Background

Systemic Lupus Erythematosus (SLE) is an autoimmune disease of unknown aetiology with multiple manifestations including musculoskeletal, renal, haematological, serosal, and neuropsychiatric involvement. Treatment for SLE to date is centred on immunosuppression and anti-inflammatory therapy, depending on the degree of end organ involvement. Pregnant and non-pregnant lupus patients benefit from the use of hydroxychloroquine (HCQ), with reductions in lupus flares, end organ damage, loss of bone mass, thrombosis, cumulative steroid usage and increased long term survival [1]. Non-steroidal anti-inflammatory drugs (NSAIDs) may be used to manage milder manifestations such as musculoskeletal or mucocutaneous manifestations. Chronic glucocorticoid (GC) therapy is associated with cumulative dose toxicity. However, given its efficacy it is often used in lower doses as a component of maintenance therapy, or in higher doses for the treatment of disease flares depending on the severity of end organ involvement. Other immunosuppressants used include mycophenolate mofetil (MMF), cyclophosphamide (CYC) and calcineurin inhibitors such as tacrolimus (TAC) and azathioprine (AZA). “Standard of care” therapy is typically defined in clinical trials to include these agents.

Multiple biological agents have recently emerged as potential novel treatments for SLE. In this review we aim to summarise the available data from randomised controlled trials for the efficacy of biologics in SLE, and to highlight potential therapies which require further data.

Methods

All phase II, and III clinical trials or randomised control trials or quasi randomised controlled trial enrolling adult patients with SLE according to standard criteria, examining biologic agent/s compared to placebo, other immunosuppressive drug/s or standard of care were examined.

Outcome measures included change in validated disease activity indices such as SLEDAI, SELENA-SLEDAI, SLEDAI-2 K, BILAG, BILAG-2004, SLICC/ACR score. Adverse events and death were also recorded.

Search methods are documented in the online supplement. Two authors independently examined all studies and extracted data. Dichotomous outcome results were expressed as risk ratios (RR) with 95% confidence intervals (CI), with data pooled using random effects models. Data with continuous outcomes were not measured in this review. The Cochrane risk of bias tool was used by both authors to independently assess the quality of included studies.

Results

One thousand eighty-seven studies were identified. Seventy-nine studies were further assessed. Forty-four studies were included with 16,889 patients, 15 distinct drug groups and 25 biological agents. Characteristics of the studies including patient characteristics and study protocols are summarised in the supplementary information. PRISMA flow diagram is shown below (Fig. 1), and the PRISMA checklist is included in the supplementary information.

Fig. 1
figure 1

PRISMA flow

Full size image

CD80/86 inhibition

CD80/86 is expressed by antigen presenting cells such as plasmacytoid dendritic cells and B cells. CD80/86 ligate CD28, a co-stimulatory receptor expressed on T cells. CD28 stimulation in conjunction with T cell receptor engagement prolongs and increases T cell differentiation and production of IL2, with subsequent B cell proliferation and differentiation into antibody producing plasma cells.

Abatacept

Abatacept is a fusion protein composed of a CTLA-4 molecule linked to the Fc portion of IgG1. This selectively and competitively antagonises CD80 and CD86 receptors on an antigen presenting cell, limiting CD28 mediated T cell activation.

Four studies [2,3,4,5] included 1017 patients. Three of the studies recruited patients with lupus nephritis whereas Merrill 2010 [2] excluded patients with renal involvement.

No outcomes achieved significance. Serious adverse events were significantly raised only in Merrill 2010 (RR 2.93, CI 1.06 to 8.05, P = 0.04) but not in the pooled data of all the Abatacept studies (RR 1.17, CI 0.87 to 1.58, P = 0.30) (Fig. 2).

Fig. 2
figure 2

Serious adverse events

Full size image

Anti-interferon monoclonal antibody

T1 IFN is considered the canonical SLE cytokine impairing immune tolerance through multiple mechanisms. Three anti-interferon monoclonal antibodies have been assessed in this review. Anifrolumab which binds to both IFN-α/β receptors, Rontalizumab and Sifalimumab which selectively bind to IFN-α receptors.

Anifrolumab

Anifrolumab is a fully human, IgG1k monoclonal antibody that binds to IFN-α/β receptor and prevents signalling by all types of I IFNs.

Three studies addressed the use of Anifrolumab in SLE: Furie 2017 [6], Furie 2019 [7] and Morand 2020 [8] and included 1124 patients. The main outcomes studied were SRI and BICLA response.

SRI 4 at 24 weeks did not achieve statistical significance (RR 1.34, CI 0.84 to 2.15, P = 0.22, 2 studies [6, 7]), though results from Furie 2017 alone were significant (RR 1.79, CI 1.12 to 2.85, P = 0.01) (Fig. 3a).

Fig. 3
figure 3

A SRI 4 at 24 weeks, B SRI 4 at 52 weeks

Full size image

SRI 4 at 52 weeks did not achieve statistical significance (RR 1.40, CI 0.94 to 2.08, P = 0.10, 2 studies [6, 7]), though results from Furie 2017 alone were significant (RR 1.76, CI 1.22 to 2.53, P = 0.002) (Fig. 3b).

In a single study [7] at 52 weeks, Anifrolumab significantly increased SRI 5 (RR 1.37, CI 1.05 to 1.78, P = 0.02) (Fig. 4), SRI 7 (RR 1.86, CI 1.27 to 2.72, P = 0.001) (Fig. 5), and SRI 8 (RR 1.97, CI 1.32 to 2.95, P = 0.0009) (Fig. 6), but not SRI 6 (RR 1.29, CI 0.99 to 1.69, P = 0.06), though the results trended towards significance (Fig. 7).

Fig. 4
figure 4

SRI 5 at 52 weeks

Full size image
Fig. 5
figure 5

SRI 7 at 52 weeks

Full size image
Fig. 6
figure 6

SRI 8 at 52 weeks

Full size image
Fig. 7
figure 7

SRI 6 at 52 weeks

Full size image

Anifrolumab significantly increased BICLA response at 52 weeks in all 3 studies (RR 1.56, CI 1.33 to 1.84, P < 0000.1) (Fig. 8).

Fig. 8
figure 8

BICLA at 52 weeks

Full size image

Prednisone dose reduction to < 10 mg/day was increased with Anifrolumab treatment (RR 1.46, CI 1.16 to 1.84, P = 0.001, 3 studies) (Fig. 9).

Fig. 9
figure 9

Change in prednisone dosages to ≤ 10 mg/day

Full size image

Adverse events were increased with Anifrolumab treatment, (RR 1.09, CI 1.04 to 1.15, P = 0.001, 3 studies) (Fig. 10), with a higher incidence of herpes zoster infections, but there were significantly fewer serious adverse events, (RR 0.68, CI 0.49 to 0.95, P = 0.02, 3 studies) compared to controls (Fig. 2). The other safety outcomes did not reach statistical significance.

Fig. 10
figure 10

Adverse events

Full size image

Rontalizumab

Rontalizumab is a human anti-IFN-α monoclonal antibody that binds to all 12 IFN-α subtypes preventing signalling through the type I IFN receptor.

One study [9] including 238 patients addressed the use of Rontalizumab in SLE: Kalunian 2016 Patients with lupus nephritis were excluded. At 24 weeks, Rontalizumab did not improve SRI 4 (RR 1.11, CI 0.83 to 1.48, P = 0.47) (Fig. 3a), though there were steroid sparing benefits with an increased number of patients tapering their steroids to a prednisone equivalent of ≤ 10 mg/day (RR 1.21, CI 1.0 to 1.46, P = 0.05) (Fig. 9). There were no significant differences in safety outcomes.

Sifalimumab

Sifalimumab is a fully human, immunoglobulin G1 κ monoclonal antibody that binds to and neutralises the majority of IFN-α subtypes.

In a single study [10], at 52 weeks, Sifalimumab improved SRI 4, (RR 1.28, CI 1.02 to 1.61, P = 0.03) (Fig. 3b) and SRI 6 (RR 1.32, CI 1.01 to 1.73, P = 0.04) (Fig. 7). SRI 5 at 52 weeks (RR1.27, CI 0.98 to 1.65, P = 0.07) (Fig. 4) and BICLA at 52 weeks (RR 1.29, CI 0.98 to 1.71, P = 0.07) (Fig. 8) trended towards but did not achieve significance. There were no steroid sparing benefits, with no difference in the reduction in prednisone < 7.5 mg/day with 25% reduction from baseline dosage (RR 1.21, CI 0.41 to 3.54, P = 0.73) (Fig. 11). Adverse events were not increased with the use of Sifalimumab, though there were higher rates of herpes zoster compared to placebo (5.9% vs 0.9%).

Fig. 11
figure 11

Change in prednisone dosages to ≤ 7.5 mg and > 25% reduction from baseline dosage

Full size image

Anti BAFF/BLyS and APRIL monoclonal antibody

BAFF and APRIL are cytokines from the TNF family, secreted by most myeloid and lymphoid cells, and bind to TACI, BCMA and BAFF receptors. Ligation of BAFF receptors promote B cell survival, immunoglobulin class switching and secretion. BAFF binds to all 3 receptors, whereas APRIL only binds to TACI and BCMA. Blisibimod and Tabalumab inhibit soluble and membrane bound BAFF and Belimumab binds to soluble human BAFF. Atacicept and Telitacicept block both BlyS and APRIL.

Belimumab

Belimumab is a human IgG1 monoclonal antibody that binds soluble human BlyS. It is currently only indicated for use in SLE not responding to standard of care therapy.

Seven studies [11,12,13,14,15,16,17] including 4022 patients. Furie 2020 [16] and Atisha Fregoso 2021 [17] included patients with lupus nephritis.

At 52 weeks, Belimumab use significantly increased SRI 4 (RR 1.27, CI 1.18 to 1.38, P < 0.0001, 4 studies) (Fig. 3b). In a single study at 52 weeks, improvements were demonstrated in SRI 5 (RR 1.56, CI 1.20 to 2.03, P = 0.0009) (Fig. 4), SRI 6 (RR 1.57, CI 1.19 to 2.08, P = 0.001) (Fig. 7) and SRI 8 (RR 1.50, CI 1.01 to 2.22, P = 0.05) (Fig. 6). In two studies, SRI 7 at 52 weeks significantly increased (RR 1.43, CI 1.13 to 1.81, P = 0.003) (Fig. 5).

Belimumab did not alter CRR/PRR at 1 year (RR 1.28, CI 0.67 to 2.45, P = 0.45, 1 study) but showed a significant effect at 2 years (RR 1.29, CI 1.04 to 1.61, P = 0.03, 2 studies) (Fig. 12b).

Fig. 12
figure 12

A Combined complete and partial renal remission at 1 year, B combined complete and partial renal remission at 2 years

Full size image

Belimumab significantly increased the number of patients able reduce prednisone dosages to ≤ 7.5 mg/day (RR 1.45, CI 1.16 to 1.80, P = 0.0009, 5 studies (Fig. 11).

There were no significant difference in serious adverse events (RR 0.88, 0.72 to 1.08, P = 0.24, 6 studies) (Fig. 2) or in other reported safety outcomes.

Blisibimod

Blisibimod is a selective inhibitor of soluble BAFF and membrane-bound BAFF, composed of a tetrameric BAFF binding domain fused to a human IgG1. Two studies [18, 19] included 988 patients. Patients with severe lupus nephritis were excluded.

Blisibimod increased SRI 4 at 24 weeks only in Merrill 2018 [19], (RR 1.19, CI 1.00 to 1.41, P = 0.05, 2 studies) (Fig. 3a), but not SRI 4 and 6 at 52 weeks or SRI 5 to 8 at 24 weeks. Blisibimod reduced prednisone dosage below 10 mg/day (RR 1.64, CI 1.07 to 2.52, P = 0.02, 1 study [19] (Fig. 9).

Infusion related adverse events were increased (RR 1.85, CI 1.21 to 2.81, P = 0.004, 1 study [19] (Fig. 13). There were no significant increase in other adverse events.

Fig. 13
figure 13

Infusion related adverse events

Full size image

Tabalumab

Tabalumab is a fully human IgG4 monoclonal antibody, that binds and neutralises both membrane and soluble BAFF. Two studies [20, 21] included 2262 patients. Patients with severe lupus nephritis were excluded.

Tabalumab significantly increased SRI 5 at 52 weeks (RR 1.23, CI 1.06 to 1.42, P = 0.005, 2 studies) (Fig. 4). Tabalumab did not significantly decrease prednisone doses (RR 1.21, CI 0.78 to 1.89, P = 0.39, 2 studies).

Infusion related adverse events were significantly higher with Tabalumab, (RR 1.63, CI 1.05 to 2.53, P = 0.03, 2 studies) (Fig. 13). Tabalumab did not increase withdrawals from the study, serious infections or death.

Atacicept

Atacicept is a recombinant fusion protein comprising the extracellular domain of the TACI receptor joined to a human IgG1 Fc domain that blocks B-cell activating factor BlyS and APRIL.

Two studies [22, 23] included 767 patients. Patients with lupus nephritis were excluded.

In one study [23], Atacicept did not increase SRI 4 (RR 1.27, CI 0.99 to 1.63, P = 0.06) (Fig. 3a), SRI 6 (RR 1.13, CI 0.79 to 1.62, P = 0.49) and BICLA (RR 1.13, CI 0.87 to 1.47, P = 0.36) at 24 weeks.

There were no steroid sparing benefits or significant differences in the safety outcomes.

Telitacicept

Telitacicept is a fusion protein comprising a recombinant TACI receptor fused to the Fc domain of human IgG, which binds to and neutralises the BLyS and APRIL, suppressing development and maturation of plasma cells and mature B cells.

One study [24] included 202 patients. Patients with severe lupus nephritis were excluded.

SRI 4 at 52 weeks was significantly increased with Telitacicept (RR 2.12, CI 1.48 to 3.03, P < 0.00001) (Fig. 3b).

There were no significant differences in reported safety outcomes of adverse events, serious adverse events and death.

Anti-CD20 monoclonal antibody

Three anti CD20 monoclonal antibodies are examined in this review, Rituximab (murine-human chimeric), Ocrelizumab and Obinutuzumab (humanised).

Obinutuzumab

Obinutuzumab is a recombinant type II anti-CD20 and IgG1 Fc-optimised humanised monoclonal antibody, which has improved mAb-FcγRIIIA interaction and direct and immune effector cell-mediated cytotoxicity compared to Rituximab.

One study [25] included 125 patients. Patients with lupus nephritis ISPN/RPS 2003 class III/IV were included in the study.

Combined CRR/PRR at 1 year was increased (RR 1.60, CI 1.27 to 2.02, P < 0.0001) (Fig. 12a) and 2 years (RR 1.83, CI 1.06 to 3.16, P = 0.03) (Fig. 12b).

There were fewer grade 3 or higher related infectious events with Obinutuzumab (RR 0.29, CI 0.10 to 0.85, P = 0.02) (Fig. 14), but no significant differences in the other safety outcomes.

Fig. 14
figure 14

Grade 3 or higher related infectious events

Full size image

Ocrelizumab

Ocrelizumab is a humanised monoclonal antibody against CD20 and may have greater antibody dependent cellular toxicity and less complement dependent cytotoxicty compared to Rituximab which is a chimeric monoclonal antibody. One study [26] included 378 patients with lupus nephritis.

Combined CRR/PRR at 1 year was not increased (RR 1.22, CI 0.97 to 1.55, P = 0.09).

There were no significant differences in the pooled safety outcomes, but a higher rate of serious infections were seen in patients receiving MMF compared to ELNT induction.

Rituximab

Rituximab is a type 1 chimeric anti-CD20 monoclonal antibody directed to the CD20 antigen on the surface of B lymphocytes, causing apoptosis, complement activation and cell mediated cytotoxicity.

Two studies [27, 28] included 401 patients. Rovin 2012 [28] only included patients with lupus nephritis class III/IV ± V. Rituximab did not increase CRR/PRR at 1 year, (RR 1.24, CI 0.90 to 1.71, P = 0.19, 1 study [28]). There was no reduction in the number of patients achieving prednisone < 10 mg/day (RR 0.81, CI 0.37 to 1.80, P = 0.60), 1 study [27].

Ustekinumab

Ustekinumab is a fully humanised monoclonal antibody against the p40 subunit found on both IL-12 and IL-23. IL-12 has a key role in inducing Th cell differentiation to Th1 cells, and IL-23 in Th17 cell activation and subsequent IL-17 secretion.

One study [29] included 102 patients. Patients with lupus nephritis class III/IV were excluded.

Ustekinumab increased SRI 4 at 24 weeks (RR 1.85, CI 1.15 to 2.97, P = 0.01) (Fig. 3a), SRI 5 at 24 weeks (RR 2.02, CI 1.06 to 3.86, P = 0.03) (Fig. 15) and SRI 6 at 24 weeks (RR 2.27, CI 1.14 to 4.52, P = 0.02) (Fig. 16), but not BICLA at 24 weeks (P = 0.86). Ustekinumab use did not increase any adverse events.

Fig. 15
figure 15

SRI 5 at 24 weeks

Full size image
Fig. 16
figure 16

SRI 6 at 24 weeks

Full size image

Group of drugs without significant results

There were no significant outcomes in the disease activity indices, composite responder rates or adverse events in the following group of drugs; anti-dsDNA complexing Abetimus [30] selective JAK 1 and 2 inhibitors Baricitinib [31], BTK inhibitors Evobrutinib [32], Fenebrutinib [33], high affinity cereblon ligand CC-220/Iberdomide [34, 35], tolerogenic peptides Edratide [36], anti CD22 monoclonal antibody Epratuzumab [37,38,39], anti IL-6 antibody PF-04326921 [40] Vobarilizumab [41] anti IL-10 monoclonal antibody BT063 [42], P140 peptide Lupuzor [43] and recombinant soluble human FcyRIIb SM101 [44].

Summary of findings

The main results of this review are presented in the summary of findings tables. Outcomes with significant results presented include composite outcomes, renal outcomes, glucocorticoid dose reduction, and adverse events. The complete GRADE tables are shown below (Refer Table 1: Composite outcomes, Table 2: Renal outcomes, Table 3: Glucocorticoid dose reduction, Table 4: Adverse events).

Table 1 Composite outcomes
Full size table
Table 2 Renal outcomes
Full size table
Table 3 Glucocorticoid dose reduction
Full size table
Table 4 Adverse events
Full size table

Risk of bias

Risk of bias graph (Fig. 17) and risk of bias summary (Fig. 18) is shown below.

Fig. 17
figure 17

Risk of bias graph

Full size image
Fig. 18
figure 18

Risk of bias summary

Full size image

Discussion

Summary of main positive and negative outcomes of our study.

We have summarised the RCT data available on 25 biological agents from 15 different drug groups in the treatment of SLE. The majority of these drugs have limited data available and will require further trials to determine their efficacy in various patient groups.

Currently Belimumab is shown to have the most significant data suggesting that it is effective in SLE without a major adverse effect profile. There is high quality evidence showing Belimumab improves composite outcomes measured by SRI. The level of evidence for other biologics with significant outcomes range from low to moderate (Summary of findings: Composite outcomes). Other newer treatments have shown significant efficacy but in more specific outcomes and will need further trials to clearly delineate their strengths and weaknesses.

The main outcomes assessed in these studies were SRI, BICLA, and combined CRR/PRR. Of the 25 biologic agents, only anti-interferon, anti BAFF/BLyS and/or APRIL, anti IL12/23 and anti CD20 monoclonal antibodies were found to improve outcomes.

Anifrolumab increased BICLA response at 52 weeks, SRI 5 to 8 in a single study (Furie 2019), decreased prednisone dosages, with increased adverse events with herpes zoster infections, but with lesser serious adverse events. Sifalimumab also improved SRI but also increased herpes zoster infections. Among the anti BAFF/Blys and/or APRIL monoclonal antibodies, Belimumab consistently improved SRI 4, decreased prednisone dosages, increased combined CRR/PRR in a single study, and had no adverse safety outcomes. Tabalumab increased SRI 5 at 52 weeks with no steroid sparing effect but was associated with increased infusion related adverse events. Telitacicept also improved SRI 4 at 52 weeks, without data on its effect on steroid dosages. Of the three anti CD-20 monoclonal antibodies, only Obinutuzumab increased combined CRR/PRR at 1 and 2 years, with lower grade 3 or higher infectious events. The single anti IL12/23 monoclonal antibody, Ustekinumab, increased SRI 4 to 6, but not BICLA at 24 weeks, with no concerning safety outcomes.

Despite positive results in some of these biologics, several of their developments have since been terminated. There are no further trials planned for Tabalumab ( two phase III trials) by the parent pharmaceutical company as it was not felt to have reached significant efficacy compared to existing treatments, and Sifalimumab's development (one phase IIb trial) has been ceased in favour of Anifrolumab. Following the completion of this review, a phase III trial of Ustekinumab involving 516 patients showed no superiority compared to placebo when measuring SRI 4 as a primary endpoint [45].

Our review did not include non-biologics such as the calcineurin inhibitor Voclosporin which has shown benefit in proteinuria reduction in patients with lupus nephritis [46].

The other remaining drug classes and biological agents did not improve any of the outcomes assessed in the study and had no other notable safety outcomes.

Difficulties with outcome measures

Prior to the introduction of SRI and BICLA, trials reported outcomes using individual BILAG, SLICC, SLEDAI based scores such as SLEDAI-2 K and SELENA-SLEDAI as their outcome measures. There were inconsistencies with how these scores were reported to denote significant results. Examples included BILAG as a numerical score determined by the study authors (and outcomes reporting changes in percentages, mean BILAG score differences compared to baseline) and differing organ domain severity scores (eg 1A and 2B, 1A and 1B, B only, C in all domains) and SLEDAI based metrics using varying decrease in points, expressed in means, medians or percentage of changes in baseline values.

SRI4 response is defined as SLEDAI improvement of 4 points or more, PGA not worsening by 0.3 points or more (10% or more), and BILAG having no new As and not having two or more new Bs. SRI 5, 6 and 7 correspond to an increase in improvement in SLEDAI points, without changes to the other criteria. BICLA response is defined as a reduction of all baseline BILAG-2004 A and B domain scores to B/C/D and C/D, no worsening in any organ system; no worsening of SLEDAI-2 K score from baseline, and no worsening ≥ 0.3 points (< 10% worsening) in Physician's Global Assessment, and no non-protocol treatment (new or increased immunosuppressives, antimalarials, corticosteroids or premature discontinuation of study treatment).

Comparing SRI and BICLA, SRI places more emphasis on SLEDAI improvement which does not evaluate the degree of individual component improvement, compared to the more comprehensive BILAG based BICLA, which does not evaluate for serological improvements. Ohmura 2021 [47] summarises the differences between the existing SLE activity indexes in clinical trials. Quality of life outcome measures also suffered from the aforementioned issues.

As SRI and BICLA incorporates a standardised change in BILAG, SELENA-SLEDAI/SLEDAI-2 K and PGA in their scoring, the authors of this study decided to omit data reporting other disease and quality of life outcomes outside of SRI and BICLA. This is mainly to maximise data that can be appropriately compared across studies, the main utility of a systematic review such as this.

Studies of lupus nephritis also did not use standardised definitions of complete or partial renal remissions (Table 5: Renal outcomes). Neither did they provide adequate reporting on other renal outcomes time to ESRD, or changes in serum creatinine/eGFR.

Table 5 Renal outcomes
Full size table

Comparison with other systematic reviews

Four other reviews examined the use of biologic agents in the treatment of SLE. A meta-analysis by Oon 2018 [48] that Belimumab, Tabalumab and Epratuzumab had steroid sparing effects, which differed from our finding of only Belimumab had a significant steroid sparing effect. We did not include the data of steroid doses in the Epratuzumab studies of Wallace (EMBLEM) 2013 and Clowse 2017 as they were reported as mean ± SD and Wallace 2013 (ALLEVIATE) which reported them as medians without sufficient IQR data. Borba 2014 [49] which assessed 7 biologic agents similarly concluded that Belimumab improved disease response in the outcomes assessed compared to placebo. Singh 2021 [50] assessed 6 RCTs of Belimumab and concluded that Belimumab was effective in increasing SELENA-SLEDAI (≥ 4 point improvement) and reduction in glucocorticoid dosages. Sciascia 2017 [51] assessed the efficacy of Belimumab in renal outcomes and reported a decrease in proteinuria in patients treated with Belimumab but were unable to arrive to any conclusions for other parameters of renal response due to differing criteria across the studies. We did not include data describing renal outcomes such as number of and time to renal flares, and proteinuria due to the heterogeneous methods of reporting them across the studies, limiting their applicability in a systematic review.

Conclusions

Recommendations for patient treatments

Based on current data, Anifrolumab, Sifalimumab, Belimumab, Tabalumab, Telitacicept, are effective treatments in the treatment of SLE without lupus nephritis. Anifrolumab and Belimumab are useful in decreasing the steroid burden in these patients when compared to other biologics. In patients with lupus nephritis, Belimumab and Obinutuzumab are effective treatments. There is insufficient data to recommend for or against the use of biologics in CNS lupus due to their exclusion from trials. Patients treated with Anifrolumab or Sifalimumab should consider herpes zoster vaccination prior to commencing treatment.

Recommendations for further research

Our review has revealed and summarised a wealth of studies in the treatment of SLE with biological agents and demonstrated the limited availability of data in many of these agents and the need for further studies to elucidate the efficacy of each agent in SLE treatment.

Comparison between agents will need to emerge as a research question in the near future. Other potential areas to consider will be the combination of treatments from different drug groups to improve the overall efficacy of disease control over time.

Studies involving biologics in SLE have heterogeneous endpoints and duration. The majority of the studies selectively excluded renal lupus involvement, though the criteria for exclusion varied widely, from active urinary sediment and mildly decreased eGFR to rapidly progressing glomerulonephritis. As lupus nephritis remains a leading cause of morbidity and mortality in SLE, a larger number of trials with a standardised definition of renal composite end points is required.

Trials in the treatment of SLE need to standardise outcomes and reporting in order that results can contribute to a coherent picture of treatment efficacy and safety.