NSAIDs are the first-line treatment for AS, exerting their anti-inflammatory effects by inhibiting cyclooxygenase (COX), also known as prostaglandin endoperoxide synthase (PGHS-1 and PGHS-2) (39). These enzymes play an essential role in the biosynthesis of prostaglandins, which are key mediators of inflammation, pain, and fever. Nevertheless, despite their extensive clinical application, NSAIDs exhibit notable limitations. A recent report in Germany revealed that only 19.1% of AS patients achieved complete remission with NSAIDs (40). 30% of patients responded to NSAIDs, but many of them experienced side effects (8, 41). Long-term use of NSAIDs can induce adverse reactions in the cardiovascular, gastrointestinal, and renal systems (8, 41). Additionally, approximately one-third of patients are completely unresponsive or intolerant to NSAIDs, necessitating alternative treatment approaches (42). Consequently, bDMARDs such as TNF-α inhibitors and IL-17 inhibitors, along with Janus kinase inhibitors (JAKi), have been adopted as second-line therapies following NSAIDs failure (43).

csDMARDs are a class of drugs that can alleviate and improve symptoms in AS, including MTX, SSZ, and hydroxychloroquine (44). However, it typically takes several months to achieve therapeutic effects (44). MTX is an anti-metabolite that competitively inhibits dihydrofolate reductase, thereby interfering with DNA synthesis and modulating the expression of various cytokines (45). Patients receiving MTX should be regularly monitored for side effects through detailed questioning and frequent blood tests (46). SSZ exerts its effects by inhibiting the synthesis of prostaglandins (47). However, a recently published guideline recommends SSZ only for patients with persistent peripheral arthritis who are intolerant to or contraindicated for TNF inhibitors (48). In addition, the administration of csDMARDs at higher doses is associated with an increased risk of various adverse events, including gastrointestinal perforations, thromboembolism, and serious infections (49).

Surgical intervention may be considered for patients with AS in cases of severe spinal deformity, spinal fractures, or other significant complications when non-surgical treatments have failed. In AS, multi-level ankylosis compromises spinal stability, leading to fractures that are 3–4 times more prevalent than in the general population and predominantly affect the cervical spine or cervical-thoracic junction (84). Given the complexity, surgery is preferred over conservative treatment for better outcomes. However, it carries high risks of complications both peri-operatively and post-operatively (85). Other conventional physical therapies include cryotherapy, ultrasound therapy, electrotherapy, kinesiotherapy, and massage (86). Systematic physical activity is essential as it effectively mitigates the progression of AS. Nevertheless, physical therapy may have certain limitations, including the requirement for consistent effort and time commitment, varying effectiveness depending on individual conditions, and potentially high costs. There is an increasing emphasis on adopting a personalized and multidimensional approach to AS treatment, which integrates diverse therapeutic modalities. In light of these limitations, there is increasing interest in investigating alternative therapeutic approaches, such as cell therapy.

MSCs are multipotent adult stem cells derived from the mesoderm during early embryonic development, characterized by their self-renewal capacity and potential for multilineage differentiation ( Figure 2 ). Initially identified in bone marrow by Friedenstein et al., MSCs have since been isolated from various tissues, including umbilical cord, dental pulp, and adipose tissue (87, 88). The International Society for Cellular Therapy (ISCT) has established standardized criteria for the identification of MSCs, which include: (1) adherence to plastic in vitro; (2) expression of specific surface markers, such as CD105, CD90, and CD73, while lacking expression of CD45, CD34, CD14 or CD11a, CD79a or CD19, and HLA II molecules; and (3) the ability to differentiate into osteoblasts, chondrocytes, and adipocytes in vitro (88).

Beyond their differentiation potential, MSCs exhibit robust immunomodulatory functions, capable of modulating both innate and adaptive immune responses. They reduce the pro-inflammatory phenotype by directly or indirectly interacting with dendritic cells, macrophages, NK cells, B cells, and T cells (89). Notably, MSCs can adapt their polarization phenotypes in response to the local microenvironment, shifting between anti-inflammatory and pro-inflammatory states according to disease conditions. This adaptability makes MSCs a promising therapeutic candidate for autoimmune diseases, including AS, where the inflammatory milieu can be dynamically targeted.

MSCs are multipotent progenitor cells with the capacity to modulate immune responses and promote tissue repair through the secretion of soluble factors and direct cell-to-cell interactions. These cells exhibit potent immunosuppressive properties by secreting a variety of molecules, including indoleamine 2,3-dioxygenase (IDO), prostaglandin E2 (PGE2), hepatocyte growth factor (HGF), transforming growth factor-beta1 (TGF-β1), insulin-like growth factor-1 (IGF-1), nitric oxide (NO), heme oxygenase-1 (HO-1), cyclooxygenase-2 (COX-2), and IL-10 (90–92).

HLA-B27 is a well-established immunogenetic marker for AS, with the arthritogenic peptide hypothesis suggesting that abnormal antigen presentation to CD8⁺ T cells by HLA class I molecules triggers a specific immune response. MSCs have the ability to regulate T cell proliferation, differentiation, and activity, and can reduce the production of pro-inflammatory cytokines. MSCs can upregulate IDO expression in response to inflammatory cytokines, notably IFN-γ. IDO catalyzes the conversion of tryptophan to kynurenine, thereby inhibiting T cell proliferation through disruption of cellular protein synthesis (13, 93). Additionally, MSCs produce inducible nitric oxide synthase (iNOS), which induces macrophages to release NO, thereby suppressing T cell function (94). MSCs also inhibit the differentiation of Th17 cells, a subset of T cells implicated in the pathogenesis of AS. Huang et al. described the inhibitory effect of human umbilical cord-derived MSCs on T cells in patients with SpA (95). In co-culture with peripheral blood mononuclear cells (PBMCs), umbilical cord-derived MSCs significantly reduced the production of IL-17, showing potential for the treatment of SpA. Regulatory T cells (Tregs) are a subset of T cells with potent immunosuppressive functions, acting by suppressing effector T cells and mitigating inflammation-induced tissue damage. Both peripheral blood and synovial fluid examinations in AS patients have shown a reduced number of Tregs, which is positively correlated with lower FOXP3 expression levels (96, 97). Multiple studies have shown that MSCs induce Treg proliferation, a key mechanism by which they limit inflammation. For instance, bone marrow-derived MSCs promote the differentiation of CD4⁺ T cells into Tregs in co-culture with PBMCs, expressing high levels of CD25 and FOXP3 (98). Moreover, bone marrow-derived MSCs induce Treg proliferation through the secretion of TGF-β1 and interaction with macrophages (99). IDO is also implicated in MSC-induced Treg generation (100). MSCs can directly interact with T cells, exhibiting the most potent inhibitory effects on activated T cells through direct cell-to-cell contact (101). This interaction is further enhanced by the upregulation of intercellular adhesion molecule-1 (ICAM-1) and vascular cell adhesion molecule-1 (VCAM-1) in MSCs, which strengthens their engagement with T cells (101).

Monocytes and macrophages in AS can polarize into pro-inflammatory (M1) or anti-inflammatory (M2) phenotypes, a process closely related to active inflammation, tissue damage, and regenerative reconstruction. In late-stage AS patients, monocytes were significantly polarized into M2 macrophages, with the M2/M1 ratio positively correlated with structural lesion damage (mSASSS) and negatively correlated with inflammatory markers (ESR, CRP) and the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) (102). MSCs influence the polarization of macrophages, which may be caused by cell-to-cell contact mechanisms and soluble factors (such as IDO, PGE2, IL-10, and COX-2) (13). For example, MSCs inhibit the proliferation of M1 macrophages and activate the production of M2 macrophages through the activation of TNF-mediated COX-2 and TNF-stimulated gene 6 (TSG-6) (13). Our previous work also demonstrated that in a mouse spondylitis model, the injection of umbilical cord-derived MSCs reduced the levels of inflammatory cytokines (TNF-α and CCL-2) in the spleen and serum of mice (103).

NK cells are a critical component of the innate immune system. HLA-B27 is specifically recognized by the inhibitory receptor KIR3DL1 on NK cells, with a correlation between KIR receptor expression and AS activity (104). This suggests that NK cells play a significant role in AS pathogenesis. MSCs can regulate NK cell phenotype through cell-to-cell interactions or secretion of factors such as TGF-β1 and PGE2, inhibiting their proliferation, cytokine secretion, and cytotoxicity (105). MSCs also suppress IL-2-stimulated NK cell proliferation (106). Interestingly, MSCs secrete HLA-G5 and IFNγ, which inhibit NK cell cytotoxicity and innate immune responses while promoting Treg proliferation (90).

Dendritic cells (DCs) are key antigen-presenting cells that synthesize IL-23, a major pro-inflammatory cytokine in AS (26, 107). IL-23 induces the differentiation of lymph node T cells into pro-inflammatory Th17 cells and stimulates IL-23R⁺ lymphocytes in the sacroiliac joints to secrete IL-22, which in turn activates osteoblasts and leads to local bone formation (108, 109). MSCs inhibit the upregulation of antigen-presenting and co-stimulatory signals (CD1a, CD40, CD80, CD86, and HLA-DR) during DC differentiation and prevent the increase in CD40, CD86, and CD83 expression during DC maturation (110). Moreover, MSCs and their supernatants interfere with DC endocytosis, reducing their ability to secrete IL-12 and activate allogeneic T cells (110). Jiang et al. also proposed a similar view that MSCs can reduce the expression of CD83 on mature DCs, indicating that DCs have lost their mature characteristics (111). MSCs can also inhibit DC maturation stimulated by CSF and IL-4 through the secretion of PGE2 (111). Additionally, MSCs inhibit DC differentiation through the production of IL-10 and cell-to-cell contact (112).

HO represents a pathological condition defined by the ectopic formation of new bone tissue in soft tissues beyond the normal skeletal system, typically evidenced by the presence of osteoblasts and chondrocytes. HO is one of most significant pathological features of AS (113). In AS, HO is predominantly manifested in soft tissues such as spinal ligaments and tendons, where the appearance of chondrocytes leads to the development of new bone (113). This process commonly occurs in conjunction with the progression of inflammation and bone erosion observed in AS patients. It can lead to joint stiffness, spinal ankylosis, and spinal deformity, and may even result in the “folded person” phenomenon. Although inflammation has long been considered a trigger for HO in AS, existing AS treatments such as NSAIDs and TNFi can rapidly alleviate inflammation and pain, but they do not significantly prevent the progression of bone lesions in AS patients.

In recent years, the immunomodulatory and regenerative properties of MSCs have garnered significant attention, prompting the initiation of several clinical trials to explore their therapeutic potential for AS ( Figure 3 ). The earliest reported use of stem cells for AS was serendipitous: a patient with acute myeloid leukemia and AS experienced marked relief of AS symptoms and improved clinical indicators following peripheral blood stem cell transplantation (129). This patient remained symptom-free from AS for approximately 3 years post-transplantation, without the need for anti-TNF or NSAID therapy (129).

In 2013, the Wang group conducted a comprehensive study to evaluate the feasibility, safety, and efficacy of bone marrow-derived MSC therapy in 31 AS patients who were intolerant to NSAIDs (130). AS patients participating in this study received four intravenous infusions of MSCs on days 0, 7, 14, and 21, with each infusion containing 1×10^6 cells/kg. The results showed that the proportion of patients achieving ASAS20 response was 77.4% at week 4, 54.8% at week 12, and 32.3% at week 16, with a mean response duration of 7.1 weeks following the fourth infusion. The mean ASDAS-CRP score decreased from 3.6 ± 0.6 at baseline to 2.4 ± 0.5 at week 4, but increased to 3.2 ± 0.8 at week 20. MRI assessments revealed a mean total inflammatory extent (TIE) of 533,482.5 at baseline, which decreased to 480,692.3 at week 4 (p > 0.05) and further to 400,547.2 at week 20 (p < 0.05). No adverse reactions were reported. In 2017, the Li group explored the therapeutic effect of umbilical cord-derived MSCs on AS (131). In this study, umbilical cord-derived MSCs were administered via intravenous infusion to five patients with AS. The cell doses ranged from 1.2 to 3.5×10^6 cells/kg, and each patient received between 1 to 3 infusions. The study revealed that following treatment, both the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and the Bath Ankylosing Spondylitis Functional Index (BASFI) demonstrated significant reductions. Specifically, BASDAI decreased from a baseline of 4.686 ± 0.999 to 1.880 ± 1.499 at the 3-month follow-up (P=0.014), while BASFI declined from 42.000 ± 21.213 at baseline to 10.900 ± 13.585 at the 3-month follow-up (P=0.062). However, the Bath Ankylosing Spondylitis Metrological Index (BASMI) increased nonsignificantly (P=0.676). The erythrocyte sedimentation rate decreased in 3 patients, and the C-reactive protein level was significantly reduced in 1 patient. Overall, symptoms of AS improved in all patients. No serious adverse reactions were noted; however, mild transient fever occurred in three patients within 2–6 hours post intravenous administration. More recently, we conducted a clinical study (NCT05962762) further confirmed the safety and efficacy of umbilical cord-derived MSCs for AS treatment. Other ongoing trials (NCT01420432, NCT01709656, NCT02809781) continue to evaluate the therapeutic potential of MSC infusion for AS ( Table 1 ).

MSC therapy has demonstrated significant potential in improving clinical symptoms and alleviating pain in patients with AS, with a favorable safety profile. This emerging therapeutic strategy offers a promising alternative to current treatments, such as biologics and JAK inhibitors, which are often associated with notable safety concerns and high costs. The immunomodulatory and regenerative properties of MSCs, which include the secretion of soluble factors and direct interactions with immune cells, may address the underlying pathogenesis of AS more effectively, with fewer adverse effects. However, several challenges remain to be addressed. Future research should focus on optimizing MSC sourcing, dosing, and administration routes, as well as conducting well-designed clinical trials to further validate their efficacy and safety in AS. Continued research and larger-scale clinical trials are anticipated to provide valuable insights and drive the development of this innovative treatment strategy, ultimately offering new hope for patients with AS.

CD19 and B cell maturation antigen (BCMA) have emerged as key B-cell surface targets, demonstrating significant therapeutic potential in conditions such as systemic lupus erythematosus (SLE), idiopathic inflammatory myopathies, and systemic sclerosis (133). CD19 is expressed throughout multiple stages of B cell development, from pro-B cells to plasmablasts, but not in plasma cells (134). This widespread expression, coupled with CD19’s multifunctional role in B cell activation, maturation, and signaling, makes it an attractive target for B cell-directed therapies in autoimmune diseases such as SLE. Molecules like BCMA, CD38, and CD138 are predominantly expressed on plasma cells, with BCMA and CD38 also present on plasmablasts (134). This differential expression pattern allows therapeutic strategies to selectively target specific subsets or broader spectra of the B cell lineage, depending on the disease context and desired therapeutic effect.

Beyond B-cell targets, CAR T-cell therapies are being developed to directly target specific autoantibodies involved in autoimmune diseases. For instance, in pemphigus vulgaris, a skin disease characterized by autoantibodies against desmoglein 3 (Dsg3), anti-Dsg3 CAR T-cell therapy is currently undergoing clinical trials (135). Additionally, CAR T-cell therapies targeting cytokines are also in development, with a focus on modulating the inflammatory milieu in autoimmune diseases. Key targets include IL-23, which plays a critical role in mediating inflammatory responses (136). By targeting these cytokines, CAR T cells may potentially disrupt the pro-inflammatory signaling pathways, leading to reduced disease activity and improved clinical outcomes.

An innovative therapeutic strategy focuses on the precise elimination of pathogenic T-cell subsets that proliferate abnormally in specific autoimmune diseases. For example, targeting TRBV9⁺ T cells in AS aims to selectively eliminate pathogenic T cells while preserving normal immune cell populations (137). This strategy enhances treatment precision and reduces adverse effects on healthy cells, potentially improving the safety and efficacy of CAR T-cell therapy in autoimmune diseases. These advancements reflect the ongoing evolution of CAR T-cell therapy, moving beyond traditional cancer applications to address the complex immunopathology of autoimmune diseases. Future research is expected to identify additional targets and refine current strategies, thereby significantly broadening the therapeutic potential of CAR T-cell therapy in this field.

Preclinical and clinical studies have demonstrated the promising therapeutic potential of CAR T-cell therapy in various autoimmune diseases, including multiple sclerosis, type 1 diabetes, inflammatory bowel disease, SLE, and pemphigus vulgaris (138). A notable case reported by the Mougiakakos group involved a woman with severe refractory SLE (SELENA score: 16) and Class III/IV lupus nephritis who received anti-CD19 CAR T-cell therapy (139). Following fludarabine lymphodepletion and CAR T-cell infusion, significant clinical improvement was observed within five weeks, characterized by normalization of dsDNA autoantibody titers and complement levels (C3 and C4). The SLE disease activity index score decreased from 16 at baseline to 0 at follow-up, and no significant adverse reactions were reported. The research team subsequently administered CAR T-cell therapy to four additional patients with refractory SLE, all of whom achieved a low lupus disease activity state (LLDAS) and successfully discontinued all SLE-specific medications (https://doi.org/10.1136/annrheumdis-2022-eular.1120). In another clinical study conducted by the Zhang group, patients with SLE and stage IV diffuse large B-cell lymphoma (DLBCL) exhibited continuous relief from disease activity following the infusion of CAR T cells targeting CD19 and BCMA (139). Follow-up examinations confirmed effective B-cell depletion, with stable disease remission lasting up to 23 months. These findings are encouraging and suggest that CAR T-cell therapy may offer a novel treatment option for patients with autoimmune diseases. However, the potential risks associated with CAR T-cell therapy, such as cytokine release syndrome (CRS) and neurotoxicity, necessitate further investigation (140). Additionally, the high cost of CAR T-cell therapy limits its widespread application.

To address these challenges, advancements in preparation techniques and diversification of cell types are being explored. Recent studies have investigated the expression of CARs in alternative cell types, such as NK cells, macrophages, regulatory T cells (Tregs), and MSCs (138). NK cells, known for their MHC-independent cytotoxicity and high safety profile, present a promising avenue for developing allogeneic therapies aimed at targeting pathogenic immune cells (141). Macrophages can phagocytose specific antigens and promote inflammatory responses, while also cross-presenting antigens to activate T cells. In contrast to the direct cytotoxic mechanisms, the activation of Tregs or MSCs through CAR-mediated antigen stimulation leverages their immunomodulatory properties to regulate immune responses. Tregs can secrete immunosuppressive molecules such as TGF-β, IL-10, and IL-35, making them suitable candidates for treating autoimmune diseases and preventing organ transplant rejection by inhibiting excessive T-cell activation (15). The Fransson group utilized CAR technology to target myelin oligodendrocyte glycoprotein (MOG) and co-express FoxP3, resulting in the generation of antigen-specific CAR Tregs (142). These MOG-CAR Tregs demonstrated the ability to inhibit effector T-cell proliferation in vitro and alleviate symptoms in experimental autoimmune encephalomyelitis (EAE) mouse models by reducing pro-inflammatory cytokine levels. Moreover, the MacDonald group reported that allogeneic HLA-A2 antigen-specific CAR Tregs (A2-CAR Tregs) maintained high expression levels of FoxP3, CD25, and CTLA-4 in vitro, effectively preventing graft-versus-host disease (GVHD) in immunodeficient mouse models (143). Recently, the Sirpilla group demonstrated the therapeutic potential of CAR-MSCs in treating GVHD (144). Specifically, E-cadherin-targeted CAR-MSCs localized to colonic cells and improved symptoms and survival rates through the upregulation of immunosuppressive genes and cytokines.

Table 2 summarizes the main clinical progress of CAR-based cell therapies for the treatment of autoimmune diseases to date. CAR-based cell therapy has emerged as a revolutionary immunotherapy, achieving significant breakthroughs in the treatment of autoimmune diseases in recent years. These studies highlight the potential of CAR-based cell therapy to induce long-term remission and reduce disease activity in patients with severe autoimmune diseases. AS is an autoimmune disease characterized by immune system dysregulation. CAR-based cell therapy may offer new treatment opportunities for AS patients by targeting abnormal immune cells. However, the application of CAR-based cell therapy for AS is still in the research and exploration stage and has not yet reached a mature stage for clinical application.

In contrast to SLE, which is primarily driven by pathogenic B cells, AS is characterized by dysregulated T cell activation (145). In recent years, significant progress has been made in CD7- and CD5-targeted CAR-T cell therapy for T-cell malignancies (146, 147). However, the efficacy of these approaches in AS remains to be demonstrated. Considering the widespread distribution and critical role of T cell antigens in normal tissues, the design of CAR-T cell therapy for AS should emphasize precision to minimize potential off-target effects and preserve the integrity of the immune system. Pathogenic T cells, such as TRBV9⁺ T cells as reported, represent promising candidates for therapeutic targeting (20). Targeting these specific T cells may offer a more refined strategy for AS treatment, thereby minimizing the risk of off-target effects.

Utilizing CAR-T cells to target and eliminate pathogenic cells represents one potential therapeutic strategy for AS. Another approach involves harnessing immune regulatory cells to precisely modulate the immune microenvironment in AS. Given the limited accessibility of the disease site in AS, employing inflammation-suppressing cells such as Tregs and MSCs, with enhanced targeting capabilities, may also hold considerable promise for effectively treating AS. CAR-based therapies warrant further investigation in future studies. Leveraging CAR-based therapies to selectively eliminate the root causes of the disease while simultaneously modulating the excessive inflammatory microenvironment, without inducing significant systemic immune suppression, may offer a new generation of safe and effective therapies for curing AS.

Extensive preclinical and clinical studies have demonstrated that MSCs exhibit high safety and efficacy in treating AS. These cells play a pivotal role in modulating overactivated immune cells, reducing chronic inflammation and promoting tissue repair through their anti-inflammatory and regenerative properties. However, before wide application of MSC treatment to AS, several challenges must be addressed.

CAR-based cell therapies represent a highly specific and targeted treatment modality that aligns well with the complex pathophysiology of AS. However, several critical questions still require clarification.