Abstract
Background
To describe real-world clinical outcomes of belimumab (BEL) treatment in patients with systemic lupus erythematosus (SLE), with a particular focus on longitudinal glucocorticoid tapering and exploratory renal analyses.
Methods
This single-center retrospective observational study included 46 patients with SLE who initiated BEL treatment between 2018 and 2025. Clinical and laboratory data, including prednisolone (PSL) dose, disease activity indices, serological markers, urine protein-to-creatinine ratio (UPCR), and estimated glomerular filtration rate (eGFR), were collected at baseline and during follow-up, when available. Longitudinal changes in concomitant background therapies were summarized descriptively. Exploratory renal analyses were performed in patients with lupus nephritis (LN) using changes in the UPCR and eGFR. Drug retention was evaluated using the Kaplan–Meier method. Renal histopathological analyses were performed in a subset with available biopsy data and were considered exploratory because of the heterogeneous clinical backgrounds and variable biopsy timing.
Results
The median PSL dose decreased from 10 mg/day at baseline to 8 mg/day at 24 weeks and 6 mg/day at 52 weeks after BEL initiation, with significant within-patient decreases at both time points (p < 0.001). Overall disease activity remained controlled, while complement levels improved during the follow-up. Among patients with LN (n = 12), UPCR showed a numerical decrease at 24 weeks (p = 0.055) and decreased at 52 weeks (p = 0.016) among patients with paired data, whereas the eGFR showed no significant change at either time point. Exploratory analyses did not identify significant associations between the available renal histopathological indices and changes in proteinuria. BEL treatment was continued in many patients during follow-up, although adverse events, including severe infections and death, were observed.
Conclusions
In this real-world cohort, PSL doses decreased during follow-up after the initiation of BEL, while the overall disease activity remained controlled. In the small LN subgroup, the UPCR decreased at 52 weeks, and no statistically significant change in eGFR was detected. Because this study lacked a comparator group and the renal analyses were exploratory and limited by the small sample size, heterogeneous treatment backgrounds, and variable biopsy timing, these findings should be interpreted as within-patient observational data and require validation in future studies.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s41927-026-00680-6.
Keywords: Systemic lupus erythematosus, Belimumab, Lupus nephritis, Glucocorticoid tapering, Proteinuria, Estimated glomerular filtration rate, Real-world study
Key Points
What is already known about this subject?
• Belimumab is effective in reducing disease activity in patients with systemic lupus erythematosus.
• Its role in glucocorticoid tapering and renal outcomes in real-world clinical settings remains insufficiently characterized.
What does this study add?
• In this single-center real-world cohort, prednisolone doses decreased during follow-up after belimumab initiation, while overall disease activity remained controlled.
• In the small lupus nephritis subgroup, the UPCR decreased at 52 weeks and no statistically significant change in eGFR was detected; however, these renal findings were exploratory and should not be interpreted as evidence of causal effects of treatment.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s41927-026-00680-6.
Introduction
Systemic lupus erythematosus (SLE) is a chronic autoimmune disease characterized by heterogeneous clinical manifestations and a relapsing–remitting disease course [1]. Lupus nephritis (LN) remains one of the most severe complications of SLE and is a major determinant of long-term morbidity and mortality [2, 3]. Despite advances in immunosuppressive therapy, long-term disease management remains challenging, particularly with regard to minimizing glucocorticoid-related toxicity while maintaining disease control.
A long-term Japanese cohort study highlights the importance of limiting organ damage and chronic complications in SLE [4]. Accordingly, treatment strategies emphasize glucocorticoid minimization while maintaining disease control [5]. Therefore, identifying treatment strategies that support glucocorticoid minimization while maintaining disease control is of substantial clinical importance.
Belimumab (BEL), a monoclonal antibody targeting B-cell activating factor, has demonstrated efficacy in reducing disease activity and preventing flares in randomized clinical trials [6, 7]. In active LN, the BLISS-LN randomized trial demonstrated improved renal responses when BEL was added to standard therapy, and a secondary analysis of the same trial evaluated additional kidney outcomes [8, 9]. However, the selected patient population and predefined treatment protocol may limit direct extrapolation to routine clinical practice.
Real-world observational studies describe BEL use in patients with varying disease activity and concomitant therapies [10, 11]. This heterogeneity reflects real-world clinical settings, but complicates the interpretation of treatment response, particularly in patients with LN, where the renal outcomes may be influenced by prior disease activity, treatment history, and the timing of renal assessment.
Renal histopathology provides information on LN classification, activity, chronicity, and prognosis [12–15]. In the present study, the available biopsies were obtained at varying times relative to BEL initiation; the corresponding histopathological analyses were therefore considered exploratory.
Recent evidence includes an indirect comparison of BEL and anifrolumab [16], as well as real-world reports on proliferative LN [17], early active SLE [18], LN treatment phases [19], and glucocorticoid reduction [20]. Patient populations and treatment backgrounds differ across these studies.
Therefore, we conducted a retrospective real-world study to describe the clinical outcomes of BEL treatment in patients with SLE, focusing on longitudinal glucocorticoid tapering and exploratory renal analyses under heterogeneous clinical conditions. In addition, available renal biopsy findings and their timing relative to the initiation of BEL were summarized to contextualize the changes in renal parameters in routine clinical practice.
Methods
Study design and patients
This single-center retrospective observational study consecutively included patients with SLE who initiated BEL therapy between 2018 and 2025 at our institution. Patients were eligible if they fulfilled the American College of Rheumatology classification criteria for SLE [21] and received BEL during the study period. Clinical data were extracted from the electronic medical records. No comparator group was included; therefore, all longitudinal analyses were conducted as within-patient observational analyses.
The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Niigata Rheumatic Center (Approval No. 2025-028). Given the retrospective observational design, informed consent was obtained using an opt-out approach and patients were provided with the opportunity to refuse participation through public disclosure.
Clinical assessments and data collection
Baseline demographic and clinical characteristics, including age, sex, disease duration, and treatment history, were collected at the time of BEL initiation. The disease activity was assessed using the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K) [22]. Flares before the initiation of BEL were ascertained from available medical records. A flare was defined as new or worsening SLE-related manifestations documented by the treating physician that required an increase in the glucocorticoid dose, intravenous methylprednisolone pulse therapy, initiation or intensification of immunosuppressive therapy, or hospitalization. To further describe the clinical context of BEL initiation, clinician-documented reasons for the introduction of BEL were reviewed post hoc and summarized. These reasons were not mutually exclusive and were not used to define formal steroid-dependent or steroid-refractory diseases. In addition, a post hoc descriptive analysis was performed in patients receiving PSL ≥ 10 mg/day at BEL initiation to evaluate PSL changes in patients with a greater need for tapering glucocorticoids.
Laboratory data, including anti-double-stranded DNA antibody titers, complement C3 and C4 levels, total hemolytic complement activity (CH50), serum creatinine, estimated glomerular filtration rate (eGFR), and urine protein-to-creatinine ratio (UPCR), were collected at baseline and 24 and 52 weeks after the initiation of BEL when available. Serum creatinine levels were measured using an enzymatic method. The eGFR was calculated using the Japanese Society of Nephrology equation: eGFR = 194 × (serum creatinine) − 1.094 × age − 0.287, with multiplication by 0.739 for female patients, where serum creatinine is expressed in mg/dL. The time windows for follow-up assessments were defined as ± 14 days around each target time point (24 and 52 weeks). Data outside these windows were treated as missing.
The prednisolone (PSL) dose was recorded at baseline and during follow-up. Longitudinal changes in the PSL dose were analyzed to describe glucocorticoid tapering during BEL treatment. Concomitant background therapies were recorded at the initiation of BEL. Longitudinal dose records for hydroxychloroquine, mycophenolate mofetil, tacrolimus, mizoribine, and azathioprine were summarized at baseline, 24 weeks, and 52 weeks, when available. Because azathioprine was used in only two patients, the detailed clinical courses of these patients are described in Supplementary Table 1. Intravenous cyclophosphamide was summarized separately because it was administered as intermittent pulse therapy before the initiation of BEL rather than as a continuous daily background therapy. No intravenous cyclophosphamide was administered after the initiation of BEL therapy. The use of proteinuria-modifying medications, including renin–angiotensin system inhibitors, sodium–glucose cotransporter 2 inhibitors, and mineralocorticoid receptor antagonists, was reviewed in patients with LN and summarized descriptively as use or non-use at baseline and at 24 and 52 weeks, when available. Percentages were calculated using the number of patients with available medication data at each time point; dose changes, adherence, and indications were not analyzed.
UPCR and eGFR measurements were obtained as part of routine clinical practice, and the follow-up timing was not strictly protocolized because of the retrospective study design. Renal parameters were analyzed at baseline and 24 and 52 weeks after the initiation of BEL, when available. The baseline was defined as the date of BEL initiation.
Assessment of LN
LN was defined as biopsy-proven LN or clinically diagnosed renal involvement attributed to SLE based on persistent proteinuria (urine protein-to-creatinine ratio ≥ 0.5 g/gCr) and/or active urinary sediment, in the absence of alternative causes, as judged by the treating physician. This definition was adopted to reflect routine clinical practice in this retrospective cohort study design. Renal parameters were evaluated in patients with LN using changes in the UPCR and eGFR over time. These renal analyses were considered exploratory because of the small sample size, heterogeneous background therapies, and retrospective nature of the data collection.
Renal biopsy data were available for a subset of patients. The timing of the renal biopsy relative to the initiation of BEL was recorded and summarized as the absolute interval between the biopsy date and the date of BEL initiation. Detailed histopathological analyses were performed only in cases evaluated at a single academic center to ensure the consistency of the pathological assessment. Eligible cases were identified based on the presence of a documented renal biopsy. A histopathological evaluation was performed using an integrated approach incorporating light microscopy, immunofluorescence, and electron microscopy.
Light microscopy findings included the modified National Institutes of Health (NIH) activity and chronicity indices described in the 2018 revision [13], as well as interstitial fibrosis and tubular atrophy (IFTA) and global glomerulosclerosis. Immunofluorescence staining intensities for IgG, IgA, IgM, C3, and C1q were semi-quantitatively graded. The presence or absence of a full-house immunofluorescence pattern was recorded as a pathological finding in the medical records. Electron microscopy was used to evaluate the presence and distribution of electron-dense immune deposits (mesangial, subendothelial, and subepithelial) and tubuloreticular inclusions [23, 24]. Exploratory subgroup analyses integrating light microscopy, immunofluorescence, and electron microscopy were conducted to examine the relationship between the renal histopathological features and changes in proteinuria.
Drug retention and safety assessment
The drug retention of BEL was evaluated using the Kaplan–Meier method. Drug retention was defined as continued BEL treatment without permanent discontinuation. The time to permanent discontinuation was calculated from the date of BEL initiation. Permanent discontinuation of BEL was considered an event in the drug-retention analysis. Temporary interruptions followed by resumption and changes in the route of administration without permanent discontinuation were not considered as events. Patients who were lost to follow-up, transferred to other institutions, died before permanent discontinuation of BEL, or were still receiving BEL at the last observation were censored at the time of the last follow-up.
Adverse events were retrospectively collected from the medical records. Serious adverse events were defined as events that required hospitalization or resulted in death. Causality between BEL and adverse events was assessed based on a retrospective chart review and the treating physician’s judgment. All subgroup analyses of the renal histopathological features were considered exploratory, and no adjustments were made for multiple comparisons.
Statistical analysis
Analyses were performed using available case data because of the retrospective study design and incomplete follow-up measurements.
Continuous variables are presented as medians with interquartile ranges or means with standard deviations, as appropriate. Categorical variables are expressed as numbers and percentages.
Within-patient changes in continuous variables from baseline to 24 and 52 weeks were assessed using the Wilcoxon signed-rank test in patients with paired data for each comparison. Comparisons between two independent groups were performed using the Mann–Whitney U test. Correlations between renal histopathological indices and changes in proteinuria were evaluated using Spearman’s rank correlation coefficient.
Drug retention was analyzed using the Kaplan–Meier method, with permanent discontinuation defined as the event described above. Statistical significance was defined as a two-sided p-value < 0.05. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria) [25]. Given the limited sample size, incomplete repeated measurements, and absence of a comparator group, linear mixed-effects models and other multivariable models were not fitted in this study. Therefore, the results should be interpreted as exploratory only.
Results
Patient characteristics at BEL initiation
A total of 46 patients with SLE who initiated BEL therapy were included in this retrospective observational study. The baseline patient characteristics are summarized in Table 1. The median age at BEL initiation was 47 years, the median age at the onset of SLE was 32.5 years, and the majority of patients were female (82.6%). The median disease duration was 8 years.
Table 1.
Baseline characteristics of patients at belimumab initiation
| Characteristic | Total (n = 46) |
|---|---|
| Demographics | |
| Age at BEL initiation, years | 47 (37.2–62.5) |
| Age at SLE onset, years | 32.5 (23.2–51.5) |
| Female sex, n (%) | 38 (82.6) |
| Disease duration at BEL initiation, years | 8 (0–16) |
| Smoking history, n (%) | |
| Never | 21 (45.7) |
| Ever | 9 (19.6) |
| Unknown | 16 (34.8) |
| Disease characteristics | |
| SLEDAI-2K score | 8 (4.2–14.2) |
| History of LN before BEL initiation, n (%) | 12 (26.1) |
| ISN/RPS class (among patients with LN, n = 12) | |
| Class II, n (%) | 1 (8.3) |
| Class III (± V), n (%) | 5 (41.7) |
| Class IV (± V), n (%) | 5 (41.7) |
| Not available, n (%) | 1 (8.3) |
| Membranous component (Class V; among patients with LN, n = 12), n (%) | 7 (58.3) |
| History of flares before BEL initiation, n (%) | |
| None | 28 (60.9) |
| Once | 9 (19.6) |
| ≥2 | 9 (19.6) |
| Laboratory findings at baseline | |
| Anti-dsDNA antibody (RIA), IU/mL (n = 29) | 1 (1–22) |
| Complement C3, mg/dL (n = 31) | 88.8 ± 27.8 |
| Complement C4, mg/dL (n = 31) | 17.5 ± 7.2 |
| CH50, U/mL (n = 37) | 33.2 ± 11.8 |
| Serum creatinine, mg/dL (n = 45) | 0.83 ± 0.45 |
| eGFR, mL/min/1.73 m² (n = 45) | 72.7 (57.3–83.3) |
| UPCR, g/gCr (patients with LN, n = 9) | 0.48 (0.22–0.86) |
| Treatment at BEL initiation | |
| Prednisolone dose, mg/day | 10 (7.1–21.2) |
| Prednisolone ≥ 10 mg/day, n (%) | 26 (56.5) |
| Hydroxychloroquine, n (%) | 21 (45.7) |
| Concomitant immunosuppressants, n (%) | 34 (73.9) |
| Mycophenolate mofetil, n (%) | 10 (21.7) |
| Tacrolimus, n (%) | 20 (43.5) |
| Mizoribine, n (%) | 10 (21.7) |
| Azathioprine, n (%) | 2 (4.3) |
| Prior intravenous cyclophosphamide before BEL initiation, n (%) | 2 (4.3) |
Data are presented as the median (interquartile range), mean ± standard deviation, or number (%), as appropriate. Percentages were calculated using the total cohort (n = 46) unless otherwise specified. A flare was defined as new or worsening SLE-related manifestations documented by the treating physician that required an increase in glucocorticoid dose, intravenous methylprednisolone pulse therapy, initiation or intensification of immunosuppressive therapy, or hospitalization. Concomitant immunosuppressants were counted when mycophenolate mofetil, tacrolimus, mizoribine, or azathioprine was used at belimumab initiation; treatment categories are not mutually exclusive. Intravenous cyclophosphamide was summarized separately as prior pulse therapy because it was administered before BEL initiation and was not administered after BEL initiation. Values below the lower detection limit were handled as the lower measurable value
Abbreviations: BEL, belimumab; RIA, radioimmunoassay; SLE, systemic lupus erythematosus; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000; LN, lupus nephritis; ISN/RPS, International Society of Nephrology/Renal Pathology Society; dsDNA, double-stranded DNA; C3, complement component 3; C4, complement component 4; CH50, total hemolytic complement activity; Cr, creatinine; eGFR, estimated glomerular filtration rate; IQR, interquartile range; UPCR, urine protein-to-creatinine ratio
At baseline, the median SLEDAI-2K score was eight, indicating moderate disease activity. Twelve patients (26.1%) were diagnosed with LN before the initiation of BEL. The timing of the renal biopsies relative to BEL initiation varied among patients, thus reflecting real-world clinical practice. Among patients with LN, ISN/RPS classes III (± V) and IV (± V) were equally represented, and more than half had a membranous component. The baseline renal function was preserved in most patients, and the median UPCR among patients with LN and available baseline UPCR data was 0.48 g/gCr.
Hydroxychloroquine (HCQ) was used by 21 patients (45.7%) at BEL initiation. Concomitant immunosuppressants were used by 34 patients (73.9%), most commonly tacrolimus, mycophenolate mofetil, and mizoribine, reflecting the heterogeneous real-world background therapies.
Changes in PSL dose after BEL initiation
During follow-up after the initiation of BEL, the PSL dose decreased over time (Fig. 1). The median PSL dose decreased from 10 mg/day at baseline to 8 mg/day at 24 weeks and 6 mg/day at 52 weeks after treatment. Within-patient comparisons using paired data showed significant decreases from baseline to 24 and 52 weeks (p < 0.001 for both comparisons). The number of available observations was 46 at baseline, 36 at 24 weeks, and 32 at 52 weeks. In a post-hoc chart review, the clinician-documented reasons for BEL initiation were not mutually exclusive. BEL was introduced primarily to facilitate PSL tapering or to address the difficulty in PSL tapering in 44 patients (95.7%), and relapse or flare prevention was documented in 22 patients (47.8%). A history of flare-ups before the initiation of BEL was documented in 18 patients (39.1%), including 9 patients with 1 flare and 9 patients with ≥ 2 flares. In a post hoc descriptive analysis of patients receiving PSL ≥ 10 mg/day at the initiation of BEL, the median PSL dose decreased from 15.0 mg/day at baseline to 6.5 mg/day at 52 weeks in patients with paired data (n = 18).
Fig. 1.
Changes in the prednisolone dose after belimumab initiation. Individual data points are shown for all available observations at baseline, 24 weeks, and 52 weeks after belimumab initiation. Boxes indicate the median and interquartile range, and whiskers extend to the most extreme values within 1.5 times the interquartile range. Open circles indicate observations beyond 1.5 times the interquartile range. Numbers below the x-axis indicate the number of available observations at each time point. Changes from baseline to 24 and 52 weeks were assessed using two-sided Wilcoxon signed-rank tests in patients with paired data for each comparison (n = 36 and n = 32, respectively)
Longitudinal changes in concomitant background therapies
Longitudinal changes in concomitant background therapies are summarized in Supplementary Table 1. Among paired observations from baseline to 52 weeks, most patients had unchanged HCQ, mycophenolate mofetil, tacrolimus, and mizoribine doses. Initiation or dose increase was observed in a minority of patients, whereas dose reduction or discontinuation was uncommon. These findings indicate that background therapies were heterogeneous but largely unchanged in paired longitudinal comparisons. After additional chart review, azathioprine was used in 2 patients at the initiation of BEL. Follow-up dose data were unavailable for one patient after transfer to another institution, whereas the other patient discontinued azathioprine 52 weeks after switching to mycophenolate mofetil. Intravenous cyclophosphamide was administered before the initiation of BEL in two patients, but no patients received intravenous cyclophosphamide after the initiation of BEL.
Changes in global disease activity and serological markers
Overall disease activity remained controlled, and several serological markers improved during the follow-up (Fig. 2). The median SLEDAI-2K scores showed a significant reduction at both 24 and 52 weeks relative to baseline (both p < 0.01). Anti-double-stranded DNA antibody titers showed a decreasing trend during the follow-up, although the changes were not statistically significant. In contrast, complement levels, including C3 and C4, and total hemolytic complement activity (CH50), significantly increased at 24 weeks and were maintained at 52 weeks. No significant worsening of serological disease activity was observed during follow-up, despite substantial glucocorticoid tapering.
Fig. 2.
Changes in the disease activity and serological markers after belimumab initiation. Individual data points are shown for all available observations at baseline, 24 weeks, and 52 weeks after belimumab initiation. Boxes indicate the median and interquartile range, and whiskers extend to the most extreme values within 1.5 times the interquartile range. Open circles indicate observations beyond 1.5 times the interquartile range. Numbers below the x-axis indicate the number of available observations at each time point. Changes from baseline to 24 and 52 weeks were assessed using two-sided Wilcoxon signed-rank tests in patients with paired data for each comparison. (A) SLEDAI-2K score; (B) anti-double-stranded DNA antibody (RIA); (C) CH50; (D) complement C3; (E) complement C4. Abbreviations: SLE, systemic lupus erythematosus; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000; dsDNA, double-stranded DNA; RIA, radioimmunoassay; C3, complement component 3; C4, complement component 4; CH50, total hemolytic complement activity; IQR, interquartile range
Changes in renal parameters in patients with LN
Among patients with LN, the UPCR showed a numerical decrease at 24 weeks relative to baseline (p = 0.055) and decreased at 52 weeks (p = 0.016) among patients with paired data (Fig. 3). In contrast, the eGFR did not significantly change from baseline to 24 or 52 weeks (p = 0.625 and p = 0.496, respectively). Thus, the UPCR decreased during follow-up in this small exploratory subgroup, whereas no statistically significant change in eGFR was detected.
Fig. 3.
Changes in renal parameters during follow-up after belimumab initiation in patients with lupus nephritis. Individual data points are shown for all available observations at baseline, 24 weeks, and 52 weeks after belimumab initiation. Boxes indicate the median and interquartile range (IQR), and whiskers extend to the most extreme values within 1.5 times the IQR. Open circles indicate observations beyond 1.5 times the IQR. Numbers below the x-axis indicate the number of available observations at each time point. Changes from baseline to 24 and 52 weeks were assessed using two-sided Wilcoxon signed-rank tests in patients with paired data for each comparison. (A) Urine protein-to-creatinine ratio (UPCR); (B) estimated glomerular filtration rate (eGFR). Abbreviations: Cr, creatinine; eGFR, estimated glomerular filtration rate; IQR, interquartile range; UPCR, urine protein-to-creatinine ratio
Paired data for both ΔPSL and ΔUPCR at 52 weeks were available for 9 of the 12 LN patients. No significant association was observed between the magnitude of PSL tapering (ΔPSL) and changes in proteinuria (ΔUPCR) (Spearman’s ρ = −0.22, p = 0.574), indicating no statistically significant relationship between the magnitude of glucocorticoid tapering and short-term changes in proteinuria in this small exploratory subgroup (Supplementary Figure S1). The use of proteinuria-modifying medications among patients with LN is summarized in Supplementary Table 1. Among the 12 patients with LN, renin–angiotensin system inhibitors were used in 4/12 (33.3%) at baseline, 6/11 (54.5%) at 24 weeks, and 7/10 (70.0%) at 52 weeks. Sodium–glucose cotransporter 2 inhibitors were used in 3/12 patients (25.0%) at baseline, 3/11 (27.3%) at 24 weeks, and 3/10 (30.0%) at 52 weeks. No patients received mineralocorticoid receptor antagonists at any time point. These medications were descriptively reviewed as potential background factors; however, dose changes, adherence, and indications were not analyzed.
Drug retention of BEL
Drug retention of BEL was evaluated using the Kaplan–Meier method (Fig. 4). BEL treatment was continued in many patients during the follow-up, although permanent discontinuation occurred because of disease exacerbation or adverse events. Temporary interruptions followed by resumption and changes in the route of administration without permanent discontinuation were not considered as events. Adverse events and deaths observed during BEL treatment are summarized in Table 2.
Fig. 4.
Drug retention after belimumab initiation. Data are shown for patients who initiated belimumab (n = 46). Drug retention was defined as continued belimumab treatment without permanent discontinuation. Permanent discontinuation of belimumab was considered an event. Temporary interruptions followed by resumption and changes in the route of administration without permanent discontinuation were not considered events. Patients who were lost to follow-up, transferred to other hospitals, died before permanent discontinuation of belimumab, or were still receiving belimumab at the last follow-up were censored at the time of the last observation. Drug retention was estimated using the Kaplan–Meier method. Plus signs indicate censored observations. Number at risk is shown below the x-axis
Table 2.
Non-fatal adverse events and deaths observed during belimumab treatment
| System organ class/outcome | Adverse event or outcome | n |
|---|---|---|
| Non-fatal infections | CMV infection | 1 |
| Pulmonary aspergilloma | 1 | |
| Kaposi varicelliform eruption during concomitant MMF treatment | 1 | |
| Deaths | Death, total | 3 |
| Infection-related death | 2 | |
| Non-infection-related death | 1 | |
| Psychiatric | Depression | 2 |
| Gastrointestinal | Diarrhea | 1 |
Data are presented as the number of adverse events or deaths. Adverse events were retrospectively collected from electronic medical records. Causality was assessed based on the treating physicians’ judgment. Serious adverse events were defined as events requiring hospitalization or resulting in death. Deaths are presented separately from non-fatal adverse events; therefore, fatal infections are included under Deaths and not under Non-fatal infections. Three deaths were identified during follow-up: two were classified as infection-related and one as non-infection-related based on retrospective chart review. A direct causal relationship with belimumab could not be established retrospectively in any of these cases
Abbreviations: CMV, cytomegalovirus; MMF, mycophenolate mofetil
Renal histopathological features and exploratory analyses
These analyses were exploratory. Renal biopsy specimens evaluated at the academic center demonstrated heterogeneous renal histopathological findings (Supplementary Figure S2 and Supplementary Table 2). Among the 12 cases with a renal biopsy, biopsies were performed before the initiation of BEL in 10 patients and after the initiation of BEL in 2 patients. The median absolute interval between the renal biopsy and the initiation of BEL was 5.5 months (IQR, 0.9–21.6 months; range, 0.5–398.3 months). Exploratory analyses did not reveal any significant associations between the available renal histopathological features and changes in proteinuria after BEL initiation.
Integrated evaluation by light microscopy, immunofluorescence, and electron microscopy revealed complex and overlapping renal pathological features (Supplementary Table 2). Most cases showed immune-complex deposition involving multiple compartments, a full-house immunofluorescence pattern, and in some cases, tubuloreticular inclusions, consistent with the multifaceted immunopathology of LN. Notably, in a representative patient who underwent a repeat renal biopsy before the initiation of BEL (in 2018 and 2022), the modified NIH activity index decreased, whereas the modified NIH chronicity index increased over time, reflecting the progression of chronic irreversible lesions despite reduced inflammatory activity. This case illustrates that the renal histopathological indices can reflect the cumulative disease burden and treatment history rather than disease activity at the time of BEL initiation.
Safety
The adverse events and deaths observed during BEL treatment are summarized in Table 2. Deaths were presented separately from non-fatal adverse events. The non-fatal adverse events included cytomegalovirus infection, pulmonary aspergilloma, Kaposi varicelliform eruption during concomitant mycophenolate mofetil treatment, psychiatric symptoms, and gastrointestinal events. Serious adverse events included severe infections and death. During follow-up, three deaths were identified: two infection-related deaths and one non-infection-related death.
The first patient was a 33-year-old woman with a 10-year history of SLE, including class II lupus nephritis, who initiated BEL while receiving PSL 10 mg/day and mizoribine 100 mg/day. She had a history of depression and was receiving outpatient psychiatric care. She died 159 weeks after the initiation of BEL, while continuing BEL; the recorded causes included hanging, suspected acute drug intoxication, and liver failure, and no infection was documented. At the time of death, she was receiving PSL 5 mg/day and mizoribine 100 mg/day. No clear causal relationship with BEL was noted in the medical records; her pre-existing depression, ongoing psychiatric follow-up, and psychosocial factors were considered relevant.
The second patient was an 88-year-old woman with a 2-year history of SLE and multiple comorbidities, including antiphospholipid syndrome, cerebral infarction, atrial fibrillation, chronic heart failure, and recurrent infection-prone conditions. BEL was initiated while she was receiving PSL 8 mg/day, HCQ 200 mg/day, and mizoribine 150 mg/day. She died 12 weeks after the initiation of BEL from infection due to urinary tract infection, dehydration, lower-leg skin erosion, and gangrene. At the time of death, she was receiving PSL 10 mg/day without concomitant immunosuppressants. Although a direct causal relationship with BEL could not be established retrospectively, this fatal infection was considered multifactorial, reflecting advanced age, poor baseline general condition, comorbidities, glucocorticoid exposure, and concomitant immunosuppressive therapy.
The third patient was a 72-year-old woman with a 28-year history of SLE and multiple comorbidities. BEL was initiated after an increase in PSL to 20 mg/day, followed by tapering to 17.5 mg/day, while she was receiving concomitant tacrolimus 1 mg/day. She developed bilateral pneumonia and died 11 weeks after the initiation of BEL, while receiving PSL 12.5 mg/day and tacrolimus 1 mg/day. Although a causal relationship with BEL could not be completely excluded retrospectively, this fatal pneumonia was considered multifactorial, including glucocorticoid exposure, concomitant tacrolimus, and the patient’s clinically vulnerable background.
Discussion
In this single-center, retrospective, real-world cohort study, prednisolone doses decreased during the 52 weeks after the initiation of BEL, while overall disease activity remained controlled. In the small LN subgroup, the UPCR decreased at 52 weeks, and no statistically significant change in eGFR was detected during follow-up. However, because this study did not include a comparator group and the background therapies were heterogeneous, these findings should be interpreted as exploratory within-patient longitudinal observations rather than evidence of causal treatment effects. The observed decrease in the PSL dose is consistent with real-world reports describing glucocorticoid reduction during BEL treatment. However, in the absence of a non-BEL comparator group, it remains possible that part of the observed tapering reflects clinical stability, physician-directed tapering, or other concurrent treatment decisions. Therefore, the present findings should be interpreted as describing glucocorticoid tapering during BEL treatment rather than demonstrating that BEL caused glucocorticoid reduction.
We also summarized the longitudinal changes in concomitant background therapies to address potential confounding factors related to treatment modification. Although most paired observations showed unchanged doses of HCQ, mycophenolate mofetil, tacrolimus, and mizoribine, the background therapies were not standardized and were modified in some patients during follow-up. Therefore, the observed changes in the PSL dose and renal parameters cannot be solely attributed to BEL.
Given the potential contribution of glucocorticoid exposure to long-term organ damage in SLE [26] and the emphasis on glucocorticoid minimization in current treatment recommendations [27], the observed decrease in the PSL dose while disease activity remained controlled is clinically relevant in routine practice. However, because this study lacked a non-BEL comparator group, the observed tapering should be interpreted as a longitudinal change during BEL treatment rather than as evidence that BEL itself caused a decrease in the PSL dose.
International recommendations emphasize HCQ as background therapy for SLE [5, 27], and the 2019 Japanese guidance also supports its consideration across disease manifestations [28]. In the present cohort, HCQ was used by 21 patients (45.7%) at BEL initiation. In a cohort of long-term HCQ users, Asian patients had a higher risk of the pericentral retinopathy pattern than non-Hispanic White patients [29]. Whether retinopathy concerns contributed to the relatively low HCQ use in this cohort cannot be determined from the present data. In a Japanese single-center observational study using data from 2019, approximately half of patients with SLE had a history of HCQ use [30]; this measure is not directly equivalent to the proportion receiving HCQ at BEL initiation in our cohort. In this context, describing glucocorticoid tapering during BEL treatment in a cohort with relatively limited HCQ use may be clinically informative, without establishing a causal effect of BEL.
BEL was initiated in heterogeneous clinical contexts in this cohort. The available biopsies were obtained at variable times relative to BEL initiation and may not represent renal activity at the start of treatment. These features limit interpretation of the exploratory histopathological analyses.
The BLISS-LN randomized trial enrolled patients with biopsy-proven active LN, including class III or IV with or without class V, and pure class V, who received standard background therapy [8]. A subsequent publication evaluated additional kidney outcomes in a secondary analysis of the same trial [9].
In contrast to BLISS-LN, the present study evaluated BEL in a real-world clinical setting. Renal biopsy data were available for patients with LN; however, the timing of biopsies relative to the initiation of BEL varied considerably. Class III (±V) and class IV (±V) were each recorded in 5 of the 12 patients with LN (41.7%), and 7 patients (58.3%) had a membranous component. Differences in biopsy timing and clinical context limit direct comparison between the present cohort and trial populations and complicate interpretation of the exploratory histopathological analyses.
BLISS-LN and its secondary analysis evaluated clinical renal responses and kidney function outcomes [8, 9]. These findings provide a clinical context for the exploratory analyses of the heterogeneous biopsy findings available in the present study. The renal biopsy findings were obtained at variable time points relative to the initiation of BEL, and histopathological activity at the time of the biopsy did not necessarily reflect the disease activity at the time of treatment initiation. In the present study, neither the modified NIH activity index nor the modified NIH chronicity index showed a significant correlation with proteinuria changes. This absence of a statistically significant association in a small, heterogeneous subgroup should not be interpreted as evidence that renal histopathology has no prognostic relevance [15]. The sample size and variable biopsy timing limit the interpretation of these exploratory analyses.
An integrated evaluation of renal pathology illustrated LN heterogeneity. Renal biopsy specimens assessed at an academic center demonstrated diverse patterns of activity, chronicity, immune-complex deposition, and ultrastructural features. These findings illustrate the range of morphological features encountered in LN [15]. Their prognostic or predictive value during BEL treatment cannot be established from the present exploratory analyses.
In the exploratory LN subgroup, the UPCR decreased at 52 weeks, whereas no statistically significant change in eGFR was detected during follow-up. These renal findings are hypothesis-generating and should be interpreted with caution because of the small sample size, lack of a comparator group, heterogeneous background therapies, and non-protocolized follow-up period. In particular, the present study could not determine whether the observed decrease in UPCR was attributable to BEL, concomitant treatments, the natural course of LN, or other clinical factors.
Serious infections and deaths were observed in a small number of patients, including two infection-related deaths and one non-infection-related death. Although the retrospective design precludes definitive attribution of these events to BEL, fatal infections occurred in clinically vulnerable patients with glucocorticoid exposure, comorbidities, infection-prone conditions, and concomitant or prior immunosuppressive therapy. These findings underscore the importance of careful infection monitoring, particularly in patients receiving glucocorticoids or other immunosuppressive therapies.
This study has several limitations. First, the retrospective single-center design and small sample size limit the statistical power and generalizability, particularly for exploratory renal and histopathological analyses. Second, the study did not include a comparator group. Therefore, causal inference is not possible, and the observed glucocorticoid tapering may partly reflect physician-directed tapering during clinical stability, regression to the mean, or other treatment decisions, rather than a direct effect of BEL. Third, the treatment regimens were not standardized. Although longitudinal changes in selected concomitant background therapies were summarized, residual confounding by immunosuppressive therapy and other treatment modifications could not be excluded. Fourth, renal biopsy data were available only for a subset of patients, and the timing of biopsies relative to the initiation of BEL varied; therefore, the available histopathological findings may not have reflected renal activity at the initiation of BEL in all patients. Finally, because of the limited sample size, incomplete repeated measurements, and absence of a comparator group, linear mixed-effects models or other multivariable models were not used. Accordingly, all longitudinal and renal analyses should be interpreted as exploratory. Although the use of proteinuria-modifying medications, including renin–angiotensin system inhibitors, sodium–glucose cotransporter 2 inhibitors, and mineralocorticoid receptor antagonists, was reviewed in patients with LN, detailed dose changes, adherence, and indications were not systematically available; therefore, their potential influence on proteinuria changes could not be fully assessed.
In conclusion, prednisolone doses decreased during follow-up after the initiation of BEL, whereas overall disease activity remained controlled in this heterogeneous real-world SLE cohort. In the small LN subgroup, the UPCR decreased at 52 weeks, and no statistically significant change in eGFR was detected; however, these renal findings were exploratory and should not be interpreted as evidence of a causal effect of BEL on proteinuria. Because this study lacked a comparator group and was limited by its retrospective design, small sample size, heterogeneous background therapies, and variable renal biopsy timing, further controlled studies are needed to validate these observations.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors thank the clinical staff involved in the care of SLE and LN patients. The authors also thank Japan Medical Communication for English language editing. During the preparation of this manuscript, the authors used ChatGPT (OpenAI) to assist with language editing, drafting and revising text, and preparing figures based on study data. After using this tool, the authors reviewed and edited all outputs as needed and take full responsibility for the content of the publication.
Author contributions
MS conceived the study, collected and analyzed the data, and drafted the manuscript. All authors contributed to data interpretation, critically revised the manuscript, and approved the final version of the manuscript.
Funding
This research received no specific grants from any funding agency in the public, commercial, or not-for-profit sector.
Data availability
The datasets generated and/or analyzed during the current study are not publicly available because of restrictions related to patient confidentiality, but are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
This study was conducted in accordance with the principles of the Declaration of Helsinki. The study protocol was approved by the Ethics Committee of Niigata Rheumatic Center (Approval No. 2025-028). Given the retrospective nature of the study, informed consent was obtained using an opt-out approach, which was made available through public disclosure.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Tsokos GC. Systemic lupus erythematosus. N Engl J Med. 2011;365:2110–21. 10.1056/NEJMra1100359. [DOI] [PubMed] [Google Scholar]
- 2.Anders HJ, Saxena R, Zhao MH, et al. Lupus nephritis. Nat Rev Dis Primers. 2020;6:7. 10.1038/s41572-019-0141-9. [DOI] [PubMed] [Google Scholar]
- 3.Almaani S, Meara A, Rovin BH. Update on lupus nephritis. Clin J Am Soc Nephrol. 2017;12:825–35. 10.2215/CJN.05780616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wada Y, Hasegawa H, Saeki T, Ito S, Kuroda T, Nakano M, Narita I. Long-term prognosis and factors associated with damage accrual in Japanese patients with systemic lupus erythematosus. Clin Exp Nephrol. 2018;22:597–602. 10.1007/s10157-017-1491-9. [DOI] [PubMed] [Google Scholar]
- 5.van Vollenhoven RF, Mosca M, Bertsias G, et al. Treat-to-target in systemic lupus erythematosus: recommendations from an international task force. Ann Rheum Dis. 2014;73:958–67. 10.1136/annrheumdis-2013-205139. [DOI] [PubMed] [Google Scholar]
- 6.Navarra SV, Guzmán RM, Gallacher AE, et al. Efficacy and safety of belimumab in patients with active systemic lupus erythematosus: a randomised, placebo-controlled, phase 3 trial. Lancet. 2011;377:721–31. 10.1016/S0140-6736(10)61354-2. [DOI] [PubMed] [Google Scholar]
- 7.Furie R, Petri M, Zamani O, et al. A phase III, randomized, placebo-controlled study of belimumab, a monoclonal antibody that inhibits B lymphocyte stimulator, in patients with systemic lupus erythematosus. Arthritis Rheum. 2011;63:3918–30. 10.1002/art.30613. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Furie R, Rovin BH, Houssiau F, et al. Two-year, randomized, controlled trial of belimumab in lupus nephritis. N Engl J Med. 2020;383:1117–28. 10.1056/NEJMoa2001180. [DOI] [PubMed] [Google Scholar]
- 9.Rovin BH, Furie R, Teng YKO, et al. A secondary analysis of the Belimumab International Study in Lupus Nephritis trial examined effects of belimumab on kidney outcomes and preservation of kidney function in patients with lupus nephritis. Kidney Int. 2022;101(2):403–13. 10.1016/j.kint.2021.08.027. [DOI] [PubMed] [Google Scholar]
- 10.Touma Z, Sayani A, Pineau CA, et al. Belimumab use, clinical outcomes and glucocorticoid reduction in patients with systemic lupus erythematosus receiving belimumab in clinical practice settings: results from the OBSErve Canada Study. Rheumatol Int. 2017;37:865–73. 10.1007/s00296-017-3682-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Iaccarino L, Bettio S, Reggia R, et al. Effects of belimumab on flare rate and expected damage progression in patients with active systemic lupus erythematosus. Arthritis Care Res (Hoboken). 2017;69(1):115–23. 10.1002/acr.22971. [DOI] [PubMed] [Google Scholar]
- 12.Weening JJ, D’Agati VD, Schwartz MM, et al. The classification of glomerulonephritis in systemic lupus erythematosus revisited. J Am Soc Nephrol. 2004;15:241–50. 10.1097/01.ASN.0000108969.21691.5D. [DOI] [PubMed] [Google Scholar]
- 13.Bajema IM, Wilhelmus S, Alpers CE, et al. Revision of the International Society of Nephrology/Renal Pathology Society classification for lupus nephritis: clarification of definitions, and modified National Institutes of Health activity and chronicity indices. Kidney Int. 2018;93:789–96. 10.1016/j.kint.2017.11.023. [DOI] [PubMed] [Google Scholar]
- 14.Hsieh C, Chang A, Brandt D, Guttikonda R, Utset TO, Clark MR. Predicting outcomes of lupus nephritis with tubulointerstitial inflammation and scarring. Arthritis Care Res (Hoboken). 2011;63(6):865–74. 10.1002/acr.20441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Parikh SV, Alvarado A, Malvar A, Rovin BH. The kidney biopsy in lupus nephritis: past, present, and future. Semin Nephrol. 2015;35(5):465–477. 10.1016/j.semnephrol.2015.08.008. [DOI] [PubMed] [Google Scholar]
- 16.Neupane B, Shukla P, Slim M, Martin A, Petri M, Bertsias GK, et al. Belimumab versus anifrolumab in adults with systemic lupus erythematosus: an indirect comparison of clinical response at 52 weeks. Lupus Sci Med. 2023;10(1):e000907. 10.1136/lupus-2023-000907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Vidal-Montal P, Fabregat A, Altabás-González I, et al. Real-world experience with belimumab-based triple therapy in proliferative lupus nephritis: data from the BEL-SPAIN Registry. Lupus Sci Med. 2026;13(1):e001825. 10.1136/lupus-2025-001825. [DOI] [PMC free article] [PubMed]
- 18.Gatto M, Cruciani C, Bellis E, Pilo PFG, Radin M, Iaccarino L, et al. Administration of belimumab prior to standard immunosuppression in patients with early active lupus hinders accrual of new EULAR/ACR criteria within the first 12 months of treatment. Lupus Sci Med. 2025;12(2):e001662. 10.1136/lupus-2025-001662. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hou S, Tang Z, Zhou X, et al. Efficacy and safety of belimumab in lupus nephritis: a real-world retrospective observational study stratified by treatment phase and age. Clin Rheumatol. 2026;45:2651–63. 10.1007/s10067-026-08026-w. [DOI] [PubMed] [Google Scholar]
- 20.Jaïs M, Roblot P, Maillot F, Hamidou M, Enfrein A, Lescoat A, et al. Belimumab corticosteroid-sparing treatment in systemic lupus erythematosus: a real-life observational study (BESST study). Rheumatol Int. 2024;44:2465–71. 10.1007/s00296-024-05589-2. [DOI] [PubMed] [Google Scholar]
- 21.Hochberg MC. Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum. 1997;40:1725. 10.1002/art.1780400928. [DOI] [PubMed] [Google Scholar]
- 22.Gladman DD, Ibañez D, Urowitz MB. Systemic Lupus Erythematosus Disease Activity Index 2000. J Rheumatol. 2002;29:288–91. [PubMed] [Google Scholar]
- 23.Truong L, Seshan SV. Lupus nephritis: the significant contribution of electron microscopy. Glomerular Dis. 2021;1(4):180–204. 10.1159/000516790. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Nossent J, Raymond W, Ognjenovic M, Kang A, Chakera A, Wong D. The importance of tubuloreticular inclusions in lupus nephritis. Pathology. 2019;51(7):727–32. 10.1016/j.pathol.2019.07.007. [DOI] [PubMed] [Google Scholar]
- 25.Kanda Y. Investigation of the freely available easy-to-use software ‘EZR’ for medical statistics. Bone Marrow Transplant. 2013;48(3):452–58. 10.1038/bmt.2012.244 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Thamer M, Hernán MA, Zhang Y, Cotter D, Petri M. Prednisone, lupus activity, and permanent organ damage. J Rheumatol. 2009;36(3):560–64. 10.3899/jrheum.080828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Fanouriakis A, Kostopoulou M, Andersen J, et al. EULAR recommendations for the management of systemic lupus erythematosus: 2023 update. Ann Rheum Dis. 2024;83(1):15–29. 10.1136/ard-2023-224762. [DOI] [PubMed] [Google Scholar]
- 28.Research Group on Autoimmune Diseases; Japan College of Rheumatology, editors. Guideline for the management of systemic lupus erythematosus 2019 (in Japanese). Tokyo: Nanzando; 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Jorge AM, Melles RB, Marmor MF, Zhou B, Zhang Y, Choi HK. Risk factors for hydroxychloroquine retinopathy and its subtypes. JAMA Netw Open. 2024;7(5):e2410677. 10.1001/jamanetworkopen.2024.10677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Manabe A, Sada RM, Miyake H, Akebo H, Tsugihashi Y, Hatta K. An observational study to identify causative factors for not using hydroxychloroquine in systemic lupus erythematosus. Sci Rep. 2024;14:7750. 10.1038/s41598-024-58463-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and/or analyzed during the current study are not publicly available because of restrictions related to patient confidentiality, but are available from the corresponding author upon reasonable request.