Adaptive Immunity and Alzheimer’s Disease: Dual Roles in Neurodegeneration and Neuroprotection with Therapeutic Implications

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder defined by amyloid-β (Aβ) plaques, tau hyperphosphorylation, and neuroinflammation. Although earlier work emphasized brain-resident glia (microglia and astrocytes), recent studies highlight adaptive immune cells, particularly T and B lymphocytes, as modulators of AD pathology. This review synthesizes animal and human findings from 2022–2025 to provide updated insights into the multifaceted roles and therapeutic potential of adaptive immunity in AD. Infiltration of peripheral T and B cells into the brain parenchyma links peripheral immunity to central nervous system (CNS) pathology. Both infiltrating lymphocytes and resident glia show context-dependent dual effects, either exacerbating neurodegeneration or promoting neuroprotection. Therapeutic strategies under active investigation include modulation of CD4⁺ T cell differentiation, adoptive transfer of regulatory T cells, and next-generation active vaccines for AD. Overall, selective modulation of discrete immune subsets may enable adaptive-immunity-based treatments, a complex yet promising avenue for AD therapy.

INTRODUCTION

Alzheimer’s disease (AD) is the most prevalent age-related neurodegenerative disorder, accounting for 60-80% of dementia cases globally (Alzheimer’s Association, 2025). AD is characterized by amyloid-β (Aβ) plaque deposition, tau hyperphosphorylation forming neurofibrillary tangles (NFTs), synaptic dysfunction, neuronal loss, and progressive cognitive decline. In recent years, neuroinflammation has garnered particular attention as a key driver of AD pathogenesis, intersecting these classical hallmarks (Heneka et al., 2024).

Multiple immune cell types contribute to neuroinflammation in AD. To date, research has primarily focused on CNS-resident glia, microglia, and astrocytes, which act as the primary immune responders by clearing misfolded protein aggregates and orchestrating inflammatory signaling. Under pathological conditions, microglia adopt a disease-associated phenotype (DAM), and polarization toward proinflammatory states (M1 microglia, A1 astrocytes) amplifies neuroinflammation, promoting Aβ accumulation, synaptic loss, and cognitive impairment (Basha et al., 2023; Deng et al., 2024; Gao et al., 2023).

Despite extensive study of CNS-resident glia, the role of the adaptive immune system, particularly T and B lymphocytes, in AD pathogenesis has only recently come into focus. Phenotypic and functional characterization of these cells in the AD brain remains incomplete, posing a barrier to targeted immunotherapy development. In this review, we outline mechanisms of adaptive immune cell infiltration into the brain parenchyma during AD and synthesize recent evidence for the dualistic roles of adaptive immune subsets, which can either mitigate or exacerbate disease progression. We also discuss emerging AD treatments that leverage adaptive-immunity-based approaches.

INFILTRATION OF T LYMPHOCYTES INTO THE AD BRAIN

Under physiological conditions, lymphocyte entry into the central nervous system (CNS) parenchyma is tightly restricted. However, neurodegenerative changes in AD can increase blood–brain barrier (BBB) permeability, permitting T lymphocyte (T cell) access to CNS compartments. During transvascular migration, T cells first tether and roll via endothelial selectins, then undergo firm adhesion through T cell integrins—very late antigen-4 (VLA-4; α4β1) binding vascular cell adhesion molecule-1 (VCAM-1) and lymphocyte function–associated antigen-1 (LFA-1; αLβ2) binding intercellular adhesion molecule-1 (ICAM-1). These interactions enable diapedesis across the endothelium into perivascular spaces and, subsequently, the brain parenchyma (Zeng et al., 2024).

Recent studies have identified additional drivers of T cell infiltration (overview in Fig. 1). In AD, BBB dysfunction elevates brain chemokine levels and enhances signaling through corresponding receptors on T cells, facilitating recruitment (Farrall and Wardlaw, 2009). In an AD neural–glial co-culture model and in the hippocampus and cortex of 6-7-month-old 5xFAD mice, astrocytic C-X-C motif chemokine ligand 10 (CXCL10) was increased, inducing chemotaxis of CD4⁺ and CD8⁺ T cells via C-X-C motif chemokine receptor 3 (CXCR3) (Jorfi et al., 2023). Interleukin-17 (IL-17) stimulation of primary mouse glia and BV2 microglia elevated CXCL1 and CXCL9/10. Consistently, crossing IL-17 receptor A (IL-17RA) knockout (KO) mice with amyloid precursor protein (APP)/PS1 mice reduced hippocampal CXCL9/10 and suppressed CD8⁺ T cell migration, likely via the choroid plexus–cerebrospinal fluid (CSF) route (Ye et al., 2023). Moreover, single-cell transcriptomic analysis of CSF from cognitively impaired patients suggested that CXCL16–CXCR6 interactions may mediate recruitment of CD8⁺ T effector memory (TEM) cells to CNS compartments (Piehl et al., 2022).

Fig. 1.

Mechanisms of T cell infiltration into the AD brain parenchyma. Interactions between integrins and adhesion molecules—VLA-4 with VCAM-1, and LFA-1 with ICAM-1—enable T cells to transmigrate across the vascular endothelium into the perivascular space. Under AD conditions, chemokine gradients such as CXCL1, CXCL9/10, and CXCL16, secreted by activated astrocytes and microglia surrounding Aβ plaques, guide T cells toward the brain parenchyma. CXCR3 and CXCR6 receptors expressed on CD4⁺ and CD8⁺ T cells, including effector memory subsets, mediate these chemokine-driven interactions. IL-17 secreted by Th17 cells further enhances T cell infiltration by stimulating glial cells to produce chemokines. In parallel, brain-derived antigens such as Aβ are conveyed via CSF drainage through meningeal lymphatic vessels to peripheral lymph nodes, where B cells present these antigens and promote the activation and recruitment of CD8⁺ T cells into the brain.

Beyond chemokine–receptor signaling intrinsic to T cells, B lymphocytes (B cells) can promote the induction of brain-infiltrating CD8⁺ T cells in the periphery by presenting brain antigens such as Aβ. In 5xFAD mice with B-cell depletion (5xFAD-BKO), the upregulation of proinflammatory and cytolytic CD8⁺ T cells in the brain was abolished (Wang et al., 2024a). Taken together, these findings illustrate diverse chemokine-dependent mechanisms by which peripheral T cells access CNS compartments under AD conditions.

INFILTRATION OF B LYMPHOCYTES INTO THE AD BRAIN

In addition to presenting brain antigens and promoting T cell recruitment, peripheral B cells can themselves infiltrate the AD brain.

Skull bone marrow is a major source of brain-surveilling immune cells, where hematopoiesis from hematopoietic stem and progenitor cells (HSPCs) occurs. Elevated Aβ deposition has been observed in skull bone marrow from patients with AD and in mouse models (APP/PS1, 5xFAD). In 3-month-old 5xFAD mice, B-cell differentiation, particularly toward age-associated B cells (ABCs), was increased, and newly produced B cells infiltrated the brain parenchyma. Intraventricular transfer of CD19⁺CD11c⁺ ABCs from 10-month-old 5xFAD donors into 3-month-old 5xFAD recipients aggravated cognitive impairment and increased Aβ burden and microglial reactivity. Conversely, IL-6 receptor blockade with tocilizumab reduced ABC levels and brain infiltration in 5xFAD mice, mitigating associated pathology (Zhang et al., 2025a).

According to Makhijani and colleagues (2025), the gut–brain axis represents another potential route for B-cell trafficking. In 5xFAD mice, myeloid cells expressing CXCL12 were increased in both brain and colon. Consistently, CXCR4⁺IgA⁺ antibody-secreting B cells (which sense CXCL12) were reduced in the colon but elevated in the brain parenchyma. Administration of the CXCR4 inhibitor AMD3100 reduced B-cell migration by blocking CXCR4–CXCL12 interactions. These findings provide mechanistic support for antibody-secreting B-cell infiltration via the gut–brain axis and highlight the colon as a peripheral reservoir influencing AD pathology (Makhijani et al., 2025).

THE DUAL ROLES OF T AND B LYMPHOCYTES IN AD PATHOLOGY

Accumulating evidence indicates that adaptive immune cells can either exacerbate neurodegeneration or promote neuroprotection. Fig. 2 overviews recent studies attributing both pathogenic and protective functions to T and B cells in AD. Overall, adaptive immunity acts as a context-dependent “double-edged sword,” highlighting the need for immunotherapies that enhance protective subsets while suppressing pathogenic responses.

Fig. 2.

Context-dependent dual roles of T and B cells in AD pathology. Adaptive immune cells exhibit context-dependent dual roles in AD, contributing to either neurodegeneration or neuroprotection. (Neurodegeneration) Metabolically dysfunctional CD38⁺CD4⁺ T cells are enriched and associated with neurodegeneration. Proinflammatory CD4⁺ effector T cells (Th1/Th17 cells) and CXCR3⁺CD127⁺ Th1 cells promote neuroinflammation, whereas brain-infiltrating CD8⁺ T cells exacerbate oligodendrocyte and myelin damage and activate IFN-responsive microglia. Expanded TEM and TEMRA cells exhibit reduced CD28, IL-7R, and JUN expression, impairing memory T-cell function. Homeostatically expanded CD8⁺ TRM cells, stimulated by perforin 1 and IFN-γ, drive neurodegeneration. MAIT cells interact with MR1 on glial cells, contributing to dystrophic neurite formation. Peripheral blood memory B cells secrete MIF, which interacts with the CD74–CD44 receptor complex on macrophages, promoting M1 polarization and impairing Aβ plaque clearance. B cells further contribute to neuroinflammation by producing cytokines such as GM-CSF, IFN-γ, and TNF-α, while reducing IL-10 production. B cells also support MAIT cell development and produce immunoglobulins (IgG and IgM). Meningeal CLIP⁺ B cells are linked to neurodegeneration, whereas pathogenic NAbs such as NAbs-p75ECD enhance Aβ aggregation and exacerbate AD pathology. (Neuroprotection) CXCR3⁺CD127⁺ Th1 cells are negatively correlated with peripheral NfL. Certain subsets of CD4⁺ effector T cells (Th2 cells) and CD8⁺ T cells (CXCR6⁺PD-1⁺ CD8⁺ T cells) are known to be neuroprotective. Brain-resident CXCR6⁺PD-1⁺ CD8⁺ T cells colocalize with plaque-associated microglia via CXCR6–CXCL16 interactions. Depletion of regulatory T cells (Tregs) reduces neuroprotective A2-like Cox2⁺ reactive astrocytes and increases neurotoxic A1-like C3⁺ reactive astrocytes. Conversely, Tregs downregulate Aβ plaque burden and CD68⁺ reactive microglia and suppress the expression of proinflammatory cytokines, complement cascade components, toll-like receptors, and microglial activation markers. B cells producing IL-35 inhibit STAT1 phosphorylation through SOCS1, thereby reducing BACE1 expression and Aβ generation. Neuroprotective NAbs targeting TREM2, PS1, and Bim reduce Aβ plaque deposition, neuronal apoptosis, and BACE1 expression, contributing to neuroprotection.

CD4⁺ AND CD8⁺ T CELLS: DISTINCT SUBSETS IN RELATION TO AD

Distinct subsets of CD4⁺ and CD8⁺ T cells play crucial, often opposing roles in AD pathology. CD4⁺ T cells differentiate into effector subsets in response to specific cytokine milieus. Proinflammatory CD4⁺ effector T cells (Teffs), including T helper 1 (Th1) and T helper 17 (Th17) cells, exacerbate AD pathology by promoting BBB disruption, activating microglia, and driving neuroinflammation. In contrast, anti-inflammatory Teffs, T helper 2 (Th2) cells, and regulatory T cells (Tregs) counteract neuroinflammatory responses and confer neuroprotection (Zhang et al., 2025c).

Recent work has begun to delineate the CD4⁺ subsets most relevant to AD. In peripheral blood of patients with AD, CXCR3⁺CD127⁺ Th1 cells negatively correlated with circulating neurofilament light chain (NfL), a neuron-specific cytoskeletal protein elevated after neuronal injury. In cerebrospinal fluid (CSF), however, both the abundance and proinflammatory activity of CXCR3⁺CD127⁺ Th1 cells were increased (Hu et al., 2025). Furthermore, metabolically impaired CD38⁺CD4⁺ T cells were enriched in presymptomatic carriers of APP duplication, a mutation associated with early-onset familial AD. Therapeutic targeting with anti-CD38 antibodies in 5xFAD mice ameliorated AD-related pathology, improving recognition memory and restoring CD4⁺ T cell metabolic fitness (Peralta Ramos et al., 2025).

CD8⁺ T cells constitute another major lymphocyte population infiltrating the AD brain. Pathological effects of infiltrating CD8⁺ T cells include neuroinflammation, cognitive deficits, and neuronal dysfunction associated with impaired synaptic plasticity (Ma et al., 2024). Notably, CD8⁺ T cells can exacerbate oligodendrocyte and myelin damage in 5xFAD mice by amplifying disease-associated microglia (DAM) responses and promoting the accumulation of aberrantly activated interferon (IFN)-responsive microglia (Kedia et al., 2024).

Conversely, certain CD8⁺ subsets may exert neuroprotective effects. Brain-resident programmed cell death protein-1 (PD-1)⁺CD8⁺ T cells colocalize with plaque-associated microglia via CXCL16–CXCR6 interactions. In CXCR6-deficient 5xFAD mice, hippocampal Aβ accumulation increased and cognitive deficits worsened alongside reduced CD8⁺ T cell accumulation in the brain, suggesting a protective role for this subset in limiting AD pathology (Su et al., 2023).

Overall, distinct CD4⁺ and CD8⁺ T cell subsets exert divergent influences on AD progression. Proinflammatory Th1/Th17 CD4⁺ cells, CXCR3⁺CD127⁺ Th1 cells enriched in CSF, metabolically impaired CD38⁺CD4⁺ cells, and cytotoxic CD8⁺ cells collectively intensify neuroinflammation, promote microglial activation, and contribute to neuronal and myelin damage. In contrast, anti-inflammatory Th2 cells, Tregs, and tissue-resident CXCR6⁺PD-1⁺CD8⁺ cells that interact with plaque-associated microglia provide counterbalancing neuroprotection. Notably, CXCR3⁺CD127⁺ Th1 cells may display context-dependent protective roles, as their peripheral abundance inversely correlates with circulating NfL, suggesting a contribution to immune homeostasis.

REGULATORY T CELLS: NEUROPROTECTIVE ROLES IN AD PATHOLOGY

Regulatory T cells (Tregs) play neuroprotective, anti-inflammatory roles in AD pathogenesis. Growing evidence indicates that Tregs modulate CNS glial function and promote polarization toward neuroprotective states (Baek et al., 2016; Liston et al., 2022; Stym-Popper et al., 2023). Ex vivo-expanded Tregs administered to immunodeficient recombination-activating gene-2 (Rag2) knockout 5xFAD mice reduced Aβ accumulation, concomitant with fewer plaque-associated CD68⁺ microglia and reactive astrocytes. This intervention was also associated with decreased expression of proinflammatory cytokines, complement components, toll-like receptors, and microglial activation markers (Faridar et al., 2022). Conversely, early Treg depletion via anti-CD25 monoclonal antibody significantly increased neurotoxic A1-like C3⁺ reactive astrocytes around amyloid deposits while reducing neuroprotective A2-like Cox2⁺ astrocytes near larger plaques (Stym-Popper et al., 2023).

MEMORY T CELLS: UPREGULATION OF T_EM AND T_EMRA CELLS IN AD PATIENTS

Memory T cells are functionally heterogeneous and can be classified by CD45 isoforms and C-C chemokine–receptor type 7 (CCR7). CD45 (protein tyrosine phosphatase receptor type C, PTPRC) is commonly referred to as the leukocyte common antigen (Al Barashdi et al., 2021). CD45RA⁺CCR7 naïve T cells differentiate into CD45RO⁺CCR7- effector memory T (T_EM) cells and CD45RA⁺CCR7- terminally differentiated effector memory T (T_EMRA) cells, the latter characterized by the re-expression of CD45RA isoforms (McGuire et al., 2025).

Clinical studies report increased frequencies of T_EM and T_EMRA subsets in AD. Clonally expanded CD8⁺ T_EMRA cells have been detected in the CSF of patients with AD, with single-cell T cell receptor (TCR) sequencing confirming that these cells represent the predominant clonally proliferating population (Gate et al., 2020). Communication between CD8⁺ T_EMRA cells and CD4⁺/CD8⁺ T_EM cells appears downregulated in AD, suggesting impaired T cell network interactions (Wang et al., 2024b).

Stage-specific alterations are also evident. PBMCs from early-stage AD patients, particularly APOE ε4 carriers, show elevated CD8⁺CD57⁺ T_EMRA cells associated with Aβ accumulation (Gericke et al., 2023). Individuals with mild cognitive impairment (MCI) exhibit increased CD8⁺ T_EM and T_EMRA cells with a proinflammatory transcriptional profile, alongside reduced immunosuppressive Tregs (Rickenbach et al., 2025); Aβ-positive MCI (APMCI) patients likewise display higher levels of these subsets (Grayson et al., 2023). Collectively, these findings suggest that expansion of T_EM and T_EMRA cells may represent an immunological hallmark of early AD progression, potentially linking peripheral immune alterations to CNS pathology.

MEMORY T CELLS: PATHOLOGICAL ALTERATIONS IN AD PROGRESSION

Emerging evidence indicates that memory T cells undergo pathological alterations during AD progression. In aging humans, CD8⁺ resident memory-like (T_RM) cells show homeostatic expansion, potentially compensating for age-related declines in CD8⁺ T cell production (Panwar et al., 2020). To model this process, Panwar and colleagues (2024) generated homeostatically induced T cell (hiT) mice via adoptive transfer: CD8⁺ T cells from 5-8-week-old C57BL/6 donors were transferred into immunocompromised 8-10-week-old B6.Foxn1 recipients, inducing age-like homeostatic expansion of CD8⁺ T_RM cells, among which APP-specific clones were enriched. This expansion produced AD-like features, including Aβ and tau pathology, neuronal loss, and cognitive impairment. Mechanistically, perforin-1 was required to initiate pathology, whereas IFN-γ further exacerbated neurodegeneration. Consistently, elevated perforin expression and accumulation of APP-specific CD8⁺ T cells were observed in human AD brain tissue (Panwar et al., 2024).

Additional evidence links memory T cell dysfunction to impaired interleukin-7 (IL-7) signaling. In late-onset AD patients carrying the APOE ε4 allele, plasma IL-7 levels were reduced, and IL-7 receptor (IL-7R) pathway genes were downregulated in PBMCs, particularly within T_EM subsets. Transcriptionally, CD4⁺ T_EM cells showed reduced CD28 and JUN, while CD8⁺ T_EM cells showed decreased IL7R and JUN expression. CD28 is a co-stimulatory receptor that augments TCR signaling to support activation and memory formation, whereas JUN, a component of the AP-1 transcription factor complex, is critical for T cell activation and downstream transcriptional programs (Zhang et al., 2024). Collectively, these findings implicate impaired homeostasis and signaling in memory T cell dysfunction during AD and suggest that rejuvenating memory T cell responses may be a promising therapeutic strategy.

MUCOSAL-ASSOCIATED INVARIANT T (MAIT) CELLS: MR1/MAIT AXIS CONTRIBUTION TO AD

Mucosal-associated invariant T (MAIT) cells are innate-like T cells enriched at mucosal sites. They recognize microbially derived vitamin B metabolite antigens presented by the MHC class I–related molecule (MR1) (Shrinivasan et al., 2024). Functional MR1 is broadly expressed across tissues and cell types, including astrocytes and microglia in the brain (Priya and Brutkiewicz, 2020), raising interest in the MR1/MAIT axis in neurodegenerative disease.

In postmortem AD brain tissue and in 5xFAD mice, microglial MR1 expression is markedly upregulated, particularly near Aβ plaques. Concordantly, MAIT cells accumulate in 5xFAD brains and display increased activation (elevated CD69 and CD25). Genetic ablation supports a functional role for this axis: crossing 5xFAD mice with MR1-deficient mice (5xFAD/MR1 KO) reduced Aβ plaque burden at 6 and 8 months (Wyatt-Johnson et al., 2023). In 8-month-old 5xFAD/MR1 KO mice, insoluble Aβ40 decreased, markers of dystrophic neurite markers such as lysosomal-associated membrane protein 1 (LAMP1), ubiquitin, and N-terminal APP were reduced, and the synaptic marker postsynaptic density protein 95 (PSD95) increased in the hippocampus (Wyatt-Johnson et al., 2025). Together, these findings suggest that the MR1/MAIT axis contributes to dystrophic neurite formation and synaptic dysfunction, whereas MR1 disruption mitigates these features.

Beyond the CNS, peripheral MAIT alterations have also been reported. Wyatt-Johnson and colleagues (2024) found that both MAIT cell proportions and CD25 expression were increased in the liver of 5xFAD mice, whereas conventional T cell populations were unchanged. Notably, mature B-cell–deficient mice (μMT) displayed reduced hepatic MAIT abundance, pointing to B-cell–MAIT interactions and motivating investigation of this axis as a potential immunotherapeutic target in AD (Wyatt-Johnson et al., 2024).

Building on the earlier dichotomy of CD4⁺ and CD8⁺ subsets, the distinct functions of Tregs, memory T cells, and MAIT cells further refine the multifaceted role of T cells in AD pathology. Tregs appear principally neuroprotective by restraining detrimental microglial and astrocytic activation. In contrast, memory T cells, particularly T_EM, T_EMRA, and CD8⁺ T_RM subsets, and MAIT cells activated through the MR1/MAIT axis can exhibit pathogenic activities characterized by expansion and impaired regulation, amplifying neuroinflammation, enhancing glial activation, and contributing to dystrophic neurite formation. These converging human and murine data underscore that the balance between protective and pathogenic T cell subsets is a critical determinant of disease trajectory and support therapeutic strategies that restore Treg-mediated regulation or limit expansion/activation of pathogenic memory and MAIT subsets.

B CELLS: CYTOKINE- AND IMMUNOGLOBULIN-MEDIATED ROLES IN AD PATHOLOGY

While T cells represent one axis of adaptive immunity in AD, B cells are also key regulators of disease progression. B-cell contributions arise through cytokine secretion and immunoglobulin (Ig) production in response to AD-associated antigens (Plantone et al., 2022).

Transgenic models highlight potential protective B-cell functions. In B-cell–deficient 5xFAD mice, Aβ accumulation, glial activation, synaptic dysfunction, and cognitive impairment were exacerbated in the frontal cortex versus controls. Interleukin-35 (IL-35) expression was elevated in the frontal cortex of 3-month-old 5xFAD mice but markedly diminished in the absence of B cells. Feng and colleagues (2023) proposed that IL-35 signals via the SOCS1 pathway to inhibit STAT1 phosphorylation; because phosphorylated STAT1 promotes transcription of β-site APP-cleaving enzyme-1 (BACE1), IL-35 activity may suppress Aβ generation. Consistently, IL-35 neutralization aggravated AD pathology and increased BACE1 expression (Feng et al., 2023). In parallel, PBMCs from patients with AD produced higher levels of proinflammatory cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF), IFN-γ, and tumor necrosis factor-α (TNF-α), while producing reduced levels of the anti-inflammatory cytokine IL-10 upon stimulation; these alterations correlated with cognitive decline, abnormal Aβ_42/40 ratios, and elevated phosphorylated tau (pTau)-181 (Wang et al., 2025).

B-cell–derived Igs also modulate AD pathology. Postmortem analyses report elevated IgG and IgM in late-stage AD brains versus controls (Lekhraj et al., 2022). In human induced pluripotent stem cell (iPSC)-derived microglia, IgG enhanced phagocytic activity under acute Aβ exposure, whereas chronic Aβ exposure reversed this effect, impairing phagocytosis and suppressing homeostatic transforming growth factor-β (TGF-β) expression (Park et al., 2022).

Collectively, these findings highlight dual B cell roles in AD: altered cytokine secretion and Ig production can shape neuroinflammation, amyloid deposition, and synaptic integrity. Clarifying these mechanisms may open avenues for immunotherapies that modulate B cell function.

B CELLS: DISTINCT SUBSETS IN RELATION TO AD

Emerging evidence suggests that distinct B-cell subsets differentially contribute to AD pathogenesis, underscoring the need for precise phenotyping. Recent single-cell RNA-sequencing analyses reported increased peripheral memory B cells and macrophage-annotated myeloid cells in patients with AD. Notably, peripheral memory B cells secreted macrophage migration inhibitory factor (MIF), which engages the CD74–CD44 receptor complex on macrophages. This signaling axis is proposed to promote M1-like macrophage polarization, thereby enhancing neuroinflammation and impairing Aβ-plaque clearance (Liu et al., 2024). Additional insight comes from Iannucci and colleagues (2024), who examined B cells expressing the MHC class II–associated invariant chain peptide (CLIP), a proteolytic fragment of CD74 previously implicated in traumatic brain injury (TBI)—a recognized risk factor for AD. In line with fluid-percussion TBI models, elevated meningeal CLIP⁺ B cells were detected in 9-month-old 5xFAD mice. Administration of a competitive CLIP antagonist peptide (CAP) reduced CD74⁺ cell abundance, restored hippocampal neurogenesis, and ameliorated neurobehavioral deficits in 5xFAD mice (Iannucci et al., 2024).

B CELLS: DUAL ROLE OF NATURALLY OCCURRING ANTIBODIES (NABS)

One prominent mechanism by which B cells influence AD is the production of naturally occurring antibodies (NAbs). In early AD, BBB dysfunction may expose brain-derived self-antigens to the periphery, facilitating NAb generation (Gu et al., 2023). Supporting this, B-cell receptor (BCR) repertoire analyses show shared class-switched BCR sequences among AD patients, suggesting recurrent activation by common antigens during disease progression (Park et al., 2022).

Endogenous NAbs have been studied for potential neuroprotection, particularly those targeting Aβ, in parallel with therapeutic monoclonal antibody development (Liu et al., 2021). More recent work indicates that NAbs recognize a broader spectrum of AD-related antigens, exerting protective or pathological effects depending on specificity.

On the neuroprotective side, Gu and colleagues (2023) identified plasma NAbs against AD-associated linear peptide antigens, among which NAbs directed against triggering receptor expressed on myeloid cells 2 (TREM2) were significantly reduced in AD patients. Importantly, these NAbs correlated positively with cognitive function, consistent with the critical role of TREM2 in regulating microglial responses to Aβ (Gu et al., 2023; Zhang et al., 2025b). Similarly, NAbs targeting presenilin-1 (PS1), the catalytic subunit of γ-secretase, were decreased in AD patients and showed a negative correlation with Aβ load and a positive correlation with cognition (Wang et al., 2022). Jian and colleagues (2022) further reported reductions in NAbs specific for B-cell lymphoma-2 (Bcl-2)–interacting mediator of cell death (Bim), a pro-apoptotic protein upregulated by Aβ toxicity (Sanphui and Biswas, 2013). Treatment with Bim-specific NAbs in APP/PS1 mice attenuated memory deficits, suppressed neuronal apoptosis and synaptic degeneration, and reduced tau hyperphosphorylation and BACE1 expression, underscoring their therapeutic potential (Jian et al., 2022).

Conversely, other NAbs appear to exacerbate AD pathology. He and colleagues (2023) reported elevated levels of NAbs against the p75 neurotrophin receptor (pNTR) extracellular domain (p75ECD) in the cerebrospinal fluid of AD patients. Although cleavage of p75NTR by TNF-α converting enzyme (TACE) generates soluble p75ECD that can attenuate Aβ aggregation, NAbs to p75ECD were associated with increased Aβ burden, gliosis, neuroinflammation, tau phosphorylation, neuronal apoptosis, and cognitive decline (He et al., 2023).

Collectively, these findings highlight the dual roles of NAbs in AD. Depending on antigenic targets, NAbs may mitigate disease by enhancing clearance and preserving neuronal function or exacerbate pathology by interfering with protective mechanisms. This antigen-driven dichotomy underscores the complexity of humoral immunity in AD and positions NAbs as candidates for biomarkers and therapeutic modulation.

ADAPTIVE IMMUNITY-BASED IMMUNOTHERAPIES FOR AD

Given the growing recognition of adaptive immune cells as key modulators of AD pathology, multiple therapeutic strategies are under investigation. Beyond efforts to pharmacologically steer CD4⁺ T cell differentiation and limit brain infiltration, clinically translatable approaches include adoptive transfer of regulatory T cells (Tregs) and active vaccination. These strategies, summarized in Table 1, highlight adaptive immunity as a promising avenue for immunotherapeutic intervention in AD.

Table 1.

Summary of adaptive immunity-based therapeutic strategies in AD

Modulators of CD4+ T cell differentiation
Agent	Target	Experimental model	Observed effects	References
CDC42	Th17 and Th2 cells	PBMCs of AD patients	Negatively correlated with Th17; positively correlated with Th2 and MMSE scores	Zhang and Niu, 2022
JKAP	Th1 and Th17 cells	PBMCs of AD patients	Negatively correlated with Th1/Th17; positively correlated with MMSE scores and Aβ42; negatively correlated with pTau	Zeng et al., 2022
PR-957 (LMP-7 inhibitor)	Th1 and Th17 cells	APP/PS1 mice; PBMCs of AD patients	Suppressed Th1/Th17 differentiation via PI3K/AKT inactivation	Li et al., 2023
Low-dose IL-2	Th17 cells and Tregs	APP/PS1 mice	Induced Tregs; restored Th17/Treg balance; reduced neuroinflammation and cognitive decline; upregulated plaque-associated microglia	Yuan et al., 2023
U. rhynchophylla alkaloids (URA)	Effector Th1/Th17 cells and Tregs	APP/PS1 mice	Inhibited effector Th1/Th17 brain infiltration; increased Tregs; reduced neuroinflammation and AD pathology	Kuang et al., 2024
Adoptive transfer therapy of Regulatory T cells
Agent	Generation strategy	Experimental model	Observed effects	References
Aβ+ Tregs	ex vivo expansion of Tregs with Aβ and bvPLA2	5xFAD mice	Reduced neuroinflammation, Aβ and pTau accumulation	Park et al., 2024
	in vivo immunization with human Aβ1-42 and CFA; in vitro incubation with Aβ and bvPLA2	3xTg-AD mice	Suppressed Aβ-specific Teffs and reactive microglia; reduced Aβ/tau pathology and neuroinflammation; mitigated cognitive impairment	Yang et al., 2022
	ex vivo expansion of Tregs with aggregated human Aβ1-42 and bvPLA2	3xTg-AD mice	Improved memory and learning; reduced neuroinflammatory responses and Aβ/tau pathology	Yang et al., 2024a
Aβ+ hTregs (VT301)	ex vivo expansion following addition of fibrillary Aβ, and bvPLA2	AD patients	Demonstrated safety and feasibility; no significant efficacy	Yang et al., 2024a
TCRAβ-transduced Tregs	Tregs transduced with a transgenic Aβ-specific TCR	TCR KO APP/PS1 mice	Reduced Aβ deposition and reactive microglia via CX3CR1-CX3CL1 interaction; ameliorated cognitive impairment	Yeapuri et al., 2023
CX3CR1+ Tregs	CX3CR1-retrovirally transduced Tregs	LPS-injected 3xTg-AD mice	Reduced microglial activation, neuroinflammation; ameliorated cognitive impairment	Yang et al., 2024b
Active immunotherapy of AD vaccines
Agent	Immunogen design	Experimental models	Observed effects	References
Aβ1-10 S8R peptide vaccine	Aβ1-10 S8R peptide B-cell epitope with OVA/KLH carrier T-cell epitopes	Aβ oligomer-injected C57BL/6J mice	Induced IgG; reduced Aβ aggregation; downregulated APP and BACE1; attenuated glial activation, neuroinflammation, and synaptotoxicity	Park et al., 2025
CRM197-scaffolded vaccine	CRM197 scaffold carrying multiple Aβ3-10 and tau294-305 peptide B-cell epitopes	APP/PS1 mice	Induced IgG; reduced Aβ deposition and proinflammatory cytokines (IL-6 and IL-1β); ameliorated memory deficits	Cui et al., 2024
Norovirus P particle-scaffolded vaccine	Norovirus P particle scaffold with Aβ1-6 and pTau peptide B-cell epitopes, CpG and MF59 adjuvants	3xTg-AD mice	Reduced Aβ/pTau deposition, glial activation, and proinflammatory cytokines (IL-1β and TNF-α); ameliorated cognitive deficits; no sign of microhemorrhage	Feng et al., 2024

THERAPEUTIC MODULATION OF CD4⁺ T CELL DIFFERENTIATION AND BRAIN INFILTRATION

Aberrant CD4⁺ T cell responses are increasingly implicated in AD pathogenesis: elevated Th1/Th17 cells with reduced Th2 cells and Tregs can exacerbate neurodegeneration. Several studies have examined molecular regulators of CD4⁺ T cell differentiation in AD. For example, cell division cycle 42 (CDC42) expression correlated negatively with Th17 cells but positively with Th2 cells and Mini–Mental State Examination (MMSE) scores in PBMCs from patients with AD (Zhang and Niu, 2022). Likewise, JNK pathway–associated phosphatase (JKAP) expression was inversely associated with Th1/Th17 proportions; higher JKAP levels correlated with better MMSE scores and lower phosphorylated tau, although, paradoxically, with increased Aβ₄₂ (Zeng et al., 2022). Pharmacological inhibition with PR-957, a selective low-molecular-mass polypeptide 7 (LMP7) inhibitor, suppressed Th1/Th17 differentiation by inactivating the PI3K–AKT pathway in APP/PS1 mice and in PBMCs from patients with AD (Li et al., 2023). Similarly, low-dose IL-2 expanded Tregs, restored the Th17/Treg balance, reduced neuroinflammation, increased plaque-associated microglia, and slowed cognitive decline in mid-stage APP/PS1 mice (Yuan et al., 2023).

Beyond differentiation, limiting effector CD4⁺ T cell infiltration into the brain has emerged as a therapeutic goal. Alkaloid compounds from Uncaria rhynchophylla (URA) reduced Th1/Th17 infiltration while increasing Tregs in APP/PS1 mice. This shift elevated anti-inflammatory cytokines (IL-10, TGF-β), suppressed proinflammatory cytokines (IFN-γ, IL-17), and improved cognition, with decreased Aβ deposition, reduced tau phosphorylation, and diminished neuronal apoptosis (Kuang et al., 2024).

ADOPTIVE TRANSFER THERAPY OF REGULATORY T CELLS

Tregs play well-established neuroprotective roles, but their immunosuppressive capacity progressively declines during AD progression (Abdelmoaty et al., 2023). To restore Treg-mediated neuroprotection, adoptive transfer therapies have been explored, particularly using Aβ-specific Tregs (Aβ⁺ Tregs) designed to enhance targeting of brain regions burdened with Aβ.

Park and colleagues (2024) generated ex vivo expanded Aβ⁺ Tregs by incubating Tregs with Aβ and bee venom phospholipase A2 (bvPLA2) in G-rex plates. Intravenous administration of these cells in 5xFAD mice mitigated neuroinflammation and reduced Aβ and phosphorylated tau accumulation in the hippocampal CA1 region (Park et al., 2024). Similarly, Yang and colleagues (2022) combined in vivo immunization with Aβ_1-42 and complete Freund’s adjuvant (CFA) and in vitro co-incubation with Aβ and bvPLA2 to generate Aβ⁺ Tregs. Adoptive transfer of these cells into 3xTg-AD mice suppressed Aβ-specific effector T cells (Teffs), restrained reactive microglia, and ameliorated cognitive impairment, Aβ accumulation, tau hyperphosphorylation, and neuroinflammation (Yang et al., 2022).

Translating these findings to the clinic, preclinical studies and a phase I clinical trial have investigated ex vivo expanded Aβ⁺ human Tregs (hTregs). Stimulation with aggregated human Aβ_1-42 and bvPLA2 allowed expansion of hTregs, which improved learning and memory upon intrathecal administration to 10-month-old 3xTg-AD mice. Hippocampal analyses revealed suppression of nitric oxide synthase 2 (NOS2), CD86, proinflammatory cytokines, chemokines, and disease-associated microglia (DAM)-related genes, alongside reduced Aβ deposition and tau phosphorylation. In the corresponding phase I, single-center, open-label trial of Aβ⁺ hTregs (VT301), intravenous administration of Aβ⁺ hTregs to patients with mild-to-moderate AD demonstrated safety and feasibility, although efficacy did not reach statistical significance (Yang et al., 2024a).

Recent studies have further refined Treg adoptive therapy by enhancing antigen specificity and microglial targeting. Yeapuri and colleagues (2023) cloned an Aβ-specific T cell receptor (TCR_Aβ) by immunizing mice with Aβ_1-42 and transduced it into Tregs for transfer into TCR-knockout APP/PS1 mice. These TCR_Aβ-transduced Tregs reduced Aβ deposition, decreased reactive microglia/DAM phenotypes, and improved cognition (Yeapuri et al., 2023). Additionally, Yang and colleagues (2024b) engineered Tregs to overexpress the fractalkine receptor (CX3CR1), enabling inhibitory crosstalk with proinflammatory microglia via CX3CL1; adoptive transfer of CX3CR1⁺ Tregs into LPS-challenged 3xTg-AD mice suppressed microglial activation, attenuated neuroinflammation, and improved cognition (Yang et al., 2024b).

NEXT-GENERATION ACTIVE AD VACCINES

While recently approved monoclonal antibody passive immunotherapies (aducanumab, lecanemab, and donanemab) do not engage endogenous B-cell activation or antigen-specific T cell responses, active immunotherapies aim to harness the patient’s adaptive immune system. Among these, active AD vaccines promote sustained endogenous antibody production and require fewer administrations than passive therapies. Early-generation vaccines, however, produced limited cognitive benefit and meningoencephalitis-like adverse events (Bhadane et al., 2024).

To address these limitations, next-generation vaccines have been developed. Park and colleagues (2025) designed an Aβ_1-10 epitope vaccine by conjugating the optimized Aβ_1-10 S8R B-cell epitope to carrier proteins (ovalbumin [OVA] or keyhole limpet hemocyanin [KLH]) to provide T cell help and enhance immunogenicity. In Aβ-oligomer-injected 8-10-week-old male C57BL/6J mice, intraperitoneal vaccination rescued memory deficits, elicited robust IgG responses, reduced Aβ aggregation, and suppressed proteins in the amyloidogenic pathway (APP, BACE1). The vaccine also attenuated glial activation, neuroinflammation, and Aβ-induced synaptotoxicity (Park et al., 2025).

Beyond Aβ-targeted strategies, combinatorial Aβ/tau vaccines have been explored. Cui and colleagues (2024) engineered CRM197-scaffolded vaccines bearing multiple copies of Aβ_3-10 and tau_294-305 B-cell epitopes. Subcutaneous administration of candidates carrying either six Aβ_3-10 insertions or four insertions of both Aβ_3-10 and tau_294-305 elicited robust IgG responses, improved memory, reduced Aβ deposition, and suppressed proinflammatory cytokines (IL-6, IL-1β) in APP/PS1 mice (Cui et al., 2024). Similarly, Feng and colleagues (2024) developed a norovirus P-particle–based combination vaccine displaying three copies of Aβ_1-6 and pTau peptides (pTau202, pTau205, pTau396, and pTau404). Formulated with CpG and MF59, intramuscular vaccination of 3xTg-AD mice—both pre- and post-onset—generated high-titer anti-Aβ and anti-pTau antibodies, reduced pathological aggregation, improved cognition, attenuated glial activation, and lowered IL-1β and TNF-α levels; no microhemorrhages were observed (Feng et al., 2024).

CONCLUSION AND FUTURE PERSPECTIVES

Recent advances have expanded our understanding of Alzheimer’s disease (AD) beyond its classical innate immune framework, positioning adaptive immune cells as central and dynamic regulators of disease initiation and progression. Blood–brain barrier dysfunction in neurodegeneration enables peripheral lymphocytes to access the CNS through chemokine–receptor interactions, cytokine signaling, and antigen-driven activation, creating a critical interface between peripheral immunity and CNS pathology. These infiltrating T and B cells display context-dependent roles—exacerbating neurotoxicity in some settings while conferring neuroprotection in others—highlighting adaptive immunity as a nuanced and highly targetable dimension of AD biology.

Moving forward, advancing this field will require a more structured and deliberate research roadmap. One priority is to disentangle age-associated immunosenescence from AD-specific immune exhaustion, two processes that may superficially resemble each other yet arise from distinct biological drivers. Immunosenescence reflects the gradual decline of immune competence with aging, whereas AD-specific exhaustion may be driven by chronic antigen exposure, sustained neuroinflammation, or persistent interactions with amyloid or tau pathology. Defining molecular signatures that differentiate these states will be essential for designing therapies that restore immune function without exacerbating inflammation.

Another critical direction is to elucidate sex-dimorphisms in adaptive immune responses, which may influence T-cell activation states, cytokine production, infiltration patterns, and responsiveness to immunotherapies. Integrating sex as a biological variable into experimental design, patient stratification, and clinical trial frameworks will improve the precision and generalizability of adaptive immunity–based interventions.

In parallel, future studies should map stage-specific alterations in adaptive immune function across the AD continuum, establish reliable biomarkers for monitoring these changes, and evaluate combinatorial treatment strategies that integrate immune modulation with existing disease-modifying approaches. Ensuring long-term safety remains essential, particularly in preventing excessive immune activation or unintended immunosuppression.

Collectively, these insights underscore adaptive immunity as both a mechanistic contributor to AD pathogenesis and a promising frontier for therapeutic innovation. By embracing a roadmap that incorporates immunological aging, disease-specific immune dysfunction, and sex-dependent variation, the field is well positioned to develop more effective and personalized interventions for Alzheimer’s disease.