REGENCY: A B-Cell Super Destroyer in Lupus

Introduction

Lupus nephritis (LN) is a formidable manifestation of systemic lupus erythematosus (SLE), affecting up to 60% of patients and standing as a leading cause of kidney failure in young individuals worldwide. At its core, LN is a battle within the glomerulus, where immune complex deposition, complement activation, and an unrelenting inflammatory cascade orchestrated by autoreactive B cells drive irreversible kidney damage. These pathogenic B cells fuel the disease by producing autoantibodies (e.g., anti-dsDNA, anti-C1q), amplifying cytokine signaling, and sustaining T-cell activation—a triad that perpetuates immune dysregulation and renal injury (Almaani et al, CJASN 2017).

The diagnosis of LN hinges on kidney biopsy, which remains the gold standard for classification, prognosis, and therapeutic decision-making (Giannico, et al, CJASN 2013). A biopsy is typically performed in patients presenting with proteinuria, abnormal urinary sediment, or rising creatinine, and findings are classified using the International Society of Nephrology and Renal Pathology Society (ISN/RPS) system. Beyond histologic class, activity and chronicity indices provide crucial insights: active lesions that suggest ongoing inflammation and may respond to aggressive immunosuppression, while chronic lesions reflect irreversible damage less amenable to therapy (see figure below).

Understanding the balance between active and chronic disease is key to predicting response to B-cell–targeted therapies, particularly anti-CD20 monoclonal antibodies (mAbs), which primarily modulate immune-mediated inflammation but not reverse fibrosis-driven progression (Chen, et al, J Autoimmun 2022, Parikh et al, AJKD 2020). mAbs have become a key strategy in LN management, but their mechanisms differ (Zavaleta-Monestel et al, Cureus 2024, Ponticelli et al, Pharmaceuticals (Basel) 2010):

  • Type I anti-CD20 mAbs (e.g., rituximab) deplete B cells via complement-dependent cytotoxicity (CDC) and antibody-dependent cellular cytotoxicity (ADCC). However, CDC-based depletion is incomplete, leaving some B cells resistant to therapy, potentially explaining the limited efficacy of rituximab in LN.

  • Type II anti-CD20 mAbs (e.g., obinutuzumab) induce stronger direct B-cell cytotoxicity and enhanced phagocytosis, leading to more profound and sustained depletion, including tissue-resident B-cell niches. This deeper depletion has been hypothesized to improve response rates in diseases like LN but may increase infection risk.

Attempts to achieve direct B-cell depletion in LN began with rituximab, which was evaluated in the LUNAR trial (Rovin et al, 2012, Arthritis Rheum). While rituximab effectively depleted circulating B cells, it failed to improve renal chronic response significantly. A post-hoc analysis of LUNAR trial (Gomez Mendez LM et al, CJASN, 2018) showed that variability in B-cell depletion dictated responses and those with a complete, lasting peripheral B-cell depletion had more complete remissions. Additionally, the EXPLORER trial also failed to demonstrate an added benefit of rituximab in SLE, reinforcing concerns about incomplete B-cell depletion with type I anti-CD20 mAbs (Merrill, et al, Arthritis Rheum 2010).

In contrast, belimumab—a BAFF/BLyS inhibitor—became the first biologic approved for LN following the BLISS-LN trial (Furie et al, NEJM 2020 | NephJC Summary). Adding belimumab to standard therapy improved renal responses and reduced flares, reinforcing the role of B-cell survival inhibition in LN management. Based on these findings, belimumab was incorporated into the updated 2024 2024 KDIGO management LN guidelines as a first-line treatment option for class III/IV LN.

This set the stage for exploring obinutuzumab, a next-generation type II anti-CD20 monoclonal antibody. The phase 2 NOBILITY trial demonstrated that obinutuzumab plus standard therapy improved renal responses, proteinuria reduction, and eGFR preservation (Furie et al, Ann Rheum Dis 2022) and led to a breakthrough therapy designation from the FDA for LN in September 2019. This set up the phase 3 REGENCY trial, designed to evaluate the efficacy and safety of obinutuzumab in assessing the renal response in patients with active LN.

The Study

Methods

Trial design: multinational, including countries from 15 countries, 4 continents (North and South America, Europe, and Africa), phase 3, randomized, double-blind, placebo-controlled, intention-to-treat trial. Stratification was done by region and race.

Inclusion criteria: The study included participants with biopsy-confirmed active class III/IV lupus nephritis with/ without class V (ISN/RPS classification) during screening or within 6 months before. Included participants had a urinary protein-to-creatinine ratio (UPCR) ≥ 1g (from 24h proteinuria) and positive ANA (ANA titer of ≥1:80 on HEp-2 cells or ≥1 equivalent positive ANA test).

Exclusion criteria: The study excluded patients with eGFR <30 ml/min/per 1.73 m2 or ESRD, active infection, recent (≤9 months) anti-CD20 therapy, and recent immunosuppressive therapies (≤2 months).

Interventions: Obinutuzumab added to standard MMF and prednisone in the intervention arm, and compared to MMF + steroids alone (i.e. with placebo) in the control arm. Obinutuzumab exploratory doses were used to optimize long-term B-cell depletion strategies. Prednisone taper targets of 5 mg/day were aimed by week 24. Patients with adequate treatment response at week 76 continued to receive blinded infusions every 6 months starting at week 80, until the study was unblinded. Patients with inadequate treatment response at week 76, or with loss of response during blinded treatment after week 76, could enter open-label treatment.

Figure S1. Study Schema, from Furie et al, NEJM, 2025

Outcomes
Primary endpoint was complete renal response at week 76, defined by proteinuria <0.5 mg/mg, eGFR ≥ 85% of baseline (calculated by 2009 CKD-EPI), and no intercurrent events including rescue therapy, treatment failure, death, or early trial withdrawal. Treatment failure included the occurrence of end-stage kidney disease, the long-term use of dialysis, kidney transplantation or rescue therapy (except for glucocorticoid-only rescue).

Secondary endpoints include patient-reported fatigue, eGFR change and partial renal response. The end points in Table S2 are listed as hypothesis testing order.

Adapted from Table S2, from Furie et al, NEJM, 2025

Safety: Adverse events were monitored with independent oversight. Descriptive safety analyses included patients who received ≥1 dose of obinutuzumab or placebo and grouped as treated.

Statistics: Power calculation was based on a 20% absolute difference in CCR. A fixed-sequence testing with fallback procedure was applied to control type I error, ensuring that subsequent hypotheses were only tested if the preceding hypothesis was statistically significant at ɑ= 0.05. If the primary or key secondary endpoint was not significant, further hypothesis testing was stopped. Missing data were imputed using multiple imputations with predicted mean matching. Prespecified subgroup analyses included patients who received at least one obinutuzumab dose.

Funding: The trial sponsor (F. Hoffmann-La Roche) designed the trial and participated in the collection, analysis and interpretation of data, as well as in preparation and submission of the manuscript. Five of the manuscript authors were employees of the sponsor (including the senior data scientist), and 5 authors have sponsor stock (including the manuscript’s first draft author). Genentech Inc is an independent subsidiary of Roche. Six of the authors are Genentech Inc employees, one of them having stocks, while two of them have stocks in F. Hoffmann-La Roche; the main author consulted for Genentech Inc. The first draft of the manuscript was written by the penultimate author (having stock in F. Hoffmann-La Roche and being an employee of Genentech, Inc.), with subsequent drafts receiving editorial assistance funded by the sponsor.

Results

From 513 screened patients, 271 were randomized to obinutuzumab (135) or placebo (136) groups.

Figure S2. Consort diagram, from Furie et al, NEJM, 2025

Baseline characteristics were similar across groups. The mean age was ～33 years, with 84% of patients being women, ～48% of patients were white, ～15% black, ~ 19% being AMerican Indian/Alaskan, and～6% Asian. The proportion of Hispanic patients (~ 57%) was higher than expected (15%) before beginning the study. Median lupus nephritis duration was approximately 36 months for both groups. ~40% patients were newly diagnosed with lupus nephritis. Baseline eGFR was similar (obinutuzumab 102.8 versus placebo 101.9 ml/min/1.73m2), and a slightly lower protein-to-creatinine ratio in the obinutuzumab group (2.13 versus 2.76 g/24-hours).

Table 1. Baseline characteristics, from Furie et al, NEJM, 2025

The study population included 40% of patients with LN class III and 60% with LN class IV. Approximately 30% had concomitant class V LN. 

Adapted from table S4. Baseline characteristics, from Furie et al, NEJM, 2025

Primary endpoints

At week 76, 46.4% of the obinutuzumab group achieved a complete renal response, compared to 33.1% in the placebo group (adjusted difference 13.4, 2.0 to 24.8).

Table 2. Primary and key secondary outcomes, from Furie et al, NEJM, 2025 (*The end points in Table 2 are listed as hypothesis testing order)

Key secondary endpoints

 A secondary analysis focusing on patients receiving prednisone ≤7.5 mg/day showed 42.7% versus 30.9% complete response in obinutuzumab versus placebo respectively. With obinutuzumab there was a greater reduction in protein-to-creatinine ratio (adjusted difference 13.7%). Proteinuric response (UPCR <0.8 g per gram) without intercurrent events was significantly higher in the obinutuzumab arm (55.5% vs 41.9%). Interestingly, in the obinutuzumab group eGFR increased  2.31 (±2.71) ml/min/1.73 m2, while in the placebo group eGFR decreased by -1.54 (adjusted difference 3.84 ml per minute per 1.73 m2 95% CI, −1.83 to 9.51). Since the eGFR endpoint did not reach statistical significance, the hierarchical testing sequence stopped, thus preventing formal statistical testing of the subsequent endpoint (renal-related event incidence). The authors report a lower incidence of renal-related events (18.9% vs 35.6%), though no difference in fatigue (FACIT-F score change).

Intercurrent events

The obinutuzumab group had a lower incidence of intercurrent events (11.1%) versus placebo group (25%). Notably, treatment failure was more frequent in the placebo group (17.6% vs 3.7%), reflecting a higher need for rescue therapy (17.6% vs 5.6%) particularly in patients receiving non-glucocorticoid-based agents. The rate of clinically significant worsening of proteinuria or eGFR was also higher in the placebo group (16.2 vs. 3.7%). The incidence of deaths was numerically higher in the obinutuzumab group (3 deaths vs 1 death). Early study withdrawals were slightly more common in the placebo group.

Table S5. Intercurrent events at week 76 in the intention to treat population, from Furie et al, NEJM, 2025

Subgroup analyses: complete renal response results were consistent across subgroups.

Figure 1. Subgroup analysis of complete renal response in intention-to-treat-population, from Furie et al, NEJM, 2025

Pharmacodynamics, serologic analysis, and immunogenicity: Obinutuzumab group had a greater reduction of C3, C4 and dsDNA antibodies, and more patients had complete CD19-positive B-cell depletion.

Figure 2. Serologic and pharmacodynamic analyses over time, from Furie et al, NEJM, 2025

Safety: The incidence of any adverse events was similar between groups, with a rate close to 90% in both arms. Serious adverse events were higher in the obinutuzumab group (33.4 vs 18.2%). The most common adverse events were infections (61.8% vs 53.8%), with serious infections occurring more frequently with obinutuzumab (11% vs 7.6%). Drug-related neutropenia was more common with obinutuzumab (12.5% vs 3.8%) while no cases of hepatitis B reactivation, progressive multifocal leukoencephalopathy, or gastrointestinal perforations were observed. (table 3, table S8)

Table 3. Adverse events, from Furie et al, NEJM, 2025

There were 4 deaths in both arms: 3 in the obinutuzumab group, 1 in the placebo group, the authors not attributing any to obinutuzumab.

Adapted from table S9. Fatal adverse events through week 76, from Furie et al, NEJM, 2025

Discussion

This trial confirms the observations of efficacy of type 2 anti-CD20 antibodies as a potential add-on to current drug regimens for active lupus nephritis in adult patients. The REGENCY trial demonstrated higher renal response rates with obinutuzumab added to standard therapy versus placebo, reinforcing findings from the phase 2 NOBILITY trial (Furie et al, Ann Rheum Dis. 2022). These results position type 2 anti-CD20 therapy as a potential next-generation treatment for LN.

Peripheral B-cell depletion (defined as CD19+ cell counts <10/microL) was sustained in 95% of patients till week 76 (19 months), a consistent observation also observed in the NOBILITY trial (Furie et al, Ann Rheum Dis. 2022), and much better than having ~80% of patients depleting at week 52 with Rituximab in the LUNAR trial (Rovin et al, 2012, Arthritis Rheum). This reflects in the anti-dsDNA antibody reduction and complement normalization in this trial. Long-lived plasma cells are found in the circulation and kidney interstitium and play a role in renal flares. The consistency and depth of B-cell depletion with obinutuzumab is a boon and is, expectedly, being explored for other immune glomerulonephritis like membranous nephropathy, AAV, and FSGS as well (a fantastic overview here). While presently evaluated only as an add-on therapy for LN (with standard lower doses of steroids and a higher, induction-regime-worthy dose of MMF), can Obinutuzumab serve to significantly reduce background immunosuppression? This is a critical question to be answered next. We would be happier if this sniper attack does its job well and doesn’t leave us with the gnawing obligation of carpet bombing LN, like we do presently. 

That said, notable strengths of this study are:

  • REGENCY is the first phase 3 trial of a type 2 anti-CD20 mAb in LN.

  • Multiethnic populations representing high-risk populations of LN progression

  • Longer observation period giving a better understanding of safety

The study has its limitations, though:

  • No post-treatment biopsies, leaving uncertainty about its impact on inflammation or fibrosis.

  • Increased mortality in patients on obinutuzumab was attributed to COVID-19, but the potential harm of greater B-cell depletion remains unclear.

  • Higher rates of SAEs and neutropenia raised infection risk concerns.

  • Comparatively lesser representation of Asians (5.9%), vs African-American (14.8%) or Hispanic (57.6%) raising the doubt of whether these results are driven by response in groups inherently more responsive to the background therapy (MMF)

  • Choice of comparator: The KDIGO 2024 guidelines allow MMF - but also CNIs, Belimumab, and in particular cyclophosphamide for induction. In severe cases (or when cost is a concern) many of us would reach for cyclophosphamide, and it is unfortunate that there is no such head:head comparison. Obinutuzumab seems like a potent and powerful B-cell depleter and a worthy, albeit expensive, replacement for cyclophosphamide but for the lack of this direct comparison.

No new adverse events were identified in this study, which itself is good news. Vaccinations, protection from infectious exposures, and timely management of cytopenias can improve the experience with obinutuzumab. How long Obinutuzumab can be safely continued in patients with LN remains to be seen. Currently priced variably between 2000-10000 USD per dose (1gm), we need more studies proving cost-effectiveness and will be watching for potent biosimilars now that the patents are expiring in many regions of the world. We also expect future trials in diverse groups (different races/ethnicities and even pediatric age groups) and at different stages of the disease (as induction or maintenance therapy, or resistant disease) to better inform us about its efficacy and adverse effects in such situations.

Conclusion

The REGENCY trial positions obinutuzumab as a promising addition to B-cell–targeted therapy in LN, demonstrating higher renal response rates. However, safety concerns, particularly infection risk and unclear long-term renal outcomes, cost, and comparative data against cheaper potent agents, may limit its widespread use.

Summary by

Cristina Popa,
Iași, Romania

Milagros Flores
Guadalajara, México

Sayali Thakare
Mumbai, India

Reviewed by 

Brian Rifkin, Swapnil Hiremath

Header Image created by AI, based on prompts by Brian Rifkin