Volume 76, Issue 2 pp. 247-254
Full Length
Open Access

Kidney Outcomes and Preservation of Kidney Function With Obinutuzumab in Patients With Lupus Nephritis: A Post Hoc Analysis of the NOBILITY Trial

Brad H. Rovin

Corresponding Author

Brad H. Rovin

The Ohio State University Wexner Medical Center, Columbus

Address correspondence via email to Brad H. Rovin, MD, at [email protected].

Search for more papers by this author
Richard A. Furie

Richard A. Furie

Northwell Health, Great Neck, New York

Search for more papers by this author
Jorge A. Ross Terres

Jorge A. Ross Terres

Genentech, Inc, South San Francisco, California

Search for more papers by this author
Sophia Giang

Sophia Giang

Genentech, Inc, South San Francisco, California

Search for more papers by this author
Thomas Schindler

Thomas Schindler

F. Hoffmann-La Roche Ltd, Basel, Switzerland

Search for more papers by this author
Armando Turchetta

Armando Turchetta

Hoffmann-La Roche Ltd, Mississauga, Ontario, Canada

Search for more papers by this author
Jay P. Garg

Jay P. Garg

Genentech, Inc, South San Francisco, California

Search for more papers by this author
William F. Pendergraft III

William F. Pendergraft III

Genentech, Inc, South San Francisco, California

Search for more papers by this author
Ana Malvar

Ana Malvar

Hospital Fernandez, Buenos Aires, Argentina

Search for more papers by this author
First published: 10 November 2023

ClinicalTrials.gov identifier: NCT02550652.

Supported by F. Hoffmann-La Roche Ltd.

Qualified researchers may request access to individual patient level data through the clinical study data request platform (https://vivli.org/). Further details on Roche's criteria for eligible studies are available at https://vivli.org/members/ourmembers/. For further details on Roche's Global Policy on the Sharing of Clinical Information and how to request access to related clinical study documents, see https://www.roche.com/research_and_development/who_we_are_how_we_work/clinical_trials/our_commitment_to_data_sharing.htm.

Author disclosures are available at https://onlinelibrary.wiley.com/doi/10.1002/art.42734.

Abstract

Objective

To determine whether adding obinutuzumab to standard-of-care lupus nephritis (LN) therapy could improve the likelihood of long-term preservation of kidney function and do so with less glucocorticoids.

Methods

Post hoc analyses of the phase II NOBILITY trial were performed. Time to unfavorable kidney outcome (a composite of treatment failure, doubling of serum creatinine, or death), LN flare, first 30% and 40% declines in estimated glomerular filtration rate (eGFR) from baseline, and chronic eGFR slope during the trial were compared between patients with active LN who were randomized to take obinutuzumab (n = 63) or placebo (n = 62) in combination with mycophenolate mofetil and glucocorticoids. The number of patients who achieved complete renal response (CRR) on 7.5 mg or less per day of prednisone was also determined.

Results

Obinutuzumab reduced the risk of developing the composite kidney outcome by 60%, LN flare by 57%, and first eGFR decline of 30% or 40% by 80% and 91%, respectively. Patients receiving obinutuzumab had a significantly slower decline in eGFR than patients receiving placebo, with an annualized eGFR slope advantage of 4.1 ml/min/1.73 m2/year (95% confidence interval 0.14–8.08). Overall, 38% of patients receiving obinutuzumab compared with 16% of patients receiving placebo achieved CRR at week 76 while receiving 7.5 mg or less per day of prednisone (P < 0.01); at week 104, the difference did not achieve significance (38% vs 22%; P = 0.06).

Conclusion

Post hoc analyses of NOBILITY demonstrated that compared with standard-of-care therapy, obinutuzumab treatment resulted in superior preservation of kidney function and prevention of LN flares. More patients achieved CRR at week 76 with less glucocorticoid use in the obinutuzumab group.

INTRODUCTION

Lupus nephritis remains a severe and therapeutically challenging manifestation of systemic lupus erythematosus.1 Despite numerous advances in the treatment of active lupus nephritis, including the regulatory approvals of belimumab in 2020 and voclosporin in 2021,2, 3 complete renal response (CRR) rates remain low.4, 5 It is also clear that lupus nephritis relapse portends a worse long-term kidney and patient prognosis.6-8

Preservation of long-term kidney function is a major therapeutic goal in lupus nephritis. As lupus nephritis clinical trials generally span 12 to 24 months, there has been an effort to identify study endpoints that serve as surrogate markers for long-term kidney health. It has been shown that reduction in proteinuria to 0.7 grams or less per day after 12 months of treatment is associated with good long-term kidney outcomes.9-11 Studies also suggest that a reduction in the rate of decline of estimated glomerular filtration rate (eGFR), measured as eGFR slope over 2 or more years, reflects improved long-term kidney survival, including for patients with lupus nephritis.12, 13 Additionally, it has become increasingly apparent that reducing glucocorticoid use is desirable in the management of patients with lupus nephritis, primarily to reduce both short- and long-term toxicities.8, 14-17

Placebo-controlled trials in patients with active lupus nephritis treated with the type I anti-CD20 antibodies rituximab and ocrelizumab were unsuccessful.18, 19 An important factor in assessing whether or not an anti-CD20 agent is efficacious is to ensure that B cell depletion has in fact occurred when an anti-CD20 agent is used. Variability in the depth and duration of B cell depletion can impact efficacy assessments.

Obinutuzumab is a recombinant, humanized type II anti-CD20 IgG1 monoclonal antibody glycoengineered to enhance antibody-dependent cell-mediated cytotoxicity and phagocytosis. It was approved in 2013 by the US Food and Drug Administration for use in patients with previously untreated chronic lymphocytic leukemia; patients with follicular lymphoma who relapsed after or are refractory to, a rituximab-containing regimen; and patients with previously untreated stage II bulky, III, or IV follicular lymphoma.20, 21 In the randomized, double-anonymized, placebo-controlled, phase II NOBILITY trial (NCT02550652), patients with active proliferative lupus nephritis receiving obinutuzumab with standard-of-care (SOC) therapy showed clinically meaningful improvements in CRR and overall renal responses at weeks 52, 76, and 104 compared with those receiving placebo and SOC therapy.22 The success of obinutuzumab was likely due, at least in part, to the depth and duration of B cell depletion compared with other anti-CD20 antibodies.23

The NOBILITY trial was conducted over 2 years. This afforded an opportunity to investigate surrogate long-term kidney survival endpoints through a post hoc analysis of eGFR slope, eGFR stability, and durability of CRR without lupus nephritis flare. Additionally, we examined the capacity of obinutuzumab to spare glucocorticoid use.

PATIENTS AND METHODS

Study design, settings, and populations

The NOBILITY trial (ClinicalTrials.gov identifier NCT02550652) was a phase II, multinational, multicenter, randomized, double-anonymized, placebo-controlled, 104-week trial to assess the safety and efficacy of obinutuzumab versus placebo in combination with mycophenolate mofetil (MMF) and glucocorticoids in adult patients with active lupus nephritis. The study design and inclusion criteria have been published previously.22 All patients met the American College of Rheumatology classification criteria for systemic lupus erythematosus24 and had biopsy-proven active lupus nephritis.25 This trial was executed in accordance with the principles of the Declaration of Helsinki and Good Clinical Practice. All patients provided informed consent.

Randomizations and interventions

Patients were randomly assigned (1:1) to receive either obinutuzumab 1,000 mg or placebo infusions on day 1 and weeks 2, 24, and 26. Patients assigned to obinutuzumab received premedication with methylprednisolone 80 mg intravenously to reduce the risk of infusion-related reactions, whereas patients assigned to placebo received placebo methylprednisolone. All patients received MMF (target dose 2–2.5 g per day or equivalent dose of mycophenolic acid). Protocol-mandated glucocorticoid treatment included methylprednisolone (1,000 mg before or during screening and up to a total of 3,000 mg before randomization) and an oral glucocorticoid regimen (initial prednisone dose: 0.5 mg per kg per day, maximum 60 mg per day, with taper to 7.5 mg per day by week 12).

Endpoint measurements and definitions

The time to an unfavorable kidney outcome was defined as the first of the following events: treatment failure (any cause), serum creatinine doubling, or death (any cause). Lupus nephritis flare was assessed from week 24 and was diagnosed if one of the following conditions occurred: (1) eGFR decrease more than 20% compared with week 24 in patients with urine protein to creatinine ratio (UPCR) greater than 1 g/g and/or cellular casts; (2) UPCR increase to greater than 1 g/g if week 24 UPCR was less than 0.2 g/g, to greater than 2.0 g/g if week 24 UPCR was 0.2 to 1 g/g, or doubling if week 24 UPCR was greater than 1 g/g; or (3) treatment failure was declared if patients received prohibited medications to treat active lupus nephritis (ie, receipt of rescue therapy). Lupus nephritis flare was assessed at week 24 to provide sufficient time for patients to have had a response (partial or complete) to treatment. Time to first eGFR decline of 30% or 40% was defined as the time to the first visit in which a 30% or 40% decline in eGFR from baseline occurred, respectively. For all time-to-event endpoints, individual missing visits were ignored, whereas participants who missed more than one consecutive visit were censored on the week of the last recorded visit.

Sustained eGFR declines of 30% or 40% were defined as 30% or 40% reductions from baseline at both of the last two visits. If one of the last two visits was missing, data imputation was performed using the last measurement carried forward. The eGFR slope decline was assessed from week 12 to week 104. This chronic eGFR slope was used to mitigate the influence of any acute changes in eGFR that may have occurred in response to treatment, for example, exposure to high-dose glucocorticoids. Chronic slope more accurately reflects the benefits of treatment and avoids including rapid eGFR fluctuations in the calculation that may be mediated by medications that can acutely influence glomerular hemodynamics or muscle creatinine production (eg, high-dose glucocorticoid).12, 13

Statistical analysis

The time-to-event endpoints (time to first unfavorable kidney outcome [treatment failure, doubling of serum creatinine, and death], lupus nephritis flare, and 30% or 40% eGFR declines from baseline) were analyzed via Cox proportional hazards regression models. The Cox regression model for the time to first kidney outcome was adjusted for region (Latin America, non-Latin America) and baseline serum creatinine. For the time to first lupus nephritis flare, the model was adjusted for region (Latin America, non-Latin America) and UPCR and eGFR at week 24. Because lupus nephritis flares were identified starting after week 24, study participants who received rescue therapy before week 24 were removed from the analysis. For the first eGFR decline, the Cox model was adjusted for region and baseline eGFR. Note that the primary analysis of the NOBILITY trial included race and a different dichotomization of the region variable as stratification factors; however, because of the limited number of events across subgroups, this analysis was adjusted for different factors. For each Cox model, the proportional hazards assumption was tested via a global chi-square test of the scaled Schoenfeld residuals. In cases in which the test suggested a violation of this assumption, a sensitivity analysis was conducted to re-estimate the confidence intervals of the hazard ratios via bootstrapping. This allows interpretation of the estimated hazard ratios from the Cox models as a weighted average of the true time-varying hazard ratios over the follow-up period.26 The eGFR slope was assessed from week 12 in a linear mixed-effects model, which included the treatment group, week visit, their interaction term, and random intercept and slope. No statistical test was performed on the sustained eGFR decline endpoint because of the limited number of events.

The dose of prednisone that defined glucocorticoid sparing was 7.5 mg or less per day. Assessment of a glucocorticoid-sparing effect while achieving CRR at weeks 76 and 104 (without use of >7.5 mg per day of prednisone between weeks 64 and 76 and Weeks 92 and 104, respectively), was calculated by the Cochran–Mantel–Haenszel test stratified for race (Afro-Caribbean/African American vs others) and country (US vs non-US). Cumulative glucocorticoid dose was evaluated from week 64 to week 76. Glucocorticoid dose was imputed by the last observation carried forward method for patients who discontinued the study and/or for missing data. All P values < 0.05 were considered to be statistically significant. Analyses were post hoc; thus, results should be considered descriptive.

RESULTS

In the NOBILITY trial, 125 patients were randomized and received obinutuzumab (n = 63, mean [±SD] age 33.1 [9.8], 87% women) or placebo (n = 62, mean [±SD] age 31.9 [10.1], 82% women) along with SOC therapy (MMF and glucocorticoids); additional baseline demographics and clinical characteristics have been published previously.22 The NOBILITY trial met its primary efficacy endpoint, and no new safety signals were identified.22

Compared with placebo, obinutuzumab significantly reduced the risk of an unfavorable kidney outcome (treatment failure, serum creatinine doubling, or death) by 60% (hazard ratio [HR] 0.40, 95% confidence interval [95% CI] 0.20–0.80) (Figure 1), a lupus nephritis flare by 57% (HR 0.43, 95% CI 0.20–0.95) (Figure 2), and first 30% or 40% eGFR declines from baseline by 80% (HR 0.20, 95% CI 0.06–0.61) and 91% (HR 0.09, 95% CI 0.01–0.73), respectively (Figure 3). The total numbers of unfavorable kidney outcomes (obinutuzumab vs placebo) were as follows: 12 versus 24 for treatment failure, 1 versus 6 for creatinine doubling, and 1 versus 4 for death, respectively.

Details are in the caption following the image
Time to unfavorable kidney outcome over 104 weeks in patients with proliferative lupus nephritis treated with obinutuzumab and standard-of-care (SOC) therapy versus placebo and SOC therapy. The dashed lines represent the 95% confidence intervals (95% CIs). *Unfavorable kidney outcomes include treatment failure, doubling of serum creatinine, and death. HR, hazard ratio.
Details are in the caption following the image
Time to lupus nephritis flare over 104 weeks in patients with proliferative lupus nephritis treated with obinutuzumab and standard-of-care (SOC) therapy versus placebo and SOC therapy. The dashed lines represent the 95% confidence intervals (95% CIs). *Lupus nephritis (LN) flare was identified from week 24 if one of the following conditions applied: impaired kidney function compared with week 24, defined as estimated glomerular filtration rate (eGFR) decrease greater than 20% from week 24 accompanied by proteinuria (urine protein to creatinine ratio [UPCR] >1 g/g) and/or cellular casts; increase in proteinuria compared with week 24, defined as an increase in UPCR over 1 if week 24 UPCR was less than 0.2 g/g, UPCR of greater than 2.0 g/g if week 24 UPCR was 0.2 to 1 g/g, or as a doubling of UPCR if week 24 UPCR was over 1 g/g; and treatment failure due to kidney disease-related intake of prohibited medication (receipt of rescue therapy). HR, hazard ratio.
Details are in the caption following the image
Time to first 30% and 40% estimated glomerular filtration rate (eGFR) decline from baseline over 104 weeks in patients with proliferative lupus nephritis treated with obinutuzumab and standard-of-care (SOC) therapy versus placebo and SOC therapy. The dashed lines represent the 95% confidence intervals (95% CIs). HR, hazard ratio.

Although the estimated scaled Schoenfeld residuals from the Cox regression models did not show any evident deviation from the proportional hazards assumption, the global test returned P values < 0.05 for the kidney outcome event and lupus nephritis flare endpoints, implying that the proportional hazards assumption was likely not met. However, the bootstrapped 95% CIs of the hazard ratios were only slightly larger than those estimated by the Cox model (0.17–0.79 for the kidney outcome and 0.15–0.98 for the lupus nephritis flare) and did not lead to different conclusions.

Sustained eGFR declines of 30% or 40% were numerically lower in the obinutuzumab treatment group; however, because of the limited number of events, no statistical testing was performed. A significant difference in attenuation of eGFR slope decline was observed between patients in the obinutuzumab arm and placebo arm, with an annualized slope advantage of 4.10 ml/min/1.73 m2/year, 95% CI 0.14–8.08, and P = 0.043 (obinutuzumab slope, −0.43 [95% CI −3.16 to 2.31] vs placebo slope, −4.52 [95% CI −7.41 to −1.66]) (Figure 4).

Details are in the caption following the image
Population-level predicted estimated glomerular filtration rate (eGFR) in patients with proliferative lupus nephritis treated with obinutuzumab (OBI) and standard-of-care (SOC) therapy versus placebo (PBO) and SOC therapy. 95% CI, 95% confidence interval.

More patients in the obinutuzumab group than the placebo group were able to maintain a daily prednisone dose of 7.5 mg or less from week 64 through week 76 and achieve CRR at week 76 (obinutuzumab: 24 of 63 patients [38%] vs placebo: 10 of 62 patients [16%]; P < 0.01) (Figure 5). The median cumulative dose of prednisone from weeks 64 through 76 was 525 mg in the obinutuzumab group versus 630 mg in the placebo group. At week 104, numerically more patients in the obinutuzumab group than the placebo group were able to achieve CRR and maintain a daily prednisone dose of under 7.5 mg from weeks 92 through 104 (38% vs 22%, respectively), but this difference did not achieve significance (P = 0.06).

Details are in the caption following the image
Complete renal response (CRR) and 7.5 mg or less per day of prednisone at weeks 76 and 104 in patients with proliferative lupus nephritis treated with obinutuzumab and standard-of-care (SOC) therapy versus placebo and SOC therapy. *At weeks 76 and 104, patients received 7.5 mg or less per day of prednisone from weeks 64 through 76 or weeks 92 through 104, respectively.

DISCUSSION

Obinutuzumab is a humanized type II anti-CD20 monoclonal antibody that was shown to be more effective than SOC alone in the treatment of lupus nephritis in the phase II NOBILITY trial.22 Significantly more patients treated with SOC plus obinutuzumab achieved a CRR than patients treated with SOC plus placebo. Because the NOBILITY trial followed patients for 2 years, we were able to determine the impact of obinutuzumab on the preservation of kidney function in this post hoc analysis. Patients treated with obinutuzumab plus SOC were significantly less likely to achieve a composite endpoint of treatment failure, doubling of serum creatinine, or death than patients receiving SOC with a risk reduction of 60%. Because there were few deaths in either arm of the NOBILITY trial, this outcome was driven by more treatment failures and potential loss of kidney function in the placebo arm. Furthermore, the risks of 30% or 40% reductions in eGFR (transient) were significantly lower in the obinutuzumab treatment group compared with the placebo group. Sustained declines in eGFR of 30% or 40% were numerically fewer in the obinutuzumab group, but the total number of events was small, which precluded statistical analysis. Although eGFR declined in both groups over the course of the trial, the rate of decline (measured as eGFR slope) was significantly faster in patients who received placebo. Investigation of the eGFR advantage for patients receiving obinutuzumab revealed an annual slope difference of 4.10 ml/min/1.73 m2/year. Finally, the risk of new lupus nephritis flares was 57% lower in patients receiving obinutuzumab. These findings suggest that the addition of obinutuzumab to SOC affords substantial protection to the kidneys in patients with lupus nephritis.

Although doubling of serum creatinine (equivalent to a 57% decline in GFR) is well-understood to be a surrogate of impending future kidney failure,27 it can take a long time for patients to reach this endpoint, and it is, therefore, an impractical endpoint for clinical trials. However, a recent consensus conference suggested that a sustained 40% decline in eGFR is a reasonable surrogate for eventual kidney failure.27 A 40% reduction in eGFR over 2 years represents sufficient chronic kidney damage to translate to a 10-fold greater risk of kidney failure among patients with impaired kidney function (ie, eGFR <60 ml/min/1.73 m2).28 Even a sustained 30% decline in eGFR is associated with enough chronic parenchymal damage to be predictive of future kidney failure, especially for trials in which the intervention does not result in acute changes in eGFR.29 Given that the main goal of lupus nephritis therapy is to preserve kidney survival, the fact that patients receiving obinutuzumab had eGFR declines of 30 or 40% significantly less often than patients receiving SOC alone suggests that the addition of obinutuzumab to initial lupus nephritis treatment regimens will be of benefit to patients’ long-term kidney health.

Another way to assess the impact of a new therapeutic on preservation of kidney function is through eGFR slope analysis. If a novel therapy can attenuate the rate of decline in eGFR (ie, eGFR slope) by 0.75 ml/min/1.73 m2 or more over 2 years, the risk of progression to kidney failure is reduced by 16% to 21%.12 The reduction in eGFR slope in patients treated with obinutuzumab was over 4 ml/min/1.73 m2/year (95% CI 0.14–8.08; P = 0.043), suggesting significant protection against eventual kidney failure.

Treatment with obinutuzumab reduced the risk of lupus nephritis flare by 57%. Disease flare is a major risk factor for loss of kidney function, and prevention of flares is an important therapeutic goal for lupus nephritis management.7, 30 With every lupus nephritis flare, damage to the kidney accumulates.31 By reducing flare risk, obinutuzumab should decrease the accumulation of chronic parenchymal kidney damage.

Our results for eGFR slope, 30% to 40% decline in eGFR, and lupus nephritis flare are similar to the findings of a post hoc analysis of the phase III BLISS-Lupus Nephritis trial that evaluated the efficacy of belimumab added to SOC for the treatment of active lupus nephritis.2, 32 Like obinutuzumab in the phase II NOBILITY trial, patients treated with belimumab demonstrated an attenuated eGFR slope, less frequently progressed to a sustained 30% or 40% loss of GFR, and had fewer lupus nephritis flares.32 These findings raise the question of whether B cell–targeted therapies in general may be effective at slowing progression of chronic kidney injury and attenuating lupus nephritis flare rate.

In this post hoc analysis, a significant glucocorticoid-sparing effect, defined as achieving CRR with daily prednisone use of 7.5 mg or less, was observed at week 76 with obinutuzumab treatment. The median cumulative dose of prednisone from weeks 64 through 76 was 100 mg lower in the obinutuzumab group than the placebo group. These data confirm that there was no imbalance in glucocorticoid use in the obinutuzumab arm that could have influenced kidney outcomes. Furthermore, a clinically meaningful reduction in glucocorticoid use is now recognized as an important short-term and long-term treatment goal in patients with lupus nephritis because risks and toxicities associated with glucocorticoid use are numerous.33

Several limitations apply to this work. Foremost, these analyses were post hoc and not prespecified. The study was not powered to address these questions, and for several analyses, the total number of events was small, precluding robust statistical confirmation. However, these analyses will be repeated in the phase III REGENCY trial to verify the results.

The totality of these data not only confirm the benefit of B cell depletion to short-term treatment outcomes of patients with lupus nephritis, but the results of our analyses have implications for preservation of long-term kidney health. Given that these conclusions are from post hoc analyses, it will be necessary to validate the associations of B cell depletion or inhibition with functional kidney protection prospectively. If validated, understanding how B cells influence kidney survival may provide insights into approaches other than immunosuppression to mitigate this influence.

In conclusion, post hoc analyses of the NOBILITY data demonstrated a favorable impact of obinutuzumab compared with placebo on several different measures that forecast long-term kidney health, including the composite unfavorable kidney outcome of treatment failure, doubling of serum creatinine, or death, transient and sustained eGFR declines of both 30% and 40%, lupus nephritis flares, and reductions in eGFR decline. In addition, a glucocorticoid-sparing effect was suggested because more patients who achieved CRR at week 76 with obinutuzumab than placebo were able to maintain a daily prednisone dose of 7.5 mg or less from weeks 64 through 76, with a similar but not statistically significant trend through week 104. Obinutuzumab, in combination with SOC therapy, is the first CD20-targeted biologic to show efficacy and safety in lupus nephritis compared with SOC therapy alone in a randomized study. For this reason, obinutuzumab is being further evaluated in patients with active proliferative lupus nephritis in the global registrational phase III REGENCY trial (NCT04221477).

ACKNOWLEDGMENTS

We would like to extend our gratitude to the patients, their families, study investigators, and site staff for their participation in the NOBILITY trial. Editorial assistance was provided by Health Interactions, Inc., and funded by F. Hoffmann-La Roche Ltd.

    AUTHOR CONTRIBUTIONS

    All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Rovin had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Study conception and design

    Rovin, Furie, Ross Terres, Turchetta, Pendergraft III.

    Acquisition of data

    Rovin, Furie, Malvar.

    Analysis and interpretation of data

    Rovin, Furie, Ross Terres, Giang, Schindler, Turchetta, Garg, Pendergraft III, Malvar.

    ROLE OF THE STUDY SPONSOR

    F. Hoffmann-La Roche Ltd participated in the study design, collection, analysis, and interpretation of the data, and the writing of the manuscript, and was involved in the decision and approval to submit the manuscript for publication. Editorial assistance was provided by Health Interactions, Inc., and funded by F. Hoffmann-La Roche Ltd.