Skip to main content

Plasma β2-microglobulin and cerebrospinal fluid biomarkers of Alzheimer’s disease pathology in cognitively intact older adults: the CABLE study



Previous studies have suggested a correlation between elevated levels of β2-microglobulin (B2M) and cognitive impairment. However, the existing evidence is insufficient to establish a conclusive relationship. This study aims to analyze the link of plasma B2M to cerebrospinal fluid (CSF) Alzheimer’s disease (AD) biomarkers and cognition.


To track the dynamics of plasma B2M in preclinical AD, 846 cognitively healthy individuals in the Chinese Alzheimer’s Biomarker and LifestylE (CABLE) cohort were divided into four groups (suspected non-AD pathology [SNAP], 2, 1, 0) according to the NIA-AA criteria. Multiple linear regression models were employed to examine the plasma B2M’s relationship with cognitive and CSF AD biomarkers. Causal mediation analysis was conducted through 10,000 bootstrapped iterations to explore the mediating effect of AD pathology on cognition.


We found that the levels of plasma B2M were increased in stages 1 (P = 0.0007) and 2 (P < 0.0001), in contrast to stage 0. In total participants, higher levels of B2M were associated with worse cognitive performance (P = 0.006 for MMSE; P = 0.012 for MoCA). Moreover, a higher level of B2M was associated with decreases in Aβ1–42 (P < 0.001) and Aβ1–42/Aβ1–40 (P = 0.015) as well as increases in T-tau/Aβ1–42 (P < 0.001) and P-tau/Aβ1–42 (P < 0.001). The subgroup analysis found B2M correlated with Aβ1–42 in non-APOE ε4 individuals (P < 0.001) but not in APOE ε4 carriers. Additionally, the link between B2M and cognition was partially mediated by Aβ pathology (percentage: 8.6 to 19.3%), whereas tau pathology did not mediate this effect.


This study demonstrated the association of plasma B2M with CSF AD biomarkers as well as a possible important role of Aβ pathology in the association between B2M and cognitive impairment, particularly in cognitively normal individuals. The results indicated that B2M could be a potential biomarker for preclinical AD and might have varied functions throughout various stages of preclinical AD progression.


Alzheimer’s disease (AD) is a prevalent form of dementia characterized by a long pre-clinical stage, multiple etiologies, and pathological progression. Despite extensive research, the complex nature of this neurodegenerative disorder has made it challenging to develop effective interventions to prevent or mitigate the pathological changes associated with AD [1, 2]. This devastating illness causes significant harm to both individuals affected by the clinical symptoms and their caretakers and imposes a significant economic burden on society [3]. As such, identifying risk factors for cognitive decline in AD is crucial for preventing the progression of this disease.

Recently, it has become widely recognized that minimally invasive blood biomarkers for screening in the preclinical stage are crucial for future Alzheimer’s disease (AD) treatment [4,5,6,7]. In order to achieve early diagnosis of AD, the search for suitable blood biomarkers should be actively pursued. Previous studies have proposed β2-microglobulin (B2M), a subunit of primary histocompatibility complex class I (MHC I) molecules, regulates behavior, synaptic plasticity, and brain development [8,9,10,11]. The results from animal studies have emphasized that the local or systemic injection of exogenous B2M in young rats leads to neurodevelopmental and hippocampus-dependent cognitive impairments and that an increase in B2M levels was observed in aged mice, which was verified by cerebrospinal fluid (CSF) or plasma from healthy populations [12, 13]. Therefore, the possibility of plasma B2M as a blood biomarker of AD deserves to be investigated. In recent studies, Dominici et al. determined that plasma B2M levels were higher in AD patients compared to the healthy control group [14]. Similar results were found in men, with B2M levels increasing with age and being higher in the dementia group compared to the healthy group [15]. Additionally, an increase in B2M was detected in the cerebrospinal fluid of HIV-related dementia patients [16, 17]. Although these results provide some evidence of a potential relationship between B2M and cognitive function and AD, the connection remains unclear. To date, there has been no systematic cohort study explaining the relationship between B2M and cognitive dysfunction and CSF AD biomarkers.

Therefore, our study used 846 cognitively normal individuals from the Chinese Alzheimer’s Biomarker and LifestylE (CABLE) study to (1) explore the relationship between B2M levels and cognition, (2) delve into the relationship between B2M levels and CSF AD biomarkers, and (3) investigate whether the effects of B2M levels on cognitive function are mediated through AD core pathology.


The CABLE study

Participants in this research who demonstrated cognitively normal were given data from the CABLE study. The project, which was started in 2017, aims to find the genetic and environmental factors that influence AD biomarkers in Han Chinese, laying the groundwork for early disease detection and illness prevention. Each patient underwent a comprehensive clinical, psychiatric-psychological, psychosocial, and psychiatric assessment and the collection of biological samples (blood and CSF samples). The CABLE database research design was authorized by the Qingdao Municipal Hospital Institutional Review Board in compliance with the Helsinki Declaration. Every subject gave their informed permission.


Participants in CABLE were Han Chinese between the ages of 40 and 90. The following conditions precluded participation: (1) central nervous system infections, multiple sclerosis, head trauma, neurodegenerative diseases other than AD (e.g., Parkinson’s, epilepsy), or other major neurological disorders; (2) significant psychiatric disorders; (3) severe systemic disease (e.g., malignancy); and (4) a history of genetic disorders in the family. The Mini-Mental State Examination (MMSE) and the Montreal Cognitive Assessment (MoCA) were used to conduct an initial cognitive screening of the subjects. Neuropsychological tests together with thorough results from CSF biomarkers and magnetic resonance imaging (MRI) were used by medical professionals with standardized training to examine all diagnoses of cognitive impairment.

There were 1844 CABLE individuals with normal cognitive functions whose covariate data were available (age, sex, years of education, and APOE genotypes). Due to a lack of B2M data, 625 people were initially disqualified, and 373 participants without CSF biomarkers and those whose data fell outside of the standard deviation (SD) of four times were also disqualified. Eventually, this cross-sectional study comprised 846 individuals.

Plasma B2M determination

B2M was determined by a latex-enhanced immunoturbidimetric assay using an automated biochemical analyzer (Beckman Coulter Automated Biochemistry Analyzer AU5800, USA) following a 12-h overnight fast, using customary regular clinical laboratory techniques at the clinical chemistry department’s lab at Qingdao Municipal Hospital in China.

CSF AD biomarker assessments

After an overnight fast, CSF specimens were collected from participants by lumbar puncture. After 2 h of collection, these samples were centrifuged at 2000 g for 10 min to remove cells and other insoluble materials, and they were then immediately frozen at − 80 °C until testing. The thaw/freeze cycle was not allowed to go longer than twice. Within 6 months, the time from sample collection to measurement was under control. CSF amyloid β plaques 42 (Aβ1–42), phosphorylated tau protein (P-tau), and total tau protein (T-tau) levels were measured by the ELISA kits (Innotest β-AMYLOID (1-42) [catalog number: 81583]; β-AMYLOID (1-40) [catalog number: 81585]; PHOSPHO-TAU (181p) [catalog number: 81581]; hTAU-Ag [catalog number: 81579]; Fujirebio, Ghent, Belgium). All measures were carried out by trained laboratory staff unaware of the clinical data. Each panel also contains blank controls, run validation controls, and internal control samples in addition to the cerebrospinal fluid samples. The plate will be reanalyzed if a quality control sample is insufficient. The mean of the duplicate analyses of standards and cerebrospinal fluid samples was utilized for further statistical analyses. The intra-batch coefficient of variation (CV) was less than 5%; the mean CVs for Aβ1–42, P-tau, and T-tau were 4.4%, 3.5%, and 4.7%, respectively. The mean inter-batch CVs for Aβ1–42, P-tau, and T-tau were less than 20%, 13.4%, and 14.2%, respectively. The levels of these CSF biomarkers were not related to the sampling collection time, storage period, and intra- and inter-batch CV, according to quality control assessments.

Evidence of AT group

The 2011 National Institute on Aging-Association Alzheimer’s (NIA-AA) committee reached the conclusion that the diagnostic criteria for preclinical AD include normal cognition but aberrant AD biomarkers [18]. Around one-third of older persons with cognitively normal brains exhibit AD pathology, according to previous amyloid imaging [11, 19, 20] and neuropathology [21, 22] investigations. Studies of Asian people have shown a similar range [23,24,25]. The three biomarkers in the procedure are neurodegeneration (T-tau), aggregated AD-tau (P-tau), and aggregated Aβ (Aβ1–42). Also, each biomarker is classified as either normal or pathological using binarization. Therefore, the cutoff values for abnormal CSF AD biomarkers were defined as Aβ1–42 < 205.82 pg/ml (lower third), P-tau > 48.73 pg/ml (upper third), and T-tau > 221.84 pg/ml (upper third) for T-tau. In order to limit the number of groups, the aggregated tau and neurodegeneration groups were merged, resulting in four distinct combinations of biomarker groups, including stage 0, stage 1, stage 2, and suspected non-AD pathology (SNAP) [26, 27]. Stage 0 individuals had normal Aβ1–42, P-tau, and T-tau readings. Stage 1 individuals had an aberrant Aβ1–42 but normal P-tau or T-tau. Stage 2 individuals had an abnormal Aβ1–42 and clear P-tau or T-tau abnormalities. Those with neurodegeneration or tau aggregation but normal amyloid levels were classified as SNAP.

APOE ε4 genotyping assessment

The QIAamp® DNA Blood Mini Kit was used to extract DNA from blood samples after a fast. The DNA was then divided and stored at − 80 °C in an enzyme-free EP tube until the study’s APOE genotyping was completed. The restriction fragment length polymorphism technique was used for genotyping two particular loci associated with APOE ε4 status (rs7412 and rs429358). Participants were split into two groups based on their APOE ε4 status (whether or not they had at least one copy of the APOE ε4 gene).

Statistical analysis

We utilized the Box-Cox transformation [28] to convert B2M into a normal distribution since it did not suit the normal distribution (Shapiro–Wilk test in R, P < 0.001). Additional file 1: Fig. S1 illustrates the Q-Q plot. We carried out the analysis after eliminating the outliers, which were identified as values that were 4 SD above or below the averages, in order to reduce the influence of extreme values.

In order to demonstrate the baseline characteristics of the recruited individuals, we computed and presented the number and percentage for the categorical variables, as well as the mean and standard deviation (SD) for the continuous variables. First, the difference between the AT groups of stages 0, 1, and 2 and SNAP was investigated using the chi-square analysis for categorical data, as well as the one-way ANOVA analysis and non-parametric testing for continuous variables [29, 30]. Then, intergroup differences in age (categorized as 65 years old, mid-life; 65 years old, late-life), sex (categorized as female and male), and APOE ε4 status (categorized as 0, 1, and 2) were studied using the GraphPad Prism software, where Student’s t-test was used to test for comparisons between the two groups. Meanwhile, for multiple comparisons of three or more groups, one-way ANOVA and Tukey Honest significant post hoc analyses of variance were used on an age-adjusted basis. After adjusting for age, sex, educational level, and APOE ε4 carrier status, a linear regression model was used to investigate the connection between B2M and CSF AD biomarkers and cognition (adjustment for MMSE was incorporated into the relationship between B2M and CSF AD biomarkers).

To evaluate whether AD pathology [31] mediated the relationship of B2M with cognition, we fitted linear regression models according to Baron and Kenny’s methodologies [32]. Regarding the three equations, (1) the CSF AD biomarkers (the mediator) were regressed on the B2M (independent variable or IV), (2) the cognitive score (dependent variable or DV) was regressed on the IV, and (3) the DV was regressed on the mediator and IV variables. The four-part criteria below must be simultaneously met to establish the mediation effects: (a) diminished relationship between B2M and cognition when incorporating the mediator (CSF AD biomarkers) into the regression model, (b) the significant correlation of CSF AD biomarkers to cognitive measures, (c) the significant correlation of B2M to cognitive measures, and (d) the significant correlation of B2M to CSF AD biomarkers. Subsequently, we estimated the indirect effects or attenuation, determining the significance by utilizing 10,000 bootstrapped iterations. For this model, each path controlled for APOE ε4 status, education, sex, and age. Interaction analysis was applied to assess the influences of age, sex, year of education, cardiovascular factors (CVF), subjective cognitive decline (SCD), and APOE ε4 status on the above relationships.

The traditional and significant two-sided P-value criterion was reported as 0.05. All the aforementioned statistical techniques and the creation of the diagrams were performed using the R Studio software (version 4.2.1), GraphPad Prism (version 9.4.2), Stata (version 15.1), and SPSS (version


Essential characteristics and intergroup comparisons

A total of 846 participants from the CABLE study were assessed, specifically for stages 0 (846), 1 (189), and 2 (92) and SNAP (250). Table 1 details the clinical characteristics and demographic features. Regarding the findings reported in the literature, the outcomes in the hypothetical group, specifically the resulting proportions of individuals, were found to be comparable [27]. With regard to the study cohort, 42% were female, aged between 40 and 90 years (mean SD = 63.1 ± 10.4), who had 9.2 ± 4.2 years of schooling and a 15.5% APOE ε4-positive rate. Moreover, the research population had an MMSE score of 27.7 ± 2.3 and a MoCA score of 22.7 ± 2.3 in terms of cognitive function. In reference to sex and APOE ε4 status, no differences were found between the two groups. Stage 2, however, had older participants and featured worse cognitive composite scores than stage 0.

Table 1 Characteristics of participants by biomarker framework

Aging is a significant risk factor for AD; thus, we wondered whether B2M levels reflect typical aging. B2M levels were found to rise with age, as shown by a statistically significant positive association (β = 0.004, P < 0.001). (Additional file 1: Fig. S2). The findings confirmed that B2M varied substantially among the different age categories (elderly: 2.04 ± 0.77 mg/l, n = 424; P = 0.005; middle age: 1.74 ± 0.69 mg/l, n = 422). (Additional file 1: Fig. S3, A). The B2M levels were not affected by sex (Additional file 1: Fig. S3, B) or APOE carrier status (all P > 0.05) (Additional file 1: Fig. S3, C).

Differences in plasma B2M levels in relation to the various biomarker classifications

The variations in B2M levels between the four stages were compared using a biomarker classification framework, particularly for stage 0 (n = 315), stage 1 (n = 189), stage 2 (n = 92), and SNAP (n = 250), to identify the relationship between B2M changes and downstream processes of Aβ deposition and amyloid cascade (i.e., tau pathology and neurodegeneration).

Significant differences in B2M levels were noted between the four groups according to the one-way ANCOVA results. Equally important, compared with the others, B2M was found to be significantly at its lowest level in the stage 0 group (Fig. 1). The B2M concentrations were significantly higher in stage 1 (P = 0.0007), stage 2 (P < 0.0001), and SNAP (P = 0.0011) groups than in stage 0 (Fig. 1). An ANCOVA model adjusting age was used since the study’s findings specified that age is a crucial factor impacting plasma B2M, and we acquired the same outcomes as before. Compared with the other groups, B2M was considerably lower in stage 0 (P < 0.001).

Fig. 1
figure 1

Plasma B2M in the biomarker classification. The cutoff values to define abnormal CSF AD biomarkers were < 205.82 pg/ml for Aβ1–42 (A +), > 48.73 pg/ml for P-tau (T +), or > 221.84 pg/ml for T-tau (T +). One-way ANOVA followed by Bonferroni post hoc analyses with adjustment for age was applied to examine the differences in plasma B2M across biomarker profiles in a biomarker framework. Abbreviations: B2M, β2-microglobulin; SNAP, suspected non-AD pathology

Differences in B2M levels between multiple groups were noted, with an increase in SNAPs compared with stage 0 (P = 0.0011). Then, B2M levels in SNAPs were found to have significantly decreased as opposed to stage 2 (P = 0.0025). These outcomes suggest elevated B2M levels are a superimposed response to amyloid and tau pathology or neuronal damage (as measured by CSF P-tau or T-tau levels).

Relationship between plasma B2M with cognition and CSF AD biomarkers

We used a linear regression model to examine the relationship between plasma B2M and cognition, adjusting for age, sex, education level, and APOE ε4 carrier status. Increased plasma B2M was linked to lower levels of MMSE (β =  − 0.094, P = 0.006) and MoCA (β =  − 0.078, P = 0.012) in the entire sample of patients (Fig. 2). Next, the link between B2M and the CSF core biomarkers of AD was evaluated using a linear regression model, adjusting for age, sex, education level, APOE ε4 status, and the MMSE score based on the dynamics of plasma B2M. Increased plasma B2M was correlated with lower levels of Aβ1–42 in the total sample of participants (n = 846) (β =  − 0.158, P < 0.001), whereas T-tau (β = 0.064, P = 0.101) and P-tau (β = 0.018, P = 0.999) were not associated (Fig. 3). Afterward, we surveyed the ratios of CSF amyloid and tau biomarkers and discovered that an increase in plasma B2M was correlated with both (a) a drop in CSF Aβ1–42/Aβ1–40 (β =  − 0.102, P = 0.015) and (b) an increase in T-tau/Aβ1–42 (β = 0.194, P < 0.001) and P-tau/Aβ1–42 (β = 0.154, P < 0.001) (Fig. 3). Upon executing additional correction factor adjustments, the sensitivity analysis exhibited that the aforementioned results remained consistent (Additional file 1: Table S1). These findings imply an early B2M response to the initial signs of neurodegeneration, with increased plasma B2M levels related to Aβ deposition independent of tau pathology and neurodegeneration.

Fig. 2
figure 2

Associations of B2M and cognitive performance. Multiple linear regression models were used to examine the associations between the plasma B2M with MMSE (A) and MoCA (B), adjusting for age, sex, education level, and APOE ɛ4 carrier statuses. Abbreviations: B2M, β2-microglobulin; MMSE, Mini-Mental State Examination; MoCA, Montreal Cognitive Assessment; APOE, apolipoprotein E

Fig. 3
figure 3

Associations between B2M and CSF AD biomarkers. Multiple linear regression models were used to examine the associations between the plasma B2M with cerebrospinal fluid (CSF) Aβ1–42 (A), T-tau (B), P-tau (C), Aβ1–42/Aβ1–40 (D) T-tau/Aβ1–42 (E), and P-tau/Aβ1–42 (F), adjusting for age, sex, education level, APOE ɛ4 carrier statuses, and MMSE. Abbreviations: B2M, β2-microglobulin; APOE, apolipoprotein E

Interactions and stratified analyses by APOE ε4 status, age, sex, CVF, and SCD

The interaction analysis expressed that the association between B2M and CSF AD biomarkers was not affected by sex, age, years of education, CVF, SCD, and APOE ε4 status (Additional file 1: Table S2). We performed additional stratified analyses and discovered that B2M was substantially connected with Aβ1–42 (β =  − 0.159, P = 0.005), Aβ1–42/Aβ1–40 (β =  − 0.129, P = 0.027), P-tau/Aβ1–42 (β = 0.181, P = 0.001), and T-tau/Aβ1–42 (β = 0.226, P < 0.001) in middle age. Meanwhile, only Aβ1–42 (β =  − 0.124, P = 0.03) was correlated with B2M in old age (Fig. 4). Then, B2M was strongly correlated with Aβ1–42 (β =  − 0.153, P < 0.001), Aβ1–42/Aβ1–40 (β =  − 0.101, P = 0.018), P-tau/ Aβ1–42 (β = 0.155, P < 0.001), and T-tau/ Aβ1–42 (β = 0.199, P < 0.001) biomarkers in non-APOE ε4 status (Fig. 4). However, B2M in APOE ε4 status did not correlate with the biomarkers. The results did not differ in the sex and CVF subgroups (Fig. 4).

Fig. 4
figure 4

Heatmap for subgroup analyses of the association between B2M and CSF AD biomarkers. Multiple linear regression models were employed with adjustment for age, sex, years of education, APOE ε4 status, and MMSE. In stratified analyses, stronger correlations were observed between B2M and biomarkers in middle-aged individuals and individuals without the APOE ε4 allele. Abbreviations: CVF, cardiovascular factors; APOE, apolipoprotein E. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001

Causal mediation analyses

The direct, indirect, and total effects of Aβ1–42 on cognition were all statistically significant (P < 0.05), highlighting that the connection of B2M with MMSE and MoCA was solely mediated by Aβ1–42, with a mediating ratio of 18.1% and 9.2%, respectively (Fig. 5). The Aβ1–42/Aβ1–40 (mediating ratio = 12.3%) and P-tau/Aβ1–42 (mediating ratio = 19.3%) mediated the relationship of B2M with MMSE, while P-tau/Aβ1–42 (mediating ratio = 8.6%) and T-tau/Aβ1–42 (mediating ratio = 16.0%) mediated the correlation of B2M with MoCA (Fig. 5).

Fig. 5
figure 5

Mediation analyses of AD biomarkers between B2M and cognition. Mediation analyses with 10,000 bootstrapped iterations were used to examine the mediation effects of Aβ pathologies on cognition. Aβ1–42, Aβ1–42/Aβ1–40, and P-tau/Aβ1–42 mediated the relationship between B2M and MMSE (AC). Aβ1–42, P-tau/Aβ1–42, and T-tau/Aβ1–42 mediated the relationship between B2M and MoCA (DF). Each model path was adjusted for age, sex, years of education, APOE ε4 status, and MMSE. a is the effect of the independent variable on mediators; b is the effect of mediators on dependent variables after controlling the influence of independent variables; c is the total effect of independent variables on dependent variables; c′ is the direct effect. Abbreviations: B2M, β2-microglobulin; P-tau, phosphorylated tau protein; T-tau, total tau protein; Aβ, amyloid β; APOE, apolipoprotein E


In this study, we aimed to assess the alterations in plasma B2M levels during the early stages of AD by incorporating a biological marker-based classification scheme. Using this classification system, we observed dynamic changes in B2M levels that progressively increased with pathological evolution. Furthermore, our study extended previous findings by exploring the relationship between B2M and the cerebrospinal fluid (CSF) AD biomarkers and cognition, which was not previously reported in the literature. Our results showed that although neurodegeneration or tau pathology was not related to B2M, the increase in B2M was related to the pathological changes of Aβ (low CSF Aβ1–42). Given the association between cognitive decline and elevated B2M levels, as well as the role of Aβ pathology in mediating the impact of B2M on cognition, the potential role of B2M in the pathogenesis of AD appears to be substantiated.

To evaluate the alterations in factor B2M related to cognitive functions and aging [13], we combined clinical analysis and biomarker-based categorization in this study. More particularly, using this categorization method enabled us to inspect the connection between pro-aging variables and the pathophysiology of AD, including Aβ1–42, tau pathology, and neurodegeneration. Our findings contribute to the existing literature by demonstrating that B2M levels dynamically change in response to certain pathogenic events. B2M was shown to be correlated with both an increase in Aβ pathology (defined as low CSF Aβ1–42) and a decline in cognitive scores (MMSE and MoCA). B2M, however, had no connection to tau pathology or neurodegeneration. Having said that, our results seem to support the theory that B2M is a pro-aging factor [33]. The accumulation of B2M in the blood promotes age-related cognitive dysfunction [13], signifying that B2M may be a therapeutic target for AD.

Our study details that even in the absence of tau pathology, higher B2M concentrations are associated with abnormal CSF Aβ1–42 levels, as indicated by the difference in B2M between stages 0 and 1. This implies that an increased B2M may induce the deposition of Aβ1–42, thereby leading to cognitive decline, as confirmed by our mediation analysis. In the absence of tau pathology, increased B2M appears to be a cause rather than a consequence of Aβ deposition, which is aligned with the findings of previous reports [8, 9, 34,35,36,37,38]. Additionally, considerable attention must be given to the most heterogeneous category, the SNAP group, given its potential to present symptoms that are not related to AD or neurodegeneration. In light of this, further research on this topic must extensively investigate the variations in B2M in neurodegenerative diseases apart from AD.

Exploring the relationship between plasma B2M and CSF AD biomarkers may generate a better overview of the pathogenicity by revealing concurrent processes in the brain. In the normal cognitive population, plasma B2M levels were negatively correlated with the CSF AD biomarkers Aβ1–42, but the correlation with T-tau and P-tau disappeared. These findings propose that alterations in plasma B2M may be associated with amyloid deposition and put forward that a facilitative response to amyloid deposition could be due to the increase in B2M during aging. Through subgroup analysis of individuals with or without APOE ε4 status, B2M was found to be significantly associated with CSF AD biomarkers in APOE ε4 carriers but not in non-carriers, suggesting a potential role for B2M in predicting future AD-related risks, particularly in individuals with non-APOE ε4 status. Considering that B2M is a renal function indicator, previous studies have found that the APOE ε4 carrier status does not affect the decline in renal function as a marker for individual dementia risk [39]. However, previous animal experiments have found that knocking out APOE ε4 in mice leads to a decrease in peripheral clearance of Aβ1–42 [40], which may to some extent affect the relationship between B2M and Aβ142. Nevertheless, in our study, the number of APOE ε4 carriers was limited to 110 (13.00%), which could lead to false negatives. Direct clinical evidence on the relationship between B2M and APOE ε4 carrier status is still lacking, and further studies with adequate sample sizes and sufficiently long follow-up periods are required.

The mechanisms underlying the increase in plasma B2M early in Aβ pathology were unclear. One explanation is that B2M fibrils can release oligomers early on and rapidly spread to the cell membranes of the neighboring cells, triggering continuous events that result in sustained neuronal and glial cell damage [41]. Another justification is that B2M acts as a pro-senescence factor [33], which may partially activate amyloidosis, hence the high levels of B2M in stage 1. Presenting intracellular antigens to CD8 + T cells (cytotoxic T cells) necessitates all cells (excluding erythrocytes) to use B2M, which is part of the MHC I complex. The expression of MHCI-like molecules increases after aging, and the high expression of MHCI-like molecules impairs neuronal plasticity, neurite growth, and neurite regeneration [36, 42,43,44,45,46,47,48]. The results underline that impaired neurogenesis and cognitive dysfunction, which are age-related, are stimulated by B2M’s systemic accumulation in aging blood. On top of that, compared with the young mice, old mice experienced elevated B2M levels in their hippocampus and plasma [13], as substantiated by our subgroup analysis on age. Known as a likely pro-aging factor [33], B2M has been discovered to contribute to age-related cognitive impairment and aging progression (a previously unrecognized function) [13]. Our study reveals that B2M may be a risk factor for the early progression of AD, which is consistent with the existing literature [14, 15]. Nevertheless, despite the trend of elevated B2M in stage 2, we noted that B2M did not correlate precisely with tau pathology and neurodegeneration, which could be due to tau pathology being driven by the Aβ deposition, and the highest concentration of B2M in stage 2 was a consequence of the action of the Aβ deposition.

Numerous factors enhance the reliability of our study. First, this investigation examined the relationship between B2M and the CSF AD biomarkers and is the largest to date. Second, we utilized the AD diagnostic criteria from the NIA-AA study to classify AD biomarkers, ensuring the high quality of the study. Additionally, the CSF data were subjected to blinded quality control, and the results allowed statistical control for various possible confounders. However, the current study has several limitations that should be considered. Firstly, this was a cross-sectional study, which means that it cannot establish causality. Secondly, the results should be verified through larger longitudinal studies. Third, considering the fact that brain imaging data (e.g., PiB-PET or Tau-PET) may better reflect the association between AD pathology and B2M, future prospective neuroimaging cohorts would facilitate validating or supplementing evidence about the role of factor B2M. Lastly, the correlation between plasma B2M and cerebrospinal fluid AD-related biomarkers in the preclinical phase of AD remains to be further explored. Despite these limitations, the protective association found in this study provides valuable insights for future research into the potential causes and mechanisms of AD. Further examination of the relationship between plasma B2M and AD biomarkers may help to clarify the role of B2M in AD.


Overall, the study identified an association between plasma B2M and CSF AD biomarkers and a possible important role of Aβ pathology in the association between B2M and cognitive impairment. This information is crucial to reveal the pathological mechanisms of B2M involvement in dementia or AD. The connection between B2M and AD pathogenesis will be further strengthened by additional research to determine its underlying processes, which might lead to the identification of novel pathogenetic pathways and treatment targets.

Availability of data and materials

The datasets used and analyzed in the current study are available from the corresponding authors upon reasonable request.





Cerebrospinal fluid


Alzheimer’s disease


Chinese Alzheimer’s Biomarker and LifestylE study

APOE ε4 :

Apolipoprotein E genotype ε4


Amyloid β


Phosphorylated tau


Total tau


Mini-Mental State Examination


Montreal Cognitive Assessment


Cardiovascular factors


  1. Scheltens P, Blennow K, Breteler MM, de Strooper B, Frisoni GB, Salloway S, et al. Alzheimer’s disease. Lancet. 2016;388(10043):505–17.

    Article  CAS  PubMed  Google Scholar 

  2. Long JM, Holtzman DM. Alzheimer disease: an update on pathobiology and treatment strategies. Cell. 2019;179(2):312–39.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  3. Reiman EM, Langbaum JB, Tariot PN, Lopera F, Bateman RJ, Morris JC, et al. CAP–advancing the evaluation of preclinical Alzheimer disease treatments. Nat Rev Neurol. 2016;12(1):56–61.

    Article  CAS  PubMed  Google Scholar 

  4. Hu H, Tan L, Bi Y-L, Xu W, Tan L, Shen X-N, et al. Association of serum Apolipoprotein B with cerebrospinal fluid biomarkers of Alzheimer’s pathology. Ann Clin Transl Neurol. 2020;7(10):1766–78.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  5. Li W-W, Shen Y-Y, Tian D-Y, Bu X-L, Zeng F, Liu Y-H, et al. Brain amyloid-β deposition and blood biomarkers in patients with clinically diagnosed Alzheimer’s disease. J Alzheimer’s Dis. 2019;69(1):169–78.

    Article  CAS  Google Scholar 

  6. Nakamura A, Kaneko N, Villemagne VL, Kato T, Doecke J, Doré V, et al. High performance plasma amyloid-β biomarkers for Alzheimer’s disease. Nature. 2018;554(7691):249–54.

    Article  CAS  PubMed  Google Scholar 

  7. Ridler C. Alzheimer disease: blood amyloid-β successfully signals AD. Nat Rev Neurol. 2018;14(4):195.

    Article  PubMed  Google Scholar 

  8. Boulanger LM, Shatz CJ. Immune signalling in neural development, synaptic plasticity and disease. Nat Rev Neurosci. 2004;5(7):521–31.

    Article  CAS  PubMed  Google Scholar 

  9. Huh GS, Boulanger LM, Du H, Riquelme PA, Brotz TM, Shatz CJ. Functional requirement for class I MHC in CNS development and plasticity. Science. 2000;290(5499):2155–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  10. Li Y, Chopp M, Chen J, Wang L, Gautam SC, Xu YX, et al. Intrastriatal transplantation of bone marrow nonhematopoietic cells improves functional recovery after stroke in adult mice. J Cereb Blood Flow Metab. 2000;20(9):1311–9.

    Article  CAS  PubMed  Google Scholar 

  11. Morris JC, Roe CM, Xiong C, Fagan AM, Goate AM, Holtzman DM, et al. APOE predicts amyloid-beta but not tau Alzheimer pathology in cognitively normal aging. Ann Neurol. 2010;67(1):122–31.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  12. Carrette O, Demalte I, Scherl A, Yalkinoglu O, Corthals G, Burkhard P, et al. A panel of cerebrospinal fluid potential biomarkers for the diagnosis of Alzheimer’s disease. Proteomics. 2003;3(8):1486–94.

    Article  CAS  PubMed  Google Scholar 

  13. Smith LK, He Y, Park JS, Bieri G, Snethlage CE, Lin K, et al. β2-microglobulin is a systemic pro-aging factor that impairs cognitive function and neurogenesis. Nat Med. 2015;21(8):932–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Dominici R, Finazzi D, Polito L, Oldoni E, Bugari G, Montanelli A, et al. Comparison of β2-microglobulin serum level between Alzheimer’s patients, cognitive healthy and mild cognitive impaired individuals. Biomarkers. 2018;23(6):603–8.

    Article  CAS  PubMed  Google Scholar 

  15. Yang R, Fu S, Zhao L, Zhen B, Ye L, Niu X, et al. Quantitation of circulating GDF-11 and β2-MG in aged patients with age-related impairment in cognitive function. Clin Sci (Lond). 2017;131(15):1895–904.

    Article  CAS  PubMed  Google Scholar 

  16. Brew BJ, Dunbar N, Pemberton L, Kaldor J. Predictive markers of AIDS dementia complex: CD4 cell count and cerebrospinal fluid concentrations of beta 2-microglobulin and neopterin. J Infect Dis. 1996;174(2):294–8.

    Article  CAS  PubMed  Google Scholar 

  17. McArthur JC, Nance-Sproson TE, Griffin DE, Hoover D, Selnes OA, Miller EN, et al. The diagnostic utility of elevation in cerebrospinal fluid beta 2-microglobulin in HIV-1 dementia. Multicenter AIDS Cohort Study. Neurology. 1992;42(9):1707–12.

    Article  CAS  PubMed  Google Scholar 

  18. Sperling RA, Aisen PS, Beckett LA, Bennett DA, Craft S, Fagan AM, et al. Toward defining the preclinical stages of Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement. 2011;7(3):280–92.

    Article  PubMed  PubMed Central  Google Scholar 

  19. Reiman EM, Chen K, Liu X, Bandy D, Yu M, Lee W, et al. Fibrillar amyloid-beta burden in cognitively normal people at 3 levels of genetic risk for Alzheimer’s disease. Proc Natl Acad Sci U S A. 2009;106(16):6820–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  20. Rowe CC, Ellis KA, Rimajova M, Bourgeat P, Pike KE, Jones G, et al. Amyloid imaging results from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging. Neurobiol Aging. 2010;31(8):1275–83.

    Article  PubMed  Google Scholar 

  21. Bennett DA, Schneider JA, Arvanitakis Z, Kelly JF, Aggarwal NT, Shah RC, et al. Neuropathology of older persons without cognitive impairment from two community-based studies. Neurology. 2006;66(12):1837–44.

    Article  CAS  PubMed  Google Scholar 

  22. Hulette CM, Welsh-Bohmer KA, Murray MG, Saunders AM, Mash DC, McIntyre LM. Neuropathological and neuropsychological changes in “normal” aging: evidence for preclinical Alzheimer disease in cognitively normal individuals. J Neuropathol Exp Neurol. 1998;57(12):1168–74.

    Article  CAS  PubMed  Google Scholar 

  23. Hong YJ, Park KW, Kang DY, Lee JH. Prediction of Alzheimer’s pathological changes in subjective cognitive decline using the self-report questionnaire and neuroimaging biomarkers. Dement Neurocogn Disord. 2019;18(1):19–29.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Hwang J, Jeong JH, Yoon SJ, Park KW, Kim EJ, Yoon B, et al. Clinical and biomarker characteristics according to clinical spectrum of Alzheimer’s disease (AD) in the validation cohort of Korean brain aging study for the early diagnosis and prediction of AD. J Clin Med. 2019;8(3):341.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Jang H, Kim HJ, Park S, Park YH, Choe Y, Cho H, et al. Application of an amyloid and tau classification system in subcortical vascular cognitive impairment patients. Eur J Nucl Med Mol Imaging. 2020;47(2):292–303.

    Article  CAS  PubMed  Google Scholar 

  26. Ma LZ, Tan L, Bi YL, Shen XN, Xu W, Ma YH, et al. Dynamic changes of CSF sTREM2 in preclinical Alzheimer’s disease: the CABLE study. Mol Neurodegener. 2020;15(1):25.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  27. Jack CR, Knopman DS, Weigand SD, Wiste HJ, Vemuri P, Lowe V, et al. An operational approach to National Institute on Aging-Alzheimer’s Association criteria for preclinical Alzheimer disease. Ann Neurol. 2012;71(6):765–75.

    Article  PubMed  PubMed Central  Google Scholar 

  28. Chen DG, Pounds JG. A nonlinear isobologram model with Box-Cox transformation to both sides for chemical mixtures. Environ Health Perspect. 1998;106 Suppl 6(Suppl 6):1367–71.

    Article  CAS  PubMed  Google Scholar 

  29. Bewick V, Cheek L, Ball J. Statistics review 9: one-way analysis of variance. Crit Care. 2004;8(2):130–6.

    Article  PubMed  PubMed Central  Google Scholar 

  30. Brace JC, Savalei V. Type I error rates and power of several versions of scaled chi-square difference tests in investigations of measurement invariance. Psychol Methods. 2017;22(3):467–85.

    Article  PubMed  Google Scholar 

  31. Dregan A, Rayner L, Davis KAS, Bakolis I, Arias de la Torre J, Das-Munshi J, et al. Associations between depression, arterial stiffness, and metabolic syndrome among adults in the UK Biobank population study: a mediation analysis. JAMA Psychiatry. 2020;77(6):598–606.

    Article  PubMed  Google Scholar 

  32. Baron RM, Kenny DA. The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. J Pers Soc Psychol. 1986;51(6):1173–82.

    Article  CAS  PubMed  Google Scholar 

  33. Villeda SA, Luo J, Mosher KI, Zou B, Britschgi M, Bieri G, et al. The ageing systemic milieu negatively regulates neurogenesis and cognitive function. Nature. 2011;477(7362):90–4.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  34. Glynn MW, Elmer BM, Garay PA, Liu X-B, Needleman LA, El-Sabeawy F, et al. MHCI negatively regulates synapse density during the establishment of cortical connections. Nat Neurosci. 2011;14(4):442–51.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  35. Goddard CA, Butts DA, Shatz CJ. Regulation of CNS synapses by neuronal MHC class I. Proc Natl Acad Sci USA. 2007;104(16):6828–33.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Lee H, Brott BK, Kirkby LA, Adelson JD, Cheng S, Feller MB, et al. Synapse elimination and learning rules co-regulated by MHC class I H2-Db. Nature. 2014;509(7499):195–200.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Loconto J, Papes F, Chang E, Stowers L, Jones EP, Takada T, et al. Functional expression of murine V2R pheromone receptors involves selective association with the M10 and M1 families of MHC class Ib molecules. Cell. 2003;112(5):607–18.

    Article  CAS  PubMed  Google Scholar 

  38. Shatz CJ. MHC class I: an unexpected role in neuronal plasticity. Neuron. 2009;64(1):40–5.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Wang M, Ding D, Zhao Q, Wu W, Xiao Z, Liang X, et al. Kidney function and dementia risk in community-dwelling older adults: the Shanghai Aging Study. Alzheimers Res Ther. 2021;13(1):21.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Sharman MJ, Morici M, Hone E, Berger T, Taddei K, Martins IJ, et al. APOE genotype results in differential effects on the peripheral clearance of amyloid-beta42 in APOE knock-in and knock-out mice. J Alzheimers Dis. 2010;21(2):403–9.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  41. Bigi A, Cascella R, Chiti F, Cecchi C. Amyloid fibrils act as a reservoir of soluble oligomers, the main culprits in protein deposition diseases. Bioessays. 2022;44(11):11.

    Article  Google Scholar 

  42. Adelson JD, Barreto GE, Xu L, Kim T, Brott BK, Ouyang Y-B, et al. Neuroprotection from stroke in the absence of MHCI or PirB. Neuron. 2012;73(6):1100–7.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Bombeiro AL, Thomé R, Oliveira Nunes SL, Monteiro Moreira B, Verinaud L, Oliveira ALRd. MHC-I and PirB upregulation in the central and peripheral nervous system following sciatic nerve injury. PloS One. 2016;11(8):0161463.

    Article  Google Scholar 

  44. Corriveau RA, Huh GS, Shatz CJ. Regulation of class I MHC gene expression in the developing and mature CNS by neural activity. Neuron. 1998;21(3):505–20.

    Article  CAS  PubMed  Google Scholar 

  45. Datwani A, McConnell MJ, Kanold PO, Micheva KD, Busse B, Shamloo M, et al. Classical MHCI molecules regulate retinogeniculate refinement and limit ocular dominance plasticity. Neuron. 2009;64(4):463–70.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  46. Götz J, Streffer JR, David D, Schild A, Hoerndli F, Pennanen L, et al. Transgenic animal models of Alzheimer’s disease and related disorders: histopathology, behavior and therapy. Mol Psychiatry. 2004;9(7):664–83.

    Article  PubMed  Google Scholar 

  47. Mangold CA, Masser DR, Stanford DR, Bixler GV, Pisupati A, Giles CB, et al. CNS-wide sexually dimorphic induction of the major histocompatibility complex 1 pathway with aging. J Gerontol A Biol Sci Med Sci. 2017;72(1):16–29.

    Article  CAS  PubMed  Google Scholar 

  48. VanGuilder Starkey HD, Van Kirk CA, Bixler GV, Imperio CG, Kale VP, Serfass JM, et al. Neuroglial expression of the MHCI pathway and PirB receptor is upregulated in the hippocampus with advanced aging. J Mol Neurosci. 2012;48(1):111–26.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

Download references


The authors thank all the participants of the present study and all colleagues who have made contributions to building the CABLE cohort.


This study was supported by grants from the National Natural Science Foundation of China (82071201 and 81971032), Science and Technology Innovation 2030 Major Projects (2022ZD0211600), Shanghai Municipal Science and Technology Major Project (No.2018SHZDZX01), Research Start-up Fund of Huashan Hospital (2022QD002), Excellence 2025 Talent Cultivation Program at Fudan University (3030277001), ZHANGJIANG LAB, Tianqiao and Chrissy Chen Institute, and the State Key Laboratory of Neurobiology and Frontiers Center for Brain Science of Ministry of Education, Fudan University.

Author information

Authors and Affiliations



Concept and design: JTY, LT; Acquisition, analysis, or interpretation of data: All authors; Drafting of the manuscript: YMH; Critical revision of the manuscript for important intellectual content: All authors; Statistical analysis: YMH, YHM, and PYG; Chart design: YMH, YHM; Obtained funding: JTY; Administrative, technical, or material support: JTY, LT. The authors read and approved the final manuscript.

Corresponding authors

Correspondence to Lan Tan or Jin-Tai Yu.

Ethics declarations

Ethics approval and consent to participate

The CABLE study was approved by the Institutional Ethics Committee of Qingdao Municipal Hospital and carried out following the Declaration of Helsinki. All subjects or their proxies gave written consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1:

Fig. S1. The Quantile-Quantile plot of B2M. Fig. S2. Associations of age and B2M. Fig. S3. Levels of B2M in the CSF biomarker classifications. Table S1. The linear relationships between β2-microglobulin and CSF biomakers. Table S2. Interaction analysis of β2-microglobulin with multiple factors.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Huang, YM., Ma, YH., Gao, PY. et al. Plasma β2-microglobulin and cerebrospinal fluid biomarkers of Alzheimer’s disease pathology in cognitively intact older adults: the CABLE study. Alz Res Therapy 15, 69 (2023).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: