Negative association of fat mass index and Alzheimer’s disease cerebrospinal fluid biomarkers in cognitively impaired patients

doi:10.21203/rs.3.rs-7980996/v1

Download PDF

Research Article

Negative association of fat mass index and Alzheimer’s disease cerebrospinal fluid biomarkers in cognitively impaired patients

https://doi.org/10.21203/rs.3.rs-7980996/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

INTRODUCTION

Alzheimer's disease (AD) is associated with early weight loss, with unclear underlying mechanisms. Using real-world data, we aimed to investigate the links between biological AD profile, body composition, and nutritional status in patients with cognitive impairment.

METHODS

Cross-sectional study of patients with cognitive decline explored with cerebrospinal fluid (CSF) AD biomarkers and bioelectrical impedance analysis. Nutritional status was assessed using GLIM criteria. Analyses were stratified by sex and adjusted for age, MMSE, comorbidities, and the presence of a caregiver at home.

RESULTS

Among 420 patients, the fat mass index was positively associated with the Aβ42/40 ratio and negatively with pTau181 levels (p < 0.05). Malnutrition was more frequent in women than men (p = 0.0059). In women, malnutrition was associated with higher t-Tau and pTau181 levels (p < 0.02). In both sexes, pTau181 levels were inversely associated with muscle mass (p = 0.041).

CONCLUSION

These results suggest possible pathophysiological links between AD CSF biomarkers and nutritional status.

Alzheimer’s disease

Neurofilament proteins

Cognitive domains performance

Neuroimaging

Neurodegenerative diseases

Cognitive decline

A large body of evidence links nutritional status and body composition with the pathophysiology of neurodegenerative disorders, specifically in Alzheimer’s disease (AD). Chronic inflammation, altered energy metabolism, and dysregulation of appetite-related hormones were suggested to explain the association between weight loss and AD [1].

In addition, reciprocal relationships between nutritional status and cognitive decline were consistently reported. Notably, unintentional weight loss or lower Body Mass Index (BMI) were described several years before the onset of clinically detectable neurocognitive disorders, as evidenced by large-scale prospective cohort studies [2–5]. In AD, accelerated cognitive decline was also reported in individuals with notable weight loss during the first year of follow-up [6] or with lower BMI at baseline [7]. Similar results were found in a general population cohort of 3035 patients, in which higher baseline BMI and smaller weight loss were associated with a slower rate of cognitive decline [8]. These findings are further supported by the results of the Dominantly Inherited Alzheimer Network (DIAN) study[9], in which asymptomatic carriers began to lose weight approximately 11 years earlier than asymptomatic non-carriers. Moreover, in the same cohort, lower BMI in the preclinical stage was associated with earlier symptom onset and higher amyloid burden.

The use of bioelectrical impedance analysis assesses the body composition by notably estimating lean mass, muscle mass, and fat mass. Indeed, low muscle mass is a phenotypic criterion for sarcopenia and for malnutrition[10] [11]. Findings about the association of incident dementia and body composition are inconsistent in the current literature. Two general population studies based on the UK Biobank cohort reported an inverse association between incident dementia and the fat-to-muscle ratio [12,13]. Other studies reported incident dementia to be associated with lower lean mass [14]. In contrast, other results found no association between incident dementia and body composition [15] or fat mass [14].

A few studies examined body composition specifically in AD. They reported lower lean mass in AD patients compared to controls [16], a higher prevalence of sarcopenia in moderate versus mild AD [17], and in AD patients compared with controls [18,19]. Interestingly, lower fat mass and lower lean mass were also associated with lower cortical thickness in AD-affected brain regions [20,21]. However, in those studies, AD diagnoses were made according to former AD diagnosis criteria (Mc Kahn 1984) without AD CSF biomarkers.

Using CSF biomarker evidence [22,23], a publication by Doorduijn et al. observed a significant inverse association between phosphorylated Tau on threonine 181 (pTau181) and total-Tau (t-tau) with fat-free mass index, but no association with amyloid biomarkers [24] in 558 participants, including AD, mild cognitive impairment, and controls. They did not explore the sex differences.

Overall, the available evidence on the relationship between cognitive decline, nutritional status, and biological biomarkers of AD remains limited, particularly regarding to sex differences and possible confounding factors.

In the present study, we aimed to examine the relationships between AD cerebrospinal fluid (CSF) biomarkers, nutritional status, and body composition. Our objectives were: (1) to investigate the prevalence of malnutrition and low muscle mass in cognitive disorders; (2) to examine the association between AD CSF biomarkers and body composition, possibly suggesting a potential pathophysiological link, and (3) to identify the factors associated with malnutrition in these patients.

2.1. Study Design and Participants

We performed a prospective multicentric study including patients evaluated for neurocognitive disorders who underwent CSF analysis for AD biomarker testing. Patients were enrolled from January 2022 to July 2025 at three centers at the Université Paris Cité GHU AP-HP Nord, Paris, France, including Lariboisière Hospital, a tertiary memory clinic, and two geriatric day-care hospitals: Bretonneau and Fernand Widal.

2.1.1. Inclusion and Exclusion Criteria

All patients who had a lumbar puncture for CSF AD biomarker measurement and signed a written informed consent for the BioCogBank study (see § 2.6) were included. Patients with conditions likely to cause inflammation, such as active cancer, infectious diseases, and chronic inflammatory diseases, and any chronic gastrointestinal condition, that adversely impact food assimilation or absorption, were excluded.

2.1.2. Cognitive disease diagnostic assessment

In each center, patients were evaluated by multidisciplinary teams composed of geriatricians, neurologists, neuropsychologists, biochemists, and neuroradiologists. Diagnoses were established after multidisciplinary team assessment including examination of CSF biomarkers, MRI and fluorodeoxyglucose positron emission tomography (PET) imaging. AD was diagnosed using the most recent diagnostic criteria for AD [25]. Other neurodegenerative diseases (OND) included dementia with Lewy Bodies (DLB) [26], frontotemporal dementia (FTD) [27], supranuclear palsy [28], corticobasal degeneration [29] or patients with cognitive decline and brain atrophy or hypometabolism on PET considered to have a neurodegenerative process after multidisciplinary case review. Patients were classified as having a non-neurodegenerative disorder (NND) if they did not meet criteria for neurodegenerative diseases (as defined above) and were considered as non-neurodegenerative by the clinician in charge, after multidisciplinary team assessment, including evaluation of CSF biomarkers, MRI, PET imaging, and cognitive follow-up. Patients without a confirmed diagnosis were classified as “unclassified dementia” and were excluded from the diagnostic-subgroup analysis.

2.2. CSF Biomarkers analysis

After lumbar puncture, CSF samples were centrifuged at 1000g for 10 min at 4°C within 2 h of collection and then aliquoted into 0.5 mL polypropylene tubes before being stored at − 80°C for further analysis.

CSF t-tau, pTau181, and Aβ42 measurements were conducted in the Biochemistry Unit of Lariboisière Hospital using Elecsys® assays on the cobas e601 analyser (Roche Diagnostics). CSF Aβ40 measurement was performed using an Innotest ELISA assay (Fujirebio, Ghent, Belgium). Patients were considered as amyloid positive (A+) according to Aβ 42/40 ratio below 0.083, tau positive (T+) if pTau181 level was over 22 pg/mL and neurodegeneration positive (N+) if t-tau was over 220 pg/mL.

Lumbar puncture for AD biomarker measurements was performed within six months of the body composition assessment (mean delay: 0.55 ± 1.94 months)

2.3. Nutritional status determination and bioelectrical impedance measurements

Anthropometric measurements (weight and height) were obtained using standardized equipment. Multifrequency bioelectrical impedance analysis (BIA) was performed within six months before or after the lumbar puncture, using the InBody S10®. The data collected on fat mass, fat-free mass, and appendicular skeletal muscle were used to determine:

Fat mass index: Fat mass (kg) / height (m²)
Fat-free mass index: Fat-free mass (kg) / height(m²)
Appendicular skeletal muscle mass (ASM, kg)
Appendicular skeletal muscle mass index: appendicular skeletal muscle (kg)/ height (m²)

Nutritional status, classified as no malnutrition, moderate malnutrition or severe malnutrition, was determined according to the 2019 Global Leadership Initiative on Malnutrition (GLIM) [10], requiring the combination of at least one phenotypic criteria and at least one aetiologic criteria. The phenotypic criteria are low BMI (< 20 for patients under 70 years old and < 22 for patients aged 70 and over), weight loss > 10% or low muscle mass. Low muscle mass was determined using the sex-specific thresholds of the revised European consensus on definition and diagnosis of sarcopenia, considering appendicular skeletal muscle mass < 20 kg in men or 15 kg in women and appendicular skeletal muscle mass index < 7 kg/m² in men or 5.5 kg/m² in women [30]. The aetiologic criteria for malnutrition were reduced appetite (using the Mini-Nutritional Assessment short form specific item), chronic diseases and major neurocognitive disorders. Based on blood tests, patients with biologically detectable inflammation or malabsorption were excluded.

2.4. Other Variables

Additional variables, potentially affecting nutritional status and body composition, were also recorded:

Functional status assessed by the 4-item Instrumental Activities of Daily Living (IADL) score [31] and the Activities of Daily Living (ADL) score [32]
Home assistance evaluated with the presence of professional help at home (yes/no), and the presence of a relative living at home with the patient (yes/no)
Comorbidities assessed using the Charlson Comorbidity Index (CCI) score [33]. The modified CCI was calculated excluding the age component to assess the comorbidity burden independently of age
Cognitive impairment severity was measured using the Mini-Mental State Examination (MMSE) score [34].

2.5. Statistical Analyses

2.5.1. Descriptive Analyses

Due to substantial sex-related differences in body composition parameters, analyses were stratified by sex. Based on a hypothesis of a moderate correlation (r=-0.2) between CSF Aβ 42/40 ratio and fat mass[24] 193 patients are needed to reach a power of 80% with an alpha-risk at 5% (using Fisher Z transformation method) in each sex, meaning a total of 386 patients needed.

Categorical variables were compared using the χ² test, or Fisher's exact test as appropriate. Continuous variables were compared using Student's t-test for variables following a parametric distribution and presented using mean ± standard deviation (SD). Variables of non-parametric distribution, of MMSE, ADL, and IADL scores were compared using the Wilcoxon test and presented using the median (1st ; 3rd quartiles).

2.5.2. Multivariate Analyses

Missing data for covariates (age, modified CCI, presence of a caregiver at home) were treated by multiple imputation. All numerical variables were centered and scaled considering the mean and standard deviation of the population, both for linear and logistic regressions. Linear regression models examining BMI, fat mass index, fat-free mass index, and appendicular skeletal muscle mass index on the one hand, and CSF Aβ-42/Aβ-40 ratio, pTau181, and t-tau on the other hand, were adjusted for age, modified CCI score, MMSE score, and the presence of a caregiver at home, as clinically relevant variables, linked to malnutrition. Diagnostic subgroup analyses were performed to evaluate the relationship between AD CSF biomarkers and body composition metrics in each diagnosis group. These analyses were also adjusted for sex, age, modified CCI score, MMSE score, and presence of a caregiver at home, but not stratified by sex.

The association between malnutrition and CSF AD biomarkers was assessed using multiple logistic regression, adjusting for age, modified CCI score, MMSE score, and presence of a caregiver at home. Analysis across the global population was also adjusted for sex. All multivariate analyses were stratified by sex to account for sex-specific differences in fat mass and lean mass.

In all analyses, the type I error risk α was set at 0.05 and the power at 80%. Statistical analyses were performed using the R software (Version 4.3.1).

2.6. Ethics

All data and samples analyzed were part of the BioCogBank study (NCT06244875). All participants gave written informed consent. This study was approved by the “Comité de Protection des Personnes Est III” (2023-A01413-42, 23-10-02) and by the "Commission Nationale Informatique et Libertés" (CNIL)

3.1. Population’s characteristics

Patients’ characteristics are presented in Table 1. A total of 420 participants were included: 240 (57.1%) women and 180 (42.9%) men. Our study sample mainly comprised mild-stage neurocognitive disorders, with a median MMSE of 23 [interquartile range 19; 26]. The most frequent diagnosis was AD (57.4%), followed by NND (13.6%), OND (12.1%), and 16.9% of the population had unclassified dementia at the time of the study. Women had a lower comorbidity index (mean CCI score 4.09 ± 1.38 vs 4.51 ± 1.66 for men, p = 0.0065) and lived alone more often than men (52.1% vs 21.1%, p < 0.001).

3.2. Prevalence of malnutrition and muscle mass loss in neurocognitive disorders

Regarding nutritional assessment, malnutrition was diagnosed in 31.7% of the patients, including 15.5% with severe malnutrition (Table 1). Low muscle mass was observed in 18.1% of the patients, low BMI in 27.1% and > 10% weight loss in 11.7%.

Women were more frequently malnourished than men (37.1% vs 24.4%, p = 0.0059). The prevalence of severe malnutrition was higher in women (20.8%) than in men (8.3%, p = 0.0034). In addition, women also had a more frequent lower muscle mass (22.9%) than men (11.7%), p = 0.0033.

3.3. Association between CSF AD biomarkers and body composition

All data are presented in Table 2 and Fig. 1.

Table 2
Association of Aβ-42/Aβ-40 ratio, pTau 181, and total Tau with body composition according to sex.
	Women							Men
		Aβ42/40 ratio		pTau 181		Total Tau		Aβ42/Aβ40 ratio		pTau 181		Total Tau
		Estimate (CI)	p	Estimate (CI)	p	Estimate (CI)	p	Estimate (CI)	p	Estimate (CI)	p	Estimate (CI)	p
BMI	Model 1	0.15 (0.03 ; 0.28)	0.018	-0.19 (-0.31 ; -0.06)	0.004	-0.19 (-0.32 ; -0.07)	0.003	0.24 (0.09 ; 0.38)	0.001	-0.11 (-0.26 ; 0.03)	0.12	-0.13 (-0.28 ; -0.01)	0.073
	Model 2	0.12 (-0.01 ; 0.26)	0.071	-0.19 (-0.32 ; -0.07)	0.003	-0.21 (-0.33 ; -0.08)	0.002	0.23 (0.08 ;0.38)	0.002	-0.10 (-0.26 ; 0.06)	0.20	-0.13 (-0.29 ; 0.03)	0.11
FMI	Model 1	0.15 (0.02 ; 0.27)	0.024	-0.15 (-0.28 ; -0.03)	0.017	-0.16 (-0.29 ; -0.03)	0.013	0.23 (0.08 ; 0.37)	0.002	-0.11 (-0.26 ; 0.04)	0.15	-0.09 (-0.24 ; 0.06)	0.22
	Model 2	0.15 (0.01 ; 0.28)	0.034	-0.17 (-0.30 ; -0.04)	0.011	-0.18 (-0.31 ; -0.05)	0.006	0.27 (0.12 ; 0.41)	< 0.001	-0.16 (-0.32 ; -0.00)	0.047	-0.15 (-0.30 ; 0.01)	0.073
FFMI	Model 1	0.07 (-0.06 ; 0.20)	0.27	-0.13 (-0.26 ; -0.001)	0.043	-0.13 (-0.26 ; -0.01)	0.040	0.11 (-0.03 ; 0.26)	0.13	-0.03 (-0.18 ; 0.12)	0.69	-0.05 (-0.20 ; 0.09)	0.48
	Model 2	0.01 (-0.12 ; 0.14)	0.89	-0.13 (-0.25 ; -0.001)	0.049	-0.13 (-0.26 ; -0.01)	0.037	0.05 (-0.09 ; 0.19)	0.48	0.06 (-0.09 ; 0.21)	0.44	0.02 (-0.13 ; 0.18)	0.75
ASMI	Model 1	0.02 (-0.10 ; 0.15)	0.70	-0.15 (-0.29 ; -0.02)	0.022	-0.15 (-0.28 ; -0.03)	0.017	0.10 (-0.04 ; 0.25)	0.17	-0.01 (-0.16 ; 0.13)	0.85	-0.02 (-0.17 ; 0.13)	0.78
	Model 2	-0.05(-0.18 ; 0.09)	0.49	-0.14 (-0.26 ; -0.01)	0.034	-0.15 (-0.27 ; -0.02)	0.022	0.05 (-0.10 ; 0.19)	0.53	0.06 (-0.09 ; 0.22)	0.42	0.05 (-0.10 ; 0.20)	0.53
Model 1: no adjustment. Model 2: adjustment for age, modified Charlson score (excluding age), MMSE and the presence of a caregiver at home. All variables have been centered and scaled. BMI: body mass index, FMI: fat mass index, ASMI: appendicular skeletal muscle mass index, FFMI: Fat free mass index
All continuous variables were standardized for analysis. Results should therefore be interpreted as “the concerned body composition parameters grows of β standard deviation units by standard deviation majoration of the concerned biomarker, after adjustment on confounding variables

In both sexes, after adjustment for confounding factors, we observed lower fat mass index to be significantly associated with lower Aβ-42/Aβ-40 ratio, reflecting higher amyloid burden (women β = 0.15, p = 0.034; men β = 0.27, p < 0.001). Similarly, higher pTau181 levels were significantly associated with lower fat mass index (women β=-0.17, p = 0.011, men β=-0.16, p = 0.047). There was no association between the Aβ-42/Aβ-40 ratio and muscle mass (fat-free mass index and skeletal muscle index).

Sex-related differences in the associations between nutritional status and CSF AD biomarkers were also detected. Lower BMI was significantly associated with a lower Aβ-42/Aβ-40 ratio (reflecting higher amyloid burden) in men (β = 0.23, p = 0.002) but not in women (β = 0.12, p = 0.071). Lower BMI was also associated with higher pTau181 level (reflecting phospho-tau accumulation) in women (β = -0.19, p = 0.003), but not in men (β = -0.10, p = 0.20). In addition, lower fat mass index was associated with higher total tau level (β=-0.18, p = 0.006) in women, but not in men (β=-0.15, p = 0.073). Conversely, lower fat-free mass index (β = − 0.13, p = 0.049) and appendicular skeletal muscle mass index (β = − 0.14, p = 0.034) were associated with higher pTau-181 level and with higher t-tau levels (β = − 0.13, p = 0.037; β = − 0.15, p = 0.022) only in women.

We performed a subgroup analysis by diagnosis for the AD, NND, and OND groups, adjusting for age, sex, modified Charlson score, MMSE, and presence of a caregiver at home (supplementary table 1 and supplementary Fig. 1). In the AD subgroup, a lower BMI was associated with higher pTau181 and t-tau levels (respectively β=-0.14, p = 0.024 and β=-0.15, p = 0.011). Lower fat mass index was associated with higher t-tau (β=-0.12, p = 0.032), and low appendicular skeletal muscle mass index was related to higher pTau and t-tau levels (respectively β=-0.10, p = 0.028 and β=-0.11, p = 0.024).

No association was observed between biomarkers and body composition in NND patients. In OND patients, a lower BMI was associated with a lower Aβ-42/Aβ-40 ratio (β = 0.33, p = 0.011), and a lower fat mass index was also associated with a lower Aβ-42/Aβ-40 ratio (β = 0.27, p = 0.021) (reflecting higher amyloid burden).

3.4. Clinical factors associated with malnutrition

Clinical and social factors associated with malnutrition in the univariate analysis are presented in Supplementary Table 2. In our total cohort, malnourished patients were older (76.5 ± 7.17 years) than non-malnourished ones (74.1 ± 8.21 years, p = 0.0031), and women were more represented in the malnourished group than in the non-malnourished group (66.9% vs 52.6%, p = 0.0059). Median IADL score was lower in malnourished patients (3.0 [2;4]) than in non-malnourished patients (4.0 [2;4] p < 0.001).

The mean CCI score was higher in malnourished patients (4.51 ± 1.35) than in non-malnourished individuals (4.16 ± 1.59, p = 0.031), and social isolation was more frequent in malnourished patients (49.6% vs 33.8%, p = 0.0035). The median MMSE score was also lower in malnourished patients (22, [19; 25]) than in their non-malnourished counterparts (24, [20; 27], p = 0.0096).

Clinical characteristics associated with malnutrition in the multivariate analysis are reported in Table 3 and Fig. 2. In the total sample, malnutrition was positively associated with T-tau (OR 1.24, [95% CI 1.005–1.56], p = 0.048). Malnutrition was not associated with Aβ42/Aβ40 ratio or pTau181 after adjustment for age, modified CCI score, MMSE, and the presence of a non-professional caregiver at home.

Table 3
*Effect of clinical outcomes in the relationship between Alzheimer’s disease CSF biomarkers and malnutrition*
		Total (n = 420)		Men (n = 180)		Women (n = 240)
		OR (CI)	p	OR (CI)	p	OR (CI)	p
Ratio Aβ 42/40	Model 1	0.73 (0.56 ; 0.93)	0.018	0.84 (0.53 ; 1.20)	0.39	0.73 (0.54 ; 0.97)	0.035
Ratio Aβ 42/40	Model 2	0.87 (0.65 ; 1.11)	0.29	0.92 (0.58 ; 1.33)	0.69	0.81 (0.59 ; 1.11)	0.20
pTau181	Model 1	1.38 (1.13 ; 1.70)	0.002	0.98 (0.69 ; 1.37)	0.92	1.49 (1.14 ; 1.98)	0.004
pTau181	Model 2	1.21 (0.98 ; 1.51)	0.078	0.85 (0.57 ; 1.24)	0.41	1.41(1.07 ; 1.88)	0.017
t-Tau	Model 1	1.42 (1.16 ; 1.75)	0.001	1.01 (0.71 ; 1.41)	0.95	1.53 (1.16 ; 2.04)	0.003
t-Tau	Model 2	1.24 (1.005 ; 1.56)	0.048	0.89 (0.60 ; 1.31)	0.58	1.45 (1.10 ; 1.96)	0.012
Multiple logistic regression based on the biomarker alone (model a), and after adjustment for age, modified Charlson score (excluding age), MMSE, and the presence of a caregiver at home (model b). Analysis on total population were further adjusted on sex
All variables were standardized for analysis. Results can be interpreted as xx% change in risk of malnutrition associated with one standard deviation majoration in the biomarker of interest, after adjustment for confounding variable

In women, malnutrition was positively associated with pTau181 (OR 1.41 [95% CI 1.07–1.88], p = 0.017) and T-tau (OR 1.45 [95% CI 1.10–1.96], p = 0.012), after adjustment for confounding factors. Despite an association of the Aβ42/Aβ40 ratio with malnutrition in our univariate analysis, this association was no longer significant after adjustment for confounding factors. No association between CSF biomarkers and malnutrition was found in men.

Logistic regression models odds ratio adjusted for age, modified CCI score, MMSE score, and presence of a caregiver at home. Global population analysis is further adjusted on sex.

In this study, based on clinical practice, using the most recent diagnosis tools and nutritional status criteria, we described the specific prevalence of malnutrition in these populations. In the whole cohort, we found in both sexes that the fat mass index was positively correlated with amyloid ratio CSF levels and inversely correlated to pTau181 levels. In summary, a low fat mass index is associated with high amyloid and pTau181 brain accumulations. In AD patients, a lower fat mass index was inversely correlated with CSF T-tau levels, reflecting brain T-tau accumulation. In women, the muscle mass was inversely associated with pTau181 and t-tau. pTau181 and t-tau were independently associated with malnutrition after adjustment for confounding factors. Finally, regarding the other factors associated with malnutrition, we observed that age, sex, social isolation, and comorbidity burden were all associated with malnutrition in the univariate analysis.

With regards to our first objective, the already reported prevalence of malnutrition in cognitively impaired individuals, as measured by the MNA questionnaire, varies widely from 2.9% to 17.4% according to previous studies [35–37]. In our results, the prevalence of malnutrition and low muscle mass was much higher, including a significant sex-related disparity in disfavor of women. Indeed, we used the GLIM criteria, which include age-specific BMI thresholds, assessment of muscle mass and weight loss over an extended period. These criteria allow for a more specific and objective determination of the nutritional status, and could explain for the discordant findings between previous studies and our results. As illustrated by Ozer et al., in a population without cognitive decline, 32.2% of older adults without dementia were malnourished using the GLIM criteria, whereas the MNA and MNA-SF only identified malnutrition risk in 12.7% and 13.1% of subjects [38]. MNA primarily identifies the risk for malnutrition [17,21,24,35,37] without any confirmation and independently of age and sex; furthermore, the reliability of the questionnaire could be inferior in patients with cognitive decline, altogether potentially leading to an underestimated malnutrition. Overall, these results underscore the important need to implement rigorous nutritional assessment and management strategies, aiming at detecting, understanding, preventing, and treating malnutrition in this population.

Our second objective was to investigate the relationships between AD CSF biomarkers and body composition parameters. Previously, a cross-sectional study from Doorduijn et al. using the MNA score in a prospective cohort from daily practice of more than 500 AD patients and controls, reported an inverse association between both pTau181 and t-tau and fat-free mass, but no association with the Aβ42 peptide. However, analyses were not stratified by sex in this study, but were only adjusted for sex.[24]. In our study, we observed similar results, but only in women associated with an inverse association of pTau 181 and t-tau with muscle mass parameters. We mainly found an association of amyloid CSF biomarker (amyloid accumulation) and CSF pTau181 (pTau accumulation) with fat mass, which was not previously described in relation to CSF biomarkers, in mild cognitive impairment.

In AD patients, a lower fat mass index was linked to total tau brain accumulation. These results imply that in neurocognitive disorders, low levels of body fat are associated with amyloid and pTau 181 brain accumulations and that in AD patients, a lower fat mass index was linked to increased neurodegeneration. Two studies in the general population have specifically examined the relationship between fat mass or muscle mass and plasma AD biomarkers and found different results. De Crom et al. showed that fat mass was positively associated with t-tau [39], while Hermesdorf et al. found fat mass was inversely associated with β-amyloid 42 levels and positively associated with Aβ-40/Aβ-42 ratio [40]. These two studies focused on the general population without cognitive impairment, younger age, and higher BMI, which can explain, at least partially, the difference with our results. Moreover, unlike CSF biomarkers, the relationships between plasma biomarkers and body composition may be confounded by variations in volume of distribution, making it difficult to establish a true pathophysiological connection [41].

Interestingly, all the reported links between AD biomarkers and nutritional status suggest a possible common pathophysiological hypothesis. We found a marked association between amyloid, tau biomarkers and fat mass, both in men and women, suggesting that AD-related pathology and neurodegeneration may be linked to a preferential loss of fat mass relative to muscle mass. Leptin deficit may be a cause of this association [42]. Leptin is mainly produced by adipocytes and is involved in the regulation of appetite and energy expenditure; it also plays a neuroprotective role, stimulating neurogenesis and participating in neurocognitive functions such as memory or learning [43]. Ishii et al. showed that AD mouse models exhibited reduced fat mass and lower leptin levels compared with wild-type controls. In response to these low leptin levels, AD mice revealed low levels of neuropeptide Y (NPY), an orexigenic peptide, and not high levels as expected [44]. In another AD mouse model, Robison et al. suggested that impaired hypothalamic signaling could explain the inadequate response to low leptin levels observed in the context of AD, thus favoring weight and fat mass loss [45].

Data from these preclinical models provide insights into potential mechanisms underlying our findings, highlighting the association of tau and amyloid CSF biomarkers with fat mass loss in both sexes. In AD patients, the links between neurodegeneration and low fat mass index could be associated with reduced leptin neuroprotection, as shown in our previous studies [42,46].

Moreover, our diagnostic subgroup results support a specific link between brain amyloid deposition and fat mass reduction, even in OND patients. Nevertheless, further studies investigating markers of neuroinflammation or neurodegeneration in relation to body composition changes across different neurodegenerative diseases are needed to better clarify the nature of this association.

Another key finding of this study is the association between malnutrition and CSF AD biomarkers in women. The sex difference observed in our study is consistent with preclinical studies in this field. Lopez-Gamboro et al. reported that neuroinflammation was associated with reduced food intake and subsequent weight loss in female mutant mice [47], in line with our results. However, in humans, other socio-environmental factors can be involved. Notably, women experienced more social isolation than men in our study, which could contribute to malnutrition.

The strengths of our study include its multicenter design, involving both geriatric and neurological centers, allowing the inclusion of patients across a broad age range. We also specifically performed sex-stratified analyses, enabling sex specificities that could lead to adapted care according to sex. However, several limitations should be acknowledged. The absence of longitudinal data and missing information on prior weight loss likely led to an underestimation of the prevalence of malnutrition. Given the hypothesis of progressive, gradual weight loss due to reduced appetite in neurodegenerative diseases, cross-sectional measurements of fat and muscle mass do not capture individuals’ changes in body composition during the disease. Moreover, despite exclusion criteria and adjustment for confounders, malnutrition is often multifactorial, and it remains challenging to account for all relevant factors, such as mood disorders, hospitalizations and environmental or caregiving conditions. Finally, limited statistical power may restrict the generalizability of our findings in men.

Our study highlights a close association between malnutrition, loss of fat and muscle mass, and AD CSF biomarkers, which could suggest a common pathophysiological link. Further longitudinal investigations, including the evaluation of appetite-regulating hormones (such as glucagon-like peptide-1 and others), are needed. In parallel, preventive strategies aimed at maintaining adequate nutritional status—through balanced diets, early screening for malnutrition, and tailored nutritional interventions—may help limit or delay cognitive decline and neurodegeneration potentially associated with malnutrition.

Ethics approval

All participants participated in the BioCogBank protocol NCT06244875), and gave written consent for CSF and plasma collection for further analysis. They also provided consent for the use of their clinical data and the results of their CSF analyses. This study was approved by local and national Ethics Committees ("Comité d'évaluation et d'Ethique pour la recherche Paris Nord" and by the "Commission Nationale Informatique et Libertés" (CNIL)).

Consent for publication

Not applicable

Competing interest

None

Funding

None

Author Contribution

Karl Götze: Conceptualization, Data curation, formal analysis, methodology, writing – original draft, writing – review and editingAgathe Vrillon: Data curation, writing – original draft, writing – review and editingManuel Sanchez : Methodology, writing – review and editingInès Petit-Damico : Data curation, writing – original draft, writing – review and editingElodie Bouaziz-Amar : formal analysis, writing – review and editingSophie Lacaille :Data curation, writing – review and editingClaire Hourrègue : Data curation, writing – review and editingEmmanuel Cognat : Data curation, writing – review and editingJacques Hugon : Data curation, writing – review and editingThéodore Decaix : Data curation, writing – review and editingJulien Dumurgier : Data curation, writing – review and editingAgathe Raynaud-Simon : Methodology, writing – review and editingMatthieu Lilamand : Conceptualization, writing – review and editingClaire Paquet : Conceptualization, writing – original draft, writing – review and editing

Acknowledgement

None

Data Availability

The dataset used and/or analysed during the current study are available from the corresponding author on reasonable request

G S, M DR, A C, Em I, E M. Weight loss and Alzheimer’s disease: temporal and aetiologic connections. The Proceedings of the Nutrition Society [Internet]. Proc Nutr Soc; 2013 [cited 2025 July 27];72. https://doi.org/10.1017/S0029665112002753
Fitzpatrick AL, Kuller LH, Lopez OL, Diehr P, O’Meara ES, Longstreth WT, et al. Midlife and late-life obesity and the risk of dementia: cardiovascular health study. Arch Neurol. 2009;66:336–42. https://doi.org/10.1001/archneurol.2008.582
Stewart R, Masaki K, Xue Q-L, Peila R, Petrovitch H, White LR, et al. A 32-year prospective study of change in body weight and incident dementia: the Honolulu-Asia Aging Study. Arch Neurol. 2005;62:55–60. https://doi.org/10.1001/archneur.62.1.55
Wagner M, Helmer C, Tzourio C, Berr C, Proust-Lima C, Samieri C. Evaluation of the Concurrent Trajectories of Cardiometabolic Risk Factors in the 14 Years Before Dementia. JAMA Psychiatry. 2018;75:1033–42. https://doi.org/10.1001/jamapsychiatry.2018.2004
Singh-Manoux A, Dugravot A, Shipley M, Brunner EJ, Elbaz A, Sabia S, et al. Obesity trajectories and risk of dementia: 28 years of follow-up in the Whitehall II Study. Alzheimers Dement. 2018;14:178–86. https://doi.org/10.1016/j.jalz.2017.06.2637
Soto ME, Secher M, Gillette-Guyonnet S, Abellan van Kan G, Andrieu S, Nourhashemi F, et al. Weight loss and rapid cognitive decline in community-dwelling patients with Alzheimer’s disease. J Alzheimers Dis. 2012;28:647–54. https://doi.org/10.3233/JAD-2011-110713
Cronk BB, Johnson DK, Burns JM, Alzheimer’s Disease Neuroimaging Initiative. Body mass index and cognitive decline in mild cognitive impairment. Alzheimer Dis Assoc Disord. 2010;24:126–30. https://doi.org/10.1097/WAD.0b013e3181a6bf3f
Liang F, Fu J, Moore JB, Zhang X, Xu Y, Qiu N, et al. Body Mass Index, Waist Circumference, and Cognitive Decline Among Chinese Older Adults: A Nationwide Retrospective Cohort Study. Front Aging Neurosci. 2022;14:737532. https://doi.org/10.3389/fnagi.2022.737532
Müller S, Preische O, Sohrabi HR, Gräber S, Jucker M, Dietzsch J, et al. Decreased body mass index in the preclinical stage of autosomal dominant Alzheimer’s disease. Sci Rep. Nature Publishing Group; 2017;7:1225. https://doi.org/10.1038/s41598-017-01327-w
Cederholm T, Jensen GL, Correia MITD, Gonzalez MC, Fukushima R, Higashiguchi T, et al. GLIM criteria for the diagnosis of malnutrition - A consensus report from the global clinical nutrition community. Clin Nutr. 2019;38:1–9. https://doi.org/10.1016/j.clnu.2018.08.002
Diagnostic de la dénutrition chez la personne de 70 ans et plus [Internet]. Haute Autorité de Santé. [cited 2025 Aug 15]. https://www.has-sante.fr/jcms/p_3165944/fr/diagnostic-de-la-denutrition-chez-la-personne-de-70-ans-et-plus. Accessed 15 Aug 2025
Wang W, Luo Y, Zhuang Z, Song Z, Huang N, Li Y, et al. Total and regional fat-to‐muscle mass ratio and risks of incident all‐cause dementia, Alzheimer’s disease, and vascular dementia. J Cachexia Sarcopenia Muscle. 2022;13:2447–55. https://doi.org/10.1002/jcsm.13054
Xu S, Wen S, Yang Y, He J, Yang H, Qu Y, et al. Association Between Body Composition Patterns, Cardiovascular Disease, and Risk of Neurodegenerative Disease in the UK Biobank. Neurology. 2024;103:e209659. https://doi.org/10.1212/WNL.0000000000209659
Cui C, Mackey RH, Shaaban CE, Kuller LH, Lopez OL, Sekikawa A. Associations of body composition with incident dementia in older adults: Cardiovascular Health Study-Cognition Study. Alzheimers Dement. 2020;16:1402–11. https://doi.org/10.1002/alz.12125
Abellan van Kan G, Rolland Y, Gillette-Guyonnet S, Gardette V, Annweiler C, Beauchet O, et al. Gait speed, body composition, and dementia. The EPIDOS-Toulouse cohort. J Gerontol A Biol Sci Med Sci. 2012;67:425–32. https://doi.org/10.1093/gerona/glr177
Burns JM, Johnson DK, Watts A, Swerdlow RH, Brooks WM. Reduced lean mass in early Alzheimer disease and its association with brain atrophy. Arch Neurol. 2010;67:428–33. https://doi.org/10.1001/archneurol.2010.38
Unsal P, Guner M, Ozsurekci C, Balli N, Bas AO, Ozturk Y, et al. Prevalence of nutrition disorders and nutrition-related conditions in older patients with Alzheimer’s disease. Nut in Clin Prac. 2023;38:1142–53. https://doi.org/10.1002/ncp.10995
Ogawa Y, Kaneko Y, Sato T, Shimizu S, Kanetaka H, Hanyu H. Sarcopenia and Muscle Functions at Various Stages of Alzheimer Disease. Front Neurol. 2018;9:710. https://doi.org/10.3389/fneur.2018.00710
Sugimoto T, Ono R, Murata S, Saji N, Matsui Y, Niida S, et al. Prevalence and associated factors of sarcopenia in elderly subjects with amnestic mild cognitive impairment or Alzheimer disease. Curr Alzheimer Res. 2016;13:718–26. https://doi.org/10.2174/1567205013666160211124828
Samuelsson J, Marseglia A, Wallengren O, Lindberg O, Dartora C, Cedres N, et al. Association of body composition with neuroimaging biomarkers and cognitive function; a population-based study of 70-year-olds. eBioMedicine. 2025;112:105555. https://doi.org/10.1016/j.ebiom.2024.105555
Verhaar BJH, de Leeuw FA, Doorduijn AS, Fieldhouse JLP, van de Rest O, Teunissen CE, et al. Nutritional status and structural brain changes in Alzheimer’s disease: The NUDAD project. Alzheimers Dement (Amst). 2020;12:e12063. https://doi.org/10.1002/dad2.12063
Jack CR, Bennett DA, Blennow K, Carrillo MC, Dunn B, Haeberlein SB, et al. NIA-AA Research Framework: Toward a biological definition of Alzheimer’s disease. Alzheimer’s & Dementia. 2018;14:535–62. https://doi.org/10.1016/j.jalz.2018.02.018
Jack CR, Andrews JS, Beach TG, Buracchio T, Dunn B, Graf A, et al. Revised criteria for diagnosis and staging of Alzheimer’s disease: Alzheimer’s Association Workgroup. Alzheimer’s & Dementia. 2024;alz.13859. https://doi.org/10.1002/alz.13859
Doorduijn AS, Visser M, van de Rest O, Kester MI, de Leeuw FA, Boesveldt S, et al. Associations of AD Biomarkers and Cognitive Performance with Nutritional Status: The NUDAD Project. Nutrients. 2019;11:1161. https://doi.org/10.3390/nu11051161
McKhann GM, Knopman DS, Chertkow H, Hyman BT, Jack CR, Kawas CH, et al. 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. 2011;7:263–9. https://doi.org/10.1016/j.jalz.2011.03.005
McKeith IG, Boeve BF, Dickson DW, Halliday G, Taylor J-P, Weintraub D, et al. Diagnosis and management of dementia with Lewy bodies: Fourth consensus report of the DLB Consortium. Neurology. 2017;89:88–100. https://doi.org/10.1212/WNL.0000000000004058
Rascovsky K, Hodges JR, Knopman D, Mendez MF, Kramer JH, Neuhaus J, et al. Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain. 2011;134:2456–77. https://doi.org/10.1093/brain/awr179
Höglinger GU, Respondek G, Stamelou M, Kurz C, Josephs KA, Lang AE, et al. Clinical diagnosis of progressive supranuclear palsy: The movement disorder society criteria: MDS Clinical Diagnostic Criteria for PSP. Mov Disord. 2017;32:853–64. https://doi.org/10.1002/mds.26987
Armstrong MJ, Litvan I, Lang AE, Bak TH, Bhatia KP, Borroni B, et al. Criteria for the diagnosis of corticobasal degeneration. Neurology. 2013;80:496–503. https://doi.org/10.1212/WNL.0b013e31827f0fd1
Cruz-Jentoft AJ, Bahat G, Bauer J, Boirie Y, Bruyère O, Cederholm T, et al. Sarcopenia: revised European consensus on definition and diagnosis. Age Ageing. 2019;48:16–31. https://doi.org/10.1093/ageing/afy169
Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist. 1969;9:179–86.
Katz S, Downs TD, Cash HR, Grotz RC. Progress in development of the index of ADL. Gerontologist. 1970;10:20–30. https://doi.org/10.1093/geront/10.1_part_1.20
Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47:1245–51. https://doi.org/10.1016/0895-4356(94)90129-5
Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12:189–98. https://doi.org/10.1016/0022-3956(75)90026-6
Guerin O, Soto ME, Brocker P, Robert PH, Benoit M, Vellas B, et al. Nutritional status assessment during Alzheimer’s disease: results after one year (the REAL French Study Group). J Nutr Health Aging. 2005;9:81–4.
Soysal P, Dokuzlar O, Erken N, Dost Günay FS, Isik AT. The Relationship Between Dementia Subtypes and Nutritional Parameters in Older Adults. J Am Med Dir Assoc. 2020;21:1430–5. https://doi.org/10.1016/j.jamda.2020.06.051
Droogsma E, van Asselt DZB, Schölzel-Dorenbos CJM, van Steijn JHM, van Walderveen PE, van der Hooft CS. Nutritional status of community-dwelling elderly with newly diagnosed Alzheimer’s disease: prevalence of malnutrition and the relation of various factors to nutritional status. J Nutr Health Aging. 2013;17:606–10. https://doi.org/10.1007/s12603-013-0032-9
Ozer NT, Akin S, Gunes Sahin G, Sahin S. Prevalence of malnutrition diagnosed by the Global Leadership Initiative on Malnutrition and Mini Nutritional Assessment in older adult outpatients and comparison between the Global Leadership Initiative on Malnutrition and Mini Nutritional Assessment energy-protein intake: A cross-sectional study. JPEN J Parenter Enteral Nutr. 2022;46:367–77. https://doi.org/10.1002/jpen.2123
de Crom TOE, Ghanbari M, Voortman T, Ikram MA. Body composition and plasma total-tau, neurofilament light chain, and amyloid-β: A population-based study. Alzheimers Dement (Amst). 2024;16:e12519. https://doi.org/10.1002/dad2.12519
Hermesdorf M, Esselmann H, Morgado B, Jahn-Brodmann A, Herrera-Rivero M, Wiltfang J, et al. The association of body mass index and body composition with plasma amyloid beta levels. Brain Communications. 2023;5:fcad263. https://doi.org/10.1093/braincomms/fcad263
Manouchehrinia A, Piehl F, Hillert J, Kuhle J, Alfredsson L, Olsson T, et al. Confounding effect of blood volume and body mass index on blood neurofilament light chain levels. Ann Clin Transl Neurol. 2020;7:139–43. https://doi.org/10.1002/acn3.50972
Lilamand M, Bouaziz-Amar E, Dumurgier J, Cognat E, Hourregue C, Mouton-Liger F, et al. Plasma Leptin Is Associated With Amyloid CSF Biomarkers and Alzheimer’s Disease Diagnosis in Cognitively Impaired Patients. J Gerontol A Biol Sci Med Sci. 2023;78:645–52. https://doi.org/10.1093/gerona/glac234
Garcia-Garcia I, Kamal F, Donica O, Dadar M, Alzheimer’s Disease Neuroimaging Initiative. Plasma levels of adipokines and insulin are associated with markers of brain atrophy and cognitive decline in the spectrum of Alzheimer’s Disease. Prog Neuropsychopharmacol Biol Psychiatry. 2024;134:111077. https://doi.org/10.1016/j.pnpbp.2024.111077
Ishii M, Wang G, Racchumi G, Dyke JP, Iadecola C. Transgenic mice overexpressing amyloid precursor protein exhibit early metabolic deficits and a pathologically low leptin state associated with hypothalamic dysfunction in arcuate neuropeptide Y neurons. J Neurosci. 2014;34:9096–106. https://doi.org/10.1523/JNEUROSCI.0872-14.2014
Robison LS, Gannon OJ, Salinero AE, Abi-Ghanem C, Kelly RD, Riccio DA, et al. Sex differences in metabolic phenotype and hypothalamic inflammation in the 3xTg-AD mouse model of Alzheimer’s disease. Biol Sex Differ. 2023;14:51. https://doi.org/10.1186/s13293-023-00536-5
Bouaziz-Amar E, Neez J, Ibrahim F, Martinet M, Hourregue C, Dumurgier J, et al. Cerebrospinal fluid leptin in Alzheimer’s disease: relationship to plasma levels and to cerebrospinal amyloid. Clin Chem Lab Med. 2025; https://doi.org/10.1515/cclm-2025-0304
López-Gambero AJ, Rosell-Valle C, Medina-Vera D, Navarro JA, Vargas A, Rivera P, et al. A Negative Energy Balance Is Associated with Metabolic Dysfunctions in the Hypothalamus of a Humanized Preclinical Model of Alzheimer’s Disease, the 5XFAD Mouse. Int J Mol Sci. 2021;22:5365. https://doi.org/10.3390/ijms22105365

Table 1 is not available with this version.

No competing interests reported.

SupplementarymaterialsBodycomp.docx

Download PDF

Editorial decision: Revision requested
17 Nov, 2025
Reviews received at journal
17 Nov, 2025
Reviews received at journal
15 Nov, 2025
Reviews received at journal
13 Nov, 2025
Reviewers agreed at journal
05 Nov, 2025
Reviews received at journal
05 Nov, 2025
Reviews received at journal
03 Nov, 2025
Reviewers agreed at journal
03 Nov, 2025
Reviewers agreed at journal
03 Nov, 2025
Reviewers agreed at journal
31 Oct, 2025
Reviewers agreed at journal
31 Oct, 2025
Reviewers invited by journal
31 Oct, 2025
Editor assigned by journal
31 Oct, 2025
Submission checks completed at journal
31 Oct, 2025
First submitted to journal
29 Oct, 2025

You are reading this latest preprint version

Negative association of fat mass index and Alzheimer’s disease cerebrospinal fluid biomarkers in cognitively impaired patients

Status:

Version 1

Abstract

Figures

1. Background

2. Methods

2.1. Study Design and Participants

2.1.1. Inclusion and Exclusion Criteria

2.1.2. Cognitive disease diagnostic assessment

2.2. CSF Biomarkers analysis

2.3. Nutritional status determination and bioelectrical impedance measurements

2.4. Other Variables

2.5. Statistical Analyses

2.5.1. Descriptive Analyses

2.5.2. Multivariate Analyses

2.6. Ethics

3. Results

3.1. Population’s characteristics

3.2. Prevalence of malnutrition and muscle mass loss in neurocognitive disorders

3.3. Association between CSF AD biomarkers and body composition

3.4. Clinical factors associated with malnutrition

4. Discussion

5. Conclusion

Declarations

Competing interest

Funding

Author Contribution

Acknowledgement

Data Availability

References

Table 1

Additional Declarations

Supplementary Files

Status:

Version 1