Safety and Effectiveness of Intravenous Belimumab in Patients with Systemic Lupus Erythematosus in South Korea: A Multicentre, Observational, Real-World Study

Abstract

Background

In South Korea, belimumab is indicated for patients with systemic lupus erythematosus (SLE), including paediatric patients aged ≥ 5 years, and for adults with lupus nephritis, based on clinical trial evidence; however, real-world data in this population remain limited.

Objective

To assess the real-world safety and effectiveness of intravenous (IV) belimumab administered in patients with SLE in South Korea.

Methods

This observational study (GSK Study 216984) enrolled patients with active SLE and patients with lupus nephritis across 18 South Korean institutions (July 2021–February 2023). Patients received on-label IV belimumab plus standard therapy (antimalarials, glucocorticoids and immunosuppressants). Data were collected for a minimum of 48 weeks as part of routine clinical practice (July 2021–March 2024). Safety was assessed by monitoring the incidence of adverse events (AEs), adverse drug reactions (ADRs), serious AEs (SAEs)/ADRs and AEs of special interest (including psychiatric events) and summarised descriptively. Changes in SELENA-SLEDAI scores, laboratory biomarkers and glucocorticoid dosage from baseline to week 24 or 48 were tested using the Wilcoxon signed-rank test.

Results

Of 126 enrolled patients, 105 were eligible for the ≥ 48-week safety evaluation (completed ≥ 24 weeks of belimumab treatment). Patients were mostly female (91.4%, n = 96/105), with mean (SD) SLE duration of 9.9 (8.1) years; 45.7% (n = 48/105) of patients had lupus nephritis. AEs and SAEs were experienced by 66.7% (n = 70/105) and 18.1% (n = 19/105) of patients; most events were mild or moderate in severity and unlikely to be related to belimumab. ADRs were reported for 3.8% (n = 4/105) of patients, and psychiatric events for 2.9% (n = 3/105). SELENA-SLEDAI and laboratory biomarkers significantly improved, and glucocorticoid dosage significantly reduced, between baseline and weeks 24 and 48.

Conclusions

This real-world study identified no new safety concerns and demonstrated effectiveness of belimumab, supporting its use in Korean patients.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40801-026-00556-8.

Plain Language Summary

Belimumab is a medication for treating lupus. It works by blocking a protein that is part of the immune system, which reduces inflammation and helps control symptoms of the disease. Because there is little current information about belimumab use in Korean patients, this study assessed if belimumab is a safe and effective medication for Korean patients with lupus. Over 48 weeks, 105 patients in South Korea received belimumab alongside their regular medications. The study found that while some patients experienced unwanted symptoms, most were mild or moderate, and in nearly all cases, these symptoms were not linked to treatment with belimumab. None of the patients experienced life-threatening symptoms. Belimumab also helped reduce the severity of lupus, which was shown with tests and lab scores developed specifically for measuring lupus. The results from this study show that there are no new safety concerns with belimumab use and it is a beneficial option for managing lupus in the real world, supporting findings from earlier clinical trials which showed similar results.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40801-026-00556-8.

Key Points

This observational study assessed the safety and effectiveness of intravenous belimumab (≥ 48 weeks) in Korean patients with systemic lupus erythematosus (SLE), where a subset also had lupus nephritis.

Safety reports indicated that 66.7% of patients experienced 181 adverse events, mostly mild or moderate and unrelated to belimumab, with no deaths or life-threatening serious adverse events; additionally, effectiveness was demonstrated by significant reductions in SELENA-SLEDAI scores, improvements in laboratory biomarkers and reduced glucocorticoid doses at weeks 24 and 48 versus baseline.

The safety profile of belimumab is consistent with global reports, with our study further demonstrating the effectiveness of belimumab in SLE and lupus nephritis among Korean patients in a real-world setting.

Introduction

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease that may affect different organ systems and is associated with higher morbidity and mortality rates compared with the healthy population [1, 2].

The prevalence of SLE and specific manifestations differ among ethnic and racial groups; Hispanic and Latino populations, as well as those of African and Asian descent, are more frequently affected by SLE and have worse long-term prognosis than Caucasians [3]. In South Korea, the annual prevalence and severity of SLE increased from 21.25 to 35.45/100,000 people in 2005–2015 [4]. Patients of Asian descent are diagnosed at a slightly younger age than other ethnicities, present with a higher rate of neuropathies and lupus nephritis than Caucasians and have higher mortality rates; however, these factors may further vary among specific Asian subpopulations [3].

Traditionally, treatment options available to patients with SLE and lupus nephritis, a severe kidney manifestation of SLE, include antimalarials, glucocorticoids, immunosuppressants and biological agents [5]. One such biological agent is belimumab, a B-cell modulator [6]. It targets the central immunopathogenic pathway in SLE by selectively inhibiting B-lymphocyte stimulator (BLyS) and reducing the autoreactive B cells that drive disease activity. By modulating autoreactive B cells, belimumab can reduce ongoing disease activity while preserving the immune system’s key functions [7, 8].

Since 2011, belimumab has been approved in many countries worldwide, including in South Korea [6, 9–11]; results of a phase 3 trial conducted in North East Asia support belimumab’s approval in adult patients with SLE from South Korea [12]. Further long-term extension study data and post hoc subgroup analyses of South Korean patients also demonstrated the efficacy of belimumab in this patient population [13, 14]. In addition, real-world studies conducted in other Asian populations, such as China, have supported the safety and effectiveness of belimumab [15, 16]. In South Korea, belimumab is also indicated alongside standard therapy for paediatric patients with SLE (aged ≥ 5) and for adults with lupus nephritis [11].

Although clinical trials demonstrated favourable safety and efficacy profiles of belimumab and real-world evidence from other Asian populations supports these findings [12–16], real-world data in Korean patients remain lacking. The aim of this observational study was to assess the safety and effectiveness of intravenous (IV) belimumab in real-world practice, among adult patients with SLE in South Korea.

Methods

Study Design

This was an open-label, non-interventional, observational study (GSK 216984) assessing the safety and effectiveness of IV belimumab in patients with SLE. Patients were recruited between July 2021 and February 2023, and outcomes were monitored between July 2021 and March 2024. The study was conducted at 18 study institutions, under the care of 22 principal investigators, who guided patients to scheduled outpatient visits, according to the on-label dosing cycle of IV belimumab (Supplementary Table 1).

Patients

Eligible patients were adults and children (≥ 5 years of age) with autoantibody-positive, active SLE and adults with active lupus nephritis, who were receiving standard therapy and initiated IV belimumab in accordance with the domestic label. In line with the belimumab reimbursement criteria of the Korea Health Insurance Review and Assessment Service (HIRA), patients with severe active lupus nephritis (defined as proteinuria > 6 g/24 h or equivalent measured by spot urine protein/creatinine ratio [uPCR] or serum creatinine > 2.5 mg/dL, or requiring haemodialysis within 90 days before belimumab initiation) are not eligible for belimumab prescriptions and were excluded from this study. Patients allergic to any component of belimumab were also excluded. Informed consent for study participation, collection and use of personal data was obtained from patients.

Outcomes

Safety was assessed for a minimum of 48 weeks since belimumab initiation, by monitoring the incidence rate of adverse events (AEs), adverse drug reactions (ADRs), serious AEs (SAEs)/ADRs, unexpected AEs/SAEs/ADRs/serious ADRs and AEs of special interest (AESI) by system organ class (SOC): psychiatric events (depression and suicidality), infections and infestations and progressive multifocal leukoencephalopathy. Unexpected AEs were defined as adverse reactions whose nature or severity was not consistent with the applicable product information (e.g., investigator’s brochure or prescribing information). An investigator assessed the causal relationship of IV belimumab with each AE, considering the natural progression of pre-existing disease, concurrent therapy and other risk factors. The causal relationship was classified as ‘certain,’ ‘fairly certain,’ ‘possible,’ ‘difficult to evaluate,’ ‘unable to evaluate’ and ‘unlikely.’ If the causal relationship with IV belimumab is ‘certain,’ ‘fairly certain,’ ‘possible,’ ‘difficult to evaluate’ or ‘unable to evaluate,’ it was defined as an ADR. Events were classified as serious ADRs if they resulted in death, were life-threatening, required hospitalisation or prolongation of existing hospitalisation, resulted in persistent or significant disability/incapacity, involved congenital anomalies/birth defects or were otherwise considered medically important.

Data were collected during scheduled visits via electronic case report forms: every 2 weeks (±7 days) until week 8 (visit 4), and at 4-weekly intervals (±7 days) thereafter. Training for investigators on the use of the electronic data capture system and on the study protocol was provided during the study initiation visit. Patients were included in all analyses if they had at least 24 weeks of belimumab treatment. A subgroup analysis was also performed for patients who received < 24 weeks of belimumab. All assessed events were reported according to Medical Dictionary for Regulatory Activities (MedDRA) Version 27.0, by SOC and preferred terms.

Safety of Estrogens in Lupus Erythematosus National Assessment-SLE Disease Activity Index (SELENA-SLEDAI) scores were assessed at baseline, week 24 and 48. Mean (standard deviation [SD]) change in SELENA-SLEDAI score was evaluated between week 24 or 48 and baseline. The proportion of patients achieving a ≥ 4-point reduction in SELENA-SLEDAI score at weeks 24 and 48 was also assessed. Presented data include patients with available data only, who received belimumab treatment for at least 24 weeks, or for a subset of patients treated for ≥ 48 weeks only.

Mean (SD) change in laboratory biomarkers, including anti-double-stranded DNA (anti-dsDNA) autoantibody, complement component 3 (C3), complement component 4 (C4) and uPCR and change in glucocorticoid dosage (mg/dose), were assessed between weeks 24 or 48 and baseline, among patients with available data. Data on laboratory testing were collected only if the tests were performed under routine medical practice.

Sociodemographic, clinical and pharmacotherapeutic data were collected at baseline (Supplementary Table 2).

Statistical Analysis

The target sample size for this study of 100 patients (or up to 130 patients, if taking into account the dropout rate). The purpose of the target sample was to detect the risk of psychiatric events (including depression and suicidality), which was identified in the risk management plan. In the BLISS-52 trial (BEL110752; NCT00424476) the incidence of psychiatric disorders was 13.8% and the incidence of depression was 3.1% [17]. Therefore, a sample size of 100 patients was deemed sufficient to provide 95% power to detect all AEs with an incidence rate of ≥ 3%.

Continuous and categorical variables were summarised descriptively (patient numbers, means and SD and patient numbers and percentages, respectively).

Missing data were handled as follows: patients with missing year of diagnosis (SLE or lupus nephritis) or year of menopause were excluded from duration analyses. Partial dialysis start dates were imputed using prespecified rules (date of informed consent for year–month entries; 1 January for missing month; first day of the month for missing day). All other variables were analysed as observed without further imputation.

The changes in SELENA-SLEDAI scores, biomarker levels and glucocorticoid dosage between week 24 or 48 and baseline were tested using Wilcoxon signed-rank test. Normality of the data was assessed using the Shapiro–Wilk test prior to statistical testing.

A multiple logistic regression analysis was conducted to identify factors influencing the occurrence of AEs. AE occurrence (yes/no) was the dependent variable. Univariable analyses (chi-square test or Fisher’s exact test, as appropriate) were first conducted to identify potential covariates, with variables with p < 0.05 included in the multivariable model. Covariates included in the final model were delivery history (yes/no), concomitant medication history (yes/no), other medical history (yes/no) and administration period relative to the median (less than the median/equal to or greater than the median). Odds ratios and corresponding 95% Wald confidence intervals (CIs) were calculated to estimate the association between factors and AE occurrence. Analyses were conducted in SAS (version 9.4 or above).

Results

Patient Population

Of 126 enrolled patients, 21 were excluded from the main analysis for having fewer than 24 weeks of belimumab administration, resulting in a total of 105 patients (83.3%) included for the safety and effectiveness evaluation. Out of those patients, 87 (82.9%) received belimumab for ≥ 48 weeks.

Patient baseline demographics and characteristics are presented in Table 1. Most patients were female (91.4%, n = 96/105), with a mean (SD) age of 35.7 (10.0) years and mean (SD) disease duration of 9.9 (8.1) years. Of all patients who had an SLE diagnosis, 45.7% (n = 48/105) were also diagnosed with lupus nephritis, with mean (SD) duration of 8.2 (7.3) years. A minority of patients had hepatic impairment including hepatitis (4.8%, n = 5/105), kidney impairment (3.8%, n = 4/105) and allergies (1.9%, n = 2/105).

Table 1.

Patient demographics, baseline characteristics and belimumab administrations

Characteristics	Patients (N = 105)
Female, n (%)	96 (91.4)
Age (years), n (%)
19–65	104 (99.0)
65+	1 (1.0)
Height, na	99
cm, mean (SD)	162.1 (7.0)
Weight, n (%)a	103 (98.1)
kg, mean (SD)	57.9 (9.5)
Body mass index (BMI), n (%)a	99 (94.3)
Mean (SD)	22.0 (3.1)
Diagnosis and disease duration
SLE, n (%)	105 (100.0)
Disease duration (years), mean (SD)	9.9 (8.1)
Lupus nephritis, n (%)	48 (45.7)
Disease duration (years), mean (SD)	8.2 (7.3)
Pregnancy status ‘no,’ n (%)b	96 (100.0)
Lactation status ‘no,’ n (%)b	96 (100.0)
Menopausal status ‘yes,’ n (%)b	10 (10.4)
Menopausal duration, n (%)c	9 (9.0)
Years, mean (SD)	6.7 (11.1)
Pregnancy history ‘yes,’ n (%)d	25 (26.6)
Times, mean (SD)	2.4 (1.0)
Delivery history ‘yes,’ n (%)d	22 (40.0)
Times, mean (SD)	1.9 (0.7)
Smoking history, n (%)
Smoker	8 (7.6)
Amount of smoking (pack-year), mean (SD)	7.3 (4.6)
Quitter	5 (4.8)
Amount of smoking (pack-year), mean (SD)	5.0 (4.4)
Cessation period (years), mean (SD)	3.8 (3.9)
Non-smoker	92 (87.6)
Lupus treatment-related prior medication history ‘yes,’ n (%)	36 (34.3)
Concomitant medication history ‘yes,’ n (%)e	105 (100.0)
Glucocorticoid	99 (94.3)
Antimalarial	87 (82.9)
Immunosuppressant	85 (81.0)
Glucocorticoid + antimalarial	82 (78.1)
Glucocorticoid + immunosuppressant	80 (76.2)
Glucocorticoid + antimalarial + immunosuppressant	64 (61.0)
Other medication	102 (97.1)
Mild hepatic impairment (including hepatitis), n (%)	5 (4.8)
Kidney impairment, n (%)	4 (3.8)
Mild	2 (50.0)
Moderate	2 (50.0)
Mild allergy, n (%)	2 (1.9)
Other medical history ‘yes,’ n (%)	95 (90.5)
Mild	89 (93.7)
Moderate	44 (46.3)
Severe	1 (1.1)
IV belimumab administration
Total administration periodf, (days), mean (SD)	346.2 (75.0)
Total number of doses, mean (SD)	13.1 (2.0)
Total dosage (mg)g, mean (SD)	7344.4 (1728.9)
Dosage (mg/1 day), mg, mean (SD)	562.3 (102.2)
Dose change/discontinuation status “yes”h, n (%)	39 (37.1)

BMI body mass index, IV intravenous, SD standard deviation, SLE systemic lupus erythematosus

^aAnalysis was limited to the patients with available data

^bThe percentage is calculated based on the total number of female patients

^cMenopausal duration = year of informed consent – year of menopause. Menopause data were analysed for nine of the 10 patients as the year of menopause wasn’t known for one patient

^dThe percentage is calculated on the basis of the patients with available data

^eAnalysis includes patients who continued to take lupus treatment-related concomitant drugs from the time prior to administration of belimumab

^fTotal administration period = (date of last dose of IV belimumab – date of first dose of the IV belimumab + 1)

^gTotal dosage = Σ Dosage (mg)/1 day

^hReasons for change of dosage or discontinuation were: weight change in 32 patients, investigator’s opinion in five patients, both weight change and investigator’s opinion in one patient and increase in dosage due to arthritis in one patient

The mean (SD) administration period of belimumab for the whole cohort (N = 105) was 346.2 (75.0) days, with a mean (SD) of 13.1 (2.0) doses and mean (SD) dosage of 562.3 (102.2) mg/1 day. Dose change or discontinuation was noted for 37.1% (n = 39/105) of patients; the reasons included weight change (n = 32/105), investigator’s opinion (n = 5/105), both weight change and investigator’s opinion (n = 1/105) and arthritis (dose increase; n = 1/105). All patients received lupus-related concomitant medications prior to the commencement of the study, with the most common being glucocorticoids (94.3%, n = 99/105), antimalarials (82.9%, n = 87/105) and immunosuppressants (81.0%, n = 85/105).

Adverse Events and Adverse Drug Reactions

All events, including AEs, SAEs, ADRs, serious ADRs and associated unexpected events, are summarised in Table 2.

Table 2.

Summary of AEs and ADRs

Classification	Patients, n (%) (N = 105)	Events, n
AEs	70 (66.7)	181
Unexpected AEs	63 (60.0)	144
SAEsa	19 (18.1)	26
Unexpected SAEs	17 (16.2)	22
ADRsb	4 (3.8)	5
Unexpected ADRs	3 (2.9)	3
Serious ADRs	1 (1.0)	1
Unexpected serious ADRs	1 (1.0)	1

ADR adverse drug reaction, AE adverse event, SAE serious adverse event

^aClassified as SAEs if the event caused death, was life-threatening, required hospitalisation or prolonged hospitalisation, caused persistent or serious disabilities or functional decline, caused congenital deformities or abnormalities or was an otherwise medically important situation

^bClassified as ADRs if the causal relationship with the intravenous belimumab is certain, likely, possible, conditional/unclassified or unassessable/unclassifiable

Adverse Events and Serious Adverse Events

Overall, 66.7% (n = 70/105) of patients experienced 181 AEs; of those, 26 were SAEs (experienced by 18.1% of patients [n = 19/105]), and all led to hospitalisation or prolonged hospitalisation (Table 3). The majority of AEs were mild (132 events, 61.0% of patients [n = 64/105]), and the rest were moderate (46 events, 22.9% [n = 24/105]) or severe (3 events, 1.0% [n = 1/105]). A causal relationship between AEs and belimumab was unlikely for 97.2% of reported AEs (176 of 181 events), with 85 events relating to underlying disease in 40.0% of patients (n = 42/105). Overall, 61.9% of patients (n = 65/105) recovered from 131 AEs (Table 3). Deaths or life-threatening SAEs were not reported.

Table 3.

Occurrence of AEs and severity

	Patientsa, n (%) (N = 105)	Events, n
AEs	70 (66.7)	181
Severity
Mild	64 (61.0)	132
Moderate	24 (22.9)	46
Severe	1 (1.0)	3
Causal relationship with belimumab
Possible	2 (1.9)	3
Conditional/unclassified	2 (1.9)	2
Unlikely	70 (66.7)	176
Causal relationship to factors other than belimumab
Underlying diseases	42 (40.0)	85
Intercurrent diseases	2 (1.9)	2
Concomitant medication	11 (10.5)	13
Trauma	1 (1.0)	1
Otherb	48 (45.7)	75
Course
Recovered	65 (61.9)	131
Recovering	24 (22.9)	41
Not recovered	7 (6.7)	9
Actions taken with belimumab
No action taken	68 (64.8)	172
Interruption	9 (8.6)	9
Re-administration status (if interrupted)
Yesc	8 (88.9)	8
Occurrence of the AE after re-administration
Nod	8 (100.0)	8
Noc	1 (11.1)	1
Remedial treatment status
Yes	50 (47.6)	98
No	48 (45.7)	83
SAEs	19 (18.1)	26
Classification of SAEs
Hospitalisation or prolonged hospitalisation	19 (18.1)	26
Life-threatening	0	0
Death	0	0
Severity
Mild	8 (7.6)	8
Moderate	11 (10.5)	15
Severe	1 (1.0)	3
Causal relationship with belimumab
Conditional/unclassified	1 (1.0)	1
Unlikely	19 (18.1)	25
Causal relationship to factors other than belimumab
Underlying diseases	12 (11.4)	16
Concomitant medication	3 (2.9)	3
Othere	6 (5.7)	6
Course
Recovered	18 (17.1)	25
Recovering	1 (1.0)	1
Actions taken with belimumab
No action taken	16 (15.2)	22
Interruption	4 (3.8)	4
Re-administration status (if interrupted)
Yesc	3 (75.0)	3
Occurrence of the AE after re-administration
Nod	3 (100.0)	3
Noc	1 (25.0)	1
Remedial treatment status
Yes	14 (13.3)	19
No	7 (6.7)	7

AE adverse event, COVID-19 coronavirus disease of 2019, SAE serious adverse event

^aPatients are counted multiple times if the same patient reports two or more AEs

^bIf ‘other’ is applicable for the causal relationship to factors other than belimumab, the detailed reasons are as follows: COVID-19 pandemic: n = 22 (22 events), seasonal factor: n = 5 (seven events), lifestyle: n = 8 (11 events), and other: n = 19 (35 events)

^cPercentage is calculated based on patients falling under ‘temporary suspension of administration’ in regard to the actions taken with belimumab (n = 9)

^dPercentage is calculated on the basis of patients who underwent re-administration of belimumab

^eIf ‘other’ is applicable for the causal relationship to factors other than belimumab, the detailed reasons are as follows: COVID-19 pandemic: n = 3 (three events) and other: n = 3 (three events)

No action was taken with IV belimumab in relation to 172 AEs experienced by 64.8% of patients (n = 68/105) and in relation to 22 SAEs experienced by 15.2% of patients (n = 16/105). Remedial treatment was reported for 98 AEs experienced by 47.6% of patients (n = 50/105) and 19 SAEs experienced by 13.3% of patients (n = 14/105). No AEs occurred in all of the 88.9% of patients (n = 8/9) with re-administration of belimumab after temporary discontinuation (Table 3).

The most common AEs by SOC were infections and infestations (48 events, 40% of patients [n = 42/105]), and by preferred term, the most common infection was coronavirus disease of 2019 (COVID-19; 22 events, 21.0% [n = 22/105]) (Supplementary Table 3). The most common SAEs by SOC were infections and infestations (nine events, 8.6% [n = 9/105]), followed by gastrointestinal disorders (eight events, 5.7% [n = 6/105]).

Unexpected Adverse Events and Serious Adverse Events

Overall, 144 unexpected AEs were experienced by 60.0% of patients (n = 63/105), of which 22 were unexpected SAEs among 16.2% of patients (n = 17/105); the severity of most unexpected AEs was mild (101 events, 52.4% of patients [n = 55/105]), with 40 moderate events (21.0%, n = 22/105) and 3 severe events (1.0%, n = 1/105). There were 68 unexpected AEs linked to underlying disease among 36.2% of patients (n = 38/105), and 56.2% of patients (n = 59/105) recovered from 99 unexpected AEs. For 137 unexpected AEs experienced by 59.0% of patients (n = 62/105), no action was taken with IV belimumab, with remedial treatment reported for 78 unexpected AEs (41.9%, n = 44/105) (Supplementary Table 4).

The most common unexpected AEs by SOC were infections and infestations (36 events, 30.5% of patients [n = 32/105]), followed by musculoskeletal and connective tissue disorders (18 events, 11.4% [n = 12/105]); the most common unexpected SAEs by SOC were also infections and infestations (eight events, 7.6% [n = 8/105]), followed by gastrointestinal disorders (seven events, 4.8% [n = 5/105]) (Supplementary Table 5).

Adverse Drug Reactions and Serious Adverse Drug Reactions

In total, five ADRs (two mild and three moderate events) were experienced by 3.8% of patients (n = 4/105); among those, one ADR was serious (by preferred term: mouth ulceration) and led to hospitalisation or prolonged hospitalisation (Table 4). Among the ADRs by preferred term, there was one event each for mouth ulceration, pyrexia, herpes zoster, skin papilloma and rash (each n = 1, 1.0%). Deaths or life-threatening ADRs were not reported.

Table 4.

Occurrence of ADRs and severity

	Patientsa, n (%) (N = 105)	Events, n
ADRs	4 (3.8)	5
Severity
Mild	1 (1.0)	2
Moderate	3 (2.9)	3
Causal relationship with belimumab
Possible	2 (1.9)	3
Conditional/unclassified	2 (1.9)	2
Course
Recovered	3 (2.9)	4
Recovering	1 (1.0)	1
Actions taken with belimumab
No action taken	4 (3.8)	5
Remedial treatment status
Yes	2 (1.9)	2
No	2 (1.9)	3
Serious ADRs	1 (1.0)	1
Serious ADR classification
Hospitalisation or prolonged hospitalisation	1 (1.0)	1
Life-threatening	0	0
Death	0	0
Severity
Moderate	1 (1.0)	1
Causal relationship with belimumab
Conditional/unclassified	1 (1.0)	1
Course
Recovered	1 (1.0)	1
Actions taken with belimumab
No action taken	1 (1.0)	1
Remedial treatment status
Yes	1 (1.0)	1

ADR adverse drug reaction

^aPatients are counted multiple times if the same patient reports two or more ADRs

Unexpected Adverse Drug Reactions and Unexpected Serious Adverse Drug Reactions

Three unexpected ADRs were experienced by 2.9% of patients (n = 3/105) (Supplementary Table 6), of which one event was an unexpected serious ADR (1.0%, n = 1/105), and all were of moderate severity. The causal relationship with belimumab was ‘conditional’ in 1.9% of patients (two events, n = 2/105) and ‘possible’ in 1.0% of patients (one event, n = 1/105); no action was taken with IV belimumab in all unexpected ADR events.

Adverse Events of Special Interest

Five psychiatric AESI events, including depression and suicidality, were experienced by 2.9% of patients (n = 3/105), but none were SAEs, ADRs or serious ADRs. At baseline, 13.3% of patients (n = 14/105) had a medical history of psychiatric disorders. A total of 40.0% (n = 42/105) experienced 48 AEs corresponding to AESI infections; of those, 8.6% of patients (n = 9/105) experienced SAEs, 1.0% (n = 1/105) experienced ADRs and none experienced serious ADRs. Progressive multifocal leukoencephalopathy AESI was not reported in this study.

Adverse Events of Patients with < 24 Weeks of Belimumab

Of the 21 patients who received belimumab for < 24 weeks, 76.2% (n = 16/21) experienced 52 AEs (Supplementary Table 7). Of those, 15 SAEs occurred in 42.9% of patients (n = 9/21); 37 unexpected AEs occurred in 71.4% of patients (n = 15/21); 10 unexpected SAEs occurred in 38.1% of patients (n = 8/21); 9 ADRs occurred in 28.6% of patients (n = 6/21); 4 serious ADRs occurred in 14.3% of patients (n = 3/21), 4 unexpected ADRs occurred in 14.3% of patients (n = 3/21), and 1 unexpected serious ADR occurred in 4.8% of patients (n = 1/21).

Factors Influencing Occurrence of AEs

Logistic regression analysis was conducted to assess factors that may influence the occurrence of AEs (Supplementary Table 8). The likelihood of AE occurrence was significantly higher in patients with ‘other medical history’ (odds ratio [95% Wald CI]: 5.24 [1.09, 25.06]; p = 0.038). None of the other factors significantly increased the likelihood of AEs.

Change in SELENA-SLEDAI Score

The mean (SD) SELENA-SLEDAI score changed significantly from 14.3 (3.9) at baseline (N = 105) to 7.0 (3.0) at Week 24 (N = 104; mean [SD] score change: − 7.3 [3.6], p < 0.0001), and to 6.1 (3.1) at week 48 (N = 72; mean [SD] score change: − 8.7 [3.9], p < 0.0001), among patients with at least 24 weeks of belimumab treatment (Table 5). The improvement rate based on ≥ 4-point reduction in SELENA-SLEDAI score was observed in 97.1% of patients (n = 101/104) at week 24, and 95.8% of patients (n = 69/72) at week 48. A similar change in SELENA-SLEDAI mean (SD) score, mean (SD) score change and improvement rate was observed in patients who received belimumab for 48 weeks and beyond.

Table 5.

SELENA-SLEDAI score

	Belimumab treatment duration group
	At least 24 weeks (N = 105)	≥ 48 weeks (N = 87)
SELENA-SLEDAI (total score)
Baselinea, n	105	87
SELENA-SLEDAI (total score), mean (SD)	14.3 (3.9)	14.5 (4.0)
Week 24, n	104	86
SELENA-SLEDAI (total score), mean (SD)	7.0 (3.0)	7.0 (3.1)
Change from baseline to week 24, n	104	86
SELENA-SLEDAI, mean (SD)	− 7.3 (3.6)	− 7.5 (3.6)
p-valueb	< 0.0001	< 0.0001
Week 48, n	72	71
SELENA-SLEDAI (total score), mean (SD)	6.1 (3.1)	6.1 (3.1)
Change from baseline to week 48, n	72	71
SELENA-SLEDAI, mean (SD)	− 8.7 (3.9)	− 8.7 (3.9)
p-valueb	< 0.0001	< 0.0001
SELENA-SLEDAI ≥-4 point reduction, n (%)
Improvement rate at week 24c	101 (97.1)	83 (96.5)
Improvement rate at week 48c	69 (95.8)	68 (95.8)

^aThe baseline used for the analysis was the result obtained at week 0 (date of informed consent)

^bThe statistical significance of the changes between week 24 or 48 and baseline was tested between patients with available data using Wilcoxon signed-rank test

^cThe improvement rate of SELENA-SLEDAI score of ≥ 4 by time point was calculated on the basis of patients with available data whose SELENA-SLEDAI values were collected at week 24 (N = 104) and week 48 (N = 72)

SD standard deviation, SELENA-SLEDAI Safety of Estrogens in Lupus Erythematosus National Assessment-SLE Disease Activity Index, SLE systemic lupus erythematosus

Laboratory Testing for Biomarkers

At baseline, levels of anti-dsDNA autoantibodies were available for n = 76/105 patients (Fig. 1A). At week 24 and 48 following belimumab initiation, the mean (SD) levels of anti-dsDNA autoantibody decreased by 136.9 (320.9) IU/mL in 60 patients and 167.3 (407.8) IU/mL in 38 patients, respectively, and were statistically significant (p < 0.0001) (Fig. 1A).

Fig. 1.

Laboratory biomarkers and glucocorticoid use monitored in patients included for safety evaluation. Anti-dsDNA anti-double-stranded DNA, C3/4 complement component 3 or 4, SD standard deviation, uPCR urine protein-creatinine ratio. Graphs were generated using Microsoft Excel (version 16.0). The statistical significance of the changes between week 24 or 48 and baseline was tested between patients with available data, using Wilcoxon signed-rank test. ^aThe baseline used for the analysis of glucocorticoid dosage was the result obtained at week 2. *p = 0.0002. **p < 0.0001

There were significant improvements in C3 and C4 levels, and reductions in uPCR, at week 24 and 48 following belimumab initiation (Fig. 1B–D).

Change in Glucocorticoid Use

The mean (SD) glucocorticoid dosage (mg/dose) significantly reduced from 9.2 (5.5) at baseline to 7.9 (6.9) at week 24 following the initiation of belimumab treatment (mean [SD] change: − 1.5 [6.9] mg/dose, p < 0.0001) in 82 patients, and to 7.7 (8.8) at week 48 (mean [SD] change: − 1.6 [9.1] mg/dose, p < 0.0001) in 78 patients (Fig. 1E).

Discussion

This observational study presents new real-world safety and effectiveness data for belimumab when used as part of standard therapy in Korean patients with SLE. Among patients eligible for the ≥ 48-week safety evaluation (i.e., those who completed ≥ 24 weeks of belimumab treatment), AEs were reported in approximately two-thirds of patients; however, most events were mild-to-moderate in severity and were considered unlikely to be related to belimumab. ADRs were reported in 3.8% of patients. Psychiatric events were reported in 2.9%, suggesting that belimumab was unlikely to contribute to psychiatric events. Significant improvements in SELENA-SLEDAI scores and laboratory biomarkers were observed, alongside significant reductions in glucocorticoid dosage, between baseline and weeks 24 and 48. Overall, these findings support the favourable real-world safety profile and clinical effectiveness of belimumab in Korean patients with SLE.

In this study, 181 AEs were experienced by 66.7% of patients, and 26 SAEs by 18.1% of patients, but most of those events were unlikely to be related to belimumab use. The proportion of patients with AEs in this study was lower than those observed in phase 3 clinical trials of belimumab in SLE (74.9–92.7% across BLISS-52 [BEL110752; NCT00424476], BLISS-76 [BEL110751; NCT00410384], North East Asia [BEL113750; NCT01345253] and North East Asia-Korean subgroup studies) [12, 14, 18, 19] and aligns with findings from observational studies [20–22]. Of note, the patient population in this study had a mean SELENA-SLEDAI score at baseline approximately 4–5 points higher than in clinical trials, with disease duration approximately 2.5–5 years longer [18, 19]. Additionally, unlike this study, the clinical trials included ethnically diverse populations, which may exhibit distinct disease characteristics, and generally excluded patients with lupus nephritis [3, 12, 14]. Future analysis should assess safety in patients with lupus nephritis in this study to enable comparison with the Asian subgroup findings from belimumab clinical trials in lupus nephritis (BLISS-LN trial; BEL114054; NCT01639339) [23].

Psychiatric events were a concern outlined in the risk management plan owing to numerical variations in previous clinical trials. The phase 4 clinical trial BASE (BEL115467; NCT01705977) reported serious depression, suicide or self-injury events in 0.9% (n = 18/2002) of patients receiving belimumab and 0.3% (n = 6/2001) receiving placebo [24]. These events were considered attributable to past medical history, as approximately 400 belimumab-treated and 370 placebo-treated patients reported current or past psychiatric disorders [24]. In this study, 14/105 patients reported current or past psychiatric disorders. Analysis of AESI suggested that belimumab was unlikely to contribute to psychiatric events (n = 3/105; including depression and suicidality) as none were classified as ADRs and all affected patients were recovering or had recovered.

Infections and infestations were the most common AEs in this study (40% of patients [42/105]), with COVID-19 as the most frequently reported infection (22 AEs and 3 SAEs). As patients with SLE may be at increased risk of severe COVID-19, particularly when receiving immunosuppressive therapies [25], these events may reflect background infection risk rather than a treatment-related safety signal. Current epidemiological data remain limited [25], and evidence from a retrospective cohort study (N = 123) suggests belimumab does not increase COVID-19 susceptibility, although larger clinical or real-world studies are needed to confirm this [26].

Among the 21 patients who received belimumab for < 24 weeks, 52 AEs occurred in 76.2% of patients, including 9 ADRs in 28.6% of patients for which a causal relationship with belimumab could not be ruled out. The percentage of patients with AEs, ADRs, SAEs, serious ADRs and unexpected events was higher than in patients who received belimumab for ≥ 24 weeks; however, this subgroup was substantially smaller (N = 21 versus N = 105, respectively). Although the proportion of patients reporting AEs in this subgroup was high, only two patients discontinued treatment due to AEs. Most discontinuations were owing to lack of perceived efficacy (n = 11) or investigator decision (n = 5).

Multiple logistic regression analysis of factors influencing the occurrence of AEs revealed that patients with ‘other medical history’ had 5.24 times higher odds of developing AEs. However, as the data were gathered from routine clinical practice, caution should be taken when interpreting those results due to potential confounding factors.

After 24 and 48 weeks of belimumab administration, significant improvements in SELENA-SLEDAI score, laboratory biomarkers, and glucocorticoid dosage were observed, supporting belimumab’s effectiveness in reducing disease activity. These findings align with previous observational [15, 20, 21, 27–31] and clinical studies [12, 14, 18, 19], although with some subtle differences. In this study, 95% of patients achieved a ≥ 4 points score improvement in SELENA-SLEDAI score, with mean scores decreasing more than two-fold over ≥ 48 weeks, compared with 58.3% in BLISS-52 and 56.1% in a South Korean patient subgroup in the North East Asia trial [14, 18]. These differences may reflect baseline disparities between studies, as clinical trial participants had lower baseline mean SELENA-SLEDAI scores (approximately 10; [12, 14, 18]), which may have limited the likelihood of achieving ≥ 4-point reductions. Additionally, SELENA-SLEDAI assessments in clinical trials may have followed stricter criteria than those used in routine clinical practice.

In South Korea, belimumab is indicated alongside standard therapy for patients with SLE, including paediatric patients aged ≥ 5 years, and for adults with lupus nephritis [11]. While clinical trials have previously demonstrated favourable safety and efficacy profiles in Korean patients [12–14], real-world findings from this study provide additional evidence supporting the safety and clinical effectiveness of belimumab in routine clinical practice among Korean patients with SLE. By addressing the previously limited real-world data in this population, these results may help inform treatment decisions and support clinician confidence in the use of belimumab in everyday clinical practice. From a pharmacovigilance perspective, these findings provide postmarketing safety data in a population underrepresented in global trials and additional context on AESI, including psychiatric events, supporting ongoing benefit–risk evaluation of belimumab in routine clinical practice.

This study has several strengths, including its prospective, multicentre, real-world design and inclusion of a clinically representative population, including paediatric patients and those with lupus nephritis, who are often underrepresented in clinical trials. The structured evaluation of psychiatric events, highlighted in belimumab risk management plans, together with simultaneous assessment of safety and effectiveness over ≥ 48 weeks, provides clinically relevant evidence on the medium-term benefit–risk profile of belimumab in routine practice.

A limitation inherent to real-world studies is data acquisition across multiple investigators and various sites as part of routine clinical practice. This may have resulted in a certain degree of variation, which should be taken into consideration when interpreting the data. In addition, the lack of a comparator cohort may limit the conclusions about belimumab’s safety and effectiveness. Furthermore, the psychiatric events were evaluated by the investigator’s medical judgement. Therefore, these items may have been collected inconsistently from all participating patients and subjected to bias. The relatively high rate of dose change or discontinuation prior to week 48, mainly owing to weight change, reduced the evaluable population and may introduce uncertainty in longer-term effectiveness estimates. Lastly, patient adherence to self-administered treatments such as glucocorticoids could have also influenced the observed outcomes.

Conclusions

This real-world study demonstrated that the safety profile of belimumab in a real-world setting in South Korea is consistent with the safety profile of belimumab demonstrated to date, with no new safety concerns, as well as effectiveness in controlling disease activity over time.

Supplementary Information

Below is the link to the electronic supplementary material.

Declarations

Funding

This study was funded by GSK (Study 216984).

Conflicts of Interests

S-BY and E-BL are employed by GSK. J-EC and D-JB are employed by GSK and hold financial equities in GSK. S-HP, S-HK, EYL, Y-AL and S-CB report no conflicting interests.

Availability of Data and Material

This manuscript has no associated data. For requests for access to anonymized subject-level data, please contact the corresponding author.

Ethics Approval

Ethical approval was obtained from the Ministry of Food and Drug Safety for the study, in accordance with the risk management plan (RMP). The RMP for IV belimumab was in accordance with Article 7-2 of the regulation on approval and examination of biological products [32]. This study was conducted in accordance with regulatory requirements including institutional review board approval (Institutional Review Board of Hanyang University Hospital; 2021-04-023-002), Declaration of Helsinki 2008 and its later amendments, patient informed consent and investigator reporting requirements. Approval was obtained from the Institutional Review Board or Ethics Committee at each of the 18 participating study sites.

Consent to Participate

Consent was obtained prior to the start of the study from each participating patient.

Author Contributions

S-BY, J-EC, D-JB and E-BL contributed to the conception or design of the study. S-HP, S-HK, EYL, Y-AL and S-CB contributed to the acquisition of data. S-BY, J-EC and E-BL contributed to data analysis and interpretation. All authors critically reviewed and approved the final version of this manuscript.

Code Availability

Not applicable

Consent for Publication

Not applicable.