Updates on the Evaluation of Lupus Arthritis

Abstract

Lupus arthritis is the most common manifestation of systemic lupus erythematosus (SLE), affecting up to 95% of patients. Clinical presentations range from inflammatory arthralgia and non-deforming non-erosive (NDNE) synovitis to Jaccoud arthropathy and erosive arthritis, reflecting substantial clinical heterogeneity. Despite its high prevalence, widely used outcome measures lack the granularity to distinguish between these phenotypes or accurately quantify disease severity. This narrative review integrates clinical phenotypes, serologic biomarkers, imaging advances, and emerging therapies to provide a framework for evaluating lupus arthritis. Besides distinct clinical phenotypes, we propose a conceptual hypothesis-generating model describing the evolution of lupus arthritis. In an early, preclinical stage, genetically predisposed individuals may develop pathogenic autoantibodies and alterations in the gut microbiome that can lead to increased inflammation within the synovium. The intermediate stage is characterized by progressive leukocyte accumulation within the joint space, leading to clinical symptoms and early abnormalities on imaging including joint effusions, synovitis, and tenosynovitis. In the late stage, persistent inflammation results in clinically evident joint damage, including synovial hypertrophy, joint capsular swelling, and bone erosions. Biomarkers such as anti-citrullinated protein antibodies (ACPAs), anti-carbamylated protein (anti-carP) antibodies, interleukin (IL)-6, IL-17, and metalloproteinases may help identify patients at risk for erosive or deforming disease and those prone to rapid progression. Advances in imaging technologies including musculoskeletal ultrasound (MSK-US), MRI, and optical tomography may enable earlier detection before clinical manifestations or physical exam findings, creating opportunities for earlier intervention. Therapeutically, hydroxychloroquine (HCQ) remains the foundation of treatment. Belimumab and anifrolumab are US Food and Drug Administration (FDA)-approved biologics now recommended as standard of care for persistent disease, alongside conventional immunosuppressants such as methotrexate, mycophenolate, or azathioprine when indicated. Additional targeted therapies including obinutuzumab and Janus kinase (JAK) inhibitors are closely following behind. Chimeric antigen receptor T cell (CAR-T) represents a promising approach that may transform the future management of SLE.

Key Summary Points

Arthritis is the most common manifestation of SLE, has a major impact on quality of life, and displays marked clinical heterogeneity ranging from inflammatory arthralgia to erosive and deforming arthritis.

Lupus arthritis may evolve through three stages: an early preclinical stage characterized by inflammatory transcriptional changes in the synovium driven by autoantibodies, microbiome alterations, and environmental triggers among genetically predisposed individuals; an intermediate stage with inflammatory symptoms and imaging abnormalities such as effusions, synovitis, and tenosynovitis; a late stage marked by evident findings on physical exam and imaging including synovial hypertrophy, erosions, and deformities.

Serologic biomarkers including ACPAs, anti-carP antibodies, IL-6, IL-17, and metalloproteinases may help identify patients at risk of developing erosive or deforming disease or progressing rapidly to advanced stages.

Advances in imaging modalities such as MSK-US, MRI, and optical tomography may detect joint inflammation and structural changes during early and intermediate stages before abnormalities become evident on physical exam.

While hydroxychloroquine remains the foundation of treatment, patients with persistent disease may benefit from conventional immunosuppressants and FDA-approved biologics such as belimumab and anifrolumab, with treatment individualized according to the clinical context. Additional therapies, including obinutuzumab and emerging cellular strategies such as CAR-T and T cell engagers, represent promising future therapeutic approaches.

Introduction

Arthritis is the most frequent manifestation of systemic lupus erythematosus (SLE), affecting up to 95% of patients, and has wide phenotypical heterogenicity, which complicates clinical management and assessment of efficacy outcomes in clinical trials [1]. Despite advances in imaging, biomarkers, and targeted therapies, important gaps persist in phenotyping, disease staging, and outcome measurement for lupus arthritis. This narrative review provides a comprehensive overview of the current evaluation of lupus arthritis by reviewing its clinical phenotypes, biomarkers, synovial pathology, imaging modalities, and therapeutic approaches.

This narrative review was based on a literature search of PubMed/MEDLINE for studies published through January 2026 using combinations of the terms systemic lupus erythematosus, lupus arthritis, arthritis, arthralgia, musculoskeletal ultrasound, MRI, biomarkers, synovial biopsy, biologics, and cellular therapy. Original studies, randomized clinical trials, observational studies, systematic reviews, and relevant guidelines published in English were considered. Additional articles were identified through reference lists of relevant publications and the authors’ expertise in the field. Given the narrative nature of this review, studies were selected on the basis of their relevance to the clinical evaluation, pathogenesis, imaging, biomarkers, and treatment of lupus arthritis.

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Clinical Phenotypes of Lupus Arthritis

The initial classification by Van Vugt et al. divided lupus arthritis into five phenotypes: arthralgia without objective evidence of arthritis on joint examination, synovitis (rheumatoid arthritis (RA)-like), Jaccoud’s arthropathy (JA), non-deforming non-erosive (NDNE) synovitis, and erosive arthritis [2]. Of note, erosive arthritis may represent either true lupus arthritis or an overlap syndrome with rheumatoid arthritis, commonly known as “rhupus.” Rhupus is typically characterized by persistent erosive polyarthritis with serologic features such as rheumatoid factor and/or anti-citrullinated protein antibodies (ACPAs), making its distinction from erosive lupus arthritis clinically relevant [3]. However, the distinction between these phenotypes is still not consistently defined in the literature, and SLE activity measures commonly used in clinical trials and research studies lack the granularity to capture the nuances of different lupus arthritis phenotypes. The SLEDAI-2K only defines arthritis if two or more joints are involved and does not consider the number of joints affected, type, or severity of arthritis into the score [4–6]. The BILAG 2004 index includes arthralgia within joint involvement across severity categories ranging from mild arthritis/arthralgia/myalgia to moderate arthritis/tendonitis/tenosynovitis, and severe arthritis, but offers limited guidance for distinguishing between these clinical phenotypes in practice [7, 8]. The Lupus Foundation of America Rapid Evaluation of Activity in Lupus (LFA-REAL) moved away from relying on glossary-defined thresholds such as in SLEDAI and BILAG to using a continuous visual analog scale (0–100 mm), which allows one to detect more granular fluctuations in severity and enables tracking of individual symptom progression and response to treatment [5, 6, 9, 10]. More recently, clinical trials have included tender and swollen joint counts as outcome measures, in an attempt to provide a more objective assessment of arthritis but fail to provide much guidance on how to evaluate this measure.

The rapidly growing application of advanced imaging technologies, including musculoskeletal ultrasound (MSK-US), MRI and optical imaging have impacted the way clinicians classify lupus arthritis. Vital et al. proposed a composite measure of arthritis (LAMDA—Lupus Arthritis and Musculoskeletal Disease Activity Instrument) that includes the swollen joints count, physician global assessment (PGA), patient global assessment (PtGA), and erythrocyte sediment rate (ESR); however, it is unclear whether this provides an advantage over the tender and swollen joint counts alone [11]. Di Matteo et al. proposed incorporating MSK-US findings into the evaluation of lupus arthritis to reflect clinical data [12]. The authors propose categorizing lupus arthritis into RA-like, NDNE synovitis with dominant synovial hypertrophy, NDNE synovitis with dominant hypoechoic fluid collections, JA, enthesitis with or without proximal tendinopathy, psoriatic arthritis (PsA)-like, and tenosynovitis with or without arthritis. The extent to which MRI and optical imaging might contribute and shape the classification of lupus arthritis is still being studied.

Serological Biomarkers for Lupus Arthritis

While multiple potential biomarkers for SLE activity have been investigated, most studies have focused on lupus nephritis or other major organ-threatening manifestations [13]. In contrast, relatively few publications specifically address lupus arthritis. Among these, the majority looked at the erosive arthritis phenotype with features overlapping with RA. ACPAs, well-known serological markers for RA, have been reported as highly specific biomarkers for erosive arthritis in SLE. A 2014 systematic review and meta-analysis including a total of 609 patients with lupus arthritis reported a pooled sensitivity of 47.8% (95% CI 26.2–70.2%) and specificity of 91.8% (95% CI 78.4–97.2%) in detection of erosive damage in lupus arthritis. Similarly, anti-CarP antibodies, a biomarker initially identified in RA, was also reported to be associated with erosive damage in lupus detected by X-ray [14] and ultrasound [15, 16]. One proposed mechanism involves synovial neutrophils releasing neutrophil extracellular traps (NETs), which may expose carbamylated autoantigens in the extracellular space. Formation of immune complexes between anti-CarP antibodies and their target antigens within the joint may then promote release of several pro-osteoclastogenic cytokines including receptor activator of nuclear factor κB ligand (RANKL), macrophage colony-stimulating factor (M-CSF), tumor necrosis factor alpha (TNFα), and interleukin (IL)-1 which ultimately promote osteoclast activation and events that favor bone erosion [17]. However, a considerable proportion of erosive arthritis in SLE is not associated with ACPAs, anti-CarP antibodies, or rheumatoid factor (RF), suggesting other pathogenic factors contribute to the development of erosive damage in SLE. Although several promising biomarkers have been identified, the available evidence is limited by the small sample sizes, substantial heterogeneity in study design, patient populations, biomarker assessment, and outcome measures. Given these limitations, the current evidence supporting biomarkers in lupus arthritis remains largely exploratory. Future prospective studies specifically designed to evaluate arthritis-associated biomarkers using validated clinical and imaging outcomes are needed before their routine clinical application can be established.

Several studies evaluated biomarkers for the JA subgroup. A 2011 study reported significant association between plasma IL-6 levels and joint deformities including both erosions and JA in patients with SLE [18]. Another 2014 study including a total of 50 patients with SLE further classified them into erosive arthritis, non-erosive arthritis, and no arthritis on the basis of the MSK-US findings, and identified a significant correlation between plasma IL-6 levels, JA hand deformities, and total ultrasound activity scores [19]. Sippl et al. [20] further analyzed T helper associated cytokine levels in acellular synovial fluids as compared to paired serum samples in 17 patients with lupus arthritis and showed significantly higher levels of IL-6 and IL-17A levels in SLE synovial fluids, suggesting their dominating role in local joint inflammation. Taken together, these studies indicated a potential therapeutic role of IL-6 and/or IL-17A blockade in the management of lupus arthritis. However, such expectations were not reflected in the clinical trials of IL-6 targeted therapy for SLE [21]. Data on the use of the anti-IL-17A secukinumab in SLE is accumulating; two studies in lupus nephritis (NCT05232864 and NCT04181762) and one in discoid lupus erythematosus are currently ongoing (NCT03866317).

Piga et al. assessed serum levels of metalloproteinases (MMP)-3 and MMP-12 among 50 patients with SLE with different arthritis phenotypes: rhupus, JA, and NDNE inflammatory joint involvement based on the presence of synovitis, hand deformities, and radiographic erosions. Patients with JA showed significantly higher MMP-3 and lower MMP-12 levels and featured MRI changes of edematous tenosynovitis and capsular swelling [22]. Ribeiro et al. investigated serum levels of CXCL13 and their association with MSK-US features in 64 patients with lupus arthritis. There was no association between levels of CXCL13 and SLEDAI arthritis, tender and swollen joint counts, or MSK-US synovitis; however, a significant association was found between high serum CXCL13 level and the presence of tenosynovitis on MSK-US [23].

Synovial Biopsy Investigation of Lupus Arthritis

Multiple studies have investigated the molecular pattern and gene expression profiles of lupus arthritis by examining the synovium and synovial fluid [24–26]. Nzeusseu Toukap et al. first performed global gene expression analyses of knee synovial tissue from untreated SLE (n = 6), RA (n = 7), and osteoarthritis (OA) (n = 6) patients with active knee involvement, and detected a characteristic upregulation of interferon-inducible (IFI) genes and downregulation of genes involved in extracellular matrix (ECM) homeostasis in SLE synovium [25]. The same group further compared high-density transcriptomic data obtained from the synovial biopsies from untreated patients with SLE (n = 4), RA (n = 7), OA (n = 5), SpA (n = 4), and crystalline arthritis (n = 5), and noted a discriminatory overexpression of IFI genes in SLE as compared to other types of arthritis, including IFI27, ISG15, RAD2, IFI6, IFIT3, and OAS1 [26].

Most recently, Hubbard et al. pivoted to proteomic and transcriptomic surveys of synovial tissues from patients with SLE (n = 4) and OA (n = 4) with active knee arthritis using weighted gene co-expression network analysis (WGCNA) [24]. Such transcriptomic profiles can be utilized to assign cell types and interestingly lupus arthritis demonstrated distinguishing enrichment of myeloid cell responses including local retention of CD11b+ macrophages in addition to neutrophils, granulocytes, activated T cells, and dendritic cells, a cellular pattern that supports leukocyte retention and accumulation (Fig. 1, highlighted in green). The study also demonstrated an association between lupus arthritis and T cell dominant injury as evidenced by the expansion of transcripts of the TNF superfamily of genes. The transcript profile of genes belonging to TNF superfamily differed between lupus arthritis and OA/RA with the majority of upregulated (n = 71) and downregulated (n = 25) transcripts not overlapping with OA/RA. Noticeably, BAFF was included in the subset of strongly upregulated genes. Expression of BAFF along with other members of the TNF superfamily in the context of a rich cellular milieu supports the amplification of autoimmunity and robust expression of drivers of autoantibody production in joints.

Fig. 1.

Hypothesis-generating conceptual model of lupus arthritis evolution across early, intermediate, and late stages

Proposed Evolution Model for Lupus Arthritis

Figure 1 proposes a conceptual, hypothesis-generating model for lupus arthritis to highlight the relationships of clinical and molecular phenotypes with joint swelling at early, intermediate, and late stages. The main findings from this proposed model are summarized in Table 1. The evaluation of biomarkers to reflect and differentiate between new onset and the different stages of lupus arthritis is limited. While it is unclear what constitutes the early stage of lupus arthritis, emerging evidence suggests that autoantibody development and environmental factors, including the gut microbiome, may contribute to its pathogenesis. As observed in RA, lupus autoantibodies, much like RF, ACPA and anti-CarPA, could bind to antigenic sites on osteoclasts and/or their precursors in genetically predisposed patients, stimulating the release of pro-inflammatory molecules with an autocrine enhancement of osteoclast maturation and activation and development of an inflammatory infiltrate. Additionally, immune complexes contribute to osteoclast activation by engaging Fc receptors. Ultimately, these steps result in erosive damage in lupus arthritis. Although these findings provide important insights into the pathogenesis of lupus arthritis, they remain largely mechanistic and should not be interpreted as clinically applicable tools for disease staging or treatment selection until validated in prospective clinical studies.

Table 1.

Main findings of the proposed conceptual model of lupus arthritis evolution across early, intermediate, and late stages

Figure element	Key finding to display	Main references
Gut microbiome pathobiont	Expansion of Streptococcus and Lachnoclostridium spp. associated with BAFF/sTNFRII signaling; gut dysbiosis may contribute to bone homeostasis and autoimmunity	Clancy et al. [27], Xiang et al. [31]
Autoantibodies	ACPAs and anti-CarP antibodies are associated with erosive lupus arthritis and osteoclast activation	Ziegelasch et al. [14], Massaro et al. [16]; O’Neil et al. [17]
Synovium transcriptomic profile	Synovium demonstrates an interferon gene signature and BAFF overexpression with enrichment of myeloid cell pathways	Hubbard et al. [24], Nzeusseu Toukap et al. [25], Lauwerys et al. [26]
T cell-mediated injury	Synovial tissue is enriched for activated T cells and TNF-superfamily transcripts	Hubbard et al. [24]
Metabolite/waste breakdown products	Impaired clearance of immune complexes and inflammatory mediators sustains synovial inflammation	Hubbard et al. [24] (conceptual model)
Leukocyte retention/accumulation	Local accumulation of macrophages, neutrophils, dendritic cells, and activated T cells	Hubbard et al. [24]
Bone abnormalities	Bone erosions associated with ACPA/anti-CarP antibodies and osteoclast activation	Ziegelasch et al. [14], Piga et al. [15], O’Neil et al. [17]
Intra-articular abnormalities	Synovitis, effusions, synovial hypertrophy, and capsular swelling detected by US/MRI	Piga et al. [22], Ossandon et al. [36], Gabba et al. [37]
Periarticular abnormalities	Tenosynovitis, tendinopathy, and enthesitis are common imaging findings.	Di Matteo et al. [12], Piga et al. [22], Gabba et al. [37]

ACPA anti-citrullinated protein antibody, anti-CarP anti-carbamylated protein antibody, BAFF B cell activating factor, MRI magnetic resonance imaging, sTNFRII soluble tumor necrosis factor receptor II, TNF tumour necrosis factor, US ultrasonography

Emerging evidence suggests that alterations in the gut microbiome and intestinal barrier play an important role in the development of SLE arthritis. Clancy et al. demonstrated an association between expansion of gut microbiome Streptococcus and Lachnochlostrium spp. and release of BAFF and soluble TNFRII, potentially mediated by mimicry to host proteins [27]. Enterococcus gallinarum has been shown to translocate across the intestinal barrier and promote TH17 polarization and anti-RNA autoantibody production [29]. More recently, increased serum zonulin levels, a marker of intestinal permeability, have been associated with gut dysbiosis, Tfh/Tfr imbalance, reduced IL-2 levels, and active SLE, suggesting that disruption of the intestinal mucosal barrier may further contribute to immune dysregulation and lupus arthritis pathogenesis [30]. Additionally, the gut pathobiont may release metabolites that modulate the complex biology of bone homeostasis, a mechanism described as “gut–joint axis” [31]. A recent systematic review and meta-analysis further supported the role of gut dysbiosis and increased intestinal permeability as shared mechanisms contributing to autoimmune disease pathogenesis, including SLE [32]. In patients with RA there is an association between gut microbiome and amplification of protein carbamylation, a process that may promote the formation of anti-CarP antibodies [28, 33].

The intermediate stage of lupus arthritis is characterized by progressive accumulation of leukocytes within the joint space, accompanied by impaired clearance of immune complexes and cellular breakdown products, and a T cell dominated inflammatory injury at the joint space. Although direct data in lupus arthritis remains limited, proteomic analyses in RA demonstrate increased expression of inflammatory and tissue remodeling pathways, supporting the concept that sustained immune activation and matrix turnover are central to joint progression [22, 20]. Clinically, these events start to manifest as stiffness, pain, and swelling in the joints, with imaging beginning to reveal effusions, synovitis, tenosynovitis, and erosive changes. However, ultrasound evidence of joint abnormalities does not always correlate with histologic injury or patient-reported symptoms, underscoring the complexity of disease assessment. Optical tomography, a novel imaging modality that assesses changes in tissue composition based on light absorption properties, has shown promise in differentiating healthy joints from those of patients with osteoarthritis, rheumatoid arthritis, and lupus arthritis [34, 35]. This technology could potentially identify the early and intermediate changes of lupus arthritis that can be missed by ultrasound assessment or clinical exam alone.

Late-stage lupus arthritis is characterized by joint abnormalities detected on clinical examination, MSK-US, MRI, and X-ray due to chronic joint inflammation. These abnormalities include intra-articular changes such as synovitis (synovial hypertrophy ± joint effusions), joint capsular swelling, and bone abnormalities (bone erosions, bone marrow edema, cysts) as well as periarticular changes such as tenosynovitis/tendinopathy and enthesitis [22, 36, 37]. The array of articular and periarticular involvement in late-stage lupus arthritis likely reflects heterogenous pathogenic mechanisms established during the early stage, which are subsequently amplified by multiple inflammatory pathways during the intermediary stage.

Therapeutic Approach to Lupus Arthritis

Hydroxychloroquine (HCQ) forms the foundation of treatment for lupus arthritis, often sufficient for inflammatory arthralgias alone. Although its precise mechanism of action is not fully elucidated, HCQ exerts its immunomodulatory effect in lupus activity through several key immunologic pathways. By accumulating within lysosomes of antigen-presenting cells, HCQ raises lysosomal pH, impairing antigen processing and presentation and thereby reducing T cell activation and downstream inflammatory cascades. It also inhibits Toll-like receptors (TLR) 7 and 9 from recognizing nucleic acid containing immune complexes, limiting type 1 interferon (IFN) signaling, a central driver of SLE pathogenesis. Additionally, B cell receptor signaling and production of other inflammatory cytokines such as IL-1, IL-6, and TNFα is also reduced [38–40]. The drug’s extensive tissue distribution and long half-life enable once-daily dosing and sustained immunomodulatory effects even with occasional missed doses [41]. Moreover, unlike traditional immunosuppressants (IS), HCQ is not associated with increased infection risk and may even be associated with a lower risk because of reduced glucocorticoid exposure [42, 43].

Alongside HCQ, acute episodes of inflammatory arthritis may be treated with short courses of non-steroidal anti-inflammatory drugs (NSAIDs) or a limited course of an oral glucocorticoid while waiting for long-term therapies to take effect. Persistent or recurrent active SLE arthritis on HCQ, regardless of prior or current NSAID or glucocorticoid therapy, warrants addition of another IS such as methotrexate (MTX), azathioprine (AZA), or mycophenolic acid analogues (MPAAs), with a low threshold to add or substitute with a biologic such as belimumab or anifrolumab [44]. The 2025 American College of Rheumatology (ACR) guidelines offer a strong recommendation to escalate therapy to a glucocorticoid-sparing agent when refractory to initial treatment, as early introduction of an IS and/or biologic minimizes glucocorticoid toxicity [44]. We have incorporated drugs and their targets into the model in Fig. 1, along with attributes and predictions, which are shown at Table 2.

Table 2.

Therapeutic targets for lupus arthritis

Agent	Status	Target	Mechanism of action	Strength of evidence
Hydroxychloroquine	SoC	APCs, B cells	Raises lysosomal pH, impairing antigen processing, inhibits TLR7 and TLR9 signaling	Guideline supported
Methotrexate	SoC	Lymphocytes	Dihydrofolate reductase inhibitor → impaired purine synthesis → reduced lymphocyte proliferation	Guideline supported
Leflunomide	Alternative	Lymphocytes	Dihydroorotate dehydrogenase inhibitor → inhibits de novo pyrimidine synthesis → reduced lymphocyte proliferation	Guideline supported
Azathioprine	SoC	Lymphocytes	Purine analogue → impaired purine synthesis → reduced lymphocyte proliferation	Guideline supported
Mycophenolate	SoC	Lymphocytes	IMP dehydrogenase inhibitor → impaired de novo purine synthesis → reduced lymphocyte proliferation	Guideline supported
Belimumab	FDA-approved	B cells	Anti-BAFF monoclonal antibody → inhibits B cell activation, proliferation, and survival	Guideline supported
Anifrolumab	FDA-approved	Type I Interferon pathway	Anti-IFNAR1 monoclonal antibody → blocks type I interferon signaling	Guideline supported
Obinutuzumab	Phase 3	B cells	Anti-CD20 monoclonal antibody → B cell depletion	Guideline supported
Baricitinib	Phase 3	JAK signaling pathway	JAK inhibitor → blocks signaling of several cytokines including interferons, IL-6, IL-12, and IL-23	Investigational
Deucravaricitinib	Phase 3	JAK signaling pathway	TYK2 inhibitor → blocks signaling of several cytokines including interferons, IL-6, IL-12, and IL-23	Investigational
Cenerimod	Phase 2	Lymphocytes	S1P receptor modulator → blocks lymphocyte egress from lymphoid organs	Investigational
Litifilimab	Phase 2	Plasmacytoid dendritic cells	Anti-BDCA2 monoclonal antibody → inhibits type I interferon production	Investigational
Iberdomide	Phase 2	Ikaros and Aiolos transcription factors	Cereblon modulator → promotes ubiquitination of Ikaros and Aiolos → multiple immunomodulatory effects including ↓type I interferon, B cell differentiation	Investigational
CAR T cell therapy	Phase 1/2	B cells, plasma cells	Genetically engineered T cells that target CD19+ B cells and/or BCMA+ plasma cells	Investigational
Bispecific T cell engagers	Phase 1/2	B cells, plasma cells	Bispecific antibodies that redirect CD3+ cytotoxic T cells to target CD19+ B cells or BCMA+ plasma cells	Investigational

SoC standard of care, APC antigen-presenting cell, FDA Food and Drug Administration, CAR chimeric antigen receptor, JAK Janus kinase, TYK2 tyrosine kinase 2, TLR Toll-like receptor, IMP inosine monophosphate, BAFF B cell activating factor, INFAR1 interferon alpha receptor 1, S1P sphingosine-1-phosphate receptor 1, BDCA1 blood dendritic cell antigen 2, BCMA B cell maturation antigen

Methotrexate was one of the first and preferred immunosuppressive agents for persistent lupus arthritis that fails to respond to HCQ alone [45, 46]. By competitively inhibiting dihydrofolate reductase, MTX disrupts purine nucleotide synthesis, thereby limiting lymphocyte proliferation [47]. A retrospective cohort study by Rahman et al. evaluating methotrexate for antimalarial-resistant lupus arthritis found that 15/17 patients treated with MTX showed ≥ 60% improvement in actively inflamed joint count at 6 months versus only 2/17 controls [48]. This study was followed by a double-blind randomized controlled trial (RCT) that compared MTX to placebo over 6 months, and found that by study end, only 1/18 patients treated with MTX had articular complaints versus 16/19 patients treated with placebo [49]. In clinical practice, MTX is generally introduced for HCQ-resistant arthritis in the absence of any active or previous major organ involvement.

Leflunomide inhibits dihydroorotate dehydrogenase, reducing de novo pyrimidine synthesis and thereby limiting activated T and B cell proliferation. Although evidence in SLE is more limited than for methotrexate or mycophenolate, small randomized and observational studies have demonstrated improvements in joint manifestations and overall disease activity. In a randomized study of 60 patients with SLE comparing leflunomide to cyclophosphamide, leflunomide was superior to cyclophosphamide in improving painful joint count, joint tenderness index, and joint swelling index [50]. Current guidelines support its use as an alternative conventional immunosuppressive for patients with persistent lupus arthritis, particularly when methotrexate is contraindicated or not tolerated [44].

Azathioprine, established as a second-line immunosuppressive agent for lupus arthritis, is metabolized to 6-mercaptopurine, a purine analogue that, when incorporated into DNA, exerts cytotoxic effects on rapidly dividing cells such as lymphocytes [51]. While it has been largely replaced by MTX as the preferred first-line immunosuppressive for musculoskeletal disease and by mycophenolate for many other SLE manifestations, AZA remains a valuable option for lupus arthritis without other organ involvement, particularly for pregnant patients or those with other contraindications for MTX use [44]. However, the evidence behind this is primarily derived from observational studies and extrapolation from trials of nonrenal SLE, as there are no dedicated RCTs specifically evaluating arthritis resolution as a primary endpoint.

Mycophenolate mofetil (MMF) inhibits inosine monophosphate dehydrogenase, the rate-limiting enzyme in de novo synthesis of purine nucleotides. This mechanism preferentially affects lymphocytes, which rely heavily on this pathway for proliferation [52]. While use of MMF is best established for lupus nephritis, accumulating evidence supports its efficacy for extrarenal manifestations [44]. Tselios et al. reported that among 72 patients with nonrenal disease, arthritis was resolved in 58.3% at 6 months and 62.5% at 12 months of treatment with MMF [53]. In a multicenter, 24-month RCT of 240 patients, in which more than 75% had arthritis, enteric-coated mycophenolate sodium achieved higher clinical remission rates than azathioprine at both 3 months (32.5% vs 19.2%) and 24 months (71.2% vs 48.3%) of therapy; however, arthritis-specific outcomes were not assessed [54].

Belimumab was the first FDA-approved biologic for the treatment of SLE in 2011 [55]. It is a fully humanized monoclonal antibody against B lymphocyte stimulator/B cell activating factor (BLyS/BAFF). By blocking BAFF, belimumab inhibits B cell activation, proliferation, and survival, thereby reducing autoantibody production and B cell-mediated inflammation and antigen presentation [56]. While belimumab has demonstrated efficacy across multiple organ domains, including lupus nephritis, musculoskeletal manifestations are among the most responsive domains [44]. Pooled data obtained from two phase III trials (BLISS-52 and BLISS-76) comparing belimumab 1 and 10 mg/kg versus placebo, plus standard therapy, found that among 1097 patients with active arthritis at baseline, those treated with belimumab 1 mg/kg had a significantly higher proportion of patients with improvements from baseline in SELENA-SLEDAI arthritis individual organ domain item score (58.3% vs 49.3%) after 52 weeks of treatment, compared to placebo [57]. Similarly, in a pooled analysis of five large RCTs evaluating belimumab (BLISS-52, BLISS-76, BLISS-NEA, BLISS-SC, and EMBRACE), including 2913 patients treated with either an approved dose of belimumab (n = 1769) or placebo (n = 1144), persistent resolution of lupus arthritis was observed in 55.9% (797/1427) of patients receiving belimumab plus standard therapy vs 47.8% (457/956) receiving placebo plus standard therapy (P < 0.0001) [58].

Anifrolumab, another FDA-approved biologic for moderate to severe SLE, is a monoclonal antibody that binds to subunit 1 of the type I interferon receptor (IFNAR1) with high specificity and affinity, thereby blocking type I interferon signaling [59]. A post hoc analysis of the phase IIb MUSE trial comparing anifrolumab 300 mg every 4 weeks with placebo showed that a greater proportion of patients treated with anifrolumab achieved improvement in arthritis by SLEDAI-2K (55/97 [56.7%] vs 42/99 [42.4%]) and BILAG (65/94 [69.1%] vs 47/95 [49.5%]) [60]. This was accompanied by a greater reduction in mean (SD) swollen and tender joint counts (− 5.5 [6.3] vs − 3.4 [5.9]). Similarly, in the phase 3 trial TULIP 1, a ≥ 50% reduction in swollen and tender joint counts from baseline was observed more frequently with anifrolumab than placebo (33/58 [47%] vs 22/68 [32%] patients [Δ = 14.7; 95% CI − 1.4 to 30.8; nominal p < 0.05]) [59, 61].

Obinutuzumab, a humanized type II anti-CD20 antibody, is approved for the treatment of lupus nephritis [62]. Unlike type I anti-CD20 antibodies such as rituximab, obinutuzumab induces more potent B cell depletion secondary to greater antibody-dependent cellular cytotoxicity, direct cell death, and antibody-dependent cellular phagocytosis. ALLEGORY, a recently published phase 3 RCT of patients with active SLE but without proliferative or membranous lupus nephritis, found that 76.7% of patients randomized to obinutuzumab 1000 mg achieved the primary endpoint of SRI-4 at week 52, compared to 53.5% in the group randomized to placebo (adjusted difference 23.1 percentage points; 95% CI 12.5–33.6; p < 0.001) [63]. While the ALLEGORY trial did not report arthritis-specific endpoints separately, over 93% of patients reported active musculoskeletal involvement at baseline. Improvements in arthritis were therefore indirectly captured through the composite endpoints used in the trial, including SRI-4, BICLA, and BILAG flare, which incorporate the musculoskeletal domains of the SLEDAI-2K and BILAG 2004 disease activity indices. Nevertheless, post hoc analyses specifically evaluating arthritis resolution would be valuable to better characterize the treatment effect in this domain.

Although these biologics have demonstrated consistent improvement in lupus arthritis across multiple clinical trials, a substantial proportion of patients fail to achieve meaningful response. Ongoing studies of agents with novel mechanisms of action and alternative pathway targets may have utility to limit the manifestations of lupus arthritis to an early stage and stop disease progression. Baricitinib is a Janus kinase (JAK) inhibitor, which has been approved for the treatment of moderate to severe RA in adults. JAKs are a family of tyrosine kinases that mediate the signaling of several pro-inflammatory cytokines, such as type l interferons, IL-6, IL-12, and IL-23, all of which are important cytokines implicated in the pathogenesis of SLE. The data supporting the use of JAK inhibitors in SLE is mixed. Baricitinib showed promise in a phase 2 trial of SLE where 4 mg daily achieved 67% resolution of arthritis or rash compared to 48% with placebo at week 24 [64]. The phase 3 trial, SLE-BRAVE-I, showed similar results where 59% of patients treated with baricitinib 4 mg daily had SLEDAI-2K remission of arthritis or rash compared to 48% in the placebo group (p = 0.019) [65]. However, in SLE-BRAVE-II, there was no difference in the primary efficacy outcome of the proportion of SRI-4 responders at week 52, and none of the major secondary endpoints were met [66]. Deucravacitinib, a selective TYK2 inhibitor, demonstrated efficacy in a phase 2 trial with 58% achieving SRI-4 response at week 32 with 3 mg twice daily versus 34% placebo (p < 0.001), while also showing improvements in BICLA, CLASI-50, and joint counts [67]. Two phase 3 trials (POETYK SLE-1 and POETYK SLE-2) are currently ongoing.

Other drugs that are currently in the pipeline for lupus arthritis include cenerimod, an S1P receptor inhibitor, which acts by restoring immune homeostasis through modulation of lymphocyte trafficking and by reversing the inflammatory environment within lymphatic vessels in lupus arthritis. S1P is a key regulator involved in the egress of lymphocytes from secondary lymphoid organs into the vascular circulation via the S1P receptor, which is highly expressed in endothelial cells and lymphocytes [68]. The S1P/S1PR signaling axis is particularly relevant to joint inflammation, as it plays a central role in regulating lymphatic endothelial barrier function within the synovium [69]. S1P receptor modulators block the movement of lymphocytes from lymphoid organs, preventing them from migrating to sites of inflammation [70]. Litifilimab, a monoclonal antibody against blood dendritic cell antigen 2 (BDCA2), a surface receptor expressed on plasmacytoid dendritic cells, thereby inhibiting type I interferon production and downstream inflammatory cascades, has shown efficacy in systemic [71] and cutaneous [72] lupus erythematosus [73, 74]. Iberdomide is a high affinity cereblon E3 ligase modulator that promotes ubiquitination and proteasomal degradation of transcription factors Ikaros (IKZF1) and Aiolos (IKZF3), which are implicated in genetic predisposition to SLE and are overexpressed in patients with the disease. Iberdomide demonstrated higher SRI-4 response over placebo in a phase 2 trial, but no significant differences were observed in swollen or tender joint counts between groups [75].

CAR-T cell therapy has recently emerged as an exciting investigational strategy for refractory SLE. In a recent systematic review, CAR-T therapy demonstrated striking clinical efficacy, with approximately 70% of treated patients achieving DORIS remission and nearly 90% reaching lupus low disease activity state (LLDAS) among the patients where these outcomes were assessed [76]. However, arthritis-specific outcomes have not yet been systematically evaluated in published CAR-T cohorts, and the degree to which this approach modifies musculoskeletal manifestations of lupus remains to be defined.

The treatment landscape for lupus arthritis has expanded beyond traditional immunosuppressants. It should be noted that much of the available evidence for biologic and cellular therapies in lupus arthritis is derived from clinical trials evaluating overall SLE disease activity, with musculoskeletal manifestations often assessed as secondary endpoints, post hoc analysis, or as components of composite disease activity indices rather than as arthritis-specific outcomes. Therefore, findings from secondary or post hoc analyses should be interpreted with caution and warrant confirmation in prospective studies specifically designed to assess arthritis-specific outcomes.

While HCQ and MTX have been considered first-line agents, belimumab and anifrolumab provide FDA-approved biologic options, and post hoc data from registrational programs suggest superiority over placebo. The investigational pipeline includes promising oral agents (deucravacitinib, baricitinib), novel biologics (obinutuzumab, litifilimab), and cellular therapies that target different pathogenic pathways. The heterogeneity of SLE suggests that personalized treatment selection based on biomarkers (such as interferon signature) and clinical phenotypes will become increasingly important.

Limitations

This review has several limitations. As a narrative review, it did not employ a systematic literature search or formal assessment of study quality or risk of bias. The available evidence is heterogeneous with respect to study design, patient populations, definitions of musculoskeletal involvement, and outcome measures. In addition, relatively few clinical trials have specifically evaluated lupus arthritis, with many therapeutic data derived from studies assessing overall SLE disease activity using composite indices rather than arthritis-specific outcomes. Finally, the proposed disease staging model is intended as a hypothesis-generating conceptual framework based on current mechanistic and clinical evidence and requires prospective validation before it can be incorporated into routine clinical practice.

Despite recent advances, several unmet needs remain in lupus arthritis research. Future research should focus on developing validated arthritis-specific outcome measures, standardized imaging-based classification systems, biomarker-guided therapeutic strategies, and clinical trials with musculoskeletal-specific primary endpoints.

Conclusion

Lupus arthritis is a highly prevalent yet understudied manifestation of SLE, with substantial phenotypic heterogeneity that is not adequately captured by current disease activity instruments. Integrating clinical assessment with serologic biomarkers, synovial transcriptomic and proteomic signatures, and advanced imaging modalities may enable more precise classification of arthritis subtypes and improve disease assessment. The proposed staging model provides a conceptual framework to integrate these findings but requires prospective validation before it can be incorporated into routine clinical practice.

Therapeutically, conventional immunomodulators and immunosuppressants remain foundational, while targeted biologics have expanded the treatment landscape for patients with persistent disease. Nonetheless, arthritis-specific evidence remains limited, with many therapeutic data derived from overall SLE clinical trials. Emerging pathway-directed therapies and immune reprogramming approaches, including CAR-T cell therapy, highlight a rapidly evolving field, but further studies with arthritis-specific outcomes and tissue-level correlates are needed to translate these advances into more precise and durable management of lupus arthritis.

Author Contributions

Wei Tang: conceptualization, methodology, software, investigation, formal analysis, data curation, visualization, validation, drafting the original manuscript, reviewing/editing the final draft. Alberto Nordmann-Gomes: conceptualization, methodology, software, investigation, formal analysis, data curation, visualization, validation, drafting the original manuscript, reviewing/editing the final draft. Leila Khalili: conceptualization, data curation, visualization, validation, reviewing/editing the final draft. Giovanna Rosas: conceptualization, methodology, software, investigation, data curation, visualization, validation, reviewing/editing the final draft. Ellen Montgomery: conceptualization, methodology, software, investigation, data curation, visualization, validation, reviewing/editing the final draft. Laurel Zhang: conceptualization, methodology, software, investigation, data curation, visualization, validation, reviewing/editing the final draft. Stephen Suh: conceptualization, methodology, software, investigation, data curation, visualization, validation, reviewing/editing the final draft. Robert Clancy: conceptualization, methodology, software, investigation, data curation, visualization, validation, drafting the original manuscript, reviewing/editing the final draft. Anca D. Askanase: conceptualization, methodology, software, investigation, data curation, visualization, validation, drafting the original manuscript, reviewing/editing the final draft.

Funding

Funding is acknowledged by Department of Defense, award numbers: W81XWH-21-1-0468 (ADA), W81XWH-22-1-0738 (ADA) and agencies at New York City Lupus Outreach and Clinical Trial Education Program, Tri-State Area Lupus Outreach and Clinical Trial Education Program. No funding was received for the publication of this article.

Data Availability

All data generated or analyzed during this study are included in this published article.

Declarations

Conflict of Interest

Wei Tang, Alberto Nordmann-Gomes, Leila Khalili, Giovana Rosas, Ellen Montgomery, Laurel Zhang, Stephen Suh, and Robert Clancy have nothing to disclose. Anca D Askanase has grants or contracts from AstraZeneca, BMS, Cabaletta, Idorsia, Genentech, NKARTA, Sana, Sanofi, and UCB; consulting fees from Abbvie, Amgen, AstraZeneca, Aurinia, Biogen, BMS, Cabaletta, Celgene, Eli Lilly, Idorsia, Janssen, Genentech, GSK, Mallinckrodt, NKARTA, Pfizer, Sana, Sanofi, and UCB. Anca Askanase is an Editorial Board member of Rheumatology and Therapy. Anca Askanase was not involved in the selection of peer reviewers for the manuscript nor any of the subsequent editorial decisions.

Ethical Approval

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.