Article Text
PDF
Statistics from Altmetric.com
After decades of failures and setbacks, the lupus community finally had a more fruitful period marked by the successful results in two phase III and one phase II randomised controlled trials (RCTs) testing belimumab (BLM), voclosporin (VCS) and obinutuzumab (OBI), respectively, in lupus nephritis (LN), as well as of anifrolumab (ANI) in general systemic lupus erythematosus (SLE) with encouraging results in LN.1–5 These trials overcame previous drawbacks in study design, introducing new approaches in the selection of endpoints and sample size, duration of follow-up and background treatment. The new developments provide the impetus for a critical appraisal of their place in the therapeutic armamentarium of SLE. This is highly timely, since these data were published after the 2019 updates of the EULAR and EULAR-ERA/EDTA recommendations for the management of SLE and LN, respectively.6 7
Recent trials in lupus nephritis: key findings
Belimumab
Based on hints from previous trials (ie, BLISS-52 and BLISS-76) for the beneficial role of add-on BLM in renal parameters, a phase III trial, BLISS-LN, tested its efficacy in LN population (table 1).1 8 At the end of follow-up, significantly more patients in the BLM group met the primary endpoint of renal response (43% vs 32%; OR, 1.6; 95% CI 1.0 to 2.3). Patients who received BLM also had a lower risk of renal-related events (a composite endpoint including end-stage kidney disease (ESKD); doubling of serum creatinine; increased proteinuria or impaired kidney function or kidney disease-related treatment failure) or death (HR, 0.51; 95% CI 0.34 to 0.77) while the safety profile was similar between groups.1 Of note, the relatively high glucocorticoid (GC) dose used in this trial raises the question whether lower GC doses would allow BLM to demonstrate its true efficacy. In this regard, in a posthoc analysis of patients who remained in the study after 24 weeks—when both mycophenolate mofetil (MMF) and GC were tapered—, patients treated with BLM had a lower risk of renal relapse (HR 0.45; 95% CI 0.28 to 0.72) and a lower rate of estimated glomerular filtration rate (eGFR) decline (eGFR slope difference 3.61; 95% CI 0.15 to 7.06), compared with placebo.9 Based on these results, BLM in combination with standard treatment was approved by both the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA), expanding its indications to include patients with LN.
Designs, characteristics and outcome measures of recent randomised controlled trials in lupus nephritis
Voclosporin
Following accumulating evidence in Asian patient populations pointing towards a beneficial role of calcineurin inhibitors (CNIs, mainly tacrolimus) in LN, two RCTs—phase II AURA-LV and phase III AURORA—examined the efficacy and safety of VCS in larger, multiethnic LN populations.2 10 VCS is structurally similar to cyclosporine A (CsA), except for a molecular modification that increases its potency up to fourfold compared with CsA and leads to fast elimination of metabolites, thus drug level monitoring is not required.11 12 In the AURORA trial, patients in the VCS group had significantly higher and earlier response rates compared with placebo (complete renal response rate at 52 weeks 41% vs 23%; OR 2.65; 95% CI 1.64 to 4.27) (table 1). Not surprisingly, this effect was mainly driven by the larger reduction in the levels of proteinuria, consistent with the antiproteinuric effect of CNIs. In January 2021, the FDA approved VCS, in combination with MMF, for the treatment of patients with LN.13
Obinutuzumab
Based on posthoc analysis of the LUNAR trial and other reports, which suggested that the degree and duration of B-cell depletion correlates with the rate of clinical response, efforts for a more potent B-cell depleting agent intensified.14 OBI, a humanised, glycoengineered, type II anti-CD20 monoclonal antibody with greater antibody-dependent cellular cytotoxicity and direct cell death potential, compared with rituximab has been recently tested in the NOBILITY trial, a phase II RCT in 125 patients with LN, with promising results.3 15 OBI was superior to placebo in achieving complete renal response at 52 and 104 weeks (35% vs 23%; difference 12%; 95% CI 3.4% to 28%; p=0.115, which was deemed statistically significant, and 41% vs 23%; percentage difference 19%; 95% CI 2.7% to 35%; p=0.026, respectively) without any significant safety concerns (table 1).3
Anifrolumab in general SLE and LN
Interferon (IFN) signalling has a key role in SLE pathogenesis. Following a successful phase II trial,16 ANI, a human monoclonal antibody to type I IFN receptor subunit 1, was investigated in two phase III RCTs, TULIP-1 and TULIP-2 in extrarenal SLE.4 17 TULIP-1 failed to reach its primary endpoint, the SLE Responder Index-4, but patients on ANI had better response rate in the British Isles Lupus Assessment Group (BILAG)-based Composite Lupus Assessment (BICLA). In contrast to the SLEDAI-based SRI, BICLA also captures partial responses and weights more skin compared with joint disease. In TULIP-2, more patients in the ANI group had a BICLA response compared with placebo at 52 weeks (47.8% vs 31.5%; difference 16.3%; 95% CI, 6.3 to 26.3; p=0.001). ANI showed a particular benefit in patients with predominant skin disease and enabled GC reduction. No major safety signals occurred; however, there were increased rates of herpes zoster with ANI compared with placebo (7.2% vs 1.1%, respectively).4 All cases were cutaneous and resolved without discontinuation of therapy.
As patients with active LN were excluded from the TULIP trials, a separate phase II RCT tested the efficacy and safety of ANI (both a basic and an intensified regimen) in LN (TULIP-LN).5 The 52-week analysis showed that the primary endpoint (relative difference in 24-hour urine protein−creatinine ratio) for the combined ANI groups vs placebo was not met (p=0.905). However, when the intensified regimen was separately compared with placebo, ANI achieved higher rates in many clinically meaningful endpoints, such as complete renal response, and GC tapering, together with improvements in extra-renal activity and serology (table 1).5
What therapy, for whom and when?
All these LN trials shared a common design adding the regimens under investigation to standard of care (SoC) which includes GC/hydroxychloroquine and an immunosuppressant. While these data are encouraging, the crucial question is whether all patients with active LN should be treated with the new agents from day one of treatment.
In support of the early use of add-on treatment, investigators have argued that LN is a severe disease with significant treatment-related and disease-related morbidity, and that standard treatment leaves many patients with incomplete response, increasing their risk of progression.18–21 Indeed, persistent proteinuria above 0.7 g/day after 1 year of treatment has been linked to adverse kidney outcomes.22 Moreover, despite the introduction of new therapies, the risk of ESKD has remained unchanged over the last 20 years and in large clinical trials, only one in three patients reaches a complete response after 1 year of SoC treatment.2 3 23–25 This has partly been attributed to adverse prognostic factors for kidney survival at the time of therapy initiation, such as low eGFR, nephrotic-range proteinuria, hypertension and relapsing disease in kidneys with accumulated damage.26 Thus, advocates of early combination treatment argue that, despite the moderate additional effect, all patients should be given the benefit of newer therapies, which may also enable faster GC tapering and decrease the risk of flares and damage accrual.27 28
On the other hand, there are valid counterarguments against a generalised use of these drugs in all patients, as add-on to SoC from day one. Although both BLISS-LN and AURORA trials allowed the inclusion of patients with low eGFR, average eGFR in the studied populations was over 90 mL/min/1.73 m2. Thus, the added value of the new agents in patients with high-risk features for ESKD needs further documentation. This, coupled with the modest treatment benefit from combination treatment (eg, 11% response difference between BLM and placebo) provides no definite reassurance, regarding its long-term impact on the risk of chronic kidney disease. Of note, in BLISS-LN, black patients and those treated with background cyclophosphamide (CYC) who had lower eGFR and higher baseline levels of proteinuria—traditionally considered as ‘high-risk’—had no additional benefit from BLM compared with placebo although both these subgroups involved a small proportion of the study population.1 Thus, one could argue that in patients with baseline kidney damage and impaired eGFR, add-on treatment with BLM may not be as efficacious, as such patients are more resistant to treatment.
Regarding the timing of treatment, while BLM with MMF/CYC were coadministered at the time of first diagnosis or flare in BLISS-LN, in the AURORA study more than 50% of patients were already treated with MMF at enrollment.1 2 In a subgroup analysis, the addition of VCS was beneficial only for patients who were already on MMF at the time of study entry, implying a beneficial effect of VCS mainly on inadequately responding/refractory disease. Finally, none of the trials addressed the important question about duration of add-on therapy and its impact on the risk of relapse. This is of special interest in the case of VCS, as CNI discontinuation has long been linked to an increased relapse risk.29
As long-term and real-life data are eagerly awaited, identification of the patients who are more likely to benefit from combination treatment is important. Acknowledging the fact that this patient stratification may be helped in the future by the discovery of –yet undefined- biomarkers, a more clinically-based approach may offer as a good starting point. In our opinion, patients with inadequate response (after the first 3–6 months), intolerance of maximum doses of standard treatment, or at increased risk for GC-related toxicity, are the best candidates. In addition, BLM may be considered in patients with concomitant serologic activity, while patients with nephrotic range proteinuria may benefit from the potent antiproteinuric effect of VCS. Combination treatment with the new agents may also be considered from the beginning in special groups. These include younger patients—who are more likely to have severe disease and have a longer, lifelong exposure to the disease and GC, thus incurring more damage—, patients with relapsing disease—who are at risk for increased damage accrual—, patients with severe proteinuria and increased thromboembolic risk and finally patients with a pressing need for rapid GC tapering, such as those with diabetes mellitus.
In the case of ANI, it is still too early to fully assess its potential impact on LN. New data suggesting a link between pathology in the skin and the kidney, with skin disease mirroring its kidney counterpart, may imply that improvements in cutaneous disease observed with ANI could be paralleled by decreased inflammation in the kidneys.30 31 In patients with general, extrarenal lupus, ANI could be added in patients with significant residual disease—especially skin and joints—to reach the current target of remission or low disease activity, and decrease GC use. The concern for herpes zoster infection may be alleviated with newer, inactivated vaccines, which will enable more patients with SLE to be vaccinated.
How could these data impact on the 2019 EULAR recommendations for SLE? Pending formal re-evaluation from the committee and based on drug-reimbursement policies in various countries, BLM and ANI may be added from the beginning on top of GC and hydroxychloroquine, with or without other conventional immunosuppressive drugs, or in patients unable to taper daily prednisone dose below 7.5 mg because of residual disease. In the case of LN, BLM and VCS may be used from the beginning especially in younger patients, patients with relapsing disease or marked proteinuria and, finally, patients with significant GC or disease-related damage.
Ethics statements
Patient consent for publication
Ethics approval
This study does not involve human participants.
References
Footnotes
Handling editor Josef S Smolen
Twitter @none
Contributors All authors contributed equally, read and approved the final manuscript.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Commissioned; externally peer reviewed.