Article Text
PDF
Abstract
Objectives To investigate the effectiveness of belimumab on active lupus nephritis (LN) and explore the predictors, including serological biomarkers, of renal response to belimumab in a real-world setting.
Methods This multicentre, real-world observational study enrolled patients with active LN receiving intravenous belimumab as an add-on therapy with 24-hour urine protein≥1 g and estimated glomerular filtration rate≥30 mL/min/1.73 m2 at baseline. Complete renal response (CRR), partial renal response (PRR), no renal response (NRR) and primary efficacy renal response (PERR) were evaluated. Multivariable logistic regression was used to identify risk factors for NRR to belimumab at 6 months.
Results Among the 122 patients enrolled, the proportions of patients achieving CRR, PRR, NRR and PERR were 35.9%, 17.1%, 47.0% and 44.4% at 6 months (n=117) and 55.6%, 19.4%, 26.4% and 58.3% at 12 months (n=72), respectively. Proteinuria, daily prednisone dosage and Systemic Lupus Erythematosus Disease Activity Index 2000 scores significantly decreased at 6 and 12 months (p<0.0001). NRR at 6 months (NRR6) was the strongest negative predictor of CRR at 12 months. Baseline anti-dsDNA positivity inversely predicted NRR6 (OR=0.32,95% CI=0.10 to 0.98, p=0.049), while anti-SSA/Ro60 positively predicted NRR6 (OR=3.16, 95% CI=1.14 to 8.74, p=0.027). The combination of anti-SSA/Ro60 and anti-dsDNA serotype quantitatively predicted belimumab renal response.
Conclusion The effectiveness of belimumab was reproducible in Chinese patients with active LN. The simple yet interesting serotype predictive model needs further validation and its possible underlying mechanistic relevance deserves further exploration.
- Lupus Erythematosus, Systemic
- Lupus Nephritis
- Autoantibodies
Data availability statement
Data are available on reasonable request.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Belimumab as an add-on therapy may improve meaningful renal response rate in patients with active lupus nephritis (LN).
WHAT THIS STUDY ADDS
The effectiveness of add-on belimumab was replicated in Chinese patients with active LN in a real-world setting. The presence of anti-dsDNA protects against no renal response (NRR), while anti-SSA positivity favours NRR at 6 months.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The combination of anti-dsDNA and anti-SSA antibody serotype may help predict the renal response of add-on belimumab, which needs further validation. Its possible underlying mechanistic relevance deserves further exploration.
Introduction
Systemic lupus erythematosus (SLE) is an autoimmune disease that can damage multiple organs, with lupus nephritis (LN) being the most frequent and severe form of organ involvement.1–3 Despite the development of new treatments, 5%–30% of patients with LN progress to end-stage renal disease within 10 years.4 5
Belimumab is a fully human IgG-1λ monoclonal antibody that impedes the maturation and development of B cells by binding to the B lymphocyte stimulator (BLyS).6 Its efficacy and safety in treating LN have been well demonstrated by the far-reaching Belimumab International Study in Lupus Nephritis (BLISS-LN), a phase III, 2-year, randomised controlled trial and its post hoc analyses.7 8 Moreover, real-world studies such as Belimumab in Real Life Setting Study (BeRLiSS)-LN9 also supported the durable renal response with add-on belimumab in patients with LN. However, the meaningful renal response to belimumab was slightly shorter than 50%, leaving more than half of patients with LN with an inadequate renal response.7 Therefore, appropriate tools to identify patients who may or may not benefit from belimumab therapy are of vital importance.
Following our first real-world multicentred study of belimumab in Chinese patients with SLE, which confirmed its effectiveness and safety,10we take a step forward in this study aiming to describe the renal effectiveness of belimumab in patients with active LN, with a special focus on exploring possible renal response predictors.
Patients and methods
This is a multicentre, real-world observational study initiated by investigators without any industry sponsorship at five referral centres in China from October 2019 through September 2023. Along with the four previously described centres,10 the fifth centre, that is, the first affiliated hospital of Wenzhou Medical University, joined this study.
Inclusion and exclusion criteria
Enrolled patients were diagnosed with SLE according to the 1997 revised American College of Rheumatology classification criteria11 or 2012 Systemic Lupus International Collaborating Clinics criteria for SLE.12 In addition to standard-of-care therapy, these patients received intravenous belimumab (10 mg/kg) treatment at weeks 0, 2, 4 and every month thereafter. On week 0, the 24-hour urine protein level should be more than 1 g and attributed to LN. In patients for whom 24-hour urine protein was not available, substitutions by the urine protein–creatinine ratio (UPCR) or urine albumin–creatinine ratio (UACR) of 1 mg/mg or more were permitted.
The exclusion criteria for this study encompassed an estimated glomerular filtration rate (eGFR) less than 30 mL/min/1.73 m2 or receiving dialysis at baseline, exposure to B-cell-targeting therapies within 6 months and insufficient baseline or follow-up data to conclude a renal outcome assessment at 6 months or at 12 months (±1 month).
Data collection
The demographic characteristics, clinical manifestations, laboratory tests and concurrent medications of the participants were collected at baseline (first dose of belimumab) and every month thereafter if available. The antinuclear antibodies profile tested closest to the baseline (±3 months) were documented, including antinuclear antibody (indirect immunofluorescent assay), anti-dsDNA antibody (ELISA, radioimmunoassay or crithidia luciliae immunofluorescence test), antinucleosome antibody (ANuA), antihistone antibody, anti-Sm, anti-RNP, antiribosomal P protein, anti-SSA/Ro60, anti-SSB and anti-Ro52 (Euroimmun line-blotting assay) antibodies. Disease activity was evaluated using the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2k) Score13 and clinical SLEDAI-2k Score (excluding complement and anti-dsDNA antibodies). Renal pathologies were classified based on the International Society of Nephrology and Renal Pathology Society criteria.
Definitions of renal responses
Complete renal response (CRR): 24-hour urine protein<0.5 g (or UPCR/UACR<0.5 mg/mg) and eGFR≥90 mL/min/1.73 m2 or stable renal function (no decline greater than 10% from the preflare value) without rescue therapy.
Partial renal response (PRR): a reduction of at least 50% in proteinuria with a maximum 24-hour urine protein≤3 g (or UPCR/UACR≤3 mg/mg) and eGFR≥90 mL/min/1.73 m2 or stable renal function (no decline greater than 10% from the preflare value) without rescue therapy.
No renal response (NRR): patients with LN who did not fulfil either CRR or PRR.
Overall renal response: patients with LN who achieved CRR or PRR.
Primary efficacy renal response (PERR): 24-hour urine protein<0.7 g (or UPCR/UACR<0.7 mg/mg) and eGFR≥60 mL/min/1.73 m2 or stable renal function (no decline greater than 20% from the preflare value) without rescue therapy.
Statistical analysis
Data were summarised using frequency counts and percentages for categorical variables and mean±SD or median (IQR) for continuous variables. Independent sample Student’s t-test, two-sample Mann-Whitney U test, Wilcoxon matched-pairs signed rank test and χ2 test were used where appropriate. With the exception of renal pathology assessment, baseline characteristics with less than 20% missing values were subjected to statistical analyses. Binary logistic regression was used to identify the independent predictors of renal response.
Statistical analyses were performed using SPSS Statistics V.24 or GraphPad Prism V.9.0 (San Diego, California, USA) software. Statistical significance was set at p<0.05.
Results
Baseline characteristics
A total of 122 patients fulfilled the inclusion/exclusion criteria (figure 1A). 33 patients (27.0%) were diagnosed as newly onset LN within 3 months before initiation of belimumab. 36 patients (29.5%) added belimumab within 3 months after LN relapse. Moreover, 53 of them (43.4%) had inadequate response (residual proteinuria≥1 g) to previous immunosuppressive agents (IS) for at least 3 months. The baseline attributes of the participants are summarised in table 1.
Flow chart, renal outcomes and changes of clinical variables over time. Flow chart of inclusion and exclusion (A). Renal response rates at 6 and 12 months (±1 month) (B). Violin plots (C, D) and box plot (E) showing changes of 24-hour urine protein, daily prednisone dosage and SLEDAI-2K at 6 and 12 months. Wilcoxon matched pairs signed rank test, ****p<0.0001. CRR, complete renal response; eGFR, estimated glomerular filtration rate; h, hour; m, month; NRR, no renal response; PERR, primary efficacy renal response; PRR, partial renal response; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000.
Baseline characteristics of patients in the real-world study
The mean (±SD) age was 34.2±11.1 years, the median follow-up time was 11.9 months (IQR, 7.4 to 16.2) and the median duration of LN was 1.9 years (IQR, 0.1 to 7.6). The average level (±SD) of 24-hour urine protein was 3.2±2.3 g, while 41.8% of the patients enrolled had nephrotic range proteinuria (≥3 g). The baseline eGFR was 101.2±36.1 mL/min/1.73 m2. Renal pathologies were available in 81 patients, and 40 biopsies were performed within 6 months before the initiation of belimumab. In this real-world study, baseline 24-hour urine protein values were replaced by UPCR in four patients and by UACR in eight patients.
As for concomitant medications, the daily prednisone dosage (or equivalent) (mean±SD) was 32.6±22.0 mg and hydroxychloroquine (HCQ) was prescribed in 77.9% of cases; other IS included mycophenolate mofetil (MMF) in 50.8% of cases, calcineurin inhibitors (CNI) in 32.8%, cyclophosphamide (CYC) in 13.9% and others (20.5%) such as azathioprine, leflunomide and iguratimod. Of note, 26.2% of patients were prescribed more than one category of classic IS, while 13.1% of patients received no classic IS other than prednisone and HCQ.
Renal response and SLE outcome measurements
At 6 months, 117 patients had renal outcomes assessment available. The proportions of patients who achieved CRR, PRR, PERR and NRR were 35.9%,17.1%, 44.4% and 47.0%, respectively. At 12 months, the CRR, PRR, PERR and NRR rates were 55.6%,19.4%, 58.3% and 26.4%, respectively (figure 1B). Notably, only 72 patients (59.0%) had renal outcomes assessment available at 12 months. Nevertheless, the baseline parameters, including the NRR rate at 6 months, were comparable between patients with and without renal outcome measurements at 12 months (online supplemental table S1). Of note, the possible differences of renal response rates between pathology classes of LN, that is, proliferative versus membranous versus mixed classes, were not captured in 81 patients with renal histology data available (online supplemental table S2).
Supplemental material
Finally, the Wilcoxon matched-pairs signed rank test showed that 24-hour urine protein, daily prednisone dosage (or equivalent) and SLEDAI-2K scores were significantly decreased at both 6 and 12 months compared with the baseline (p<0.0001) (figure 1C–E).
NRR at 6 months (NRR6) is the strongest negative predictor of CRR at 12 months (CRR12)
The characteristics of patients with LN reached CRR12 (n=40) or failed to reach CRR12 (non-CRR12) (n=32) are listed in online supplemental table S3. Patients who achieved CRR12 had lower baseline levels of urine protein (2.7±1.9 g vs 4.1±2.6 g, p=0.020) and a lower proportion of nephrotic range proteinuria (30.0% vs 53.1%, p=0.047). Importantly, patients achieving CRR12 were less likely to have NRR6 (16.2% vs 70.0%, p=0.000). Furthermore, NRR6 (OR=0.05, 95% CI=0.01 to 0.22, p=0.000) and baseline proteinuria (OR=0.63, 95% CI=0.45 to 0.89, p=0.009) remained independent negative predictors of CRR12 after adjusting confounding factors (figure 2A).
Risk factors for renal response to add-on belimumab. Forrest map in logarithmic scale showing predictors for CRR12 (A) and NRR6 (B) by multivariate binary logistic regression analysis. NRR6 rates in four serotypes based on the anti-dsDNA and anti-SSA/Ro60 antibody status in the entire cohort and in subgroups with eGFR≥60 and ≥90 mL/min/1.73 m2 (C). CRR12, complete renal response at 12 months; eGFR, estimated glomerular filtration rate (mL/min/1.73 m2); LN, lupus nephritis; NRR6, no renal response at 6 months; ORR6, overall renal response at 6 months; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000.
The presence of anti-SSA/Ro60 antibody positively predicted NRR6 while anti-dsDNA antibody inversely predicted NRR6
Considering the high data attrition rate at 12 months, NRR6 was determined to be the surrogate endpoint for CRR12. Subsequently, four sets of putative predictors were selected for regression modelling based on univariable analysis (table 1) and the investigators’ judgement, that is, general (age, clinical SLEDAI-2K), renal (LN duration, proteinuria, eGFR), serology (anti-dsDNA, anti-SSA/Ro60, low complement) and concomitant treatment (daily prednisone dosage, exposure to HCQ, CNI, MMF or CYC).
Of the 13 baseline parameters selected, clinical SLEDAI-2K, anti-dsDNA antibody, low complement and daily prednisone dosage were significant by univariate analysis (p<0.05). Although with significance by univariable analysis, ANuA was excluded due to the consideration of its overlapping with anti-dsDNA antibody. Age, renal function (eGFR) and exposure to CNI which had marginal p values (<0.1) were also included. With special interest based on subgroup analyses, anti-SSA/Ro60 antibody was predictive for NRR6 in patients with eGFR≥60 mL/min/1.73 m2 (OR=2.49, 95% CI=1.064 to 5.832, p=0.035) and with eGFR≥90 mL/min/1.73 m2 (OR=3.47, 95% CI=1.212 to 9.955, p=0.020). Therefore, anti-SSA/Ro60 was added to the serology set along with anti-dsDNA and low complement. For renal set, proteinuria has been linked to belimumab renal response according to Belimumab International Study in Lupus Nephritis (BLISS-LN) post hoc analyses,14 thus, was enclosed to further modelling. Because median (IQR) LN duration was relatively shorter among patients displaying positive anti-SSA/Ro60 (1.0 (0.0–4.9) vs 3.5 (0.1–8.0), p=0.088), and new-onset LN was reported to predict good renal response,15 LN duration was also included. To be thorough, since prednisone and CNI had already been included as putative predictors, HCQ, MMF or CYC exposures were not left behind and all subjected to multivariable analyses.
In the end, only anti-SSA/Ro60 antibody (OR=3.16, 95% CI=1.14 to 8.74, p=0.027), anti-dsDNA antibody (OR=0.32, 95% CI=0.10 to 0.98, p=0.049) and impaired renal function (eGFR 30–60 mL/min/1.73 m2) (OR=4.62, 95% CI=1.02 to 20.9, p=0.047) remained significant (figure 2B).
Predictive value of combining anti-dsDNA and anti-SSA/Ro60 antibody for NRR6
Patients with LN were divided into four serotypes based on their anti-dsDNA and anti-SSA/Ro60 antibody status: double negative (dsDNA−/SSA−), isolated anti-dsDNA positive (dsDNA+/SSA−), isolated anti-SSA/Ro60 positive (dsDNA−/SSA+) and double positive (dsDNA+/SSA+). Interestingly, it was observed that the dsDNA−/SSA+ group (n=20) had a significantly higher NRR6 rate compared with the dsDNA+/SSA− group (n=31) in the entire cohort (75% vs 29%, p=0.001), and this difference was maintained in subgroups with eGFR≥60 mL/min/1.73 m2 (77.7% vs 22.2%, p=0.000) and with eGFR≥90 mL/min/1.73 m2 (69.2% vs 19.0%, p=0.003) (figure 2C, online supplemental table S4).
Adverse events
19 patients (15.6%) reported adverse events, leading to belimumab discontinuation in 6 patients (table 2). Infectious diseases were most common (n=12), including recurrent upper respiratory tract infection with leucopenia, COVID-19 infection, urinary tract infection, gastroenteritis, cytomegaloviremia in one case each, bacteria pneumonia in two cases and herpes zoster infection in five cases. In addition, five patients experienced hypogammaglobulinemia and needed intravenous immunoglobulin supplement. Moreover, another one patient had sustained leucopenia, one patient reported infusion-related fever, one patient had hip replacement and one patient reported optic neuritis. No alarming safety signals were revealed in this study.
Adverse events (AE)
Discussion
This multicentre, real-world study of 122 patients provided further evidence of the effectiveness of belimumab as an add-on therapy for the Chinese LN population whose baseline characteristics were similar to those in the BLISS-LN trial 7. At 6 and 12 months, there was a significant reduction in proteinuria, SLEDAI-2K Score and daily prednisone dosage. At 6 months (week 24), the percentage of patients who achieved PERR (44.4%) was comparable to that in the BLISS-LN trial (approximately 40% in the belimumab group). However, the CRR rate was slightly higher in this study (35.9%) than that in the BLISS-LN trial (approximately 23% in the belimumab group), which may be partially explained by the more stringent glucocorticoid taper regimen in the BLISS-LN trial. In our study, 48 patients (39.3%) still had an oral prednisone (or equivalent) dosage of more than 10 mg/day at 6 months. Recently, three real-world studies have addressed the efficacy and safety of belimumab as an add-on medication in Chinese patients with LN. Two studies included patients with LN (n=61 and n=112, respectively) with more stable and milder disease activity. The median level of proteinuria was 0.5–1 g/day and up to half of the participants had achieved CRR or PRR at baseline.16 17 The other enclosed patients with refractory LN (n=45) who had failed at least 6 months induction therapy with glucocorticoids and conventional IS including CYC, MMF or multitarget treatment.18 Thus, our study by far largely replicated the BLISS-LN trial design and the effectiveness of belimumab in Chinese LN is confirmatory.
More importantly, we pinpointed a simple yet robust finding that the combination of anti-dsDNA and anti-SSA/Ro60 serotype may help to foretell the patient’s renal response to belimumab. Of note, NRR6 was used as the endpoint in our analyses. This was not only a data-driven choice but also was in line with standard protocol in practice. As an example, according to the 2019 update of European League Against Rheumatism and European Renal Association-European Dialysis and Transplant Association recommendations, the management of LN should aim for at least a PRR by 6 months so as to achieve a CRR by 12 months.19 Otherwise, the induction regimen will be deemed as suboptimal.
Studies suggested that BLyS-targeted treatments selectively reduced naïve B cells and anti-dsDNA producing plasmablasts resulted from the extrafollicular pathway, but spared germinal centre pathway-generated memory B cells and long-lived plasma cells.20–23 By using NRR6, it was hardly a surprise that anti-dsDNA is predictive in terms of a better belimumab response. In the pooled post hoc analysis of BLISS-52/76 trial,14 patients with LN with serologic activity at baseline, that is, positive anti-dsDNA and/or hypocomplementemia, had greater renal benefit with belimumab. In addition, the evidence of belimumab suppressing the titers of anti-dsDNA antibody is substantial in patients with SLE.20 24 25 On the other hand, our data suggested that anti-SSA/Ro60 antibody was a strong risk factor associated with NRR6, which may turn out to be ‘the elephant in the room’ for predicting belimumab inadequate response in LN. It has been demonstrated that anti-SSA/Ro60 was resistant to even profound B-cell depletion by anti-CD19 CAR T cell therapy, which suggested that long-lived plasma cells might actually be responsible for this serotype.26 27 Therefore, it is a legitimate postulation that patients with anti-SSA/Ro60-dominant LN might be resistant to anti-B-cell therapy such as BLISS-LN protocol. Unfortunately, we did not have immunophenotyping data in the current real-world study to directly address this intriguing question.
Indeed, the combination of anti-dsDNA and anti-SSA/Ro60 antibody serotypes had yielded a good predicting performance in NRR6. It has been well conceived that a better-preserved renal function at baseline indicated a better response by definition. Intriguingly, even adjusted for different eGFR, the presence of anti-SSA/Ro60 contributed to a stable NRR6 probability, which further suggested its independent predictive value.
This study has several limitations. First, the sample size is small and an external validation cohort is lacking to confirm our finding. We have inquired (correspondence) to conduct a post hoc analysis for BLISS-LN trial, but unfortunately, data of anti-ENA (anti-SSA/Ro60) was not collected in the first place as per trial protocol. Second, heavily impacted by COVID-19 pandemic/endemic and China’s lockdown policy during the follow-up period of the study, the median drug retention time was only 11.9 months in our study instead of 2 years, which would be ideal to emulate the treatment arm of BLISS-LN trial and to generate more eGFR-oriented analysis. Consequently, the data attrition resulted in a surrogate endpoint of NRR6 instead of a typical CRR12. Third, renal pathology was missing in 33.6% of patients, and only 40 patients had renal biopsy performed within 6 months before the initiation of belimumab. This makes the analysis on renal pathology effects14 impossible. Last but not least, due to the real-world layout without support from a central lab, the anti-dsDNA assays were diverse and the results were merely dichotomized as either positive or negative.
In conclusion, this study demonstrated the effectiveness of belimumab among patients with active LN in real-world clinical settings in China. Moreover, we identified that LN patients with the dsDNA−/SSA+ serotype were less likely to benefit from the BLISS-LN regimen.
Data availability statement
Data are available on reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants. This study was approved by the central ethics committee of Renji Hospital, Shanghai Jiao Tong University School of Medicine (KY2021-059-B). Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors appreciated the contributions of all the participants in this study.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
LZ, WW and LW contributed equally.
Contributors LZ, JZ, LS and SY contributed to design the study and conception. LZ and SY drafted the manuscript. LZ, WW, LW, TW, JT, XW, FS, HD, NS and HW contributed to collection, analysis and interpretation of the data. All the authors reviewed and approved the final submitted manuscript. SY is responsible for the overall content as guarantor.
Funding This research is supported by grant from the Clinical Research Plan of Shanghai Hospital Development Center (Project No. SHDC2020CR1015B and Project No. SHDC2024CRI019), foundation from Renji Hospital, School of Medicine, Shanghai Jiaotong University (Project No. 2019NYLYCP0202) and The Key Research and Development Project of Xinjiang Uygur Autonomous Region (No. 2022B03002-1).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.