Association of APOE ε4 with cerebrospinal fluid protein and amyloid-β₄₂ levels in Alzheimer's disease

Abstract

Background

Alzheimer's disease (AD) is a neurodegenerative disease, with APOE ε4 as the most common genetic risk factor. This protein has been implicated in amyloid-β peptide (Aβ)-mediated pathological processes in patients with AD. However, the mechanism by which APOE ε4 regulates cerebrospinal fluid (CSF) protein levels and Aβ₄₂ concentration in CSF during the pathogenesis of AD has not been clarified.

Objective

To determine whether APOE ε4 contributes to the pathogenesis of AD by altering Aβ₄₂ concentrations and protein levels in CSF.

Methods

A total of 228 AD patients were enrolled and assigned into the APOE non-ε4 carrier group (146 patients) and the APOE ε4 carrier group (82 patients). The concentration of Aβ₄₂ and protein levels in CSF, as well as APOE genotypes were quantified. Subsequently, the association of APOE ε4 with CSF biomarkers was explored through multiple logistic regression analysis.

Results

In APOE ε4 carriers, CSF Aβ₄₂ (p = 0.016) and the protein (p = 0.040) levels in CSF were lower compared with levels in the non-carrier group. Moreover, APOE ε4 carriage was associated with lower levels of CSF Aβ₄₂ (odds ratio 0.998, 95% confidence interval 0.995–1.000, p = 0.048) and CSF protein (odds ratio 0.004, 95% confidence interval 0.000–0.224, p = 0.007) in APOE ε4 carriers.

Conclusions

APOE ε4 modulates the CSF protein levels and CSF Aβ₄₂ concentrations in AD patients. This indicates that the APOE ε4 genotype and CSF protein levels may help monitor the progression of AD.

Introduction

Alzheimer's disease (AD) is the most common neurodegenerative disease that contributes to cognitive decline in aged patients. The disease is characterized by excessive formation of senile plaque and neurofibrillary tangles, accumulation of extracellular neurotoxic amyloid-β (Aβ) peptide plaques, and intracellular neurofibrillary tangles enriched with hyperphosphorylated tau. ¹ Studies have shown that AD is the most common cause of dementia, accounting for approximately 60–80% of all cases. ² In the brain, Aβ₄₂ is negatively correlated with Aβ₄₂ concentrations in the cerebrospinal fluid (CSF), with reduced Aβ_42/40 in CSF considered a marker of AD. The concentration of CSF (p)Tau is increased in AD patients, and this correlates with intracerebral (p)Tau pathology. In addition, researchers have demonstrated that pTau is more specifically correlated with tau tangles in patients with AD than total tau.^3,4

Compelling evidence shows that the APOE ε4 allele is the most significant genetic risk factor influencing the development of late-onset AD (LOAD). ⁵ APOE is a glycoprotein composed of 299 amino acids, with a molecular weight of 34 kilodaltons (kDa). This protein is primarily produced in hepatocytes and intrahepatic macrophages within peripheral tissues. ⁶ Evidence indicates that the APOE protein is highly expressed in astrocytes, activated glial cells, vascular endothelial cells, choroid plexus cells, and in neurons to some degree under high stress.^7,8 In humans, three alleles of APOE have been reported: APOE2 (ε2), APOE3 (ε3), and APOE4 (ε4), and these alleles carry different disease risk levels. The APOE ε4 influences the risk of AD, exhibiting a gene-dose-dependent increase in the risk of homozygotes by up to 15-fold. ⁹ Conversely, APOE ε2 reduces the risk of AD by nearly half and prolongs the patient's lifespan. ¹⁰ Notably, APOE ε4 carriers have higher concentrations of Aβ deposits in the brain.^11,12 Besides forming Aβ plaques, APOE facilitates Aβ clearance via various mechanisms, including modulating receptor-mediated uptake and protease degradation. The interaction between APOE and pathological Aβ deposition has been postulated as the primary mechanism mediating the effects of APOE on AD development.

Notably, the APOE ε4 carriers exhibit low levels of Aβ in the CSF, suggesting that it can potentially contribute to disease progression. ¹³ Data indicate that the APOE ε4 regulates Aβ in various pathological mechanisms.^14–16 However, the mechanisms linking APOE ε4, CSF protein levels and CSF Aβ₄₂ are not fully understood.

In this study, we analyzed and compared differences in CSF protein and Aβ₄₂ levels between APOE ε4 carriers and non-carriers. The aim was to determine whether these associations vary by age, gender, and cognitive function by leveraging Mini-Mental State Examination (MMSE). The findings of this study expand the current understanding into the mechanisms via which APOE ε4 contributes to AD pathology.

Methods

Participants

From January 2022 to October 2024, 228 consecutive participants were recruited from the cognitive impairment clinic of Tianjin Huanhu Hospital. The study was conducted in accordance with local clinical research regulations and informed consent was obtained all participants. In addition, the study followed the relevant national regulations and institutional policies, and followed the principles outlined in the Declaration of Helsinki. All patients completed the clinical scale assessment and agreed to undergo lumbar puncture examination. Among the participants, 168 participants fulfilled the biomarker criteria for AD. However, due to incomplete data and other related factors, 148 patients were ultimately included in the study. A detailed flowchart outlining the participant selection process is provided below.

Inclusion criteria

The inclusion criteria for patient enrollment was that all patients met the diagnostic guidelines for AD. ¹⁷ Specifically, AD patients presenting with an insidious onset, a well-documented history of cognitive decline, and according to the 2018 NIA-AA AT(N) diagnostic framework for AD biomarkers, met the criteria for AD biomarker positivity. ¹⁸

Exclusion criteria

Patients with increased risk of bleeding (such as thrombocytopenia, coagulopathy, or utilization of anticoagulant medications), a history of spinal surgery, inability to obtain sufficient quantities of CSF, and traumatic or hemorrhagic lumbar puncture, Malnutrition, neoplastic, dyslipidemia and hepatic disease, were excluded from the study. In addition, patients with mixed neurodegeneration conditions, particularly vascular dementia, dementia with Lewy bodies, frontotemporal lobar degeneration, and Parkinson's disease dementia co-occurrence, were excluded from the study.

Standardized collection of the clinical data

The eligible patients were comprehensively assessed following well-established protocols. Data were retrieved from patients’ records and using questionnaires. The data comprised variables such as years of education, current smoking and alcohol status (yes/no), hypertension, type 2 diabetes mellitus, body mass index (BMI), stroke and coronary heart disease. Various tests, including a MMSE, were applied. ¹⁹ For each patient, cerebrospinal fluid biomarkers, including cerebrospinal protein, phosphorylated tau (p-tau181), and Aβ_1–42, as well as the ratio of Aβ_1–42 to Aβ_1–40 (refer to “Cerebrospinal fluid biomarkers” section) were recorded. Blood samples from were collected from all patients for APOE genotype testing, and they were categorized based on the number of APOE ε4 (0/non-APOE ε4 carrier, 1/APOE ε4 carrier).

The study was approved by the Medical Ethics Committee of Tianjin Huanhu Hospital (ID: 2019-40).

APOE genotyping

The Maxwell^®16 Instrument (Promega) and the Maxwell^®16 Blood DNA Purification Kit were used to extract genomic DNA from whole peripheral blood. Next, the APOE rs429358 and rs7412 regions were amplified through PCR utilizing Promega's GoTaq^® hot-start polymerase. The PCR products were purified using 0.5 mL centrifugal filter from Merck Millipore's Amicon^®Ultra. Cycle sequencing was performed using the AB Prism Big Dye Terminator Sequencing Kit 3.1 (Life Technologies), following the manufacturer's guidelines. Subsequently, the obtained sequence was purified utilizing a MicroSEQ™ID sequencing cleanup kit (Life Technologies) and loaded onto a 3500 gene analyzer (Life Technologies). The sequences were analyzed using Chromas software (Technelysium Pty Ltd).

Cerebrospinal fluid biomarkers

CSF samples were collected from the patients using a standardized protocol during a routine lumbar puncture diagnosis. All patients provided informed consent to participate in the study. The samples were collected in outpatient settings from 09:30 to 10:30 h. ²⁰ The CSF samples were obtained into sterile polypropylene tubes and gently mixed to prevent gradient effects, for further measurement. The remaining CSF sample was centrifuged at 3000 rpm for 3 min and promptly aliquoted in equal volumes of 0.5 mL, and then immediately stored at −80°C or liquid nitrogen for subsequent analysis. The concentrations of p-tau181 and Aβ_1–42 in CSF were quantified using Lumipulse (Fujirebio) by a skilled technician who was blinded to the diagnosis or APOE genotype. In accordance with the accepted standards, CSF AD-like characteristics were defined based on the following predetermined cutoff values: Aβ_1–42 ≤ 650 pg/mL, p-Tau181 ≥ 58 pg/mL, or Aβ_1–42/Aβ_1–40<0.068. ²⁰

The concentrations of protein in CSF were measured using the SPAplus^® turbidity analyzer and Freelite^® serum free light chain immunoassay, in line with the manufacturer's instructions.

Statistical analysis

The data were analyzed using IBM SPSS 27.0 software. Differences were considered statistically significant at p-value < 0.05.

The distribution of the data was determined using the Shapiro-Wilk test based on a p-value >0.05.

Normality tests were performed for continuous variables, and all variables followed a normal distribution. Continuous variables were presented as the mean value plus or minus the standard deviation, while categorical variables were expressed as numbers and percentages (n, %). The association and correlation between APOE genotype and biological markers and differences in CSF proteins among various APOE genotype classified by gender, age, and cognitive status was analyzed using the independent-samples t test. Categorical variables were compared using the Fisher's exact test. In addition, categorical variables of different APOE genotype were analyzed using the chi-square test. Finally, multivariate logistic regression analysis was conducted to explore the association between various APOE carriers and CSF biomarkers.

Results

Participant characteristics

A total of 228 patients diagnosed with AD were enrolled in this study (Figure 1), with 146 patients in the APOE non-ε4 carrier group (mean ± SD, age 62.90 ± 8.64 years, male: 81 [55.5%]) and 82 in the APOE ε4 carrier group (mean ± SD, age 64.12 ± 8.08 years old, male 44 [53.7%]). Among them, 125 (54.82%) were males and 103 (45.18%) were female. In addition, 47 (25.97%) patients were smokers, 21 (9.21%) consumed alcohol, 28 (12.28%) had hypertension, 19 (8.33%) had diabetes, 15 (6.58%) had coronary disease, and 14 (6.14%) had stroke. The demographic and clinical characteristics of all patients are presented in Table 1. Notably, there were no significant differences between the APOE ε4 carrier group and non-ε4 carrier group in terms of gender, age, smoking status, drinking status, and education level. Further comparison of clinical features between the two groups found no significant differences in cognitive status (as measured by MMSE score), hypertension, diabetes, coronary heart disease, stroke or BMI.

Figure 1.

Flowchart depicting the participant recruitment process.

Table 1.

Clinical and demographic characteristics of AD patients.

Characteristics	Non-ε4 carriers n = 146	ε4 carriers n = 82	p
Gender, n (%)
male	81 (55.5)	44 (53.7)	0.185
Age, y (mean ± SD)	62.90 ± 8.64	64.12 ± 8.08	0.262
MMSE, scores (mean ± SD)	15.28 ± 7.43	14.64 ± 6.67	0.521
MoCA, scores (mean ± SD)	10.52 ± 6.13	11.21 ± 5.48	0.316
BMI (mean ± SD)	23.10 ± 2.68	22.53 ± 3.05	0.055
Education, y (mean ± SD)	9.76 ± 4.15	10.70 ± 4.00	0.370
Alcohol, n (%)	14 (9.6)	7 (8.5)	0.758
Smoking, n (%)	31 (21.2)	16 (19.5)	0.792
Hypertension, n (%)	17 (11.6)	11 (13.4)	0.984
Diabetes, n (%)	14 (9.6)	5 (6.1)	0.226
Coronary disease, n (%)	10 (6.8)	5 (6.1)	0.826
Stroke, n (%)	8 (5.5)	6 (7.3)	0.579

MMSE: Mini-Mental State Exam; T2DM: type 2 diabetes mellitus; BMI: body mass index.

The association between APOE genotype and CSF biomarkers

In the AD cohort, the APOE ε4 carrier group demonstrated significantly lower levels of CSF Aβ₄₂ and total protein, as well as significantly higher levels of P-tau181, compared to the non-ε4 carriers. We analyzed the levels of CSF biomarkers between APOE ε4 carriers and non-ε4 carriers, and the results are presented in Table 2, the mean values of CSF Aβ₄₂ (p = 0.016) and CSF Aβ_42/40 (p = 0.035) were lower in the APOE ε4 carrier group than in the non-ε4 carrier group. The mean values of p-tau181 in the CSF of the APOE ε4-carrying group exceeded that of the APOE non-ε4-carrying group (p = 0.018). Compared with the non-ε4 carrier group, the CSF protein level was lower in the APOE ε4 carrier group (p = 0.040).

Table 2.

Analysis of the association of APOE genotype with CSF biomarkers in AD.

CSF-biomarkers	Non-ε4 carriers n = 146	ε4 carriers n = 82	p
CSF-Aβ42	639.31 ± 479.11	434.02 ± 163.81	0.016*
p-tau181	85.20 ± 108.49	107.38 ± 85.94	0.018*
Aβ42/40	0.06 ± 0.03	0.05 ± 0.02	0.035*
CSF protein	0.40 ± 0.14	0.34 ± 0.10	0.040*

APOE: apolipoprotein E; CSF: cerebrospinal fluid; Aβ₄₂: amyloid beta 42; p-tau181: hyperphosphorylation tau protein. *p < 0.05.

The association of APOE genotype with the CSF biomarkers in 138 age-matched, patients with non-Alzheimer's dementia was investigated. It was observed that, compared to the non-APOE ε4 carriers, the APOE ε4 carriers tended to have lower levels of CSF Aβ₄₂ and total protein, although the differences were not statistically significant. The APOE ε4 carriers tended to have higher Aβ_42/40 ratios and P-tau181 levels in CSF, it did not reach statistical significance (Table 3).

Table 3.

Analysis of the association between APOE genotype and CSF biomarkers in patients with non-Alzheimer's dementia.

CSF-biomarkers	Non-ε4 carriers (n = 48)	ε4 carriers (n = 19)	p
CSF-Aβ42	512.8	466.88	0.572
Aβ42/40	0.08	0.06	0.065
P-tau181	46.66	56.94	0.262
CSF protein	0.390	0.385	0.578

Regression analysis of the APOE genotype and levels of biomarkers in CSF

The correlation results between the APOE genotype and various CSF biomarkers, including CSF protein, CSF Aβ₄₂, CSF Aβ_42/40 ratio, and CSF p-tau181 is presented in Table 4. Moreover, in the model comprising CSF protein, CSF Aβ₄₂, CSF Aβ_42/40 ratio, and p-tau181 (hereafter referred to as the base model), APOE ε4 was strongly correlated with lower CSF protein levels (odds ratio, 0.006, 95%CI,0–0.308, p = 0.001). Furthermore, after adjusting for potential confounders such as sex, age, smoking, alcohol consumption, hypertension, type 2 diabetes, cerebral infarction, and heart disease in the final model, the presence of the APOE ε4 was significantly associated with reduced CSF Aβ₄₂ levels (odds ratio, 0.998; 95% CI, 0.995–1.000; p = 0.048). Furthermore, the correlation of CSF protein levels was stronger than previously reported (odds ratio, 0.004; 95% CI, 0.000–0.224; p = 0.007). However, no significant differences were observed in the association between APOE ε4 carriage and CSF Aβ_42/40 or p-tau181 levels.

Table 4.

Regression analysis of the APOE genotype and levels of biomarkers in CSF.

Biological markers	No. (%) of subjects		Basic model a		Final model b
	Non-ε4 carriers (n = 146)	ε4 carriers (n = 82)	OR95%CI	p	OR95%CI	p
CSF protein	130 (89.04)	76 (92.68)	0.006 (0.000–0.308)	0.001*	0.004 (0.000–0.224)	0.007*
CSF-Aβ42	90 (61.64)	58 (70.73)	0.998 (0.996–1.000)	0.091	0.998 (0.995–1.000)	0.048*
Aβ42/40	84 (64.61)	54 (65.85)	0.000 (0–2474.883)	0.186	0.000 (0–2376982.55)	0.415
p-tau181	100 (68.49)	63 (76.83)	1.001 (0.997–1.004)	0.712	1.001 (0.998–1.005)	0.502

APOE: apolipoprotein; CSF: cerebrospinal fluid.

^a incorporated cerebrospinal fluid (CSF) protein, CSF Aβ₄₂, CSF Aβ_42/40 ratio, and phosphorylated tau181 (p-tau181).

^b Adjusted for age, sex, education level, current smoking, alcohol, stroke, hypertension, type 2 diabetes mellitus and heart disease.

*p < 0.05.

Analysis of the association between APOE ε4 carriers and levels of protein in CSF

Difference in the APOE ε4 carriers and CSF protein levels were compared among genders (Figure 2(a)), age groups (Figure 2(b)), and cognitive statuses (Figure 2(c)). It was found that in female patients, the CSF protein levels of APOE ε4 carriers were significantly lower than those of non-APOE ε4 carriers (p = 0.004). In male patients, a trend toward lower CSF protein levels was observed in APOE ε4 carriers compared to non-APOE ε4 carriers; however, this difference was not statistically significant (p = 0.142).

Figure 2.

Analysis of the association between APOE ε4 carriers and levels of protein in CSF: (a) relationship between APOE and CSF protein in different gender, (b) relationship between APOE and CSF protein in different age, (c) relationship between APOE and CSF protein in different cognition levels. *p < 0.05; **p < 0.001. MMSE: Mini-Mental State Examination.

In patients aged 65–70 years, the CSF protein levels were lower in the APOE ε4 carrier group than in the non-APOE ε4 carrier group (p = 0.016). Similarly, for patients over 70 years old, the APOE ε4 carrier group exhibited lower CSF protein levels compared to the non-ε4 carrier group (p = 0.019).

Based on the MMSE scores, among the individuals with MMSE scores greater than 21 and between 10 and 20, APOE ε4 carriers showed significantly lower levels of CSF proteins than non-ε4 carriers (p = 0.024 and p = 0.004, respectively). However, in patients with MMSE scores below 10, the CSF protein levels were not significantly different between APOE ε4 carriers and non-ε4 carriers (p = 0.366).

Discussion

Key findings

This study found that the CSF levels of Aβ₄₂, the Aβ_42/40 ratio, and CSF protein were significantly lower in AD APOE ε4 carriers than in AD non-ε4 carriers (Table 2). Notably, this signature was not significant in age-matched, patients with non-Alzheimer's dementia (Table 3). Analysis of the correlation between APOE ε4 carriers and CSF biomarkers (Table 4) revealed a strong correlation between APOE ε4 carriers and decreased CSF Aβ₄₂ and CSF protein levels.

Moreover, the APOE ε4 carriers had lower concentrations of CSF Aβ₄₂, consistent with findings of prior studies, which revealed that Aβ₄₂ levels in the CSF were lower in APOE ε4 carriers than in non-ε4 carriers.^21–23

Mechanisms underlying reduced CSF Aβ₄₂

Research has shown that APOE is highly co-deposited with Aβ in amyloid plaques. ²⁴ Furthermore, emerging amyloid pathology, which is characterized by marginally decreased CSF Aβ₄₂ or CSF Aβ₄₂ levels above the conventional threshold for amyloid positivity, may negatively affect the brain structure, function, and cognition.^25–28 Moreover, APOE ε4 carriers exhibited significantly higher and rapid amyloid deposition in the brain compared with non-APOE ε4 carriers.^16,29 Notably, the rate of Aβ clearance is modulated by its competitive binding to Aβ receptors, such as LDLR-associated protein 1 (LRP1), on astrocyte surfaces, thereby blocking Aβ uptake. ³⁰ Evidence from prior investigations show that APOE ε4 can disrupt the integrity of the blood-brain barrier (BBB) via multiple pathways: APOE ε4 binds and interacts with the LRP1 receptor on the pericyte membrane ³¹ ; as well as stimulates the loss and activation of the LRP1-dependent cyclophilin A-MMP9 pathway in pericytes and endothelial cells. ³² In conclusion, the results indicated that Aβ₄₂ concentration in the CSF was significantly lower in APOE ε4 carriers.

Mechanisms underlying reduced CSF total protein

This analysis uncovered that the CSF protein concentration was lower in APOE ε4 carriers than in non-carriers. Currently, the association of APOE ε4 with CSF protein levels is a subject of debate. Previous research has shown that APOE ε4 carriers exhibit significantly increased protein levels in the CSF and a higher CSF-to-serum albumin ratio, suggesting a strong association between the APOE ε4 allele and increased BBB permeability. ²³ However, evidence from research is inconsistent with this assertion, with some studies demonstrating that APOE ε4 may regulate CSF protein levels via other mechanisms besides compromising the integrity of the BBB. Other scholars have shown that APOE ε4 modulates the clearance and deposition of Aβ, altering other proteins in CSF.^33,34 Researchers have shown that APOE ε4 can induce mitochondrial dysfunction and neurodegenerative changes, potentially reducing CSF protein levels.^35,36 Furthermore, APOE ε4 can disrupt lipid transport and metabolism, compromising the normal physiological functions of glial cells such as astrocytes, leading to a decrease in CSF protein levels. ³⁷ A previous investigation reported that APOE ε4 altered synaptic function and plasticity, increasing the risk of neurodegenerative disorders. ¹⁵ Although advanced techniques were used for both blood APOE genotyping and cerebrospinal fluid biomarker detection, several practical limitations were encountered during the implementation. These included potential sampling errors, individual variability in sampling outcomes, and temporal constraints linked to sampling procedures. Additionally, we performed immunoassays, which may be affected by errors arising from reagent quality and instrument calibration issues, which affect the accuracy and precision of the results.

Influence of demographic and cognitive factors

We also explored the association between APOE ε4 carriers and CSF protein with regard to sex, age group, and cognitive status. Results indicated that CSF protein levels were significantly different between female APOE ε4 carriers and male non-ε4 carriers (Figure 2(a)). In prior studies, a negative correlation was observed between longitudinal changes in APOE and Aβ levels in female patients, which is consistent with our observations. ³⁸ This may be due to sex-specific associations between APOE and AD pathology. For instance, it was demonstrated that CSF sex hormone-binding globulin (SHBG) can regulate bioactive testosterone and estradiol. Moreover, higher plasma SHBG levels were detected in AD patients, inactivating functionally neuroprotective testosterone and estradiol. ³⁹ Furthermore, APOE expression level is modulated by estrogen, ⁴⁰ as well as its receptor. ⁴¹ Therefore, we hypothesize that the sex-specific association between CSF APOE and AD biomarkers may be partially influenced by female sex hormones.^38,42 Moreover, the APOE ε4 can trigger an inflammatory response in the female brain, activating microglia and astrocytes, thereby releasing inflammatory mediators, and causing neuronal damage. ⁴³

Among individuals aged 65 years and older, APOE ε4 carriers had significantly lower concentrations of CSF protein compared to non-ε4 carriers (Figure 2(b)).

Available evidence indicates that APOE ε4 exacerbates age-related lipid metabolism disorders in the brain. Notably, aging is accompanied by a decline in physiological functions of enzymes (arachidonate cytidylyltransferase 2 (AACT2), etc.) and transporters (cellular cholesterol transporter ABCA1). ⁴³ The presence of APOE ε4 further aggravates this decline, causing disruptions in intracellular and intercellular lipid transport and metabolism. This further compromises normal function of neurons and glial cells, ⁴⁴ and interferes with protein synthesis, transport, and secretion in CSF,^45–47 resulting is reduced CSF protein levels.

Previous investigations revealed that aging influenced the capacity of APOE ε4 to promote the HMGB1 (critical pro-neuroinflammatory) translocation in the brain, leading to glial cell activation and gliosis. ⁴⁸ This aggravates the inflammatory response, altering the expression and stability of proteins in the CSF. This can potentially induce neuronal damage, inhibit the synthesis and secretion of proteins, consequently reducing CSF protein levels.

Among patients with MMSE scores ranging from 10 to 20 and greater than 21, APOE ε4 carriers had significantly lower CSF protein levels compared with non-carriers (Figure 2(c)). Mechanistically, APOE ε4 contributed to cognitive decline, particularly during the early stages of the disease. Moreover, the Aβ₄₂ levels were decreased in patients with AD, and this phenomenon decreased as the condition progressed from very mild to moderate dementia. ⁴⁹ In addition, APOE ε4 carriers exhibited lower CSF proteins at this stage, suggesting that decreased CSF proteins could serve as an early indicator of AD progression.

In this study, we investigated the association of APOE ε4 with the CSF protein and CSF Aβ₄₂ levels. These associations were particularly pronounced in women, individuals aged ≥65 years, and those with mild-to-moderate cognitive impairment (MMSE 10–20 or >21). Although the exact mechanisms are not clear, the observed reductions may reflect accelerated amyloid deposition and impaired protein homeostasis in APOE ε4 carriers.^50–52 Longitudinal studies combining receptor-binding assays and proteomics are advocated to clarify the pathways linking APOE ε4 to CSF protein dynamics. These findings underscore the critical role of APOE ε4 and CSF protein levels in the pathogenesis and progression of AD.

Limitation and outlook

Nevertheless, this study has some limitations that should be acknowledged. Firstly, this was a cross-sectional study with a limited sample size, which limited our capacity to confirm causality. Moreover, practical limitations of CSF sampling as a routine disease monitoring tool in AD patients should be acknowledged.

In future, large-scale, longitudinal studies should be performed to comprehensively investigate the relationship between the APOE ε4 and CSF protein levels, as well as the associated pathological mechanisms.

Conclusion

In conclusion, this study demonstrates that AD APOE ε4 carriers have reduced CSF Aβ₄₂ and CSF proteins relative to non-APOE ε4 carriers. Moreover, our results show that the APOE ε4 and CSF protein levels may serve as biomarkers for evaluating the progression of AD.