Introduction: autoimmunity and current unmet needsAutoimmune diseases constitute a substantial and growing global clinical burden1. Despite notable therapeutic advances over recent decades, clinical management of these conditions continues to rely largely on well-established strategies, including corticosteroids, conventional disease-modifying anti-rheumatic drugs, biologics, and more recently, immune cell-depleting approaches adapted from CAR T cancer immunotherapy2,3,4. While these interventions can provide symptomatic improvement and disease control, a considerable proportion of patients experience incomplete responses, disease relapse, or treatment-limiting toxicities5. In addition, prolonged immune suppression is frequently associated with impaired host defence and an increased risk of infections and malignancies6.At a conceptual level, many current therapeutic strategies are grounded in a reductionist paradigm, targeting individual cytokines, signalling pathways, or discrete immune cell populations implicated in disease pathogenesis. Such approaches have demonstrated substantial clinical benefit, particularly in settings where a limited number of well-defined mechanisms dominate disease activity. However, systemic autoimmune diseases are intrinsically multifactorial, involving extensive functional redundancy, compensatory pathways, and dynamic interactions among immune and tissue-resident cell populations7,8. As a result, interventions directed against single components of this high-dimensional network may result in incomplete or transient efficacy, as alternative pathogenic processes may persist or emerge6.Taken together, these observations highlight opportunities to complement established therapeutic strategies with experimental and conceptual frameworks capable of resolving immune organisation at the systems level. Recent advances in high-dimensional immune profiling technologies now enable more comprehensive and system-level interrogation of immune inflammation and tolerance9. Here, we discuss how such approaches can refine and extend current understanding of immune pathology and provide a basis for complementing and informing therapeutic strategies beyond single-pathway intervention (Fig. 1).Fig. 1: Restoring high-dimensional immune architecture in autoimmune disease.Full size imageConceptual model illustrating the transition from a healthy immunome to a dysregulated autoimmune state and a potential pathway towards immune restoration. Autoimmune disease is associated with distortion of immune architecture and the emergence of pathogenic network features. Therapeutic interventions can be considered within a Reset-Rebuild framework. Mechanism-based therapies may provide rapid control of dominant inflammatory pathways or pathogenic cell populations and can contribute to an initial immune reset. High-dimensional immune profiling, together with the Architecture-Anomaly-Target-Lead (AATL) framework, may then be used to monitor immune reconstitution, identify residual or emerging network abnormalities, and inform tolerance-oriented interventions aimed at reinforcing regulatory balance. These approaches are presented as complementary components of a broader therapeutic framework rather than mutually exclusive strategies. Points represent functional cell subsets, lines represent mechanistic interactions, and distance between points reflects their relative positioning within the immunome. Created in part with BioRender.com.In health, the immune system functions as a balanced, interconnected network maintained by regulatory-effector feedback loops that preserve tolerance. In autoimmunity, this network becomes rewired, with loss of regulation and emergence of dominant pro-inflammatory hubs. Current therapeutic strategies often target individual molecules and pathways, reducing inflammatory activity, but not fully restoring broader network stability or regulatory balance. High-dimensional immunomics enables mapping of immune architecture and identification of dysregulated modules, providing a framework to better understand how interventions impact immune organisation and guiding tolerance-based strategies aimed at re-establishing more stable immune homoeostasis (Box 1).Box 1 Conceptual and operational framework for architecture-aware immunotherapySeveral conceptual constructs introduced here organise our argument and are referenced in the relevant sections. The definitions below provide both conceptual meaning and how each construct can be measured.Defining the autoimmunome: organisation of immune architecture in autoimmunityImmune architecture refers to the measurable organisation of immune function across interacting dimensions, including cellular composition, clonal structure, regulatory-effector coupling, spatial localisation and intercellular communication. Within this framework, the autoimmunome is the disease-specific distortion of this architecture in autoimmunity. Systematic efforts to map immune “wiring” show that immune function is distributed across structured intercellular networks rather than governed by linear pathways10,11,12.A defining property of the autoimmunome is compositional heterogeneity. Autoreactivity is typically polyclonal and poly-epitopic, involving diverse TCR and BCR lineages that vary across individuals, disease stages, and tissue sites13,14. In rheumatoid arthritis (RA), single-cell profiling of synovial tissue has revealed inflammatory and regulatory cellular states that extend beyond conventional subsets and stratify patients into distinct “endotypes”15. Integrating single-cell transcriptomics with immune repertoire sequencing further shows that clonally related lymphocytes can be tracked between blood and synovium yet occupy divergent functional states depending on compartment16.A second core feature is functional plasticity, whereby immune cells occupy context-dependent and reversible states17,18. Canonical lineage definitions (e.g., Th17 versus Treg) are less distinct in human inflammatory tissues than in murine models19,20. In autoimmune arthritis, inflammatory environments can drive hybrid or transitioning phenotypes (such as Th17/1 states), emphasising that the pathogenic unit is not a static subset but a dynamic trajectory shaped by cues in situ21. This plasticity has important therapeutic implications, as inflammatory environments may promote continued transitions between immune cell states, with the potential to limit sustained therapeutic responses.Third, the autoimmunome is spatially organised. Tissues affected by autoimmune disease are patterned into microenvironments or niches where stromal scaffolds, antigen presentation, cytokine gradients, and cell-cell interactions converge to amplify or limit inflammation22. Spatial transcriptomic studies of RA synovium reveal localised immune and stromal neighbourhoods with distinct cellular compositions and interaction circuits, rather than a uniform tissue-wide distribution23. Pathogenic and regulatory processes are localised within tissue contexts that peripheral measurements may not capture, making the periphery-tissue interface a central translational challenge22,23,24.The autoimmunome links immune cells, cytokines, checkpoints, antigens and tissue programmes through coupled feedback loops. In RA, synovial fibroblast subsets can function as tissue organisers that partition inflammation and tissue-remodelling processes, revealing non-immune nodes that are integral to the autoimmune network architecture25. This integrated view highlights that pathogenic processes are distributed across interacting cellular and tissue compartments, rather than confined to individual cell types or pathways.Together, this high-dimensional perspective exposes a translational gap. While clinical trials rely on limited longitudinal endpoints and “single-compartment” sampling, it remains challenging to determine whether an intervention reshapes immune organisation or merely modulates downstream inflammatory output.The reductionist paradigm: strengths and complementary opportunitiesA central premise of contemporary immunotherapy is that chronic immune-mediated disease can be effectively controlled by inhibiting a dominant cytokine, blocking a key signalling pathway, or depleting a pathogenic immune lineage. Such mechanism-based targeting, which we refer to here as a reductionist therapeutic paradigm, has substantially advanced clinical care, with targeted biologic therapies transforming outcomes across RA, inflammatory bowel disease, psoriasis and spondyloarthritis26. These agents provide reliable disease control during active treatment, and for many patients remission is well maintained on therapy. In some settings, however, remission is not consistently sustained once treatment is withdrawn. The HONOR study, for instance, reported that only 48% of RA patients sustained DAS28-ESR remission one year after adalimumab discontinuation, even after achieving deep remission on therapy27. Long-term therapeutic strategy is also informed by safety considerations that vary across drug classes: for example, in the ORAL Surveillance trial, tofacitinib did not meet non-inferiority criteria versus TNF inhibitors for cardiovascular and malignancy endpoints28. Together, these observations indicate that, alongside their well-established benefits, there may be opportunities in some patients to complement reductionist interventions with strategies aimed at reinforcing regulatory balance and promoting sustained, treatment-minimised disease control5.The Reset-Rebuild model, in which lymphodepletion or lineage ablation first attenuates dominant pathogenic immune compartments (a reset phase), followed by reconstitution of a more tolerant immune architecture (a rebuild phase), provides a useful conceptual frame. Lineage-depleting approaches have transformed the management of multiple autoimmune diseases, including anti-CD20-mediated B-cell depletion and CD52-targeted lymphocyte depletion with alemtuzumab. In ANCA-associated vasculitis, the RAVE trial showed that rituximab was non-inferior to cyclophosphamide for induction of remission, with complete remission off glucocorticoids achieved in 64% versus 53% of patients at 6 months, and superior efficacy in relapsing disease29. In relapsing multiple sclerosis, the phase 3 OPERA I and OPERA II trials demonstrated that ocrelizumab reduced annualised relapse rates by approximately 45% compared with interferon β-1a (0.16 versus 0.29 per year in both trials)30, and ocrelizumab subsequently became the first disease-modifying therapy approved for primary progressive multiple sclerosis31. Next-generation anti-CD20 agents such as obinutuzumab further improved complete renal response rates in active lupus nephritis in the phase 3 REGENCY trial32. Autologous haematopoietic stem cell transplantation in relapsing-remitting multiple sclerosis produced durable disease control in the MIST trial, with substantially lower progression rates than standard disease-modifying therapy over 5-year follow-up33. The durability of disease control following CD20-directed therapy varies by disease, clinical endpoint, treatment schedule, and therapeutic context29,30,31,32. Prolonged responses have been observed in some settings, whereas other indications use continued or repeated treatment to maintain disease control. Collectively, these findings underscore the substantial clinical impact of B-cell depletion while illustrating that the extent of immune reconstitution differs across therapeutic contexts.CD19-directed CAR T-cell therapies extend the reset principle further by combining lymphodepleting preconditioning with CD19 lineage ablation. Pioneering case series and multi-disease follow-up reports have documented drug-free remission in patients with refractory systemic lupus erythematosus, idiopathic inflammatory myositis, and systemic sclerosis at median follow-up of over a year3. However, outcomes appear heterogeneous across cohorts and indications. Long-lived plasma cells reside in protective bone marrow niches and may evade CD19-directed depletion34, and the relative contributions of lymphodepleting preconditioning, B-cell lineage depletion, and subsequent immune reconstitution to long-term clinical responses remain difficult to disentangle. Similarly, the extent to which broader regulatory networks are re-established following B-cell reconstitution remains incompletely characterised. These observations suggest that reset-based approaches can achieve substantial clinical benefit through profound lineage depletion and immune remodelling, while the depth and stability of longer-term immune restoration continue to be defined across diseases and treatment contexts.A useful frame for reconciling these observations is the long-established clinical paradigm of induction and maintenance therapy. In autoimmune practice, as in oncology, high-intensity reductionist interventions are often used to achieve rapid disease control, followed by a maintenance phase in which therapy is tapered to the lowest effective regimen. This pattern is well established in paediatric autoimmune disease, where high-intensity induction with intravenous corticosteroids in juvenile dermatomyositis, or cyclophosphamide in severe vasculitis, is routinely followed by transition to lower-intensity maintenance regimens. Within this frame, reductionist and architecture-aware approaches may be complementary rather than mutually exclusive. Reductionist interventions, including pathway inhibitors, CAR T-cell therapy, and CD20-directed depletion, can provide rapid and substantial control of dominant disease processes, with their roles in induction, maintenance, or repeated treatment varying across clinical settings. High-dimensional, tolerance-oriented strategies may add particular value during immune recovery or longer-term disease management, where the goal extends from controlling inflammation to understanding and, where possible, reinforcing a more stable regulatory balance35. The duration of the maintenance phase will likely depend on the stability of the restored immune architecture. In some patients, transient reinforcement of immune regulation may be sufficient until the risk of pathogenic clonal re-emergence has subsided, whereas persistent genetic susceptibility, epigenetic programmes or other disease-promoting factors may sustain autoimmune risk and require long-term tolerance-oriented therapy. In these settings, maintenance strategies informed by high-dimensional monitoring of the recovering immunome, including antigen-specific peptide therapy, regulatory cell infusion, or tolerogenic combination regimens, may be particularly relevant to sustained, treatment-minimised disease control because they are directly aligned with the architecture of immune regulation itself. This framework motivates the discussion of tolerance-oriented modalities in upcoming sections.High-dimensional approaches and their applications in autoimmune diseasesThe recognition of the autoimmunome as a high-dimensional, network-structured system necessitates a corresponding shift in experimental and translational strategies. Traditional immune profiling approaches designed to measure limited parameters in bulk or in single compartments are poorly suited to capture the compositional heterogeneity, functional plasticity, and spatial organisation that define autoimmune pathology. High-dimensional approaches address this gap by enabling simultaneous, multi-layered interrogation of immune systems, providing the resolution required to reconstruct immune architecture and identify pathogenic subnetworks. Importantly, these technologies are increasingly being applied in translational settings to support patient stratification, target identification, and monitoring of therapeutic responses across diseases (Table 115,23,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70).Table 1 High-dimensional technologies for immune profiling in autoimmunity: scope, value and limitationsFull size tableSingle-cell and multimodal immune profilingSingle-cell transcriptomic technologies have become foundational for dissecting immune heterogeneity in autoimmunity. scRNA-seq resolves immune populations into discrete but interconnected cellular states, revealing pathogenic and regulatory programmes that are obscured by conventional subset-based analyses71. Multimodal extensions such as CITE-seq further integrate transcriptomic and surface protein information, improving immune state annotation and resolution of rare populations44,72.Critically, the integration of scRNA-seq with T-cell and B-cell receptor sequencing links functional states to clonal lineage, allowing autoreactive clones to be tracked across tissues and disease stages73,74. Such approaches have demonstrated that clonally related lymphocytes can adopt divergent phenotypes depending on tissue context, reinforcing the need for tissue-aware immune profiling.Single-cell epigenomic profiling, including scATAC-seq and multiome platforms, adds a regulatory dimension by capturing chromatin accessibility and lineage potential36,75.High-dimensional proteomics and longitudinal immune monitoringHigh-dimensional cytometry (CyTOF) enables longitudinal, single-cell-resolved quantification of dozens of proteins across immune populations using serial cellular samples, capturing activation states, checkpoint expression, and signalling responses over time47,48,76,77. In parallel, affinity-based plasma proteomics platforms enable longitudinal measurement of cytokines, chemokines, and pathway-level signals, facilitating temporal mapping of immune responses during disease progression or therapeutic intervention78.Spatial immunomics and tissue-resolved immune statesThe emergence of spatial transcriptomic and proteomic technologies has fundamentally altered the understanding of autoimmune pathology. By preserving tissue context, these approaches reveal that immune responses are organised into discrete microenvironments where immune cells, stromal elements, antigen presentation, and cytokine gradients converge. Foundational work in tissue immunology established that immune decisions are embedded within spatially structured niches, rather than occurring uniformly in circulation23,55.More recently, systematic efforts to map immune wiring across human tissues have reinforced the view that immune function is distributed across structured intercellular networks, providing a conceptual and technical basis for spatially informed immune profiling11.Antigen-specific immune mappingAntigen-specific mapping approaches, including epitope libraries and peptide-MHC multimer technologies, enable direct identification of T-cell specificities in autoimmune disease. By linking immune cell states to their cognate antigens, these approaches provide a critical layer of resolution for defining pathogenic immune responses and support the development of antigen-specific therapeutic strategies42,46.Functional perturbation and metabolic profilingHigh-dimensional profiling increasingly extends beyond descriptive mapping toward functional and causal interrogation. Perturbation-based approaches such as Perturb-seq combine genetic or controlled extracellular stimuli with single-cell readouts, enabling systematic identification of regulatory circuits that control immune cell states40.Regulatory versus effector states are stabilised by distinct metabolic pathways, which can be resolved using high-dimensional metabolic profiling, positioning metabolism as a key axis of immune regulation and a potential lever for tolerance induction79.Integration, modelling, and therapeutic translationA major challenge is integrating high-dimensional data across biological and clinical layers. Pathogenic mechanisms arise through interactions among immune-cell states, tissue microenvironments, and environmental influences, including the microbiome. This requires integrative models that connect these dimensions.Artificial intelligence and machine-learning frameworks are increasingly being applied to this integrative step, enabling scalable analysis of complex, multimodal immune datasets and supporting precision immunomonitoring by linking architecture to diagnosis, prognosis, and therapeutic response in autoimmune disease80.Conceptually, these approaches support what we term the Architecture-Anomaly-Target-Lead (AATL) framework for immunotherapy development. Immune architecture is first mapped at high resolution; pathological anomalies are then identified as network perturbations; therapeutic targets are defined as connected immune compartments; and leads are selected or engineered to restore balance within those compartments, with the aim of correcting the underlying anomalies9,10,11. This framework provides a structured pathway for organising high-dimensional immune data around therapeutic target and lead selection, with the broader aim of coordinating modulation across dysregulated immune compartments.To operationalise this framework, each component can be linked to explicit biological and translational criteria.ArchitectureHigh-dimensional mapping of the patient immunome benchmarked against age- and sex-matched healthy reference atlases.AnomalyReproducible deviations from reference distributions, defined by pre-specified statistical or biological criteria such as altered network topology, expanded pathogenic clones, depleted regulatory states, or compartment-specific inflammatory modules, with cross-cohort replication required before clinical use.TargetSelection of dysregulated immune compartments, rather than isolated molecular nodes, whose modulation is hypothesised to restore immune balance, with explicit identification of the module being engaged.LeadTherapeutic candidates chosen or engineered to modulate the targeted compartment, with high-dimensional readouts of immune rebalancing used as pre-specified pharmacodynamic endpoints alongside clinical outcomes.High-dimensional immune profiling may help connect mechanistic observations with therapeutic evaluation by identifying dysregulated immune compartments, stratifying patients, and tracking immune-state changes during treatment. For example, in clinically quiescent juvenile idiopathic arthritis, residual immunome perturbation was associated with subsequent relapse following anti-TNFα withdrawal47. Such observations illustrate how immune profiling can provide information beyond conventional clinical disease measures and establish a quantitative baseline for evaluating tolerance-oriented interventions. Prospective validation will be required before these approaches can routinely inform therapeutic decisions.Immune tolerance as a therapeutic objectiveImmune tolerance is not the property of any single cell type but emerges from a multi-layered regulatory architecture. Central and deletional mechanisms, including thymic negative selection, B-cell receptor editing, and clonal deletion, eliminate most autoreactive clones before they enter the periphery. FOXP3⁺ regulatory T cells provide active peripheral suppression through cytokine secretion, metabolic competition, and contact-dependent inhibition. Tr1 cells represent a distinct FOXP3-independent regulatory population characterised by high IL-10 production and suppressive activity81, while iTr35 cells mediate suppression via IL-35 and operate independently of both FOXP3 and IL-1082. Tolerogenic antigen-presenting cells and tissue-specific regulatory niches provide additional layers of peripheral immune regulation83. Together, these mechanisms indicate that immune tolerance emerges from a distributed, multi-layered regulatory system rather than a single dominant pathway, providing a rationale for therapeutic strategies capable of reinforcing coordinated regulatory effects across multiple immune compartments. This multi-layered view may, in some settings, support a Reset-Rebuild architectural model of tolerance restoration, in which lymphodepleting or lineage-targeted modalities first remove entrenched effector hubs, and tolerogenic programmes are then reinforced during immune reconstitution35.Unlike pathway-specific therapies, tolerance-inducing strategies have the potential to remodel multiple immune compartments. Through coordinated effects on antigen-presenting cells, autoreactive T and B cells, cytokine networks, and regulatory circuits, these approaches may progressively restore immune architecture rather than simply suppress inflammatory pathways81. For example, Treg-based approaches can modulate dendritic-cell activation and antigen presentation, including through CTLA-4-dependent regulation of APC costimulatory capacity84, suppress pathogenic effector T-cell responses, and reshape local cytokine environments. Regulatory T-cell subsets can also influence humoral immunity by modulating T follicular helper-B-cell interactions and germinal-centre responses85. Antigen-directed tolerogenic approaches may selectively modulate disease-relevant T-cell responses86. By limiting the expansion and persistence of dominant autoreactive responses, these interventions may also alter clonal dominance and repertoire composition, although direct evidence for restoration of TCR or BCR repertoire diversity in humans remains limited. Through these coordinated effects on dendritic cells, T cells, B cells, cytokine networks, and clonal repertoires, tolerance-oriented therapies may progressively reinforce regulatory immune architecture.Cellular tolerance therapies: polyclonal Tregs and low-dose IL-2The translational feasibility of autologous Treg therapy has been tested across autoimmune disease and transplantation. In type 1 diabetes, phase 1 trials demonstrated that ex vivo-expanded polyclonal Tregs are safe, persist long-term, and support β-cell function in selected patients87,88. In kidney transplantation, the multicentre ONE Study showed that regulatory cell therapy permitted ~40% of recipients to be weaned off mycophenolate mofetil onto tacrolimus monotherapy with fewer infections and comparable rejection rates to standard-of-care89. Low-dose IL-2, which preferentially expands endogenous Tregs, has shown clinical signals in systemic lupus erythematosus and across multiple autoimmune diseases90,91.These studies establish the feasibility of strengthening regulatory immunity, while also highlighting two practical limitations: limited antigen specificity and variability in the quality of patient-derived Tregs. In several autoimmune settings, autologous Tregs may carry disease-associated defects in suppressive function or lineage stability, particularly under inflammatory stress92. Rapamycin-based expansion can favour FOXP3⁺ Treg outgrowth93, but more direct approaches seek either to stabilise regulatory identity through engineering or to reduce dependence on patient-derived Tregs through renewable cell sources. Engineered and iPSC-derived Tregs can therefore be viewed as strategies to overcome patient-intrinsic Treg instability or dysfunction, rather than simply as alternative manufacturing platforms.Engineered antigen-specific cellular therapies: CAR-TregsTo address the lack of antigen specificity, chimeric antigen receptor (CAR) engineering has been adapted to Tregs. First-in-human CAR-Treg trials are underway, including STEADFAST (NCT04817774), evaluating an HLA-A2-specific CAR-Treg in kidney transplantation, and LIBERATE (NCT05234190), evaluating an HLA-A2-directed engineered Treg product in liver transplantation94,95. These trials provide early clinical evaluation of the safety and feasibility of engineered Treg products. Parallel programmes are exploring disease- or tissue-relevant targets in other immune-mediated settings. Important challenges remain, particularly in target selection and the durability of regulatory function after transfer. These challenges have motivated broader efforts to develop more stable and scalable regulatory-cell products.Toward a next-generation universal regulatory cell platformCurrent efforts to develop more broadly applicable regulatory-cell therapies address several limitations of conventional Treg products. Enforced FOXP3 expression can improve the stability of human CAR-Treg products96, whereas CAR- and TCR-based redirection can concentrate regulatory activity in defined antigenic or tissue contexts. iPSC platforms address the cell-source problem by providing a renewable and editable starting population from which functional regulatory T cells, including CAR-engineered variants, can be generated97,98.In parallel, hypoimmunogenic engineering of pluripotent cells provides a general route towards immune-evasive allogeneic cell products. Strategies combining reduced MHC class I and II expression with CD47 overexpression have progressed to studies in immunocompetent non-human primates99,100. Regulatory-cell platforms beyond conventional FOXP3⁺ Tregs, including Tr1 cells and tolerogenic dendritic cells, further broaden the field81,101.PANTreg represents one conceptual example that seeks to combine several of these principles in a single regulatory-cell design, including inflammation-responsive activity, enforced regulatory identity, and inducible delivery of tolerogenic mediators102. The individual design principles are supported by the broader engineering literature discussed above, although the integrated platform itself remains to be validated. More generally, human evidence that engineered regulatory-cell therapies can durably restore immune-network architecture remains limited, and this will require longitudinal clinical evaluation.Peptide-based antigen-specific tolerancePeptide-based strategies complement cell-based approaches by delivering tolerogenic signals through defined autoantigen epitopes rather than engineered cell products. The heat shock protein-derived peptide dnaJP1, tested in RA, induces re-engagement of regulatory pathways and cytokine rebalancing consistent with tolerance induction rather than effector deletion86,103,104,105. Allergen immunotherapy, the most clinically mature tolerance paradigm, induces Treg expansion, IgG4 class switching, and durable suppression of pathogenic effector responses that persist after treatment cessation in some settings106, providing proof-of-principle that sustained, antigen-specific tolerance can be achieved in humans. Broader translation remains limited by incomplete identification of dominant antigens, epitope spreading, HLA restriction, tissue access, and inter-patient heterogeneity, all of which complicate patient selection.Connecting high-dimensional profiling to tolerance-oriented therapyAcross these three modality classes, evidence for broader immune modulation beyond single-pathway suppression is emerging: dnaJP1 responses involve re-engagement of regulatory pathways rather than pathway suppression86,103,104,105; allergen immunotherapy induces durable Treg expansion, IgG4 class switching, and suppression that persists after treatment cessation106; and ONE Study regulatory cell therapy permits immunosuppression minimisation without excess rejection89. However, direct evidence that these interventions fully restore human immune architecture remains limited and will require longitudinal high-dimensional studies linking network-level immune restoration with long-term clinical outcomes. Pretreatment high-dimensional profiling provides a framework for addressing this gap by guiding the selection and design of tolerance-directed therapies according to the dominant immune anomaly and, where identifiable, the relevant antigenic or tissue context. Longitudinal profiling during treatment may further inform dose, dosing interval, or the need for repeat administration by tracking the magnitude and persistence of regulatory immune states and the re-emergence of pathogenic cellular or clonal programmes. Identification of residual or compensatory immune modules may support rational combination strategies, while changes in regulatory-effector balance, pathogenic clonal expansion, cytokine networks, and tissue-associated immune states may serve as pre-specified pharmacodynamic and immune-restoration endpoints alongside clinical outcomes. Although prospective use of high-dimensional profiling to direct these therapeutic decisions remains to be clinically validated, existing longitudinal immunome studies suggest that high-dimensional immune states may provide information beyond conventional measures of clinical disease activity47.Architecture-aware immunotherapeutics: future directions and challenges in autoimmune diseasesThe emergence of high-dimensional immunology increasingly supports viewing autoimmunity through the lens of immune architecture, inspiring the development of ‘architecture-aware’ immunotherapeutic concepts that seek to reinforce network balance alongside control of dominant inflammatory pathways11. However, despite rapid methodological innovation, cross-platform and analytical variability continue to complicate the standardisation and reproducible interpretation of high-dimensional single-cell data107. Bridging this gap from architectural insight to clinical implementation presents a series of intertwined conceptual, technical, and regulatory challenges.A first hurdle is the standardisation of high-dimensional biomarkers. While single-cell, spatial, and proteomic platforms can sensitively capture immune states, their outputs are complex, multidimensional, and often cohort-specific108. Defining reproducible, clinically interpretable signatures of immune tolerance rather than disease activity or transient inflammatory suppression will be essential. This requires harmonised analytical pipelines, reference atlases, and consensus endpoints that regulators and clinicians can apply across trials and diseases109.Second, regulatory frameworks for tolerance-inducing and engineered immune therapies are still evolving but remain a practical bottleneck. Unlike conventional anti-inflammatory agents, tolerance-based interventions, particularly those involving cellular or antigen-specific modalities, place greater emphasis on potency, comparability, and long-term follow-up, issues that become more complex when the intended mechanism is immune reprogramming rather than cytotoxicity. Regulators will need validated biomarkers that demonstrate restoration of immune balance without compromising host defence, as well as long-term monitoring strategies tailored to immune reprogramming rather than acute suppression110.Third, manufacturing constraints remain a major barrier to scalability and clinical deployment. For antigen-specific regulatory T cells and engineered products such as CAR-Tregs, low precursor frequencies, phenotypic instability, product heterogeneity, and individualised manufacturing pipelines complicate scale-out, reproducibility, and cost-effectiveness101,111. These limitations have catalysed growing interest in hypoimmunogenic iPSC-derived regulatory immune cells as off-the-shelf alternatives, offering the potential for standardised, renewable, and scalable tolerance-inducing products while reducing donor variability and manufacturing burden102,112.These practical constraints also motivate the exploration of combination regimens in which tolerogenic agents augment or stabilise endogenous regulation rather than acting as standalone interventions101. Notably, dnaJP1 has been evaluated clinically, and post hoc analyses have supported potential synergistic benefit with background immunomodulatory therapy in RA, including concomitant use of hydroxychloroquine86. Such regimens may offer a pragmatic route to enhance sustainability of response while mitigating the limitations of single-modality tolerance induction.A further challenge is capturing tissue-specific immune states that may not be fully represented by peripheral blood, while direct tissue sampling remains difficult in routine care23,55. The development of minimally invasive, spatial, and surrogate biomarkers will be important for capturing tissue-relevant immune states.Finally, the translational value of high-dimensional immunology depends on the integration of multiomic datasets into clinically actionable frameworks113,114. The scale and complexity of high-dimensional datasets, including multiomic and spatial data, present challenges for data integration, interpretation and standardisation, which currently limit their routine clinical implementation. Addressing analytical complexity will require harmonised preprocessing and quality-control standards, benchmarked integration methods, reference datasets and prospective multi-centre validation107,108,115. In parallel, longitudinal and spatial sampling will be needed to capture temporal and tissue-specific immune dynamics that are missed by cross-sectional peripheral profiling. Computational models will also be needed to relate complex immune-state patterns to reproducible clinical endpoints and therapeutic decisions. Addressing these challenges will determine whether high-dimensional immunology can meaningfully inform the development and evaluation of more durable, tolerance-oriented therapies for autoimmune disease.Data availabilityNo datasets were generated or analysed during the current study.ReferencesChen, C. et al. Global, regional and national disparities and temporal trends of common autoimmune disease burdens among children and adolescents from 1990 to 2019. BMJ Glob. Health 10, e017187 https://doi.org/10.1136/bmjgh-2024-017187 (2025).Mackensen, A. et al. Anti-CD19 CAR T cell therapy for refractory systemic lupus erythematosus. Nat. Med. 28, 2124–2132 https://doi.org/10.1038/s41591-022-02017-5 (2022).Article CAS PubMed Google Scholar Muller, F. et al. CD19 CAR T-cell therapy in autoimmune disease - a case series with follow-up. N. Engl. J. Med. 390, 687–700 https://doi.org/10.1056/NEJMoa2308917 (2024).Wu, D. et al. CAR T-cell therapy in autoimmune diseases: a promising frontier on the horizon. Front. Immunol. 16, 1613878 https://doi.org/10.3389/fimmu.2025.1613878 (2025).Article CAS PubMed PubMed Central Google Scholar McInnes, I. B. & Gravallese, E. M. Immune-mediated inflammatory disease therapeutics: past, present and future. Nat. Rev. Immunol. 21, 680–686 https://doi.org/10.1038/s41577-021-00603-1 (2021).Article CAS PubMed PubMed Central Google Scholar Fugger, L., Jensen, L. T. & Rossjohn, J. Challenges, progress, and prospects of developing therapies to treat autoimmune diseases. Cell 181, 63–80 https://doi.org/10.1016/j.cell.2020.03.007 (2020).Article CAS PubMed Google Scholar Nagafuchi, Y., Yanaoka, H. & Fujio, K. Lessons from transcriptome analysis of autoimmune diseases. Front. Immunol. 13, 857269 https://doi.org/10.3389/fimmu.2022.857269 (2022).Article CAS PubMed PubMed Central Google Scholar Song, Y., Li, J. & Wu, Y. Evolving understanding of autoimmune mechanisms and new therapeutic strategies of autoimmune disorders. Signal Transduct. Target Ther. 9, 263 https://doi.org/10.1038/s41392-024-01952-8 (2024).Article CAS PubMed PubMed Central Google Scholar Bonaguro, L. et al. A guide to systems-level immunomics. Nat. Immunol. 23, 1412–1423 https://doi.org/10.1038/s41590-022-01309-9 (2022).Article CAS PubMed Google Scholar Kumar, P. et al. Single-cell transcriptomics and surface epitope detection in human brain epileptic lesions identifies pro-inflammatory signaling. Nat. Neurosci. 25, 956–966 https://doi.org/10.1038/s41593-022-01095-5 (2022).Article CAS PubMed PubMed Central Google Scholar Shilts, J. et al. A physical wiring diagram for the human immune system. Nature 608, 397–404 https://doi.org/10.1038/s41586-022-05028-x (2022).Article CAS PubMed PubMed Central Google Scholar Kumar, P. et al. CyNET — a network analysis framework for high dimensional, system level analyses of the functional immunome. J. Immunol. 215, vkag064 https://doi.org/10.1093/jimmun/vkag064 (2026).Banchereau, R., Cepika, A. M., Banchereau, J. & Pascual, V. Understanding human autoimmunity and autoinflammation through transcriptomics. Annu. Rev. Immunol. 35, 337–370 https://doi.org/10.1146/annurev-immunol-051116-052225 (2017).Article CAS PubMed PubMed Central Google Scholar Zaslavsky, M. E. et al. Disease diagnostics using machine learning of B cell and T cell receptor sequences. Science 387, eadp2407 https://doi.org/10.1126/science.adp2407 (2025).Article CAS PubMed PubMed Central Google Scholar Zhang, F. et al. Defining inflammatory cell states in rheumatoid arthritis joint synovial tissues by integrating single-cell transcriptomics and mass cytometry. Nat. Immunol. 20, 928–942 https://doi.org/10.1038/s41590-019-0378-1 (2019).Article CAS PubMed PubMed Central Google Scholar Dunlap, G. et al. Clonal associations between lymphocyte subsets and functional states in rheumatoid arthritis synovium. Nat. Commun. 15, 4991 https://doi.org/10.1038/s41467-024-49186-0 (2024).Article CAS PubMed PubMed Central Google Scholar Galli, S. J., Borregaard, N. & Wynn, T. A. Phenotypic and functional plasticity of cells of innate immunity: macrophages, mast cells and neutrophils. Nat. Immunol. 12, 1035–1044 https://doi.org/10.1038/ni.2109 (2011).Article CAS PubMed PubMed Central Google Scholar Leong, J. Y. et al. Epigenetic instability and hypofunctionality of fetal Tregs allow a permissive regulatory environment for T effector memory maturation. Proc. Natl. Acad. Sci. USA 122, e2506673122 https://doi.org/10.1073/pnas.2506673122 (2025).Article CAS PubMed PubMed Central Google Scholar Voo, K. S. et al. Identification of IL-17-producing FOXP3+ regulatory T cells in humans. Proc. Natl. Acad. Sci. USA 106, 4793–4798 https://doi.org/10.1073/pnas.0900408106 (2009).Article PubMed PubMed Central Google Scholar Kleinewietfeld, M. & Hafler, D. A. The plasticity of human Treg and Th17 cells and its role in autoimmunity. Semin. Immunol. 25, 305–312 https://doi.org/10.1016/j.smim.2013.10.009 (2013).Article CAS PubMed PubMed Central Google Scholar Nistala, K. et al. Th17 plasticity in human autoimmune arthritis is driven by the inflammatory environment. Proc. Natl. Acad. Sci. USA 107, 14751–14756 https://doi.org/10.1073/pnas.1003852107 (2010).Article PubMed PubMed Central Google Scholar Buckley, C. D., Ospelt, C., Gay, S. & Midwood, K. S. Location, location, location: how the tissue microenvironment affects inflammation in RA. Nat. Rev. Rheumatol. 17, 195–212 https://doi.org/10.1038/s41584-020-00570-2 (2021).Article PubMed Google Scholar Vickovic, S. et al. Three-dimensional spatial transcriptomics uncovers cell type localizations in the human rheumatoid arthritis synovium. Commun. Biol. 5, 129 https://doi.org/10.1038/s42003-022-03050-3 (2022).Article CAS PubMed PubMed Central Google Scholar Leong, J. Y. et al. POS0156 CD4+HLADR+ T cells with autoimmune reactivity that resist regulatory control and perpetuate disease inflammation in juvenile idiopathic arthritis. Ann. Rheum. Dis. 83, 527 https://doi.org/10.1136/annrheumdis-2024-eular.4615 (2024).Article Google Scholar Croft, A. P. et al. Distinct fibroblast subsets drive inflammation and damage in arthritis. Nature 570, 246–251 https://doi.org/10.1038/s41586-019-1263-7 (2019).Article CAS PubMed PubMed Central Google Scholar Gibbons, J. B., Laber, M. & Bennett, C. L. Humira: the first $20 billion drug. Am. J. Manag Care 29, 78–80 https://doi.org/10.37765/ajmc.2023.89315 (2023).Article PubMed Google Scholar Tanaka, Y. et al. Discontinuation of adalimumab after achieving remission in patients with established rheumatoid arthritis: 1-year outcome of the HONOR study. Ann. Rheum. Dis. 74, 389–395 https://doi.org/10.1136/annrheumdis-2013-204016 (2015).Article CAS PubMed Google Scholar Ytterberg, S. R. et al. Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis. N. Engl. J. Med. 386, 316–326 https://doi.org/10.1056/NEJMoa2109927 (2022).Article CAS PubMed Google Scholar Stone, J. H. et al. Rituximab versus cyclophosphamide for ANCA-associated vasculitis. N. Engl. J. Med. 363, 221–232 https://doi.org/10.1056/NEJMoa0909905 (2010).Article CAS PubMed PubMed Central Google Scholar Hauser, S. L. et al. Ocrelizumab versus interferon Beta-1a in relapsing multiple sclerosis. N. Engl. J. Med. 376, 221–234 https://doi.org/10.1056/NEJMoa1601277 (2017).Article CAS PubMed Google Scholar Montalban, X. et al. Ocrelizumab versus placebo in primary progressive multiple sclerosis. N. Engl. J. Med. 376, 209–220 https://doi.org/10.1056/NEJMoa1606468 (2017).Article CAS PubMed Google Scholar Furie, R. A. et al. Efficacy and safety of obinutuzumab in active lupus nephritis. N. Engl. J. Med. 392, 1471–1483 https://doi.org/10.1056/NEJMoa2410965 (2025).Article CAS PubMed Google Scholar Burt, R. K. et al. Effect of nonmyeloablative hematopoietic stem cell transplantation vs continued disease-modifying therapy on disease progression in patients with relapsing-remitting multiple sclerosis: a randomized clinical trial. JAMA 321, 165–174 https://doi.org/10.1001/jama.2018.18743 (2019).Article PubMed PubMed Central Google Scholar Halliley, J. L. et al. Long-lived plasma cells are contained within the CD19(-)CD38(hi)CD138(+) subset in human bone marrow. Immunity 43, 132–145 https://doi.org/10.1016/j.immuni.2015.06.016 (2015).Article CAS PubMed PubMed Central Google Scholar Albani, S., Koffeman, E. C. & Prakken, B. Induction of immune tolerance in the treatment of rheumatoid arthritis. Nat. Rev. Rheumatol. 7, 272–281 https://doi.org/10.1038/nrrheum.2011.36 (2011).Article CAS PubMed Google Scholar Buenrostro, J. D. et al. Single-cell chromatin accessibility reveals principles of regulatory variation. Nature 523, 486–490 https://doi.org/10.1038/nature14590 (2015).Article CAS PubMed PubMed Central Google Scholar Banchereau, R. et al. Personalized immunomonitoring uncovers molecular networks that stratify lupus patients. Cell 165, 551–565 https://doi.org/10.1016/j.cell.2016.03.008 (2016).Article CAS PubMed PubMed Central Google Scholar Bengtsson, A. A. et al. Metabolic profiling of systemic lupus erythematosus and comparison with primary sjogren’s syndrome and systemic sclerosis. PLoS ONE 11, e0159384 https://doi.org/10.1371/journal.pone.0159384 (2016).Article CAS PubMed PubMed Central Google Scholar Delong, T. et al. Pathogenic CD4 T cells in type 1 diabetes recognize epitopes formed by peptide fusion. Science 351, 711–714 https://doi.org/10.1126/science.aad2791 (2016).Article CAS PubMed PubMed Central Google Scholar Dixit, A. et al. Perturb-seq: dissecting molecular circuits with scalable single-cell RNA profiling of pooled genetic screens. Cell 167, 1853–1866.e17 https://doi.org/10.1016/j.cell.2016.11.038 (2016).O’Neill, L. A., Kishton, R. J. & Rathmell, J. A guide to immunometabolism for immunologists. Nat. Rev. Immunol. 16, 553–565 https://doi.org/10.1038/nri.2016.70 (2016).Article CAS PubMed PubMed Central Google Scholar Bentzen, A. K. & Hadrup, S. R. Evolution of MHC-based technologies used for detection of antigen-responsive T cells. Cancer Immunol. Immunother. 66, 657–666 https://doi.org/10.1007/s00262-017-1971-5 (2017).Article CAS PubMed PubMed Central Google Scholar Pianta, A. et al. Evidence of the immune relevance of prevotella copri, a gut microbe, in patients with rheumatoid arthritis. Arthritis Rheumatol. 69, 964–975 https://doi.org/10.1002/art.40003 (2017).Article CAS PubMed PubMed Central Google Scholar Stoeckius, M. et al. Simultaneous epitope and transcriptome measurement in single cells. Nat. Methods 14, 865–868 https://doi.org/10.1038/nmeth.4380 (2017).Article CAS PubMed PubMed Central Google Scholar Damond, N. et al. A map of human type 1 diabetes progression by imaging mass cytometry. Cell Metab. 29, 755–768.e5 https://doi.org/10.1016/j.cmet.2018.11.014 (2019).Kula, T. et al. T-Scan: a genome-wide method for the systematic discovery of T cell epitopes. Cell 178, 1016–1028.e13 https://doi.org/10.1016/j.cell.2019.07.009 (2019).Leong, J. Y. et al. Immunome perturbation is present in patients with juvenile idiopathic arthritis who are in remission and will relapse upon anti-TNFalpha withdrawal. Ann. Rheum. Dis. 78, 1712–1721 https://doi.org/10.1136/annrheumdis-2019-216059 (2019).Article CAS PubMed PubMed Central Google Scholar Yeo, J. G. et al. The extended polydimensional immunome characterization (EPIC) web-based reference and discovery tool for cytometry data. Nat. Biotechnol. 38, 679–684 https://doi.org/10.1038/s41587-020-0532-1 (2020).Article CAS PubMed Google Scholar King, H. W. et al. Integrated single-cell transcriptomics and epigenomics reveals strong germinal center-associated etiology of autoimmune risk loci. Sci. Immunol. 6, eabh3768 https://doi.org/10.1126/sciimmunol.abh3768 (2021).Article CAS PubMed PubMed Central Google Scholar Wieber, K., Zimmer, C. L. & Hertl, M. Detection of autoreactive CD4+ T cells by MHC class II multimers in HLA-linked human autoimmune diseases. J. Clin. Invest. 131, e148674 https://doi.org/10.1172/JCI148674 (2021).Argyriou, A. et al. Single cell sequencing identifies clonally expanded synovial CD4(+) T(PH) cells expressing GPR56 in rheumatoid arthritis. Nat. Commun. 13, 4046 https://doi.org/10.1038/s41467-022-31519-6 (2022).Article CAS PubMed PubMed Central Google Scholar Guo, C. et al. Single-cell transcriptome profiling and chromatin accessibility reveal an exhausted regulatory CD4+ T cell subset in systemic lupus erythematosus. Cell Rep. 41, 111606 https://doi.org/10.1016/j.celrep.2022.111606 (2022).Article CAS PubMed Google Scholar Huang, X. et al. Multi-platform omics analysis reveals molecular signatures for pathogenesis and activity of systemic lupus erythematosus. Front. Immunol. 13, 833699 https://doi.org/10.3389/fimmu.2022.833699 (2022).Akesson, J. et al. Proteomics reveal biomarkers for diagnosis, disease activity and long-term disability outcomes in multiple sclerosis. Nat. Commun. 14, 6903 https://doi.org/10.1038/s41467-023-42682-9 (2023).Article CAS PubMed PubMed Central Google Scholar Castillo, R. L. et al. Spatial transcriptomics stratifies psoriatic disease severity by emergent cellular ecosystems. Sci. Immunol. 8, eabq7991 https://doi.org/10.1126/sciimmunol.abq7991 (2023).Article CAS PubMed PubMed Central Google Scholar Gupta, A. et al. Dynamic regulatory elements in single-cell multimodal data implicate key immune cell states enriched for autoimmune disease heritability. Nat. Genet. 55, 2200–2210 https://doi.org/10.1038/s41588-023-01577-7 (2023).Article CAS PubMed PubMed Central Google Scholar Wu, Y. et al. Immunometabolomics provides a new perspective for studying systemic lupus erythematosus. Int. Immunopharmacol. 118, 109946 https://doi.org/10.1016/j.intimp.2023.109946 (2023).Article CAS PubMed Google Scholar Baker, T. et al. Type I interferon blockade with anifrolumab in patients with systemic lupus erythematosus modulates key immunopathological pathways in a gene expression and proteomic analysis of two phase 3 trials. Ann. Rheum. Dis. 83, 1018–1027 https://doi.org/10.1136/ard-2023-225445 (2024).Article CAS PubMed PubMed Central Google Scholar Fan, Y. et al. Immune features revealed by single-cell RNA and single-cell TCR/BCR sequencing in patients with rheumatoid arthritis receiving COVID-19 booster vaccination. J. Med. Virol. 96, e29573 https://doi.org/10.1002/jmv.29573 (2024).Article CAS PubMed Google Scholar Figueiredo, M. L. Applications of single-cell RNA sequencing in rheumatoid arthritis. Front. Immunol. 15, 1491318 https://doi.org/10.3389/fimmu.2024.1491318 (2024).Article CAS PubMed PubMed Central Google Scholar Kante, A. et al. Mass cytometry: exploring the immune landscape of systemic autoimmune and inflammatory diseases in the past fourteen years. Front. Immunol. 15, 1509782 https://doi.org/10.3389/fimmu.2024.1509782 (2024).Article CAS PubMed Google Scholar Niarakis, A. et al. Immune digital twins for complex human pathologies: applications, limitations, and challenges. NPJ Syst. Biol. Appl. 10, 141 https://doi.org/10.1038/s41540-024-00450-5 (2024).Article PubMed PubMed Central Google Scholar Tian, C. et al. Single-cell RNA sequencing of peripheral blood links cell-type-specific regulation of splicing to autoimmune and inflammatory diseases. Nat. Genet. 56, 2739–2752 https://doi.org/10.1038/s41588-024-02019-8 (2024).Article CAS PubMed PubMed Central Google Scholar Wang, X. et al. Emerging role of gut microbiota in autoimmune diseases. Front. Immunol. 15, 1365554 https://doi.org/10.3389/fimmu.2024.1365554 (2024).Article CAS PubMed PubMed Central Google Scholar Wu, M. et al. Spatial proteomics: unveiling the multidimensional landscape of protein localization in human diseases. Proteome Sci. 22, 7 https://doi.org/10.1186/s12953-024-00231-2 (2024).Article CAS PubMed PubMed Central Google Scholar Ho, C. H. et al. Genetic and epigenetic screens in primary human T cells link candidate causal autoimmune variants to T cell networks. Nat. Genet. 57, 2536–2545 https://doi.org/10.1038/s41588-025-02301-3 (2025).Article CAS PubMed PubMed Central Google Scholar Rood, J. E. et al. The Human Cell Atlas from a cell census to a unified foundation model. Nature 637, 1065–1071 https://doi.org/10.1038/s41586-024-08338-4 (2025).Article CAS PubMed Google Scholar Shimagami, H. et al. Single-cell analysis reveals immune cell abnormalities underlying the clinical heterogeneity of patients with systemic sclerosis. Nat. Commun. 16, 4949 https://doi.org/10.1038/s41467-025-60034-7 (2025).Article CAS PubMed PubMed Central Google Scholar Liu, Q. et al. Single-cell profiling unveils nephritis-related circulating immunological signatures in systemic lupus erythematosus patients. Commun. Biol. 9, 155 https://doi.org/10.1038/s42003-025-09431-8 (2026).Article CAS PubMed PubMed Central Google Scholar Medina-Serpas, M. A. et al. Spatial transcriptomics from pancreas and local draining lymph node tissue reveals a lymphotoxin-beta signature in human type 1 diabetes. Cell Rep. 45, 117144 https://doi.org/10.1016/j.celrep.2026.117144 (2026).Article CAS PubMed PubMed Central Google Scholar Papalexi, E. & Satija, R. Single-cell RNA sequencing to explore immune cell heterogeneity. Nat. Rev. Immunol. 18, 35–45 https://doi.org/10.1038/nri.2017.76 (2018).Article CAS PubMed Google Scholar Hao, Y. et al. Integrated analysis of multimodal single-cell data. Cell 184, 3573–3587.e29 https://doi.org/10.1016/j.cell.2021.04.048 (2021).Stubbington, M. J. T. et al. T cell fate and clonality inference from single-cell transcriptomes. Nat. Methods 13, 329–332 https://doi.org/10.1038/nmeth.3800 (2016).Article CAS PubMed PubMed Central Google Scholar Lindeman, I. et al. BraCeR: B-cell-receptor reconstruction and clonality inference from single-cell RNA-seq. Nat. Methods 15, 563–565 https://doi.org/10.1038/s41592-018-0082-3 (2018).Article CAS PubMed Google Scholar Buenrostro, J. D., Wu, B., Chang, H. Y. & Greenleaf, W. J. ATAC-seq: a method for assaying chromatin accessibility genome-wide. Curr. Protoc. Mol. Biol. 109, 21.29.1–21.29.9 https://doi.org/10.1002/0471142727.mb2129s109 (2015).Spitzer, M. H. & Nolan, G. P. Mass cytometry: single cells, many features. Cell 165, 780–791 https://doi.org/10.1016/j.cell.2016.04.019 (2016).Article CAS PubMed PubMed Central Google Scholar Leong, J. Y., Lajam, A., Ng, C. T. & Albani, S. POS0170 circulating pathogenic like lymphocytes are mechanistic targets for the tolerogenic re-education in rheumatoid arthritic patients. Ann. Rheum. Dis. 83, 300 https://doi.org/10.1136/annrheumdis-2024-eular.4619 (2024).Article Google Scholar Nakayasu, E. S. et al. Plasma protein biomarkers predict the development of persistent autoantibodies and type 1 diabetes 6 months prior to the onset of autoimmunity. Cell Rep. Med. 4, 101093 https://doi.org/10.1016/j.xcrm.2023.101093 (2023).Article CAS PubMed PubMed Central Google Scholar Pearce, E. L., Poffenberger, M. C., Chang, C. H. & Jones, R. G. Fueling immunity: insights into metabolism and lymphocyte function. Science 342, 1242454 https://doi.org/10.1126/science.1242454 (2013).Article CAS PubMed PubMed Central Google Scholar Yaung, K. N. et al. Artificial intelligence and high-dimensional technologies in the theragnosis of systemic lupus erythematosus. Lancet Rheumatol. 5, e151–e165 https://doi.org/10.1016/S2665-9913(23)00010-3 (2023).Article CAS PubMed Google Scholar Roncarolo, M. G., Gregori, S., Bacchetta, R., Battaglia, M. & Gagliani, N. The biology of T regulatory type 1 cells and their therapeutic application in immune-mediated diseases. Immunity 49, 1004–1019 https://doi.org/10.1016/j.immuni.2018.12.001 (2018).Article CAS PubMed Google Scholar Collison, L. W. et al. IL-35-mediated induction of a potent regulatory T cell population. Nat. Immunol. 11, 1093–1101 https://doi.org/10.1038/ni.1952 (2010).Article CAS PubMed PubMed Central Google Scholar Steinman, R. M. & Nussenzweig, M. C. Avoiding horror autotoxicus: the importance of dendritic cells in peripheral T cell tolerance. Proc. Natl. Acad. Sci. USA 99, 351–358 https://doi.org/10.1073/pnas.231606698 (2002).Article CAS PubMed PubMed Central Google Scholar Qureshi, O. S. et al. Trans-endocytosis of CD80 and CD86: a molecular basis for the cell-extrinsic function of CTLA-4. Science 332, 600–603 https://doi.org/10.1126/science.1202947 (2011).Article CAS PubMed PubMed Central Google Scholar Linterman, M. A. et al. Foxp3+ follicular regulatory T cells control the germinal center response. Nat. Med. 17, 975–982 https://doi.org/10.1038/nm.2425 (2011).Article CAS PubMed PubMed Central Google Scholar Koffeman, E. C. et al. Epitope-specific immunotherapy of rheumatoid arthritis: clinical responsiveness occurs with immune deviation and relies on the expression of a cluster of molecules associated with T cell tolerance in a double-blind, placebo-controlled, pilot phase II trial. Arthritis Rheum. 60, 3207–3216 https://doi.org/10.1002/art.24916 (2009).Article CAS PubMed Google Scholar Marek-Trzonkowska, N. et al. Administration of CD4+CD25highCD127- regulatory T cells preserves beta-cell function in type 1 diabetes in children. Diabetes Care 35, 1817–1820 https://doi.org/10.2337/dc12-0038 (2012).Bluestone, J. A. et al. Type 1 diabetes immunotherapy using polyclonal regulatory T cells. Sci. Transl. Med. 7, 315ra189 https://doi.org/10.1126/scitranslmed.aad4134 (2015).Article CAS PubMed PubMed Central Google Scholar Sawitzki, B. et al. Regulatory cell therapy in kidney transplantation (The ONE Study): a harmonised design and analysis of seven non-randomised, single-arm, phase 1/2A trials. Lancet 395, 1627–1639 https://doi.org/10.1016/S0140-6736(20)30167-7 (2020).Article CAS PubMed PubMed Central Google Scholar Rosenzwajg, M. et al. Immunological and clinical effects of low-dose interleukin-2 across 11 autoimmune diseases in a single, open clinical trial. Ann. Rheum. Dis. 78, 209–217 https://doi.org/10.1136/annrheumdis-2018-214229 (2019).Article CAS PubMed Google Scholar Humrich, J. Y. et al. Low-dose interleukin-2 therapy in active systemic lupus erythematosus (LUPIL-2): a multicentre, double-blind, randomised and placebo-controlled phase II trial. Ann. Rheum. Dis. 81, 1685–1694 https://doi.org/10.1136/ard-2022-222501 (2022).Article CAS PubMed Google Scholar Buckner, J. H. Mechanisms of impaired regulation by CD4(+)CD25(+)FOXP3(+) regulatory T cells in human autoimmune diseases. Nat. Rev. Immunol. 10, 849–859 https://doi.org/10.1038/nri2889 (2010).Article CAS PubMed PubMed Central Google Scholar Battaglia, M., Stabilini, A. & Roncarolo, M. G. Rapamycin selectively expands CD4+CD25+FoxP3+ regulatory T cells. Blood 105, 4743–4748 https://doi.org/10.1182/blood-2004-10-3932 (2005).Article CAS PubMed Google Scholar ClinicalTrials.gov. Safety & tolerability study of chimeric antigen receptor T-reg cell therapy in living donor renal transplant recipients (STEADFAST). https://clinicaltrials.gov/ct2/show/NCT04817774 (2021).ClinicalTrials.gov. Safety and clinical activity of QEL-001 in A2-mismatch liver transplant patients (LIBERATE). https://clinicaltrials.gov/ct2/show/NCT05234190 (2022).Henschel, P. et al. Supraphysiological FOXP3 expression in human CAR-Tregs results in improved stability, efficacy, and safety of CAR-Treg products for clinical application. J. Autoimmun. 138, 103057 https://doi.org/10.1016/j.jaut.2023.103057 (2023).Article CAS PubMed Google Scholar Yano, H. et al. Human iPSC-derived CD4(+) Treg-like cells engineered with chimeric antigen receptors control GvHD in a xenograft model. Cell Stem Cell 31, 795–802.e6 https://doi.org/10.1016/j.stem.2024.05.004 (2024).Fong, H. et al. A serum- and feeder-free system to generate CD4 and regulatory T cells from human iPSCs. Stem Cells 43, sxaf001 https://doi.org/10.1093/stmcls/sxaf001 (2025).Deuse, T. et al. Hypoimmunogenic derivatives of induced pluripotent stem cells evade immune rejection in fully immunocompetent allogeneic recipients. Nat. Biotechnol. 37, 252–258 https://doi.org/10.1038/s41587-019-0016-3 (2019).Article CAS PubMed PubMed Central Google Scholar Hu, X. et al. Hypoimmune induced pluripotent stem cells survive long term in fully immunocompetent, allogeneic rhesus macaques. Nat. Biotechnol. 42, 413–423 https://doi.org/10.1038/s41587-023-01784-x (2024).Article CAS PubMed Google Scholar Mansilla, M. J., Hilkens, C. M. U. & Martinez-Caceres, E. M. Challenges in tolerogenic dendritic cell therapy for autoimmune diseases: the route of administration. Immunother. Adv. 3, ltad012 https://doi.org/10.1093/immadv/ltad012 (2023).Article CAS PubMed PubMed Central Google Scholar Albani, S., Zhao, Y., Guo, D. & Chew, M. POS0459 PANTREG: a cell-therapy, system-biology - based approach for therapy of autoimmunity. Ann. Rheum. Dis. 83, 1106 https://doi.org/10.1136/annrheumdis-2024-eular.2800 (2024).Albani, S., Tuckwell, J. E., Esparza, L., Carson, D. A. & Roudier, J. The susceptibility sequence to rheumatoid arthritis is a cross-reactive B cell epitope shared by the Escherichia coli heat shock protein dnaJ and the histocompatibility leukocyte antigen DRB10401 molecule. J. Clin. Invest. 89, 327–331 https://doi.org/10.1172/JCI115580 (1992).Article CAS PubMed PubMed Central Google Scholar Albani, S. et al. Positive selection in autoimmunity: abnormal immune responses to a bacterial DNAJ antigenic determinant in patients with early rheumatoid arthritis. Nat. Med. 1, 448–452 https://doi.org/10.1038/nm0595-448 (1995).Article CAS PubMed Google Scholar Prakken, B. J. et al. Epitope-specific immunotherapy induces immune deviation of proinflammatory T cells in rheumatoid arthritis. Proc. Natl. Acad. Sci. USA 101, 4228–4233 https://doi.org/10.1073/pnas.0400061101 (2004).Article CAS PubMed PubMed Central Google Scholar Akdis, C. A. & Akdis, M. Mechanisms of allergen-specific immunotherapy and immune tolerance to allergens. World Allergy Organ. J. 8, 17 https://doi.org/10.1186/s40413-015-0063-2 (2015).Chen, W. et al. A multicenter study benchmarking single-cell RNA sequencing technologies using reference samples. Nat. Biotechnol. 39, 1103–1114 https://doi.org/10.1038/s41587-020-00748-9 (2021).Article CAS PubMed Google Scholar Stuart, T. et al. Comprehensive integration of single-cell data. Cell 177, 1888–1902.e21 https://doi.org/10.1016/j.cell.2019.05.031 (2019).Regev, A. et al. The Human Cell Atlas. eLife 6, e27041 https://doi.org/10.7554/eLife.27041 (2017).Radi, Z. A. & Wynn, T. A. Opinion on immune tolerance therapeutic development. Toxicol. Pathol. 48, 712–717 https://doi.org/10.1177/0192623320948805 (2020).Article PubMed Google Scholar Requejo Cier, C. J., Valentini, N. & Lamarche, C. Unlocking the potential of Tregs: innovations in CAR technology. Front. Mol. Biosci. 10, 1267762 https://doi.org/10.3389/fmolb.2023.1267762 (2023).Article CAS PubMed PubMed Central Google Scholar Diop, M. P. & van der Stegen, S. J. C. The pluripotent path to immunotherapy. Exp. Hematol. 139, 104648 https://doi.org/10.1016/j.exphem.2024.104648 (2024).Article CAS PubMed Google Scholar Ivanisevic, T. & Sewduth, R. N. Multi-omics integration for the design of novel therapies and the identification of novel biomarkers. Proteomes 11, 34 https://doi.org/10.3390/proteomes11040034 (2023).Baião, A. R. et al. A technical review of multi-omics data integration methods: from classical statistical to deep generative approaches. Brief. Bioinform. 26, bbaf355 https://doi.org/10.1093/bib/bbaf355 (2025).Luecken, M. D. & Theis, F. J. Current best practices in single-cell RNA-seq analysis: a tutorial. Mol. Syst. Biol. 15, e8746 https://doi.org/10.15252/msb.20188746 (2019).Article PubMed PubMed Central Google Scholar Download referencesAcknowledgementsV.R.C. is supported by the SingHealth Duke-NUS Academic Medicine Research Grant AM/SU114/2025 (SRDUKAMR25S4). Grant support to S.A. from the Agency for Science, Technology and Research (A*STAR) Singapore’s H24J4a0019, H25J6a0065, and from Duke-NUS and SingHealth AMC core funding is gratefully acknowledged. This research is also supported by the Singapore Ministry of Health’s National Medical Research Council under its Centre Grant Programme (MOH-000988).Author informationAuthor notesThese authors contributed equally: Vasuki-Ranjani Chellamuthu, Yunqian Zhao.Authors and AffiliationsTranslational Immunology Institute, SingHealth/Duke-NUS Academic Medical Centre, Singapore, SingaporeVasuki-Ranjani Chellamuthu, Yunqian Zhao & Salvatore AlbaniDuke-NUS Medical School, Singapore, SingaporeSalvatore AlbaniAuthorsVasuki-Ranjani ChellamuthuView author publicationsSearch author on:PubMed Google ScholarYunqian ZhaoView author publicationsSearch author on:PubMed Google ScholarSalvatore AlbaniView author publicationsSearch author on:PubMed Google ScholarContributionsV.R.C. and Y.Z. contributed equally to this work. V.R.C. and Y.Z. performed the literature review, synthesised the evidence, and wrote the initial draft of the manuscript. S.A. conceived and structured the review, provided critical intellectual input, and revised the manuscript. All authors contributed to discussions, edited the manuscript, and approved the final version.Corresponding authorCorrespondence to Salvatore Albani.Ethics declarationsCompeting interestsThe authors declare no competing interests.Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Rights and permissionsOpen Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.Reprints and permissionsAbout this article