Peripheral immune cells as a window on Alzheimer’s disease and its treatment
Characterizing the immune response in AD might be clinically significant for at least two reasons: peripheral immune-cell phenotypes are emerging as non-invasive indicators of disease progression and treatment efficacy, and changes in peripheral immune cells are associated with central inflammation in patients with AD.
Figure 1 at a glance
- 6
- peripheral immune-cell types with changes reported in AD
- 18
- reported changes: 9 decreases and 9 increases
- 73
- human patient studies (1981–2020) showing T-cell dysfunction in AD, per Dai & Shen
A growing disease burden
The rate of new diagnoses increases as the population ages — increasing longevity correlates with rising annual incidence rates of AD diagnoses.
All figures as cited in the CogniMune Phase I proposal.
Treatments that clear amyloid — and a side effect we do not yet understand
The gold standard for diagnosing AD uses MRI scans alongside amyloid detection in cerebrospinal fluid (CSF). The broader context of peripheral changes in AD patients — particularly immune-cell alterations — remains significantly underappreciated.
There is no cure for AD, and available treatments slow disease progression. The recently approved amyloid-targeting therapies lecanemab and donanemab are designed to reduce amyloid deposits in the brain, but can lead to side effects known as Amyloid-Related Imaging Abnormalities (ARIA). ARIA is detectable through MRI and is associated with compromised blood–brain barrier (BBB) integrity and immune-inflammatory responses.
The approvals of lecanemab and donanemab highlight our limited understanding of the factors that may heighten ARIA risk. Conversely, their clinical use is a valuable opportunity to identify individuals who may be at risk — by focusing on changes in peripheral immune cells.
A proposed mechanism
Hypothesis- 1Amyloid plaques are removedby lecanemab and/or donanemab.
- 2Pre-existing BBB damage is exposedthe barrier may already be compromised.
- 3Peripheral immune cells flow inARIA may be associated with elevated immune markers.
- 4ARIA-associated neuroinflammationvisible as imaging abnormalities on MRI.
Supporting evidence. In animal models of cerebral amyloid angiopathy (CAA), damage to blood-vessel walls activates the endothelium, increases BBB permeability, facilitates the infiltration of peripheral cells and enhances the production of pro-inflammatory cytokines.
It is conceivable that anti-amyloid antibodies (1) indirectly influence the activation of certain peripheral immune cells in response to Aβ-related inflammation, and (2) shift the balance between pro- and anti-inflammatory responses.
Changes in peripheral immune cells reported in AD
Immune-system alterations in AD patients indicate a systemic immune dysregulation beyond the central nervous system. Select a cell type to see the changes reported in the literature and the panel populations that measure them.
T cells
Adaptive lymphocytes: CD4 helper and regulatory T cells, CD8 cytotoxic T cells.
-
Naïve CD4decreased
-
Activated T cellsdecreased
-
Apoptotic T cellsdecreasedNo dedicated population in the 40-color panel.
-
CD4⁺CD38⁺decreased
-
Tregdecreased
-
Effector CD4 cellsincreased
-
Effector CD4 memoryincreased
-
CD4⁺CD8dimincreased
Dendritic cells
Antigen-presenting cells that link innate and adaptive immunity.
-
CD11cdecreased
Monocytes
Circulating myeloid cells: classical, intermediate and non-classical subsets.
-
Non-classical monocytesdecreased
-
Expression of TREM2increased
-
Classical monocytesincreased
Neutrophils
The most abundant blood leukocytes; mature and immature forms.
-
Neutrophil-to-lymphocyte ratioincreased
-
Senescent to immunosuppressive neutrophil ratioincreased
NK cells
Innate cytotoxic lymphocytes (CD3⁻ CD56⁺).
-
CD3⁻CD56⁺ NKdecreased
-
CXCR1⁺ NKincreasedNo dedicated population in the 40-color panel.
B cells
Antibody-producing lymphocytes, including regulatory B cells (Breg).
-
Bregdecreased
-
HLA-DR expressionincreased
Source: proposal Figure 1, “The changes in peripheral immune cells in patients with AD reported in the literature.” The mapping of each change to panel populations and markers is the CogniMune reference definition used in the research portal.
The immune system can reflect a patient’s AD status
Non-invasive indicators
Peripheral immune-cell phenotypes are emerging as valuable, non-invasive indicators of both disease progression and treatment efficacy.
A link to central inflammation
Recent studies have shown an association between changes in peripheral immune cells and central inflammation in patients with AD.
Decades of evidence
Dai & Shen identified T-cell dysfunction across 73 human patient studies in AD, spanning 1981 to 2020 and using the best available tools of their respective eras.
Conventional versus spectral flow cytometry
Flow cytometry (FC) measures protein expression on single cells with antibodies labeled by fluorescent dyes. The intensity of those dyes can be quantified with conventional or spectral instruments — the emission of each dye is detected by detectors and this allows us to characterize the immunological signature of each patient.
Conventional FC
PolychromaticSchematic: one detector (bandpass window) per dye; the rest of each emission spectrum is lost or spills into neighboring detectors.
- Traditional instruments can measure up to 30 parameters, depending on their configuration.
- All FC data in the AD space have been collected with this older polychromatic technology — limiting data complexity while requiring much higher volumes of blood.
Spectral FC
Full spectrumSchematic: the full emission spectrum of every dye is recorded across many detectors.
- Spectral flow cytometers have recently achieved 50 different markers, with virtually no hardware limitations on the emission wavelengths of fluorescent probes.
- The cost of assay implementation in spectral cytometry has been reduced dramatically, based on Dr. Robinson's innovative patent US7,280,204.
Cross-sectional snapshots versus longitudinal trajectories
Interpreting FC studies of peripheral immune changes faces a critical limitation: most existing data derive from cross-sectional studies. They provide temporal snapshots without establishing causality — unlike longitudinal approaches, which can determine whether variables influence one another.
The proposed study follows each treated patient from before therapy initiation through the treatment cycle, so every patient serves as their own reference.
APOE ε4: more than an AD risk gene
ARIA is more likely to be present in patients carrying the genetic variant APOE ε4 (over 50% of all cases). A recent study of a large dataset of patients with AD, Parkinson's disease (PD) or amyotrophic lateral sclerosis (ALS) found that the APOE ε4 proteomic signature is strongly associated with pro-inflammatory immune and infection pathways, and with immune-cell populations including monocytes, T cells and NK cells.
Because the association appears consistent regardless of the neurodegenerative disease, APOE ε4 may function as a pleiotropic immune modulator — a disease-independent biological vulnerability to neurodegeneration. Findings from this study could therefore extend beyond AD, although testing in PD or ALS is outside the scope of the current proposal.
Treated patients are genotyped for APOE ε4 during screening at the IU School of Medicine Brain Health Program.
Monitoring demand in the era of anti-amyloid therapy
Estimates as stated in the proposal, based on the number of patients diagnosed with AD in 2011 and the length of the new therapy cycle.
Cost of 1,800,000 tests per year
Per the proposal.
Tests that require mass spectrometry or MRI can exceed $1,000 per individual and are not particularly convenient to patients.
Blood-based tests today
Current tests mostly indicate the presence or absence of AD, without a comprehensive understanding of its mechanistic progression. Plasma-based tests draw anticoagulated blood — the same draw could serve CogniMune if the two assays are coordinated.
Three critical innovations in AD monitoring
From episodic assessment to continuous, quantitative immune surveillance.
Technological
High-parameter spectral flow cytometry combined with optimal-transport theory — proposed as the first application of this approach in neurological disease monitoring, to detect subtle immune changes previously undetectable with conventional 10–15 parameter systems. The proposal cites spectral instruments reaching 50 parameters; the CogniMune panel uses 40 colors.
Methodological
Graph-based immune fingerprinting integrated with automated cloud-based analysis: a standardized platform for real-time immune monitoring across multiple clinical sites, with OT Sinkhorn distances that quantify immune-profile changes objectively.
Clinical
Dynamic treatment-response monitoring throughout therapy rather than a binary test of AD presence — potentially identifying ARIA risk before clinical manifestation, integrated with existing blood draws for plasma-based AD tests.
See how Phase I will test this
Three Specific Aims: validate the 40-color panel, follow treated patients longitudinally, and automate optimal-transport analysis.