Treating Frequently Relapsing and Steroid-Dependent Nephrotic Syndrome

Nephrotic syndrome is a common childhood kidney disease associated with high treatment-related morbidity.¹ Most cases are steroid-sensitive, yet up to half develop frequently relapsing or steroid-dependent disease. These children receive multiple steroid courses and often steroid-sparing immunosuppressive medications to prevent relapses and avoid steroid toxicity. Rituximab, a chimeric mouse/human monoclonal antibody targeting CD20, is used in those that remain resistant to common second-line agents (cyclophosphamide, tacrolimus, or mycophenolate).¹ Rituximab effectively depletes B cells, which leads to a sustained remission in nephrotic syndrome and is superior to placebo or tacrolimus in randomized controlled trials (RCTs). Controversies regarding the optimal dosing, frequency, and monitoring of rituximab therapy remain, and there is significant global practice variation. Importantly, there is treatment effect heterogeneity resulting in some children experiencing no B-cell depletion, rapid B-cell repopulation, or relapsing despite B-cell depletion. One third form antidrug antibodies, which may reduce efficacy and lead to infusion reactions.² Rituximab is a type 1 antibody, acting primarily through complement- and antibody-dependent cytotoxicity, with minimal direct cell death.

To overcome rituximab's limitations, several novel anti–B-cell therapies (ofatumumab, obinutuzumab, and daratumumab) were developed to treat B-cell malignancies. In a recent single-center Italian RCT, no benefit was shown for ofatumumab (versus rituximab) among 140 children and young adults with steroid-dependent nephrotic syndrome.³ There is evidence that obinutuzumab is more efficacious (versus rituximab) in preclinical and oncology RCTs, which has led to interest in using obinutuzumab to treat autoimmune kidney diseases. Obinutuzumab is a fully humanized monoclonal antibody against CD20. It is a type 2 antibody, with greater direct and antibody-dependent cytotoxicity through caspase-independent mechanisms.⁴ The glycoengineered Fc region also has greater affinity for effector cell binding in individuals with certain genetic polymorphisms.

In this issue of CJASN, Dossier et al. report a retrospective, single-center cohort study including 41 children with frequently relapsing or steroid-dependent nephrotic syndrome treated with a single low-dose obinutuzumab infusion (300 mg/1.73 m²).⁵ Obinutuzumab's indication was rituximab resistance (lack of B-cell depletion by 1 month) in 20%, short duration (<3 months) of B-cell depletion in 15%, or relapse after rituximab in 66%. For the third group, obinutuzumab was used at clinician discretion for children that relapsed early after B-cell repopulation or continued to relapse after several rituximab courses. All children achieved B-cell depletion postobinutuzumab (including all five cases with antirituximab antibodies) for a median 8-month duration. Sustained remission was present in 92% of children by 1 year and 68% by 2 years postobinutuzumab. Adverse events were comparable with rituximab; 12% experienced infusion reactions and 21% neutropenia (all resolved after stopping trimethoprim-sulfamethoxazole prophylaxis). Hypogammaglobulinemia was common at baseline. Obinutuzumab resulted in significant decreases in IgM, but no change in IgG. One child experienced a severe drug-related acute lung injury, requiring intensive care unit admission.

The study was conducted at a single academic center in France with significant experience using novel anti–B-cell therapies. Strengths include the research question's novelty, relatively large sample size of children with complicated frequently relapsing or steroid-dependent nephrotic syndrome and rituximab treatment failure, the ≥2-year follow-up without any attrition, and the monthly B-cell subset and immunoglobulin monitoring. However, several limitations should be considered when interpreting their results. First, this is a retrospective observational study comparing obinutuzumab versus prior rituximab courses (i.e., without a true control arm). Two thirds of cases received obinutuzumab for relapse after B-cell repopulation postrituximab. These children were selected at clinician discretion. The median number of rituximab infusions preobinutuzumab was only one, and rituximab was given as a single 375 mg/m² infusion. Many of these children may have responded to higher rituximab doses, repeated courses, and/or concurrent maintenance immunosuppression, which have been shown to substantially reduce relapse risk postrituximab.⁶ Furthermore, few cases were tested for antirituximab antibodies, which could have provided justification for obinutuzumab use. Because obinutuzumab was indicated on the basis of rituximab treatment failure, comparison against previous rituximab courses creates a serious selection bias. These analyses should be considered exploratory, justifying future RCTs comparing obinutuzumab directly to current standard of care. Novel observational research methods, such as target trial emulation, could have been used to align the analysis of this study with that of a hypothetical RCT, using strict selection criteria, a concurrent control arm, and adjustment for measured confounders.

The preliminary data on obinutuzumab are promising. One-year relapse-free survival after obinutuzumab was 92%, compared with 50%–95% after rituximab in recent RCTs.⁷ After 2 years, 68% of children remained in remission after obinutuzumab versus only 10%–25% after rituximab in several observational studies.⁸ An RCT is now needed to compare obinutuzumab versus rituximab in nephrotic syndrome. The Efficacy and Safety of Obinutuzumab Versus Rituximab in Childhood Steroid Dependant and Frequent Relapsing Nephrotic Syndrome (OBIRINS) trial (NCT05786768) is a double-blind, multicenter RCT currently planned (anticipated enrollment=88 participants) by this study's authors to evaluate effectiveness of obinutuzumab in childhood frequently relapsing or steroid-dependent nephrotic syndrome. If obinutuzumab is superior, cost will be a key consideration in treatment decision making. With biosimilars available, the cost of rituximab has decreased considerably, making obinutuzumab approximately 35× more expensive ($110 versus $3900 USD per dose for an average-sized 10-year-old child, based on Canadian data).⁹ Therefore, even if superior, obinutuzumab use might remain restricted to children who have failed rituximab treatment (i.e., absent or short B-cell depletion±antirituximab antibodies). Another industry-sponsored RCT of obinutuzumab versus mycophenolate mofetil in children and young adults with nephrotic syndrome (INShore, NCT05627557) is also underway.

The current study raises questions about the optimal dosing of obinutuzumab in nephrotic syndrome. Recent obinutuzumab RCTs for other autoimmune kidney diseases have typically administered two 1000 mg/1.73 m² infusions 2 weeks apart, on the basis of the oncology literature. However, this study demonstrates that a significantly lower obinutuzumab dose (300 mg/1.73 m²) still effectively depletes B cells, which could mitigate hypogammaglobulinemia and other adverse events. Another question is whether all children with suboptimal responses to rituximab should be tested for antidrug antibodies. Although present in up to one third of children with nephrotic syndrome receiving rituximab, there are conflicting data regarding their effect on the duration of B-cell depletion and relapse-free survival. Like rituximab, there is also the potential for formation of antidrug antibodies to obinutuzumab. Although obinutuzumab is fully humanized and, therefore, less immunogenic, it is generated in a Chinese hamster ovarian cell line. Antibodies may form against attached glycans as a result.¹⁰ These are both key questions that should be addressed by future RCTs.

This study is emblematic of a larger issue in child health research, which is the lack of RCTs evaluating novel treatment strategies. In pediatrics, new treatments are often clinically introduced off-label, on the basis of data from adult RCTs or retrospective observational studies. This creates barriers to the subsequent conduct of pediatric RCTs to answer key research questions and establish dosing in children. We need to consider pilot trials before RCT initiation, to test feasibility with strict selection criteria and dosing strategies. This would provide important information for larger, more generalizable clinical trials. Less than 60% of initiated pediatric RCTs are ever completed and published, and the most common reason for discontinuation is poor study recruitment.¹¹ This is particularly relevant to therapeutic trials in nephrotic syndrome, where recruitment difficulties have historically led to underpowered studies (e.g., the National Institutes of Health FSGS clinical trial). Potential solutions to overcome these challenges include pilot trials, design of adaptive Bayesian RCTs to increase efficiency, other nontraditional study designs to optimize sample size (e.g., crossover trials), development of international clinical trial networks and disease registries to facilitate recruitment, and increased training of pediatric clinical trialists.

In childhood nephrotic syndrome, we have a pattern of adopting medications into clinical practice on the basis of small observational studies. For example, tacrolimus use was first reported in two cases of childhood nephrotic syndrome in 2003, which quickly led to widespread practice change in the absence of RCTs comparing effectiveness against other steroid-sparing medications such as cyclophosphamide. With rituximab, 10 years passed between case reports of its effectiveness in nephrotic syndrome and the first published RCTs. We now have another opportunity to rigorously evaluate a new nephrotic syndrome treatment before it becomes adopted into practice without a strong evidence basis. This step is critical because long-term rates of remission and safety after years of treatment with rituximab and obinutuzumab remain unknown.

Although the comparative efficacy of obinutuzumab versus rituximab is important to understand, the results are only applicable to a small fraction of children with steroid-sensitive nephrotic syndrome. RCTs designed to answer broader questions, including optimizing steroid treatment regimens for relapses, evaluating novel relapse prevention strategies (e.g., dietary modification, galactose therapy), identifying the most effective first-line steroid-sparing medications, and reporting on long-term patient-centered outcomes (i.e., prolonged remission) should be prioritized. Most steroid-sparing medications effectively prevent relapses, but do not modify the long-term disease trajectory or likelihood of permanent remission into adulthood. Until treatments are developed that effectively induce long-term remission, the focus of management is to reduce relapse frequency, minimize steroid exposure, and monitor for spontaneous disease remission.

The decision to move to new antibody-directed therapies in childhood nephrotic syndrome should be driven by evidence. It is important for pediatric nephrologists to know there is an alternative for children with complicated, unremitting disease. However, we must await more definitive data from upcoming RCTs before we can answer whether to Obi or not to Obi.

Disclosures

R.S. Parekh is a site PI for the Vertex trial of APOL1 and CKD and reports ownership interest in Coramed stock, SpineFx, and Synaptive-stock; research funding from Canadian Institute of Health Research (CIHR), NIH, and Ministry of Colleges and Universities; patents or royalties from IZI and SpineFx; serving as an Associate Editor of CJASN, a Board Member of Baycrest Academy, a Board Member of Baycrest Hospital, a Board Member of Bishop Strachan School, and a Board Member of Conference of Independent Schools of Ontario; and an advisory or leadership role for ISN Council. R.S. Parekh’s spouse reports patents or royalties from Coramed and serves as an officer of Coramed and SpineFx. The remaining author has nothing to disclose.

Funding

None.

Author Contributions

Conceptualization: Rulan S. Parekh, Cal H. Robinson.

Writing – original draft: Cal H. Robinson.

Writing – review & editing: Rulan S. Parekh, Cal H. Robinson.