Research and therapeutics—traditional and emerging therapies in systemic lupus erythematosus

Abstract

This review summarizes traditional and emerging therapies for SLE. Evidence suggests that the heterogeneity of SLE is a crucial aspect contributing to the failure of large clinical trials for new targeted therapies. A clearer understanding of the mechanisms driving disease pathogenesis combined with recent advances in medical science are predicted to enable accelerated progress towards improved SLE diagnosis and personalized approaches to treatment.

Rheumatology key messages

• New therapies for SLE target B cells, T cells and cytokines and block key signalling pathways

• Consideration of disease heterogeneity is predicted to improve SLE prognosis and treatment.

• Future treat-to-target trials might apply immune, genetic and clinical phenotypes to classify SLE patient groups.

Introduction

SLE is a lifelong disease, with complex underlying pathogenic mechanisms and no complete cure. A variety of immunological defects contribute to SLE, and the generation of autoantibodies has been associated with the pathogenesis of disease, which manifests in multiple organs. Current therapies are designed to manage inflammation, stave off relapses and reduce clinical symptoms, with the goal of preventing permanent organ injury. This brief review highlights immune mechanisms prevalent in traditional and some emerging targeted therapies for SLE. We also discuss how discovery of immunological defects has accelerated approaches to targeted therapies and impacted SLE diagnosis.

Traditional therapies

Before the era of clinical trials with predetermined end points and powered to detect statistical significance, the US Food and Drug Administration (FDA) in the 1950s approved three drugs for treatment of SLE: aspirin, CSs and HCQ [1]. To this day, all three agents retain a role as non-specific drugs for certain SLE features. CSs are one of the most effective therapies for immediately dampening inflammation; however, long-term use of CSs can cause serious side-effects [2, 3].

In order to develop new agents for lupus, SLE clinical trials awaited further standardization and development of methodology. The SLEDAI was introduced in 1984, and the BILAG index was published in 1993, followed by expansions and revisions, instruments for flares and damage and patient response measures. The FDA provided and revised guidance documents for SLE drug development in 2010 (http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm064981.htm).

Multiple synthetic drugs were evaluated for SLE over almost five decades (1950–1990s) in non-definitive reports, often with limited patient numbers, from single centres, with retrospective study design and using a large variety of non-validated disease activity and outcomes measures. In this way, the uses of AZA, MTX, nitrogen mustard, chlorambucil, CYC, CSA, apheresis and CSs were disseminated, although clear demonstrations of clinical efficacy were not achieved [1].

SLE clinical trial design improved markedly in the 1970s with the US National Institutes of Health’s landmark studies of LN. These studies demonstrated that i.v. CYC with oral CSs was superior to other regimens that included oral CYC and AZA [4]. CYC is an alkylating agent that crosslinks DNA and results in activated lymphocyte death and is protective to glomeruli [5, 6]. A meta-analysis demonstrated a decreased risk of end-stage renal disease (1970s to mid-1990s), which coincided with the implementation of CYC as standard-of-care therapy for LN [6]. CYC does have several potential side-effects, including leukopenia, infection risk, bladder toxicity and increased risk for malignancy [3, 7]. As a result, CYC is used clinically as an induction treatment for severe lupus and is replaced by agents such as mycophenolic acid or MMF or AZA for long-term maintenance therapy (Table 1).

Table 1.

Traditional therapies and supplements for SLE

Therapy	Agent type	Description	References
AZA	DMARD	Purine analogue, inhibits DNA, RNA and protein synthesis, mitosis	[1, 7–13]
HCQ	Anti-malarial, DMARD	Broad targets, raises pH of lysosomes, inhibits TLRs, antigen processing	[14–17]
CSs	Steroid hormones	Broad targets, suppress inflammatory mediators and immune cell activity	[1–4, 7, 18, 19]
CYC	DMARD	Alkylating agent, inhibits autoimmune T lymphocyte proliferation	[1, 4–11, 20, 21]
CSA	DMARD	Calcineurin inhibitor, inhibits autoimmune T lymphocyte proliferation	[7, 22–24]
LEF	DMARD	Pyrimidine synthesis inhibitor, inhibits dihydroorotase dehydrogenase	[1, 24, 25]
MTX	DMARD	Folate analogue, inhibits autoimmune T lymphocyte proliferation	[1, 24, 26, 27]
MMF	DMARD	Inhibits IMPDH, inhibits autoimmune T lymphocyte proliferation	[7–13, 20, 21, 28, 29]
Prasterone	Immunomodulator	Synthetic dehydroepiandrosterone	[30–32]
Tacrolimus	DMARD	FKBP inhibitor, inhibits autoimmune T lymphocyte proliferation	[1, 7, 18, 19, 24]
Vitamin D	Immunomodulator	Vitamin supplement, active form is 25-hydroxyvitamin D3	[33–39]

Traditional therapies and supplements for SLE along with the agent type, a description of each agent type, their targets and key references are shown. IMPDH: inosine monophosphate dehydrogenase; TLR: Toll-like receptor; FKBP: FK506 binding protein.

By applying the clinical trial tools developed over many years, older drugs have also received new consideration. For example, a double-blind, placebo-controlled trial in moderately active SLE patients showed that MTX was a steroid-sparing agent [26]. Likewise, an LEF study demonstrated that it significantly reduced SLEDAI scores [25]. Recent clinical trials have focused on management strategies for refractory SLE and LN incorporating standard therapies [7, 40–42].

MMF is an inhibitor of purine synthesis acting to inhibit lymphocyte proliferation and nitric oxide production by activated macrophages [28]. Recent reports suggest that when compared with CYC, MMF more rapidly reduced numbers of peripheral plasmablasts, activated B cells and levels of free immunoglobulin light chains, with these effects persisting over time [6]. Careful studies in LN have established the equivalence of MMF to i.v. CYC and its equivalence or superiority to AZA in the maintenance phase of treatment (Aspreva Lupus Management Study (ALMS) trial, Mycophenolate Mofetil versus Azathioprine for Maintenance Therapy of Lupus Nephritis (MAINTAIN) trial) [8–11, 20, 21].

In non-renal lupus, the evidence for MMF effectiveness is backed by fewer large studies. A systematic review of 20 case series and open-label trials suggested a benefit in patients with haematological manifestations or refractory dermatological involvement [12]. Merrill et al. [29] performed a double-blind, placebo-controlled trial of MMF vs placebo in 27 patients with lupus arthritis and achieved a statistically significant response in the MMF group. Finally, a post hoc analysis of mucocutaneous, musculoskeletal and haematological extrarenal manifestations in the ALMS trial found significant improvement in BILAG and Safety of Estrogens in Lupus Erythematosus National Assessment modification of the SLEDAI (SELENA-SLEDAI) measures in the MMF and the i.v. CYC groups [20]. MMF side-effects include gastrointestinal symptoms, bone marrow suppression, infection risk and long-term risk of neoplasia from immunosuppression. Overall, MMF is less toxic than CYC therapy, except in an Asian patient subgroup who have increased sensitivity to this agent when combined with high-dose CSs [12, 43, 44]. In cases of severe lupus involving major organs, such as cerebritis or nephritis, CYC therapy is favoured over MMF for its long track record and potency [7].

The use of the calcineurin inhibitors tacrolimus and CSA in lupus has built on the experience gained in organ transplantation. These agents have multiple mechanisms of actions, most notably in the inhibition of calcineurin’s serine/threonine specific phosphatase activity [22]. These drugs suppress production of cytokines, inhibit T- and B cell activation and preserve the renal podocyte actin cytoskeleton, which lessens proteinuria [23]. A recent summary of the literature on non-renal SLE suggests that several agents, including AZA, MTX, LEF, MMF and CSA, exhibited steroid-sparing effects as they reduced disease activity and flares in some cases [24]. Tacrolimus is preferred over CSA for LN because it has fewer side-effects and better long-term outcomes [7]. A combination of low-dose MMF, tacrolimus and CSs was the only induction therapy with a better short-term outcome than CYC in Chinese patients [7, 18, 19]. More studies are required to determine the most successful regimens for the treatment of LN subgroups.

HCQ remains a mainstay of SLE therapy, with potential benefits for dermatological manifestations, arthritis, preventing lupus flares, reducing thrombosis in APS, atherosclerotic risk and type II diabetes risk [14–16]. HCQ accumulates in lysosomes and autophagosomes, downregulating overactive responses in these compartments [17]. The immunological effects of HCQ include decreased HLA class II expression and antigen presentation, decreased production of pro-inflammatory cytokines (IL-1β and TNF), control of Toll-like receptor (TLR)9 activation and decreased generation of reactive oxygen species by immune cells [15, 17].

Recent efforts to reduce toxicity of traditional agents have employed supplementation with adjunct therapies. Prasterone and vitamin D represent two other immunomodulatory agents that are not immunosuppressive and may be used as supplements to control disease activity or reduce use of CSs. Prasterone is a synthetic form of the steroid hormone dehydroepiandrosterone [30]. SLE patients given oral prasterone tolerated taper of CS dosages better than controls [31], stabilizing SLE disease activity for patients on CSs [32]. Vitamin D has favourable actions on the vasculature, decreasing insulin resistance, decreasing pro-inflammatory cytokine production and down-regulating the renin–angiotensin system [33, 34]. Adequate levels of 25-hydroxy-vitamin-D were associated with improvements in measures of SLE disease activity [35, 36], whereas low levels of vitamin D were associated with increased risks for vascular disease in non-lupus patients [34]. Most but not all recent clinical studies reported that supplementation with vitamin D improved SLE disease activity [36–38]. Recent genotyping studies revealed that vitamin D deficiency combined with specific polymorphisms in CYP24A1, the gene for the cytochrome P450 enzyme that converts vitamin D to vitamin D₃, was associated with increased risk of patients transitioning to full-blown SLE [39].

Targeted therapies

Multiple clinical trials have incorporated biologics into treatment regimens for SLE patients with poor responses or major side-effects to standard therapies [1, 7, 40, 41]. Examples of targeted therapies in use and under study are presented in Table 2. Here we highlight some of the targeted therapies for SLE and insights gained from these studies. The original goal of biologics, such as treatment with B cell depleting mAbs, was to induce remission of disease and re-establish self-tolerance. No therapeutic regimen has achieved this goal. The heterogeneity of disease mechanisms inherent in SLE suggests that cell- and cytokine- or pathway-specific therapies will be most effective in subgroups of patients with evidence of alterations in targeted pathways, as illustrated in Fig. 1.

Table 2.

Targeted therapies for SLE

Targeted therapy	Agent type	Description	References	ClinicalTrials.gov identifiers
B Cell
Rituximab	Anti-CD20 mAb	B cell depletion	[40, 41, 45–52]	NCT00282347, NCT00137969, NCT02260934, NCT02284984
Ofatumumab	Anti-CD20 mAb	B cell depletion; shown efficacy in RA	[27]
Obinutuzumab	Anti-CD20 mAb	B cell depletion; in phase II trials for proliferative LN	[53]	NCT02550652
Ocrelizumab	Anti-CD20 mAb	B cell depletion; studies halted for SLE	[40, 54–58]	NCT00626197, NCT00539838
MEDI551	Anti-CD19 mAb	B cell and plasma cell depletion; spares Tregs	[59–61]
MDX-1342	Anti-CD19 mAb	B cell and plasma cell depletion		NCT00639834
Epratuzumab	Anti-CD22 mAb	B cell inhibition of activation and proliferation, reduces B cell numbers	[62–64]	NCT01261793, NCT01262365
Cytokine
Belimumab	Anti-CD257 (BAFF) mAb	Inhibits B cell survival and differentiation; FDA approved for SLE	[65–72]	NCT00424476, NCT00410384, NCT01639339, NCT02260934, NCT02284984
		Blocks soluble BAFF		NCT01632241 (EMBRACE), NCT00712933, NCT00583362, NCT01597622, NCT01729455, NCT02119156
Atacicept	TACI-Ig fusion protein	Inhibits B cell survival and growth factors; trials halted	[65, 70, 73–77]	NCT00624338, NCT00573157, NCT01972568, NCT02070978, NCT01369628, NCT01440231
Blisibimod	BAFF peptibody antagonist	Inhibits B cell survival and differentiation; blocks soluble and membrane BAFF	[40, 65, 72, 78–80]	NCT01395745, NCT02514967
Tabalumab	Anti-CD257 (BAFF) mAb	Inhibits B cell survival and growth factors; trials halted	[66–69]	NCT01205438, NCT01196091, NCT02041091 and NCT01488708
Sifalimumab, MEDI-545	Anti-IFNα	Blocks inflammation induced by type I IFN	[1, 41, 65, 81]	NCT01031836
Rontalizumab	Anti-IFNα	Blocks inflammation induced by type I IFN	[1, 41, 65, 82]	NCT00962832
AGS-009	Anti-IFNα	Blocks inflammation induced by type I IFN		NCT00960362
IFN kinoid	Immunogen	Human IFN-α2b-KLH immunogen induced anti-IFNα response	[1, 65, 83]	NCT02665364
Anifrolumab (MEDI-546)	Anti-IFNAR mab	Neutralizes type I IFN activity by receptor blockade	[1, 65, 84]	NCT01753193, NCT02446912, NCT02547922, NCT02446899, NCT02794285, NCT01438489, NCT01753193, NCT01559090
Ustekinumab	Anti-p40 IL-12/IL-23	p40 subunit of interleukin IL-12/IL-23		NCT02349061
Tocilizumab	Anti-IL-6R	Blocks IL-6 binding to the IL-6R to decrease inflammation	[1, 85]	NCT00046774
Sirukumab (CNTO1 36)	Anti-IL-6	Neutralizes IL-6 to decrease inflammation	[1, 86–88]	NCT01273389
Tregs
Low-dose IL-2	Recombinant cytokine	Induction and expansion of Tregs	[89–102]	NCT02084238
Costimulators
Abatacept	CTLA4-Fc fusion protein	Blocks CD28 binding to CD80/CD86, blocks T and B cell costimulation	[40, 41, 103–106]	NCT02270957, NCT02429934, NCT00119678, NCT01714817
CDP7657	Anti-CD154 mAb, CD40 ligand	Blocks T cell activation, Th cell activity	[107, 108]	NCT01093911, NCT01764594
Anti-CD275	Anti-CD275 mAb, ICOS ligand	Blocks T follicular helper cell activation/differentiation	[109]	NCT01683695, NCT02391259, NCT00774943
BIIB023	Anti-CD255 mAb, TWEAK	Blocks pleiotropic effects on mesenchymal and epithelial renal cells, cytokines		NCT01499355, NCT01930890
Anti-CXCR5	Anti-CXCR5 mAb, CD185	Blocks CXCR5 binding to CXCL13 and cell migration to LN and target organs		NCT02321709, NCT02331810
Eculizumab	Anti-C5 mAb	Blocks C5, prevents C5a, membrane attack complex formation (C5b–C9)	[110–120]
Other
Lupuzor	Possible altered peptide ligand	Blocks activation of autoreactive T cells, may induce apoptosis		NCT02504645
Stem Cells	Autologous adult stem cells	May induce immunosuppression of pro-inflammatory cells		NCT02741362

Targeted therapies for SLE along with the type of biologic, a description of the targets, references related to the biology of targets or clinical studies and clinical trial identifiers available through ClinicalTrials.gov are shown. BAFF: B Cell Activating Factor of the TNF family; TACI: Transmembrane Activator and CAML Interactor; CAML: calmodulin modulating ligand; ICOS: inducible T cell costimulator; TWEAK: TNF-like weak inducer of apoptosis; FDA: US Food and Drug Administration.

Fig. 1.

Biologic therapies and their targets in the immune system

Shown are some of the common cytokines, pathways and functions of the immune system (in blue) known to be dysregulated in SLE and targeted by biologic therapies (in red). Plasmacytoid dendritic cells release type I IFNs, such as IFNα, in response to Toll-like receptor (TLR)9 stimulation induced by immune complexes. In SLE, immune complexes may be formed by autoantibodies, such as anti-DNA antibodies complexed to self-DNA. IFN-primed neutrophils release DNA in the form of neutrophil extracellular traps (NETs; [121]) or oxidized mitochondrial DNA after activation by anti-Sm or anti-RNP autoantibodies, which trigger the Fcγ receptor, TLR7 and reactive oxygen species pathways [122]. Studies suggest that SLE disease activity is associated with type I IFN activity, which is elevated in the periphery of most SLE patients. Sifalimumab, a human anti-IFNα mAb, and rontalizumab, a humanized IgG1 anti-IFNα mAb, have been developed to target this pathway. BAFF (CD257, BLyS, TNFSF13B) is a key B cell maturation and survival factor that is produced predominantly by myeloid lineage cells. BAFF has been assessed as a therapeutic target, diagnostic marker and prognostic indicator for SLE. Therapeutic agents include a unique peptibody, blisibimod, which has four domains that bind to BAFF fused to a human IgG1 Fc domain, and belimumab, a humanized IgG1 mAb, which has achieved US Food and Drug Administration approval for the treatment of SLE. Both agents block BAFF from binding to its receptors. Other B cell-targeted therapies are directed at CD19, CD20 and CD22 cell surface proteins. Rituximab is a humanized anti-CD20 mAb used for B cell depletion therapy and might be most effective in combination with anti-CD257 (BAFF) therapies. MEDI-551 is a humanized IgG1 mAb to CD19. MEDI-551 is predicted to be a more effective B cell depletion therapy than rituximab because CD19 is expressed by plasmablasts and long-lived plasma cells, unlike CD20, and might spare B regulatory cells. Epratuzumab, a humanized IgG1 mAb, enhances the inhibitory function of CD22 on B cells. Epratuzumab, acting through CD22 (Siglec-2), reduced B cell activity as determined by a number of parameters, including decreased expression of cell surface proteins, such as adhesion molecules, reduced cytokine production (e.g. IL-6 and TNF) and diminished migration and cell growth. Abatacept is composed of a human cytotoxic T-lymphocyte-associated antigen-4 (CTLA4) protein joined to a human IgG1 Fc domain. Abatacept out-competes the CD28 receptor for binding to CD80 and CD86 ligands expressed by antigen-presenting myeloid cells, including B cells. CD28 is a key costimulatory molecule that delivers the ‘second signal’ required for antigen-induced T cell activation. Abatacept treatment can result in diminished T cell responses and T-dependent B cell antibody production. CD154 (CD40 ligand) is expressed on activated T cells and binds to the CD40 receptor on antigen-presenting cells. The CD154–CD40 interaction is important for T cell responses and T-dependent B cell responses. Thus, treatment with anti-CD154 can result in diminished T cell and humoral responses in an antigen-specific manner. IL-6 can be produced by a variety of cells and promotes inflammation and immune responses. Sirukumab, a human anti-IL-6 mAb, binds with high specificity and affinity to IL-6. Sirukumab prevents IL-6 from binding to the soluble and transmembrane forms of the IL-6 receptor, thus inhibiting IL-6-driven inflammation and immune responses. Tocilizumab, a humanized anti-IL-6 receptor mAb, likewise dampens IL-6-dependent responses, and it has been approved for the treatment of RA. Eculizumab is an anti-C5 mAb under study for SLE. Eculizumab may be beneficial for lupus nephritis in cases resistant to other therapies with evidence of C5 activation. This figure was adapted from Petri, M: Treat to Target(s)—the Lupus 12-Step Programme. A CME Programme for Rheumatologists 2015.

B cell-targeted therapies

One-tenth of registered lupus clinical trials are focused on B cells. Approaches include targeting the pan B cell specific surface proteins, such as CD19, CD20 and CD22, or suppressing the B cell antigen receptor cell signalling pathway, specifically by targeting spleen tyrosine kinase or phosphoinositide-3-kinase or by targeting costimulatory receptor/ligands, including CD40/CD40-ligand, CD30/CD30 ligand or inducible costimulator ICOS (CD278)/ICOS ligand interactions [59, 62–64, 123–128]. Alternatively, agents have been developed to block TLR stimulation, with TLR7 and TLR9 particularly relevant to lupus, whereas other therapies target proteins that inhibit B cell survival and/or differentiation into immunoglobulin-secreting plasma cells, including the cytokines IL-6, IL-21, IL-17, CD257 (B Lymphocyte Stimulator (BLyS,) , B cell Activating Factor (BAFF)), CD256 (A Proliferation-Inducing Ligand (APRIL)) and type I IFNs, or target homing receptors necessary for B cell migration to germinal centres or effector niches, such as chemokine receptors/chemokines including CXCR4/CXCL12, CXCR5/CXCL13 and CXCR3/CXCL9. As techniques to detect pathogenic autoreactive B cells improve, future therapies will be designed to target an individual’s autoreactive B cell pool while sparing regulatory B cells and the majority of B cells that provide protective immunity.

Rituximab is the best characterized of the anti-CD20 mAbs and is a prototypical biologic [127]. Rituximab is a chimeric mAb with human IgG1 domains and murine variable regions to CD20. Rituximab is a B cell depletion therapy and is approved for RA treatment. However, the LUNAR and EXPLORER trials for LN or extrarenal lupus, respectively, did not reveal additive effects of using rituximab on top of the background immunosuppressants [45, 46]. In studies without background immunosuppression beyond CSs, small reports indicate that rituximab is at least equivalent to non-specific immunosuppressants in treating active LN. Rituximab was compared with MMF and with CYC in a trial of 54 LN patients and shown to be as effective as the other agents [47]. B cell depletion is marked but not complete, because early B cells and plasma cells lack CD20 [51]. Some normalization of B cell subsets has been observed in rituximab-treated SLE patients [123]. Initially, it was suggested that complete B cell depletion might result in a better outcome for SLE [123]. However, SLE flares observed after repeated rituximab infusions were correlated with elevated circulating CD257 (BLyS) levels and high anti-dsDNA levels [48, 49]. This finding is consistent with the elevated serum CD257 levels observed in patients with low numbers of circulating B cells [50]. Thus, it was proposed that B cell depletion with rituximab induced a surge in CD257 levels that exacerbated disease in some SLE patients [51]. In these individuals, rituximab depletion was followed by rapid peripheral B cell reconstitution, with increased frequencies of circulating plasmablasts. It has been suggested that these plasmablasts might stimulate autoreactive T follicular helper (Tfh) cells, which promote more autoantibody production and drive a positive feedback loop promoting disease activity [51]. Of ongoing trials, one LN trial will examine the efficacy of rituximab with or without belimumab (anti-CD257 mAb) therapy and CYC. Another trial for refractory SLE will evaluate rituximab-treated patients receiving low- or high-dose belimumab. These strategies use sequential biologic therapies to address the predicted spikes in CD257 levels and autoantibody formation observed in some patients after rituximab depletion therapy.

Rituximab has been shown to be effective for certain refractory SLE patients, especially refractory neuropsychiatric SLE [52]. Thus, a personalized approach to treatment has been proposed for rituximab in this patient subgroup. Other anti-CD20 mAb, such as obinutuzumab, also efficiently mediated B cell depletion, and obinutuzumab is under study for SLE [27, 53–58].

Anti-CD19 mAbs may be more effective depletion therapies than anti-CD20 mAbs for SLE because CD19 is expressed on plasmablasts, long-lived plasma cells and early B cells that do not express CD20 [59]. Moreover, SLE B cells expressing high levels of CD19 were correlated with neurological or renal involvement, and patients demonstrated less benefit from rituximab therapy [129, 130]. MEDI551, a humanized IgG1 κ anti-CD19 mAb has been examined in a humanized Sle1.hCD19-Tg murine lupus autoantibody model [60] and is currently in trials for multiple sclerosis, scleroderma and B cell malignancies [59]. This anti-CD19 mAb spared regulatory B cells in an animal model of multiple sclerosis [61]. The depletion of regulatory B cells is a potential explanation for autoimmunity observed in rituximab-treated individuals [131–135]. Thus, anti-CD19 antibody might be a more effective therapy than rituximab for SLE in cases of refractory disease.

CD22 forms a complex with the B cell antigen receptor and functions as an inhibitory co-receptor. The anti-CD22 mAb, epratuzumab, is under study for SLE [63, 125, 126]. Anti-CD22 therapy appeared to attenuate B cell responses and promoted IL-10 anti-inflammatory cytokines while dampening the inflammatory cytokines, IL-6 and TNF, and altering B cell trafficking patterns [62, 63, 125]. Thus, anti-CD22 appears to be an effective and less extreme alternative to B cell depletion therapies for select patients.

Cytokine-targeted therapies

Belimumab, the anti-CD257 mAb, acts as a soluble CD257 antagonist and was the first drug approved in 56 years by the FDA for SLE and has been extensively reviewed elsewhere [65–72]. The cytokine CD257 (also known as B-lymphocyte stimulator, BLyS, BAFF or TNFSF13B) promotes B cell survival and differentiation and can be produced largely by myeloid cells and a variety of other cells. Interestingly, blood CD257 and CD256 levels decrease with age in healthy individuals [50]. Low peripheral B cell numbers in mice or immunodeficient patients can be correlated with increases in circulating CD257 levels [50]. Initially, it was suggested that an increase in circulating CD257 levels might slow B cell differentiation, resulting in fewer plasmablasts after B cell depletion in SLE [66, 124]. The B cell depletion studies did not support this notion [51]. As discussed earlier, B cell depletion with rituximab appears to alter CD257 homeostasis. Thus, studies combining rituximab and belimumab are in progress.

Blisibimod is a fusion protein composed of a peptide that targets CD257 with high affinity and an IgG Fc domain [40, 65, 72, 78–80]. Blisibimod antagonizes the membrane and soluble forms of CD257. Clinical trials are ongoing [80]. Two other inhibitors that target the CD257/BLyS axis, atacicept and tabalumab, have been studied and are currently not being pursued for SLE [65, 70, 73–77].

IFN-induced genes have been investigated as biomarkers, and IFN-targeted therapies have been studied for SLE [81, 136, 137]. Sifalimumab and rontalizumab are humanized anti-IFNα mAbs [1, 65, 81]. Initial results of rontalizumab in a trial for SLE indicated that the patients with low IFN signatures (IFN gene expression in peripheral blood cells) received the most benefit from this therapy [1, 65]. The observation of better responses in Interferon Signature Metric (ISM)-Low patients was an unexpected finding that emerged from an exploratory analysis, not a primary end point. ISM-Low patients had lower titres of anti-dsDNA and ENA antibodies, less profound hypocomplementaemia, but similar baseline disease activity compared with ISM-High patients. ISM-Low patients achieved higher mean through concentrations of rontalizumab, so that differences in drug exposure might have affected results [82]. Anifrolumab, a humanized anti-IFNα receptor mAb, is predicted to be more effective in targeting IFNα [84]. The safety and efficacy of immunization against IFNα with an IFNα-kinoid (IFN kinoid) has been demonstrated [83]. A trial of the immunogen IFN-kinoid, which consists of inactivated IFNα coupled to keyhole limpet haemocyanin as a carrier protein, is in progress (NCT02665364). Thus IFN-targeted therapies remain an active area of investigation for SLE.

IL-6 contributes to a plethora of immune cell activities, such as cell activation, proliferation, differentiation and cytokine secretion. IL-6 acts in concert with IL-1β and TNFα to drive inflammation. IL-6 stimulates the differentiation of potent inflammatory Th17 cells and B cell differentiation into plasma cells. However, IL-6 is key for certain homeostatic mechanisms as well as the acute phase response. Tocilizumab is a humanized anti-IL-6 receptor (IL-6R) mAb that blocks IL-6 binding to IL-6R. It has been tested in a trial for SLE [85]. Although it is an FDA-approved therapy for RA, an increased risk of infection may limit the use of tocilizumab for SLE. Sirukumab, an anti-IL-6 mAb, was tested in LN [86–88]. A recent report suggests that no future lupus studies are planned for sirukumab [1].

Low-dose IL-2 has been shown to promote the expansion of Tregs that assist in the maintenance of peripheral tolerance [89–91]. High expression of Foxp3 identifies Tregs in some but not all cases [92]. Mice and humans develop severe autoimmune syndromes with Foxp3 deficiency. In humans, Foxp3 deficiency is called immune dysregulation, polyendocrinopathy, enteropathy or X-linked syndrome [93, 94]. In murine models, the deficiency causes scurfy mice [95]. Treg numbers and/or function are often diminished in chronic SLE [96–99], whereas Treg percentages can be increased in new-onset SLE and correlated with anti-dsDNA titres [98]. Methylprednisolone was reported to induce Treg expansion in SLE [100]. The conditions that promote the pathogenic conversion of unstable Tregs in humans have not been completely elucidated [101, 102]. Current therapies used for SLE patients could decrease Treg numbers and/or function. Initial results of a clinical trial assessing low-dose IL-2 for Treg expansion in SLE were promising, and no serious adverse events were reported (NCT02084238).

Costimulator-targeted therapies

Abatacept is a fusion protein of CTLA4-Fc, with CTLA4 fused to the Fc domain of IgG1. CTLA4 (CD152) is a homologue of CD28 that negatively regulates costimulatory signals through the CD28–CD80/CD86 pathway. CTLA4 deficiencies result in lymphoproliferative disorders [103, 104]. Abatacept has been approved for RA and it is being tested for SLE and LN [105, 106]. New abatacept trials have focused on SLE arthritis.

CD154/CD40 provides important costimulatory signals for normal T- and B cell immune responses. CD40 is expressed by antigen-presenting cells, including B cells, monocytes and dendritic cells. CD154 is usually expressed by activated CD4⁺ T cells, but in SLE CD154 can be expressed by CD8⁺ and CD4⁺ T cells. Early trials with anti-CD154 mAbs were discontinued because of thromboembolic events [138] or poor efficacy [139, 140]. A new PEGylated anti-CD154 antibody monovalent fragment lacking Fc, CDP7657, does not induce thrombotic events and is now in clinical trials [107]. Single doses of CDP7657 were well tolerated in a randomized, double-blind, dose-escalation phase I trial [108]. Further testing is needed to determine the efficacy of CDP7657 for SLE.

CD275/CD278 (Inducible T cell Costimulator Ligand-ICOSL/ICOS) is another important receptor/ligand pair required for Tfh cell differentiation. Tfh cells provide help for B cell antibody production and maintain germinal centres [141]. Targeting this pathway is predicted to reduce autoantibodies, while complete inhibition is predicted to cause immunodeficiency. Conversely, increased expression of ICOS is correlated with lupus-like disease in murine models [142, 143]. An anti-ICOSL mAb is under study in lupus clinical trials [109]. Future studies will establish the potential efficacy of anti-ICOSL therapy for SLE.

The chemokine receptor CXCR5 has been shown to play an important role in B cell migration to lymphoid organs, B- and T cell migration to germinal centres and enhancing B cell antigen receptor-mediated responses [144, 145]. CXCR5 deficiency increases survival in lupus-prone mice [146]. An initial clinical trial of anti-CXCR5 mAb has been completed (NCT02321709).

Complement-targeted therapies

A phase I clinical trial of an anti-C5a receptor mAb for treatment of chronic autoimmune diseases has been completed (NCT02151409), but study results are not available. Another mAb targeting complement is eculizumab, which is a humanized recombinant IgG2/IgG4 mAb that targets complement factor C5. Eculizumab blocks C5 from converting to C5a and C5b, which prevents the membrane attack complex formation and C5a chemotactic activity. Eculizumab has been approved for paroxysmal nocturnal haemoglobinuria and similar disorders in the USA [110–115]. Eculizumab has been used in catastrophic APS (CAPS) not responding to conventional therapies, with mixed results in several case reports [116–120, 147]. In addition, case reports have proposed that eculizumab may be useful after kidney transplantation in preventing ischaemia–reperfusion injury and antibody-mediated rejection as well as APS-related thrombotic microangiopathy [120]. Randomized trials are under way. Further studies of eculizumab for SLE will determine whether it will be a viable alternative for select patients.

The biologic therapies discussed above highlight the variety of immune targets, such as B cells, T cells and myeloid cells and their cytokines, that contribute to lupus pathogenesis. Other SLE therapies are listed in Table 2 and demonstrate the expansion of our knowledge base and understanding of disease mechanisms over the past two decades. As discussed below, a number of novel therapies for SLE have been proposed or are available for lupus trials.

Future therapeutics

Several potential SLE therapies target pathways not tested in previous lupus clinical trials. For example, ustekinumab is an mAb directed against the p40 subunit of IL12/IL23 and has been FDA approved to treat psoriasis and is now in trials for SLE. Other agents include an anti-Fcγ-receptor-IIb, TLR inhibitors, Jak inhibitors, kinase inhibitors specifically targeting spleen tyrosine kinase or phosphoinositide-3-kinase, proteasome inhibitors and histone deacetylase inhibitors as candidate therapies for SLE [40, 65]. Recently, there has also been a drive to accelerate repurposing drugs approved for other diseases for SLE, such as those described at www.linkedin.com/in/lrxlstat (LRxL, Lupus Treatment List-STAT, SLE Treatment Acceleration Trials) [1]. Given the rapid advances in antisense oligonucleotide therapies (aptamers) for monogenic diseases and cancer, it is likely that these agents will be used in targeted approaches for specific subsets of SLE patients [148].

New approaches to assess therapeutics: unmet needs

Studies on the efficacy of single targets suggest that combination therapies designed to strike two or more complementary pathways could be the most effective approach to treatment for many SLE patients. Unfortunately, this approach is not often observed in current clinical trial designs, possibly because an elevated side-effect risk was previously seen in RA trials that combined biologics. Exceptions include the two trials that evaluated the efficacy of rituximab with belimumab therapy. The failure of new SLE therapies to meet primary outcome measures in large clinical trials has drawn attention to the need to allow for the heterogeneous nature of the disease [149]. As an example, TNF inhibitors, such as etanercept and infliximab, can be associated with the induction of anti-dsDNA; however, these agents could be of potential therapeutic benefit for the SLE patient subgroup with arthritis [1, 150].

The heterogeneity observed in SLE was addressed in a study by Banchereau et al. [151] that identified seven distinct paediatric SLE patient subgroups. Banchereau et al. [151] used weighted gene co-expression network analysis to identify transcriptional signatures from the blood of individual patients assessed in longitudinal studies. These data were correlated with clinical measures [151]. The stratification of patients into seven distinct subgroups was bolstered by data from immune genotypes that overlapped with known expression quantitative trait loci. Results from the transcriptional profiling and immune genotypes were combined with clinical data and immunophenotypes to establish the SLE subgroups. In the absence of longitudinal studies, a patient’s disease phenotype might be derived from combining select genotype data, transcriptome analysis, immunophenotypes and/or other assays to confirm immune dysregulation.

Another example, a biomarker assay, similar to the assay performed by Lu et al. [152] that incorporates measures of serum cytokines, chemokines and soluble receptors or other mediators dysregulated in preclinical disease or at the onset of SLE and followed over time in individual patients, could improve prognosis and treatment. This information could be used to stratify patients for clinical trials designed to strike specific targets. Thus, improvements in clinical trial design combined with personalized approaches to disease management are predicted to lead to more rapid advances in targeted therapies for SLE in the near future.

Other improvements that could facilitate clinical trial design include central randomization of patients from multiple centres and changes to add-on drug trial designs [149]. New therapies that synergize with each other could be chosen over addition of agents with redundant mechanisms. A clear example noted is rituximab, which is effective as a monotherapy but fails in trials that include background therapy with non-specific immunosuppressants.

Improvements in outcome measures and appropriate study end points might include the incorporation of more practical composite outcome measures that are feasible for short-term end points, as well as more definitive outcome measures, for example, improved biomarkers, which could be uniformly evaluated at a centralized location [149]. Such improvements in clinical trial design are predicted to lead to more rapid advances in targeted therapies for SLE.

Personalized medicine has become routine for certain cancers; however, forming personalized treatment regimens for SLE is far more complex given the variety of genetic and environmental factors that contribute to disease pathogenesis. Yet assembly of strategies toward a more personalized approach to diagnose and treat SLE is under development. Current research models integrate clinical data and immune system genetic and transcriptional profiles to impute a patient’s SLE subtype [151, 153].

One recent proposal for SLE clinical trial design suggests incorporating a personalized approach by looking at an individual’s disease activity combined with the related treatable activated target. For example, an open-label dose-escalation trial would be used to determine the optimal dose for each individual that could potentially respond with a 50% reduction in target to a given biologic therapy [154]. Responders could then enter a blind randomization phase at their optimal dose or be evaluated for flares as a primary end point. Combination therapies might be investigated for non-responders, who might have other active pathways or feedback mechanisms that could be targeted.

Future stratification algorithms might incorporate such information for potential application to personalized disease management [151, 153, 154]. Alterations in clinical trial design are underway and will require further modifications in order to account better for disease heterogeneity and to translate current discoveries and targeted therapies efficiently to clinical practice.