Synaptic dysfunction and glial activation markers throughout aging and early neurodegeneration: a longitudinal CSF biomarker-based study

Synaptic dysfunction and glial activation markers throughout aging and early neurodegeneration: a longitudinal CSF biomarker-based study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Synaptic dysfunction and glial activation markers throughout aging and early neurodegeneration: a longitudinal CSF biomarker-based study Mariana I. Muñoz-García, Yuetiva Deming, Ferran Lugo Hernández, and 13 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6982788/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 17 Oct, 2025 Read the published version in Molecular Neurodegeneration → Version 1 posted 9 You are reading this latest preprint version Abstract Background Synaptic homeostasis, maintained by microglia and astroglia, is disrupted throughout aging and early on in neurodegenerative diseases. Our aim was to study the relationship between TREM2-dependent microglial reactivity, astroglial response and synaptic dysfunction in two longitudinal cohorts of cognitively healthy volunteers and determine whether this relationship is influenced by AD core biomarkers. Methods We analyzed cross-sectional and longitudinal associations between cerebrospinal fluid levels of soluble TREM2 (sTREM2), astroglial markers (GFAP, S100B), and synaptic markers (neurogranin, α-synuclein) in cognitively unimpaired participants from the Wisconsin Registry for Alzheimer’s Prevention (WRAP) and the Alzheimer’s and Families (ALFA+) cohort. Biomarkers were quantified using validated immunoassays (NeuroToolKit, Roche), with sTREM2 measured using an in-house MSD-based assay in the WRAP cohort. Linear regression and linear mixed-effects models were used, both unadjusted and adjusted for Aβ42 and p-tau. Subgroup analyses were performed based on AT classification, APOE -ε4 status, and median splits of Aβ42/Aβ40 ratio and p-tau, to capture profiles suggestive of early AD-related neuropathogenesis. Results We found significant cross-sectional associations between sTREM2 and α-synuclein, as well as between sTREM2 and S100B, in subgroups exhibiting AD-related biomarker profiles. Longitudinally, lower baseline neurogranin and α-synuclein and higher S100B predicted greater increases in sTREM2 over time independently of AD-related markers in the WRAP cohort (β = −0.02, p = 0.006; β = −0.02, p = 0.01; β = 0.02, p = 0.03, respectively). In ALFA+, lower baseline α-synuclein also predicted a greater subsequent longitudinal increase in sTREM2, but only among individuals with Aβ42/Aβ40 ratio above the median (β = -0.01, p = 0.05). Notably, higher baseline sTREM2 was associated with a smaller longitudinal increase in neurogranin in both cohorts (β = -0.01, p = 0.03 for WRAP, β = -0.01, p = 0.04 in ALFA+). Conclusions Synaptic dysfunction markers at baseline influence the longitudinal dynamics of CSF sTREM2 independently of AD-pathology related biomarkers throughout aging and earliest stages of neurodegeneration. In turn, higher baseline sTREM2 is associated with more stable neurogranin levels over time. These results suggest an independent interaction between synaptic dysfunction and TREM2-dependent microglial activation throughout aging and early neurodegeneration beyond AD pathology. Microglia synaptic function aging neurodegeneration Figures Figure 1 Figure 2 Figure 3 Background Synaptic dysfunction and glial reactivity play crucial roles in the early stages and progression of neurodegenerative diseases, as well as throughout the aging process [ 1 , 2 ]. Experimental evidence suggests that microglia and astrocytes play critical roles in regulating synaptic plasticity during both physiological aging and disease [ 3 – 5 ]. Microglial-derived proteins such as TGF-β1 and C1q have been identified as crucial signals that impact synaptic protein homeostasis and influence synaptic maintenance in the aging brain [ 6 , 7 ]. Astrocytes contribute through neurotransmitter clearance, ion balance regulation, and gliotransmitter release, which affect short- and long-term synaptic plasticity [ 3 , 8 , 9 ]. Thus, both glial cell types are well-positioned to sense early disruptions in synaptic activity and potentially interact with synaptic failure throughout both aging and disease [ 3 ]. Microglial and astroglial responses to early protein aggregation have also been identified as key processes in neurodegenerative diseases [ 10 , 11 ]. Single-cell sequencing technologies have identified dynamic microglial populations transitioning from homeostatic to disease-associated profiles in response to initial protein aggregation [ 12 ]. Loss-of-function genetic variants affecting the microglial protein triggering receptor expressed on myeloid cells 2 (TREM2) increase AD risk by preventing microglia from transitioning to a protective disease-associated state [ 13 ]. Similarly, disease-associated astrocytes are often found surrounding pathological protein aggregates in various neurodegenerative diseases and appear to be induced by activated microglia through pro-inflammatory cytokines [ 14 ]. This activation leads to a loss of astrocytic homeostatic roles, including maintaining synaptic functionality, as shown in animal-based studies [ 15 ]. These findings suggest active synapse regulation by microglia and astroglia in neurodegeneration [ 16 ]. Nonetheless, the association between glial responses and synaptic dysfunction remains largely unexplored in human cohorts in the framework of aging and neurodegeneration. Synaptic function can be assessed in live human cohorts by measuring synapse-related biomarkers in cerebrospinal fluid (CSF). Neurogranin is a postsynaptic protein widely studied in CSF. Reduced CSF levels of neurogranin have been described in Parkinson disease (PD), dementia with Lewy bodies (DLB) and preclinical Alzheimer’s disease (AD) [ 17 – 19 ]. However, most studies in AD report elevated CSF neurogranin across the continuum of disease, as well as correlation with CSF p-tau, likely reflecting progressive dendritic pathology and/or neuronal hyperactivity [ 20 – 23 ]. CSF alpha-synuclein (α-syn), a crucial presynaptic protein, has different changes in concentration depending on the disease context, stage, and analytical method [ 24 – 26 ]. Discrepancies in CSF α-syn levels highlight the multifactorial nature of α-syn biology, which encompasses synaptic dysfunction, aggregation, neurodegeneration, and axonal remodeling [ 18 , 27 – 29 ]. Thus, these synaptic proteins are probably appropriate synapse-related markers during healthy aging or early stages of neurodegenerative processes but should be interpreted carefully in the context of advanced neurodegeneration. Glial function can also be studied in humans through biomarkers in biofluids, offering a translational bridge between experimental findings and clinical observations. The soluble cleavage product of the microglial protein TREM2 (sTREM2) serves as a marker of the cell-autonomous microglial activation response [ 28 , 29 ]. Longitudinal biomarker-based data from autosomal-dominant AD suggests that the main trigger of TREM2-dependent microglial response is initial Aβ aggregation, decades before the first symptoms appear [ 11 ]. Higher baseline levels and longitudinal increase of sTREM2 in CSF are associated with slower cognitive decline and reduced amyloid accumulation in AD patients, which supports the protective role of microglia [ 11 , 30 ]. Additionally, astroglial reactivity marker GFAP has been proposed to mediate the transition between soluble and insoluble Aβ and represent a possible connection between amyloidosis and tau-related neurodegeneration in sporadic AD [ 10 , 31 ]. Another important astroglial activation marker measurable in CSF is S100 calcium-binding protein B (S100B). This marker is proposed to play a role in modulating synaptic plasticity and microglial response [ 32 , 33 ]. These biomarkers are useful for monitoring the dynamic roles of microglia and astroglia in clinical cohorts throughout aging and disease stages. Understanding the interplay between synaptic dysfunction and glial responses is essential for characterizing the biological underpinnings of cognitive decline in aging and neurodegenerative diseases. These processes reflect key mechanisms underlying cognitive impairment and represent some of the earliest events driving disease progression, potentially opening new windows for therapeutic intervention. Nonetheless, the temporal relationships among presynaptic (α-syn), postsynaptic (neurogranin), microglial (sTREM2), and astroglial (GFAP, S100B) markers remain poorly understood, particularly in asymptomatic individuals, as very few studies directly investigated this association. In this study, we leverage longitudinal CSF biomarker data from two independent cohorts of cognitively unimpaired, late-middle-aged individuals to investigate how synaptic and glial signals crosstalk with TREM2-dependent microglial activation. By integrating cross-sectional and longitudinal analyses across populations with differing AD risk profiles, we aim to uncover early synapse-to-glia signaling dynamics that distinguish physiological aging from the preclinical phase of neurodegeneration—ultimately informing the development of precision strategies for early intervention. Materials and Methods Patient cohorts We conducted a longitudinal observational study using CSF biomarker data from two independent cohorts of cognitively unimpaired adults, aiming to assess associations between synaptic and glial biomarkers and microglial activation over time. The first cohort of participants stemmed from the Wisconsin Registry for Alzheimer’s Prevention (WRAP) study; an observational cohort first established in 2001. The WRAP study initially enrolled participants at midlife with a mean age of 54 and parental history of probable AD dementia and thus, at risk for late onset dementia. Since 2004, it also included participants without parental history of dementia to better understand its role in the risk of dementia. Genetic and clinical data was gathered initially, and subjects were followed longitudinally with neuropsychological evaluation, self-reported medical and lifestyle data, laboratory tests, and optional lumbar puncture (LP) at different time points (approximately every two years). Further details about the cohort are discussed elsewhere [ 34 ]. This study uses a subset of 239 WRAP participants, with available CSF samples, 116 of them with available longitudinal data, out of the 1561 total participants[ 34 ]. We repeated analyses in a confirmation cohort, the ALFA + cohort, a longitudinal observational cohort nested within the ALFA (for ALzheimer and FAmilies) parent cohort. The ALFA study was established between 2013 and 2014 and recruited 2,743 cognitively unimpaired individuals, primarily first-degree descendants of patients with sporadic AD, aged 45 to 75 years. Participants were extensively characterized at baseline, including sociodemographic, clinical, lifestyle, and cognitive measures, with additional data collected on modifiable risk factors and APOE genotype. Within the ALFA cohort, a subset was selected for the ALFA + study based on risk profile ( APOE and family history status), and 400 of those participants were available for cross-sectional analyses. ALFA + involves more detailed longitudinal phenotyping, including fluid biomarker collection. Baseline visits occurred between 2016 and 2019, with follow-up assessments every three years. In the present study, at baseline, 15 blood-based biomarkers were analyzed, and during the second wave (V2, begun in 2019), CSF biomarkers such as Aβ42, p-tau, t-tau, GFAP, neurogranin, S100, α-syn, sTREM2, and others were measured [ 35 ]. A subset of 259 had available longitudinal data at the time of analyses. CSF biomarker quantification Synapse-related biomarkers including neurogranin and α-syn, and astroglial markers S100B and GFAP, as well as Aβ40 were measured using the NeuroToolKit (NTK) in both the WRAP and ALFA + cohorts, as a panel of exploratory prototype assays [ 25 , 36 , 37 ]. The specific cleaved sTREM2 isoform was measured in the CSF of participants included in the WRAP cohort by an in-house immunoassay in the MSD platform as previously reported [ 11 , 37 ]. Total sTREM2 was quantified in the ALFA + cohort by the NTK [ 36 ]. AD core biomarkers including Aβ42 and total tau (t-tau) and tau phosphorylated at threonine 181 (p-tau) in CSF were quantified by the commercially available Elecsys® immunoassays, as described elsewhere [ 36 ]. In the WRAP cohort, we defined positivity in the AT classification as Aβ42/Aβ40 ratio 24.8 pg/mL; in the ALFA + cohort, as Aβ42/Aβ40 ratio 24 pg/mL. Biomarkers Aβ42, Aβ40, pTau, t-tau and S100B were measured using the Cobas® e 601 analyzer, and the remaining on a Cobas® e 411 analyzer (both Roche Diagnostics International Ltd) [ 36 ]. All NTK measurements for both cohorts were performed in singlicate at the Clinical Neurochemistry Laboratory at the University of Gothenburg (Sweden). Statistical analysis Only participants with complete data (all measurements for biomarkers and covariate data for each described model) were included in the analyses; no imputation was performed. We first tested for normality of the distribution for each biomarker using the Shapiro test. CSF Aβ42, Aβ42/Aβ40 ratio, p-tau, t-tau, sTREM2, neurogranin, α-syn, S100B, and GFAP did not follow a normal distribution and were thus log10-transformed. To describe the data, we stratified participants according to APOE status, medians of Aβ42/Aβ40 ratio and p-tau, and we used χ2 tests for categorical variables, and t-student or ANOVA for continuous variables to compare groups. We stratified the cohort according to medians of Aβ42/Aβ40 ratio and p-tau to better approach the contribution of the earliest AD-related pathological changes, since this cohort was composed of healthy participants with a low percentage of amyloid and p-tau positivity. This methodology is supported by findings showing differential longitudinal cognitive profiles based on preclinical AD-biomarker changes, when dividing the Aβ42 and p-tau values according to cohort-specific tertiles and medians [ 38 ]. For cross-sectional analysis, we calculated partial correlations adjusted by age across all biomarkers with the Pearson method. We then performed linear regression analysis to study the association between sTREM2 and synaptic function biomarkers, using TREM2 both as the main independent variable and the dependent outcome variable. For each analysis, we use two linear regression models: Model 1, adjusted for age and gender, and Model 2, further adjusted for baseline Aβ42 and p-tau levels. Analyses were conducted in the entire cohort and stratified by subgroups based on the median Aβ42/Aβ40 ratio, p-tau levels, Aβ or p-tau marker positivity (AT classification), and APOE carriage status, to better capture different preclinical stages of the AD continuum. Furthermore, we repeated the models including interaction terms between the independent biomarker and Aβ42 and p-tau, as continuous variables. The overall aim was to evaluate whether these associations were influenced by an underlying initial AD pathological process. For longitudinal analyses, we performed linear mixed effects models with random intercepts to account for within-subject variability over time, based on longitudinal data from 116 participants in the WRAP cohort and 259 in the ALFA + cohort. The WRAP cohort participants had mainly one follow-up visit (20 had 3 and 6 had 4 visits), while all participants in the ALFA + cohort had only one follow-up visit. For this reason, the statistical model included only random intercepts (not random slopes) to avoid convergence issues, which allowed to model between-subject variability without overfitting. Linear mixed-effects models were implemented using the lme4 package. Univariate models were used to assess the effect of baseline biomarkers (predictor) on the longitudinal change of the outcome biomarker. We included an interaction term between the baseline biomarker and time to understand how the relationship between biomarkers changed over time. The time variable was modeled as a continuous variable centered at baseline (time of first LP). We used sTREM2 both as the outcome biomarker and as the predictor for each synaptic biomarker. We adjusted the models by age, gender, and continuous values of Aβ42, and p-tau. We also calculated the interaction between time and baseline biomarker levels divided by medians. We performed sensitivity analysis by performing analysis in previously described subgroups. Given the exploratory nature of the study, p-values were considered nominally significant at p < 0.05 without correction for multiple comparisons. All statistical analyses were performed with R software ( http://www.r-project.org/ ) and RStudio (last updated version 2024.12.0 + 467). Results WRAP cohort Demographic and baseline biomarker characteristics of WRAP participants are summarized in Table 1 . Participants with both an Aβ42/Aβ40 ratio below the median and p-tau levels at or above the median were older at the time of LP and had a higher frequency of APOE ε4 carriers, compared to other subgroups. No differences were observed in gender, parental history of dementia, ethnicity, or MMSE score. As expected, groups defined by the median of Aβ42/Aβ40 ratio differed in percentage of amyloid positivity according to Aβ42/Aβ40 ratio and p-tau/Aβ42 ratio, as well as tau positivity according to pre-established cut-offs. Groups defined by median p-tau, also differed in percentage of amyloid positivity according to Aβ42/Aβ40 ratio and p-tau/Aβ42 ratio. The biomarker profiles adjusted by age are also shown in Table 1 . The group of participants with an Aβ42/Aβ40 ratio below the median showed biomarkers congruent with first stages of amyloid aggregation: lower Aβ42, higher t-tau and p-tau, lower Aβ42/Aβ40 ratio than participants with an Aβ42/Aβ40 ratio above the median. When stratifying by p-tau median, participants with p-tau levels above the median also had a biomarker profile suggestive of first stages within the AD continuum, except for having higher baseline Aβ42, but this was compensated with a significantly lower Aβ42/Aβ40 ratio. Cross-sectionally, participants with p-tau levels above the median had higher levels of sTREM2, GFAP, S100B, neurogranin and ⍺-syn than participants with p-tau levels below the median (Table 1 ). In contrast, levels of the studied proteins did not significantly differ between groups stratified according to Aβ42/Aβ40 ratio. Partial correlations adjusted by age between studied markers in the entire cohort are summarized in Supplementary Fig. 1 . Table 1 Demographics and baseline biomarkers in the WRAP (Wisconsin's Registry for Alzheimer Prevention) cohort Overall Aβ42/Aβ40 ratio ≥ median (2) Aβ42/Aβ40 ratio < median p p-tau ≥ median (4) p-tau < median p n 239 120 119 120 119 Age at time of LP [mean (SD)] 61.4 (7.2) 60.1 (7.47) 62.8 (6.6) 0.003 62.5 (7.30) 60.4 (6.94) 0.021 Gener [Male (%)] 89 (37.2) 45 (37.5) 44 (37.0) 1.000 39 (32.5) 50 (42.0) 0.165 PH of dementia [Yes (%)] 177 (74.1) 91 (75.8) 86 (72.3) 0.631 89 (74.2) 88 (73.9) 1.000 Ethnicity (%) 0.454 0.556 White 230 (96.2) 115 (95.8) 115 (96.6) 116 (96.7) 114 (95.8) Other (1) 9 (3.8) 5 (4.2) 4 (3.4) 4 (3.3) 4 (4.1) ApoE 𝛆4[Non-carrier (%)] 151 (63.2) 91 (75.8) 60 (50.4) < 0.001 67 (55.8) 84 (70.6) 0.026 MMSE [mean (SD)] 29.3 (0.9) 29.3 (0.89) 29.3 (0.94) 0.518 29.29 (1.00) 29.3 (0.82) 0.902 Education [mean (SD)] 16.3 (2.4) 16.1 (2.45) 16.4 (2.24) 0.213 16.27 (2.35) 16.2 (2.36) 0.896 Amyloid + (%) (3) 46 (19.2) 0 (0.0) 46 (38.7) < 0.001 41 (34.2) 5 (4.2) < 0.001 Tau + (%) (5) 25 (10.5) 4 (3.3) 21 (17.6) 0.001 25 (20.8) 0 (0.0) < 0.001 Positive ptau/ab42 ratio (%) 35 (14.6) 0 (0.0) 35 (29.4) < 0.001 31 (25.8) 4 (3.4) < 0.001 p (adj by age) p (adj by age) Aβ42 [mean (SD)] 881 (375) 1097 (337) 663 (272) < 0.001 1009 (419) 751 (269) < 0.001 Aβ40 [mean (SD)] 14024 (4395) 14494 (4072) 13550 (4667) 0.027 16995 (3570) 11028 (2841) < 0.001 Aβ42/Aβ40 ratio [mean (SD)] 0.06 (0.02) 0.08 (0.01) 0.05 (0.01) < 0.001 0.06 (0.02) 0.07 (0.01) < 0.001 T- tau [mean (SD)] 196 (63) 183.3 (52.8) 208 (77.6) 0.0374 246 (56.6) 145 (25.9) < 0.001 P-tau [mean (SD)] 17.3 (6.75) 15.9 (4.93) 18.6 (7.99) 0.0224 22.2 (6.12) 12.3 (2.22) < 0.001 csTREM2 [mean (SD)] 7.63 (2.88) 7.57 (2.78) 7.69 (2.99) 0.539 8.69 (2.95) 6.56 (2.37) < 0.001 sTREM2 [mean (SD)] 7.78 (2.29) 7.82 (2.18) 7.73 (2.41) 0.152 8.92 (2.34) 6.62 (1.54) < 0.001 Neurogranin [mean (SD)] 771 (298) 741 (248) 801 (339) 0.263 986 (255) 554 (138) < 0.001 𝛂- synuclein [mean (SD)] 152 (64.2) 149 (56.4) 155 (71.2) 0.860 193 (61.9) 110 (30.9) < 0.001 S100B [mean (SD)] 1.16 (0.30) 1.15 (0.29) 1.17 (0.30) 0.949 1.20 (0.28) 1.12 (0.30) 0.065 GFAP [mean (SD)] 8.69 (3.03) 8.67 (3.22) 8.72 (2.84) 0.253 9.72 (3.14) 7.66 (2.52) < 0.001 (1) Other: American Indian or Alaska Native, Asian, Black or African American and other (2) Aβ42/Aβ40 median = 0.067 (3) The cut-off for amyloid (A) positivity according to the Aβ42/Aβ40 ratio is 0.046. (4) P-tau median = 15.94 pg/mL. (5) The cut-off for tau (T) positivity according to P-tau is 24.8 pg/mL. LP: lumbar puncture. PH: Parental history. MMSE: mini-mental state examination score In cross-sectional analyses, we performed linear regression models adjusted for age and gender (Model 1), and then further adjusted for baseline Aβ42 and p-tau levels (Model 2), as well as after stratification by subgroups based on the median Aβ42/Aβ40 ratio, p-tau levels, Aβ or p-tau marker positivity (A/T classification), and APOE carriage status. These are summarized in Supplementary Table 1. We found significant cross-sectional associations between sTREM2 and GFAP (β = 0.26, p = 0.0001), S100B (β = 0.28, p = 0.002), neurogranin (β = 0.37, p < 0.0001), and ⍺-syn (β = 0.37, p < 0.0001), in the whole cohort using Model 1 (Supplementary Table 1). In contrast, after adjusting for Aβ42 and p-tau baseline levels (Model 2), we only found a trend for an association in the whole sample between sTREM2 and S100B (β = 0.16, p = 0.06), and between sTREM2 and ⍺-syn (β = 0.16, p = 0.09), indicating an influence of AD related markers on the previous associations. The results for Model 2 are shown in Fig. 1 . In subgroups stratified by AD biomarker profiles, we found a significant cross-sectional association between sTREM2 and ⍺-syn CSF levels in T + participants even after AD-related markers adjustment, indicating that the cross-sectional relationship is not influenced by AD-related biomarkers (β = 0.83, p = 0.02) shown in Fig. 1 B and Supplementary Table 1. Additionally, we found a significant association between sTREM2 and S100B in participants with Aβ42/Aβ40 ratio below the median (β = 0.26, p = 0.03) and in participants with p-tau levels above the median (β = 0.28, p = 0.01) (Fig. 1 C and Supplementary Table 1) after AD-related marker adjustment. This suggests that the association between sTREM2 and S100B is present in individuals with a biomarker profile indicative of first stages of an AD pathology and is not mediated by Aβ42 or p-tau levels. We did not find any other significant cross-sectional association between sTREM2 and neurogranin or GFAP in models adjusted by Aβ42 and p-tau. The cross-sectional associations between sTREM2 and studied biomarkers were not affected by the APOE 𝛆4 allele carriage status, whether included as a covariate in regression models or assessed in stratified analyses (Supplementary Table 1). Furthermore, interaction terms were tested using continuous values of Aβ42 and p-tau; none were statistically significant and are not shown. Adjusted by age, gender, AΒ42 and p-tau. Then, we examined whether baseline astroglial response and synapse-related markers influenced the longitudinal dynamics of sTREM2 using linear mixed models. Two models were applied: Model 1, adjusted for age and gender, and Model 2, further adjusted for baseline Aβ42 and p-tau levels. Among the 116 participants, follow-up data were available for 90 individuals with one visit, 20 with two visits, and 6 with three visits. The adjustment for baseline Aβ42 and p-tau did not alter the association between baseline astroglial and synaptic markers and the longitudinal change in sTREM2. Nevertheless, to evaluate the independent associations between astroglial and synaptic markers and the longitudinal change in sTREM2, we focused on the adjusted models (Model 2). A summary of these models is provided in Supplementary Table 2. Concerning the synapse-related markers, lower levels of baseline neurogranin (β-coefficient for interaction with time = -0.04, p = 0.0002) and ⍺-synuclein (β-coefficient for interaction with time = -0.03, p = 0.004) significantly predicted a larger subsequent longitudinal increase in sTREM2 CSF levels over time. When stratifying baseline neurogranin and ⍺-synuclein by their median values, we consistently observed that levels below the median were significantly associated with an increase in sTREM2 over time (Fig. 2 A and 2 B). Regarding the astroglial markers GFAP and S100B, we only observed a significant association with baseline S100B when stratifying by its median, while no significant associations were found between baseline GFAP and the longitudinal change of sTREM2 levels. Participants with baseline S100B levels above the median showed a significantly greater subsequent longitudinal increase in CSF sTREM2 levels (β-coefficient for the interaction between time and S100B above median = 0.02, p = 0.03), as shown in Fig. 2 C and Supplementary Table 2. These longitudinal associations remained significant when stratifying by Aβ42/Aβ40 ratio medians, p-tau medians, amyloid or p-tau positivity cut-offs, or APOE 𝜀4 carrier status (Supplementary Table 2). Finally, we evaluated whether baseline sTREM2 levels influenced the subsequent longitudinal changes in synapse-related markers and astroglial response. After adjusting for Aβ42 and p-tau, we found that higher baseline sTREM2 levels were significantly associated with a diminished longitudinal increase in neurogranin over time (β-coefficient for interaction with time =-0.03, p = 0.0001) (Fig. 3 A). No other significant associations were observed between baseline sTREM2 levels and the longitudinal changes of ⍺-synuclein, S100B or, GFAP (Fig. 3 B, 3 C and 3 D). ALFA + cohort To replicate our findings in an independent sample, we applied the same analytic pipeline to the ALFA + cohort. Demographics of this cohort are shown in Table 2 . Demographically, the main differences between the cohorts were the proportion of APOE 𝛆4 carriers (54% in ALFA + vs. 37% WRAP), which influenced the proportion of A + participants (33.8% vs. 19%), but not T+ (11.9% vs. 10.5%). Subgroup analysis revealed that participants with an Aβ42/Aβ40 ratio below the median and p-tau above the median had profiles closer to the AD continuum. We repeated the same cross-sectional and longitudinal analyses used in the WRAP cohort. Partial correlation results are summarized in Supplementary Fig. 1 , showing a similar profile to correlations in the WRAP cohort. Table 2. Demographics and baseline biomarkers for the Alzheimer and Families (ALFA+) cohort Overall Aβ42/Aβ40 ratio ≥ median (2) Aβ42/Aβ40 ratio < median p p-tau ≥ median (4) p-tau < median p n 400 200 200 177 176 Age at time of LP [mean (SD)] 61.2 (4.69) 60.4 (4.34) 61.9 (4.90) 0.001 61.7 (4.37) 59.9 (4.82) < 0.001 Gener [Male (%)] 245 (62.5) 123 (62.8) 122 (62.2) 1.000 114 (65.5) 110 (64.3) 0.906 Ethnicity (%) 0.343 0.716 White 394 (98.5) 197 (98.5) 197 (98.5) 175 (98.9) 172 (97.7) Other (1) 6 (1.5) 3 (1.5) 3 (1.5) 2 (1.1) 4 (2.3) ApoE 𝛆4[Non-carrier (%)] 184 (46.0) 128 (64.0) 56 (28.0) < 0.001 80 (45.2) 84 (47.7) 0.712 MMSE [mean (SD)] 29.1 (0.99) 29.2 (0.92) 29.1 (1.05) 0.206 29.1 (1.10) 29.2 (0.88) 0.450 Education [mean (SD)] 4.49 (0.91) 4.55 (0.88) 4.42 (0.94) 0.170 4.43 (0.97) 4.51 (0.88) 0.406 Amyloid + (%) (3) 135 (33.8) 0 (0.0) 135 (67.5) < 0.001 73 (41.2) 43 (24.4) 0.001 Tau + (%) (5) 42 (11.9) 10 (5.4) 32 (19.0) < 0.001 42 (23.7) 0 (0.0) < 0.001 Positive ptau/ab42 ratio (%) 50 (14.2) 0 (0.0) 50 (29.8) < 0.001 44 (24.9) 6 (3.4) < 0.001 p (adj by age) p (adj by age) Aβ42 [mean (SD)] 1322 (597) 1661 (601) 983 (350) < 0.001 1607 (713) 1143 (341) < 0.001 Aβ40 [mean (SD)] 17403 (4997) 17971 (5120) 16834 (4817) 0.023 21191 (4207) 14521 (2874) < 0.001 Aβ42/Aβ40 ratio [mean (SD)] 0.08 (0.02) 0.09 (0.01) 0.06 (0.02) < 0.001 0.07 (0.02) 0.08 (0.01) 0.059 T- tau [mean (SD)] 201 (72.9) 192 (56.5) 212 (86) 0.009 253 (67.8) 149 (24.0) < 0.001 P-tau [mean (SD)] 16.5 (7.59) 15.4 (5.18) 17.8 (9.43) 0.003 21.6 (7.78) 11.5 (2.01) < 0.001 sTREM2 [mean (SD)] 7956 (2257) 8070 (2220) 7842 (2293) 0.313 8964 (2399) 7161 (1711) < 0.001 Neurogranin [mean (SD)] 800 (331) 787 (296) 812 (362) 0.452 1047 (313) 587 (134) < 0.001 𝛂- synuclein [mean (SD)] 234 (254) 235 (227) 232 (279) 0.905 298 (329) 184 (164) < 0.001 S100B [mean (SD)] 1024 (236) 999 (209) 1049 (258) 0.031 1087 (250) 966 (206) < 0.001 GFAP [mean (SD)] 7720 (2638) 7675 (2842) 7766 (2424) 0.731 8814 (2837) 6718 (2091) < 0.001 (1) Other: Gypsy ethnic, latin american and not evaluated (2) Aβ42/Aβ40 median = 0.08057 (3) The cut-off for amyloid (A) positivity according to the Aβ42/Aβ40 ratio is 0.071. (4) P-tau median = 14.75 pg/mL. (5) The cut-off for tau (T) positivity according to P-tau is 24 pg/mL. LP: lumbar puncture. PH: Parental history. MMSE: mini-mental state examination score The linear regression models showed associations between sTREM2 and synapse-related and astroglial biomarkers, summarized in Supplementary Table 3 . As in the WRAP cohort, these associations were attenuated after adjusting for Aβ42 and p-tau levels, indicating partial dependence on AD pathology (Model 2). However, negative associations remained between sTREM2 and neurogranin (β = -0.21, p = 0.04), as well as in the subgroups with below-median p-tau (β = -0.37, p = 0.002) and T- (β = -0.21, p = 0.04). In contrast to WRAP cohort results, sTREM2 and ⍺-synuclein showed positive adjusted associations only in the below-median Aβ42/Aβ40 ratio and A + group (β = 0.15, p = 0.02 and β = 0.20, p = 0.007, respectively). Consistent with WRAP results, the ALFA + cohort showed a significant positive association between sTREM2 and S100B in the whole sample after adjusting for Aβ42 and p-tau (β = 0.28, p = 0.00002). This association was also significant in almost all subgroups, particularly in the A + group (β = 0.51, p = 0.000001. Furthermore, there was a significant interaction between S100B and Aβ42 levels in the whole cohort (p = 0.03). In contrast to the WRAP results, the ALFA + cohort demonstrated a positive, significant adjusted association between GFAP and sTREM2 in the whole sample (β = 0.24, p = 0.000002), which was also significant in most subgroups, except for T + and below-median Aβ42/Aβ40 ratio. Remaining interaction terms were tested using continuous values of Aβ42 and p-tau; none were statistically significant and are not shown. Longitudinally, the linear mixed models revealed similar trends, with associations found only between synapse-related biomarkers and sTREM2, but not between glial activation biomarkers and sTREM2. They are summarized in Supplementary Table 4. Importantly, in contrast to the WRAP cohort results, the adjustment for Aβ42 and p-tau influenced the coefficients in Model 2. For neurogranin, we found an association between lower baseline neurogranin and a larger longitudinal increase in sTREM2 in Model 1, only in the above-median Aβ42/Aβ40 ratio group (β-coefficient for interaction with time = -0.02, p = 0.02) and the A- group (β-coefficient for interaction with time = -0.002, p = 0.008), suggesting an association in participants without evidence of amyloid pathology. However, these associations did not remain significant after adjusting for Aβ42 and p-tau (Model 2), suggesting confounding by AD pathology. Similarly, we observed that lower baseline α-syn was associated with a larger longitudinal increase in sTREM2 over time in the above-median Aβ42/Aβ40 ratio group for both model 1 (β-coefficient for interaction with time = -0.01, p = 0.05) and model 2 (β-coefficient for interaction with time = -0.01, p = 0.05). As in the WRAP cohort, sTREM2 levels above the median were associated with a diminished longitudinal increase in neurogranin (β-coefficient for interaction with time = -0.01, p = 0.04). In sum, while both cohorts demonstrated cross-sectional associations between sTREM2 and markers of synaptic and glial function, the longitudinal patterns were more robust and AD-independent in the WRAP cohort compared to ALFA+. Discussion This study offers new insights into the interplay between synaptic dysfunction, TREM2-dependent microglial response, and astroglial activation, through a CSF-based biomarker approach applied to two independent longitudinal cohorts of cognitively normal, late-middle-aged individuals. Longitudinally, lower baseline levels of synaptic proteins and higher levels of S100B predicted a larger subsequent increase in sTREM2, independent of AD-related biomarkers. These findings suggest that early synaptic dysfunction may act as a trigger for TREM2-dependent microglial activation, regardless of AD pathology. Additionally, higher baseline levels of sTREM2 were associated with more stable neurogranin levels over time, further supporting the role of TREM2 as a modulator of synaptic function and potentially protective against synaptic dysregulation and cognitive decline throughout aging and early stages of neurodegenerative processes. Cross-sectionally, we found an association between sTREM2 and α-syn, specifically in participants with neurodegeneration-related biomarker profiles (T + group in the WRAP cohort, and Aβ42/Aβ40 below median and A + groups in ALFA+). Interpreting CSF α-syn levels remains challenging due to its dual role in pathological processes, such as aggregation and neurodegeneration in synucleinopathies, and physiological processes, including synaptic function and axonal remodeling during normal aging [ 18 , 27 ]. In neurodegenerative diseases, total CSF α-syn is often considered a marker of neurodegeneration rather than synaptic dysfunction [ 24 , 39 ]. This is supported by previously reported significant correlations between ⍺-syn, p-tau and t-tau in CSF, in concordance with our results [ 25 , 40 ]. However, in PD, most studies report reduced total CSF ⍺-syn levels during the early stages, likely reflecting initial synaptic dysfunction or early α-syn aggregation. In later stages, higher levels have been reported, which may signal overt neuronal injury [ 41 ]. The variations in α-syn levels clearly illustrate that it must be interpreted within its biological context. We interpret the observed cross-sectional relationship between sTREM2 and α-syn in participants with neurodegeneration-related biomarker profiles as reflective of a shared cross-sectional association with incipient neurodegeneration. Regarding the cross-sectional relationship between sTREM2 and astroglial activation markers, we found a significant association between sTREM2 and S100B in the ALFA + cohort and a trend toward an association in the WRAP cohort. These associations remained robust after adjusting for AD-related biomarkers and were stronger in subgroups with a biochemical profile suggestive of early phases of AD pathology. This observation is consistent with previous studies that show stronger sTREM2–S100B correlations in asymptomatic participants with elevated p-tau/Aβ42 ratios[ 25 ] and across the symptomatic AD continuum [ 40 , 42 ]. This suggests increased interaction between astroglial and microglial responses in early AD. Interestingly, we observed no cross-sectional association between GFAP and sTREM2 in the WRAP cohort. Interestingly, both S100B and GFAP demonstrated significant cross-sectional associations with sTREM2 across subgroups in the ALFA + cohort. Previous research has already described a relatively low correlation between S100B and other astroglial biomarkers [ 25 ], which suggests that S100B represents a distinct response specifically related to synaptic dysfunction rather than general astrocyte activation. Given its theoretical dual role—neurotrophic at nanomolar and pro-inflammatory at micromolar concentrations—S100B may reflect a neuroprotective astroglial response to early synaptic dysfunction in asymptomatic late-middle-aged individuals [ 43 , 44 ]. This interpretation is reinforced by the weak correlation between S100B and GFAP, suggesting S100B secretion occurs independently from the pathological astrocytic hyperactivation indicated by increased GFAP. Differences between the WRAP and ALFA + cohorts could stem from their distinct participant characteristics, as the ALFA + cohort includes healthy volunteers enriched for AD risk factors, likely leading to earlier GFAP elevations as an initial astroglial response to pathology. In fact, the observed cross-sectional association between GFAP and sTREM2 in the ALFA + cohort probably reflects the early interplay between astroglial and microglial activation as AD-related changes begin. Overall, our findings indicate distinct astroglial response patterns involving S100B and GFAP across cohorts with varying AD risk profiles. This underscores a complex interplay between astroglial and microglial activation that may differentially reflect synaptic dysfunction and initial amyloid-related pathology. Longitudinally, we observed that a biomarker profile suggestive of early synaptic dysfunction at baseline —characterized by lower levels of α-syn and neurogranin along with higher levels of S100B in CSF— was associated with a greater subsequent longitudinal increase of sTREM2 in CSF over time. These associations were independent of AD-related biomarker status in the WRAP cohort of cognitively healthy, late-middle-aged individuals. As discussed, interpreting CSF α-syn levels is challenging due to its involvement in different pathological and physiological processes [ 18 , 27 ]. In participants of the WRAP cohort, who are mainly individuals without manifest amyloid deposition nor neurodegeneration, we interpret lower CSF α-syn levels as indicative of age-related synaptic dysfunction rather than incipient neurodegeneration. In contrast, in individuals with overt neurodegenerative processes or more evident AD pathology, α-syn levels may represent the underlying neurodegeneration, which our cross-sectional findings support. In the ALFA + cohort, lower levels of α-syn at baseline were associated to greater increases in sTREM2 specifically among participants with higher Aβ42/Aβ40 ratios. This highlights how synaptic dysfunction could be a booster of microglial activation within aging or non-AD neurodegeneration rather than within the AD continuum. Similarly, lower neurogranin levels in individuals without biomarker evidence of AD likely reflect reduced postsynaptic activity. This interpretation is consistent with prior findings in PD, where neurogranin is reduced and correlates with cortical hypometabolism and cognitive deficits [ 45 ]. However, elevated CSF neurogranin has been reported even at presymptomatic stages in the AD continuum, likely reflecting neurodegeneration or neuronal hyperactivity rather than isolated synaptic dysfunction [ 17 , 21 , 22 , 46 – 48 ]. The strong correlation between neurogranin and tau-related markers supports its role as a disease-stage specific injury marker of AD [ 22 , 46 ]. Notably, increased network excitability—along with heightened seizure susceptibility—has been observed in early AD and may contribute to regional increases in synaptic density and elevated neurogranin levels [ 49 ], highlighting its dynamic, context-dependent interpretation. In the ALFA + cohort, lower baseline neurogranin was also associated with increased sTREM2 over time. However, only in subgroups without amyloid pathology and prior to adjustment for p-tau, suggesting that tau pathology may mask this association in a cohort with a high AD-risk profile. Further supporting early synaptic dysfunction as an independent trigger for TREM2-dependent microglial activation, we found that individuals with higher S100B levels at baseline exhibited a greater longitudinal increase in sTREM2 in the WRAP cohort. Interestingly, baseline GFAP levels were not predictive of longitudinal changes in sTREM2 in the WRAP nor in the ALFA + cohort. The absence of longitudinal associations with GFAP suggests that generalized astroglial activation does not independently boost microglial responses over time. Instead, the selective longitudinal relationship between elevated S100B and subsequent sTREM2 increase reinforces our interpretation of S100B as a synapse-coupled astroglial signal that can prime microglia independently of GFAP-defined astrocytic activation during physiological aging and non-AD neurodegenerative processes. Mechanistically, astrocytic S100B may mark dendritic-spine stress that, in turn, triggers a TREM2-dependent pruning response, echoing the complement-mediated synapse-elimination pathway observed in recent mouse and human studies [ 50 ]. In the context of AD pathology, however, the strong effect of Aβ aggregation in boosting TREM2-dependent microglial response may override the subtler modulatory influence of synaptic dysfunction throughout aging and non-AD neurodegeneration processes. And finally, we found that higher baseline CSF sTREM2 predicted a slower rise in neurogranin yet had no influence on α-syn or S100B trajectories. This pattern is congruent with recent findings that show TREM2-competent microglia identify phosphatidylserine-tagged, hyperactive spines and remove them via a complement pathway, thereby normalizing circuit activity both in disease models and clinical cohorts [ 50 – 55 ]. Instead, loss-of-function TREM2 variants in mice increase spine density, drive cortical hyperexcitability, and heighten seizure susceptibility [ 53 ]. This might be mirrored in electrophysiological phenotypes of patients with early AD who experience a higher incidence of subclinical and overt epileptic events [ 49 ]. Together, our findings and external evidence support a modulatory effect of TREM2-activated microglia on excessive synaptic activity: pruning superfluous spines, stabilizing neurogranin release, and reducing network hyperexcitability and seizure risk. The absence of longitudinal effects on the presynaptic marker α-syn or on astroglial S100B emphasizes that this microglial feedback loop is largely postsynaptic-specific. This study has several limitations. First, the inclusion of only cognitively unimpaired individuals limits generalizability to symptomatic stages, though it enables the study of preclinical processes. Second, CSF biomarkers do not fully reflect regional synaptic or neuroinflammatory dynamics. Third, modest sample size and limited longitudinal data reduce statistical power and temporal resolution. Additionally, strong inter-biomarker correlations may obscure independent effects, despite the use of complementary sTREM2 assays. The use of cohort-specific Aβ and tau cut-offs may affect comparability across studies. Finally, despite adjusting for p-tau, residual confounding cannot be excluded due to its collinearity with synaptic markers. Despite these limitations, our study has several strengths. One key strength is the availability of longitudinal CSF biomarker data from cognitively normal individuals. This allows us to examine biomarker trajectories over time, providing valuable insights into the evolution of synaptic and glial markers in aging and the preclinical stages of neurodegeneration. Another strength is the validation of findings in a secondary cohort of cognitively normal individuals, enhancing the robustness of our results. Furthermore, we incorporated multiple biomarkers to capture the complexity of the biological processes underlying synaptic dysfunction and microglial activation, ensuring a more comprehensive assessment of their interplay. Together, our cross-sectional and longitudinal findings suggest that synaptic stress activates a TREM2-dependent microglial response across aging and in non-AD neuropathological contexts. Once Aβ aggregation begins along the AD-continuum, its potent effect on activating the TREM2 pathway may dominate, diminishing the relative contribution of synaptic dysfunction. Astroglial–microglial coupling also changes with AD biomarker status, emphasizing the dynamic and coordinated glial response to early neuropathological changes. Our findings support ongoing therapeutic strategies that target the synapse-to-glia axis, boosting TREM2 signaling or protecting synapses [ 29 , 56 ]. The data highlights that the clinical benefit of modulating TREM2 will hinge on whether synaptic stress or established Aβ/tau pathology is the predominant biological driver. Abbreviations AD- Alzheimer’s disease ALFA- ALzheimer and Families cohort Aβ- amyloid-β CSF- cerebrospinal fluid DLB- dementia with Lewy bodies GFAP- glial fibrillary acid protein LP- lumbar puncture NTK- NeuroToolKit PD-Parkinson’s disease S100B- S100 calcium-binding protein B sTREM2- soluble cleavage product of the microglial protein TREM2 TREM2- triggering receptor on myeloid cells 2 WRAP- Wisconsin Registry for Alzheimer’s Prevention α-syn- alpha-synuclein Declarations Ethics approval and consent to participate The WRAP cohort study was conducted in compliance with the ethical principles for human subjects’ research defined in the Declaration of Helsinki, including approval by the University of Wisconsin-Madison Institutional Review Board. The ALFA study protocol was approved by the Independent Ethics committee Parc de Salut Mar and registered at clinicaltrials.gov (identifier: NCT01835717). It was conducted in accordance with the directives of the Spanish Law 14/2007, of 3 rd of July, on Biomedical Research. All participants in the ALFA study accepted the study procedures by signing an informed consent form and had a close relative volunteering to participate in the functional assessment procedure of the participant, who also granted his or her consent. Availability of data and materials Data utilized in the WRAP cohort consist of sensitive, human research participant data and participants have not signed informed consent to have their data shared in public repositories for publications. Therefore, data deposition is unethical. Data are available upon request for authorized researchers who meet the criteria for access to confidential data. The data underlying the results presented in this study are available from http://www.wai.wisc.edu/research/. Data from the ALFA+ cohort is also available upon request from https://www.barcelonabeta.org/es/estudio-alfa/sobre-el-estudio-alfa. Competing interests M.I.M.-G has received in the past 36mo speaker fees by Almirall. E.M.-R. has given lectures and symposia sponsored by KRKA Farmaceutica SL and Laboratorios Esteve SA. M.S.-C. has received in the past 36mo consultancy/speaker fees (paid to the institution) from by Almirall, Eli Lilly, Quanterix, Novo Nordisk, and Roche Diagnostics. He has received consultancy fees or served on advisory boards (paid to the institution) of Eli Lilly, Grifols, Novo Nordisk, and Roche Diagnostics. He was granted a project and is a site investigator of a clinical trial (funded to the institution) by Roche Diagnostics. In-kind support for research (to the institution) was received from ADx Neurosciences, Alamar Biosciences, ALZPath, Avid Radiopharmaceuticals, Eli Lilly, Fujirebio, Janssen Research & Development, Meso Scale Discovery, and Roche Diagnostics; MS-C did not receive any personal compensation from these organizations or any other for-profit organization. H.Z. has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp & Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of MicThera (outside submitted work). S.C.-J. serves as a consultant to Eli Lilly, Merck, AlzPath and Enigma Biomedical. G.K. is a full-time employee of Roche Diagnostics GmbH, Penzberg, Germany. CQ-R is a full-time employee of Roche Diagnostics International Ltd, Rotkreuz, Switzerland. Funding E.M.-R. receives funding by the Instituto de Salud Carlos III (ISCIII) under the Juan Rodés Program (Grant number JR21/00014) and through the project PI22/00215; she is also funded by Eugenio Rodríguez Pascual Foundation through the project FERP-2024-091. M.I.M.-G receives funding by the Instituto de Salud Carlos III (ISCIII) under the Rio Hortega Program (Grant number CM24/00130). The Wisconsin Registry for Alzheimer’s Prevention is funded by the National Institute on Aging R01 AG027161 with additional funding from AG021155 and the CSF collection service of the Wisconsin Alzheimer’s Disease Research Center P30AG062715. MSC receives funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (Grant agreement No. 948677); ERA PerMed-ERA NET and the Generalitat de Catalunya (Departament de Salut) through the project SLD077/21/000001; Project "PI19/00155" and “PI22/00456, funded by Instituto de Salud Carlos III (ISCIII) and co-funded by the European Union; and from a fellowship from ”la Caixa” Foundation (ID 100010434) and from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 847648 (LCF/BQ/PR21/11840004). HZ is a Wallenberg Scholar and a Distinguished Professor at the Swedish Research Council supported by grants from the Swedish Research Council (#2023-00356, #2022-01018 and #2019-02397), the European Union’s Horizon Europe research and innovation programme under grant agreement No 101053962, Swedish State Support for Clinical Research (#ALFGBG-71320), the Alzheimer Drug Discovery Foundation (ADDF), USA (#201809-2016862), the AD Strategic Fund and the Alzheimer's Association (#ADSF-21-831376-C, #ADSF-21-831381-C, #ADSF-21-831377-C, and #ADSF-24-1284328-C), the European Partnership on Metrology, co-financed from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States (NEuroBioStand, #22HLT07), the Bluefield Project, Cure Alzheimer’s Fund, the Olav Thon Foundation, the Erling-Persson Family Foundation, Familjen Rönströms Stiftelse, Familjen Beiglers Stiftelse, Stiftelsen för Gamla Tjänarinnor, Hjärnfonden, Sweden (#FO2022-0270), the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 860197 (MIRIADE), the European Union Joint Programme – Neurodegenerative Disease Research (JPND2021-00694), the National Institute for Health and Care Research University College London Hospitals Biomedical Research Centre, the UK Dementia Research Institute at UCL (UKDRI-1003), and an anonymous donor. Authors' contributions E.M.-R., Y.D, and B.B.B. contributed to the conceptualization of the study. M.I.M.-G. was responsible for data analysis and wrote the initial draft of the manuscript. E.M.-R. co-wrote the manuscript. F.L.H. contributed to data analysis in the ALFA+ cohort. Y.D., S.J., S.A., C.C., O.C.O., and B.B.B. contributed to data acquisition, WRAP cohort management, and interpretation of results. E.M.-R. contributed to biomarker measurements for the WRAP cohort. M.S.-C. and F.L.H. contributed to data acquisition and interpretation for the ALFA cohort. G.K. and C.Q.R. contributed to biomarker measurements for the ALFA cohort. Y.D., S.J., S.A., C.C., O.C.O., D.P.-M., A.V.-G., K.B., M.S.-C., H.Z., and B.B.B. contributed to critical revision of the manuscript. All authors reviewed and approved the final version of the manuscript. Acknowledgements: This publication is part of the ALFA study (ALzheimers and Families). The authors would like to express their most sincere gratitude to the ALFA project participants and relatives without whom this research would not have been possible. The authors thank Roche Diagnostics International Ltd for providing the kits to measure CSF biomarkers. The Roche NeuroToolKit is a panel of exploratory prototype assays designed to robustly evaluate biomarkers associated with key pathologic events characteristic of AD and other neurological disorders, used for research purposes only and not approved for clinical use. Elecsys Phospho-Tau (181P) CSF and Elecsys Total-Tau CSF assays are approved for clinical use. COBAS and COBAS and ELECSYS are trademarks of Roche. All other product names and trademarks are the property of their respective owners. Collaborators of the ALFA Study are: Annabella Beteta, Anna Brugulat-Serrat, Alba Cañas, Irene Cumplido-Mayoral, Carme Deulofeu, Ruth Dominguez, Maria Emilio, Karine Fauria, Ana Fernández-Arcos, Sherezade Fuentes, Patricia Genius, Laura Hernández, Gema Huesa, Jordi Huguet, Paula Marne, Tania Menchón, Wiesje Pelkmans, Albina Polo, Sandra Pradas, Blanca Rodríguez-Fernández, Anna Soteras, Laura Stankeviciute, and Marc Vilanova The NeuroToolKit is a panel of exploratory prototype assays designed to robustly evaluate biomarkers associated with key pathologic events characteristic of AD and other neurological disorders, used for research purposes only and not approved for clinical use (Roche Diagnostics International Ltd, Rotkreuz, Switzerland). COBAS and ELECSYS are trademarks of Roche. Elecsys β-Amyloid (1–42) CSF, Elecsys Phospho-Tau (181P) CSF and Elecsys Total-Tau CSF assays are approved for clinical use. References Wilson III DM, Cookson MR, Van L, Bosch D, Zetterberg H, Holtzman DM, et al. Hallmarks of neurodegenerative diseases. Cell. 2023;186:693–714. Simuni T, Chahine LM, Poston K, Brumm M, Buracchio T, Campbell M, et al. A biological definition of neuronal α-synuclein disease: towards an integrated staging system for research. Lancet Neurol. 2024;23:178–90. Chung WS, Welsh CA, Barres BA, Stevens B. Do Glia Drive Synaptic and Cognitive Impairment in Disease? Nat Neurosci. 2015;18:1539. Li L, Lu S, Zhu J, Yu X, Hou S, Huang Y, et al. Astrocytes Excessively Engulf Synapses in a Mouse Model of Alzheimer’s Disease. Int J Mol Sci. 2024;25. Gómez-Gonzalo M, Martin-Fernandez M, Martínez-Murillo R, Mederos S, Hernández-Vivanco A, Jamison S, et al. Neuron–astrocyte signaling is preserved in the aging brain. Glia. 2017;65:569–80. Soreq L, Rose J, Soreq E, Hardy J, Trabzuni D, Cookson MR, et al. Major Shifts in Glial Regional Identity Are a Transcriptional Hallmark of Human Brain Aging. Cell Rep. 2017;18:557–70. Scott-Hewitt N, Mahoney M, Huang Y, Korte N, Yvanka de Soysa T, Wilton DK, et al. Microglial-derived C1q integrates into neuronal ribonucleoprotein complexes and impacts protein homeostasis in the aging brain. Cell. 2024;187:4193-4212.e24. Sancho L, Contreras M, Allen NJ. Glia as sculptors of synaptic plasticity. Neurosci Res. 2021;167:17–29. Santello M, Toni N, Volterra A. Astrocyte function from information processing to cognition and cognitive impairment. Nat Neurosci. 2019;22:154–66. Bellaver B, Povala G, Ferreira PCL, Ferrari-Souza JP, Leffa DT, Lussier FZ, et al. Astrocyte reactivity influences amyloid-β effects on tau pathology in preclinical Alzheimer’s disease. Nat Med. 2023;29:1775–81. Morenas-Rodríguez E, Li Y, Nuscher B, Franzmeier N, Xiong C, Suárez-Calvet M, et al. Soluble TREM2 in CSF and its association with other biomarkers and cognition in autosomal-dominant Alzheimer’s disease: a longitudinal observational study. Lancet Neurol. 2022;21:329–41. Keren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, et al. A Unique Microglia Type Associated with Restricting Development of Alzheimer’s Disease. Cell. 2017;169:1276–90. Mazaheri F, Snaidero N, Kleinberger G, Madore C, Daria A, Werner G, et al. TREM 2 deficiency impairs chemotaxis and microglial responses to neuronal injury . EMBO Rep. 2017;18:1186–98. Habib N, McCabe C, Medina S, Varshavsky M, Kitsberg D, Dvir-Szternfeld R, et al. Disease-associated astrocytes in Alzheimer’s disease and aging. Nat Neurosci. 2020;23:701. Liddelow SA, Guttenplan KA, Clarke LE, Bennett FC, Bohlen CJ, Schirmer L, et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature. 2017;541:481. Tzioras M, McGeachan RI, Durrant CS, Spires-Jones TL. Synaptic degeneration in Alzheimer disease. Nat Rev Neurol. 2023;19:19–38. Barba L, Abu-Rumeileh S, Halbgebauer S, Bellomo G, Paolini Paoletti F, Gaetani L, et al. CSF Synaptic Biomarkers in AT(N)-Based Subgroups of Lewy Body Disease. Neurology. 2023;101:e50. Paolini Paoletti F, Gaetani L, Bellomo G, Chipi E, Salvadori N, Montanucci C, et al. CSF neurochemical profile and cognitive changes in Parkinson’s disease with mild cognitive impairment. NPJ Parkinsons Dis. 2023;9:68. Lehmann S, Schraen-Maschke S, Buée L, Vidal JS, Delaby C, Hirtz C, et al. Clarifying the association of CSF Aβ, tau, BACE1, and neurogranin with AT(N) stages in Alzheimer disease. Mol Neurodegener. 2024;19:66. Nilsson J, Binette AP, Palmqvist S, Brum WS, Janelidze S, Ashton NJ, et al. Cerebrospinal fluid biomarker panel for synaptic dysfunction in a broad spectrum of neurodegenerative diseases. Brain. 2024; Portelius E, Zetterberg H, Skillbäck T, Törnqvist U, Andreasson U, Trojanowski JQ, et al. Cerebrospinal fluid neurogranin: relation to cognition and neurodegeneration in Alzheimer’s disease. Brain. 2015;138:3373–85. Lista S, Hampel H. Synaptic degeneration and neurogranin in the pathophysiology of Alzheimer’s disease. Expert Rev Neurother. 2017;17:47–57. Therriault J, Schindler SE, Salvadó G, Pascoal TA, Benedet AL, Ashton NJ, et al. Biomarker-based staging of Alzheimer disease: rationale and clinical applications. Nat Rev Neurol. 2024;20:232–44. Camporesi E, Nilsson J, Brinkmalm A, Becker B, Ashton NJ, Blennow K, et al. Fluid Biomarkers for Synaptic Dysfunction and Loss. Biomark Insights. 2020;15. Van Hulle C, Jonaitis EM, Betthauser TJ, Batrla R, Wild N, Kollmorgen G, et al. An examination of a novel multipanel of CSF biomarkers in the Alzheimer’s disease clinical and pathological continuum. Alzheimer’s and Dementia. 2021;17:431–45. Salvadó G, Larsson V, Cody KA, Cullen NC, Jonaitis EM, Stomrud E, et al. Optimal combinations of CSF biomarkers for predicting cognitive decline and clinical conversion in cognitively unimpaired participants and mild cognitive impairment patients: A multi-cohort study. Alzheimer’s and Dementia. 2023;19:2943–55. Bereczki E, Bogstedt A, Höglund K, Tsitsi P, Brodin L, Ballard C, et al. Synaptic proteins in CSF relate to Parkinson’s disease stage markers. NPJ Parkinsons Dis. 2017;3. Wunderlich P, Glebov K, Kemmerling N, Tien NT, Neumann H, Walter J. Sequential proteolytic processing of the triggering receptor expressed on myeloid cells-2 (TREM2) protein by ectodomain shedding and γ-secretase-dependent intramembranous cleavage. J Biol Chem. 2013;288:33027–36. Schlepckow K, Morenas-Rodríguez E, Hong S, Haass C. Stimulation of TREM2 with agonistic antibodies-an emerging therapeutic option for Alzheimer’s disease. Lancet Neurol. 2023;22:1048–60. Ewers M, Franzmeier N, Suárez-Calvet M, Morenas-Rodriguez E, Angel M, Caballero A. Increased soluble TREM2 in cerebrospinal fluid is associated with reduced cognitive and clinical decline in Alzheimer’s disease for the Alzheimer’s Disease Neuroimaging Initiative. Sci Transl Med. 2019;11:6221. Pelkmans W, Shekari M, Brugulat-Serrat A, Sánchez-Benavides G, Minguillón C, Fauria K, et al. Astrocyte biomarkers GFAP and YKL-40 mediate early Alzheimer’s disease progression. Alzheimers Dement. 2024;20:483–93. Michetti F, Clementi ME, Di Liddo R, Valeriani F, Ria F, Rende M, et al. The S100B Protein: A Multifaceted Pathogenic Factor More Than a Biomarker. Int J Mol Sci. 2023;24:9605. Nishiyama H, Knö Pfel † T, Endo S, Itohara S. Glial protein S100B modulates long-term neuronal synaptic plasticity. Proc Natl Acad Sci U S A. 99:4037–42. Johnson SC, Koscik RL, Jonaitis EM, Clark LR, Mueller KD, Berman SE, et al. The Wisconsin Registry for Alzheimer’s Prevention: A review of findings and current directions. Alzheimer’s and Dementia: Diagnosis, Assessment and Disease Monitoring. 2018;10:130–42. Molinuevo JL, Gramunt N, Gispert JD, Fauria K, Esteller M, Minguillon C, et al. The ALFA project: A research platform to identify early pathophysiological features of Alzheimer’s disease. Alzheimer’s and Dementia: Translational Research and Clinical Interventions. 2016;2:82–92. Johnson SC, Suárez-Calvet M, Suridjan I, Minguillón C, Gispert JD, Jonaitis E, et al. Identifying clinically useful biomarkers in neurodegenerative disease through a collaborative approach: the NeuroToolKit. Alzheimers Res Ther. 2023;15. Argiris G, Akinci M, Peña-Gómez C, Palpatzis E, Garcia-Prat M, Shekari M, et al. Data-driven CSF biomarker profiling: imaging and clinical outcomes in a cohort at risk of Alzheimer’s disease. Alzheimer’s Research and Therapy . 2024;16:274. Soldan A, Pettigrew C, Cai Q, Wang MC, Moghekar AR, O’Brien RJ, et al. Hypothetical preclinical Alzheimer disease groups and longitudinal cognitive change. JAMA Neurol. 2016;73:698–705. Milà-Alomà M, Salvadó G, Gispert JD, Vilor-Tejedor N, Grau-Rivera O, Sala-Vila A, et al. Amyloid beta, tau, synaptic, neurodegeneration, and glial biomarkers in the preclinical stage of the Alzheimer’s continuum. Alzheimers Dement. 2020;16:1358–71. Salvadó G, Shekari M, Falcon C, Operto GDS, Milà-Alomà M, Sánchez-Benavides G, et al. Brain alterations in the early Alzheimer’s continuum with amyloid-β, tau, glial and neurodegeneration CSF markers. Brain Commun. 2022;4::fcac134. doi: 10.1093/braincomms/fcac134. Eusebi P, Giannandrea D, Biscetti L, Abraha I, Chiasserini D, Orso M, et al. Diagnostic utility of cerebrospinal fluid α-synuclein in Parkinson’s disease: A systematic review and meta-analysis. Movement Disorders. 2017;32:1389–400. Bonomi CG, Assogna M, Di Donna MG, Bernocchi F, De Lucia V, Nuccetelli M, et al. Cerebrospinal Fluid sTREM-2, GFAP, and β-S100 in Symptomatic Sporadic Alzheimer’s Disease: Microglial, Astrocytic, and APOE Contributions Along the Alzheimer’s Disease Continuum. Journal of Alzheimer’s Disease. 2023;92:1385–97. Steiner J, Bogerts B, Schroeter ML, Bernstein HG. S100B protein in neurodegenerative disorders. Clin Chem Lab Med. 2011. p. 409–24. Baecker J, Wartchow K, Sehm T, Ghoochani A, Buchfelder M, Kleindienst A. Treatment with the Neurotrophic Protein S100B Increases Synaptogenesis after Traumatic Brain Injury. J Neurotrauma. 2020;37:1097–107. Selnes P, Stav AL, Johansen KK, Bjørnerud A, Coello C, Auning E, et al. Impaired synaptic function is linked to cognition in Parkinson’s disease. Ann Clin Transl Neurol. 2017;4:700–13. Kester MI, Teunissen CE, Crimmins DL, Herries EM, Ladenson JKH, Scheltens P, et al. Neurogranin as a cerebrospinal fluid biomarker for synaptic loss in symptomatic Alzheimer disease. JAMA Neurol. 2015;72:1275–80. Salvadó G, Milà-Alomà M, Shekari M, Minguillon C, Fauria K, Niñerola-Baizán A, et al. Cerebral amyloid-β load is associated with neurodegeneration and gliosis: Mediation by p-tau and interactions with risk factors early in the Alzheimer’s continuum. Alzheimer’s and Dementia. 2021;17:788–800. Nilsson J, Gobom J, Sjödin S, Brinkmalm G, Ashton NJ, Svensson J, et al. Cerebrospinal fluid biomarker panel for synaptic dysfunction in Alzheimer’s Disease. Alzheimer’s and Dementia: Diagnosis, Assessment and Disease Monitoring. 2021;13. Kamondi A, Grigg-Damberger M, Löscher W, Tanila H, Horvath AA. Epilepsy and epileptiform activity in late-onset Alzheimer disease: clinical and pathophysiological advances, gaps and conundrums. Nat Rev Neurol. 2024;20:162–82. Rachmian N, Medina S, Cherqui U, Akiva H, Deitch D, Edilbi D, et al. Identification of senescent, TREM2-expressing microglia in aging and Alzheimer’s disease model mouse brain. Nat Neurosci. 2024;27:1116–24. Rueda‐Carrasco J, Sokolova D, Lee S, Childs T, Jurčáková N, Crowley G, et al. Microglia‐synapse engulfment via PtdSer‐TREM2 ameliorates neuronal hyperactivity in Alzheimer’s disease models. EMBO J. 2023;42:e113246. Rim C, You MJ, Nahm M, Kwon MS. Emerging role of senescent microglia in brain aging-related neurodegenerative diseases. Transl Neurodegener. 2024;13:10. Das M, Mao W, Voskobiynyk Y, Necula D, Lew I, Petersen C, et al. Alzheimer risk-increasing TREM2 variant causes aberrant cortical synapse density and promotes network hyperexcitability in mouse models. Neurobiol Dis. 2023;186:106263. Tzioras M, Daniels MJD, Davies C, Baxter P, King D, McKay S, et al. Human astrocytes and microglia show augmented ingestion of synapses in Alzheimer’s disease via MFG-E8. Cell Rep Med. 2023;4:101175. Rim C, You MJ, Nahm M, Kwon MS. Emerging role of senescent microglia in brain aging-related neurodegenerative diseases. Transl Neurodegener. 2024;13. Dejanovic B, Sheng M, Hanson JE. Targeting synapse function and loss for treatment of neurodegenerative diseases. Nat Rev Drug Discov. Nature Research; 2024. p. 23–42. Additional Declarations Competing interest reported. M.I.M.-G has received in the past 36mo speaker fees by Almirall. E.M.-R. has given lectures and symposia sponsored by KRKA Farmaceutica SL and Laboratorios Esteve SA. M.S.-C. has received in the past 36mo consultancy/speaker fees (paid to the institution) from by Almirall, Eli Lilly, Quanterix, Novo Nordisk, and Roche Diagnostics. He has received consultancy fees or served on advisory boards (paid to the institution) of Eli Lilly, Grifols, Novo Nordisk, and Roche Diagnostics. He was granted a project and is a site investigator of a clinical trial (funded to the institution) by Roche Diagnostics. In-kind support for research (to the institution) was received from ADx Neurosciences, Alamar Biosciences, ALZPath, Avid Radiopharmaceuticals, Eli Lilly, Fujirebio, Janssen Research & Development, Meso Scale Discovery, and Roche Diagnostics; MS-C did not receive any personal compensation from these organizations or any other for-profit organization. H.Z. has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp & Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of MicThera (outside submitted work). S.C.-J. serves as a consultant to Eli Lilly, Merck, AlzPath and Enigma Biomedical. G.K. is a full-time employee of Roche Diagnostics GmbH, Penzberg, Germany. CQ-R is a full-time employee of Roche Diagnostics International Ltd, Rotkreuz, Switzerland. Supplementary Files Supplementarymaterial.docx Cite Share Download PDF Status: Published Journal Publication published 17 Oct, 2025 Read the published version in Molecular Neurodegeneration → Version 1 posted Editorial decision: Revision requested 20 Aug, 2025 Reviews received at journal 20 Aug, 2025 Reviews received at journal 04 Aug, 2025 Reviewers agreed at journal 21 Jul, 2025 Reviewers agreed at journal 17 Jul, 2025 Reviewers invited by journal 14 Jul, 2025 Editor assigned by journal 10 Jul, 2025 Submission checks completed at journal 27 Jun, 2025 First submitted to journal 26 Jun, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6982788","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":485301040,"identity":"ec4d92ba-132d-4d4d-8170-f73570e0e04f","order_by":0,"name":"Mariana I. Muñoz-García","email":"","orcid":"","institution":"Hospital Universitario 12 de Octubre","correspondingAuthor":false,"prefix":"","firstName":"Mariana","middleName":"I.","lastName":"Muñoz-García","suffix":""},{"id":485301041,"identity":"55f1a9aa-e70d-4abf-ae7f-11dfef8ebd53","order_by":1,"name":"Yuetiva Deming","email":"","orcid":"","institution":"University of Wisconsin, University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Yuetiva","middleName":"","lastName":"Deming","suffix":""},{"id":485301042,"identity":"c25ab9e9-a8a6-457a-b47e-d6c34952505e","order_by":2,"name":"Ferran Lugo Hernández","email":"","orcid":"","institution":"Pasqual Maragall Foundation","correspondingAuthor":false,"prefix":"","firstName":"Ferran","middleName":"Lugo","lastName":"Hernández","suffix":""},{"id":485301043,"identity":"590dd524-06a2-4df3-a631-bbbd204b33e5","order_by":3,"name":"Sterling Johnson","email":"","orcid":"","institution":"University of Wisconsin, University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Sterling","middleName":"","lastName":"Johnson","suffix":""},{"id":485301044,"identity":"88e319ca-5136-468f-81e5-e0a34652d232","order_by":4,"name":"Sanjay Asthana","email":"","orcid":"","institution":"University of Wisconsin, University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Sanjay","middleName":"","lastName":"Asthana","suffix":""},{"id":485301045,"identity":"562b6b4c-fc91-43b5-9c6d-318b5139e4d7","order_by":5,"name":"Gwendlyn Kollmorgen","email":"","orcid":"","institution":"Roche Diagnostics GmbH","correspondingAuthor":false,"prefix":"","firstName":"Gwendlyn","middleName":"","lastName":"Kollmorgen","suffix":""},{"id":485301046,"identity":"ee73c31b-cfb2-42b8-b89a-4004076e4b36","order_by":6,"name":"Clara Quijano-Rubio","email":"","orcid":"","institution":"Roche Diagnostics International Ltd","correspondingAuthor":false,"prefix":"","firstName":"Clara","middleName":"","lastName":"Quijano-Rubio","suffix":""},{"id":485301047,"identity":"41592a1d-9c51-43ea-be40-0cba59eb1aab","order_by":7,"name":"Cynthia Carlsson","email":"","orcid":"","institution":"University of Wisconsin, University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Cynthia","middleName":"","lastName":"Carlsson","suffix":""},{"id":485301048,"identity":"0d4df7a0-3c50-46eb-999a-29cc7fd85f9f","order_by":8,"name":"Ozioma C. Okonkwo","email":"","orcid":"","institution":"University of Wisconsin, University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Ozioma","middleName":"C.","lastName":"Okonkwo","suffix":""},{"id":485301049,"identity":"f69252bc-438b-42eb-a9cc-7a1078643964","order_by":9,"name":"David Pérez-Martinez","email":"","orcid":"","institution":"Hospital Universitario 12 de Octubre","correspondingAuthor":false,"prefix":"","firstName":"David","middleName":"","lastName":"Pérez-Martinez","suffix":""},{"id":485301050,"identity":"ca8d8f3c-8624-4e50-b2ec-26e03d54e6e1","order_by":10,"name":"Alberto Villarejo-Galende","email":"","orcid":"","institution":"Hospital Universitario 12 de Octubre","correspondingAuthor":false,"prefix":"","firstName":"Alberto","middleName":"","lastName":"Villarejo-Galende","suffix":""},{"id":485301051,"identity":"450afe09-5bfe-40a1-8622-f5e6a4fd39f0","order_by":11,"name":"Kaj Blennow","email":"","orcid":"","institution":"University of Gothenburg","correspondingAuthor":false,"prefix":"","firstName":"Kaj","middleName":"","lastName":"Blennow","suffix":""},{"id":485301052,"identity":"d162b163-208c-41d0-addf-09f98a8a8dc1","order_by":12,"name":"Marc Suárez-Calvet","email":"","orcid":"","institution":"Pasqual Maragall Foundation","correspondingAuthor":false,"prefix":"","firstName":"Marc","middleName":"","lastName":"Suárez-Calvet","suffix":""},{"id":485301053,"identity":"4f4c4e27-97d2-4923-a755-9921a3e9d057","order_by":13,"name":"Henrik Zetterberg","email":"","orcid":"","institution":"University of Gothenburg","correspondingAuthor":false,"prefix":"","firstName":"Henrik","middleName":"","lastName":"Zetterberg","suffix":""},{"id":485301054,"identity":"382ff450-b4e9-4b15-bdfd-2ac1d0e786a5","order_by":14,"name":"Barbara B Bendlin","email":"","orcid":"","institution":"University of Wisconsin, University of Wisconsin-Madison","correspondingAuthor":false,"prefix":"","firstName":"Barbara","middleName":"B","lastName":"Bendlin","suffix":""},{"id":485301055,"identity":"140a8f1d-b09b-428e-ae3a-516b66c9fdb9","order_by":15,"name":"Estrella Morenas-Rodríguez","email":"data:image/png;base64,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","orcid":"","institution":"Hospital Universitario 12 de Octubre","correspondingAuthor":true,"prefix":"","firstName":"Estrella","middleName":"","lastName":"Morenas-Rodríguez","suffix":""}],"badges":[],"createdAt":"2025-06-26 11:08:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6982788/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6982788/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s13024-025-00901-5","type":"published","date":"2025-10-17T15:57:17+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":87043579,"identity":"e3540bc6-ca35-4ee1-a9e3-1af99a1b5af3","added_by":"auto","created_at":"2025-07-18 14:18:25","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":96249,"visible":true,"origin":"","legend":"\u003cp\u003eCross-sectional associations between biomarkers and sTREM2.\u003c/p\u003e\n\u003cp\u003eAdjusted by age, gender, AΒ42 and p-tau.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-6982788/v1/315e88a5007db6c0dded3e2b.png"},{"id":87042358,"identity":"43304ea8-be0b-4124-861b-7a7538ae6fb1","added_by":"auto","created_at":"2025-07-18 14:10:25","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":116737,"visible":true,"origin":"","legend":"\u003cp\u003eLongitudinal associations between baseline biomarkers and sTREM2.\u003c/p\u003e\n\u003cp\u003eModels are adjusted by age, gender, AB42 and p-tau.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eNeurogranin\u003c/em\u003e: Beta-coefficient for interaction with time = -0.04, p= 0.0001 (continuous values), Beta-coefficient for interaction with time = -0.02, p= 0.006 (comparing \u0026gt;median vs \u0026lt;median). \u003cem\u003eAlfa-synuclein\u003c/em\u003e: Beta-coefficient for interaction with time = -0.03, p= 0.002 (continuous values), Beta-coefficient for interaction with time = -0.02, p= 0.01 (comparing \u0026gt;median vs \u0026lt;median). \u003cem\u003eS100b\u003c/em\u003e: Beta-coefficient for interaction with time = 0.02, p= 0.32 (continuous values), Beta-coefficient for interaction with time = 0.02, p= 0.03 (comparing \u0026gt;median vs \u0026lt;median). \u003cem\u003eGFAP\u003c/em\u003e: Beta-coefficient for interaction with time = -0.004, p= 0.73 (continuous values). Beta-coefficient for interaction with time = -0.01, p= 0.12 (comparing \u0026gt;median vs \u0026lt;median).\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-6982788/v1/38103ea3215c0845eded9f15.png"},{"id":87042360,"identity":"7d2ffc35-a113-4268-b30a-e29bff8bf982","added_by":"auto","created_at":"2025-07-18 14:10:25","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":107826,"visible":true,"origin":"","legend":"\u003cp\u003eLongitudinal associations between baseline sTREM2 and biomarkers.\u003c/p\u003e\n\u003cp\u003eModels are adjusted by age, gender, AB42 and p-tau. \u003cem\u003eNeurogranin\u003c/em\u003e: \"Beta-coefficient for interaction with time = -0.03, p= 0.0001 (continuous values). Beta-coefficient for interaction with time = -0.01, p= 0.03 (comparing \u0026gt;median vs \u0026lt;median). \u003cem\u003e⍺-syn\u003c/em\u003e: Beta-coefficient for interaction with time = -0.03, p= 0.08 (continuous values). Beta-coefficient for interaction with time = -0.01, p= 0.33 (comparing \u0026gt;median vs \u0026lt;median). \u003cem\u003eS100b\u003c/em\u003e: Beta-coefficient for interaction with time = -0.006, p= 0.36 (continuous values). Beta-coefficient for interaction with time = -0.003, p= 0.55 (comparing \u0026gt;median vs \u0026lt;median). \u003cem\u003eGFAP\u003c/em\u003e: Beta-coefficient for interaction with time = -0.006, p= 0.40 (continuous values). Beta-coefficient for interaction with time = -0.004, p= 0.46 (comparing \u0026gt;median vs \u0026lt;median).\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-6982788/v1/cea91a0f6837f1b6da7ae8c7.png"},{"id":93956092,"identity":"ed8461d1-db8b-4059-a8cf-2f047100ae06","added_by":"auto","created_at":"2025-10-20 16:10:23","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1580291,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6982788/v1/ac2daccf-506a-45cf-9894-bb421759ad88.pdf"},{"id":87042363,"identity":"d846bfa3-dd2a-493d-ad4f-e8ef9511cc72","added_by":"auto","created_at":"2025-07-18 14:10:25","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":280189,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterial.docx","url":"https://assets-eu.researchsquare.com/files/rs-6982788/v1/1a5261da388d57da4a49dbd9.docx"}],"financialInterests":"Competing interest reported. M.I.M.-G has received in the past 36mo speaker fees by Almirall. \nE.M.-R. has given lectures and symposia sponsored by KRKA Farmaceutica SL and Laboratorios Esteve SA.\nM.S.-C. has received in the past 36mo consultancy/speaker fees (paid to the institution) from by Almirall, Eli Lilly, Quanterix, Novo Nordisk, and Roche Diagnostics. He has received consultancy fees or served on advisory boards (paid to the institution) of Eli Lilly, Grifols, Novo Nordisk, and Roche Diagnostics. He was granted a project and is a site investigator of a clinical trial (funded to the institution) by Roche Diagnostics. In-kind support for research (to the institution) was received from ADx Neurosciences, Alamar Biosciences, ALZPath, Avid Radiopharmaceuticals, Eli Lilly, Fujirebio, Janssen Research \u0026 Development, Meso Scale Discovery, and Roche Diagnostics; MS-C did not receive any personal compensation from these organizations or any other for-profit organization.\nH.Z. has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp \u0026 Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of MicThera (outside submitted work).\nS.C.-J. serves as a consultant to Eli Lilly, Merck, AlzPath and Enigma Biomedical.\nG.K. is a full-time employee of Roche Diagnostics GmbH, Penzberg, Germany. \nCQ-R is a full-time employee of Roche Diagnostics International Ltd, Rotkreuz, Switzerland.","formattedTitle":"Synaptic dysfunction and glial activation markers throughout aging and early neurodegeneration: a longitudinal CSF biomarker-based study","fulltext":[{"header":"Background","content":"\u003cp\u003eSynaptic dysfunction and glial reactivity play crucial roles in the early stages and progression of neurodegenerative diseases, as well as throughout the aging process [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Experimental evidence suggests that microglia and astrocytes play critical roles in regulating synaptic plasticity during both physiological aging and disease [\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Microglial-derived proteins such as TGF-β1 and C1q have been identified as crucial signals that impact synaptic protein homeostasis and influence synaptic maintenance in the aging brain [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Astrocytes contribute through neurotransmitter clearance, ion balance regulation, and gliotransmitter release, which affect short- and long-term synaptic plasticity [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Thus, both glial cell types are well-positioned to sense early disruptions in synaptic activity and potentially interact with synaptic failure throughout both aging and disease [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eMicroglial and astroglial responses to early protein aggregation have also been identified as key processes in neurodegenerative diseases [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Single-cell sequencing technologies have identified dynamic microglial populations transitioning from homeostatic to disease-associated profiles in response to initial protein aggregation [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Loss-of-function genetic variants affecting the microglial protein triggering receptor expressed on myeloid cells 2 (TREM2) increase AD risk by preventing microglia from transitioning to a protective disease-associated state [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eSimilarly, disease-associated astrocytes are often found surrounding pathological protein aggregates in various neurodegenerative diseases and appear to be induced by activated microglia through pro-inflammatory cytokines [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. This activation leads to a loss of astrocytic homeostatic roles, including maintaining synaptic functionality, as shown in animal-based studies [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. These findings suggest active synapse regulation by microglia and astroglia in neurodegeneration [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Nonetheless, the association between glial responses and synaptic dysfunction remains largely unexplored in human cohorts in the framework of aging and neurodegeneration.\u003c/p\u003e\u003cp\u003eSynaptic function can be assessed in live human cohorts by measuring synapse-related biomarkers in cerebrospinal fluid (CSF). Neurogranin is a postsynaptic protein widely studied in CSF. Reduced CSF levels of neurogranin have been described in Parkinson disease (PD), dementia with Lewy bodies (DLB) and preclinical Alzheimer\u0026rsquo;s disease (AD) [\u003cspan additionalcitationids=\"CR18\" citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. However, most studies in AD report elevated CSF neurogranin across the continuum of disease, as well as correlation with CSF p-tau, likely reflecting progressive dendritic pathology and/or neuronal hyperactivity [\u003cspan additionalcitationids=\"CR21 CR22\" citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e]. CSF alpha-synuclein (α-syn), a crucial presynaptic protein, has different changes in concentration depending on the disease context, stage, and analytical method [\u003cspan additionalcitationids=\"CR25\" citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Discrepancies in CSF α-syn levels highlight the multifactorial nature of α-syn biology, which encompasses synaptic dysfunction, aggregation, neurodegeneration, and axonal remodeling [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan additionalcitationids=\"CR28\" citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Thus, these synaptic proteins are probably appropriate synapse-related markers during healthy aging or early stages of neurodegenerative processes but should be interpreted carefully in the context of advanced neurodegeneration.\u003c/p\u003e\u003cp\u003eGlial function can also be studied in humans through biomarkers in biofluids, offering a translational bridge between experimental findings and clinical observations. The soluble cleavage product of the microglial protein TREM2 (sTREM2) serves as a marker of the cell-autonomous microglial activation response [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Longitudinal biomarker-based data from autosomal-dominant AD suggests that the main trigger of TREM2-dependent microglial response is initial Aβ aggregation, decades before the first symptoms appear [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Higher baseline levels and longitudinal increase of sTREM2 in CSF are associated with slower cognitive decline and reduced amyloid accumulation in AD patients, which supports the protective role of microglia [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eAdditionally, astroglial reactivity marker GFAP has been proposed to mediate the transition between soluble and insoluble Aβ and represent a possible connection between amyloidosis and tau-related neurodegeneration in sporadic AD [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. Another important astroglial activation marker measurable in CSF is S100 calcium-binding protein B (S100B). This marker is proposed to play a role in modulating synaptic plasticity and microglial response [\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. These biomarkers are useful for monitoring the dynamic roles of microglia and astroglia in clinical cohorts throughout aging and disease stages.\u003c/p\u003e\u003cp\u003eUnderstanding the interplay between synaptic dysfunction and glial responses is essential for characterizing the biological underpinnings of cognitive decline in aging and neurodegenerative diseases. These processes reflect key mechanisms underlying cognitive impairment and represent some of the earliest events driving disease progression, potentially opening new windows for therapeutic intervention. Nonetheless, the temporal relationships among presynaptic (α-syn), postsynaptic (neurogranin), microglial (sTREM2), and astroglial (GFAP, S100B) markers remain poorly understood, particularly in asymptomatic individuals, as very few studies directly investigated this association.\u003c/p\u003e\u003cp\u003eIn this study, we leverage longitudinal CSF biomarker data from two independent cohorts of cognitively unimpaired, late-middle-aged individuals to investigate how synaptic and glial signals crosstalk with TREM2-dependent microglial activation. By integrating cross-sectional and longitudinal analyses across populations with differing AD risk profiles, we aim to uncover early synapse-to-glia signaling dynamics that distinguish physiological aging from the preclinical phase of neurodegeneration\u0026mdash;ultimately informing the development of precision strategies for early intervention.\u003c/p\u003e"},{"header":"Materials and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e\u003ch2\u003ePatient cohorts\u003c/h2\u003e\u003cp\u003eWe conducted a longitudinal observational study using CSF biomarker data from two independent cohorts of cognitively unimpaired adults, aiming to assess associations between synaptic and glial biomarkers and microglial activation over time. The first cohort of participants stemmed from the Wisconsin Registry for Alzheimer\u0026rsquo;s Prevention (WRAP) study; an observational cohort first established in 2001. The WRAP study initially enrolled participants at midlife with a mean age of 54 and parental history of probable AD dementia and thus, at risk for late onset dementia. Since 2004, it also included participants without parental history of dementia to better understand its role in the risk of dementia. Genetic and clinical data was gathered initially, and subjects were followed longitudinally with neuropsychological evaluation, self-reported medical and lifestyle data, laboratory tests, and optional lumbar puncture (LP) at different time points (approximately every two years). Further details about the cohort are discussed elsewhere [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. This study uses a subset of 239 WRAP participants, with available CSF samples, 116 of them with available longitudinal data, out of the 1561 total participants[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eWe repeated analyses in a confirmation cohort, the ALFA\u0026thinsp;+\u0026thinsp;cohort, a longitudinal observational cohort nested within the ALFA (for ALzheimer and FAmilies) parent cohort. The ALFA study was established between 2013 and 2014 and recruited 2,743 cognitively unimpaired individuals, primarily first-degree descendants of patients with sporadic AD, aged 45 to 75 years. Participants were extensively characterized at baseline, including sociodemographic, clinical, lifestyle, and cognitive measures, with additional data collected on modifiable risk factors and \u003cem\u003eAPOE\u003c/em\u003e genotype. Within the ALFA cohort, a subset was selected for the ALFA\u0026thinsp;+\u0026thinsp;study based on risk profile (\u003cem\u003eAPOE\u003c/em\u003e and family history status), and 400 of those participants were available for cross-sectional analyses. ALFA\u0026thinsp;+\u0026thinsp;involves more detailed longitudinal phenotyping, including fluid biomarker collection. Baseline visits occurred between 2016 and 2019, with follow-up assessments every three years. In the present study, at baseline, 15 blood-based biomarkers were analyzed, and during the second wave (V2, begun in 2019), CSF biomarkers such as Aβ42, p-tau, t-tau, GFAP, neurogranin, S100, α-syn, sTREM2, and others were measured [\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. A subset of 259 had available longitudinal data at the time of analyses.\u003c/p\u003e\u003c/div\u003e\n\u003ch3\u003eCSF biomarker quantification\u003c/h3\u003e\n\u003cp\u003eSynapse-related biomarkers including neurogranin and α-syn, and astroglial markers S100B and GFAP, as well as Aβ40 were measured using the NeuroToolKit (NTK) in both the WRAP and ALFA\u0026thinsp;+\u0026thinsp;cohorts, as a panel of exploratory prototype assays [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. The specific cleaved sTREM2 isoform was measured in the CSF of participants included in the WRAP cohort by an in-house immunoassay in the MSD platform as previously reported [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Total sTREM2 was quantified in the ALFA\u0026thinsp;+\u0026thinsp;cohort by the NTK [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. AD core biomarkers including Aβ42 and total tau (t-tau) and tau phosphorylated at threonine 181 (p-tau) in CSF were quantified by the commercially available Elecsys\u0026reg; immunoassays, as described elsewhere [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. In the WRAP cohort, we defined positivity in the AT classification as Aβ42/Aβ40 ratio\u0026thinsp;\u0026lt;\u0026thinsp;0.046 and p-tau\u0026thinsp;\u0026gt;\u0026thinsp;24.8 pg/mL; in the ALFA\u0026thinsp;+\u0026thinsp;cohort, as Aβ42/Aβ40 ratio\u0026thinsp;\u0026lt;\u0026thinsp;0.071 and p-tau\u0026thinsp;\u0026gt;\u0026thinsp;24 pg/mL. Biomarkers Aβ42, Aβ40, pTau, t-tau and S100B were measured using the Cobas\u0026reg; e 601 analyzer, and the remaining on a Cobas\u0026reg; e 411 analyzer (both Roche Diagnostics International Ltd) [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. All NTK measurements for both cohorts were performed in singlicate at the Clinical Neurochemistry Laboratory at the University of Gothenburg (Sweden).\u003c/p\u003e\u003cdiv id=\"Sec5\" class=\"Section2\"\u003e\u003ch2\u003eStatistical analysis\u003c/h2\u003e\u003cp\u003eOnly participants with complete data (all measurements for biomarkers and covariate data for each described model) were included in the analyses; no imputation was performed. We first tested for normality of the distribution for each biomarker using the Shapiro test. CSF Aβ42, Aβ42/Aβ40 ratio, p-tau, t-tau, sTREM2, neurogranin, α-syn, S100B, and GFAP did not follow a normal distribution and were thus log10-transformed. To describe the data, we stratified participants according to \u003cem\u003eAPOE\u003c/em\u003e status, medians of Aβ42/Aβ40 ratio and p-tau, and we used χ2 tests for categorical variables, and t-student or ANOVA for continuous variables to compare groups. We stratified the cohort according to medians of Aβ42/Aβ40 ratio and p-tau to better approach the contribution of the earliest AD-related pathological changes, since this cohort was composed of healthy participants with a low percentage of amyloid and p-tau positivity. This methodology is supported by findings showing differential longitudinal cognitive profiles based on preclinical AD-biomarker changes, when dividing the Aβ42 and p-tau values according to cohort-specific tertiles and medians [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eFor cross-sectional analysis, we calculated partial correlations adjusted by age across all biomarkers with the Pearson method. We then performed linear regression analysis to study the association between sTREM2 and synaptic function biomarkers, using TREM2 both as the main independent variable and the dependent outcome variable. For each analysis, we use two linear regression models: Model 1, adjusted for age and gender, and Model 2, further adjusted for baseline Aβ42 and p-tau levels. Analyses were conducted in the entire cohort and stratified by subgroups based on the median Aβ42/Aβ40 ratio, p-tau levels, Aβ or p-tau marker positivity (AT classification), and \u003cem\u003eAPOE\u003c/em\u003e carriage status, to better capture different preclinical stages of the AD continuum. Furthermore, we repeated the models including interaction terms between the independent biomarker and Aβ42 and p-tau, as continuous variables. The overall aim was to evaluate whether these associations were influenced by an underlying initial AD pathological process.\u003c/p\u003e\u003cp\u003eFor longitudinal analyses, we performed linear mixed effects models with random intercepts to account for within-subject variability over time, based on longitudinal data from 116 participants in the WRAP cohort and 259 in the ALFA\u0026thinsp;+\u0026thinsp;cohort. The WRAP cohort participants had mainly one follow-up visit (20 had 3 and 6 had 4 visits), while all participants in the ALFA\u0026thinsp;+\u0026thinsp;cohort had only one follow-up visit. For this reason, the statistical model included only random intercepts (not random slopes) to avoid convergence issues, which allowed to model between-subject variability without overfitting. Linear mixed-effects models were implemented using the lme4 package. Univariate models were used to assess the effect of baseline biomarkers (predictor) on the longitudinal change of the outcome biomarker. We included an interaction term between the baseline biomarker and time to understand how the relationship between biomarkers changed over time. The time variable was modeled as a continuous variable centered at baseline (time of first LP).\u003c/p\u003e\u003cp\u003eWe used sTREM2 both as the outcome biomarker and as the predictor for each synaptic biomarker. We adjusted the models by age, gender, and continuous values of Aβ42, and p-tau. We also calculated the interaction between time and baseline biomarker levels divided by medians. We performed sensitivity analysis by performing analysis in previously described subgroups. Given the exploratory nature of the study, p-values were considered nominally significant at p\u0026thinsp;\u0026lt;\u0026thinsp;0.05 without correction for multiple comparisons. All statistical analyses were performed with R software (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://www.r-project.org/\u003c/span\u003e\u003cspan address=\"http://www.r-project.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and RStudio (last updated version 2024.12.0\u0026thinsp;+\u0026thinsp;467).\u003c/p\u003e\u003c/div\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec7\" class=\"Section2\"\u003e\u003ch2\u003eWRAP cohort\u003c/h2\u003e\u003cp\u003eDemographic and baseline biomarker characteristics of WRAP participants are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. Participants with both an Aβ42/Aβ40 ratio below the median and p-tau levels at or above the median were older at the time of LP and had a higher frequency of \u003cem\u003eAPOE\u003c/em\u003e ε4 carriers, compared to other subgroups. No differences were observed in gender, parental history of dementia, ethnicity, or MMSE score. As expected, groups defined by the median of Aβ42/Aβ40 ratio differed in percentage of amyloid positivity according to Aβ42/Aβ40 ratio and p-tau/Aβ42 ratio, as well as tau positivity according to pre-established cut-offs. Groups defined by median p-tau, also differed in percentage of amyloid positivity according to Aβ42/Aβ40 ratio and p-tau/Aβ42 ratio.\u003c/p\u003e\u003cp\u003eThe biomarker profiles adjusted by age are also shown in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. The group of participants with an Aβ42/Aβ40 ratio below the median showed biomarkers congruent with first stages of amyloid aggregation: lower Aβ42, higher t-tau and p-tau, lower Aβ42/Aβ40 ratio than participants with an Aβ42/Aβ40 ratio above the median. When stratifying by p-tau median, participants with p-tau levels above the median also had a biomarker profile suggestive of first stages within the AD continuum, except for having higher baseline Aβ42, but this was compensated with a significantly lower Aβ42/Aβ40 ratio. Cross-sectionally, participants with p-tau levels above the median had higher levels of sTREM2, GFAP, S100B, neurogranin and ⍺-syn than participants with p-tau levels below the median (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). In contrast, levels of the studied proteins did not significantly differ between groups stratified according to Aβ42/Aβ40 ratio. Partial correlations adjusted by age between studied markers in the entire cohort are summarized in \u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eDemographics and baseline biomarkers in the WRAP (Wisconsin's Registry for Alzheimer Prevention) cohort\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"8\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eOverall\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eAβ42/Aβ40 ratio\u0026thinsp;\u0026ge;\u0026thinsp;median (2)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eAβ42/Aβ40 ratio\u0026thinsp;\u0026lt;\u0026thinsp;median\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e\u003cp\u003ep-tau\u0026thinsp;\u0026ge;\u0026thinsp;median (4)\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u003cp\u003ep-tau\u0026thinsp;\u0026lt;\u0026thinsp;median\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003en\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e239\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e120\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e119\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e120\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e119\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge at time of LP [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e61.4 (7.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e60.1 (7.47)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e62.8 (6.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e62.5 (7.30)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e60.4 (6.94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e0.021\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGener [Male (%)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e89 (37.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e45 (37.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e44 (37.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e1.000\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e39 (32.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e50 (42.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.165\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePH of dementia [Yes (%)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e177 (74.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e91 (75.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e86 (72.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.631\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e89 (74.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e88 (73.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e1.000\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEthnicity (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.454\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.556\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWhite\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e230 (96.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e115 (95.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e115 (96.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e116 (96.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e114 (95.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9 (3.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e5 (4.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4 (3.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4 (3.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4 (4.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eApoE \u0026#120518;4[Non-carrier (%)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e151 (63.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e91 (75.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e60 (50.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e67 (55.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e84 (70.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e0.026\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMMSE [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e29.3 (0.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e29.3 (0.89)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e29.3 (0.94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.518\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e29.29 (1.00)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e29.3 (0.82)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.902\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16.3 (2.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e16.1 (2.45)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16.4 (2.24)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.213\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e16.27 (2.35)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e16.2 (2.36)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.896\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmyloid + (%) (3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e46 (19.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0 (0.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e46 (38.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e41 (34.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e5 (4.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTau + (%) (5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e25 (10.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4 (3.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e21 (17.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e25 (20.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0 (0.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePositive ptau/ab42 ratio (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e35 (14.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0 (0.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e35 (29.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e31 (25.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4 (3.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003ep (adj by age)\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003ep (adj by age)\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAβ42 [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e881 (375)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1097 (337)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e663 (272)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1009 (419)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e751 (269)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAβ40 [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e14024 (4395)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e14494 (4072)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e13550 (4667)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.027\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e16995 (3570)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e11028 (2841)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAβ42/Aβ40 ratio [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.06 (0.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.08 (0.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.05 (0.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.06 (0.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.07 (0.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eT- tau [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e196 (63)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e183.3 (52.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e208 (77.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0374\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e246 (56.6)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e145 (25.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP-tau [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e17.3 (6.75)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15.9 (4.93)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e18.6 (7.99)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.0224\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e22.2 (6.12)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e12.3 (2.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ecsTREM2\u0026nbsp; [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.63 (2.88)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7.57 (2.78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7.69 (2.99)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.539\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8.69 (2.95)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e6.56 (2.37)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003esTREM2\u0026nbsp; [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7.78 (2.29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7.82 (2.18)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7.73 (2.41)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.152\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8.92 (2.34)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e6.62 (1.54)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNeurogranin [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e771 (298)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e741 (248)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e801 (339)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.263\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e986 (255)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e554 (138)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u0026#120514;- synuclein \u0026nbsp; [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e152 (64.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e149 (56.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e155 (71.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.860\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e193 (61.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e110 (30.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eS100B\u0026nbsp; [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1.16 (0.30)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.15 (0.29)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.17 (0.30)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.949\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1.20 (0.28)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1.12 (0.30)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.065\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGFAP [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e8.69 (3.03)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8.67 (3.22)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e8.72 (2.84)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.253\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e9.72 (3.14)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e7.66 (2.52)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"7\" nameend=\"c7\" namest=\"c1\"\u003e\u003cp\u003e(1) Other: American Indian or Alaska Native, Asian, Black or African American and other (2) Aβ42/Aβ40 median\u0026thinsp;=\u0026thinsp;0.067 (3) The cut-off for amyloid (A) positivity according to the Aβ42/Aβ40 ratio is 0.046. (4) P-tau median\u0026thinsp;=\u0026thinsp;15.94 pg/mL. (5) The cut-off for tau (T) positivity according to P-tau is 24.8 pg/mL. LP: lumbar puncture. PH: Parental history. MMSE: mini-mental state examination score\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colspan=\"1\" nameend=\"c8\" namest=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eIn cross-sectional analyses, we performed linear regression models adjusted for age and gender (Model 1), and then further adjusted for baseline Aβ42 and p-tau levels (Model 2), as well as after stratification by subgroups based on the median Aβ42/Aβ40 ratio, p-tau levels, Aβ or p-tau marker positivity (A/T classification), and \u003cem\u003eAPOE\u003c/em\u003e carriage status. These are summarized in \u003cb\u003eSupplementary Table\u0026nbsp;1.\u003c/b\u003e We found significant cross-sectional associations between sTREM2 and GFAP (β\u0026thinsp;=\u0026thinsp;0.26, p\u0026thinsp;=\u0026thinsp;0.0001), S100B (β\u0026thinsp;=\u0026thinsp;0.28, p\u0026thinsp;=\u0026thinsp;0.002), neurogranin (β\u0026thinsp;=\u0026thinsp;0.37, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), and ⍺-syn (β\u0026thinsp;=\u0026thinsp;0.37, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001), in the whole cohort using Model 1 (Supplementary Table\u0026nbsp;1). In contrast, after adjusting for Aβ42 and p-tau baseline levels (Model 2), we only found a trend for an association in the whole sample between sTREM2 and S100B (β\u0026thinsp;=\u0026thinsp;0.16, p\u0026thinsp;=\u0026thinsp;0.06), and between sTREM2 and ⍺-syn (β\u0026thinsp;=\u0026thinsp;0.16, p\u0026thinsp;=\u0026thinsp;0.09), indicating an influence of AD related markers on the previous associations. The results for Model 2 are shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. In subgroups stratified by AD biomarker profiles, we found a significant cross-sectional association between sTREM2 and ⍺-syn CSF levels in T\u0026thinsp;+\u0026thinsp;participants even after AD-related markers adjustment, indicating that the cross-sectional relationship is not influenced by AD-related biomarkers (β\u0026thinsp;=\u0026thinsp;0.83, p\u0026thinsp;=\u0026thinsp;0.02) shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB and Supplementary Table\u0026nbsp;1.\u003c/p\u003e\u003cp\u003eAdditionally, we found a significant association between sTREM2 and S100B in participants with Aβ42/Aβ40 ratio below the median (β\u0026thinsp;=\u0026thinsp;0.26, p\u0026thinsp;=\u0026thinsp;0.03) and in participants with p-tau levels above the median (β\u0026thinsp;=\u0026thinsp;0.28, p\u0026thinsp;=\u0026thinsp;0.01) (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC and Supplementary Table\u0026nbsp;1) after AD-related marker adjustment. This suggests that the association between sTREM2 and S100B is present in individuals with a biomarker profile indicative of first stages of an AD pathology and is not mediated by Aβ42 or p-tau levels. We did not find any other significant cross-sectional association between sTREM2 and neurogranin or GFAP in models adjusted by Aβ42 and p-tau. The cross-sectional associations between sTREM2 and studied biomarkers were not affected by the \u003cem\u003eAPOE\u003c/em\u003e \u0026#120518;4 allele carriage status, whether included as a covariate in regression models or assessed in stratified analyses (Supplementary Table\u0026nbsp;1). Furthermore, interaction terms were tested using continuous values of Aβ42 and p-tau; none were statistically significant and are not shown.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eAdjusted by age, gender, AΒ42 and p-tau.\u003c/p\u003e\u003cp\u003eThen, we examined whether baseline astroglial response and synapse-related markers influenced the longitudinal dynamics of sTREM2 using linear mixed models. Two models were applied: Model 1, adjusted for age and gender, and Model 2, further adjusted for baseline Aβ42 and p-tau levels. Among the 116 participants, follow-up data were available for 90 individuals with one visit, 20 with two visits, and 6 with three visits. The adjustment for baseline Aβ42 and p-tau did not alter the association between baseline astroglial and synaptic markers and the longitudinal change in sTREM2. Nevertheless, to evaluate the independent associations between astroglial and synaptic markers and the longitudinal change in sTREM2, we focused on the adjusted models (Model 2). A summary of these models is provided in \u003cb\u003eSupplementary Table\u0026nbsp;2.\u003c/b\u003e\u003c/p\u003e\u003cp\u003eConcerning the synapse-related markers, lower levels of baseline neurogranin (β-coefficient for interaction with time = -0.04, p\u0026thinsp;=\u0026thinsp;0.0002) and ⍺-synuclein (β-coefficient for interaction with time = -0.03, p\u0026thinsp;=\u0026thinsp;0.004) significantly predicted a larger subsequent longitudinal increase in sTREM2 CSF levels over time. When stratifying baseline neurogranin and ⍺-synuclein by their median values, we consistently observed that levels below the median were significantly associated with an increase in sTREM2 over time (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB). Regarding the astroglial markers GFAP and S100B, we only observed a significant association with baseline S100B when stratifying by its median, while no significant associations were found between baseline GFAP and the longitudinal change of sTREM2 levels. Participants with baseline S100B levels above the median showed a significantly greater subsequent longitudinal increase in CSF sTREM2 levels (β-coefficient for the interaction between time and S100B above median\u0026thinsp;=\u0026thinsp;0.02, p\u0026thinsp;=\u0026thinsp;0.03), as shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC and Supplementary Table\u0026nbsp;2. These longitudinal associations remained significant when stratifying by Aβ42/Aβ40 ratio medians, p-tau medians, amyloid or p-tau positivity cut-offs, or \u003cem\u003eAPOE\u003c/em\u003e \u0026#120576;4 carrier status (Supplementary Table\u0026nbsp;2).\u003c/p\u003e\u003cp\u003eFinally, we evaluated whether baseline sTREM2 levels influenced the subsequent longitudinal changes in synapse-related markers and astroglial response. After adjusting for Aβ42 and p-tau, we found that higher baseline sTREM2 levels were significantly associated with a diminished longitudinal increase in neurogranin over time (β-coefficient for interaction with time =-0.03, p\u0026thinsp;=\u0026thinsp;0.0001) (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). No other significant associations were observed between baseline sTREM2 levels and the longitudinal changes of ⍺-synuclein, S100B or, GFAP (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB, \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eC and \u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD).\u003c/p\u003e\u003c/div\u003e\u003cdiv id=\"Sec8\" class=\"Section2\"\u003e\u003ch2\u003eALFA\u0026thinsp;+\u0026thinsp;cohort\u003c/h2\u003e\u003cp\u003eTo replicate our findings in an independent sample, we applied the same analytic pipeline to the ALFA\u0026thinsp;+\u0026thinsp;cohort. Demographics of this cohort are shown in \u003cb\u003eTable\u0026nbsp;2\u003c/b\u003e. Demographically, the main differences between the cohorts were the proportion of \u003cem\u003eAPOE\u003c/em\u003e \u0026#120518;4 carriers (54% in ALFA\u0026thinsp;+\u0026thinsp;vs. 37% WRAP), which influenced the proportion of A\u0026thinsp;+\u0026thinsp;participants (33.8% vs. 19%), but not T+ (11.9% vs. 10.5%). Subgroup analysis revealed that participants with an Aβ42/Aβ40 ratio below the median and p-tau above the median had profiles closer to the AD continuum. We repeated the same cross-sectional and longitudinal analyses used in the WRAP cohort. Partial correlation results are summarized in \u003cb\u003eSupplementary Fig.\u0026nbsp;1\u003c/b\u003e, showing a similar profile to correlations in the WRAP cohort.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"No\" id=\"Taba\" border=\"1\"\u003e\u003ccolgroup cols=\"8\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c8\" colnum=\"8\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colspan=\"6\" nameend=\"c6\" namest=\"c1\"\u003e\u003cp\u003eTable\u0026nbsp;2. Demographics and baseline biomarkers for the Alzheimer and Families (ALFA+) cohort\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e\u003cb\u003eOverall\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e\u003cb\u003eAβ42/Aβ40 ratio\u0026thinsp;\u0026ge;\u0026thinsp;median (2)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e\u003cb\u003eAβ42/Aβ40 ratio\u0026thinsp;\u0026lt;\u0026thinsp;median\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003ep\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e\u003cb\u003ep-tau\u0026thinsp;\u0026ge;\u0026thinsp;median (4)\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e\u003cb\u003ep-tau\u0026thinsp;\u0026lt;\u0026thinsp;median\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003en\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e400\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e200\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e177\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e176\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge at time of LP [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e61.2 (4.69)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e60.4 (4.34)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e61.9 (4.90)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e61.7 (4.37)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e59.9 (4.82)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGener [Male (%)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e245 (62.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e123 (62.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e122 (62.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e1.000\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e114 (65.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e110 (64.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.906\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEthnicity (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.343\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.716\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eWhite\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e394 (98.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e197 (98.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e197 (98.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e175 (98.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e172 (97.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eOther (1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e6 (1.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3 (1.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e3 (1.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e2 (1.1)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4 (2.3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eApoE \u0026#120518;4[Non-carrier (%)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e184 (46.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e128 (64.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e56 (28.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e80 (45.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e84 (47.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.712\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMMSE [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e29.1 (0.99)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e29.2 (0.92)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e29.1 (1.05)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.206\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e29.1 (1.10)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e29.2 (0.88)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.450\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e4.49 (0.91)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e4.55 (0.88)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e4.42 (0.94)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.170\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e4.43 (0.97)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e4.51 (0.88)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.406\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAmyloid + (%) (3)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e135 (33.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0 (0.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e135 (67.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e73 (41.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e43 (24.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTau + (%) (5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e42 (11.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e10 (5.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e32 (19.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e42 (23.7)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0 (0.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003ePositive ptau/ab42 ratio (%)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e50 (14.2)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0 (0.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e50 (29.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e44 (24.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e6 (3.4)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003ep (adj by age)\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003ep (adj by age)\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAβ42 [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1322 (597)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1661 (601)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e983 (350)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1607 (713)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e1143 (341)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAβ40 [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e17403 (4997)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e17971 (5120)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e16834 (4817)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.023\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e21191 (4207)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e14521 (2874)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAβ42/Aβ40 ratio [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.08 (0.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.09 (0.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.06 (0.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e0.07 (0.02)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e0.08 (0.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cem\u003e0.059\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eT- tau [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e201 (72.9)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e192 (56.5)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e212 (86)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.009\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e253 (67.8)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e149 (24.0)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eP-tau [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e16.5 (7.59)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e15.4 (5.18)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e17.8 (9.43)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.003\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e21.6 (7.78)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e11.5 (2.01)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003esTREM2\u0026nbsp; [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7956 (2257)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8070 (2220)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7842 (2293)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.313\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8964 (2399)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e7161 (1711)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNeurogranin [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e800 (331)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e787 (296)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e812 (362)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.452\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1047 (313)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e587 (134)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u0026#120514;- synuclein \u0026nbsp; [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e234 (254)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e235 (227)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e232 (279)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.905\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e298 (329)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e184 (164)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eS100B\u0026nbsp; [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1024 (236)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e999 (209)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1049 (258)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cb\u003e0.031\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e1087 (250)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e966 (206)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eGFAP [mean (SD)]\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e7720 (2638)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e7675 (2842)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e7766 (2424)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e\u003cp\u003e\u003cem\u003e0.731\u003c/em\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e\u003cp\u003e8814 (2837)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c7\"\u003e\u003cp\u003e6718 (2091)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c8\"\u003e\u003cp\u003e\u003cb\u003e\u0026lt;\u0026thinsp;0.001\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003e(1) Other: Gypsy ethnic, latin american and not evaluated (2) Aβ42/Aβ40 median\u0026thinsp;=\u0026thinsp;0.08057 (3) The cut-off for amyloid (A) positivity according to the Aβ42/Aβ40 ratio is 0.071. (4) P-tau median\u0026thinsp;=\u0026thinsp;14.75 pg/mL.\u003c/p\u003e\u003cp\u003e(5) The cut-off for tau (T) positivity according to P-tau is 24 pg/mL. LP: lumbar puncture. PH: Parental history. MMSE: mini-mental state examination score\u003c/p\u003e\u003cp\u003eThe linear regression models showed associations between sTREM2 and synapse-related and astroglial biomarkers, summarized in \u003cb\u003eSupplementary Table\u0026nbsp;3\u003c/b\u003e. As in the WRAP cohort, these associations were attenuated after adjusting for Aβ42 and p-tau levels, indicating partial dependence on AD pathology (Model 2). However, negative associations remained between sTREM2 and neurogranin (β = -0.21, p\u0026thinsp;=\u0026thinsp;0.04), as well as in the subgroups with below-median p-tau (β = -0.37, p\u0026thinsp;=\u0026thinsp;0.002) and T- (β = -0.21, p\u0026thinsp;=\u0026thinsp;0.04). In contrast to WRAP cohort results, sTREM2 and ⍺-synuclein showed positive adjusted associations only in the below-median Aβ42/Aβ40 ratio and A\u0026thinsp;+\u0026thinsp;group (β\u0026thinsp;=\u0026thinsp;0.15, p\u0026thinsp;=\u0026thinsp;0.02 and β\u0026thinsp;=\u0026thinsp;0.20, p\u0026thinsp;=\u0026thinsp;0.007, respectively). Consistent with WRAP results, the ALFA\u0026thinsp;+\u0026thinsp;cohort showed a significant positive association between sTREM2 and S100B in the whole sample after adjusting for Aβ42 and p-tau (β\u0026thinsp;=\u0026thinsp;0.28, p\u0026thinsp;=\u0026thinsp;0.00002). This association was also significant in almost all subgroups, particularly in the A\u0026thinsp;+\u0026thinsp;group (β\u0026thinsp;=\u0026thinsp;0.51, p\u0026thinsp;=\u0026thinsp;0.000001. Furthermore, there was a significant interaction between S100B and Aβ42 levels in the whole cohort (p\u0026thinsp;=\u0026thinsp;0.03). In contrast to the WRAP results, the ALFA\u0026thinsp;+\u0026thinsp;cohort demonstrated a positive, significant adjusted association between GFAP and sTREM2 in the whole sample (β\u0026thinsp;=\u0026thinsp;0.24, p\u0026thinsp;=\u0026thinsp;0.000002), which was also significant in most subgroups, except for T\u0026thinsp;+\u0026thinsp;and below-median Aβ42/Aβ40 ratio. Remaining interaction terms were tested using continuous values of Aβ42 and p-tau; none were statistically significant and are not shown.\u003c/p\u003e\u003cp\u003eLongitudinally, the linear mixed models revealed similar trends, with associations found only between synapse-related biomarkers and sTREM2, but not between glial activation biomarkers and sTREM2. They are summarized in \u003cb\u003eSupplementary Table\u0026nbsp;4.\u003c/b\u003e Importantly, in contrast to the WRAP cohort results, the adjustment for Aβ42 and p-tau influenced the coefficients in Model 2. For neurogranin, we found an association between lower baseline neurogranin and a larger longitudinal increase in sTREM2 in Model 1, only in the above-median Aβ42/Aβ40 ratio group (β-coefficient for interaction with time = -0.02, p\u0026thinsp;=\u0026thinsp;0.02) and the A- group (β-coefficient for interaction with time = -0.002, p\u0026thinsp;=\u0026thinsp;0.008), suggesting an association in participants without evidence of amyloid pathology. However, these associations did not remain significant after adjusting for Aβ42 and p-tau (Model 2), suggesting confounding by AD pathology.\u003c/p\u003e\u003cp\u003eSimilarly, we observed that lower baseline α-syn was associated with a larger longitudinal increase in sTREM2 over time in the above-median Aβ42/Aβ40 ratio group for both model 1 (β-coefficient for interaction with time = -0.01, p\u0026thinsp;=\u0026thinsp;0.05) and model 2 (β-coefficient for interaction with time = -0.01, p\u0026thinsp;=\u0026thinsp;0.05). As in the WRAP cohort, sTREM2 levels above the median were associated with a diminished longitudinal increase in neurogranin (β-coefficient for interaction with time = -0.01, p\u0026thinsp;=\u0026thinsp;0.04). In sum, while both cohorts demonstrated cross-sectional associations between sTREM2 and markers of synaptic and glial function, the longitudinal patterns were more robust and AD-independent in the WRAP cohort compared to ALFA+.\u003c/p\u003e\u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis study offers new insights into the interplay between synaptic dysfunction, TREM2-dependent microglial response, and astroglial activation, through a CSF-based biomarker approach applied to two independent longitudinal cohorts of cognitively normal, late-middle-aged individuals. Longitudinally, lower baseline levels of synaptic proteins and higher levels of S100B predicted a larger subsequent increase in sTREM2, independent of AD-related biomarkers. These findings suggest that early synaptic dysfunction may act as a trigger for TREM2-dependent microglial activation, regardless of AD pathology. Additionally, higher baseline levels of sTREM2 were associated with more stable neurogranin levels over time, further supporting the role of TREM2 as a modulator of synaptic function and potentially protective against synaptic dysregulation and cognitive decline throughout aging and early stages of neurodegenerative processes.\u003c/p\u003e\u003cp\u003eCross-sectionally, we found an association between sTREM2 and α-syn, specifically in participants with neurodegeneration-related biomarker profiles (T\u0026thinsp;+\u0026thinsp;group in the WRAP cohort, and Aβ42/Aβ40 below median and A\u0026thinsp;+\u0026thinsp;groups in ALFA+). Interpreting CSF α-syn levels remains challenging due to its dual role in pathological processes, such as aggregation and neurodegeneration in synucleinopathies, and physiological processes, including synaptic function and axonal remodeling during normal aging [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. In neurodegenerative diseases, total CSF α-syn is often considered a marker of neurodegeneration rather than synaptic dysfunction [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. This is supported by previously reported significant correlations between ⍺-syn, p-tau and t-tau in CSF, in concordance with our results [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. However, in PD, most studies report reduced total CSF ⍺-syn levels during the early stages, likely reflecting initial synaptic dysfunction or early α-syn aggregation. In later stages, higher levels have been reported, which may signal overt neuronal injury [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. The variations in α-syn levels clearly illustrate that it must be interpreted within its biological context. We interpret the observed cross-sectional relationship between sTREM2 and α-syn in participants with neurodegeneration-related biomarker profiles as reflective of a shared cross-sectional association with incipient neurodegeneration.\u003c/p\u003e\u003cp\u003eRegarding the cross-sectional relationship between sTREM2 and astroglial activation markers, we found a significant association between sTREM2 and S100B in the ALFA\u0026thinsp;+\u0026thinsp;cohort and a trend toward an association in the WRAP cohort. These associations remained robust after adjusting for AD-related biomarkers and were stronger in subgroups with a biochemical profile suggestive of early phases of AD pathology. This observation is consistent with previous studies that show stronger sTREM2\u0026ndash;S100B correlations in asymptomatic participants with elevated p-tau/Aβ42 ratios[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] and across the symptomatic AD continuum [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. This suggests increased interaction between astroglial and microglial responses in early AD. Interestingly, we observed no cross-sectional association between GFAP and sTREM2 in the WRAP cohort. Interestingly, both S100B and GFAP demonstrated significant cross-sectional associations with sTREM2 across subgroups in the ALFA\u0026thinsp;+\u0026thinsp;cohort.\u003c/p\u003e\u003cp\u003ePrevious research has already described a relatively low correlation between S100B and other astroglial biomarkers [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e], which suggests that S100B represents a distinct response specifically related to synaptic dysfunction rather than general astrocyte activation. Given its theoretical dual role\u0026mdash;neurotrophic at nanomolar and pro-inflammatory at micromolar concentrations\u0026mdash;S100B may reflect a neuroprotective astroglial response to early synaptic dysfunction in asymptomatic late-middle-aged individuals [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e, \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. This interpretation is reinforced by the weak correlation between S100B and GFAP, suggesting S100B secretion occurs independently from the pathological astrocytic hyperactivation indicated by increased GFAP. Differences between the WRAP and ALFA\u0026thinsp;+\u0026thinsp;cohorts could stem from their distinct participant characteristics, as the ALFA\u0026thinsp;+\u0026thinsp;cohort includes healthy volunteers enriched for AD risk factors, likely leading to earlier GFAP elevations as an initial astroglial response to pathology. In fact, the observed cross-sectional association between GFAP and sTREM2 in the ALFA\u0026thinsp;+\u0026thinsp;cohort probably reflects the early interplay between astroglial and microglial activation as AD-related changes begin. Overall, our findings indicate distinct astroglial response patterns involving S100B and GFAP across cohorts with varying AD risk profiles. This underscores a complex interplay between astroglial and microglial activation that may differentially reflect synaptic dysfunction and initial amyloid-related pathology.\u003c/p\u003e\u003cp\u003eLongitudinally, we observed that a biomarker profile suggestive of early synaptic dysfunction at baseline \u0026mdash;characterized by lower levels of α-syn and neurogranin along with higher levels of S100B in CSF\u0026mdash; was associated with a greater subsequent longitudinal increase of sTREM2 in CSF over time. These associations were independent of AD-related biomarker status in the WRAP cohort of cognitively healthy, late-middle-aged individuals. As discussed, interpreting CSF α-syn levels is challenging due to its involvement in different pathological and physiological processes [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. In participants of the WRAP cohort, who are mainly individuals without manifest amyloid deposition nor neurodegeneration, we interpret lower CSF α-syn levels as indicative of age-related synaptic dysfunction rather than incipient neurodegeneration. In contrast, in individuals with overt neurodegenerative processes or more evident AD pathology, α-syn levels may represent the underlying neurodegeneration, which our cross-sectional findings support. In the ALFA\u0026thinsp;+\u0026thinsp;cohort, lower levels of α-syn at baseline were associated to greater increases in sTREM2 specifically among participants with higher Aβ42/Aβ40 ratios. This highlights how synaptic dysfunction could be a booster of microglial activation within aging or non-AD neurodegeneration rather than within the AD continuum.\u003c/p\u003e\u003cp\u003eSimilarly, lower neurogranin levels in individuals without biomarker evidence of AD likely reflect reduced postsynaptic activity. This interpretation is consistent with prior findings in PD, where neurogranin is reduced and correlates with cortical hypometabolism and cognitive deficits [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. However, elevated CSF neurogranin has been reported even at presymptomatic stages in the AD continuum, likely reflecting neurodegeneration or neuronal hyperactivity rather than isolated synaptic dysfunction [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan additionalcitationids=\"CR47\" citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. The strong correlation between neurogranin and tau-related markers supports its role as a disease-stage specific injury marker of AD [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. Notably, increased network excitability\u0026mdash;along with heightened seizure susceptibility\u0026mdash;has been observed in early AD and may contribute to regional increases in synaptic density and elevated neurogranin levels [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e], highlighting its dynamic, context-dependent interpretation. In the ALFA\u0026thinsp;+\u0026thinsp;cohort, lower baseline neurogranin was also associated with increased sTREM2 over time. However, only in subgroups without amyloid pathology and prior to adjustment for p-tau, suggesting that tau pathology may mask this association in a cohort with a high AD-risk profile.\u003c/p\u003e\u003cp\u003eFurther supporting early synaptic dysfunction as an independent trigger for TREM2-dependent microglial activation, we found that individuals with higher S100B levels at baseline exhibited a greater longitudinal increase in sTREM2 in the WRAP cohort. Interestingly, baseline GFAP levels were not predictive of longitudinal changes in sTREM2 in the WRAP nor in the ALFA\u0026thinsp;+\u0026thinsp;cohort. The absence of longitudinal associations with GFAP suggests that generalized astroglial activation does not independently boost microglial responses over time. Instead, the selective longitudinal relationship between elevated S100B and subsequent sTREM2 increase reinforces our interpretation of S100B as a synapse-coupled astroglial signal that can prime microglia independently of GFAP-defined astrocytic activation during physiological aging and non-AD neurodegenerative processes. Mechanistically, astrocytic S100B may mark dendritic-spine stress that, in turn, triggers a TREM2-dependent pruning response, echoing the complement-mediated synapse-elimination pathway observed in recent mouse and human studies [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. In the context of AD pathology, however, the strong effect of Aβ aggregation in boosting TREM2-dependent microglial response may override the subtler modulatory influence of synaptic dysfunction throughout aging and non-AD neurodegeneration processes.\u003c/p\u003e\u003cp\u003eAnd finally, we found that higher baseline CSF sTREM2 predicted a slower rise in neurogranin yet had no influence on α-syn or S100B trajectories. This pattern is congruent with recent findings that show TREM2-competent microglia identify phosphatidylserine-tagged, hyperactive spines and remove them via a complement pathway, thereby normalizing circuit activity both in disease models and clinical cohorts [\u003cspan additionalcitationids=\"CR51 CR52 CR53 CR54\" citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e]. Instead, loss-of-function TREM2 variants in mice increase spine density, drive cortical hyperexcitability, and heighten seizure susceptibility [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. This might be mirrored in electrophysiological phenotypes of patients with early AD who experience a higher incidence of subclinical and overt epileptic events [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. Together, our findings and external evidence support a modulatory effect of TREM2-activated microglia on excessive synaptic activity: pruning superfluous spines, stabilizing neurogranin release, and reducing network hyperexcitability and seizure risk. The absence of longitudinal effects on the presynaptic marker α-syn or on astroglial S100B emphasizes that this microglial feedback loop is largely postsynaptic-specific.\u003c/p\u003e\u003cp\u003eThis study has several limitations. First, the inclusion of only cognitively unimpaired individuals limits generalizability to symptomatic stages, though it enables the study of preclinical processes. Second, CSF biomarkers do not fully reflect regional synaptic or neuroinflammatory dynamics. Third, modest sample size and limited longitudinal data reduce statistical power and temporal resolution. Additionally, strong inter-biomarker correlations may obscure independent effects, despite the use of complementary sTREM2 assays. The use of cohort-specific Aβ and tau cut-offs may affect comparability across studies. Finally, despite adjusting for p-tau, residual confounding cannot be excluded due to its collinearity with synaptic markers.\u003c/p\u003e\u003cp\u003eDespite these limitations, our study has several strengths. One key strength is the availability of longitudinal CSF biomarker data from cognitively normal individuals. This allows us to examine biomarker trajectories over time, providing valuable insights into the evolution of synaptic and glial markers in aging and the preclinical stages of neurodegeneration. Another strength is the validation of findings in a secondary cohort of cognitively normal individuals, enhancing the robustness of our results. Furthermore, we incorporated multiple biomarkers to capture the complexity of the biological processes underlying synaptic dysfunction and microglial activation, ensuring a more comprehensive assessment of their interplay.\u003c/p\u003e\u003cp\u003eTogether, our cross-sectional and longitudinal findings suggest that synaptic stress activates a TREM2-dependent microglial response across aging and in non-AD neuropathological contexts. Once Aβ aggregation begins along the AD-continuum, its potent effect on activating the TREM2 pathway may dominate, diminishing the relative contribution of synaptic dysfunction. Astroglial\u0026ndash;microglial coupling also changes with AD biomarker status, emphasizing the dynamic and coordinated glial response to early neuropathological changes. Our findings support ongoing therapeutic strategies that target the synapse-to-glia axis, boosting TREM2 signaling or protecting synapses [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e]. The data highlights that the clinical benefit of modulating TREM2 will hinge on whether synaptic stress or established Aβ/tau pathology is the predominant biological driver.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAD- Alzheimer\u0026rsquo;s disease\u003c/p\u003e\n\u003cp\u003eALFA- ALzheimer and Families cohort\u003c/p\u003e\n\u003cp\u003eA\u0026beta;- amyloid-\u0026beta;\u003c/p\u003e\n\u003cp\u003eCSF- cerebrospinal fluid\u003c/p\u003e\n\u003cp\u003eDLB- dementia with Lewy bodies\u003c/p\u003e\n\u003cp\u003eGFAP- glial fibrillary acid protein\u003c/p\u003e\n\u003cp\u003eLP- lumbar puncture\u003c/p\u003e\n\u003cp\u003eNTK- NeuroToolKit\u003c/p\u003e\n\u003cp\u003ePD-Parkinson\u0026rsquo;s disease\u003c/p\u003e\n\u003cp\u003eS100B- S100 calcium-binding protein B\u003c/p\u003e\n\u003cp\u003esTREM2- soluble cleavage product of the microglial protein TREM2\u003c/p\u003e\n\u003cp\u003eTREM2- triggering receptor on myeloid cells 2\u003c/p\u003e\n\u003cp\u003eWRAP- Wisconsin Registry for Alzheimer\u0026rsquo;s Prevention\u003c/p\u003e\n\u003cp\u003e\u0026alpha;-syn- alpha-synuclein\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003eEthics approval and consent to participate\u003c/p\u003e\n\u003cp\u003eThe WRAP cohort study was conducted in compliance with the ethical principles for human subjects\u0026rsquo; research defined in the Declaration of Helsinki, including approval by the University of Wisconsin-Madison Institutional Review Board.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe ALFA study protocol was approved by the Independent Ethics committee \u003cem\u003eParc de Salut Mar\u0026nbsp;\u003c/em\u003eand registered at clinicaltrials.gov (identifier: NCT01835717). It was conducted in accordance with the directives of the Spanish Law 14/2007, of 3\u003csup\u003erd\u003c/sup\u003e of July, on Biomedical Research. All participants in the ALFA study accepted the study procedures by signing an informed consent form and had a close relative volunteering to participate in the functional assessment procedure of the participant, who also granted his or her consent.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eAvailability of data and materials\u003c/p\u003e\n\u003cp\u003eData utilized in the WRAP cohort consist of sensitive, human research participant data and participants have not signed informed consent to have their data shared in public repositories for publications. Therefore, data deposition is unethical. Data are available upon request for authorized researchers who meet the criteria for access to confidential data. The data underlying the results presented in this study are available from http://www.wai.wisc.edu/research/. Data from the ALFA+ cohort is also available upon request from https://www.barcelonabeta.org/es/estudio-alfa/sobre-el-estudio-alfa.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCompeting interests\u003c/p\u003e\n\u003cp\u003eM.I.M.-G\u0026nbsp;has received in the past 36mo speaker fees by Almirall.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eE.M.-R. has given lectures and symposia sponsored by KRKA Farmaceutica SL and Laboratorios Esteve SA.\u003c/p\u003e\n\u003cp\u003eM.S.-C. has received in the past 36mo consultancy/speaker fees (paid to the institution) from by Almirall, Eli Lilly, Quanterix, Novo Nordisk, and Roche Diagnostics. He has received consultancy fees or served on advisory boards (paid to the institution) of Eli Lilly, Grifols, Novo Nordisk, and Roche Diagnostics. He was granted a project and is a site investigator of a clinical trial (funded to the institution) by Roche Diagnostics. In-kind support for research (to the institution) was received from ADx Neurosciences, Alamar Biosciences, ALZPath, Avid Radiopharmaceuticals, Eli Lilly, Fujirebio, Janssen Research \u0026amp; Development, Meso Scale Discovery, and Roche Diagnostics; MS-C did not receive any personal compensation from these organizations or any other for-profit organization.\u003c/p\u003e\n\u003cp\u003eH.Z. has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp \u0026amp; Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of MicThera (outside submitted work).\u003c/p\u003e\n\u003cp\u003eS.C.-J. serves as a consultant to Eli Lilly, Merck, AlzPath and Enigma Biomedical.\u003c/p\u003e\n\u003cp\u003eG.K. is a full-time employee of Roche Diagnostics GmbH, Penzberg, Germany.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eCQ-R is a full-time employee of Roche Diagnostics International Ltd, Rotkreuz, Switzerland.\u003c/p\u003e\n\u003cp\u003eFunding\u003c/p\u003e\n\u003cp\u003eE.M.-R. receives funding by the Instituto de Salud Carlos III (ISCIII) under the Juan Rod\u0026eacute;s Program (Grant number JR21/00014) and through the project PI22/00215; she is also funded by Eugenio Rodr\u0026iacute;guez Pascual Foundation through the project FERP-2024-091.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eM.I.M.-G\u0026nbsp;receives funding by the Instituto de Salud Carlos III (ISCIII) under the Rio Hortega Program (Grant number\u0026nbsp;CM24/00130).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe Wisconsin Registry for Alzheimer\u0026rsquo;s Prevention is funded by the National Institute on Aging R01 AG027161 with additional funding from AG021155 and the CSF collection service of the Wisconsin Alzheimer\u0026rsquo;s Disease Research Center P30AG062715.\u003c/p\u003e\n\u003cp\u003eMSC receives funding from the European Research Council (ERC) under the European Union\u0026rsquo;s Horizon 2020 research and innovation programme (Grant agreement No. 948677); ERA PerMed-ERA NET and the Generalitat de Catalunya (Departament de Salut) through the project SLD077/21/000001; Project \u0026quot;PI19/00155\u0026quot; and \u0026ldquo;PI22/00456, funded by Instituto de Salud Carlos III (ISCIII) and co-funded by the European Union; and from a fellowship from \u0026rdquo;la Caixa\u0026rdquo; Foundation (ID 100010434) and from the European Union\u0026rsquo;s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 847648 (LCF/BQ/PR21/11840004).\u003c/p\u003e\n\u003cp\u003eHZ is a Wallenberg Scholar and a Distinguished Professor at the Swedish Research Council supported by grants from the Swedish Research Council (#2023-00356, #2022-01018 and #2019-02397), the European Union\u0026rsquo;s Horizon Europe research and innovation programme under grant agreement No 101053962, Swedish State Support for Clinical Research (#ALFGBG-71320), the Alzheimer Drug Discovery Foundation (ADDF), USA (#201809-2016862), the AD Strategic Fund and the Alzheimer\u0026apos;s Association (#ADSF-21-831376-C, #ADSF-21-831381-C, #ADSF-21-831377-C, and #ADSF-24-1284328-C), the European Partnership on Metrology, co-financed from the European Union\u0026rsquo;s Horizon Europe Research and Innovation Programme and by the Participating States (NEuroBioStand, #22HLT07), the Bluefield Project, Cure Alzheimer\u0026rsquo;s Fund, the Olav Thon Foundation, the Erling-Persson Family Foundation, Familjen R\u0026ouml;nstr\u0026ouml;ms Stiftelse, Familjen Beiglers Stiftelse, Stiftelsen f\u0026ouml;r Gamla Tj\u0026auml;narinnor, Hj\u0026auml;rnfonden, Sweden (#FO2022-0270), the European Union\u0026rsquo;s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 860197 (MIRIADE), the European Union Joint Programme \u0026ndash; Neurodegenerative Disease Research (JPND2021-00694), the National Institute for Health and Care Research University College London Hospitals Biomedical Research Centre, the UK Dementia Research Institute at UCL (UKDRI-1003), and an anonymous donor.\u003c/p\u003e\n\u003cp\u003eAuthors\u0026apos; contributions\u003c/p\u003e\n\u003cp\u003eE.M.-R., Y.D, and B.B.B. contributed to the conceptualization of the study. M.I.M.-G. was responsible for data analysis and wrote the initial draft of the manuscript. E.M.-R. co-wrote the manuscript. F.L.H. contributed to data analysis in the ALFA+ cohort. Y.D., S.J., S.A., C.C., O.C.O., and B.B.B. contributed to data acquisition, WRAP cohort management, and interpretation of results. E.M.-R. contributed to biomarker measurements for the WRAP cohort. M.S.-C. and F.L.H. contributed to data acquisition and interpretation for the ALFA cohort. G.K. and C.Q.R. contributed to biomarker measurements for the ALFA cohort. Y.D., S.J., S.A., C.C., O.C.O., D.P.-M., A.V.-G., K.B., M.S.-C., H.Z., and B.B.B. contributed to critical revision of the manuscript. All authors reviewed and approved the final version of the manuscript.\u003c/p\u003e\n\u003cp\u003eAcknowledgements:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThis publication is part of the ALFA study (ALzheimers and Families). The authors would like to express their most sincere gratitude to the ALFA project participants and relatives without whom this research would not have been possible. The authors thank Roche Diagnostics International Ltd for providing the kits to measure CSF biomarkers. The Roche NeuroToolKit is a panel of exploratory prototype assays designed to robustly evaluate biomarkers associated with key pathologic events characteristic of AD and other neurological disorders, used for research purposes only and not approved for clinical use. Elecsys Phospho-Tau (181P) CSF and Elecsys Total-Tau CSF assays are approved for clinical use. COBAS and COBAS and ELECSYS are trademarks of Roche. All other product names and trademarks are the property of their respective owners. Collaborators of the ALFA Study are: Annabella Beteta, Anna Brugulat-Serrat, Alba Ca\u0026ntilde;as, Irene Cumplido-Mayoral, Carme Deulofeu, Ruth Dominguez, Maria Emilio, Karine Fauria, Ana Fern\u0026aacute;ndez-Arcos, Sherezade Fuentes, Patricia Genius, Laura Hern\u0026aacute;ndez, Gema Huesa, Jordi Huguet, Paula Marne, Tania Mench\u0026oacute;n, Wiesje Pelkmans, Albina Polo, Sandra Pradas, Blanca Rodr\u0026iacute;guez-Fern\u0026aacute;ndez, Anna Soteras, Laura Stankeviciute, and Marc Vilanova\u003c/p\u003e\n\u003cp\u003eThe NeuroToolKit is a panel of exploratory prototype assays designed to robustly evaluate biomarkers associated with key pathologic events characteristic of AD and other neurological disorders, used for research purposes only and not approved for clinical use (Roche Diagnostics International Ltd, Rotkreuz, Switzerland). COBAS and ELECSYS are trademarks of Roche. Elecsys \u0026beta;-Amyloid (1\u0026ndash;42) CSF, Elecsys Phospho-Tau (181P) CSF and Elecsys Total-Tau CSF assays are approved for clinical use.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eWilson III DM, Cookson MR, Van L, Bosch D, Zetterberg H, Holtzman DM, et al. Hallmarks of neurodegenerative diseases. Cell. 2023;186:693\u0026ndash;714. \u003c/li\u003e\n\u003cli\u003eSimuni T, Chahine LM, Poston K, Brumm M, Buracchio T, Campbell M, et al. A biological definition of neuronal \u0026alpha;-synuclein disease: towards an integrated staging system for research. Lancet Neurol. 2024;23:178\u0026ndash;90. \u003c/li\u003e\n\u003cli\u003eChung WS, Welsh CA, Barres BA, Stevens B. Do Glia Drive Synaptic and Cognitive Impairment in Disease? Nat Neurosci. 2015;18:1539. \u003c/li\u003e\n\u003cli\u003eLi L, Lu S, Zhu J, Yu X, Hou S, Huang Y, et al. Astrocytes Excessively Engulf Synapses in a Mouse Model of Alzheimer\u0026rsquo;s Disease. Int J Mol Sci. 2024;25. \u003c/li\u003e\n\u003cli\u003eG\u0026oacute;mez-Gonzalo M, Martin-Fernandez M, Mart\u0026iacute;nez-Murillo R, Mederos S, Hern\u0026aacute;ndez-Vivanco A, Jamison S, et al. Neuron\u0026ndash;astrocyte signaling is preserved in the aging brain. Glia. 2017;65:569\u0026ndash;80. \u003c/li\u003e\n\u003cli\u003eSoreq L, Rose J, Soreq E, Hardy J, Trabzuni D, Cookson MR, et al. Major Shifts in Glial Regional Identity Are a Transcriptional Hallmark of Human Brain Aging. Cell Rep. 2017;18:557\u0026ndash;70. \u003c/li\u003e\n\u003cli\u003eScott-Hewitt N, Mahoney M, Huang Y, Korte N, Yvanka de Soysa T, Wilton DK, et al. Microglial-derived C1q integrates into neuronal ribonucleoprotein complexes and impacts protein homeostasis in the aging brain. Cell. 2024;187:4193-4212.e24. \u003c/li\u003e\n\u003cli\u003eSancho L, Contreras M, Allen NJ. Glia as sculptors of synaptic plasticity. Neurosci Res. 2021;167:17\u0026ndash;29. \u003c/li\u003e\n\u003cli\u003eSantello M, Toni N, Volterra A. Astrocyte function from information processing to cognition and cognitive impairment. Nat Neurosci. 2019;22:154\u0026ndash;66. \u003c/li\u003e\n\u003cli\u003eBellaver B, Povala G, Ferreira PCL, Ferrari-Souza JP, Leffa DT, Lussier FZ, et al. Astrocyte reactivity influences amyloid-\u0026beta; effects on tau pathology in preclinical Alzheimer\u0026rsquo;s disease. Nat Med. 2023;29:1775\u0026ndash;81. \u003c/li\u003e\n\u003cli\u003eMorenas-Rodr\u0026iacute;guez E, Li Y, Nuscher B, Franzmeier N, Xiong C, Su\u0026aacute;rez-Calvet M, et al. Soluble TREM2 in CSF and its association with other biomarkers and cognition in autosomal-dominant Alzheimer\u0026rsquo;s disease: a longitudinal observational study. Lancet Neurol. 2022;21:329\u0026ndash;41. \u003c/li\u003e\n\u003cli\u003eKeren-Shaul H, Spinrad A, Weiner A, Matcovitch-Natan O, Dvir-Szternfeld R, Ulland TK, et al. A Unique Microglia Type Associated with Restricting Development of Alzheimer\u0026rsquo;s Disease. Cell. 2017;169:1276\u0026ndash;90. \u003c/li\u003e\n\u003cli\u003eMazaheri F, Snaidero N, Kleinberger G, Madore C, Daria A, Werner G, et al. TREM 2 deficiency impairs chemotaxis and microglial responses to neuronal injury . EMBO Rep. 2017;18:1186\u0026ndash;98. \u003c/li\u003e\n\u003cli\u003eHabib N, McCabe C, Medina S, Varshavsky M, Kitsberg D, Dvir-Szternfeld R, et al. Disease-associated astrocytes in Alzheimer\u0026rsquo;s disease and aging. Nat Neurosci. 2020;23:701. \u003c/li\u003e\n\u003cli\u003eLiddelow SA, Guttenplan KA, Clarke LE, Bennett FC, Bohlen CJ, Schirmer L, et al. Neurotoxic reactive astrocytes are induced by activated microglia. Nature. 2017;541:481. \u003c/li\u003e\n\u003cli\u003eTzioras M, McGeachan RI, Durrant CS, Spires-Jones TL. Synaptic degeneration in Alzheimer disease. Nat Rev Neurol. 2023;19:19\u0026ndash;38. \u003c/li\u003e\n\u003cli\u003eBarba L, Abu-Rumeileh S, Halbgebauer S, Bellomo G, Paolini Paoletti F, Gaetani L, et al. CSF Synaptic Biomarkers in AT(N)-Based Subgroups of Lewy Body Disease. Neurology. 2023;101:e50. \u003c/li\u003e\n\u003cli\u003ePaolini Paoletti F, Gaetani L, Bellomo G, Chipi E, Salvadori N, Montanucci C, et al. CSF neurochemical profile and cognitive changes in Parkinson\u0026rsquo;s disease with mild cognitive impairment. NPJ Parkinsons Dis. 2023;9:68. \u003c/li\u003e\n\u003cli\u003eLehmann S, Schraen-Maschke S, Bu\u0026eacute;e L, Vidal JS, Delaby C, Hirtz C, et al. Clarifying the association of CSF A\u0026beta;, tau, BACE1, and neurogranin with AT(N) stages in Alzheimer disease. Mol Neurodegener. 2024;19:66. \u003c/li\u003e\n\u003cli\u003eNilsson J, Binette AP, Palmqvist S, Brum WS, Janelidze S, Ashton NJ, et al. Cerebrospinal fluid biomarker panel for synaptic dysfunction in a broad spectrum of neurodegenerative diseases. Brain. 2024; \u003c/li\u003e\n\u003cli\u003ePortelius E, Zetterberg H, Skillb\u0026auml;ck T, T\u0026ouml;rnqvist U, Andreasson U, Trojanowski JQ, et al. Cerebrospinal fluid neurogranin: relation to cognition and neurodegeneration in Alzheimer\u0026rsquo;s disease. Brain. 2015;138:3373\u0026ndash;85. \u003c/li\u003e\n\u003cli\u003eLista S, Hampel H. Synaptic degeneration and neurogranin in the pathophysiology of Alzheimer\u0026rsquo;s disease. Expert Rev Neurother. 2017;17:47\u0026ndash;57. \u003c/li\u003e\n\u003cli\u003eTherriault J, Schindler SE, Salvad\u0026oacute; G, Pascoal TA, Benedet AL, Ashton NJ, et al. Biomarker-based staging of Alzheimer disease: rationale and clinical applications. Nat Rev Neurol. 2024;20:232\u0026ndash;44. \u003c/li\u003e\n\u003cli\u003eCamporesi E, Nilsson J, Brinkmalm A, Becker B, Ashton NJ, Blennow K, et al. Fluid Biomarkers for Synaptic Dysfunction and Loss. Biomark Insights. 2020;15. \u003c/li\u003e\n\u003cli\u003eVan Hulle C, Jonaitis EM, Betthauser TJ, Batrla R, Wild N, Kollmorgen G, et al. An examination of a novel multipanel of CSF biomarkers in the Alzheimer\u0026rsquo;s disease clinical and pathological continuum. Alzheimer\u0026rsquo;s and Dementia. 2021;17:431\u0026ndash;45. \u003c/li\u003e\n\u003cli\u003eSalvad\u0026oacute; G, Larsson V, Cody KA, Cullen NC, Jonaitis EM, Stomrud E, et al. Optimal combinations of CSF biomarkers for predicting cognitive decline and clinical conversion in cognitively unimpaired participants and mild cognitive impairment patients: A multi-cohort study. Alzheimer\u0026rsquo;s and Dementia. 2023;19:2943\u0026ndash;55. \u003c/li\u003e\n\u003cli\u003eBereczki E, Bogstedt A, H\u0026ouml;glund K, Tsitsi P, Brodin L, Ballard C, et al. Synaptic proteins in CSF relate to Parkinson\u0026rsquo;s disease stage markers. NPJ Parkinsons Dis. 2017;3. \u003c/li\u003e\n\u003cli\u003eWunderlich P, Glebov K, Kemmerling N, Tien NT, Neumann H, Walter J. Sequential proteolytic processing of the triggering receptor expressed on myeloid cells-2 (TREM2) protein by ectodomain shedding and \u0026gamma;-secretase-dependent intramembranous cleavage. J Biol Chem. 2013;288:33027\u0026ndash;36. \u003c/li\u003e\n\u003cli\u003eSchlepckow K, Morenas-Rodr\u0026iacute;guez E, Hong S, Haass C. Stimulation of TREM2 with agonistic antibodies-an emerging therapeutic option for Alzheimer\u0026rsquo;s disease. Lancet Neurol. 2023;22:1048\u0026ndash;60. \u003c/li\u003e\n\u003cli\u003eEwers M, Franzmeier N, Su\u0026aacute;rez-Calvet M, Morenas-Rodriguez E, Angel M, Caballero A. Increased soluble TREM2 in cerebrospinal fluid is associated with reduced cognitive and clinical decline in Alzheimer\u0026rsquo;s disease for the Alzheimer\u0026rsquo;s Disease Neuroimaging Initiative. Sci Transl Med. 2019;11:6221. \u003c/li\u003e\n\u003cli\u003ePelkmans W, Shekari M, Brugulat-Serrat A, S\u0026aacute;nchez-Benavides G, Minguill\u0026oacute;n C, Fauria K, et al. Astrocyte biomarkers GFAP and YKL-40 mediate early Alzheimer\u0026rsquo;s disease progression. Alzheimers Dement. 2024;20:483\u0026ndash;93. \u003c/li\u003e\n\u003cli\u003eMichetti F, Clementi ME, Di Liddo R, Valeriani F, Ria F, Rende M, et al. The S100B Protein: A Multifaceted Pathogenic Factor More Than a Biomarker. Int J Mol Sci. 2023;24:9605. \u003c/li\u003e\n\u003cli\u003eNishiyama H, Kn\u0026ouml; Pfel \u0026dagger; T, Endo S, Itohara S. Glial protein S100B modulates long-term neuronal synaptic plasticity. Proc Natl Acad Sci U S A. 99:4037\u0026ndash;42. \u003c/li\u003e\n\u003cli\u003eJohnson SC, Koscik RL, Jonaitis EM, Clark LR, Mueller KD, Berman SE, et al. The Wisconsin Registry for Alzheimer\u0026rsquo;s Prevention: A review of findings and current directions. Alzheimer\u0026rsquo;s and Dementia: Diagnosis, Assessment and Disease Monitoring. 2018;10:130\u0026ndash;42. \u003c/li\u003e\n\u003cli\u003eMolinuevo JL, Gramunt N, Gispert JD, Fauria K, Esteller M, Minguillon C, et al. The ALFA project: A research platform to identify early pathophysiological features of Alzheimer\u0026rsquo;s disease. Alzheimer\u0026rsquo;s and Dementia: Translational Research and Clinical Interventions. 2016;2:82\u0026ndash;92. \u003c/li\u003e\n\u003cli\u003eJohnson SC, Su\u0026aacute;rez-Calvet M, Suridjan I, Minguill\u0026oacute;n C, Gispert JD, Jonaitis E, et al. Identifying clinically useful biomarkers in neurodegenerative disease through a collaborative approach: the NeuroToolKit. Alzheimers Res Ther. 2023;15. \u003c/li\u003e\n\u003cli\u003eArgiris G, Akinci M, Pe\u0026ntilde;a-G\u0026oacute;mez C, Palpatzis E, Garcia-Prat M, Shekari M, et al. Data-driven CSF biomarker profiling: imaging and clinical outcomes in a cohort at risk of Alzheimer\u0026rsquo;s disease. Alzheimer\u0026rsquo;s Research and Therapy . 2024;16:274. \u003c/li\u003e\n\u003cli\u003eSoldan A, Pettigrew C, Cai Q, Wang MC, Moghekar AR, O\u0026rsquo;Brien RJ, et al. Hypothetical preclinical Alzheimer disease groups and longitudinal cognitive change. JAMA Neurol. 2016;73:698\u0026ndash;705. \u003c/li\u003e\n\u003cli\u003eMil\u0026agrave;-Alom\u0026agrave; M, Salvad\u0026oacute; G, Gispert JD, Vilor-Tejedor N, Grau-Rivera O, Sala-Vila A, et al. Amyloid beta, tau, synaptic, neurodegeneration, and glial biomarkers in the preclinical stage of the Alzheimer\u0026rsquo;s continuum. Alzheimers Dement. 2020;16:1358\u0026ndash;71. \u003c/li\u003e\n\u003cli\u003eSalvad\u0026oacute; G, Shekari M, Falcon C, Operto GDS, Mil\u0026agrave;-Alom\u0026agrave; M, S\u0026aacute;nchez-Benavides G, et al. Brain alterations in the early Alzheimer\u0026rsquo;s continuum with amyloid-\u0026beta;, tau, glial and neurodegeneration CSF markers. Brain Commun. 2022;4::fcac134. doi: 10.1093/braincomms/fcac134. \u003c/li\u003e\n\u003cli\u003eEusebi P, Giannandrea D, Biscetti L, Abraha I, Chiasserini D, Orso M, et al. Diagnostic utility of cerebrospinal fluid \u0026alpha;-synuclein in Parkinson\u0026rsquo;s disease: A systematic review and meta-analysis. Movement Disorders. 2017;32:1389\u0026ndash;400. \u003c/li\u003e\n\u003cli\u003eBonomi CG, Assogna M, Di Donna MG, Bernocchi F, De Lucia V, Nuccetelli M, et al. Cerebrospinal Fluid sTREM-2, GFAP, and \u0026beta;-S100 in Symptomatic Sporadic Alzheimer\u0026rsquo;s Disease: Microglial, Astrocytic, and APOE Contributions Along the Alzheimer\u0026rsquo;s Disease Continuum. Journal of Alzheimer\u0026rsquo;s Disease. 2023;92:1385\u0026ndash;97. \u003c/li\u003e\n\u003cli\u003eSteiner J, Bogerts B, Schroeter ML, Bernstein HG. S100B protein in neurodegenerative disorders. Clin Chem Lab Med. 2011. p. 409\u0026ndash;24. \u003c/li\u003e\n\u003cli\u003eBaecker J, Wartchow K, Sehm T, Ghoochani A, Buchfelder M, Kleindienst A. Treatment with the Neurotrophic Protein S100B Increases Synaptogenesis after Traumatic Brain Injury. J Neurotrauma. 2020;37:1097\u0026ndash;107. \u003c/li\u003e\n\u003cli\u003eSelnes P, Stav AL, Johansen KK, Bj\u0026oslash;rnerud A, Coello C, Auning E, et al. Impaired synaptic function is linked to cognition in Parkinson\u0026rsquo;s disease. Ann Clin Transl Neurol. 2017;4:700\u0026ndash;13. \u003c/li\u003e\n\u003cli\u003eKester MI, Teunissen CE, Crimmins DL, Herries EM, Ladenson JKH, Scheltens P, et al. Neurogranin as a cerebrospinal fluid biomarker for synaptic loss in symptomatic Alzheimer disease. JAMA Neurol. 2015;72:1275\u0026ndash;80. \u003c/li\u003e\n\u003cli\u003eSalvad\u0026oacute; G, Mil\u0026agrave;-Alom\u0026agrave; M, Shekari M, Minguillon C, Fauria K, Ni\u0026ntilde;erola-Baiz\u0026aacute;n A, et al. Cerebral amyloid-\u0026beta; load is associated with neurodegeneration and gliosis: Mediation by p-tau and interactions with risk factors early in the Alzheimer\u0026rsquo;s continuum. Alzheimer\u0026rsquo;s and Dementia. 2021;17:788\u0026ndash;800. \u003c/li\u003e\n\u003cli\u003eNilsson J, Gobom J, Sj\u0026ouml;din S, Brinkmalm G, Ashton NJ, Svensson J, et al. Cerebrospinal fluid biomarker panel for synaptic dysfunction in Alzheimer\u0026rsquo;s Disease. Alzheimer\u0026rsquo;s and Dementia: Diagnosis, Assessment and Disease Monitoring. 2021;13. \u003c/li\u003e\n\u003cli\u003eKamondi A, Grigg-Damberger M, L\u0026ouml;scher W, Tanila H, Horvath AA. Epilepsy and epileptiform activity in late-onset Alzheimer disease: clinical and pathophysiological advances, gaps and conundrums. Nat Rev Neurol. 2024;20:162\u0026ndash;82. \u003c/li\u003e\n\u003cli\u003eRachmian N, Medina S, Cherqui U, Akiva H, Deitch D, Edilbi D, et al. Identification of senescent, TREM2-expressing microglia in aging and Alzheimer\u0026rsquo;s disease model mouse brain. Nat Neurosci. 2024;27:1116\u0026ndash;24. \u003c/li\u003e\n\u003cli\u003eRueda‐Carrasco J, Sokolova D, Lee S, Childs T, Jurč\u0026aacute;kov\u0026aacute; N, Crowley G, et al. Microglia‐synapse engulfment via PtdSer‐TREM2 ameliorates neuronal hyperactivity in Alzheimer\u0026rsquo;s disease models. EMBO J. 2023;42:e113246. \u003c/li\u003e\n\u003cli\u003eRim C, You MJ, Nahm M, Kwon MS. Emerging role of senescent microglia in brain aging-related neurodegenerative diseases. Transl Neurodegener. 2024;13:10. \u003c/li\u003e\n\u003cli\u003eDas M, Mao W, Voskobiynyk Y, Necula D, Lew I, Petersen C, et al. Alzheimer risk-increasing TREM2 variant causes aberrant cortical synapse density and promotes network hyperexcitability in mouse models. Neurobiol Dis. 2023;186:106263. \u003c/li\u003e\n\u003cli\u003eTzioras M, Daniels MJD, Davies C, Baxter P, King D, McKay S, et al. Human astrocytes and microglia show augmented ingestion of synapses in Alzheimer\u0026rsquo;s disease via MFG-E8. Cell Rep Med. 2023;4:101175. \u003c/li\u003e\n\u003cli\u003eRim C, You MJ, Nahm M, Kwon MS. Emerging role of senescent microglia in brain aging-related neurodegenerative diseases. Transl Neurodegener. 2024;13. \u003c/li\u003e\n\u003cli\u003eDejanovic B, Sheng M, Hanson JE. Targeting synapse function and loss for treatment of neurodegenerative diseases. Nat Rev Drug Discov. Nature Research; 2024. p. 23\u0026ndash;42. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"molecular-neurodegeneration","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mond","sideBox":"Learn more about [Molecular Neurodegeneration](http://molecularneurodegeneration.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/mond/default.aspx","title":"Molecular Neurodegeneration","twitterHandle":"@MolNeuro","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Microglia, synaptic function, aging, neurodegeneration","lastPublishedDoi":"10.21203/rs.3.rs-6982788/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6982788/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eSynaptic homeostasis, maintained by microglia and astroglia, is disrupted throughout aging and early on in neurodegenerative diseases. Our aim was to study the relationship between TREM2-dependent microglial reactivity, astroglial response and synaptic dysfunction in two longitudinal cohorts of cognitively healthy volunteers and determine whether this relationship is influenced by AD core biomarkers.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eWe analyzed cross-sectional and longitudinal associations between cerebrospinal fluid levels of soluble TREM2 (sTREM2), astroglial markers (GFAP, S100B), and synaptic markers (neurogranin, α-synuclein) in cognitively unimpaired participants from the Wisconsin Registry for Alzheimer\u0026rsquo;s Prevention (WRAP) and the Alzheimer\u0026rsquo;s and Families (ALFA+) cohort. Biomarkers were quantified using validated immunoassays (NeuroToolKit, Roche), with sTREM2 measured using an in-house MSD-based assay in the WRAP cohort. Linear regression and linear mixed-effects models were used, both unadjusted and adjusted for Aβ42 and p-tau. Subgroup analyses were performed based on AT classification, \u003cem\u003eAPOE\u003c/em\u003e-ε4 status, and median splits of Aβ42/Aβ40 ratio and p-tau, to capture profiles suggestive of early AD-related neuropathogenesis.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eWe found significant cross-sectional associations between sTREM2 and α-synuclein, as well as between sTREM2 and S100B, in subgroups exhibiting AD-related biomarker profiles. Longitudinally, lower baseline neurogranin and α-synuclein and higher S100B predicted greater increases in sTREM2 over time independently of AD-related markers in the WRAP cohort (β = \u0026minus;0.02, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.006; β = \u0026minus;0.02, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.01; β\u0026thinsp;=\u0026thinsp;0.02, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;=\u0026thinsp;0.03, respectively). In ALFA+, lower baseline α-synuclein also predicted a greater subsequent longitudinal increase in sTREM2, but only among individuals with Aβ42/Aβ40 ratio above the median (β = -0.01, p\u0026thinsp;=\u0026thinsp;0.05). Notably, higher baseline sTREM2 was associated with a smaller longitudinal increase in neurogranin in both cohorts (β = -0.01, p\u0026thinsp;=\u0026thinsp;0.03 for WRAP, β = -0.01, p\u0026thinsp;=\u0026thinsp;0.04 in ALFA+).\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e\u003cp\u003eSynaptic dysfunction markers at baseline influence the longitudinal dynamics of CSF sTREM2 independently of AD-pathology related biomarkers throughout aging and earliest stages of neurodegeneration. In turn, higher baseline sTREM2 is associated with more stable neurogranin levels over time. These results suggest an independent interaction between synaptic dysfunction and TREM2-dependent microglial activation throughout aging and early neurodegeneration beyond AD pathology.\u003c/p\u003e","manuscriptTitle":"Synaptic dysfunction and glial activation markers throughout aging and early neurodegeneration: a longitudinal CSF biomarker-based study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-07-18 14:10:20","doi":"10.21203/rs.3.rs-6982788/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-08-20T20:07:03+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-20T20:00:19+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-08-05T01:27:17+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"255785990465351167000874458998609241067","date":"2025-07-21T19:57:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"135597252651546059855645999050439150826","date":"2025-07-17T12:42:04+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-07-14T13:44:45+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-07-10T17:26:22+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-06-27T13:20:07+00:00","index":"","fulltext":""},{"type":"submitted","content":"Molecular Neurodegeneration","date":"2025-06-26T10:57:25+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"molecular-neurodegeneration","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"mond","sideBox":"Learn more about [Molecular Neurodegeneration](http://molecularneurodegeneration.biomedcentral.com)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/mond/default.aspx","title":"Molecular Neurodegeneration","twitterHandle":"@MolNeuro","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"8bcfa8b4-647e-4f2a-9542-ec6cf3226efe","owner":[],"postedDate":"July 18th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-10-20T16:04:46+00:00","versionOfRecord":{"articleIdentity":"rs-6982788","link":"https://doi.org/10.1186/s13024-025-00901-5","journal":{"identity":"molecular-neurodegeneration","isVorOnly":false,"title":"Molecular Neurodegeneration"},"publishedOn":"2025-10-17 15:57:17","publishedOnDateReadable":"October 17th, 2025"},"versionCreatedAt":"2025-07-18 14:10:20","video":"","vorDoi":"10.1186/s13024-025-00901-5","vorDoiUrl":"https://doi.org/10.1186/s13024-025-00901-5","workflowStages":[]},"version":"v1","identity":"rs-6982788","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6982788","identity":"rs-6982788","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}