (age, sex, race, years of education), clinical information (medical comorbidities, smoking and alcohol exposure) and measures (MMSE, PSMS, IADL, GDS, NPI), and biomarkers (above mentioned panel of serum molecules). Preliminary analyses investigated measures of central tendency and frequency for all study variables. Biomarker data were preprocessed in four steps: 1) values below the lower limit of detection were set to missing; 2) values were log-transformed; 3) missing values were filled in via quantile regression imputation for left-censored data (QRILC) [41]; and finally 4) values were standardized (z-scored) to place all log-biomarkers on a common metric (M = 0; SD = 1).

Non-parametric tests (Mann-Whitney U; Spearman’s rho) were used to identify potentially confounding sample characteristics via guidelines from the statistical literature: any characteristic demonstrating a relationship with both apathy and a given biomarker met criteria [42] to be considered a potential confounder. Sample characteristics evaluated as potential confounders included sex, race, ethnicity, current living situation (e.g., lives alone, lives with spouse), marital status, primary language, diabetes diagnosis (present versus absent), and tobacco use in the past 30 days (present versus absent), and other medical comorbidities (e.g., hypertension, hypercholesterolemia). Of these, three variables demonstrated a relationship with both apathy and at least one given biomarker under investigation: sex, diabetes diagnosis (present versus absent), and tobacco use in the past 30 days (present versus absent). These variables were included as covariates in parametric models of the biomarkers for which the potential for confounding was demonstrated (see Table 1 notes for specific details).

Table 1

Sociodemographic and clinical variables of patients with Alzheimer’s disease (AD) with and without apathy

Variable	AD +	AD –
	Apathy	Apathy
	(n = 373)	(n = 946)
	M (SD)	M (SD)	p
Age	74.85 (7.9)	75.52 (8.6)	0.134
Years of Education	14.17 (3.8)	14.04 (3.6)	0.288
MMSE	22.20 (4.1)	21.28 (5.1)	0.022
GDS	6.83 (5.6)	4.25 (4.8)	< 0.001
PSMS	7.91 (2.7)	3.99 (3.9)	< 0.001
IADLs	16.78 (6.1)	7.77 (8.1)	< 0.001
CDR-SOB	5.75 (2.8)	5.70 (3.4)	0.075
	N (%)	N (%)	p
Sex	0.024
Male	182 (31.4)	397 (68.6)
Female	191 (25.8)	549 (74.2)
Race	0.411
American Indian/Native Alaskan/Islander	0 (0.0)	4 (100.0)
Asian American	7 (38.9)	11 (61.1)
Black	14 (25.0)	42 (75.0)
White	352 (28.4)	889 (71.6)
Independence	0.024
Able to Live Independently	46 (20.2)	182 (79.8)
Requires Assistance with Complex Activities	237 (29.5)	567 (70.5)
Requires Assistance with Basic Activities	84 (31.1)	186 (68.9)
Completely Dependent	6 (35.3)	11 (64.7)
Congestive Heart Failure	0.223
Absent/Inactive	370 (28.5)	929 (71.5)
Recent/Active	3 (15.8)	16 (84.2)
Hypertension	0.817
Absent/Inactive	154 (28.6)	384 (71.4)
Recent/Active	219 (28.0)	562 (72.0)
Hypercholesterolemia	0.152
Absent/Inactive	136 (26.1)	385 (73.9)
Recent/Active	237 (29.7)	560 (70.3)
Diabetes	0.010
Absent/Inactive	306 (27.0)	827 (73.0)
Recent/Active	67 (36.2)	118 (63.8)
Thyroid disease	0.621
Absent/Inactive	293 (28.6)	733 (71.4)
Recent/Active	78 (30.1)	181 (69.9)
Tobacco Use	0.019
No	349 (27.8)	905 (72.2)
Yes	24 (42.1)	33 (57.9)
Alcohol Abuse	0.818
No	370 (29.8)	871 (70.2)
Yes	3 (33.3)	6 (66.7)

MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale; PSMS, Lawton-Brody Physical Self-Maintenance Scale; IADL, Lawton-Brody Instrumental Activities of Daily Living; CDR-SOB, Clinical Dementia Rating Scale Sum of Boxes.

Analyses investigating relationships between each of 16 biomarkers and dichotomous apathy (absent versus present) proceeded in three stages. First, for each biomarker, the non-parametric Mann-Whitey U test was used to broadly characterize differences between apathy groups. Second, logistic regression was used to separately model apathy as a function of each biomarker, adjusted for the potential confounders where necessary [43] (see above) for each apathy/biomarker relationship. Given the cross-sectional nature of the present data, these models were enabled to characterize either apathy or the biomarker as the outcome for each model. Even after data preprocessing, the biomarker data rarely conformed to readily identified statistical distributions (e.g., Gaussian, lognormal). Therefore, the apathy variable was operationalized as the outcome variable; however, it should be noted that this is not meant to imply a directional hypothesis between the variables. For both the Mann-Whitney U tests and the logistic regression models, unadjusted and adjusted p-values are provided. Adjusted p-values were derived to address potential Type I error via false discovery rate (FDR) [43]. These adjusted values provide a conservative interpretation of the present findings. However, given the exploratory nature of the present work, the unadjusted values provide a complementary (albeit less conservative) interpretation that may reduce the potential for Type II error.

Finally, to evaluate the relationship between each biomarker and apathy in the presence of the other biomarkers, the outcome was fit as a function of the entire set of biomarkers (i.e., multiple logistic regression), adjusted for sex, diabetes (present versus absent), and tobacco use in the past 30 days (present versus absent). This model was developed using regularization via the least absolute shrinkage and selection operator (LASSO) [44] with post-selection inference [45]. LASSO was used because a non-regularized model (i.e., non-LASSO) using the same data demonstrated high variance inflation factor values due to strong correlations between biomarkers (i.e., multicollinearity). The LASSO procedure was used to reduce the high variance in parameter estimation that results from this multicollinearity by shrinking model coefficients towards zero. Further, with respect to variable selection, LASSO may reduce parameter estimates all the way to zero, effectively removing those variables. Finally, recent developments in post-selection inference have derived a method for traditional statistical inference (i.e., p-values) when using LASSO regression. This method has resolved the biasing effects of the LASSO that, to date, have not been resolved in alternative variable selection routines (e.g., stepwise selection).

RESULTS

From 1,319 participants with AD (M/F: 579/740, mean age ± SD 75.3 ± 8.4) evaluated at the baseline of TARCC study, 373 (28.3%) exhibited apathy. Table 1 shows the comparison between patients with and without apathy regarding sociodemographic and clinical variables. Groups were significantly different with respect to sex, levels of independence, diabetes diagnosis (present versus absent), and tobacco use in the past 30 days (present versus absent); conversely, the groups were not different with respect to age, education, and other medical comorbidities. The apathy group also had worse cognitive performance (as assessed by the MMSE) and functional status (according to PSMS, IADL) than the non-apathy group.

Raw (i.e., pre-standardization) measures of central tendency (median; interquartile range) are provided for the log-transformed set of biomarkers in Table 2.

Table 2

Biomarker descriptive statistics (log-transformed)

Marker	Overall	Apathy Absent	Apathy Present
	N = 1,319	N = 946	N = 373
	Median [IQR]	Median [IQR]	Median [IQR]
CRP	13.32 [9.77, 14.97]	13.02 [9.68, 14.86]	13.70 [10.13, 15.22]
IFN-γ	2.18 [1.22, 3.08]	2.05 [1.19, 3.05]	2.44 [1.36, 3.10]
TNF-α	2.03 [1.36, 2.51]	1.87 [1.34, 2.50]	2.17 [1.43, 2.57]
IL-2	–0.79 [–1.72, 0.17]	–0.82 [–1.74, 0.17]	–0.73 [–1.67, 0.16]
IL-6	0.71 [–0.20, 1.55]	0.49 [–0.28, 1.47]	1.01 [–0.02, 1.74]
IL-8	3.66 [2.61, 4.28]	3.54 [2.57, 4.25]	3.87 [2.74, 4.34]
IL-10	–0.09 [–0.95, 0.57]	–0.25 [–0.98, 0.51]	0.13 [–0.84, 0.69]
IL-12	5.87 [4.06, 6.59]	5.77 [4.03, 6.57]	6.00 [4.14, 6.63]
MIP-1β	5.75 [4.20, 6.30]	5.63 [4.15, 6.28]	5.85 [4.28, 6.32]
IP-10	6.66 [4.94, 7.37]	6.57 [4.91, 7.34]	6.83 [4.98, 7.45]
Leptin	7.80 [6.41, 9.38]	7.58 [6.29, 9.36]	8.20 [6.76, 9.43]
Insulin	5.35 [4.40, 6.51]	5.21 [4.38, 6.47]	5.64 [4.43, 6.63]
VEGF	5.28 [3.67, 6.27]	5.16 [3.61, 6.24]	5.46 [3.80, 6.32]
SAA	14.94 [11.31, 16.19]	14.83 [11.26, 16.22]	15.06 [11.70, 16.13]
sICAM-1	13.01 [10.51, 13.40]	12.97 [10.43, 13.38]	13.07 [10.70, 13.46]
sVCAM-1	13.58 [10.90, 13.97]	13.54 [10.80, 13.95]	13.65 [11.03, 14.00]

CRP, C-reactive protein; IFN-γ, Interferon gamma; IL, interleukin; MIP-1β, macrophage inflammatory protein-1β; IP-10, interferon gamma-induced protein 10; SAA, serum amyloid A; sICAM-1, soluble intercellular adhesion molecule-1; sVCAM-1, soluble vascular cell adhesion molecule-1; TNF-α, tumor necrosis factor; VEGF, vascular endothelial growth factor.

Table 3

Model Results

Biomarker	Mann-Whitney U Test Non-Parametric Models			Logistic Regression Single-Biomarker Adjusted Models			LASSO Multiple-Biomarker Model^**
	Z	p	FDR-Adjusted p	Odds Ratio	p	FDR-Adjusted p	p
CRP	2.285	0.022	0.071	1.106	0.107	0.245	0.738
IFN-γ ^*	1.617	0.106	0.141	1.080	0.211	0.328	0.066
TNF-α	2.007	0.045	0.102	1.107	0.099	0.245	0.287
IL-2	0.445	0.656	0.668	1.010	0.874	0.874	0.141
IL-6	3.505	< 0.001	0.007	1.224	0.001	0.008	0.027
IL-8	2.375	0.018	0.070	1.131	0.049	0.196	0.315
IL-10	3.307	0.001	0.008	1.228	0.001	0.008	0.016
IL-12_P40^*‡	1.660	0.097	0.141	1.119	0.073	0.234	0.771
MIP-1B‡	1.443	0.149	0.170	1.075	0.246	0.328	0.290
IP-10	1.674	0.094	0.141	1.069	0.283	0.348	0.953
Leptin^*†	2.704	0.007	0.037	1.190	0.006	0.032	0.031
Insulin^†	1.818	0.069	0.138	1.091	0.157	0.314	0.234
VEGF_ANGIO_PLATE	1.654	0.098	0.141	1.086	0.183	0.325	0.467
SAA^*	0.428	0.668	0.668	1.027	0.672	0.717	0.060
sICAM-1^*	2.099	0.036	0.096	1.079	0.234	0.328	0.563
sVCAM-1^‡	1.491	0.136	0.167	1.061	0.350	0.400	0.723

CRP, C-reactive protein; IFN-γ, nterferon gamma; IL, interleukin; MIP-1β, macrophage inflammatory protein-1β; IP-10, interferon gamma-induced protein 10 kDa; SAA, serum amyloid A; sICAM-1, soluble intercellular adhesion molecule-1; sVCAM-1, soluble vascular cell adhesion molecule-1; TNF-α, tumor necrosis factor-alpha; VEGF, vascular endothelial growth factor. ^*Logistic regression model adjusted for sex. ^† Logistic regression model adjusted for diabetes (present versus absent). ^‡ Logistic regression model adjusted for tobacco use (present versus absent). ^**LASSO model results adjusted for sex, diabetes (present versus absent), and tobacco use (present versus absent).

Table 3 provides a summary of the results from each of the inferential analyses, including Mann-Whitney U tests (Z-statistic; unadjusted and adjusted p-values), logistic regression models (odds ratios (OR); unadjusted and adjusted p-values), and the p-values provided from the LASSO with post-selection inference. Conservatively, three biomarkers were significantly related to higher odds of apathy (i.e., apathy-present versus apathy-absent) after Type-I error correction across all models: IL-6 (OR = 1.22; adjusted p = 0.008), IL-10 (OR = 1.23; adjusted p = 0.008), and leptin (OR = 1.19; adjusted p = 0.031). For these analyses, the OR provides a meaningful effect size: a one-standard deviation increase in any of these log-biomarkers was related to 19–23% higher odds of apathy-present AD.

Preliminary evidence also supported IL-8/CCL8 in both the Mann-Whitney (Z = 2.38; unadjusted p = 0.018) and logistic regression models (OR = 1.13; unadjusted p = 0.049); however, this biomarker was not significant after Type I error correction in either case. Three biomarkers found limited support by the non-parametric Mann-Whitney test, but not the logistic regression model, before Type I error rate correction: CRP (Z = 2.29; unadjusted p = 0.022), TNF-α (Z = 2.01; unadjusted p = 0.045), and sICAM-1 (Z = 2.10; unadjusted p = 0.036). The set of factors that were not supported across any of the models included IFN-γ, IL-2, IL-12 p40, MIP-1β/CCL4, IP-10/CXCL10, VEGF-A, sVCAM-1, SAA, and insulin.

DISCUSSION

The analysis of the TARCC baseline showed that apathy is a frequent condition in patients with AD. The finding of ∼30% of apathy in participants with AD is consistent with other cross-sectional studies [5, 6]. Furthermore, subjects with apathy had worse cognition and functional status, corroborating the literature and the emerging view that apathy is associated with the progression of AD [9, 10]. The original contribution of this study refers to the finding of association between a set of inflammation-related serum biomarkers and apathy. More specifically, higher odds of apathy were found for patients who demonstrated higher serum levels of IL-6, IL-10, and leptin.

The pathophysiology of apathy in AD is highly complex, involving multiple mechanisms. It has been viewed as the result of ‘positive’—deposition of amyloid-β plaques and tau-containing neurofibrillary tangles, activated glia—and negative phenomena, i.e., loss of neurons and synapses, leading to circuit dysfunction [46]. Inflammatory mechanisms have been implicated in the pathophysiology of AD, contributing not only to neurodegenerative changes, but also to synaptic and circuit dysfunction [47 –49]. Our group and others have reported increased peripheral (i.e., serum or plasma) levels of inflammatory molecules, such as Il-1β and IL-6, in patients with AD compared to cognitively intact controls [37 , 51].

Apathy has also been associated with high CRP levels, a classical marker of inflammatory response, in community-dwelling older adults [26, 27]. In a previous study, our group found that peripheral levels of soluble TNF-α receptors 1 (sTNFR1) and 2 (sTNFR2), markers of inflammatory activity, correlated with the severity of apathy in patients with AD [52]. However, this study assessed only a very limited set of inflammatory molecules in a small group of patients (N = 27). The main contribution of the current study is to extend these previous findings to a larger cohort with greater generalizability. Our results clearly show that AD patients with apathy had higher levels of inflammatory/immune molecules, and that IL-6, IL-10, and leptin were associated with apathy. Of note, preliminary analyses also linked apathy in AD to higher levels of CRP, TNF-α, IL-8/CCL8, and sICAM-1, but these associations did not remain after Type I error correction or multivariate analysis.

IL-6 is a pleiotropic cytokine with multiple effects on inflammatory response and immunity, such as the induction of acute phase proteins as CRP. IL-6 has been implicated in the pathophysiology of AD. For instance, Lyra et al. reported that circulating levels of IL-6 correlated inversely with cognitive performance and hippocampal volumes in AD patients [51]. IL-10 also displays pleiotropic effects on inflammatory response and immunity, mainly antagonizing the effects and/or inhibiting the production of pro-inflammatory cytokines, such as TNF-α, IL-1β, IL-6, and IFN-γ, and chemokines like IL-8/CCL8 and IP-10/CXCL10. Accordingly, IL-10 is conceptualized as an ‘anti-inflammatory cytokine’ [53]. As IL-6, circulating levels of IL-10 are usually increased in patients with AD compared to controls, suggesting that this increase might be a compensatory phenomenon to the chronic low-grade peripheral inflammation characteristic of AD [54]. Leptin is a hormone predominantly produced by adipose cells involved in the regulation of energy metabolism and appetite [55]. Several studies have implicated leptin in the pathogenesis of chronic low-grade inflammation which makes obese individuals more susceptible to increased risk of developing cardiovascular diseases, type II diabetes, and neurodegenerative diseases, such as AD [56].

While the current study has several strengths, notably sample size and relatively broad panel of biomarkers reflecting different pathophysiological pathways, some limitations must be acknowledged. First, the diagnosis of AD was based on clinical grounds and lacked valid neuroimaging and cerebrospinal fluid biomarkers of the disease. Therefore, it is not possible to rule out completely that patients have other types of dementia, including mixed or vascular dementia. In our study, apathetic patients were more likely to have diabetes than non-apathetic ones and, possibly, more microvascular changes in the brain [57]. This was not specifically investigated in the TARCC cohort, but differences in the levels of biomarkers between apathy versus non-apathy groups remained statistically significant when controlling for the presence of medical comorbidities, including diabetes. Second, the definition of presence versus absence of apathy was based on a single item of the NPI-Q. Apathy is a complex construct, and a comprehensive assessment with dedicated tools, such as the Apathy Evaluation Scale [58], the Dimensional Apathy Scale [59], among others, could have provided a more reliable definition and, more importantly, a granular view of the association between different dimensions of apathy (affective versus cognitive versus behavioral) and inflammation biomarkers [60]. In this regard, it is worth mentioning that specific reward-related deficits or dimensions seem to be more associated with neuroendocrine stress response and inflammation than others [60, 61]. Biomarkers were collected at a single time point, and thus we cannot make any assessments regarding change over time. Additionally, because this analysis relied on cross-sectional data, there is no way to infer the direction of the relationship between the biomarkers and apathy. Finally, the magnitude of the effect sizes (OR ≈1.20) described by the present study were modest. Inferences drawn from the present study may be better considered hypothesis-generating, wherein the signal detected here provides a direction for future research to build on these results by evaluating higher-order interactions between these and other biomarkers.

To our knowledge, no previous study has specifically investigated a set of blood-based molecules reflecting different pathophysiological processes as biomarkers of apathy. Our final model taking into account potential confounding factors showed that higher levels of inflammation-related molecules (IL-6, IL-10, and leptin) were associated with apathy. While the clinical meaning of this finding remains to be determined, given the association between apathy and development and/or progression of AD reported in previous studies [9, 10], the investigation of inflammation-related molecules in the context of apathy seems promising and deservers further studies. It is tempting to speculate that this finding might open new venues for therapeutic intervention against apathy based on anti-inflammatory strategies, as has been proposed for mood disorders and other neurodegenerative diseases [62 –64].

Footnotes

ACKNOWLEDGMENTS

The Neuropsychiatry Program is supported by UT Health Houston Department of Psychiatry & Behavioral Sciences, and the Texas Alzheimer’s Research and Care Consortium (TARCC).

Authors’ disclosures available online ().

References

Canevelli

, Adali

, Voisin

, Soto

, Bruno

, Cesari

, Vellas

(2013) Behavioral and psychological subsyndromes in Alzheimer’s disease using the Neuropsychiatric Inventory. Int J Geriatr Psychiatry 28, 795–803.

Cummings

, Ritter

, Rothenberg

(2019) Advances in management of neuropsychiatric syndromes in neurodegenerative diseases. Curr Psychiatry Rep 21, 79.

Deardorff

, Grossberg

(2019) Behavioral and psychological symptoms in Alzheimer’s dementia and vascular dementia. Handb Clin Neurol 165, 5–32.

Robert

, Lanctot

, Aguera-Ortiz

, Aalten

, Bremond

, Defrancesco

, Hanon

, David

, Dubois

, Dujardin

, Husain

, Konig

, Levy

, Mantua

, Meulien

, Miller

, Moebius

, Rasmussen

, Robert

, Ruthirakuhan

, Stella

, Yesavage

, Zeghari

, Manera

(2018) Is it time to revise the diagnostic criteria for apathy in brain disorders? The 2018 international consensus group. Eur Psychiatry 54, 71–76.

Steinberg

, Shao

, Zandi

, Lyketsos

, Welsh-Bohmer

, Norton

, Breitner

, Steffens

, Tschanz

, Cache County

(2008) Point and 5-year period prevalence of neuropsychiatric symptoms in dementia: The Cache County Study. Int J Geriatr Psychiatry 23, 170–177.

Zhao

, Tan

, Wang

, Jiang

, Tan

, Xu

, Li

, Wang

, Lai

, Yu

(2016) The prevalence of neuropsychiatric symptoms in Alzheimer’s disease: Systematic review and meta-analysis. J Affect Disord 190, 264–271.

van der Linde

, Matthews

, Dening

, Brayne

(2017) Patterns and persistence of behavioural and psychological symptoms in those with cognitive impairment: The importance of apathy. Int J Geriatr Psychiatry 32, 306–315.

Connors

, Seeher

, Crawford

, Ames

, Woodward

, Brodaty

(2018) The stability of neuropsychiatric subsyndromes in Alzheimer’s disease. Alzheimers Dement 14, 880–888.

Bock

, Bahorik

, Brenowitz

, Yaffe

(2020) Apathy and risk of probable incident dementia among community-dwelling older adults. Neurology 95, e3280–e3287.

10.

Roberto

, Portella

, Marquie

, Alegret

, Hernandez

, Mauleon

, Rosende-Roca

, Abdelnour

, de Antonio

, Gil

, Tartari

, Vargas

, Espinosa

, Ortega

, Perez-Cordon

, Sanabria

, Orellana

, de Rojas

, Moreno-Grau

, Montrreal

, Alarcon-Martin

, Ruiz

, Tarraga

, Boada

, Valero

(2021) Neuropsychiatric profiles and conversion to dementia in mild cognitive impairment, a latent class analysis. Sci Rep 11, 6448.

11.

Hudon

, Escudier

, De Roy

, Croteau

, Cross

, Dang-Vu

, Zomahoun

HTV

, Grenier

, Gagnon

, Parent

, Bruneau

, Belleville

, Consortium for the Early Identification of Alzheimer’s Disease – Quebec (2020) Behavioral and psychological symptoms that predict cognitive decline or impairment in cognitively normal middle-aged or older adults: A meta-analysis. Neuropsychol Rev 30, 558–579.

12.

Cummings

(2021) The role of neuropsychiatric symptoms in research diagnostic criteria for neurodegenerative diseases. Am J Geriatr Psychiatry 29, 375–383.

13.

Teixeira

, Jr , Caramelli

(2006) [Apathy in Alzheimer’s disease]. Braz J Psychiatry 28, 238–241.

14.

Nijsten

JMH

, Leontjevas

, Pat-El

, Smalbrugge

, Koopmans

, Gerritsen

(2017) Apathy: Risk factor for mortality in nursing home patients. J Am Geriatr Soc 65, 2182–2189.

15.

Nobis

, Husain

(2018) Apathy in Alzheimer’s disease. Curr Opin Behav Sci 22, 7–13.

16.

Manera

, Abrahams

, Aguera-Ortiz

, Bremond

, David

, Fairchild

, Gros

, Hanon

, Husain

, Konig

, Lockwood

, Pino

, Radakovic

, Robert

, Slachevsky

, Stella

, Tribouillard

, Trimarchi

, Verhey

, Yesavage

, Zeghari

, Robert

(2020) Recommendations for the nonpharmacological treatment of apathy in brain disorders. Am J Geriatr Psychiatry 28, 410–420.

17.

Mortby

, Adler

, Aguera-Ortiz

, Bateman

, Brodaty

, Cantillon

, Geda

, Ismail

, Lanctot

, Marshall

, Padala

, Politis

, Rosenberg

, Siarkos

, Sultzer

, Theleritis

, Istaart Nps

PIA

(2021) Apathy as a treatment target in Alzheimer’s disease: Implications for clinical trials. Am J Geriatr Psychiatry, 10.1016/j.jagp.2021.06.016.

18.

Husain

, Roiser

(2018) Neuroscience of apathy and anhedonia: A transdiagnostic approach. Nat Rev Neurosci 19, 470–484.

19.

Theleritis

, Politis

, Siarkos

, Lyketsos

(2014) A review of neuroimaging findings of apathy in Alzheimer’s disease. Int Psychogeriatr 26, 195–207.

20.

Marshall

, Fairbanks

, Tekin

, Vinters

, Cummings

(2006) Neuropathologic correlates of apathy in Alzheimer’s disease. Dement Geriatr Cogn Disord 21, 144–147.

21.

Jonsson

, Edman

, Lind

, Rolstad

, Sjogren

, Wallin

(2010) Apathy is a prominent neuropsychiatric feature of radiological white-matter changes in patients with dementia. Int J Geriatr Psychiatry 25, 588–595.

22.

Munro

, Donovan

, Guercio

, Wigman

, Schultz

, Amariglio

, Rentz

, Johnson

, Sperling

, Marshall

(2015) Neuropsychiatric symptoms and functional connectivity in mild cognitive impairment. J Alzheimers Dis 46, 727–735.

23.

Jones

, De Marco

, Manca

, Bell

, Blackburn

, Wilkinson

, Soininen

, Venneri

(2019) Altered frontal and insular functional connectivity as pivotal mechanisms for apathy in Alzheimer’s disease. Cortex 119, 100–110.

24.

Hall

, Wiechmann

, Johnson

, Edwards

, Barber

, Cunningham

, Singh

, O’Bryant

(2014) Total cholesterol and neuropsychiatric symptoms in Alzheimer’s disease: The impact of total cholesterol level and gender. Dement Geriatr Cogn Disord 38, 300–309.

25.

Hall

, Wiechmann

, Johnson

, Edwards

, Barber

, Winter

, Singh

, O’Bryant

(2013) Biomarkers of vascular risk, systemic inflammation, and microvascular pathology and neuropsychiatric symptoms in Alzheimer’s disease. J Alzheimers Dis 35, 363–371.

26.

Eurelings

, Richard

, Eikelenboom

, van Gool

, Moll van Charante

(2015) Low-grade inflammation differentiates between symptoms of apathy and depression in community-dwelling older individuals. Int Psychogeriatr 27, 639–647.

27.

Yao

, Mizoguchi

, Monji

, Yakushiji

, Takashima

, Uchino

, Yuzuriha

, Hashimoto

(2019) Low-grade inflammation is associated with apathy indirectly via deep white matter lesions in community-dwelling older adults: The Sefuri Study. Int J Mol Sci 20, 1905.

28.

Waring

, O’Bryant

, Reisch

, Diaz-Arrastia

, Knebl

, Doody

, for the Texas Alzheimer’s Research Consortium (2008) The Texas Alzheimer’s Research Consortium longitudinal research cohort: Study design and baseline characteristics. Texas Public Health J 60, 9–13.

29.

O’Bryant

, Xiao

, Barber

, Reisch

, Doody

, Fairchild

, Adams

, Waring

, Diaz-Arrastia

, Texas Alzheimer’s Research Consortium (2010) A serum protein-based algorithm for the detection of Alzheimer disease. Arch Neurol 67, 1077–1081.

30.

McKhann

, Drachman

, Folstein

, Katzman

, Price

, Stadlan

(1984) Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 34, 939–944.

31.

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.

32.

Lawton

, Brody

(1969) Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 9, 179–186.

33.

O’Bryant

, Waring

, Cullum

, Hall

, Lacritz

, Massman

, Lupo

, Reisch

, Doody

, Texas Alzheimer’s Research C (2008) Staging dementia using Clinical Dementia Rating Scale Sum of Boxes scores: A Texas Alzheimer’s research consortium study. Arch Neurol 65, 1091–1095.

34.

Yesavage

, Brink

, Rose

, Lum

, Huang

, Adey

, Leirer

(1982) Development and validation of a geriatric depression screening scale: A preliminary report. J Psychiatr Res 17, 37–49.

35.

Cummings

, Mega

, Gray

, Rosenberg-Thompson

, Carusi

, Gornbein

(1994) The Neuropsychiatric Inventory: Comprehensive assessment of psychopathology in dementia. Neurology 44, 2308–2314.

36.

Kaufer

, Cummings

, Ketchel

, Smith

, MacMillan

, Shelley

, Lopez

, DeKosky

(2000) Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. J Neuropsychiatry Clin Neurosci 12, 233–239.

37.

Forlenza

, Diniz

, Talib

, Mendonca

, Ojopi

, Gattaz

, Teixeira

(2009) Increased serum IL-1beta level in Alzheimer’s disease and mild cognitive impairment. Dement Geriatr Cogn Disord 28, 507–512.

38.

Diniz

, Teixeira

, Ojopi

, Talib

, Mendonca

, Gattaz

, Forlenza

(2010) Higher serum sTNFR1 level predicts conversion from mild cognitive impairment to Alzheimer’s disease. J Alzheimers Dis 22, 1305–1311.

39.

Lai

KSP

, Liu

, Rau

, Lanctot

, Kohler

, Pakosh

, Carvalho

, Herrmann

(2017) Peripheral inflammatory markers in Alzheimer’s disease: A systematic review and meta-analysis of 175 studies. J Neurol Neurosurg Psychiatry 88, 876–882.

40.

, Zhao

, Xia

, Xu

(2019) Peripheral inflammatory biomarkers in Alzheimer’s disease and mild cognitive impairment: A systematic review and meta-analysis. Psychogeriatrics 19, 300–309.

41.

Lazar

(2015) imputeLCMD: A collection of methods for left-censored missing data imputation. Version: 2.0.

42.

Pocock

, Assmann

, Enos

, Kasten

(2002) Subgroup analysis, covariate adjustment and baseline comparisons in clinical trial reporting: Current practice and problems. Stat Med 21, 2917–2930.

43.

Benjamini

, Hochberg

(1995) Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Stat Soc Series B 57, 289–300.

44.

Tibshirani

(1996) Regression shrinkage and selection via the Lasso. J R Stat Soc Series B 58, 267–288.

45.

Tibshirani

, Taylor

, Lockhart

, Tibshirani

(2016) Exact post-selection inference for sequential regression procedures. J Am Stat Assoc 111, 600–620.

46.

Knopman

, Amieva

, Petersen

, Chetelat

, Holtzman

, Hyman

, Nixon

, Jones

(2021) Alzheimer disease. Nat Rev Dis Primers 7, 33.

47.

Teixeira

, Reis

, Coelho

, Carneiro

, Teixeira

, Vieira

, Mukhamedyarov

, Zefirov

, Janka

, Palotas

(2008) All-or-nothing type biphasic cytokine production of human lymphocytes after exposure to Alzheimer’s beta-amyloid peptide. Biol Psychiatry 64, 891–895.

48.

Van Eldik

, Carrillo

, Cole

, Feuerbach

, Greenberg

, Hendrix

, Kennedy

, Kozauer

, Margolin

, Molinuevo

, Mueller

, Ransohoff

, Wilcock

, Bain

, Bales

(2016) The roles of inflammation and immune mechanisms in Alzheimer’s disease. Alzheimers Dement (N Y) 2, 99–109.

49.

Hayley

, Hakim

, Albert

(2021) Depression, dementia and immune dysregulation. Brain 144, 746–760.

50.

Morgan

, Touchard

, Leckey

, O’Hagan

, Nevado-Holgado

; NIMA Consortium Barkhof

, Bertram

, Blin

, Bos

, Dobricic

, Engelborghs

, Frisoni

, Frölich

, Gabel

, Johannsen

, Kettunen

, Kłoszewska

, Legido-Quigley

, Lleó

, Martinez-Lage

, Mecocci

, Meersmans

, Molinuevo

, Peyratout

, Popp

, Richardson

, Sala

, Scheltens

, Streffer

, Soininen

, Tainta-Cuezva

, Teunissen

, Tsolaki

, Vandenberghe

, Visser

, Vos

, Wahlund

, Wallin

, Westwood

, Zetterberg

, Lovestone

, Morgan

; Annex: NIMA–Wellcome Trust Consortium for Neuroimmunology of Mood Disorders and Alzheimer’s Disease (2019) Inflammatory biomarkers in Alzheimer’s disease plasma. Alzheimers Dement 15, 776–787.

51.

Lyra

ESNM

, Goncalves

, Pascoal

, Lima-Filho

RAS

, Resende

EPF

, Vieira

ELM

, Teixeira

, de Souza

, Peny

, Fortuna

JTS

, Furigo

, Hashiguchi

, Miya-Coreixas

, Clarke

, Abisambra

, Longo

, Donato

, Jr. , Fraser

, Rosa-Neto

, Caramelli

, Ferreira

, De Felice

(2021) Pro-inflammatory interleukin-6 signaling links cognitive impairments and peripheral metabolic alterations in Alzheimer’s disease. Transl Psychiatry 11, 251.

52.

Guimaraes

, Caramelli

, Fialho

PPA

, Franca

, Afonso

MPD

, Teixeira

(2013) Serum levels of soluble TNF-alpha receptors but not BDNF are associated with apathy symptoms in mild Alzheimer’s disease and amnestic mild cognitive impairment. Dement Neuropsychol 7, 298–303.

53.

Couper

, Blount

, Riley

(2008) IL-10: The master regulator of immunity to infection. J Immunol 180, 5771–5777.

54.

D’Anna

, Abu-Rumeileh

, Fabris

, Pistis

, Baldi

, Sanvilli

, Curcio

, Gigli

, D’Anna

, Valente

(2017) Serum interleukin-10 levels correlate with cerebrospinal fluid amyloid beta deposition in Alzheimer disease patients. Neurodegener Dis 17, 227–234.

55.

McGuire

, Ishii

(2016) Leptin dysfunction and Alzheimer’s disease: Evidence from cellular, animal, and human studies. Cell Mol Neurobiol 36, 203–217.

56.

Iikuni

, Lam

, Lu

, Matarese

, La Cava

(2008) Leptin and inflammation. Curr Immunol Rev 4, 70–79.

57.

van Sloten

, Sedaghat

, Carnethon

, Launer

, Stehouwer

CDA

(2020) Cerebral microvascular complications of type 2 diabetes: Stroke, cognitive dysfunction, and depression. Lancet Diabetes Endocrinol 8, 325–336.

58.

Marin

, Biedrzycki

, Firinciogullari

(1991) Reliability and validity of the Apathy Evaluation Scale. Psychiatry Res 38, 143–162.

59.

Radakovic

, Abrahams

(2014) Developing a new apathy measurement scale: Dimensional Apathy Scale. Psychiatry Res 219, 658–663.

60.

Teixeira

, Gonzales

, de Souza

, Weisenbach

(2021) Revisiting apathy in Alzheimer’s disease: From conceptualization to therapeutic approaches. Behav Neurol 2021, 6319826.

61.

Stanton

, Holmes

, Chang

SWC

, Joormann

(2019) From stress to anhedonia: Molecular processes through functional circuits. Trends Neurosci 42, 23–42.

62.

Rocha

, de Miranda

, Teixeira

(2015) Insights into neuroinflammation in Parkinson’s disease: From biomarkers to anti-inflammatory based therapies. Biomed Res Int 2015, 628192.

63.

Colpo

, Stimming

, Rocha

, Teixeira

(2017) Promises and pitfalls of immune-based strategies for Huntington’s disease. Neural Regen Res 12, 1422–1425.

64.

Colpo

, Leboyer

, Dantzer

, Trivedi

, Teixeira

(2018) Immune-based strategies for mood disorders: Facts and challenges. Expert Rev Neurother 18, 139–152.