AbstractThe application of chimeric antigen receptor T (CAR-T) cells in hematologic malignancies has driven significant advancements in this form of immunotherapy. The therapeutic strategy of CAR-T cells targeting specific cell populations has opened new avenues for treating non-oncological diseases, such as autoimmune diseases, aging-related conditions, and infections. For instance, in non-oncological diseases like systemic lupus erythematosus (SLE), abnormal B cell development or dysfunction leads to the production of autoantibodies, triggering localized deposition of immune complexes and resulting in tissue or organ damage and dysfunction. Relevant studies have identified disease-specific surface antigens or pathogenic cell subsets, making CAR-T cell therapy a feasible treatment approach. Moreover, in non-oncological diseases, CAR-T cells can mediate immune remodeling for certain conditions, thereby achieving long-term therapeutic remission. Although no CAR-T therapies have yet been approved for non-oncological diseases, multiple clinical trials have been initiated, with some achieving interim successes (e.g., allogeneic CAR-T cells have demonstrated efficacy in treating rheumatic diseases). Meanwhile, ongoing research into the pathogenesis of non-neoplastic diseases further supports the potential application of CAR-T cells. With this background, this article aims to introduce the latest research, mechanisms, and applications of CAR-T cell therapy in non-oncological diseases, summarize the pathogenesis of related disorders, and discuss the advantages, challenges, and future prospects of CAR-T cell therapy.IntroductionChimeric Antigen Receptor T (CAR-T) cells are genetically engineered T lymphocytes expressing synthetic transmembrane receptors that confer antigen-specific targeting capabilities. Structurally, these chimeric receptors integrate an extracellular antigen-binding domain (typically a single-chain fragment variable, scFv) with intracellular signaling modules capable of activating T cell effector functions. Upon antigen engagement, CAR-T cells initiate cytotoxic programs through perforin-granzyme secretion and Fas/FasL pathways, while concurrently recruiting innate immune components via cytokine release (e.g., interferon (IFN)-γ, interleukin (IL)-2), thereby orchestrating coordinated antitumor immunity.1,2,3,4,5 The conceptual foundation of CAR technology emerged from early attempts to reconfigure immune receptor architectures. In 1987, TCR-antibody fusion enabled MHC-independent antigen recognition, decoupling specificity from TCR signaling.6,7 By 1989, chimeric TCRs targeting folate-binding protein lysed tumor cells, validating engineered T-cell therapy.8,9 In 1993, Eshhar et al. created the first CAR by linking an anti-TNP scFv to CD3ζ, combining antibody-like targeting with TCR activation. This modular design paved the way for later CAR generations with costimulatory domains (e.g., CD28, 4-1BB).10,11,12The clinical application of chimeric antigen receptor (CAR) T-cell therapy achieved a major milestone in 2017 with the first Food and Drug Administration (FDA) approval of tisagenlecleucel for B-cell malignancies. As of November 2024, regulatory agencies including both the U.S. FDA and European Medicines Agency (EMA) have approved seven distinct CAR-T cell products. These therapies primarily target two key antigens: CD19 for B-cell malignancies (including relapsed/refractory diffuse large B-cell lymphoma (DLBCL), follicular lymphoma, mantle cell lymphoma (MCL), and B-cell acute lymphoblastic leukemia (ALL)), and B-cell maturation antigen (BCMA) for relapsed/refractory multiple myeloma.13,14,15,16,17,18,19,20,21,22,23,24 Mechanistically, CAR-T cells recognize CD19 or BCMA via their engineered chimeric receptors, initiating dual cytotoxic pathways: (1) perforin/granzyme-mediated direct target cell lysis, and (2) Fas/FasL interaction-induced apoptosis. Concurrently, activated CAR-T cells secrete pro-inflammatory cytokines (e.g., IFN-γ, Tumor necrosis factor (TNF)-α), recruiting innate immune effectors (macrophages, Natural killer (NK) cells) to amplify tumoricidal activity. This multi-modal attack achieves sustained depletion of malignant CD19+ B-cell populations.25The transformative success of CAR-T cell therapy in oncology has catalyzed its exploration across immune-mediated pathologies, leveraging its precision targeting to reprogram aberrant immune activity. In autoimmune conditions, pathogenic B cells play a central role by producing autoantibodies that form immune complexes, thereby driving aberrant immune activation and tissue damage.26 Beyond autoimmunity, CAR-T cells may also address other pathological processes: senescent cells, which exacerbate inflammation through senescence-associated secretory phenotypes (SASP)27; cardiomyocyte injury, which triggers fibroblast activation and progressive fibrosis28; and chronic infections (e.g., HIV, HBV, Aspergillus), which perpetuate immune dysregulation. In the context of infectious diseases like HIV and HCMV, where immune deficiency dominates the clinical picture, CAR-T cells are designed to directly eliminate infected cells and restore immune competence.29,30 Similarly, in graft-versus-host disease (GvHD)—a complication of allogeneic transplantation—regulatory CAR-T cells hold promise for suppressing excessive alloimmune responses while preserving protective immunity.31 This expanding repertoire of applications underscores the versatility of CAR-T technology in recalibrating disrupted immune homeostasis across diverse disease states.32,33,34While CAR-T therapy offers distinct advantages for non-malignant diseases, including potential long-term remission through immune memory formation35 and reduced cytokine release syndrome risk compared to cancer applications, several challenges remain: Target specificity: Many senescence-associated antigens are also expressed on healthy tissues.36 Precision requirements: Chronic diseases demand exceptionally specific targeting to avoid off-tissue effects. Cost barriers: High expenses associated with autologous/allogeneic manufacturing.37This article comprehensively examines CAR-T applications across non-oncologic diseases, including: autoimmune disorders, infectious diseases, aging and senescence, cardiovascular and metabolic disorders and neurological diseases. Analyzing their design principles, mechanisms of action, therapeutic potential, the aim is to provide insights into target selection, pathogenesis research, CAR design optimization, and strategies for enhancing the safety and controllability of CAR-T cell therapy in non-oncological diseases. This review seeks to contribute to the clinical application and broader adoption of CAR-T cells in non-oncological diseases.Structure and mechanism of CAR-TStructure and generations of CARsChimeric antigen receptors (CARs) are fusion proteins consisting of an intracellular signal transduction domain (such as CD3ζ), a transmembrane domain, and an extracellular domain which is single-chain variable fragment (scFv) derived from the antibody. The specificity of CARs is mainly determined by the extracellular scFv domain, which specifically recognizes target cells, followed by transmitting activation signal to CAR-T cells for cytotoxic effects.11,12,38 (Fig. 1).Fig. 1Full size imageCAR structure and iteration diagram. scFv: single-chain variable fragment, the smallest functional fragment with full antigen activity, containing the antibody recognition and antigen-binding sites; CD3ζ: cluster of differentiation 3, the CD3ζ chain typically exists as a ζ-ζ homodimer connected to the intracellular segment of the T cell receptor (TCR) and associates non-covalently with the TCR to form the TCR/CD3 receptor complex. In this complex, CD3 is responsible for mediating the transmission of signals from the extracellular domain to the intracellular domain upon TCR stimulation, thereby activating intracellular signal transduction; TM: transmembrane domain. 1st generation CAR-T cells use intracellular CD3ζ as the signaling domain, with an extracellular scFv domain that recognizes and binds to the target antigen. 2nd generation CAR-T cells incorporate an additional co-stimulatory domain, such as CD28, in the intracellular structure, enhancing the activation and persistence of the engineered T cells. 3rd generation CAR-T cells build upon the 2nd generation by adding another co-stimulatory domain, such as 4-1BB, further synergistically enhancing T cell activation and persistence. 4th generation, also named as TRUCKs, are genetically engineered to secrete cytokines, such as IL-12, which recruit NK cells and macrophages to kill target cells that are not directly bound by the CAR. This generation also incorporates various gene regulatory elements through genetic modification to further enhance the function of CAR-T cells. 5th generation CAR-T cells introduce intracellular co-stimulatory structures that activate the JAK-STAT signaling pathway, thereby enhancing the cytotoxic function of CAR-T cells. TRUCKs T cells Redirected for Antigen-Unrestricted Cytokine-mediated Killing, NK cells natural killing cellsThe development of CAR constructs has evolved through five distinct generations, each addressing critical limitations of previous designs. First-generation CARs, incorporating only the CD3ζ signaling domain, demonstrated restricted clinical efficacy primarily due to insufficient T cell expansion and rapid exhaustion upon repeated antigen exposure.39 This key limitation prompted the development of second-generation CARs, which incorporated additional co-stimulatory domains (CD28, 4-1BB, or OX40) alongside CD3ζ, resulting in significantly enhanced T cell activation, expansion, cytotoxicity, and persistence.40,41,42,43,44,45 Building on this advancement, third-generation CARs combined multiple co-stimulatory domains (typically CD3ζ, CD28, and 4-1BB) to further optimize signaling pathways, though comparative studies have shown variable impacts on in vivo persistence depending on domain combinations.46,47 The emergence of fourth-generation CARs, termed TRUCKs (T cells Redirected for Antigen-Unrestricted Cytokine-mediated Killing), introduced two major innovations: (1) capacity for cytokine secretion (particularly IL-12) to recruit and activate innate immune cells against antigen-negative tumor cells, and (2) incorporation of safety mechanisms including suicide genes and logic-gated control systems for enhanced regulation. Most recently, fifth-generation CARs have integrated novel co-stimulatory domains capable of activating JAK-STAT signaling pathways, offering potential advantages in T cell proliferation and long-term persistence.38,46,48,49,50,51,52,53,54 This progressive evolution of CAR designs reflects an ongoing effort to balance potent antitumor activity with improved safety and durability of cellular responses (Fig. 1).Engineering strategiesThis therapeutic gap has driven intensive research into optimizing the chimeric antigen receptor (CAR) structure itself, with several innovative strategies currently under investigation to enhance precision and functionality. A key advancement involves the development of logic-gated systems, where CAR activation requires specific input combinations (OR, AND, or NOT gates).55,56,57,58,59,60 These synthetic biology approaches create multi-signal responsive circuits that improve targeting accuracy while maintaining system reliability—if one activation pathway fails, alternative pathways can still trigger the desired response.55,56,57,58,59,60 Such OR logic gates can be implemented through various designs including bivalent,55,56,57 bicistronic,58 two vectors,59 co-infusion.60Parallel efforts focus on cellular fitness enhancement through genetic modifications. These include overexpression of pro-survival factors like c-Jun,61 knockout of inhibitory receptors (PD1),62 TGFBR2,63 HPK1 (also known as MAP4K1)64), and incorporation of safety switches (EGFRt,65 CD20 epitope,66 HER2t,67 HSV-tk,68 iCasp969) for controlled elimination if needed. Regulatable platforms represent another critical innovation, enabling precise spatiotemporal control of CAR activity through external inducers like AP1903-inducible costimulation,70 antibody-coupled,71 fluorescein-CAR,72 switchable CAR T cells,73 or antigen receptor complex (ARC) T cells.74To combat the immunosuppressive tumor microenvironment, “armored” CAR-T cells have been engineered with enhanced defensive capabilities. These include dominant-negative receptors (TGFβR,75 PD-1,76 FAS77), checkpoint inhibitors (PD-1-Fc, anti-PD-1 scFv78,79), cytokine modifications (IL-12,80 IL-18,81 NFAT-induced IL-12,82 IL-15 and IL-2183), and switch receptors (PD-1–CD28, IL-4R–IL-2Rβ84). Complementary strategies aim to engage endogenous immunity through chemokine receptor overexpression (CXCR5, CCR4, CCL19, CCR2, CX3CR185,86) or immune modulator expression (CD40L overexpression87).Finally, improving CAR-T cell expansion and persistence remains a major focus, with approaches including IL-15 secretion,88 sushi domain,89 tethered,90 IL-7: secretion,91 or mutant constitutive.92 These multifaceted optimization strategies aim to overcome current limitations and expand CAR-T therapy’s potential across broader disease indications.Mechanisms of action in non-oncological diseasesMechanism of CAR-T cellsCAR-T cells exert therapeutic effects through an antigen-driven activation cascade (Fig. 2). Target recognition begins when the CAR’s scFv domain engages cognate antigens, triggering intracellular signaling via costimulatory domains (CD28/4-1BB) and CD3ζ ITAMs.40,42,43,93,94,95 This initiates three core effector mechanisms96,97: 1. direct Cytotoxicity: perforin/granzyme-mediated pore formation and caspase activation; and Fas-FasL induced apoptosis98; 2. immune modulation includes: secretion of IFN-γ and TNF-α promote the infiltration of monocytes/macrophages while enhancing macrophage activation and the secretion of chemokines such as CCL2. These cytokines also stimulate IL-12 production, which activates downstream signaling pathways to induce IFN-γ secretion by T cells and NK cells. The released IFN-γ further amplifies the activation of macrophages and dendritic cells (DCs), establishing a positive feedback loop that sustains and intensifies the immune response,99,100 and establishment of pro-inflammatory microenvironments; 3. immunological memory formation: generation of central/effector memory T cell subsets for sustained surveillance.101Fig. 2Full size imageMechanism of CAR-T to target cells. After the extracellular scFv domain of the CAR binds to the antigen on the surface of the target cell, the intracellular CD3ζ domain mediates ITAM phosphorylation. This is followed by signal transduction via ZAP70 to the intracellular signaling domains, leading to CAR-T cell activation and the initiation of immune responses. These immune effects occur primarily through: direct secretion of perforin and granzymes, which induce target cell death; production of cytokines that recruit and activate other immune cells such as macrophages, T cells, and dendritic cells at the target site, triggering a localized immune response; generation of memory T cells, which enable a rapid immune response upon re-exposure to the target antigen. ITAM: immunoreceptor tyrosine-based activation motif; ZAP70: 70 kDa c-related proteinThe functional heterogeneity among T cell subsets (memory T cells and effector T cells) within individual patients provides critical insights for the efficacy of CAR-T cell immunotherapy. T cells mature in the thymus and express T cell receptors (TCRs). When expressing CD8 on their surface, they are termed CD8+ T cells (cytotoxic T cells), while those expressing CD4 are called CD4+ T cells (helper T cells).102 CD4+ T cells further differentiate into distinct subsets: Th1, Th2, Th9, Th17, Th22, Treg (regulatory T cells), and Tfh (follicular helper T cells), each characterized by unique cytokine profiles that play pivotal roles in immune and effector responses. All CD4+ Th subsets originate from naive CD4+ T cells through specific cytokine induction: Th1 differentiation is driven by IL-12 and IFN-γ, Th2 by IL-4, and Treg by IL-2 and TGF-β.103,104 Each Th subset releases characteristic cytokines that mediate pro-inflammatory, anti-inflammatory, survival, or protective functions. For instance, Th1 cells secrete IFN-γ and TNF; Th2 cells produce IL-4 (a crucial survival factor for B lymphocytes), IL-5 and IL-13; Th9 generates IL-9; Treg secretes IL-10 (an immunosuppressive cytokine that maintains FOXP3 transcription factor expression required for Treg-mediated suppression) and TGF-β; while Th17 produces IL-17 (a cytokine essential for host defense against bacteria and fungi).105Notably, cytokines from Th1 and Th2 CD4+ T cell subsets have been shown to differentially modulate cytotoxicity in second-generation CD28-costimulated CAR-T cells—high concentrations of Th1 cytokines induce acute toxicity, whereas elevated Th2-type cytokines cause chronic autotoxicity in animals receiving CD19-targeted CAR-T therapy. Another example involves engineered CAR-T cells producing inducible IL-12, which recruits macrophages, reprograms immunosuppressive cells, and modifies tumor stroma to enhance tumor killing.106T cell differentiation, particularly the balance between memory and effector T cells, plays a fundamental role in pathogen-directed immunity. Effector cells are short-lived, while memory cells possess long-term survival potential. The persistence of CAR-T therapy has been demonstrated to depend on the infusion product’s content of CD4+ cells and central memory cells (CD45RO+CD62L+).107 Naive CD8+ T cells can differentiate into stem cell memory T cells (TSCM), central memory T cells (TCM), effector memory T cells (TEM), and terminal effector cells (TEFF). During CD8+ T cell differentiation, effector functions increase while memory potential and proliferative capacity decline, with distinct biomarkers including L-selectin, CD45RO, CD45RA, and CCR-7 marking different stages.108 Non-human primate studies have shown that CD8+ clones derived from central memory T cells (rather than effector cells) exhibit prolonged persistence after adoptive transfer.109 Furthermore, combined CD8+ and CD4+ subsets significantly enhance T cell transfer efficacy, with CD4+ cells supporting the development of CD8+ memory function. These findings underscore the importance of specific T cell subsets and their combinations in immunotherapy.Immune reconstitutionCAR-T cell therapy represents a transformative approach for immune reconstitution in autoimmune diseases through two complementary mechanisms: targeted depletion of autoreactive lymphocytes and restoration of regulatory functions. The pathogenesis of autoimmune disorders stems from complex interactions between genetic predisposition and environmental triggers that disrupt immune tolerance.110,111,112,113 This breakdown manifests most prominently as aberrant activation of B and T lymphocytes against self-antigens, resulting in pathogenic autoantibody production. These autoantibodies form immune complexes with self-antigens, triggering inflammatory cascades that ultimately cause tissue damage. The precision of CAR-T technology enables selective targeting of these autoreactive lymphocytes through chimeric antigen receptors designed to recognize lymphocyte surface markers. Upon engagement, CAR-T cells induce apoptosis of pathogenic B and T cells through both direct cytotoxicity (perforin/granzyme-mediated lysis) and death receptor signaling. Clinical applications demonstrate this strategy’s potential: CD19-directed CAR-T cells have achieved durable B-cell depletion and clinical improvement in systemic lupus erythematosus (SLE),114,115 and KYV-101, a CD19-targeting CAR-T product, has shown efficacy in depleting pathogenic lymphocytes in both multiple sclerosis and rheumatoid arthritis.116,117,118Equally important is CAR-T therapy’s capacity to augment impaired regulatory mechanisms. The immune system maintains homeostasis through balanced activation and suppression, and autoimmune diseases often feature regulatory T cell (Treg) dysfunction. Engineered CAR-Tregs address this defect by: suppressing pathogenic cytokine production (e.g., IL-23R CAR-Tregs reduce IFN-γ and TNF in inflammatory bowel disease).119,120 Controlling alloreactive responses in graft-versus-host disease through HLA-A2-specific, CD19-targeted, or OX40L-directed CAR-Tregs.121,122,123,124 These regulatory CAR-T cells promote long-term immune tolerance without causing generalized immunosuppression.Senescent cell clearanceCellular senescence represents a fundamental biological process marked by irreversible cell cycle arrest.125 This physiological state is accompanied by profound alterations in cellular architecture and function, including chromatin remodeling, metabolic reprogramming, and the development of a distinctive senescence-associated secretory phenotype (SASP). While serving as an important tumor suppression mechanism, the pathological accumulation of senescent cells contributes significantly to chronic inflammation, tissue dysfunction, and malignant transformation through SASP-mediated bystander effects. The therapeutic potential of senescent cell clearance has been established in preclinical studies, prompting the development of targeted elimination strategies. A key advancement in this field involves the identification of urokinase-type plasminogen activator receptor (uPAR) as a nearly universal surface marker of senescent cells. This discovery has enabled the design of precision CAR-T therapies capable of specifically recognizing and eliminating senescent cell populations. Preclinical evidence demonstrates the efficacy of this approach: uPAR-targeted CAR-T cells effectively reduce senescent cell burden in vivo, treatment extends survival in murine models of lung adenocarcinoma, therapeutic benefits include restoration of tissue homeostasis in liver fibrosis models.126,127,128,129,130 These findings highlight the potential of CAR-T technology to address senescence-related pathologies through targeted cell clearance, offering advantages over traditional senolytic approaches by providing: antigen-specific senescent cell recognition, in vivo persistence for sustained effects, capacity for tissue penetration and immune memory against recurring senescent cells.The success of uPAR-directed CAR-T cells in preclinical models establishes a proof-of-concept for immunologically targeting cellular senescence, with potential applications across age-related diseases, fibrotic disorders, and senescence-associated malignancies.Pathogen-specific targetingCAR-T cell therapy has demonstrated versatility in targeting pathogen-infected cells through specialized receptor designs. This approach leverages pathogen-specific surface markers to enable precise immune recognition and clearance of infected host cells. Several engineered CAR-T cells have shown promising results against diverse infectious agents: HIV-specific bNAbs-CAR-T cells effectively identify and eliminate HIV-infected cells by targeting conserved viral epitopes,131 for hepatitis B virus (HBV) infection: S/L-CAR-T cells specifically recognize HBV surface antigens or demonstrating potent activity against HBV-infected hepatocytes,79,132,133 in cytomegalovirus (HCMV) infection: gB-specific CAR-T cells target the viral glycoprotein B and show specific cytotoxicity against HCMV-infected cells, and antifungal applications: Dectin-1 CAR-T cells recognize β-glucans on Aspergillus species, CD8+ Af-CAR-T cells target fungal surface antigens and both demonstrate effective clearance of fungal pathogens.50,134,135 These cells directly act on infected cells to form an immune clearance effect. The development of these targeted therapies represents a significant advancement in antimicrobial immunotherapy, offering potential advantages over conventional treatments through their specificity, adaptability, and capacity for persistent surveillance. Each CAR construct is carefully designed to recognize conserved pathogen structures and minimizing off-target effects.Applications by disease categoryAutoimmune diseasesSystemic diseasesSystemic lupus erythematosus (SLE)Systemic lupus erythematosus (SLE) is a heterogeneous autoimmune disease driven by autoreactive B cells producing antinuclear antibodies (ANA), which form immune complexes, triggering inflammation and organ damage.136,137 Innate immune dysregulation involves excessive type I interferon (IFN) signaling, neutrophil extracellular traps (NETs), and dendritic cell (DC) activation via TLRs, promoting BAFF production and chronic immune activation.138,139,140Adaptive immune dysfunction includes T-cell abnormalities, leading to aberrant cytokine release and CD40L-mediated B-cell activation, while Treg and Breg impairment disrupts immune tolerance, sustaining autoantibody production and inflammation.137,141,142,143,144Significant progress has been made in the treatment of SLE over the past few decades; existing treatment options include corticosteroids, antimalarial drugs, immunosuppressants, and biologics. However, these treatments require long-term use, which can have significant side effects, and their effectiveness often diminishes over time.145,146 Current methods for eliminating alloantibodies include plasma exchange, intravenous immunoglobulin, and monoclonal antibodies. However, these approaches cannot eradicate memory B cells or plasma cells, resulting in limited therapeutic efficacy. The demonstrated success of CAR-T cells in B-cell malignancies—where CD19 CAR-T cells have achieved long-term remission in chronic lymphocytic leukemia patients and eliminated tumor cells in acute lymphoblastic leukemia and B-cell non-Hodgkin lymphoma-suggests potential applications for systemic lupus erythematosus (SLE). The primary mechanism of CD19 CAR-T cells involves binding to CD19 on B-cell surfaces, triggering a cascade of immune responses and curing B-ALL. Theoretically, this long-term B-cell depletion mechanism could be adapted for SLE treatment.147 In-depth investigation of B-cell depletion mechanisms following CAR-T therapy in SLE patients has confirmed selective CAR T cell-mediated resetting of responsive B-cell populations, while revealing transcriptional changes in monocyte and T-cell subsets—particularly marked reduction in type I IFN signaling that correlates with SLE pathogenesis, providing rationale for CAR-T therapy in SLE. Current clinical trials have progressed to Phase I, demonstrating symptom relief (reduced proteinuria, decreased anti-dsDNA antibody levels) in SLE patients. For instance, Li et al. have found a woman with refractory thrombocytopenia complicating SLE had a response to CD19 CAR T cells with an increase in platelet counts and minimal symptoms of cytokine release syndrome (CRS),148 Mackensen et al. treated 5 SLE patients with CD19 CAR-T cells following lymphodepletion,114 while Müller et al. treated 8 patients, observing sustained clinical improvement for up to two years with only mild CRS.149 Unlike in malignancies, SLE patients showed rapid immune cell recovery post-treatment, suggesting preserved immune competence. A case report by Mougiakakos et al. described rapid resolution of acute symptoms without adverse events in a severely affected 20-year-old SLE patient, confirming that pathogenic autoantibodies originate from CD19+ B cells and supporting CD19 CAR-T efficacy.150 However, pre-treatment lymphodepletion in these studies may confound accurate assessment of CAR-T effects,149 though remains necessary to ensure therapeutic efficacy of adoptive immunotherapy.151 Uppin et al. proposed BAFF-targeting CAR-T cells (against BAFF-R, BCMA, and TACI antigens on mature B cells) based on observed clinical improvement with BAFF-specific belimumab and CD20-specific rituximab in SLE patients. Humanized mouse models demonstrated symptom relief with such CAR-T cells, with theoretical advantages of selectively depleting pathogenic mature B cells while sparing immature B cells and CD19- plasma cells.152 However, current evidence remains limited to in vitro and mouse studies—significant interspecies differences (e.g., stronger BAFF-BCMA binding in humans) may enhance clinical efficacy, but human SLE heterogeneity necessitates further safety and effectiveness studies. Notably, BCMA expression on mature B cells increases with serum BAFF levels (particularly on CD19+ B cells),153 and memory B cells/plasma cells resistant to CD19 CAR-T may reactivate to produce autoantibodies.154 BAFF CAR-T cells could enable sustained remission if achieving long-term persistence.152Zhang et al. developed combo CD19/BCMA (APRIL) CAR-T cells that effectively cleared B-cell and plasma-cell subpopulations in murine models. In desensitization experiments, combo CAR-T outperformed single-target therapies, reducing donor-specific antibody (DSA) levels below baseline, and suggesting potential to eliminate vaccine-induced protective antibodies and HLA alloantibodies which minimizing graft-versus-host reactions.155 This contrasts with CD19 CAR-T’s inability to clear such antibodies.156 While BCMA CAR-T alone shows limited efficacy against B cells (allowing plasma-cell repopulation),157 combining with Treg CAR-T may maintain therapeutic effects through immune modulation.157Wang et al. initiated Phase I trials of combo CAR-T in SLE patients, demonstrating effective pathogenic cell clearance, immune reset, and symptom improvement with good safety—particularly promising for refractory cases.115 However, long-term persistence data (inferred from depletion levels) lack clonal tracking post-treatment. The impact on other antibody types (e.g., anti-MHC I/II or drug antibodies) remains unaddressed—particularly relevant given SLE’s diverse autoantibody profile beyond just antinuclear antibodies.157 The challenge of long-lived plasma cells (LLPCs) lacking surface immunoglobulin expression requires mechanistic investigation to improve targeting, as this underlies treatment resistance in late-stage SLE. CAR-T’s antigen-specific precision could prove transformative if surface markers on pathogenic plasma-cell subsets can be identified.CAR-T cells applied to the treatment of systemic lupus erythematosus (SLE) still face several challenges, such as: differences in B cell kinetics between SLE and leukemia patients: B cell reconstitution in SLE patients occurs more rapidly than in leukemia patients, so the kinetics of CD19 CAR-T cells in SLE need further exploration, particularly regarding their targeting efficiency and persistence.114,115,158 B cell regeneration impairment: CAR-T therapies targeting B cell lineage markers (such as CD19) or B cell maturation antigen (BCMA) have shown efficacy but have chance to lead to long-term B cell depletion.159,160 To improve specificity, chimeric autoantibody receptor T (CAAR-T) cells have been developed to selectively eliminate autoreactive B cells while sparing healthy ones. However, CAAR-T therapy is limited in its long-term effectiveness due to its inability to effectively target long-lived plasma cells.161 Target selection challenges: While 100% of SLE patients test positive for antinuclear antibodies (ANA), the presence of other autoantibodies varies. Identifying the optimal autoantibody target for CAR-T therapy remains a critical and unresolved challenge162 (Fig.3).Fig. 3Full size imageMechanisms of autoimmune diseases. 1, Systemic Lupus Erythematosus (SLE) is triggered by environmental and genetic factors, leading to immune dysregulation. This manifests as excessive formation of neutrophil extracellular traps (NETs), abnormal expression of the type I interferon gene phenotype causing over-secretion, and activation of Toll-like receptors (TLRs). These processes induce dendritic cell (DC) activation and the production of B cell-activating factors (BAFF), creating a positive feedback loop between these factors and cells. This ultimately leads to B cell activation and the production of autoantibodies. These autoantibodies form immune complexes that abnormally deposit in tissues and are presented by antigen-presenting cells to T cells, which due to inherent dysfunction are activated alongside B cells, leading to tissue damage. CAR-T therapies such as BAFF (targeted to BAFFR and BCMA) CAR-T cells, CD19 CAR-T cells, and Combo CAR-T cells can recognize and eliminate the autoantibody-producing B cells. CAR-Treg help regulate the immune environment and support the function of BCMA CAR-T cells. 2, Graft-versus-host disease (GVHD) typically arises from host tissue damage caused by radiation or chemotherapy. Damage-associated molecules like TNF and type I interferons are recognized by host antigen-presenting cells (APCs), which activate donor T cells, triggering a cascade of immune-mediated damage. HLA-A2-specific CAR-Tregs, anti-human CD19 CAR-Tregs, and OX40L-CAR-Tregs are used to suppress excessive T cell responses. CD7 CAR-T cells eliminate CD3⁺, CD7⁺ T cells and NK cells. 3, Systemic sclerosis (SSc) is a complex autoimmune disease with a multifactorial etiology involving genetic predisposition, environmental triggers, immune dysregulation, vascular damage, and fibrosis. The primary driver is immune dysregulation, characterized by Th2 cell polarization and hyperactivated B cells; B cells produce anti-centromere antibodies (ACA), anti-Scl-70 antibodies, and anti-RNA polymerase III antibodies. Dysregulated cytokines, including TGF-β, IL-4, IL-6, and IL-13, contribute to excessive pro-fibrotic signaling. These pathological processes lead to fibrosis in internal organs and skin, vasculopathy, non-healing wounds (ulcers), pruritus, and complications such as pulmonary arterial hypertension (PAH). CD19 CAR-T cell targets and eliminates pathogenic B cells, this approach has shown potential in reversing fibrosis and restoring immune balance. 4, Myasthenia Gravis involves Th1 and Th2 cells causing direct damage to acetylcholine receptors (AChRs) under the effect of anti-AChR autoantibodies, along with BAFF-mediated B cell activation, leading to autoantibody production. These mechanisms collectively damage AChRs at the neuromuscular junction, resulting in muscle weakness. Descartes-08 (anti-BCMA RNA CAR-T cells) and MuSK-CAART specifically target and eliminate pathogenic B cells producing these antibodies. 5, Rheumatoid Arthritis (RA) is characterized by autoantibodies such as rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPAs). These antibodies interact with T cells, recognize APCs or bind to antigens on B cells, triggering cell-to-cell interactions (e.g., inflammation), which cause joint damage and functional impairment. KYV-101 and FITC CAR-T cells are used to target and eliminate autoreactive B cells. 6, Idiopathic Inflammatory Myopathies show elevated expression of MHC class I and II molecules in muscle tissue, which activate NK cells and T cells, leading to direct muscle damage or cytokine production. Complement activation causes immune complex deposition, damaging perimysial blood vessels and muscle cells. CD19 CAR-T cells are used to target and eliminate pathogenic B cells. 7, Asthma is associated with a Th2-mediated immune response, where T cells secrete cytokines (inflammatory mediators). These cytokines activate antibody-dependent cytotoxic cells (neutrophils, macrophages, NK cells) and the complement pathway, resulting in local inflammation or direct tissue damage. In asthma, Treg function is often impaired, reducing their ability to regulate Th2 responses. 5TIF4-CAR-T cells target and eliminate eosinophils. Interactions with CAR-NKT cells disrupt allergen–mast cell/basophil interactions. CD8⁺ Af-CAR-T cells target fungal infections that trigger asthma, eliminating the associated T cells. 8, Inflammatory Bowel Disease (IBD) is driven by overproduction of IL-23 by macrophages and dendritic cells, which activates Th17 cells. The resulting cytokines cause local inflammation and recruit additional immune cells via positive feedback. CD7 CAR-T cells target and eliminate pathogenic T cells, while IL-23R CAR-Tregs and regulatory CD4⁺CD25⁺FoxP3⁺ T cells (Tregs) modulate the immune response and reduce levels of IFN-γ and TNF. 9, Pemphigus Vulgaris is primarily caused by IgG autoantibodies against desmoglein (Dsg) core proteins in desmosomes. B cells that produce these antibodies form immune complexes, activate immune cells, and induce cytokine release, recruiting immune cells and promoting inflammation and memory B cell formation, leading to recurrent inflammation. Dsg3 CAAR-T cells specifically target B/plasma cells expressing Dsg3-specific B cell receptors. CAR-T cells expressing a recombinant Dsg3–CD137–CD3ζ T cell receptor eliminate these pathogenic cells. BCMA B-cell maturation antigen, NMJ neuromuscular junction, MP/Mø macrophagesCurrently, there are numerous clinical trials targeting SLE, most of which are still in Phase I and primarily focused on assessing adverse drug reactions. The promising early-phase trial data position CAR-T therapy as a potential paradigm shift in SLE treatment. However, addressing SLE’s profound heterogeneity remains the central challenge. The varying autoantibody profiles among patients question the viability of universal CAR-T products and point toward future personalized targeting strategies. Beyond initial B-cell depletion, the field is advancing toward precision approaches like CAAR-T cells. Yet the persistent barrier of long-lived plasma cells (LLPCs)—resistant to current therapies due to lack of surface targets—demands novel antigen discovery or combinatorial strategies. Notably, the rapid immune reconstitution and sustained remission observed in early trials suggest CAR-T may enable an “immune reset” in autoimmune contexts, distinct from its oncological mechanism. Future success hinges on long-term clonal tracking of CAR-T cells and detailed monitoring of the reconstituted immune repertoire to verify durability. Ultimately, transitioning CAR-T therapy into clinical practice will require trials that stratify patients by immunopathological signatures, moving toward truly personalized cellular therapy for autoimmune disease.Graft-versus-host disease (GvHD)Graft-versus-host disease (GvHD) is a condition that occurs following allogeneic hematopoietic stem cell transplantation (allo-HSCT). It arises when the surviving allogeneic immune effector cells in the recipient’s body attack the recipient’s organs, leading to complications in various tissues. On a microscopic level, GvHD is triggered when donor T cells encounter allogeneic stimuli, become activated, secrete cytokines, and proliferate into effector cells. This initial donor anti-host response induces a systemic effect that manifests as a multi-organ syndrome. There are two types of GvHD: acute GvHD(aGvHD) and chronic GvHD(cGvHD). Acute GvHD progresses through three phases: tissue damage from conditioning releases DAMPs/PAMPs, activating host APCs; donor CD4+/CD8+ T cells are primed by APCs; and effector T cells and cytokines (e.g., TNF, IFN-γ) target epithelial cells, causing apoptosis and organ damage.163 Chronic GvHD develops through overlapping stages: early inflammation, thymic injury leading to immune dysregulation (escape of autoreactive T cells), and fibrosis driven by IFN-γ and cytokine imbalance (e.g., IL-6, TNF-α). Unlike aGvHD, cGvHD involves both donor alloreactive and host autoreactive T cells, perpetuating tissue fibrosis.164,165T cells are the primary pathogenic factor. The broad expression of CD7 on T cells and their precursors provides a therapeutic target.166 Li et al. reported a case study where a patient received universal CD7 CAR-T cell therapy after chemotherapy preconditioning, resulting in complete elimination of tumor cells. The therapy effectively expanded CD7 CAR-T cells and eradicated CD3+, CD7+ T cells and natural killer (NK) cells, with therapeutic effects lasting up to four months. The patient subsequently underwent hematopoietic stem cell transplantation to restore hematopoiesis and remained free of graft-versus-host disease (GvHD) for three months without immunosuppressive drugs.167 This suggests that universal CD7 CAR-T therapy may potentially reduce GvHD risk while creating a favorable environment for transplantation. However, other studies have reported GvHD occurrence following universal CD7 CAR-T cell therapy.167 Therefore, the efficacy of universal CD7 CAR-T against GvHD remains uncertain and requires further investigation. Key questions include whether prolonged CD7+ T cell depletion may lead to post-transplant relapse and/or increased infection risk, and whether CD7- T cells retain graft-versus-leukemia (GVL) effects and anti-infection capabilities.166,168Regulatory T cells (Tregs) show promising potential for GvHD treatment by preventing or mitigating unwanted immune responses in transplantation or autoimmune conditions.169 Recent advances demonstrate that HLA-A2-specific CAR-Treg cells can effectively suppress HLA-A2+ peripheral blood mononuclear cell (PBMC) proliferation in vitro and prevent xenogeneic GvHD induced by HLA-A2+ PBMCs in mice.170 Dawson et al. also observed HLA-A2+ cell-mediated rejection and accordingly developed a humanized Treg cell product (hA2-CAR Tregs) that reduced graft rejection, providing supportive evidence for clinical application of engineered Tregs.31 Additionally, anti-human CD19 CAR-Tregs have been shown to suppress antibody production in immunodeficient mice reconstituted with human PBMCs, achieving GvHD inhibition without compromising graft-versus-tumor (GVT) effects.171While these findings strongly support Treg applications, it should be noted that CAR-T cell therapy itself may induce GvHD. Consequently, managing CAR-T-induced GvHD has gained attention, primarily through genetic engineering approaches. One strategy involves TCR knockout (particularly TRAC gene disruption) to prevent T cell recognition and attack of host tissues, such as utilizing non-αβ T cells or genetically eliminating TCR from αβ T cells.172 Another approach incorporates an inducible caspase-9 (iCasp9) suicide gene system, enabling pharmacological induction of apoptosis in case of severe toxicity or off-target effects.173 The safety profile of CAR-T cell therapy for GvHD requires further validation. Given the high costs of autologous T cell modification, allogeneic CAR-T cell research is actively progressing. Alternative strategies include (Fig. 3).CAR-T cells therapy for GvHD is evolving along two complementary paths: universal CAR-T cells targeting markers like CD7 aim to eliminate pathogenic immune cells, while engineered Tregs focus on restoring immune tolerance. Despite promising early results, key challenges remain. Preserving graft-versus-leukemia effects and anti-infection capacity during targeted T-cell depletion requires careful balance. Additionally, while genetic engineering strategies such as TCR knockout and suicide switches have improved the safety profile of allogeneic CAR-T products, concerns regarding delayed immune reconstitution and long-term immunosuppression persist. Looking ahead, the field should prioritize developing precision-targeting strategies, including bispecific CARs or conditionally activated systems that distinguish pathogenic from protective immune cells. Combining engineered Tregs with targeted depletion therapies may establish a dual-layer “suppression-and-clearance” network for sustained tolerance. Furthermore, smart CAR-T cells capable of dynamically responding to biomarkers could pave the way for personalized GvHD treatment. As our understanding of GvHD pathogenesis deepens and gene-editing technologies advance, cell therapy holds potential to transition from an adjunct intervention to a curative modality for GvHD.Systemic sclerosis (SSc)Systemic sclerosis (SSc) is a complex autoimmune disease characterized by widespread organ fibrosis. Clinical manifestations include mobility impairment, non-healing wounds (ulcers), pruritus, pulmonary arterial hypertension, and organ failure, among others. Current conventional treatments for SSc primarily involve low-dose glucocorticoids combined with immunosuppressants such as cyclophosphamide and azathioprine. For patients with comorbid interstitial pneumonia, antifibrotic agents like pirfenidone or nintedanib are often used. Some patients may benefit from biologic therapies such as rituximab. However, overall, the efficacy of currently available treatments for SSc remains limited, often only slowing disease progression. Existing therapies cannot reverse established organ fibrosis, and some patients show poor response to all available drugs, resulting in an unfavorable prognosis.174Recent studies suggest that B cells play a crucial role in the pathogenesis of SSc, contributing to immune dysregulation, fibrosis, and vascular dysfunction. B cells secrete large quantities of autoantibodies and promote the release of pro-inflammatory or pro-fibrotic cytokines (particularly IL-6), interacting with other SSc-related cells such as vascular endothelial cells, vascular smooth muscle cells, and fibroblasts. This leads to extracellular matrix deposition, ultimately causing fibrosis in the skin, muscles, and internal organs.175,176 In a groundbreaking study, Schett et al. first employed anti-CD19 CAR-T cell therapy to treat a severely refractory SSc patient. The patient, a 60-year-old male, exhibited severe diffuse myocardial fibrosis, pulmonary fibrosis with pulmonary hypertension, Raynaud’s phenomenon, and wrist arthritis, with no response to immunosuppressive therapy. Following infusion of CAR-T cells (1 × 106 cells/kg), the patient showed improvement in skin fibrosis and resolution of wrist arthritis. Both the EUSTAR AI activity score and the modified Rodnan Skin Score (mRSS) decreased. The patient also reported reduced frequency and severity of Raynaud’s episodes.177Subsequently, the team enrolled four additional refractory SSc patients in a clinical trial. After CD19 CAR-T cell infusion, analysis of SSc-related autoantibodies revealed a moderate reduction in anti-Scl-70 antibodies, while autoantibodies targeting RP11, RP155, PM-Scl70, and CENP-B disappeared post-treatment. Correspondingly, these patients exhibited significant reductions in EUSTAR AI activity scores, mRSS, glucocorticoid dependence, and the need for immunosuppressive therapy.149Serial studies observed the re-emergence of naive, non-class-switched B cells, the disappearance of circulating plasmablasts, and downregulation of disease-associated immunoglobulin heavy and light chains, supporting the concept that CD19 CAR-T cell therapy may induce a “reset” of pathological autoimmunity in these patients. Although CAR-T cells target both pathogenic and non-pathogenic B cells, the reconstitution of the B-cell compartment after CAR-T cell depletion appears to occur in the absence of autoreactive B-cell clones, potentially explaining long-term disease remission. However, the contribution of concomitant immunosuppressive therapy to the short-term effects of CAR-T treatment cannot be entirely ruled out, necessitating further observation of drug-free disease progression in SSc. Given the limited immunosuppressive intensity of the conditioning regimen (lower than standard SSc treatment protocols), additional studies involving more patients—particularly those with poor prognostic factors such as anti-Scl-70 positivity—are warranted.In another study, Merkt et al. utilized third-generation CD19 CAR-T cells to treat a 38-year-old woman with anti-Scl-70-positive SSc and rapidly progressive nonspecific interstitial pneumonia (NSIP). The addition of CD19 CAR-T cells to her existing mycophenolate/nintedanib regimen led to sustained improvement in pulmonary function, accompanied by significant radiological regression. This study was the first to describe the disappearance of Fc receptor-activating immune complexes in an autoimmune disease following CAR-T therapy. In SSc, such immune complexes are thought to contribute to pathology via Fcγ receptor activation. This innovative approach improved pulmonary outcomes and optimized immunosuppressive therapy, though the underlying mechanisms require further investigation. Additionally, whether the persistence of CAR-T cells plays a deterministic role remains to be validated in larger cohorts178 (Fig. 3).CD19 CAR-T therapy shows promising results in refractory systemic sclerosis, demonstrating improvement in skin fibrosis, arthritis, and pulmonary function with reduced clinical scores. The treatment appears to achieve immunological reset through B-cell depletion and reconstitution, evidenced by plasmablast disappearance and autoantibody profile improvement without autoreactive clone recurrence. However, key questions require further investigation: the contribution of concomitant immunosuppression to therapeutic outcomes needs clarification, particularly given the mild preconditioning regimen. The mechanistic relationship between Fc receptor-activating immune complex disappearance and clinical improvement warrants exploration. Additionally, the correlation between CAR-T persistence and long-term efficacy remains unverified. Addressing these issues through larger studies, especially in high-risk populations, will be crucial for standardizing and optimizing this innovative approach for SSc treatment.Musculoskeletal systemMyasthenia gravis (MG)Myasthenia gravis (MG) is an autoimmune disorder where antibodies target nicotinic acetylcholine receptors (AChRs) at neuromuscular junctions, impairing synaptic transmission and causing muscle weakness via complement activation, AChR blockade, and accelerated degradation.179,180,181 CD4+ T helper cells (Th1, Th2, Th17) drive pathogenic anti-AChR antibody production through pro-inflammatory cytokines (IFN-γ, IL-17A) and B-cell activation, while Tregs (CD4+CD25+FOXP3+) and Th3 cells (TGF-β) regulate immune tolerance.182,183,184,185,186 A rare MuSK-MG subtype involves anti-MuSK antibodies, which disrupt AChR clustering and synaptic folding, distinct from AChR-MG pathogenesis.187Anti-BCMA RNA CAR-T cells (rCAR-T) called Descartes-08 are being explored for treating autoimmune diseases.188,189 In MG, muscle-specific tyrosine kinase (MuSK) antibodies, primarily IgG4, disrupt the physiological functions of MuSK in synaptic maintenance and adaptation, causing disintegration of the neuromuscular junction. MuSK is a transmembrane tyrosine kinase with an extracellular domain that includes three immunoglobulin-like (Ig1-Ig3) and frizzled-like (Fz) domains. Studies show that sera from MuSK MG patients can recognize Ig1, Ig2, and Ig3-Fz domains at rates of 100%, 58%, and 23%, respectively.190 Researchers have designed T cells to express MuSK-chimeric autoantibody receptor (MuSK-CAART) with a CD137-CD3ζ signaling domain to specifically target B cells that express anti-MuSK antibodies. MuSK-CAART has demonstrated efficacy similar to anti-CD19 CAR-T cells in eliminating anti-MuSK B cells and has retained cytolytic activity even in the presence of soluble anti-MuSK antibodies. The designed MuSK-CAAR structure, which contains the full extracellular domain of MuSK linked to CD137-CD3ζ co-stimulatory and activation domains, has shown comparable activity, indicating that it can be successfully expressed. MuSK-CAART treatment depletes antigen-specific IgG without depleting all B cells and achieves therapeutic effects without prior lymphodepletion. Current research demonstrates promising targeted approaches for myasthenia gravis treatment. Descartes-08, an anti-BCMA RNA CAR-T cell therapy, is under investigation for autoimmune applications. More notably, MuSK-CAART has been engineered to specifically eliminate anti-MuSK antibody-producing B cells while sparing normal B cells. This approach shows comparable efficacy to anti-CD19 CAR-T in depleting target cells, maintains activity despite soluble antibodies, achieves therapeutic effects without lymphodepletion, and has demonstrated no off-target toxicity in preclinical studies. However, these findings remain at the animal testing stage, requiring validation through clinical trials before clinical application can be considered187 (Fig. 3).Rheumatoid arthritis (RA)Rheumatoid arthritis (RA) is a chronic autoimmune disease characterized by progressive joint inflammation and destruction, driven by dysregulated immune responses involving T/B lymphocytes, macrophages, and inflammatory cytokines.191,192,193,194,195 Key mechanisms include: T-cell activation by citrullinated antigens via HLA-DRB1, triggering synovitis and anti-citrullinated protein antibody (ACPA) production196; macrophage polarization toward proinflammatory M1 phenotypes (TNF-α, IL-6, IL-1β) with impaired M2-mediated repair196,197; and B-cell dysfunction, producing autoantibodies (rheumatoid factor, anti-CCP) and proinflammatory cytokines while regulatory B/T cells (Tregs, Bregs) fail to suppress autoimmunity.198,199,200,201,202,203Current treatment options include rituximab, an anti-CD20 monoclonal antibody that demonstrates clinical efficacy by depleting B cells.204,205 However, this broad depletion strategy carries significant risks of infection and compromises protective immune memory responses.206,207 More targeted approaches using CAR-T cell therapy are emerging as potential alternatives. The fully humanized CD19-CAR-T product KYV-101, previously effective in neurologic autoimmune conditions, has shown promise in inducing remission in RA patients following lymphodepleting chemotherapy,118 though the required immunosuppression remains a limiting factor.208 Li et al. applied fourth-generation CAR-T cell therapy to three rheumatoid arthritis (RA) patients, innovatively integrating three functional modalities: tumor necrosis factor (TNF) inhibition, IL-6 suppression, and B-cell depletion. This approach not only eliminates B cells but also provides localized anti-inflammatory effects by targeting cytokine networks at B-cell infiltration sites in RA joints. The study demonstrated: robust in vivo expansion of CAR-T cells, complete eradication of CD19+ B cells, clearance of RA-associated autoantibodies, and sustained high-level clinical responses with persistent CAR-T cell activity.209 Another innovative approach utilizes universal FITC-CAR-T cells designed to target FITC-labeled RA-specific peptides, which have demonstrated dose-dependent cytotoxicity against autoreactive B cell subsets in vitro. While these CAR-T strategies show encouraging preliminary results, their application in RA remains largely conceptual, awaiting validation through preclinical in vivo studies and clinical trials.210,211 Future progress in autoantigen discovery and multi-omics technologies (combining serological, proteomic, and genetic analyses) may enable the development of precisely targeted CAR-T designs capable of selectively eliminating pathogenic lymphocyte populations while preserving protective immunity, building on current experimental successes in this rapidly evolving field (Fig. 3).KYV-101 shows clinical potential despite requiring immunosuppressive preconditioning. More innovative designs include fourth-generation CAR-T cells combining B-cell depletion with localized TNF and IL-6 suppression, demonstrating multi-modal efficacy in early studies. Additionally, universal FITC-CAR-T platforms enable targeting of specific autoreactive B-cell subsets. While these approaches present promising alternatives to conventional therapies like rituximab, they remain primarily conceptual for RA treatment. Further validation through preclinical in vivo studies and clinical trials is essential to establish their therapeutic potential and safety profile in rheumatoid arthritis.Idiopathic inflammatory myositis (IIM)Idiopathic inflammatory myositis (IIM) comprises six autoimmune subtypes characterized by muscle weakness and cutaneous manifestations, classified by myositis-specific autoantibodies.207,211 Key pathogenic mechanisms include: MHC I/II upregulation on muscle tissue, activating T/NK cells and causing myofiber damage; complement deposition in capillaries (dermatomyositis (DM), polymyositis (PM), anti-synthetase syndrome (ASS), immune-mediated necrotizing myopathy (IMNM)) and proinflammatory cytokines (IL-1, TNF, interferons) amplifying inflammation; and TREM-1-mediated immune cell migration and TNF-α secretion.207,210,211,212,213,214 Distinct pathological mechanisms operate across different IIM subtypes. In IMNM, autoantibodies directly contribute to myotube and muscle fiber destruction,215 while ASS is characterized by autoantibodies targeting aminoacyl-tRNA synthetases, with antibody titers often correlating with clinical disease severity.216Emerging clinical evidence demonstrates the therapeutic potential of CD19-targeted CAR-T cell therapy in refractory cases. Two male ASS patients who had failed multiple conventional immunosuppressive therapies, including rituximab and intravenous immunoglobulin, showed rapid and sustained clinical improvement following CAR-T treatment.217,218 Both patients achieved seroconversion of anti-Jo-1 antibodies, normalization of creatine kinase levels, and significant improvements in muscle strength and respiratory function. Similar outcomes were observed in a 45-year-old female patient with refractory myositis, arthritis, and pulmonary involvement who had previously failed nine different treatment regimens. Following CD19 CAR-T therapy, she demonstrated marked improvement across all disease manifestations according to ACR/EULAR criteria and maintained drug-free remission during 150 days of follow-up.219 Additional case reports have corroborated these findings, showing consistent efficacy of CD19 CAR-T therapy in ASS patients.176,219,220 While these clinical observations are promising, several key questions remain unanswered. The precise mechanism by which CD19-targeted CAR-T cells ameliorate IIM symptoms requires further elucidation, particularly given that some patients received combination therapies making it challenging to isolate the specific contribution of CAR-T treatment.206,221,222 Current research efforts include an ongoing Phase I/II clinical trial (NCT06154252) focused on evaluating adverse events and treatment efficacy223 (Fig. 3).CD19-directed CAR-T cell therapy demonstrates promising clinical potential in refractory idiopathic inflammatory myopathies, particularly in anti-synthetase syndrome (ASS). Multiple case reports document rapid and sustained improvement in muscle strength, respiratory function, and serological markers following treatment, with some patients achieving drug-free remission. However, the precise therapeutic mechanism remains unclear, complicated by concurrent immunosuppressive therapies in reported cases. Future research should focus on elucidating mechanisms of action, optimizing CAR constructs, establishing patient selection criteria, and conducting long-term safety monitoring to fully assess this innovative therapeutic approach.Respiratory systemAsthmaAsthma is a chronic respiratory disease affecting over 300 million people worldwide, characterized by airway inflammation, hyperresponsiveness (AHR), and reversible airflow obstruction.224,225,226,227 The pathophysiology of allergic asthma involves: epithelial alarmins (IL-25, IL-33, TSLP) activating ILC2s and Th2 cells, driving IL-4/5/9/13 production, eosinophilia, and IgE-mediated inflammation.228,229,230 And Treg suppression via CTLA-4 and granzyme-perforin mechanisms to balance Th2 responses.231,232,233,234 Current therapies (steroids, biologics) face limitations, prompting exploration of CAR-T cells for durable immunomodulation.235,236,237,238Innovative CAR-T strategies are being developed to precisely target these pathogenic pathways. A breakthrough approach involves engineered 5TIF4 cells—IL-5Rα-targeting CAR-T cells with BCOR/ZC3H12A knockout that co-express an IL-4 mutein to simultaneously block IL-4/IL-13 signaling.35,239 This design achieved durable suppression of allergic asthma in murine models with a single administration, eliminating the need for chronic treatment or lymphodepleting preconditioning. Alternative strategies leverage NKT cells’ unique immunomodulatory properties, with CAR-NKT constructs targeting IgE receptors to disrupt allergen–mast cell interactions while secreting anti-inflammatory cytokines (IFN-γ, TGF-β, IL-10).240,241,242,243,244,245,246,247 For IgE-driven disease, EMPD-specific CAR-T cells selectively deplete membrane IgE+ B cells without affecting circulating IgE or non-IgE producing B cell subsets, demonstrating precise targeting in preclinical models.248,249,250 Fungal asthma presents distinct challenges being addressed through dectin-1 CAR constructs that recognize β-glucans on Aspergillus and other fungi, with CD8+ Af-CAR-T cells showing potent antifungal activity via granzyme B and perforin release.50,251,252,253,254,255,256,257 Despite these promising developments, significant hurdles remain for clinical translation. The requirement for lymphodepleting chemotherapy poses challenges in non-oncological diseases like asthma, where the toxicity-risk profile must be exceptionally favorable.35 Additional concerns include potential on-target/off-tissue effects given shared antigen expression between pathogenic and normal cells,258 risks of immune surveillance impairment from prolonged B cell depletion, and theoretical oncogenic potential from genetic modifications.230,249,259,260 Most current evidence derives from murine models, underscoring the need for rigorous preclinical optimization and clinical validation to establish safety and efficacy profiles suitable for chronic respiratory conditions. Future directions will need to address these challenges while refining cellular persistence, specificity, and controllability to realize the potential of CAR-based therapies for asthma and other allergic diseases (Fig. 3).Digestive systemInflammatory bowel disease (IBD)Inflammatory bowel disease (IBD) arises from disrupted intestinal barrier function and dysregulated immune responses, creating a vicious cycle of inflammation.261,262,263,264,265,266,267,268,269 Key mechanisms include: IL-23/Th17 axis activation (via dendritic cells/macrophages) driving IL-17A/F and IL-22 production, amplifying T-cell-mediated inflammation270,271; and pathogenic B-cell involvement, where mucosal B cells and plasma cells in refractory cases produce pro-inflammatory cytokines and autoantibodies against epithelial cells, exacerbating tissue damage and treatment resistance.272Novel CAR-T cell strategies are being explored to target these pathogenic mechanisms. CD7-directed CAR-T cells offer potential by eliminating specific immune subsets (T cells, NK cells) involved in intestinal inflammation, as CD7 is expressed on these cell surfaces.273 An alternative approach utilizes engineered IL-23R CAR-Tregs, where regulatory T cells expressing the IL-23 receptor chimeric antigen receptor can suppress mucosal inflammation. These modified Tregs have demonstrated therapeutic potential in colitis models by reducing proinflammatory cytokine production (IFN-γ, TNF) while maintaining their immunoregulatory functions.274,275 However, IBD therapy development faces several challenges. The disease’s complex etiology complicates target selection, with current research primarily focusing on universal CD7 CAR-T cells and regulatory T cell approaches.273 Notably, observations from oncology research reveal that CD19 CAR-T cells can accumulate in intestinal tissue, inducing IBD-like symptoms.276 This phenomenon warrants careful consideration for two reasons: it may inform strategies to mitigate gastrointestinal toxicity in lymphoma treatment, while simultaneously providing insights into IBD pathogenesis through the study of CAR-T cell interactions with intestinal mucosa. These findings also highlight potential safety concerns regarding CAR-T cell localization and activation patterns in non-malignant conditions. As the field progresses, further investigation is needed to elucidate optimal cellular targets, improve specificity to avoid off-tissue effects, and develop strategies to control CAR-T cell trafficking and persistence. The dual potential of CAR-T therapies—to either treat or paradoxically induce intestinal inflammation—underscores the importance of rigorous preclinical evaluation and careful clinical translation for IBD applications (Fig. 3).Skin systemPemphigus vulgaris (PV)Pemphigus vulgaris (PV) is a severe autoimmune blistering disorder caused by IgG autoantibodies against desmoglein, leading to skin/mucosal blistering and increased mortality.277 Key pathogenic mechanisms include: CD4+ T cell-dependent activation of autoreactive B cells, generating Dsg-specific IgG through IL-5/6, TNF-α, and SDF-1α-mediated inflammation278,279,280,281; and Autoantibody-mediated disruption of epidermal adhesion, with IgG deposition directly causing acantholysis. Current therapies (corticosteroids or/and immunosuppressants) improve outcomes but carry significant risks (infections or cancer),282,283,284,285,286,287,288 underscoring the need for targeted treatments.Research on autoantibody production has opened new avenues for CAR-T cell therapy applications. Ellebrecht et al. developed Dsg3 CAAR-T cells by fusing the PV autoantigen with CD137-CD3ζ, which showed specific cytotoxicity against Dsg3-expressing B cells/plasma cells. In mouse models where experimental pemphigus was induced through Dsg3 hybridoma cell transfer, administration of Dsg3 CAAR-T cells prevented disease development.289 These engineered cells demonstrate targeted cytotoxicity against PV autoantigen-presenting cells in vitro, even in the presence of circulating autoantibodies, while maintaining minimal toxicity.289,290 Ongoing work by the same group aims to extend this approach to other pemphigus autoantigens like Dsg1,291 with related clinical trials currently underway. Therapeutic strategies targeting antibody-secreting B cells offer precise elimination of long-lived plasma cells, potentially enabling long-term remission. However, while CAAR-T cell therapy provides specific targeting of a B-cell populations, its effectiveness against short-lived plasma cells is limited by their lack of target antigen expression. Consequently, these cells can only be cleared indirectly, potentially increasing relapse risk.292 Further research is needed to address challenges posed by short-lived plasma cells, particularly given the current limitations in clinical trial data. Additionally, investigation of other potential therapeutic targets requires continued advancement to improve treatment outcomes (Fig. 3).Neurological diseasesMultiple sclerosis (MS)Multiple sclerosis (MS) is a chronic inflammatory disorder driven by an aberrant immune response against myelin protein peptides.11 The disease is initiated when peripherally activated CD4+ T cells migrate into the central nervous system (CNS), where they are reactivated by antigen-presenting cells (APCs) and recruit additional immune cells, including T cells and macrophages, to form inflammatory lesions. Th1 cells secrete pro-inflammatory cytokines such as IFN-γ and TNF-α, while Th17 cells produce IL-17, IL-21, and IL-22. CD8+ T cells further contribute to CNS inflammation by releasing IL-17 and IFN-γ, which activate resident immune cells (microglia, astrocytes, and macrophages) and sustain local inflammatory damage.293,294,295 A key feature of MS is the dysfunction of Foxp3+ regulatory T cells (Tregs), which exhibit impaired suppressive activity compared to those in healthy individuals. This deficiency allows autoreactive T cells and activated B cells to drive myelin destruction through cytokine production and autoantibody release.296,297,298 Studies suggest that induced Tregs (iTregs) play a critical role in suppressing pathogenic immune responses in MS, and their dysfunction exacerbates disease progression.299 Promisingly, myelin oligodendrocyte glycoprotein (MOG)-specific CAR-Tregs have demonstrated therapeutic potential in experimental autoimmune encephalomyelitis (EAE), a mouse model of MS. CARαMOG-Foxp3-expressing iTregs effectively suppressed T cell responses, particularly in the presence of MOG antigen and macrophages. Brain imaging revealed preferential localization of these engineered Tregs in regions such as the granular layer, lateral septal nucleus, and piriform cortex. Mice treated with CARαMOG-Foxp3 Tregs exhibited reduced CNS levels of IFN-γ and IL-12 compared to controls, highlighting their anti-inflammatory effects.116,300Despite these advances, significant challenges remain in Treg-based therapies for MS: Limited cell migration to target sites: Adoptively transferred Tregs often diminish in number during migration, with potential off-target accumulation in other organs. Delivery hurdles: While the olfactory pathway has been explored as a potential route for CNS delivery, systemic distribution via nasal vasculature remains a concern. Current tracking methods, such as olfactory nerve transmission monitoring, are limited to short-term observation (about 1 hour). And safety considerations: Exogenous Treg infusion may inadvertently suppress beneficial immune responses following CNS injury and even impair neuronal survival, necessitating caution in chronic neurodegenerative conditions.300,301 These findings underscore both the promise and complexities of CAR-Treg therapy for MS, emphasizing the need for refined delivery strategies and long-term safety assessments (Fig. 4).Fig. 4Full size imageMechanisms of neurological diseases. 1, Multiple sclerosis (MS) is triggered by self-antigens presented via MHC molecules that activate Th2 cells. These cells migrate to the central nervous system (CNS), where they are further activated by APCs, leading to the recruitment of additional immune cells and the initiation of an inflammatory response that damages neural tissue. CARαMOG-Foxp3 Tregs can suppress T cell activity, reduce cytokine levels, and alleviate inflammation. 2, Neuromyelitis optica spectrum disorder (NMOSD) is caused by plasmablasts that produce AQP4-IgG antibodies. These antibodies can cross the blood-brain barrier and mediate complement-dependent cellular damage within the CNS. Accumulation of neutrophils and eosinophils further contributes to the inflammatory response and demyelination. Anti-BCMA CAR-T cells and proliferating cytotoxic-like CD8⁺ CAR-T cells have been shown to effectively eliminate abnormally expanded plasmablasts and plasma cells in the cerebrospinal fluid. 3, Autoimmune encephalitis is triggered by antigens derived from virus-infected neurons or apoptotic tumor cells. This leads to a cascade that activates B cells, resulting in the production of autoantibodies. These antibodies (primarily targeting GABA, NMDAR, or LGI1) bind to neurons and cause neural injury. Currently, NMDAR-specific chimeric autoantibody receptor T cells (NMDAR CAAR-T cells) are being developed to selectively recognize and eliminate B cells and plasma cells that produce anti-NMDAR antibodies. GABA γ-aminobutyric acid, NMDAR N-methyl-D-aspartate-receptor, LGI1 leucine-rich gliom-nactivated-1CAR-Treg therapy represents a promising approach for multiple sclerosis, with MOG-specific CAR-Tregs demonstrating targeted migration to CNS regions and suppression of pro-inflammatory cytokines in EAE models. However, significant challenges require resolution before clinical translation. These include limited trafficking efficiency to intended sites with potential off-target organ accumulation, difficulties in reliable CNS delivery despite exploration of olfactory pathways, and safety concerns regarding potential suppression of beneficial immune responses and neuronal survival. Current tracking methods remain constrained to short-term observation. Future development must address these delivery limitations and conduct thorough long-term safety evaluations to realize the therapeutic potential of this approach.Neuromyelitis optica spectrum disorder (NMOSD)Neuromyelitis optica spectrum disorder (NMOSD) is a variant of MS characterized by severe, recurrent episodes and significant residual effects.302 Aquaporin-4-IgG antibodies (AQP4-IgG) are pathogenic in NMOSD, where plasma blasts producing AQP4-IgG and the antibodies themselves can breach the blood-brain barrier. This leads to complement-dependent cytotoxicity, immune cell chemotaxis, and various pathological outcomes. Recent clinical investigations of CT103A, a novel BCMA-targeting CAR-T therapy, have shown promising safety and efficacy profiles in AQP4-IgG-seropositive NMOSD patients.303Study had revealed that abnormal expansion of B cell lineages, including plasma blasts, exist in both blood and cerebrospinal fluid (CSF) of NMOSD patients. Notably, approximately 13% of CSF B cells shared Ig-VH (immunoglobulin heavy chain variable region) sequences with those in blood samples, suggesting a central nervous system (CNS) origin. Single-cell transcriptomics and TCR/B-cell receptor (BCR) sequencing further characterized the clonal dynamics and transcriptional profiles of anti-BCMA CAR-T cells, highlighting the key role of proliferating cytotoxic-like CD8+ CAR-T cell clones in autoimmunity. These cells were derived mainly from circulating T cell subpopulations in the infusion product. Interestingly, CD8+ CAR-T cells showed increased transformation into NK-like T cells with early dysfunction rather than exhaustion. The study also demonstrated that CAR-T cells expressing CXCR3 and enriched for chemotactic gene programs could effectively cross the blood-CSF barrier and infiltrate CSF. These infiltrating CAR-T cells efficiently eliminated abnormally expanded plasmablasts and plasma cells in CSF, leading to reduced soluble BCMA levels and suppressed neuroinflammation. Moreover, the study observed that CD4+ CAR-T cells expressing CD44 and lower levels of L-selectin before infusion were associated with prolonged persistence, suggesting that an early memory phenotype may contribute to sustained CAR-T cell activity.304 Whether the cell can cross the blood-brain barrier, whether it can achieve effective concentration in the central nervous system, and whether its clearance of target cells is precise all require further investigation. Additionally, the extent of central nervous system damage caused by CRS (cytokine release syndrome) and how to monitor it remain unresolved issues.305 These findings highlight both the therapeutic potential and remaining knowledge gaps in applying CAR-T cell therapy for NMOSD, particularly regarding CNS-specific pharmacokinetics and safety profiling. The demonstrated ability of engineered T cells to traffic into CSF and modulate autoimmune B cell populations offers a promising foundation for further development, while underscoring the need for additional mechanistic and clinical investigations (Fig. 4).Autoimmune encephalitis (AE)Autoimmune encephalitis (AE) is a form of encephalitis driven by autoimmune mechanisms. Since the identification of anti-N-methyl-D-aspartate receptor (NMDAR) encephalitis in 2007, numerous autoantibodies targeting neuronal cell-surface or synaptic proteins have been discovered.306 AE accounts for approximately 10–20% of all encephalitis cases, with anti-NMDAR encephalitis being the most prevalent subtype, representing 54–80% of AE cases.307 Other notable forms include encephalitis associated with antibodies against leucine-rich glioma-inactivated protein 1 (LGI1) and γ-aminobutyric acid type B receptor (GABABR).308,309,310 Brain tissue biopsies from patients with NMDAR AE have revealed significant compartmentalized enrichment of CD20+ B cells and CD138+ plasma cells, suggesting that autoantibodies found in the cerebrospinal fluid (CSF) may be produced by activated immune cells infiltrating the central nervous system.311,312,313 Reincke et al. developed NMDAR-specific chimeric autoantibody receptor T cells (NMDAR CAAR-T), which consist of an NMDAR autoantigen fragment fused with a CD8 hinge and 41BB/CD3ζ intracellular signaling domains. In vitro, NMDAR CAAR-T cells were activated and expanded using K562 and Nalm6 cell lines co-expressing human monoclonal antibodies and luciferase (ffluc) as target cells. Upon activation, these CAAR-T cells secreted interferon-γ and granzyme B, leading to the selective lysis of target cells. The therapeutic efficacy of NMDAR CAAR-T cells was evaluated using ffluc as a reporter. In vivo, immunodeficient mice lacking natural killer cells and lymphocytes were co-injected with NMDAR antibodies and Nalm6 cells, followed by treatment with NMDAR CAAR-T cells. This resulted in a reduction of Nalm6 cells and elimination of the pathogenic antibodies, without any observed off-target toxicity or adverse effects.314 Current related experiments remain at the preclinical stage, and further investigation is required to determine their potential applicability. Moreover, there is a lack of experimental evidence regarding their extension to other antibody-based AE therapies (Fig. 4).Infectious diseasesSeveral persistent infectious agents employ sophisticated strategies to evade host immunity and establish chronic infections (Fig. 5). HIV primarily targets CD4+ T cells, integrating its genome into host DNA and escaping detection by dendritic cells, macrophages, and cytotoxic T cells, leading to progressive immune dysfunction.315 Similarly, HBV exploits the NTCP receptor on hepatocytes to deliver its genome, forming stable cccDNA that enables persistent viral replication and propagation.316,317,318 Herpesviruses demonstrate remarkable latency strategies: EBV maintains lifelong infection in B lymphocytes through episomal persistence and expression of EBNA proteins that promote immune evasion and oncogenesis,319,320,321,322,323,324 while HCMV incorporates viral proteins into extracellular vesicles to trigger autoimmunity and multi-organ damage.325,326,327 Unlike these viral pathogens, Aspergillus fungi invade immunocompromised hosts through airborne spores that germinate into invasive hyphae, overcoming respiratory defenses and releasing tissue-damaging toxins.328,329 Common to all these infections is their ability to subvert normal immune surveillance—whether through viral latency programs (HIV, HBV, EBV, HCMV) or structural adaptations (Aspergillus)—creating persistent reservoirs that challenge conventional therapies. This shared pathogenesis highlights the need for innovative approaches like CAR-T cells that can recognize and eliminate these evasive pathogens through engineered immune recognition.Fig. 5Full size imageMechanisms of infectious diseases. 1, After HIV infection, the virus is presented by dendritic cells (DCs) to CD4⁺ T cells, activating them while inserting fragments of viral genetic material into the host DNA. These infected cells become unrecognizable to the immune system and lose normal function, while continuing to produce and release new viruses. D.66.α CD4-CAR T cells and bNAbs (NKG2D receptor)-CAR-T cells can target and bind to HIV-infected CD4⁺ T cells. 2, Hepatitis B Virus (HBV) binds to NTCP receptors on the surface of hepatocytes, enters the cells via endocytosis, and uncoats to deliver its nucleic acids into the nucleus. There, the viral genome integrates into the host DNA, forming covalently closed circular DNA (cccDNA), which serves as a template for viral gene expression and replication. Newly assembled viruses are released to infect other hepatocytes. S-CAR-T cells and L-CAR-T cells can recognize and eliminate HBV-infected hepatocytes. 3, Epstein-Barr Virus (EBV) enters the body through the skin and infects B cells. Through cell-to-cell transmission, the virus spreads and reprograms B cells, resulting in immune dysregulation. CAR-T cells targeting CD30, CD7, and CD19 can identify and kill EBV-infected B cells. 4, Human Cytomegalovirus (HCMV) integrates viral proteins into extracellular vesicles of host cells. The virus causes damage to target organs by generating autoantibodies, recruiting immune cells, and injuring vascular endothelial cells. HCMV-gB CAR-T cells specifically recognize glycoprotein gB on the surface of HCMV and eliminate infected cells. In immunocompromised or immunodeficient patients, Aspergillus spores may evade immune recognition and develop into invasive hyphae that penetrate lung tissue, releasing toxins and enzymes that damage the lungs. Dectin-1 CAR-T cells recognize β-glucan on the surface of Aspergillus. Aspergillus-specific CAR (Af-CAR) T cells specifically target conserved antigens on Aspergillus, while CD8⁺ and CD4⁺ Af-CAR-T cells also recognize these conserved fungal antigens and help eliminate the infection. NTCP sodium taurocholate co-transporting polypeptide, CTLs cytotoxic T lymphocyte, rcDNA relaxed circular DNA, EV extracellular vesiclesRecent advances in CAR-T cell therapy have shown remarkable potential for treating persistent viral and fungal infections through innovative engineering approaches. For HIV infection, two primary strategies have emerged: CD4-based CAR-T cells utilizing truncated D1D2 domains to maintain HIV recognition while avoiding MHC class II interactions,330,331,332,333,334,335,336,337 and convertible CAR designs combining broadly neutralizing antibodies (bNAbs) with NKG2D receptors for selective targeting of infected cells.330,331,338 Structural optimizations, including LYC motif incorporation and hinge region modifications, have improved surface expression and antigen recognition.339 However, clinical translation faces challenges including variable viral reservoir targeting, CAR-T cell persistence, and tissue homing efficiency.330,340,341,342,343,344,345,346In HBV, researchers have developed L-CAR-T and S-CAR-T cells targeting envelope proteins, demonstrating dual functionality through direct hepatocyte killing and IFN-γ/IL-2 secretion to combat cccDNA persistence.347 The S-CAR design shows particular promise with its conserved S-domain recognition, inducing rapid viral clearance in mouse models while exhibiting favorable liver tropism.348 Though only one clinical trial (NCT06251115) is currently evaluating HBV-specific CAR-T cells, highlighting the need to address viral escape and on-target/off-tumor effects. EBV-directed CAR-T therapies have primarily focused on malignancies, with CD19, CD30, and CD7-targeting constructs in clinical trials,349,350 while non-malignant EBV infections remain understudied. HCMV-specific CARs targeting glycoprotein complexes (particularly gB) demonstrate cytokine-mediated antiviral activity despite viral evasion mechanisms like UL37x1-mediated inhibition of cytotoxicity.351,352,353 Costimulatory domain incorporation (e.g., 4-1BB) has enhanced functionality354,355,356,357 though GVHD risks require careful evaluation.For invasive aspergillosis, CAR-T strategies have evolved from broad-spectrum Dectin-1 targeting50 to more specific AB90-E8-directed designs that show superior fungal clearance through perforin/granzyme release and macrophage activation.358 However, unidentified antigen identity and manufacturing delays for acute infections remain substantial barriers.While these CAR-T approaches demonstrate impressive preclinical results across HIV,359 HBV,347 EBV,360 HCMV,48 and Aspergillus50 infections, several shared limitations must be addressed: antigenic variability and escape, tissue-specific trafficking and persistence, manufacturing complexities for acute infections, safety concerns including cytokine release and on-target/off-tumor effects, limited clinical trial activity, and some trials have only shown effectiveness in animal models. The field now requires optimized CAR designs combining multiple effector functions, improved delivery strategies for viral reservoirs, and accelerated manufacturing platforms to realize the full therapeutic potential against these challenging infections.CAR-T cell therapy demonstrates significant potential against persistent infections through pathogen-specific engineering, including CD4-based CARs for HIV, envelope-targeting constructs for HBV, glycoprotein-directed approaches for herpesviruses, and Dectin-1/AB90-E8 CARs for Aspergillus. However, these developments face substantial challenges: antigenic variability, limited tissue homing to reservoir sites, manufacturing complexities for acute infections, and safety concerns including on-target/off-tumor effects. Current clinical evaluation remains limited, with only one active HBV CAR-T trial (NCT06251115) identified. Future progress requires optimized CAR designs integrating multiple effector functions, enhanced delivery strategies to viral reservoirs, and improved manufacturing platforms to address these persistent infections effectively.Aging and senescenceCellular senescence-associated diseasesCellular aging emerges from sustained cell cycle arrest triggered by diverse stressors including replicative stress, oxidative damage, mitochondrial dysfunction, radiation exposure, chemotherapy, and traumatic injury. These insults activate two primary pathways: (1) the INK4/ARF locus, where ARF inhibits MDM2 to stabilize p53, while p16INK4a blocks CDK4/6-mediated Rb phosphorylation, maintaining E2F binding and G1 arrest; and (2) DNA damage responses, where ATM/ATR kinases activate p53 to induce p21, subsequently inhibiting Cdk2-mediated pRb phosphorylation and similarly enforcing G1 arrest.361 Senescent cells subsequently develop the senescence-associated secretory phenotype (SASP), releasing cytokines, chemokines, and matrix metalloproteinases that drive pathology through multiple mechanisms: immune cell recruitment (macrophages, dendritic cells, neutrophils), chronic inflammation, senescence reinforcement, paracrine senescence induction, and cancer progression.362 Age-related accumulation of senescent cells results from progressive tissue damage and declining immune surveillance, contributing to organismal aging.363,364 Genetic ablation studies demonstrate that selective removal of senescent cells can ameliorate age-related metabolic dysfunction and physical decline.127,365,366,367 While senolytic drugs partially replicate these benefits, their mechanisms often remain unclear and require continuous administration.36,368Recent advances identify urokinase-type plasminogen activator receptor (uPAR) as a promising target, with uPAR-positive cells increasing proportionally with age.36 Engineered uPAR CAR-T cells (m.uPAR-h.28z) incorporating anti-muPAR scFv with hCD28/CD3ζ signaling domains demonstrate remarkable properties368: specific elimination of senescent cells in juvenile and aged mice without tissue toxicity, improved glucose homeostasis and metabolic function in both physiological aging and high-fat diet models, long-term persistence (> 15 months) following single low-dose administration, prevention of age- and diet-induced metabolic decline when administered prophylactically, enhanced pancreatic β-cell function and insulin sensitivity in metabolic syndrome models, and reduced senescent cell burden across multiple tissues (pancreas, liver, adipose). These findings corroborate earlier work using GPNMB vaccination against senescent cells.369 Notably, uPAR CAR-T cells show particularly strong activity in aged mouse livers and spleens, with protective effects lasting over half the murine lifespan.368 The durability of response after single-dose administration contrasts favorably with conventional senolytics. The field faces significant challenges in target identification due to senescent cell heterogeneity, making universal surface antigens unlikely.370,371 However, this diversity may enable precision targeting of specific senescent subpopulations. Alternative targets under investigation include NKG2D ligands (MICA, MICB, ULBP1-5), though their broad expression patterns present manufacturing and toxicity control challenges.372,373,374 These developments highlight the potential of immunotherapeutic approaches to address fundamental aging mechanisms while underscoring the need for continued target discovery and validation (Fig. 6).Fig. 6Full size imageMechanisms of aging and senescence. Various damaging factors can lead to DNA damage or activation of the INK4/ARF gene, resulting in the upregulation of downstream effectors such as p16INK4a, p53, and p21. This process weakens the inhibitory effect of MDM2, leading to increased binding between RB and E2F, ultimately causing cell cycle arrest. Arrested cells develop a senescence-associated secretory phenotype (SASP), which includes cytokines, chemokines, and matrix metalloproteinases. These factors contribute to pathogenesis by recruiting immune cells, triggering inflammation, reinforcing senescence in an autocrine and paracrine manner, and potentially promoting cancer progression. uPAR CAR-T cells, which combine a human intracellular co-stimulatory domain with a murine extracellular antigen-recognition receptor, can specifically recognize and eliminate senescent cellsuPAR-targeting CAR-T cells represent a significant advancement in senolytic therapy, demonstrating specific elimination of senescent cells across multiple tissues with durable effects from single-dose administration. These cells improve metabolic function and extend healthspan in aging models, outperforming conventional senolytics through their persistent activity. However, the field faces the fundamental challenge of senescent cell heterogeneity, which complicates universal target identification. While uPAR shows promise, alternative targets like NKG2D ligands present manufacturing and toxicity concerns. Future progress depends on discovering additional surface markers specific to senescent subpopulations and validating their therapeutic potential, balancing precision with practicality in clinical translation.Cardiovascular and metabolic disordersCardiac fibrosisCardiac fibrosis, a key pathological feature of heart failure, develops through two distinct patterns: reparative fibrosis forming organized scar tissue post-MI to stabilize necrotic areas, and reactive fibrosis characterized by interstitial ECM deposition in non-ischemic cardiomyopathy or surviving myocardium, disrupting tissue architecture.375,376,377,378,379,380 The process is driven by mast cell-derived mediators activating fibroblasts, leading to excessive ECM accumulation that impairs contractility and promotes arrhythmias.378,379,380,381,382,383,384,385 While initially adaptive (e.g., post-MI scar stabilization), fibrosis becomes maladaptive, though evidence suggests reversibility particularly in HFpEF.386,387 Therapeutic strategies focus on targeting ECM degradation, myofibroblast elimination, and fibroblast activation pathways.28,388,389,390In mouse models of hypertensive heart injury and fibrosis, researchers designed CAR-T cells targeting fibroblast activation protein (FAP) to specifically eliminate myofibroblasts. Adoptive transfer of FAP CAR-T cells significantly reduced cardiac fibrosis and restored cardiac function, with effects observed exclusively in the heart.28,84,391,392,393,394,395,396 These findings build upon existing angiotensin pathway inhibitors (ACEi/ARBs) that indirectly modulate fibroblast activity,388 while offering potential advantages through direct, targeted myofibroblast elimination. The success of these approaches hinges on the dynamic nature of cardiac fibrosis and the abundance of targetable myofibroblasts in fibrotic hearts,28 which first be identified as therapeutic targets in foundational preclinical work (Fig. 7).Fig. 7Full size imageMechanisms of cardiovascular and metabolic disorders. 1, Cardiac fibrosis occurs following tissue injury, during which the number of mast cells increases. These cells undergo degranulation and release various fibrotic mediators that activate, differentiate, and promote the proliferation of fibroblasts. This leads to excessive local collagen deposition, resulting in impaired contractile function or arrhythmia. FAP-CAR-T cells can recognize fibroblast surface antigens and exert a fibroblast-clearing effect. 2, Atherosclerosis is characterized by chronic inflammation and localized autoimmune responses that cause lipid accumulation in blood vessels. These deposits further trigger recurring vascular inflammation. uPAR-CAR-T cells can recognize and eliminate damaged vascular endothelial cells, preventing downstream inflammatory reactions. CAR-Tregs can suppress local immune responses, thereby reducing inflammation and autoimmunity in affected vascular regions. 3, Type 1 diabetes is caused by defective development of FOXP3⁺ Tregs, leading to impaired immune tolerance. As a result, pancreatic β-cells are targeted and destroyed by the immune system, resulting in insulin deficiency. The lack of immune suppression allows infection or stress-related inflammatory factors to further activate T and B cells, amplifying the autoimmune attack on β-cells. Universal CD7 CAR-T cells and 287 CAR CD8⁺ T cells can target and eliminate autoreactive immune cells, thereby reducing the autoimmune responseAtherosclerosisAtherosclerosis develops through a complex interplay of chronic inflammation and autoimmune responses. The pathogenic cascade initiates with intimal injury, triggering recruitment of immune cells (monocytes/macrophages, neutrophils, and lymphocytes) into the vascular wall. This infiltration promotes lipid accumulation and sustains vascular inflammation, driving disease progression.397 While most recruited immune cells exacerbate atherosclerosis, CD4+ regulatory T cells (Tregs) serve a protective function by maintaining immune homeostasis through suppression of pro-inflammatory responses. Despite their relatively low abundance within atherosclerotic plaques, Tregs play a crucial role in preserving immune tolerance.398,399,400There are currently two proposed solutions. One is uPAR CAR-T, where endothelial cell-induced chronic inflammation itself triggers a series of inflammatory responses due to subsequent degenerative changes. By eliminating the abnormally displayed endothelial cells, it can reduce the immune response triggered.401 The other is regulatory CD4+ Tregs. Engineered Tregs: These counteract hyperlipidemia-induced endothelial damage by suppressing autoimmune responses and preserving vascular endothelial function.402 However, significant challenges remain in translating these approaches. Infusion of polyclonally expanded Treg cells with unknown antigen specificity cannot effectively suppress target cells and inhibit undesirable immune responses and may lead to unwanted side effects such as systemic immunosuppression and reactivation of latent infections. Therefore, the use of polyclonal infusions in clinical trials is often the main reason for the low efficacy of the adopted therapies. CAR Treg cell infusion may induce so-called cytokine storm and neurotoxicity.402 Furthermore, the field faces ongoing difficulties in target identification and engineering approaches suitable for precise immune modulation.402 These limitations highlight the need for continued refinement of cellular therapies to achieve selective immunomodulation without compromising protective immune functions in atherosclerosis (Fig. 7).Type 1 diabetesType 1 diabetes mellitus (T1D) develops through autoimmune destruction of pancreatic β-cells, primarily driven by excessive activation of effector T cells within pancreatic islets.11 This process involves multiple pathological mechanisms: (1) deficient regulatory T cell (Treg) activity fails to control autoreactive T cell expansion against islet autoantigens (insulin, ZnT8, GAD65); (2) impaired thymic selection or defective FOXP3+ Treg development leads to loss of self-tolerance; and (3) β-cell-specific T cells migrate to pancreatic lymph nodes where they encounter antigen-presenting cells (APCs) displaying pancreatic antigens, subsequently differentiating into inflammatory effector T cells that infiltrate islets and mediate β-cell destruction.403,404,405,406 Current treatments involve CD7 CAR-T (universal CAR-T), with at least one positive islet autoantibody (e.g., GADA, IA-2A, IAA, ZnT8). In 2021, a study demonstrated that CAR-T could restore normal insulin secretion in mice. In 2022, CAR-Tregs were found to reverse Type 1 diabetes. Major histocompatibility complex (MHC) II gene polymorphisms, especially HLA-DQ, are crucial in Type 1 diabetes development.407,408 The disease’s strong genetic component involves MHC II polymorphisms, particularly HLA-DQ variants. In non-obese diabetic (NOD) mice, the insulin B chain peptide 9-23 serves as a critical autoantigen when presented by MHCII (I-Ag7), activating pathogenic T cells.409 A proof-of-concept study engineered CAR-T cells targeting I-A g7-B:9-23, evaluating the cytotoxicity of CAR-engineered CD8+ T cells against APCs presenting I-A g7-B:9-23. Results showed 287-CAR-T CD8+ T cells were specific in detecting and killing antigen-presenting cells (APCs) presenting I-A g7-B:9-23, with higher proliferation and homing rates in pancreatic lymph nodes compared to spleens and inguinal lymph nodes.47 Although a single infusion of CAR-T CD8+ cells did not achieve complete cure, disease onset was delayed, indicating reversible protection.47In these studies, allogeneic infusion of CAR-T cells can induce GvHD, especially chronic GvHD that persists over the long term. Despite these advances, significant challenges remain. Further research is needed in the design of CAR-T therapies.410 Additionally, the issue of Treg persistence after in vivo infusion also needs to be addressed.411 Moreover, the balance between effective autoimmunity suppression and general immune competence requires precise modulation. These findings highlight both the promise and complexities of cellular therapies for T1D, where successful translation will require solutions to persistence, specificity, and safety challenges while leveraging the unique capacity of engineered immune cells to target autoimmune processes at their source (Fig. 7).CAR-T therapies demonstrate promising applications across cardiovascular and metabolic diseases. FAP-targeted CAR-T cells effectively reduce cardiac fibrosis in murine models, while uPAR CAR-T and engineered Tregs offer potential approaches for atherosclerosis. In type 1 diabetes, CAR-T cells targeting specific MHC-peptide complexes show capability to delay autoimmune destruction. However, these approaches face significant challenges including target identification precision, potential for graft-versus-host disease, limited persistence of therapeutic cells, and the critical balance between effective autoimmunity suppression and preserved immune function. The field requires continued refinement of CAR designs and thorough investigation of long-term safety profiles before clinical translation can be realized.Clinical translation and trialsThe latest clinical trial results and registered clinically relevant experiments and advancements.Although no clinical drugs for non-tumor diseases have been approved yet, related clinical trials are underway, most of which are in Phase I or pre-Phase I stages (Table 1). Several of these experiments have yielded significant breakthroughs, for example, the allogeneic universal CD19 CAR-T cell therapy developed by Xu’s team was administered to three patients with severe autoimmune diseases.412 Specifically, an SRP-IMNM patient who received TyU19 cell therapy did not experience any fever or cytokine storm symptoms. Additionally, the patient’s Total Improvement Score (TIS) rapidly increased from a baseline of 72.5 to 100 within two months and remained stable during follow-up. Imaging and pathological examinations also showed a significant reduction in muscle inflammation. Two diffuse cutaneous systemic sclerosis (dcSSc) patients who received TyU19 cell therapy showed a significant increase in CRISS scores within a few months. Regarding skin fibrosis, the modified Rodnan skin scores (mRSS) of both patients decreased significantly.413 Similarly, Yang et al. evaluated the safety and efficacy of allogeneic CD19-targeted CAR-T cells (TyU19) in refractory SLE. The study enrolled four young female patients (aged 22–24) with treatment-refractory lupus. All four patients demonstrated sustained clinical improvement in signs and symptoms. At the 3-month follow-up, all met the Systemic Lupus International Collaborating Clinics (SLICC) criteria for sustained remission (SRI-4). Notably, peripheral BCMA+ and CD19−BCMA+ plasma cells declined in all patients, suggesting that TyU19 infusion may facilitate plasma cell reconstitution. Consistent with observations in autologous CAR-T therapies, most reconstituted B cells were naive B cells, with marked reductions in memory B cells and plasma cells. One patient who discontinued post-intervention immunosuppressants maintained durable drug-free remission, highlighting the potential advantages of allogeneic CAR-T cells as a promising therapy for refractory SLE (NCT05988216).414Table 1 Ongoing clinical trials for non-oncological diseasesFull size tableCD19 CAR-T therapy has shown promise in autoimmune diseases, particularly systemic lupus erythematosus (SLE). Some case reports have indicated that patients experienced symptom relief without significant adverse reactions. This includes remission of SLE according to DORIS criteria and improvements in laboratory parameters, such as seroconversion of anti-double-stranded DNA antibodies.114 Additionally, low C3 and C4 levels normalized, and the Systemic Lupus Erythematosus Disease Activity Index score (with the SELENA [Safety of Estrogens in Lupus National Assessment] modification) decreased from 16 at baseline to 0 at follow-up.150Challenges and risk-benefit analysis of CAR-T cell therapy in non-oncological diseasesSafetyCytokine release syndrome (CRS): After CAR-T cells are infused into the body, they recognize and activate against the body’s tumor cells, releasing large amounts of cytokines. This further activates a large number of immune cells (T cells, B cells, macrophages, monocytes, etc.) and rapidly proliferates, causing an excessive cascade release of cytokines such as interleukin (IL)-1, IL-6, tumor necrosis factor-alpha (TNF-α), and granulocyte-macrophage colony-stimulating factor (GM-CSF), forming a “cytokine storm”.415,416,417 CRS typically occurs within hours to 14 days after CAR-T cell infusion, with diverse clinical manifestations that can lead to shock, multi-organ dysfunction, and even death, underscoring the importance of early recognition and management of CRS.418,419 CRS is less common in non-tumor diseases because the target cell load is much lower than in tumor diseases. Therefore, the application prospects are better compared to tumor diseases, although CRS is not completely absent.114,177,218 In clinical trials targeting IIM patients, it has been mentioned that during the first three days, patients showed transient increases in creatine kinase and inflammation-related markers.218 Although these markers later returned to normal levels, reaching treatment standards, effective management of CRS is still necessary. Some trials also observed moderate CRS in non-tumor diseases, which exacerbated pre-existing respiratory difficulties.420 Therefore, CRS still requires attention in patients with underlying conditions.L-CRS and LICATS: B-cell non-Hodgkin lymphoma (B-NHL) exhibits distinct pathophysiological features and clinical manifestations compared to other hematologic malignancies such as B-cell acute lymphoblastic leukemia (B-ALL). A notable difference is that B-NHL lesions are typically localized. In B-NHL patients undergoing CAR T-cell therapy, compartmental inflammation can be observed, presenting as redness, swelling, and enlargement around the local lymphoma or lesions. This localized inflammatory response is defined as local CRS (L-CRS).421 The activation of monocytes/macrophages and the progression of CRS are likely associated with the in vivo distribution of CAR T cells, and the aggregation of these cells triggered by direct contact activation is a key pathogenic mechanism. Recently, a study comprehensively analyzed existing data on CAR-T therapy-specific adverse events in autoimmune diseases, documenting 54 cases of localized reactions termed Local Immune effector Cell-Associated Toxicity Syndrome (LICATS). LICATS is observed in autoimmune disease patients receiving CD19-targeted CAR T-cell therapy, most likely stemming from the clearance of immune cells from affected organs. It is characterized by self-limiting, organ-specific manifestations that are typically mild in intensity. LICATS reflects differences in toxicity profiles between CAR T-cell therapy for B-cell malignancies and autoimmune diseases. Given the high burden of target cells in lymphoma—which is associated with CRS and ICANS—and the significantly greater tumor cell load compared to the B-cell burden in autoimmune diseases, these CAR-driven inflammatory syndromes are more prevalent and severe in B-cell malignancies than in autoimmune diseases. Consequently, higher-grade CRS and ICANS are uncommon in autoimmune disorders. These self-limiting inflammatory events may arise from the direct effects of CAR T cells, localized cytokine release following target cell killing, or the activation of autoreactive T cells that are reinfused as non-engineered T cells and expand alongside CAR T cells in the patient. Alternatively, they may result from repopulating immune cells after lymphodepletion, including but not limited to the effects of fludarabine on myeloid cells and regulatory T cells. However, given the localized nature of LICATS, its pathogenesis could also involve inflammatory responses related to the clearance of dead B cells and debris in affected tissues. LICATS represents an immune reset-mediated toxicity that is organ-specific, typically emerging days to weeks after CAR T-cell therapy, and most importantly, it is self-limiting and serves as a significant indicator of actual disease recurrence. Unlike CRS, LICATS is likely driven by immune-related clearance processes in tissues previously affected by autoimmune disease prior to CAR T-cell therapy.422 The underlying mechanisms of L-CRS and LICATS remain unclear, and further studies with larger sample sizes are needed to advance understanding in this area.Immune effector cell-associated neurotoxicity syndrome (ICANS): ICANS manifests clinically as toxic encephalopathy, and in severe cases, it can progress to seizures, cerebral edema, motor dysfunction, altered consciousness, and coma.415,423 It is a disorder characterized by a pathologic process involving the central nervous system following any immune therapy that results in the activation or engagement of endogenous or infused T cells and/or other immune effector cells. Symptoms or signs can be progressive and may include aphasia, altered level of consciousness, impairment of cognitive skills, motor weakness, seizures, and cerebral edema.415 ICANS is commonly observed in tumor diseases due to the target cell load being much lower than in tumor diseases. It is important to note that in non-tumor diseases, particularly neurological disorders, the pathogenesis involves pathogenic cells and antibodies crossing the blood-brain barrier to reach the central nervous system, triggering an inflammatory response and causing cellular dysfunction.113,302,424 Therefore, when designing CAR-T therapy, consideration is given to the ability to cross the blood-brain barrier and generate related immune responses in the central nervous system,304 which theoretically could lead to ICANS. Although there are currently no clinical trials targeting the nervous system, this remains an important factor to consider.Differences in peak expansion: compared to the robust expansion observed in hematologic malignancies, CAR-T cell expansion in non-oncologic indications typically exhibits a more moderate peak. This attenuated response can be attributed to several factors: lower antigen burden: in non-oncologic diseases (e.g., systemic lupus erythematosus, systemic sclerosis), target antigens (such as CD19 or BCMA) are expressed at lower levels than on tumor cells, reducing CAR-T cell activation.176 Distinct immune microenvironment: Non-oncologic tissues lack the strong co-stimulatory signals (e.g., CD28/CD80 pathway) characteristic of tumor microenvironments, limiting CAR-T cell activation.425Cardiovascular toxicity: Cardiovascular toxicity is not uncommon in patients receiving CAR-T cell therapy. A recent pharmacovigilance study based on the FDA Adverse Event Reporting System reported 2.8% arrhythmias, 2.6% cardiomyopathy, 1.8% cardiogenic shock, 1.7% pleural disease, 0.4% pericardial disease, and 1.6% venous thromboembolism events.426 Serious events such as myocardial infarction, atrial arrhythmias, congestive heart failure, and cardiac arrest are generally rare (