BLISS in the Treatment of Lupus Nephritis

• lupus nephritis
• B cell
• biologic

The optimal treatment of lupus nephritis remains a vexing challenge for nephrologists. Induction therapy for patients with active disease usually includes cyclophosphamide or mycophenolate mofetil (MMF) in combination with corticosteroids. These treatments have improved outcomes. Yet, only 30%–50% of patients achieve a complete remission after induction therapy with these drugs, and >10% of patients with proliferative lupus nephritis progress to kidney failure despite treatment. Furthermore, these medications have significant side effects, including a higher risk of infections and loss of fertility. Because of these shortcomings, there has been a concerted effort to find better and safer treatments. In particular, investigators have sought agents that target the key molecular and cellular drivers of disease.

A large body of research indicates that lupus nephritis is mediated by immune-complex deposits within the glomeruli, or by autoantibodies that bind to glomerular antigens. There is, therefore, strong justification for evaluating drugs that deplete B cells. Unfortunately, B cell–targeted therapies have failed to improve patient outcomes in several clinical trials (1). This may be, at least in part, because of the limited efficacy of the drugs. A post hoc analysis of the Lupus Nephritis Assessment with Rituximab study, for example, revealed that patients in the rituximab group who achieved full depletion of their peripheral B cells were more likely to have a complete response (2). Obinutuzumab is more effective than rituximab, and unpublished data from a recent clinical trial indicate that addition of this drug to MMF improves the rate of complete response. Thus, B cells remain a promising therapeutic target.

Trial design may also have contributed to previous failures of B cell–targeted drugs. The experimental drugs were added to MMF- and steroid-containing regimens in most studies, and testing on a background of potent immunosuppression makes it more difficult to discern a drug’s specific benefits. Furthermore, the outcomes in these trials were reported at 48–52 weeks, which may not be a long enough period to detect differences between the treatment and placebo groups.

Given the previous disappointments, it is encouraging that the recently published Belimumab International Study in Lupus Nephritis (BLISS-LN) showed a significant benefit when belimumab was added to standard-of-care treatment in patients with lupus nephritis (3). Belimumab is an mAb to B cell–activating factor (BAFF), a B cell survival factor that also promotes the development of autoreactive B cells. There is a strong scientific rationale for targeting BAFF in lupus nephritis, and for using belimumab in combination with cytotoxic agents. Serum BAFF levels are elevated in patients with active lupus, and belimumab was previously shown to improve control of nonkidney lupus (4). A study in mice also demonstrated that BAFF levels increase during B cell reconstitution after treatment with cyclophosphamide, and that BAFF blockade prevented the regeneration of autoreactive B cells in this setting (5).

In the BLISS-LN study, 448 patients with proliferative or membranous lupus nephritis were randomized to treatment with belimumab or placebo, on top of induction treatment with cyclophosphamide- or MMF-containing regimens. The choice of cyclophosphamide or MMF was at the discretion of the treating physician. Induction treatment with cyclophosphamide was followed by maintenance therapy with azathioprine instead of MMF. Thus, the immunosuppressive regimens used in the cyclophosphamide and MMF subgroups were different throughout the entire follow-up period. Investigators could also use high-dose intravenous methylprednisolone at the start of treatment, although the study required that prednisone be tapered to 10 mg/d by 24 weeks.

Belimumab or placebo was added to these induction protocols as biweekly infusions, and continued for 2 years. The original primary outcome for this study used an ordinal end point (complete, partial, or no kidney response). Partway through the study, however, the investigators changed the primary outcome to a novel composite end point, termed “primary efficacy renal response” (PERR). The PERR comprised a urine protein-to-creatinine ratio of <0.7, an eGFR within 20% of the preflare value or ≥60 ml/min per 1.73 m², and no use of rescue therapy. The proteinuria threshold of 0.7 in the PERR is supported by the observation that a reduction of the urine protein-to-creatinine ratio to <0.8 predicts favorable kidney outcomes (6).

More patients in the belimumab group than in the placebo group achieved a PERR by 24 weeks, and the benefit was maintained to 104 weeks (43 versus 32%). Treatment was also associated with an increase in complete kidney response as originally defined (30% versus 20% at 104 weeks). When secondary end points were examined, the belimumab group had a greater reduction in proteinuria and stability of eGFR. Treatment with belimumab was also associated with greater normalization of disease biomarkers, including levels of anti–double-stranded DNA antibodies, anti-C1q antibodies, C3, and C4. In subgroup analysis, treatment with belimumab was associated with improved outcomes in Black patients. It also appeared to be beneficial in both the MMF and cyclophosphamide subgroups, although the increase in PERR was not statistically significant in the cyclophosphamide group. The rate of infections was similar in patients who received belimumab or placebo. This is important, as a previous study of atacicept in lupus nephritis was stopped early because of a higher risk of infection (7). Atacicept has a similar mechanism of action to belimumab, except that it blocks both BAFF and APRIL (“a proliferation-inducing ligand”), another B cell growth factor.

Why was belimumab effective for lupus nephritis, whereas rituximab and ocrelizumab have not been? It is possible that belimumab is a more effective drug for this disease, but the larger size and longer duration of BLISS-LN also undoubtedly gave the study greater power than the previous trials. It also probably helped that BLISS-LN mandated tapering of the steroids. In addition, the use of PERR as the end point may have been an advantage. Although the more liberal proteinuria cutoff increased the number of responders in both the drug and placebo groups, post hoc analyses of other lupus trials have shown that such changes can increase the ability to discern differences between treatments (8).

It is not yet clear how nephrologists should incorporate the results of the BLISS-LN into their clinical practice. The addition of belimumab to the standard induction protocols for lupus nephritis is safe and improves outcomes. Unfortunately, even with this three-drug regimen, more than half of the patients still had an unsatisfactory clinical response. Rather than changing the way we treat all of our patients with lupus nephritis, it is tempting to add belimumab to treatment regimens for those patients who have the most severe disease, or have an inadequate response to induction treatment. However, treatment with belimumab did not increase the rate of complete response in the cyclophosphamide subgroup. These patients had slightly worse kidney disease at baseline than the MMF group, arguing against simply reserving belimumab for patients who have severe disease. The recent Combination of Antibodies in Lupus Nephritis: Belimumab and Rituximab Assessment of Tolerance and Efficacy study tested whether belimumab was safe and beneficial when administered after treatment with cyclophosphamide, rituximab, and steroids in 43 patients with recurrent or refractory disease (9). The addition of belimumab reduced the generation of autoreactive cells during B cell reconstitution, but it did not significantly improve clinical outcomes in that study.

Several other new drugs have looked promising in recent lupus nephritis trials. As mentioned above, obinutuzumab may be effective when added to the standard of care. Similarly, the addition of voclosporin (a new calcineurin inhibitor) to MMF improved outcomes in the Aurinia Urinary Protein Reduction Active Lupus with Voclosporin (10) and unpublished Aurinia Renal Response in Active Lupus with Voclosporin studies (ClinicalTrials.gov; NCT03021499). Although the design and end points varied slightly among BLISS-LN and these other trials, the addition of the test drug to standard-of-care treatment led to a complete response in approximately 40% of participants across these studies. The drugs also seemed to benefit only a subgroup of patients in each case, and no specific histologic findings or serologic biomarkers have been identified to predict a response (or lack of response) to the new treatments.

The BLISS-LN investigators and the patients should be commended for completing such a challenging study. A total of 797 patients were screened at 107 sites in 21 countries. Patients received intravenous infusions for 2 years, and only five patients were lost during follow-up. As has been pointed out many times, belimumab is the first drug in more than 50 years to be approved for the treatment of lupus, and BLISS-LN is an important advancement in the field. Even with these promising results and those seen for other new drugs, it is critical that additional studies be performed to determine how these medications can be used most effectively. Because many patients do not respond to the add-on therapies, we urgently need predictive biomarkers that identify those patients most likely to benefit from their use. We also need a better mechanistic understanding of how the various drugs affect autoimmunity, to provide a rational basis for choosing combinations of drugs that are most likely to be effective and safe. Otherwise, clinicians will simply be left to unsystematically add more and more immunosuppressive drugs when patients fail to respond to standard treatment.

• Copyright © 2021 by the American Society of Nephrology