APOE ε4 carriers share immune-related proteomic changes across neurodegenerative diseases

Shvetcov, Artur; Johnson, Erik C. B.; Winchester, Laura M.; Walker, Keenan A.; Wilkins, Heather M.; Thompson, Terri G.; Rothstein, Jeffrey D.; Krish, Varsha; Imam, Farhad B.; Burns, Jeffrey M.; Swerdlow, Russell H.; Slawson, Chad; Finney, Caitlin A.

doi:10.1038/s41591-025-03835-z

Download PDF

Article
Open access
Published: 15 July 2025

APOE ε4 carriers share immune-related proteomic changes across neurodegenerative diseases

Nature Medicine volume 31, pages 2590–2601 (2025) Cite this article

61k Accesses
59 Citations
45 Altmetric
Metrics details

Subjects

Abstract

The APOE ε4 genetic variant is the strongest genetic risk factor for late-onset Alzheimer’s disease (AD) and is increasingly being implicated in other neurodegenerative diseases. Using the Global Neurodegeneration Proteomics Consortium SomaScan dataset covering 1,346 cerebrospinal fluid (CSF) and 9,924 plasma samples, we used machine learning-based proteome profiling to identify an APOE ε4 proteomic signature shared across individuals with AD, frontotemporal dementia (FTD), Parkinson’s disease dementia (PDD), Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS) and nonimpaired controls. This signature was enriched in pro-inflammatory immune and infection pathways as well as immune cells, including monocytes, T cells and natural killer cells. Analysis of the dorsolateral prefrontal cortex proteome for 262 donors from the Accelerating Medicines Partnership for AD UPenn Proteomics Study revealed a consistent APOE ε4 phenotype, independent of neurodegenerative pathology, including amyloid-β tau and gliosis for all diseases, as well as TDP-43 in ALS and FTD cases, and α-synuclein in PD and PDD cases. While systemic proteomic changes were consistent across APOE ε4 carriers, their relationship with clinical and lifestyle factors, such as hypertension and smoking, varied by disease. These findings suggest APOE ε4 confers a systemic biological vulnerability that is necessary but not sufficient for neurodegeneration, emphasizing the need to consider gene–environment interactions. Overall, our study reveals a conserved APOE ε4-associated pro-inflammatory immune signature persistent across the brain, CSF and plasma irrespective of neurodegenerative disease, highlighting a fundamental, disease-independent biological vulnerability to neurodegeneration. This work reframes APOE ε4 as a pleiotropic immune modulator rather than an AD-specific risk gene, providing a foundation for precision biomarker development and early intervention strategies across neurodegenerative diseases.

The Global Neurodegeneration Proteomics Consortium: biomarker and drug target discovery for common neurodegenerative diseases and aging

Article Open access 15 July 2025

Proteomic signatures of the APOE ε4 and APOE ε2 genetic variants and Alzheimer’s disease

Article Open access 15 May 2026

Cerebrospinal fluid proteomic signatures in cognitively normal individuals identify distinct clusters linked to neurodegeneration

Article 14 October 2025

Main

The ε4 variant of the apolipoprotein E (APOE ε4) gene is well-established as the largest genetic risk factor of late-onset Alzheimer’s disease (AD)¹. Growing evidence, however, indicates that APOE ε4 carriage may also have a role in other age-associated neurodegenerative diseases. Studies have linked APOE ε4 to increased risk and lower age of onset of frontotemporal dementia (FTD)^2,3,4,5,6, Parkinson’s disease (PD)^5,7,8 and amyotrophic lateral sclerosis (ALS)^9,10. APOE ε4 is also linked to a faster rate of cognitive decline and poor cognition in PD, increasing the risk of PD dementia (PDD)^{11,12,13,14,15,16}. Despite the deleterious impact of APOE ε4, little is known about the biological mechanisms underlying this effect and if, or how, it changes across the different neurodegenerative diseases. We recently showed that APOE ε4 carriers, irrespective of cognitive status in AD and mild cognitive impairment, had the same proteomic signature in the cerebrospinal fluid (CSF) associated with a pro-inflammatory immune molecular phenotype¹⁷. However, whether this extends to other neurodegenerative diseases is unknown.

To identify systemic proteomic changes associated with APOE ε4 carriers who develop neurodegenerative diseases, we used the Global Neurodegeneration Proteomics Consortium (GNPC) dataset. Here the plasma and CSF proteome were assessed using the SomaScan assay for 11,270 APOE ε4 carriers and noncarriers with AD, FTD, PD, PDD, ALS and nonimpaired controls. Using supervised machine learning, we identified and characterized an APOE ε4 proteome signature across the CSF and plasma. We then confirmed whether the same APOE ε4-enriched pathways in the periphery were mirrored in the brains of carriers and noncarriers. The dorsolateral prefrontal cortex (dlPFC) proteome of 262 AD, FTD, PDD, PD, ALS and nonimpaired control donors from the Accelerating Medicines Partnership for AD (AMP-AD) UPenn Proteomics Study was measured using label-free mass spectrometry (MS). In these samples, we also examined postmortem histopathological markers, including the presence of amyloid-β plaques, tau neurofibrillary tangles, gliosis and angiopathy. In FTD and ALS cases, we also examined TDP-43, while in PD and PDD, we examined α-synuclein. Lastly, to assess potential interactions between proteins in the APOE ε4 signature and environmental variables across the diseases, we performed a correlation network analysis between proteins and 18 clinical and lifestyle variables collected in the GNPC dataset, such as hypertension, smoking, and diabetes (Supplementary Table 1 and Fig. 1a).

Fig. 1: Study design and characterization of the CSF proteome signature in APOE ε4 carriers. — **Fig. 1: Study design and characterization of the CSF proteome signature in *APOE* ε4 carriers.**

Results

CSF proteome profiling reveals a distinct signature in APOE ε4 carriers

We used SomaScan (6,340 proteins measured per sample) proteomic data from the GNPC dataset to profile the CSF proteome of 526 AD, 247 PD and 573 nonimpaired control individuals (Supplementary Table 1). An initial principal component analysis (PCA) revealed that there was no clustering across all 6,340 proteins (Fig. 1b). We used mutual information (>0.01) to identify APOE ε4 proteins. Unlike traditional correlation coefficients, which capture only linear relationships, mutual information can detect both linear and nonlinear associations. This makes it well suited for complex high-dimensional biological data, where relationships between variables may not follow simple patterns. Using this, we identified 229 CSF proteins that were APOE ε4 associated in nonimpaired controls (Methods; Supplementary Table 2 and Extended Data Fig. 1a). A subsequent PCA showed that these proteins led to distinct clustering of groups based on APOE genotype (Fig. 1c) and the number of APOE ε4 alleles (Fig. 1d) independently of disease. This effect was further visualized using a heatmap showing that APOE ε4 proteins were upregulated or downregulated based on genotype but not disease (Fig. 1e). Many of these APOE ε4 proteins were also identified in our earlier work profiling the CSF proteome of APOE ε4 carriers with mild cognitive impairment and AD from the AD Neuroimaging Initiative cohort¹⁷ further highlighting the robustness and generalizability of our finding.

Using classification and regression trees (CART) modeling, we showed that these 229 CSF proteins were able to reliably (performance metrics > 0.95) predict APOE ε4 carriers from noncarriers across AD and PD (Extended Data Table 1 and Fig. 1f). We also found that there were no sex differences in the ability of our identified proteins to distinguish between APOE ε4 carriers and noncarriers (Extended Data Table 1 and Fig. 1f). To determine if the APOE ε4 signature might be due to differences in amyloid-β levels between carriers and noncarriers, we examined CSF amyloid-β A4 protein levels measured in the SomaScan assay. A Wilcoxon test indicated that there were no significant differences (P > 0.05) in CSF amyloid-β A4 protein levels between APOE ε4 carriers and noncarriers with AD, PD or nonimpaired controls (Extended Data Fig. 1b).

A functional enrichment analysis in the protein analysis through evolutionary relationships (PANTHER) database¹⁸ of the CSF APOE ε4 proteins revealed significant (false discovery rate (FDR) < 0.05) enrichment for viral processes, apoptosis, rhythmicity, cellular processes, protein phosphorylation and folding and RNA/DNA processes (Fig. 1g and Supplementary Table 3). Given that the most significant enrichment was observed for viral processes, we also performed enrichment analysis for immune pathway-specific processes. Immune-specific Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways revealed that APOE ε4 proteins were enriched in numerous infection-related pathways, including hepatitis, herpes, measles, Epstein–Barr virus (EBV), and influenza A. There was also significant enrichment for T cell, B cell and inflammatory signaling cascades, including Toll‑like receptor (TLR), tumor necrosis factor (TNF), interleukin 17 (IL-17), JAK/STAT and nuclear factor-κB (NF-κB; Fig. 1h and Supplementary Table 3). Using single-cell RNA-sequencing data from the Human Protein Atlas¹⁹, we performed an immune cell subtype enrichment analysis on APOE ε4 proteins. Across innate immune cells, APOE ε4 proteins were the most enriched in nonclassical and intermediate monocytes. In adaptive immune cells, memory CD8 T cells were the most enriched for APOE ε4 proteins, followed by T_regs and memory CD4 T cells. In innate-like T cells and lymphoid cells (mixed), both natural killer (NK) cells and γδ T cells showed APOE ε4 enrichment (Fig. 1i). Given that we found enrichment for hepatitis KEGG pathways, we also performed a cell-type-specific enrichment analysis in the liver. Across parenchymal cells, APOE ε4 proteins were the most enriched in hepatocytes and Kupffer cells, in line with both cell types being the primary producers of the APOE4 proteoform in the liver²⁰ (Fig. 1j).

These results show that APOE ε4 carriers have a distinct CSF proteomic signature characterized by enriched viral processes and pro-inflammatory immune pathways and cells. We find enrichment across hepatocytes and Kupffer cells in the liver, further implicating APOE4 proteoform synthesis sites in the periphery. This finding may also be reflective of brain-liver signaling and liver responses to neuroinflammation²¹. Notably, these changes were independent of neurodegenerative disease and sex, suggesting that APOE ε4 carriers share a common molecular phenotype.

Plasma proteome profiling reveals a similar APOE ε4-specific signature

We next sought to determine whether the APOE ε4 CSF proteome changes observed were also reflected in the plasma. Further leveraging the GNPC dataset, we performed plasma proteome profiling of 2,929 AD, 75 FTD, 169 PDD, 422 PD, 230 ALS and 6,099 nonimpaired control individuals with and without an APOE ε4 allele (Supplementary Table 1). A PCA of all 6,340 proteins revealed no group clustering (Fig. 2a). Using mutual information, we identified 58 plasma proteins in nonimpaired controls (Supplementary Table 4 and Extended Data Fig. 2a) that were associated with APOE genotype (Fig. 2b) and led to clustering based on the number of APOE ε4 alleles (Fig. 2c) rather than by neurodegenerative disease. A heatmap also revealed that the 58 APOE ε4 plasma changes were upregulated or downregulated based on genotype (Fig. 2d). Two of these proteins, TBCA and LRRN1, were also identified in the serum of healthy APOE ε4 centenarians²², providing further external validation of our finding and the importance of these proteins in APOE ε4 carriers.

Fig. 2: Identification and characterization of the plasma proteome signature in APOE ε4 carriers. — **Fig. 2: Identification and characterization of the plasma proteome signature in *APOE* ε4 carriers.**

CART models using these 58 proteins as predictors showed a strong ability to differentiate between APOE ε4 carriers and noncarriers across each of the neurodegenerative disease groups (performance metrics > 0.85; Extended Data Table 2). CART modeling revealed that there were no sex differences in the APOE ε4 plasma signature (Extended Data Table 2 and Fig. 2e). We extended this to show that there were no effects of race. Here we trained our models on White individuals and tested them using proteomic data from either Black/African American or American Indian/Alaskan Native individuals. In both cases, our models were reliably able to predict APOE ε4 carriers from noncarriers (Extended Data Table 2 and Fig. 2e). We also compared amyloid-β A4 protein levels between APOE ε4 carriers and noncarriers across neurodegenerative diseases to determine if the plasma APOE ε4 signature was related to amyloid-β pathology. A Wilcoxon test revealed that there were no significant differences in plasma amyloid-β A4 protein between APOE ε4 carriers and noncarriers (Extended Data Fig. 2b).

APOE ε4 plasma processes were significantly enriched for biological processes, including apoptosis, cellular processes, protein processes and RNA/DNA processes (Supplementary Table 5). As in the CSF, viral processes were the most significantly enriched biological process (Fig. 2f). This was further supported by similar KEGG immune and infection pathway enrichments, including EBV and hepatitis (Fig. 2g). There was also significant enrichment for inflammatory and cytokine signaling pathways, including TNF, IL-17, TLR and NF-κB (Fig. 2g and Supplementary Table 5). Among immune cell subtypes, nonclassical and intermediate monocytes were implicated in APOE ε4 proteins, as seen in the CSF. Unlike the CSF, however, basophils were enriched for APOE ε4 proteins. Of adaptive immune cell subtypes, plasma APOE ε4 proteins were enriched for memory CD8 T cells, T_regs and naive CD8 T cells. NK cells and γδ T cells were also associated with APOE ε4 (Fig. 2h). In the liver, we found cell-type-specific enrichment primarily for Kupffer cells and T cells and, unlike the CSF, very little enrichment for hepatocytes.

Our results demonstrate that genotype-specific proteomic changes observed in the CSF are also reflected in the plasma of APOE ε4 carriers and noncarriers, consistently indicating pro-inflammatory immune dysregulation across multiple pathways and immune cell populations. The APOE ε4 proteomic signature remained independent of neurodegenerative disease status and sex, with CART models showing strong predictive power for carrier status. We further showed that this signature generalizes across racial groups, underscoring its robustness and broad applicability. Unlike in the CSF, plasma APOE ε4 proteins were not enriched in hepatocytes but were significantly enriched in Kupffer and T cells, suggesting that peripheral immune activation is distinct from central nervous system (CNS) immune-metabolic signaling along the liver-brain axis in APOE ε4 carriers.

Key features of the peripheral immune signature are mirrored in the brains of APOE ε4 carriers

We then sought to determine whether the proteomic changes observed in the periphery were reflective of central changes in APOE ε4 carriers and to further validate our findings from the GNPC cohort. To do this, we leveraged label-free MS proteomic and postmortem histopathological data from the AMP-AD UPenn Proteomics study for the dlPFC of 49 AD, 31 FTD, 47 PDD, 33 PD, 55 ALS and 47 nonimpaired individual donors (Supplementary Table 6). SomaScan and label-free MS proteomic assays had different coverage of specific proteins; therefore, we focused on confirming enrichment for biological processes and pathways across the CSF, plasma, and dlPFC of carriers and noncarriers. We again used mutual information to identify APOE ε4 proteins from the dlPFC within each group independently. Across all neurodegenerative disease groups, we identified 248 APOE ε4 proteins (Supplementary Table 7). This was confirmed with PCAs showing clustering based on APOE genotype (Extended Data Fig. 3a–l and Supplementary Table 7). Functional enrichment analyses revealed that three of the main biological processes (viral processes, negative regulation of apoptosis, and protein folding) identified in the CSF and plasma were also significantly enriched in the dlPFC of APOE ε4 carriers across all neurodegenerative diseases (Fig. 3a,b and Supplementary Table 8). Further, four of the most significantly enriched KEGG immune pathways in the CSF and plasma were also identified in APOE ε4 carriers in a disease-independent manner, including EBV, hepatitis B, viral carcinogenesis and pathogenic Escherichia coli infection (Fig. 3b,c and Supplementary Table 8).

Fig. 3: Key overlapping enrichments across the CSF, plasma, and brains of APOE ε4 carriers across neurodegenerative diseases. — **Fig. 3: Key overlapping enrichments across the CSF, plasma, and brains of *APOE* ε4 carriers across neurodegenerative diseases.**

Previous research suggests that APOE ε4 carriers may develop neurodegenerative pathology, including amyloid-β plaques and tau neurofibrillary tangles, even in the absence of symptomatic disease^{8,23,24,25,26}. We therefore sought to determine if the changes seen in APOE ε4 carriers may be due to the presence of APOE ε4-driven brain pathology. There was no relationship between APOE ε4 and tau or thio-S-positive plaques across nonimpaired controls and disease groups (Supplementary Table 9). In FTD, PDD, and PD, there was also no relationship between APOE ε4 and α-synuclein and, in FTD and ALS, no relationship with TDP-43 (Supplementary Table 9). Although none of the neurodegenerative disease groups showed a relationship between APOE ε4 and angiopathy, there was a small statistically significant association found in nonimpaired controls (Supplementary Table 9).

Taken together, these findings suggest a relationship between peripheral and central proteomic changes. Both are characterized by pro-inflammatory immune responses, further highlighting that APOE ε4 carriers have systemic immune dysregulation. This demonstrates that the major proteome changes seen in the periphery of APOE ε4 carriers are indeed reflective of ongoing processes in the brain and that peripheral processes may actively contribute to or even drive changes in the brain proteome. Notably, APOE ε4-associated changes are independent of neurodegenerative disease, whether measured simply by diagnosis or by specific pathological changes in the brain.

Peripheral APOE ε4-associated proteins are differentially correlated to clinical variables according to disease

Given that APOE ε4 carriers across neurodegenerative diseases share major underlying systemic proteomic changes, we sought to better understand additional drivers that may be disease specific. To do this, we leveraged the GNPC dataset and a correlation network analysis to identify the relationship between APOE ε4 proteins and demographic, clinical, and lifestyle variables, such as age, blood pressure, and smoking (Supplementary Table 1 and see Extended Data Fig. 4 for a stratification by comorbidities for APOE ε4 carriers and noncarriers). Of the 40 proteins that overlapped across the CSF and plasma of APOE ε4 carriers, we chose 16 of these that had more than 20 functional connections (Fig. 4a and Supplementary Table 10). These proteins represent central ‘nodes’ in the protein–protein network, suggesting that any changes are more likely to disrupt diverse pathways and functions²⁷. Of note, two of the overlapping proteins, APOE and PHGDH, were differentially expressed in opposite directions across the plasma and CSF. Both proteins were upregulated in the CSF but downregulated in the plasma. As we had a higher number of individuals with plasma samples, thereby improving statistical power, the correlation analysis reflects the decreased APOE and PHGDH levels observed in plasma.

Fig. 4: Correlation analyses between APOE ε4 CSF and plasma central node proteins and demographic, lifestyle, and clinical variables. — **Fig. 4: Correlation analyses between *APOE* ε4 CSF and plasma central node proteins and demographic, lifestyle, and clinical variables.**

We first calculated Spearman’s rank correlation (continuous variables) or correlation ratio (categorical variables) for APOE and demographic, clinical, and lifestyle variables in nonimpaired, AD, PDD and PD APOE ε4 carriers. Due to low numbers, we were unable to calculate similar correlations for FTD and ALS. Our analyses revealed unique, neurodegenerative disease-specific significant relationships with APOE (Fig. 4b and Supplementary Table 11). APOE was associated with sex and race in AD, with diabetes in PD, with resting heart rate and chronic obstructive pulmonary disease (COPD) in PDD, and with age, body mass index, and hypertension in nonimpaired controls. A key limitation is the lack of longitudinal data for nonimpaired APOE ε4 carriers, who may later develop neurodegenerative disease. Thus, observed associations with clinical variables may reflect preclinical disease. We extended these correlation analyses to the remaining 15 central node APOE ε4 proteins. This analysis further identified that APOE ε4 central node proteins are significantly correlated (P < 0.05) with demographic, clinical, and lifestyle variables in a disease-specific way in APOE ε4 carriers (Fig. 4c and Supplementary Table 11).

Together, these findings indicate that although all APOE ε4 carriers have the same underlying proteomic signature, specific proteins within this signature are differentially correlated with demographic, lifestyle, and clinical variables. Critically, these correlations were neurodegenerative disease specific.

Discussion

The GNPC dataset represents a substantial advancement in neurodegenerative disease research, by providing a real-world clinical proteomic dataset that comprises over 35,000 (11,270 with APOE genotype) individuals across AD, FTD, PDD, PD, ALS and normal aging from across more than 20 clinical sites in the US, UK and Europe²⁸. This enabled us to ask whether the APOE ε4-associated proteomic signature is shared across multiple neurodegenerative diseases. Our results demonstrate that all APOE ε4 carriers, irrespective of neurodegenerative disease, have a unique proteome signature that extends across the plasma and CSF. Unlike prospectively designed cohorts, the GNPC dataset reflects real-world clinical heterogeneity, highlighting the robustness and generalizability of our findings. This signature is associated with pro-inflammatory immune dysregulation and an enrichment for circulating immune cells, including monocytes, memory CD8 and γδ T cells, T_regs and NK cells. This molecular phenotype extends to the brains of APOE ε4 carriers in a similar disease-independent manner and is not associated with the presence of any disease-specific brain pathology (Fig. 5). Although all APOE ε4 carriers have a systemic immune-related proteome signature, we find that the relationships between proteins within this signature are uniquely associated with demographic, lifestyle, and clinical variables in a neurodegenerative disease-specific manner. Notably, this suggests that although the biological changes associated with APOE ε4 carriage are essential for neurodegeneration, broadly, interactions with underlying biological vulnerability and the environment may be key for driving the pathogenesis of the specific neurodegenerative disease.

**Fig. 5: Summary of the study’s findings.**

There is evidence that the APOE ε4 genotype is a modern-day example of antagonistic pleiotropy. In younger adults, APOE ε4 is associated with increased survival and fertility in environments with high levels of infectious disease^29,30,31,32. For example, healthy individuals who are APOE ε4 heterozygotes exhibit heightened cytokine release, increased plasma TNF levels, a more pronounced hyperthermic response and an earlier onset of IL-6 production following immune challenge with TLR2/TLR4/TLR5 ligands or lipopolysaccharide³³. While this immune response protects younger APOE ε4 carriers from infectious diseases, prolonged states of inflammation and cytokine release are likely deleterious with age^34,35. Although this study identified a proteomic signature indicative of a pro-inflammatory phenotype, a limitation is the absence of direct measures of routine inflammatory markers such as C-reactive protein or cytokines. Future studies should incorporate these markers in prospectively designed cohorts to clarify their association with APOE ε4.

To date, the biological effects of APOE ε4 carriage have largely only been studied in the context of AD. A notable finding of our study is that the pro-inflammatory molecular phenotype associated with APOE ε4 extends to individuals with other neurodegenerative diseases, including FTD, PDD, PD and ALS. This raises two key considerations. First, our findings underscore the need to shift focus from the continued identification of genetic risk loci via genome-wide association studies toward functional characterization of established variants³⁶. Notably, the absence of a statistically significant association between a variant and a specific disease phenotype does not preclude biological relevance. Despite APOE ε4 being overrepresented in the AD cohort, we observe a consistent molecular signature associated with APOE ε4 across multiple neurodegenerative diseases, highlighting an underrecognized role for this variant beyond AD that may have been overlooked due to its historically strong link with AD risk. Second, our data support reconceptualizing APOE ε4 not only as a disease-specific risk factor but also as a broader susceptibility allele contributing to shared pathogenic mechanisms across neurodegenerative diseases. It remains unclear, however, why APOE ε4 is more strongly associated with AD in terms of prevalence despite exhibiting systemic biological effects across neurodegenerative diseases. One possibility is that interactions between APOE ε4 and additional age-related, environmental, or comorbid factors may selectively amplify neurodegenerative pathways characteristic of AD. Another possibility is that CNS-specific vulnerabilities or cellular contexts (for example, in hippocampal circuits or dopaminergic neurons) modulate how the APOE ε4 inflammatory phenotype manifests clinically. Thus, while APOE ε4 confers a shared biological susceptibility, disease expression likely depends on a combination of genetic background, cellular context(s), and lifetime exposures. Large-scale integrative efforts, such as the GNPC, which harmonize data across distinct disease cohorts into unified datasets, provide a powerful framework for advancing this line of inquiry. By enabling cross-disease comparisons, such efforts may help delineate modifiers that influence why some APOE ε4 carriers develop AD while others remain healthy or develop different neurodegenerative diseases. This has important implications for prognosis and risk stratification in midlife individuals who carry APOE ε4.

A limitation of our study is the absence of validated biomarkers (for example, CSF p-tau217) to confirm clinical diagnoses, which reflects both the nature of the GNPC dataset and global heterogeneity in clinical practice²⁸. However, several features mitigate concerns regarding potential misdiagnosis. First, the APOE ε4 signature was derived from nonimpaired individuals. While it is possible that a minority of these nonimpaired individuals harbored asymptomatic pathology (for example, asymptomatic AD), the majority would be free of overt disease, reducing the likelihood of confounding results. Second, the consistent presence of the signature across all APOE ε4 carriers, irrespective of clinical diagnosis, supports its generalizability and suggests it reflects a broader APOE ε4-related biological phenotype rather than disease-specific changes. Finally, we validate our findings using postmortem brain proteomics and histopathology, where diagnostic certainty is highest. Here APOE ε4 status was not associated with hallmark pathologies, including amyloid-β, tau, TDP-43 or α-synuclein, across respective disease groups. This postmortem validation reinforces the robustness of our findings. Future studies would benefit from prospective cohorts incorporating validated CSF or plasma biomarkers to confirm and extend these observations.

An unexpected finding of our study was that plasma neurofilament light (NEFL) levels were lower in APOE ε4 carriers across neurodegenerative diseases, despite NEFL’s growing recognition as a biomarker of neurodegeneration³⁷. Prior studies have reported conflicting results—some found no association^38,39, another reported increased levels in APOE ε4 carriers⁴⁰, while others using the SomaScan⁴¹ and Simoa⁴² assays observed decreased levels, consistent with our findings. These discrepancies may reflect differences in sample size, with studies reporting decreased NEFL levels generally including larger cohorts (~600 to 5,000 participants, and 9,924 in our study). Alternatively, APOE ε4-related blood–brain barrier (BBB) or metabolic dysfunction may alter peripheral clearance of NEFL and affect its plasma or CSF concentration. These results raise important questions about the reliability of NEFL as a stand-alone biomarker and underscore the need for mechanistic studies and a shift toward precision biomarkers that integrate genetic and environmental context.

In our study, key peripheral inflammatory states in APOE ε4 carriers were mirrored in the CNS. Notably, brain proteomics performed using label-free MS orthogonally validated findings from SomaScan-based CSF and plasma analyses. While SomaScan aptamer technology offers high-throughput protein quantification, it is relatively insensitive to proteoform diversity, including post-translational modifications⁴³, a limitation when examining signaling pathways and protein networks where such modifications are functionally significant. In contrast, MS enables the detection of broader proteoforms, offering an independent assessment of protein abundance and pathway enrichment. The concordance of APOE ε4-associated protein signatures and enrichment patterns across plasma, CSF and brain proteomic datasets suggests that our findings are not artifacts of platform-specific biases but likely reflect underlying biology. Supporting this interpretation, we observed enrichment of APOE ε4-associated proteins in both hepatocytes and Kupffer cells in the CSF, but only in Kupffer cells and T cells in plasma. This pattern implies that plasma reflects chronic peripheral immune activation, whereas the CSF may capture liver-brain axis-mediated inflammatory signaling in APOE ε4 carriers²¹. Further experimental studies are warranted to validate these mechanisms. Prospective cohort studies incorporating cross-platform, multi-tissue analyses from the same individuals will be essential to confirm and expand upon these observations.

While the exact mechanisms underlying the mirroring of peripheral and central pro-inflammatory states remain unclear, they may involve interactions between pro-inflammatory peripheral immune cells and the BBB. A recent study of individuals with long COVID-19 demonstrated that hyperactive peripheral blood mononuclear cells adhere to the endothelial cells of the BBB, driving inflammation, degradation, and symptoms of brain fog⁴⁴. In support of this, cognitively healthy APOE ε4 carriers exhibit early markers of BBB dysfunction in the hippocampus and medial temporal lobe⁴⁵. As the BBB becomes increasingly compromised, it allows the infiltration of blood-derived proteins and pro-inflammatory immune cells into the brain, contributing to exacerbated neuroinflammation and neurodegeneration⁴⁶. In our study, we identify enriched pro-inflammatory immune cell subpopulations, including nonclassical and intermediate monocytes^47,48, memory CD8⁺ T cells⁴⁹, γδ T cells⁵⁰ and NK cells^51,52. These immune cell types may represent peripheral blood mononuclear cells that interact with the BBB to promote neurodegeneration in APOE ε4 carriers. Given that APOE ε4 is linked to a shared molecular phenotype across neurodegenerative diseases, this hypothesis is consistent with current understanding. Healthy, nonimpaired APOE ε4 carriers have a systemic pro-inflammatory phenotype¹⁷ (and as we have shown here), early signs of BBB disruption⁴⁵, alterations in brain activity and connectivity^53,54, and sleep disturbances⁵⁵, all of which are known to be risk factors for neurodegenerative diseases, broadly^56,57,58. Further supporting this is the recent evidence that vaccinations protect against dementia^59,60, including the shingles vaccines Zostavax⁶¹ and Shingrix⁶². Although none of these studies explicitly examined the interactive effects between vaccination status and APOE ε4 genotype, the fact that more than 60% of individuals with dementia are APOE ε4 carriers⁶³ suggests that APOE ε4 may represent a critical modifier of the observed associations, warranting further investigation. Future studies in human disease-relevant models, such as patient stem cell-derived organoids, are needed to further elucidate these mechanisms.

If all APOE ε4 carriers share a common underlying biological vulnerability to neurodegeneration, an important question is what determines which specific neurodegenerative disease they develop. This likely reflects complex interactions between genetic risk and environmental exposures⁶⁴. In our study, proteins within the APOE ε4 signature were differentially correlated with demographic, lifestyle, and clinical variables, in a neurodegenerative disease-specific manner. This suggests that genotype–environment interactions may modulate disease trajectories in APOE ε4 carriers. While limited by the variables consistently collected across the GNPC’s global real-world cohorts, our findings highlight meaningful associations that warrant further investigation. Future prospective studies should incorporate more detailed environmental and behavioral measures, such as physical activity, diet, sleep, substance use and immunization history, to better characterize modifiers of neurodegenerative disease risk in APOE ε4 carriers. Given the cross-sectional nature of our study, we cannot infer causality or perform mediation analyses, as clinical variables like hypertension may have arisen after disease onset. One plausible mechanism is that APOE ε4 carriers exhibit heightened inflammatory responses to environmental or pathological stressors, including comorbidities such as hypercholesterolemia and ischemic heart disease, to which they are predisposed^65,66,67. This suggests that the pro-inflammatory phenotype of APOE ε4 carriers may be, at least in part, independent of cardiovascular comorbidities. Our findings further support this interpretation. Our PANTHER biological processes enrichment analyses of the APOE ε4 CSF proteome showed no significant enrichment of cardiovascular-related processes. In the plasma, we found only a single enrichment for heart development; however, this was modest relative to the top enrichment of viral processes. These results suggest that additional environmental or pathological insults may act synergistically with a genetically primed pro-inflammatory state to drive neurodegeneration in APOE ε4 carriers. Another interpretation of our data, however, is that individuals with pre-existing neurodegenerative diseases are more likely to develop comorbidities^68,69, such as hypertension, which may be a consequence of neurodegenerative processes rather than a precursory event. Longitudinal studies tracking APOE ε4 carriers over time will be essential to disentangle causal pathways and advance personalized prevention strategies.

Neurodegenerative disease risk is reflective of a complex polygenic architecture and co-occurring variants in genes, such as TREM2, MAPT, GRN, GBA and LRRK2, may modulate or interact with APOE ε4-associated molecular pathways. For example, joint carriage of APOE ε4 and MAPT mutations resulted in a significantly lower age of FTD onset⁶. APOE ε4 and GBA and LRRK2 carriers also have the highest risk of PDD¹³. Genome-wide genetic data beyond APOE genotype, however, are not currently available within the GNPC cohort and therefore could not be included in the present study. Despite this, a key strength of our study is the demonstration that a well-established risk variant for one neurodegenerative disease (APOE ε4 in AD) also exerts a conserved molecularsignature across multiple neurodegenerative conditions. These findings provide proteomic evidence that a single risk variant is implicated across the neurodegenerative disease spectrum. This indicates the need for a conceptual shift in the field, moving away from viewing genetic risk variants as disease specific and toward recognizing their potential as shared, pleiotropic modifiers of neurodegenerative vulnerability. Future work, therefore, would greatly benefit from a focus on the role of other genetic variants across the neurodegenerative disease spectrum, even if they are enriched in only one disease.

Overall, our study provides a conceptual and translational advance in understanding APOE ε4-mediated risk for neurodegeneration. We identify a conserved, systemic pro-inflammatory immune proteomic signature associated with APOE ε4 across plasma, CSF and brain, irrespective of neurodegenerative disease or pathology. Notably, this signature is also present in asymptomatic individuals, suggesting it precedes clinical symptom onset and reflects an intrinsic biological vulnerability. The ability to detect this signature in plasma, a minimally invasive and clinically accessible biofluid, also offers promising opportunities for blood-based precision biomarkers to identify individuals at risk before symptom onset. Together, these findings demonstrate that APOE ε4 confers an intrinsic biological vulnerability characterized by a chronic pro-inflammatory immune phenotype that is independent of downstream disease processes. While this chronic immune activation may predispose APOE ε4 carriers to neurodegeneration, it is unlikely to be sufficient alone to drive the development of neurodegenerative disease. Instead, the transition to clinical disease likely involves complex interactions between genetic susceptibility and modifiable environmental, lifestyle, and clinical factors. By linking the APOE ε4 signature to such factors, our findings provide actionable insights for early intervention and prevention. They also establish a roadmap for future exploration of gene–gene and gene–environment interactions, including polygenic and epistatic effects. More broadly, our findings reframe APOE ε4 as a pleiotropic immune modulator rather than an AD-specific gene, underscoring the need for cross-disease, genetically informed models of risk stratification and therapy. Finally, this work lays the foundation for future wet laboratory-based mechanistic studies to elucidate how APOE ε4 shapes immune function and neuroinflammatory signaling, a critical step toward developing targeted and individualized immunotherapies for early and preclinical disease.

Methods

Participants

GNPC cohort

The GNPC cohort represents the largest collection of SomaScan proteomic data for individuals with neurodegenerative diseases and nonimpaired controls sourced from study sites across the USA, UK and Europe. In the present study, we included 11,270 individuals from the GNPC cohort, with 6,672 nonimpaired controls, 3,455 AD, 75 FTD, 169 PDD, 669 PD and 230 ALS individuals. Of these, 4,325 individuals were APOE ε4 carriers (either heterozygous or homozygous) and 6,945 were non-APOE ε4 carriers. All included individuals in the study provided either CSF or plasma samples, but none provided both. Individuals were diagnosed based on diagnostic criteria from each research group, as described in the GNPC cohort summary paper¹. Cognitive impairment for AD and PDD patients was further assessed using a clinical dementia rating score of ≥1, Mini-Mental State Exam score of ≤24 and/or Montreal Cognitive Assessment score of ≤23. Most individuals with AD had cognitive impairment scores indicative of mild or moderate AD. We excluded individuals with mild cognitive impairment in the current study due to its heterogeneity as a clinical diagnosis, and because it does not necessarily reflect neurodegenerative pathology. CSF or plasma samples were collected from all participants at a single timepoint along with demographic and clinical variables (Supplementary Table 1). Participants from each of the included study sites in the GNPC cohort provided written informed consent and studies were approved by the relevant institution’s ethics committee²⁸.

AMP-AD UPenn Proteomics Study cohort

The AMP-AD UPenn Proteomics Study is a cohort of autopsy-collected samples from the dlPFC of 49 AD, 31 FTD (with TDP-43 inclusions), 47 PDD, 33 PD, 55 ALS and 47 nonimpaired individual donors from the University of Pennsylvania School of Medicine Brain Bank (https://www.synapse.org/Synapse:syn21438414). Individual diagnoses were confirmed through postmortem neuropathological evaluation for neuritic plaque distribution according to the Consortium to Establish a Registry for AD criteria⁷⁰ and neurofibrillary tangle pathology according to the Braak staging system⁷¹. Other neuropathologic assessments, including for α-synuclein, TDP-43, gliosis and angiopathy, were made in line with established criteria and guidelines^72,73. Demographic variables for the cohort are listed in Supplementary Table 6. Human postmortem tissues were acquired under proper institutional review board protocols.

Proteomics

GNPC cohort

CSF and plasma proteomics in the GNPC cohort were performed using the SomaScan v4.1 assay, which measures approximately 7,000 proteins^74,75. This technology uses aptamer-based approaches, specifically slow off-rate modified aptamers, which incorporate chemically modified nucleotides that bind with high specificity and affinity to target proteins⁷⁴. SomaLogic provides ‘raw’ data that has been standardized, normalized, and calibrated, including using adaptive normalization by maximum likelihood. Protein measurements are provided in relative fluorescent units. Aptamers were mapped to Uniprot before being included in the GNPC cohort dataset. Details on the creation and harmonization of this dataset are described elsewhere²⁸. Before our own analyses, we log₂ transformed and standardized training and testing datasets separately.

AMP-AD UPenn Proteomics Study cohort

Proteins from postmortem brain homogenates from the dlPFC were measured using label-free MS on a Q-Exactive Plus mass spectrometer as previously described⁷⁶. Proteins included in this study were identified using MaxQuant’ MaxLFQ (label-free quantification) algorithm. Normalized label-free quantitation protein intensities were extracted from the proteinGroups.txt MaxQuant output file using a custom R script (LoaderAndBatchCorrection-UPenn354cases.R)⁷⁶. To account for missing protein quantitation values in the dataset, we imputed values using the median for proteins with ≤20% missing values and excluded those proteins with ≥20% missing values.

Statistical analyses

Feature selection

We identified APOE ε4-associated proteins using mutual information, a statistical measure that quantifies the amount of information one variable provides about another⁷⁷. In the context of our study, it measures how informative each protein is for distinguishing APOE ε4 carriers from noncarriers. A mutual information score of zero indicates no association, while higher values indicate greater dependency between a protein and APOE ε4 status. We used a cutoff of ≥0.01 to indicate proteins that were likely to be associated with APOE ε4 carrier status. For CSF and plasma, we used nonimpaired controls to identify APOE ε4-associated proteins. This ensured that we were identifying proteins in only those individuals who are not symptomatic for any one neurodegenerative disease. Appropriate feature selection (identification of proteins specifically associated with APOE ε4) was further confirmed using PCA and heatmap. In the brain, we used mutual information to identify APOE ε4 proteins specific to each group of individuals. PCA was then used to confirm that within each group, APOE ε4-identified proteins clustered based on APOE ε4 carrier status. For visualization of differential protein abundance, volcano plots were constructed using log₂ (fold change(FC)) on the x axis and −log₁₀-transformed adjusted P values on the y axis. Plots show fold change and significance thresholds that were applied at log₂(FC) = ± 0.585 (corresponding to a 1.5-fold change) and FDR-adjusted P < 0.05, respectively. Mutual information was calculated in R (v4.4.1) using the package ‘FSelectorRcpp’⁷⁸ and PCA plots and heatmaps were made using ‘ggplot2’⁷⁹ and ‘pheatmap’, respectively.

CART

The generalizability of our identified APOE ε4 proteins across different neurodegenerative diseases was tested. We performed this analysis using supervised machine learning with CART via the ‘caret’ package in R (v4.4.1). For all CART models, we first split the dataset into a 70% training and validation dataset and a 30% withheld (unseen) testing dataset. Model training and evaluation were conducted using ‘caret’ with a fivefold cross-validation procedure repeated five times. Where classes (groups) were imbalanced, we addressed this by using upsampling. Model performance was assessed using the 30% unseen dataset and several metrics, including sensitivity, specificity, positive predictive value, negative predictive value and area under the curve. To identify potential sex-specific differences in the generalizability of APOE ε4 proteins, we trained and tested the CART models on a mixed sample of male and female APOE ε4 carriers and noncarriers. To assess any race-specific effects in plasma, we trained the CART models on White APOE ε4 carriers and noncarriers and then tested them using only individuals of another race (Black/African American or American Indian/Alaskan Native).

Linear regression

To identify a potential relationship between APOE ε4 carrier status and postmortem histopathology in the dlPFC, we used a linear regression. Postmortem histopathology data for all donors included tau, thio-S-positive plaques, and angiopathy. Additional measures for AD and nonimpaired donors included antibody-positive plaques, Thal amyloid score, and gliosis. FTD, PDD and PD donors also had postmortem histopathological assessments for α-synuclein. FTD and ALS donors included TDP-43 assessments.

Correlation network analysis

To determine a potential relationship between overlapping CSF and plasma proteins identified as being APOE ε4-specific and demographic, clinical and lifestyle variables, we performed a correlation network analysis. Protein levels were based on those measured in the plasma, although only overlapping APOE ε4 identified across both CSF and plasma were included. Before computing any correlations, rows with missing values or encoded as an indeterminate response (for example, ‘unsure’, coded as −1 or NA) were excluded on a pairwise basis. This ensured that each test was conducted only on complete and valid data. For associations between two continuous variables, we used Spearman’s rank correlation coefficient (ρ). For associations between categorical variables and continuous protein expression values, we employed a correlation ratio (η²), which measures the proportion of variance in the continuous variable explained by the grouping variable. This was calculated by fitting a linear model with the continuous variable as the outcome and the categorical variable as the predictor. The coefficient of determination (R²) from the model was interpreted as η². The statistical significance of the association was determined using ANOVA (P < 0.05). Categorical variables included binary indicators (for example, presence/absence of comorbidities) as well as variables with more than two levels (for example, ‘yes’, ‘no’, ‘unsure’). All categorical variables were explicitly converted to factors, and associations were computed only where at least two factor levels were present in the subset of complete cases. The correlation network analysis figure was created using GraphPad Prism (v10.0.0 for Windows).

Enrichment analyses

Protein–protein enrichment analyses to assess the functions enriched for APOE ε4 proteins were done using NetworkAnalyst (v3.0)^80,81,82. Here generic protein–protein interactions were identified using a first-order network from the International Molecular Exchange Consortium interactome database⁸³ and InnateDB⁸⁴. Network enrichment for biological processes was performed using the PANTHER classification system¹⁸ and KEGG pathways⁸⁵. Significance was determined by an FDR of >0.05.

Immune cell-type-specific enrichment analyses for immune and liver cells were done using single-cell RNA-sequencing data from the Human Protein Atlas (https://www.proteinatlas.org/, v23; Ensembl v109)¹⁹. For each APOE ε4 protein, we identified the corresponding protein-coding transcripts per million counts. We then normalized the expression for each cell type using min–max scaling and used heatmaps to visually represent these enrichments. Enrichment bar graphs and heatmaps were created using GraphPad Prism (v10.0.0 for Windows).

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The harmonized GNPC data used to generate these findings was provided to Consortium Members in June 2024 and will be made available for public request by the AD Data Initiative by 15 July 2025. Members of the global research community will be able to access the metadata and place a data use request via the AD Discovery Portal (https://discover.alzheimersdata.org/). Access is contingent on adherence to the GNPC Data Use Agreement and the Publication Policies.

The AMP-AD UPenn Proteomics Study data is available through the AD Knowledge Portal (https://adknowledgeportal.synapse.org/). Researchers who wish to access this controlled dataset are required to submit a Data Use Agreement. More information can be found here: https://adknowledgeportal.synapse.org/Data%20Access.

Code availability

All code used for analyzing the proteomic and clinical data are available at https://github.com/Art83/gnpc_apoe.

References

Belloy, M. E., Napolioni, V. & Greicius, M. D. A quarter century of APOE and Alzheimer’s disease: progress to date and the path forward. Neuron 101, 820–838 (2019).
Article CAS PubMed PubMed Central Google Scholar
Fabre, S. F. et al. Clinic-based cases with frontotemporal dementia show increased cerebrospinal fluid tau and high apolipoprotein E ε4 frequency, but no tau gene mutations. Exp. Neurol. 168, 413–418 (2001).
Article CAS PubMed Google Scholar
Bernardi, L. et al. The effects of APOE and tau gene variability on risk of frontotemporal dementia. Neurobiol. Aging 27, 702–709 (2006).
Article CAS PubMed Google Scholar
Rubino, E. et al. Apolipoprotein E polymorphisms in frontotemporal lobar degeneration: a meta-analysis. Alzheimer’s Dement. 9, 706–713 (2013).
Article Google Scholar
Ferrari, R. et al. Genetic architecture of sporadic frontotemporal dementia and overlap with Alzheimer’s and Parkinson’s diseases. J. Neurol. Neurosurg. Psychiatry 88, 152–164 (2017).
Article PubMed Google Scholar
Koriath, C. et al. ApoE4 lowers age at onset in patients with frontotemporal dementia and tauopathy independent of amyloid-β copathology. Alzheimer’s Dement. (Amst.) 11, 277–280 (2019).
Article Google Scholar
Li, Y. J. et al. Apolipoprotein E controls the risk and age at onset of Parkinson’s disease. Neurology 62, 2005–2009 (2004).
Article CAS PubMed Google Scholar
Ibanez, L. et al. Functional genomic analyses uncover APOE-mediated regulation of brain and cerebrospinal fluid β-amyloid levels in Parkinson disease. Acta Neuropathol. Commun. 8, 196 (2020).
Article CAS PubMed PubMed Central Google Scholar
Drory, V. E., Birnbaum, M., Korczyn, A. D. & Chapman, J. Association of APOE ε4 allele with survival in amyotrophic lateral sclerosis. J. Neurol. Sci. 190, 17–20 (2001).
Article CAS PubMed Google Scholar
Zetterberg, H., Jacobsson, J., Rosengren, L., Blennow, K. & Andersen, P. A. Association of APOE with age at onset of sporadic amyotrophic lateral sclerosis. J. Neurol. Sci. 273, 67–69 (2008).
Article CAS PubMed Google Scholar
Real, R. et al. Association between the LRP1B and APOE loci and the development of Parkinson’s disease dementia. Brain 146, 1873–1887 (2022).
Article PubMed Central Google Scholar
Mata, I. F. et al. APOE, MAPT, and SNCA genes and cognitive performance in Parkinson disease. JAMA Neurol. 71, 1405–1412 (2014).
Article PubMed PubMed Central Google Scholar
Szwedo, A. A. et al. GBA and APOE impact cognitive decline in Parkinson’s disease: a 10-year population-based study. Mov. Disord. 37, 1016–1027 (2022).
Article CAS PubMed PubMed Central Google Scholar
Tan, M. M. X. et al. Genome-wide association studies of cognitive and motor progression in Parkinson’s disease. Mov. Disord. 36, 424–433 (2021).
Article CAS PubMed Google Scholar
Tsuang, D. et al. APOE ε4 increases risk for dementia in pure synucleinopathies. JAMA Neurol. 70, 223–228 (2013).
Article PubMed PubMed Central Google Scholar
Jo, S. et al. The role of APOE in cognitive trajectories and motor decline in Parkinson’s disease. Sci. Rep. 11, 7819 (2021).
Article CAS PubMed PubMed Central Google Scholar
Shvetcov, A. et al. Proteome profiling of cerebrospinal fluid using machine learning shows a unique protein signature associated with APOE4 genotype. Aging Cell 24, e14439 (2025).
Article CAS PubMed Google Scholar
Mi, H., Muruganujan, A., Casagrande, J. T. & Thomas, P. D. Large-scale gene function analysis with PANTHER classification system. Nat. Protoc. 8, 1551–1566 (2019).
Article Google Scholar
Uhlen, M. et al. A genome-wide transcriptomic analysis of protein-coding genes in human blood cells. Science 366, 6472 (2019).
Article Google Scholar
Martinez-Martinez, A. B. et al. Beyond the CNS: the many peripheral roles of APOE. Neurobiol. Dis. 138, 104809 (2020).
Article CAS PubMed PubMed Central Google Scholar
Butterworth, R. F. The liver-brain axis in liver failure: neuroinflammation and encephalopathy. Nat. Rev. Gastroenterol. Hepatol. 10, 522–528 (2013).
Article CAS PubMed Google Scholar
Sebastiani, P. et al. A serum protein signature of APOE genotypes in centenarians. Aging Cell 18, e13023 (2019).
Article CAS PubMed PubMed Central Google Scholar
Fouquet, M., Besson, F. L., Gonneaud, J., La Joie, R. & Chetelat, G. Imaging brain effects of APOE4 in cognitively normal individuals across the lifespan. Neuropsychol. Rev. 24, 290–299 (2014).
Article PubMed Google Scholar
Morris, J. C. et al. APOE predicts amyloid-beta but not tau Alzheimer pathology in cognitively normal aging. Ann. Neurol. 67, 122–131 (2010).
Article CAS PubMed PubMed Central Google Scholar
Mishra, S. et al. Longitudinal brain imaging in preclinical Alzheimer disease: impact of APOE ε4 genotype. Brain 141, 1828–1839 (2018).
Article PubMed PubMed Central Google Scholar
Robinson, J. L. et al. Neurodegenerative disease concomitant proteinopathies are prevalent, age-related, and APOE4-associated. Brain 141, 2181–2193 (2018).
Article PubMed PubMed Central Google Scholar
Han, J.-D. J. et al. Evidence for dynamically organized modularity in the yeast protein–protein interaction network. Nature 430, 88–93 (2004).
Article CAS PubMed Google Scholar
Imam, F. et al. The Global Neurodegeneration Proteomics Consortium: biomarker and drug target discovery for common neurodegenerative diseases and aging. Nat. Med. https://doi.org/10.1038/s41591-025-03834-0 (2025).
Van Exel, E. et al. Effect of APOE ε4 allele on survival and fertility in an adverse environment. PLoS ONE 12, e0179497 (2017).
Article PubMed PubMed Central Google Scholar
Martin, G. M. APOE alleles and lipophylic pathogens. Neurobiol. Aging 20, 441–443 (1999).
Article CAS PubMed Google Scholar
Finch, C. E. & Sapolsky, R. M. The evolution of Alzheimer disease, the reproductive schedule, and apoE isoforms. Neurobiol. Aging 20, 407–428 (1999).
Article CAS PubMed Google Scholar
Tuminello, E. R. & Han, D. S. The apolipoprotein E antagonistic pleiotropy hypothesis: review and recommendations. Int. J. Alzheimer’s Dis. 2011, 726197 (2011).
Article Google Scholar
Gale, S. C. et al. APOε4 is associated with enhanced in vivo innate immune responses in humans. J. Allergy Clin. Immunol. 134, 127–134 (2015).
Article Google Scholar
Fernandez-Calle, R. et al. APOE in the bullseye of neurodegenerative diseases: impact of the APOE genotype in Alzhiemer’s disease pathology and brain diseases. Mol. Neurodegener. 17, 62 (2022).
Article CAS PubMed PubMed Central Google Scholar
Tao, Q. et al. Association of chronic low-grade inflammation with risk of Alzheimer disease in ApoE4 carriers. JAMA Netw. Open 1, e183597 (2018).
Article PubMed PubMed Central Google Scholar
Gallagher, M. D. & Chen-Plotkin, A. S. The post-GWAS era: from association to function. Am. J. Hum. Genet. 102, 717–730 (2018).
Article CAS PubMed PubMed Central Google Scholar
Ashton, N. J. et al. A multicentre validation study of the diagnostic value of plasma neurofilament light. Nat. Commun. 12, 3400 (2021).
Article CAS PubMed PubMed Central Google Scholar
Gotze, K. et al. Plasma neurofilament light chain as prognostic marker of cognitive decline in neurodegenerative diseases, a clinical setting study. Alzheimer’s Res. Ther. 16, 231 (2024).
Article Google Scholar
Lin, Y.-S., Lee, W.-J., Wang, S.-J. & Fuh, J.-L. Levels of plasma neurofilament light chain and cognitive function in patients with Alzheimer or Parkinson disease. Sci. Rep. 8, 17368 (2018).
Article PubMed PubMed Central Google Scholar
Malek-Ahmadi, M. et al. Plasma NfL is associated with the APOE ε4 allele, brain imaging measurements of neurodegeneration, and lower recall memory scores in cognitively unimpaired late-middle-aged and older adults. Alzheimer’s Res. Ther. 15, 74 (2023).
Article CAS Google Scholar
Frick, E. A. et al. Serum proteomics reveal APOE-ε4-dependent and APOE-ε4-independent protein signatures in Alzheimer’s disease. Nat. Aging 4, 1446–1464 (2024).
Article CAS PubMed PubMed Central Google Scholar
Mattson, N., Andreassen, O. A., Zetterberg, H., Blennow, K. & Alzheimer’s Disease Neuroimaging Initiative. Association of plasma neurofilament light with neurodegeneration in patients with Alzheimer disease. JAMA Neurol. 74, 557–566 (2017).
Article Google Scholar
Rutledge, J. et al. Comprehensive proteomics of CSF, plasma, and urine identify DDC and other biomarkers of early Parkinson’s disease. Acta Neuropathol. 147, 52 (2024).
Article CAS PubMed PubMed Central Google Scholar
Greene, C. et al. Blood–brain barrier disruption and sustained systemic inflammation in individuals with long COVID-associated cognitive impairment. Nat. Neurosci. 27, 421–432 (2024).
Article CAS PubMed PubMed Central Google Scholar
Montagne, A. et al. APOE4 leads to blood–brain barrier dysfunction predictive cognitive decline. Nature 581, 71–76 (2020).
Article CAS PubMed PubMed Central Google Scholar
Sweeney, M. D., Montagne, A., Sagare, A. P., Nation, D. A. & Zlokovic, B. V. Broken blood–brain barrier in Alzheimer’s disease and other neurodegenerative disorders. Nat. Rev. Neurol. 15, 133–150 (2019).
Google Scholar
Ziegler-Heitbrock, L. The CD14⁺ and CD16⁺ blood monocytes: their role in infection and inflammation. J. Leukoc. Biol. 81, 584–592 (2007).
Article CAS PubMed Google Scholar
Wong, K. L. et al. The three human monocyte subsets: implications for health and disease. Immunol. Res. 53, 41–57 (2012).
Article CAS PubMed Google Scholar
Akbar, A. N. & Henson, S. M. Are senescence and exhaustion intertwined or unrelated processes that compromise immunity. Nat. Rev. Immunol. 11, 289–295 (2011).
Article CAS PubMed Google Scholar
Hayday, A. C. γδ Cells: a right time and a right place for a conserved third way of protection. Annu. Rev. Immunoloy 18, 975–1026 (2000).
Article CAS Google Scholar
Vivier, E., Tomasello, E., Baratin, M., Walzer, T. & Ugolini, S. Functions of natural killer cells. Nat. Immunol. 9, 503–510 (2008).
Article CAS PubMed Google Scholar
Lanier, L. L. NK cell recognition. Annu. Rev. Immunoloy 23, 225–274 (2005).
Article CAS Google Scholar
Filippini, N. et al. Distinct patterns of brain activity in young carriers of the APOE-ε4 allele. Proc. Natl Acad. Sci. USA 106, 7209–7214 (2009).
Article CAS PubMed PubMed Central Google Scholar
Dennis, N. A. et al. Temporal lobe functional activity and connectivity in young adult APOE ε4 carriers. Alzheimer’s Dement. 6, 303–311 (2010).
Article Google Scholar
Andre, C. et al. Reduced rapid eye movement sleep in late middle-aged and older apolipoprotein E ε4 allele carriers. Sleep 47, zsae094 (2024).
Article PubMed PubMed Central Google Scholar
Sweeney, M. D., Zhao, Z., Montagne, A., Nelson, A. R. & Zlokovic, B. V. Blood–brain barrier dysfunction across neurodegeneration. Nat. Rev. Neurosci. 20, 36–53 (2019).
Google Scholar
Seeley, W. W., Crawford, R. K., Zhou, J., Miller, B. L. & Greicius, M. D. Neurodegenerative diseases target large-scale human brain networks. Neuron 62, 42–52 (2009).
Article CAS PubMed PubMed Central Google Scholar
Ramanan, V. K. & Saykin, A. J. Pathways to neurodegeneration: mechanistic insights from GWAS in Alzheimer’s disease, Parkinson’s disease, and related disorders. Am. J. Neurodegener. Dis. 2, 145–175 (2013).
PubMed PubMed Central Google Scholar
Harris, K. et al. The impact of routine vaccinations on Alzheimer’s disease risk in persons 65 years and older: a claims-based cohort study using propensity score matching. J. Alzheimer’s Dis. 95, 703–718 (2023).
Article CAS Google Scholar
Wu, X. et al. Adult vaccination as a protective factor for dementia: a meta-analysis and systematic review of population-based observational studies. Front. Immunol. 13, 872542 (2022).
Article CAS PubMed PubMed Central Google Scholar
Eyting, M. et al. A natural experiment on the effect of herpes zoster vaccination on dementia. Nature 641, 438–446 (2025).
Article CAS PubMed PubMed Central Google Scholar
Taquet, M., Dercon, Q., Todd, J. A. & Harrison, P. J. The recombinant shingles vaccine is associated with lower risk of dementia. Nat. Med. 30, 2777–2781 (2024).
Article CAS PubMed PubMed Central Google Scholar
Mattson, N. et al. Prevalence of the apolipoprotein E ε4 allele in amyloid β positive subjects across the spectrum of Alzheimer’s disease. Alzheimer’s Dement. 14, 913–924 (2018).
Article Google Scholar
Argentieri, A. M. et al. Integrating the environmental and genetic architectures of aging and mortality. Nat. Med. 31, 1016–1025 (2025).
Article CAS PubMed PubMed Central Google Scholar
Lumsden, A. L., Mulugeta, A., Zhou, A. & Hypponen, E. Apolipoprotein E (APOE) genotype-associated disease risks: a phenome-wide, registry-based, case-control study utilising the UK Biobank. eBioMedicine 59, 102954 (2020).
Article CAS PubMed PubMed Central Google Scholar
Dunk, M. M. et al. Relationships between APOE, type 2 diabetes, and cardiovascular disease in postmenopausal women. J. Gerontol. A Biol. Sci. Med. Sci. 80, glae246 (2025).
Article PubMed Google Scholar
Lahoz, C. et al. Apolipoprotein E genotype and cardiovascular disease in Framingham Heart Study. Atherosclerosis 154, 529–537 (2001).
Article CAS PubMed Google Scholar
Bunn, F. et al. Comorbidity and dementia: a scoping review of the literature. BMC Med. 12, 192 (2014).
Article PubMed PubMed Central Google Scholar
Carroll, C. et al. Addressing comorbidities in people with Parkinson’s disease: considerations from an expert panel. J. Parkinson’s Dis. 14, 53–63 (2024).
Article Google Scholar
Mirra, S. S. et al. The Consortium to Establish a Registry for Alzheimer’s disease (CERAD). Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 41, 479–486 (1991).
Article CAS PubMed Google Scholar
Braak, H. & Braak, E. Neuropathological stageing of Alzheimer-related changes. Acta Neuropathol. 82, 239–259 (1991).
Article CAS PubMed Google Scholar
Gelb, D. J., Oliver, E. & Gilman, S. Diagnostic criteria for Parkinson disease. Arch. Neurol. 56, 33–39 (1999).
Article CAS PubMed Google Scholar
Ince, P. G., Lowe, J. & Shaw, P. J. Review. Neuropathol. Appl. Neurobiol. 24, 104–117 (1998).
Article CAS PubMed Google Scholar
Gold, L. et al. Aptamer-based multiplexed proteomic technology for biomarker discovery. PLoS ONE 5, e15004 (2010).
Article CAS PubMed PubMed Central Google Scholar
Williams, S. A. et al. Plasma protein patterns as comprehensive indicators of health. Nat. Med. 25, 1851–1857 (2019).
Article CAS PubMed PubMed Central Google Scholar
Johnson, E. C. B. et al. Large-scale proteomic analysis of Alzheimer’s disease brain and cerebrospinal fluid reveals early changes in energy metabolism associated with microglia and astrocyte activation. Nat. Med. 26, 769–780 (2020).
Article CAS PubMed PubMed Central Google Scholar
Ross, B. C. Mutual information between discrete and continuous data sets. PLoS ONE 9, e87357 (2014).
Article PubMed PubMed Central Google Scholar
Fayyad, U. M. & Irani, K. B. Multi-interval discretization of continuous-valued attributes for classification learning. In Proc. 13th International Joint Conference on Uncertainty in Artificial Intelligence 1022–1027 (1993).
Wickham, H. ggplot2: Elegant Graphics for Data Analysis (Springer-Verlag, 2016).
Xia, J., Benner, M. J. & Hancock, R. E. W. NetworkAnalyst—integrative approaches for protein–protein interaction network analysis and visual exploration. Nucleic Acids Res. 42, W167–W174 (2014).
Article CAS PubMed PubMed Central Google Scholar
Xia, J., Gill, E. & Hancock, R. E. W. NetworkAnalyst for statistical, visual and network-based approaches for meta-analysis of expression data. Nat. Protoc. 10, 823–844 (2015).
Article CAS PubMed Google Scholar
Zhou, G. et al. NetworkAnalyst 3.0: a visual analytics platform for comprehensive gene expression profiling and meta-analysis. Nucleic Acids Res. 47, W234–W241 (2019).
Article CAS PubMed PubMed Central Google Scholar
Orchard, S. et al. Protein interaction data curation—the International Molecular Exchange Consortium (IMEx). Nat. Methods 9, 345–350 (2013).
Article Google Scholar
Breuer, K. et al. InnateDB: systems biology of innate immunity and beyond—recent updates and continuing curation. Nucleic Acids Res. 41, 1228–1233 (2013).
Article Google Scholar
Kanehisa, M. & Goto, S. KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res. 38, 27–30 (2000).
Article Google Scholar

Download references

Acknowledgements

We are grateful to the cohort contributors, patients, donors and families who helped to make this research possible. We are also grateful to the members of the Global Neurodegeneration Proteomics Consortium, who provided ongoing feedback. This work was supported by the Australian Government’s Medical Research Future Fund (MRF2040081 to C.A.F. and A.S.); philanthropic funding from Neil & Norma Hill Foundation, Annemarie & Arturo Gandioli-Fumagali Foundation, Perpetual Foundation—John Williams Endowment and Hillcrest Foundation (to C.A.F.); the NIH (K08AG08604, R01AG089497 and P50AG025688 to E.C.B.J.; R35NS132179 to J.D.R.; P30AG072973 to H.M.W., J.M.B., R.H.S. and C.S.; R01AG064227 to C.S.; and R21TR003589, R01AG07816 and U19AG068054 to H.M.W.), Alzheimer’s Research UK (to L.M.W.); Alzheimer’s Association (23AARG-1023 to H.M.W.); Answer ALS Foundation (to J.D.R.) and Robert Packard Center for ALS Research at Johns Hopkins University (to J.D.R.). The results published here are in whole or in part based on data obtained from the AD Knowledge Portal (https://adknowledgeportal.org). Study data were provided through the Accelerating Medicine Partnership for AD (U01AG046161 and U01AG061357), based on specimens provided by the Emory Goizueta Alzheimer’s Disease Research Center (P50AG025688) where data collection was supported through funding by NIA (grants R01AG053960, R01AG057911, R01AG061800, RF1AG057471, RF1AG057470, R01AG061800, R01AG057911 and R01AG057339). The funders of this work had no role in the design of the study, the running of experiments and analyses, the interpretation of the results and the writing of the manuscript.

Author information

A list of authors and their affiliations appears at the end of the paper.
A full list of members and their affiliations appear in the Supplementary Information.

Authors and Affiliations

Neurodegeneration and Disease Modelling Research Group, Westmead Institute for Medical Research, The University of Sydney, Westmead, New South Wales, Australia
Artur Shvetcov & Caitlin A. Finney
School of Medical Sciences, Faculty of Medicine and Health, The University of Sydney, Sydney, New South Wales, Australia
Artur Shvetcov & Caitlin A. Finney
Department of Neurology, Emory University, Atlanta, GA, USA
Erik C. B. Johnson
Goizueta Alzheimer’s Disease Research Centre, Emory University, Atlanta, GA, USA
Erik C. B. Johnson
Department of Psychiatry, University of Oxford, Oxford, UK
Laura M. Winchester
Laboratory of Behavioral Neuroscience, National Institute on Aging, Intramural Research Program, Baltimore, MD, USA
Keenan A. Walker
University of Kansas Alzheimer’s Disease Research Centre, University of Kansas Medical Center, Kansas City, KS, USA
Heather M. Wilkins, Jeffrey M. Burns, Russell H. Swerdlow & Chad Slawson
Department of Neurology, University of Kansas School of Medicine, Kansas City, KS, USA
Heather M. Wilkins, Jeffrey M. Burns & Russell H. Swerdlow
Department of Biochemistry and Molecular Biology, University of Kansas Medical Center, Kansas City, KS, USA
Heather M. Wilkins, Russell H. Swerdlow & Chad Slawson
OnPoint Scientific, Inc., San Diego, CA, USA
Terri G. Thompson
Brain Science Institute, Johns Hopkins University School of Medicine, Baltimore, MD, USA
Jeffrey D. Rothstein
Department of Neurology, Johns Hopkins University School of Medicine, Baltimore, MD, USA
Jeffrey D. Rothstein
Gates Ventures, Seattle, WA, USA
Varsha Krish & Farhad B. Imam
Department of Cell Biology and Physiology, University of Kansas Medical Center, Kansas City, KS, USA
Russell H. Swerdlow
University of Kansas Cancer Center, University of Kansas Medical Center, Kansas City, KS, USA
Chad Slawson

Authors

Artur Shvetcov
View author publications
Search author on:PubMed Google Scholar
Erik C. B. Johnson
View author publications
Search author on:PubMed Google Scholar
Laura M. Winchester
View author publications
Search author on:PubMed Google Scholar
Keenan A. Walker
View author publications
Search author on:PubMed Google Scholar
Heather M. Wilkins
View author publications
Search author on:PubMed Google Scholar
Terri G. Thompson
View author publications
Search author on:PubMed Google Scholar
Jeffrey D. Rothstein
View author publications
Search author on:PubMed Google Scholar
Varsha Krish
View author publications
Search author on:PubMed Google Scholar
Farhad B. Imam
View author publications
Search author on:PubMed Google Scholar
Jeffrey M. Burns
View author publications
Search author on:PubMed Google Scholar
Russell H. Swerdlow
View author publications
Search author on:PubMed Google Scholar
Chad Slawson
View author publications
Search author on:PubMed Google Scholar
Caitlin A. Finney
View author publications
Search author on:PubMed Google Scholar

Consortia

The Global Neurodegeneration Proteomics Consortium (GNPC)

Artur Shvetcov
, Erik C. B. Johnson
, Laura M. Winchester
, Keenan A. Walker
, Terri G. Thompson
, Jeffrey D. Rothstein
, Varsha Krish
, Farhad B. Imam
, Jeffrey M. Burns
, Chad Slawson
& Caitlin A. Finney

Contributions

A.S. and C.A.F. conceptualized the study, led the study design, produced all the figures and wrote the paper. A.S. led the data analysis. E.C.B.J. provided additional genotypes and postmortem histopathological data for the AMP-AD UPenn Proteomics Study cohort. L.M.W. provided key insights into the analytical strategy and PD. E.C.B.J., K.A.W., H.M.W., R.H.S. and C.A.F. provided key insights into the effects of APOE genotypes. T.G.T., J.D.R., F.B.I. and V.K. provided genotype data and T.G.T. and J.D.R. key insights for amyotrophic lateral sclerosis. E.C.B.J., L.M.W., K.A.W., H.M.W., J.M.B., R.H.S., C.S. and C.A.F. provided key insights into AD, FTD and PDD. The GNPC participated in creating v1 of the harmonized GNPC dataset. C.A.F. supervised the study. All authors revised the paper for intellectual content and read and approved the final version of the paper.

Corresponding authors

Correspondence to Artur Shvetcov or Caitlin A. Finney.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Medicine thanks Boris Decourt, Omar El-Agnaf and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Liam Messin and Anna Maria Ranzoni, in collaboration with the Nature Medicine team.

Additional information

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

Extended data

Extended Data Fig. 1 Further characterization of the APOE ε4 CSF protein signature.

(a) Volcano plots showing fold change and adjusted p values for the APOE ε4 CSF proteins. Red dots indicate the proteins with the highest mutual information value (>0.1) for reference to the feature selection method used to identify APOE ε4 proteins. Differential protein abundance was assessed using linear modeling and empirical Bayes moderation. All tests were two-sided, and p-values were corrected for multiple comparisons using the Benjamini-Hochberg FDR method. Volcano plots display log₂ fold change on the x-axis and -log₁₀-transformed adjusted p-values on the y-axis, with thresholds set at log₂(FC) = ± 0.585 and FDR-adjusted p < 0.05. (b) Box plots showing amyloid beta A4 protein CSF levels in relative fluorescent units (RFU) across APOE ε4 carriers and non-carriers with AD, PD, and non-impaired controls. The center line represents the median, the bounds of the box indicate the 25th and 75th percentiles (interquartile range; IQR), and the whiskers extend to the minimum and maximum values within 1.5× IQR. Data points outside this range are plotted individually as outliers. N = 526 AD, 247 PD, and 573 non-impaired control individuals. Abbreviations: AD: Alzheimer’s disease; NI: non-impaired controls; PD: Parkinson’s disease.

Extended Data Fig. 2 Further characterization of the APOE ε4 plasma protein signature.

(a) Volcano plots showing fold change and adjusted p values for the APOE ε4 plasma proteins. Red dots indicate the proteins with the highest mutual information value (>0.1) for reference to the feature selection method used to identify APOE ε4 proteins. Differential protein abundance was assessed using linear modeling and empirical Bayes moderation. All tests were two-sided, and p-values were corrected for multiple comparisons using the Benjamini-Hochberg FDR method. Volcano plots display log₂ fold change on the x-axis and −log₁₀-transformed adjusted p-values on the y-axis, with thresholds set at log₂(FC) = ± 0.585 and FDR-adjusted p < 0.05. (b) Box plots showing amyloid beta A4 protein plasma levels in relative fluorescent units (RFU) across APOE ε4 carriers and non-carriers with AD, ALS, FTD, PDD, PD, and non-impaired controls. The center line represents the median, the bounds of the box indicate the 25th and 75th percentiles (interquartile range; IQR), and the whiskers extend to the minimum and maximum values within 1.5x IQR. Data points outside this range are plotted individually as outliers. N = 2,929 AD, 75 FTD, 169 PDD, 422 PD, 230 ALS, and 6,099 non-impaired control individuals. Abbreviations: AD: Alzheimer’s disease; ALS: amyotrophic lateral sclerosis; FTD: frontotemporal dementia; NI: non-impaired controls; PD: Parkinson’s disease; PDD: Parkinson’s disease dementia.

Extended Data Fig. 3 Stratification of comorbidities across APOE ε4 carriers and non-carriers.

Bar graph representation of the proportion of APOE ε4 and non-carriers with comorbidity across the clinical variables included in our correlation network analysis between APOE ε4 proteins and demographic, lifestyle, and clinical variables. Only those individuals who said ‘yes’ to having comorbidity are included in the graph (Supplementary Table 1). APOE4+: APOE ε4 carrier; APOE4−: APOE ε4 non-carrier; afib: atrial fibrillation; chf: congestive heart failure; copd: chronic obstructive pulmonary disease; tbi: traumatic brain injury; tia: transient ischemic attack; mi: myocardial infarction.

Extended Data Fig. 4 Identification of APOE ε4 proteins in the dorsolateral prefrontal cortex of individuals with neurodegenerative diseases and non-impaired controls.

APOE ε4 proteins identified using label-free MS from the AMP-AD UPenn Proteomics Study cohort and mutual information. (a,c,e,g,i,k) PCA showing all label-free MS measured proteins in each donor group and no clear clustering across APOE ε4 carriers and non-carriers. (b,d,f,h,j,l) PCA showing only APOE ε4-identified proteins in each donor group, which leads to clustering in each group based on the presence or absence of an APOE ε4 allele. AD: Alzheimer’s disease; ALS: amyotrophic lateral sclerosis; FTD: frontotemporal dementia; PD: Parkinson’s disease; PDD: Parkinson’s disease dementia.

Extended Data Table 1 Performance metrics of CART models for predicting APOE ε4 carriers from non-carriers in AD, PD, and by sex using APOE ε4 CSF proteins (n = 229) as predictive features

Full size table

Extended Data Table 2 Performance metrics of CART models for predicting APOE ε4 carriers from non-carriers in AD, PD, and by sex using APOE ε4 plasma proteins (n = 58) as predictive features

Full size table

Supplementary information

Supplementary Information (download PDF )

List of GNPC members.

Reporting Summary (download PDF )

Supplementary Tables 1–11 (download XLSX )

Supplementary Tables 1–11.

Rights and permissions

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

Reprints and permissions

About this article

Cite this article

Shvetcov, A., Johnson, E.C.B., Winchester, L.M. et al. APOE ε4 carriers share immune-related proteomic changes across neurodegenerative diseases. Nat Med 31, 2590–2601 (2025). https://doi.org/10.1038/s41591-025-03835-z

Download citation

Received: 20 April 2025
Accepted: 16 June 2025
Published: 15 July 2025
Version of record: 15 July 2025
Issue date: August 2025
DOI: https://doi.org/10.1038/s41591-025-03835-z

This article is cited by

A paradigm shift: from precision medicine to precision health in neurodegenerative conditions and aging
- Andrés D. Klein
- Arnold Hoppe
Journal of Precision Health (2026)
Protective mutations associated with APOE in Alzheimer’s disease
- Yuejia Ma
- Yanxi Li
- Dayong Wang
Molecular Psychiatry (2026)
Sex and APOE genotype specific brain regional vulnerability to Alzheimer's Disease
- Qi Zeng
- Minghui Wang
- Kenneth Spicer
GeroScience (2026)
Large-scale proteomics project seeks cures for neurodegenerative diseases
- Boris Decourt
- Marwan N. Sabbagh
- Michelle M. Mielke
Nature Medicine (2025)
One potent gene raises risk of Alzheimer’s, Parkinson’s and other brain diseases
- Smriti Mallapaty
Nature (2025)

Subjects

Abstract

Similar content being viewed by others

Main

Results

Plasma proteome profiling reveals a similar APOE ε4-specific signature

Key features of the peripheral immune signature are mirrored in the brains of APOE ε4 carriers

Peripheral APOE ε4-associated proteins are differentially correlated to clinical variables according to disease

Discussion

Methods

Participants

GNPC cohort

AMP-AD UPenn Proteomics Study cohort

Proteomics

GNPC cohort

AMP-AD UPenn Proteomics Study cohort

Statistical analyses

Feature selection

CART

Linear regression

Correlation network analysis

Enrichment analyses

Reporting summary

Data availability

Code availability

References

Acknowledgements

Author information

Authors and Affiliations

Consortia

The Global Neurodegeneration Proteomics Consortium (GNPC)

Contributions

Corresponding authors

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Extended data

Supplementary information

Rights and permissions

About this article

Cite this article

Share this article

This article is cited by

Search

Quick links