Role of IL-6 and IL-6 targeted therapy in systemic lupus erythematosus

Abstract

Interleukin-6 (IL-6) is one of the cytokines implicated in murine and human SLE. Only a few small studies have investigated IL-6 inhibition in human SLE. Currently, there are no studies registered in clinicaltrials.gov to assess the IL-6 targeted therapy in SLE, yet its role in the future remains to be defined. This narrative review analyses these and potential areas of future studies with IL-6 targeted therapy in SLE.

Video Abstract

A video abstract is available for this article and can be viewed at https://doi.org/10.1093/rheumatology/kead416.

Introduction

SLE is a heterogeneous chronic autoimmune disease characterized by loss of immune tolerance, autoantibody formation, immune complex deposition and complement activation leading to multisystem inflammation and organ injury [1]. The pathogenesis of SLE is complex, involving various cell types, including plasmacytoid dendritic cells, neutrophils, T cells, B cells and cytokines including IL-6, IL-10, IL-12, IFN-γ and IL-2 [2]. The pleiotropic cytokine IL-6 is involved in immune regulation and tolerance. IL-6 deficiency has been noted in immunodeficiency and abnormal inflammatory response [3]. IL-6 inhibition strategies have been successfully used in various rheumatic disease management, including rheumatoid arthritis [4], systemic juvenile idiopathic arthritis [5], giant cell arteritis [6] and adult-onset Still’s disease [7, 8]; however, it remains an area of uncertain future in SLE [9]. IL-6 is of particular interest in SLE as it helps differentiate B cells into antibody-producing plasma cells [10, 11], and various antibodies often preceding the clinical diseases are implicated in SLE pathogenesis [12]. In addition, murine and human studies have shown a positive association between IL-6 and systemic lupus, including in lupus nephritis [13]. In this narrative review, we will introduce IL-6 and its role in murine and human lupus, and we will also review the evolving literature and discuss future directions about targeting the IL-6 pathway in SLE.

Signalling mechanism of IL-6

IL-6 was first identified as 184 AA containing soluble human B-cell differentiation factor (BCDF or BSF-2) that helped induce B cells’ final maturation into immunoglobulin-secreting plasma cells [14]. Various cell types, including T cells, B cells, monocytes, fibroblasts, keratinocytes, endothelial cells, mesangial cells, adipocytes and some tumour cells produce IL-6. Of these, T cells [15], B cells [16], monocytes [17] and endothelial cells [18] are also implicated in SLE pathogenies. IL-6 exerts its biological activities through two molecules: IL-6R and glycoprotein 130 (gp130). IL-6R is mainly expressed in haemopoietic cells, such as T cells, monocytes, activated B cells and neutrophils. IL-6 can bind to either a membrane-bound form of IL-6R or soluble IL-6R and cause a homodimerization of gp130. Homodimerization of gp130 initiates signal transduction via the JAK/STAT and the SHP-2/ERK MAPK pathways ( Fig. 1) [19, 20].

IL-6 has been implicated in T cell-dependent B cell activation and antibody production [21], decreased NK cell cytolytic function and amplification of a pro-inflammatory cytokine cascade [22]. During active inflammation, IL-6 also acts on the hepatocyte as a physiological target to regulate hepatic acute phase response [23]. IL-6 is responsible for most acute phase proteins at the time of infection or inflammation, including serum amyloid A, C-reactive protein, alpha 1-acid glycoprotein, alpha 1-antichymotrypsin, haptoglobin, alpha 1-antitrypsin, fibrinogen, complement component C3 and ceruloplasmin [24]. IL-6 also aids in the recruitment of cells like neutrophils to the inflammation site [25], haemopoiesis (reviewed in [20]) and angiogenesis [26].

IL-6 in murine models of lupus

The studies in murine models of SLE provide sound evidence of the direct role of IL-6 in lupus pathogenesis. Treatment with a monoclonal antibody against IL-6R has been shown to reduce anti-dsDNA antibody levels significantly, preserve glomerular function and structure [27], reduce proteinuria and prolong life [28]. Similarly, in another study using IL-6 deficient mice to study the effect of IL-6 on LN, Cash et al. showed MRL-Fas(lpr) IL-6 -/- mice had delayed onset of proteinuria, haematuria and improved survival rate compared with IL-6-intact control mice [29]. The absence of IL-6 resulted in a statistically significant reduction of infiltrating macrophages in the kidney, decreased renal IgG and C3 deposition, and decreased CD4+ and CD8+ lymphocytes. In another study, a lack of B cell-derived IL-6 abrogates spontaneous germinal centre formation in mouse SLE, resulting in the loss of class-switched autoantibodies and protection from systemic autoimmunity [30].

IL-6 in human lupus

Elevated IL-6 levels have been observed in the preclinical stage of lupus concurrently or before the first positive autoantibody [31]. Additionally, lupus patients have elevated serum IL-6 levels compared with healthy control [32]. Furthermore, active SLE patients have higher IL-6 levels than inactive SLE patients [32]. In addition, studies looking into the individual lupus manifestations and cytokine profile have noted a higher level of serum IL-6 in NPSLE [33], cognitive fatigue [34], SLE-Sjogren syndrome [35], anaemia [36], arthritis [37], cutaneous involvement [38] and in lupus patients with high cardiovascular risk factors [39]. Likewise, urinary IL-6 levels have been higher in patients with lupus nephritis [40, 41]. In addition, increased B-cell response to IL-6 has been reported in lupus patients [11], and IL-6 inhibition in established lupus patients helps restore these naive B cell and T cell balances [42]. However, there have been conflicting reports on whether IL-6 levels correlate with lupus disease activity [43, 44]. A recent meta-analysis of 1817 SLE patients and 874 healthy individuals evaluated this. The IL-6 level was higher in SLE than in healthy individuals and using the SLEDAI >4 as a definition of active lupus, IL-6 level correlated with the disease activity in lupus. However, when using SLEDAI-2K >4 to define the active disease, such an association between IL-6 level and lupus activity was not seen [45]. Additionally, IL-6 genetic polymorphism has been linked to SLE, particularly discoid lupus [46].

Anti-IL 6 studies in human lupus

Despite the broad use of IL-6 inhibition in other rheumatologic illnesses and strong evidence of the role of IL-6 in SLE pathogenesis both in murine lupus and human lupus, there is a relatively small volume of literature on the use of IL-6 inhibition in SLE.

In a Phase 1 study, Illei et al. assessed the safety and efficacy of tocilizumab, a humanized monoclonal antibody against the IL-6 receptor, among 16 lupus patients (LN or extrarenal lupus) with mild-moderate disease activity [47]. Patients received tocilizumab intravenously every other week for 12 weeks at three different doses, along with prednisone (94% of patients) and hydroxychloroquine (75% of patients). Among patients who received either 4 mg/kg or 8 mg/kg dose, five of nine patients had >45% decrease in anti-dsDNA antibody at week 14. Other antibodies (ANA, anti-SSA, anti-SSB and anti-cardiolipin antibodies) did not change with treatment. Among clinical manifestations, arthritis appeared to be the most responsive disease manifestation: during treatment, all seven patients with arthritis showed improvement, while complete arthritis resolution occurred in four of seven patients. On the other hand, proteinuria did not seem to respond in all five patients. A modest but significant decrease in systemic disease activity as measured by Systemic Lupus Activity Measure (SLAM) scores and modified Safety of Estrogens in SLE National Assessment (m-SELENA)-SLEDAI was observed. Improvement in SLAM scores was related to changes in ESR, fatigue and hematological parameters, whereas improved m-SELENA-SLEDAI was due to differences in arthritis and rash.

A second phase 1 study examined sirukumab, another human anti-IL-6 monoclonal antibody, in a two-part study of cutaneous lupus erythematosus (CLE) and SLE [48]. In part B, patients with mild-moderate SLE with primarily mucocutaneous and musculoskeletal manifestations got 10 mg/kg of sirukumab every two weeks for a total of four doses vs placebo (10 patients in sirukumab vs five patients in placebo). Among sirukumab-treated SLE patients, both ANA and anti-dsDNA titres decreased with treatment. The patient-reported outcome, the Short Form 36 (SF-36) physical component summary at weeks 10 and 22 and SF-36 mental component summary at week 10 showed improvements with sirukumab. Dermatologic Life Quality Index scores worsened in placebo groups while they remained stable in the sirukumab treatment group [48].

Rovin et al. conducted a Phase 1 trial in LN patients with 10 mg/kg sirukumab infusion [49]. Patients with Class III or IV LN who had completed induction therapy with MMF or CYC and had persistent disease activity while on MMF or AZA maintenance therapy were included [49]. In this multicentre, randomized, double‐blind, placebo‐controlled study, 25 patients were randomized in a 5:1 study design. Patients received either additional sirukumab (10 mg/kg IV) or a placebo every four weeks through week 24, while the follow-up was done up to week 40. Overall, there was worsening proteinuria in the placebo group by 43.3% compared with no change in proteinuria in the treatment group. However, the primary end point of reduction in proteinuria (defined as protein creatinine ratio reduction of <0.5 if the baseline ratio was equal or <3.0; improvement by > or equal to 50% if the baseline ratio was >3.0) was not achieved. Subset analysis of five sirukumab-treated patients showed ≥ 50% improvement in their protein creatinine ratio through week 28 [49]. Interpretations of this Phase 1 data have raised the possibility that the lack of improvement in proteinuria may reflect patient selection and irreversible structural damage rather than a lack of efficacy.

In 2017, Wallace et al. published a phase II trial evaluating the efficacy and safety of an anti-IL-6 monoclonal antibody (PF-04236921) in SLE [50]. A total of 183 patients with active disease (SLEDAI-2K score of ≥6 and the BILAG 2004 A disease in ≥1 organ system or BILAG B disease in ≥2 organ systems if no level A disease activity was present) were randomized to placebo or PF-04236921 10 mg, 50 mg or 200 mg, subcutaneously, every eight weeks with stable background therapy (placebo, n = 45; 10 mg, n = 45; 50 mg, n = 47; 200 mg, n = 46). SLE Responder Index (SRI-4; primary end point) and British Isles Lupus Assessment Group-based Composite Lupus Assessment (BICLA) were assessed at week 24. The 200 mg dose was discontinued due to safety concerns and was not included in the primary efficacy analysis. The majority (>70%) of study participants in each group were white, and most patients (>85%) had mucocutaneous and musculoskeletal involvement. Even though the primary end point was not achieved with any dose, the BICLA response rate was statistically significant for 10 mg (P = 0.026). This study also noted a drop in antibody levels with treatment, although it was only >10%. In this study, around 60% of patients had detectable anti-dsDNA, and an even smaller percentage of patients (<25%) had significantly elevated titres of >120IU/ml at baseline. The post hoc analysis identified enriched patients who would respond better to treatment. Enriched patients had more significant baseline disease activity [by the presence of one or more of the following characteristics: SLEDAI-2K score ≥10, corticosteroids ≥7.5 mg/day, anti-dsDNA ≥28 IU/ml or hypocomplementemia (C3 and C4)]. Among the enriched study group, the SRI-4 and BICLA response rates were significantly different with 10 mg vs placebo. The authors also evaluated the incidence of severe SLE flares as an outcome measure using SLE Flare Index (SFI) and BILAG. At week 24, the incidence of severe flares was significantly reduced with a 10 mg dose group (n = 0) and 50 mg (n = 2) combined vs placebo (n = 8) using SFI.

Safety profile of IL-6 inhibition in lupus

Safety profiles related to study agents using IL-6 inhibition in SLE, including the background therapy used, are summarized in Table 1. Common adverse events were infection and GI distress. Infections accounted for the most serious adverse events, with severe neutropenia observed in a few patients. At 200 mg dose, PF-04236921 had increased deaths reported, likely due to significant immunosuppression at higher doses.

Hypocomplementemia is seen with IL-6 inhibitors due to decreased complement production rather than consumption (due to complement pathway activation) [47]. Neutropenia [51, 52] and thrombocytopenia [52] can occur with IL-6 inhibitors. However, these adverse effects (hypocomplementemia, neutropenia, thrombocytopenia) will pose a challenge to distinguish from active SLE/SLE flare. Dyslipidaemia (increased triglycerides with a minor transient increase in total cholesterol) occurs with IL-6 inhibition. While long-term cardiovascular (CV) risk implications of dyslipidaemia in SLE patients (who are already at heightened CV risk) cannot be answered by these short-term studies, tocilizumab, when compared with etanercept did not have an increased incidence of major atherosclerotic cardiovascular events in rheumatoid arthritis (also a patient population with increased CV risk) [53].

Despite these limitations, history of diverticulitis and advanced age, the risk factors for GI perforation associated with IL-6 inhibition [54] are not common in SLE, where patients are mostly young.

Future directions

Currently, there are no studies registered on clinicaltrials.gov assessing the role of IL-6 inhibition in SLE. This perhaps is due to lack of clear success of the above studies that failed to replicate the findings of murine studies. However, the objective improvement and protective effect of IL-6 pathway inhibition in murine studies may not have been replicated in the above human studies due to the heterogeneous nature of human SLE. The mouse models of SLE are genetically homogeneous for disease initiation and progression, and likely represent only the subset of heterogeneous human SLE [55, 56]. Hence, the previously discussed human studies do not provide enough information to put an end to the question of what role IL-6 inhibition has in SLE management. If at all, these studies indicate the need to select a homogeneous population for future studies. All seven patients with arthritis showed improvement, with complete arthritis resolution occurring in four of the seven patients [47]. When >85% of patients had mucocutaneous or musculoskeletal manifestations, BICLA response was better with a 10 mg dose of tocilizumab [50]. Checking the IL-6 level to guide the patient selection or selecting musculoskeletal manifestation of SLE may be a few strategies. Similarly, proteinuria was worse in the placebo than in the intervention arm, which had no proteinuria change [49]. This may point to the renal protective effect (without the reversal). Thus, targeting IL-6 in SLE patients without LN to prevent LN may be another important strategy. An additional area of interest with anti-IL-6 agents in SLE may be in association with the metabolic syndrome that has been well recognized in SLE [57]. IL-6 is one of the inflammatory cytokines implicated in chronic inflammation leading to metabolic syndrome [58]. In addition, SLE patients also receive a high cumulative dose of glucocorticoids over the years, and glucocorticoid itself is associated with obesity and insulin resistance, a key component of metabolic syndrome [59]. Both glucocorticoid exposure and metabolic syndrome have been shown to predict the damage accrual in SLE [60, 61]. Furthermore, as cardiovascular (CV) disease is the main driver of mortality in established SLE, and CV morbidity is directly related to metabolic syndrome and glucocorticoid exposure, anti-IL-6 therapy would be an attractive preventative strategy. Direct evidence of improved cardiovascular outcomes with the IL-6 pathway alternation in the experimental model is also emerging [62]. Putting together the possible renal protective effect and decreased CV events with IL-6 inhibition and recruiting newly diagnosed patients without any end organ damage to assess the clinical outcome of the progression of renal disease or cardiovascular outcome will be an exciting opportunity for research in SLE. SLE patients without significant comorbidities who are typically only on hydroxychloroquine will make it easier for patient recruitment and comparative effectiveness research in SLE while minimizing adverse effects that occur when often combined with additional immunosuppression when studies involve patients with severe SLE. Additionally, as IL-6 is implicated in osteoporosis [63], the anti-osteoclastogenic effect of anti-IL-6 therapy and the glucocorticoid-sparing effect will help minimize osteoporosis and osteonecrosis. This becomes especially important since there is no good way to assess osteoporosis among young individuals. Even in established patients with osteoporotic fractures, bisphosphonate is not indicated if patients have childbearing potential [64]. On the other hand, tocilizumab can be used in patients with childbearing potential, with the recommendation to stop it once pregnancy is confirmed [65]. These, along with the cost advantage of an established drug like tocilizumab, provide the added advantage of an anti-IL-6 strategy over Blys inhibitors, interferon inhibitors, and ciclosporin inhibitors. In summary, the anti-IL-6 strategy still has a great deal of potential in SLE management and is summarized in Table 2.

Conclusion

Il-6, a pleiotropic cytokine, has been implicated in lupus pathogenesis by both murine and human studies. Therapy targeted against IL-6 has shown promising results in murine experiments and mixed results in limited human studies. Despite the promise, it remains to be seen if/when therapy targeted against IL-6/IL-6R translates into routine clinical practice in the care of lupus patients.

Data availability

No new data were generated or analysed in coming up with this manuscript. Available data from cited references were used where appropriate.

Funding

No specific funding was received from any bodies in the public, commercial or not-for-profit sectors to carry out the work described in this article.

Disclosure statement: The authors have declared no conflicts of interest.

Acknowledgement

The authors would like to acknowledge Michael Putman, MD, MSCi, Assistant Professor of Medicine, Medical College of Wisconsin, for reviewing the manuscript.

References