Abstract
Objective. To evaluate the use and efficacy of belimumab in academic practices. Belimumab is a human monoclonal antibody that inhibits soluble B lymphocyte stimulator and has been approved for the treatment of adults with systemic lupus erythematosus (SLE).
Methods. Invitations to participate and complete a 1-page questionnaire for each patient prescribed belimumab were sent to 16 physicians experienced in SLE phase III clinical trials. The outcome was defined as the physician’s impression of improvement in the initial manifestation(s) being treated without worsening in other organ systems.
Results. Of 195 patients treated with belimumab at 10 academic centers, 96% were taking background medications for SLE at initiation of belimumab, with 74% taking corticosteroids. The main indications for initiation of belimumab were arthritis, rash, and/or worsening serologic activity, with 30% of patients unable to taper corticosteroids. Of the 120 patients taking belimumab for at least 6 months, 51% responded clinically and 67% had ≥ 25% improvement in laboratory values. While numbers are limited, black patients showed improvement at 6 months. In a subset of 39 patients with childhood-onset SLE, 65% responded favorably at 6 months, and 35% discontinued corticosteroids.
Conclusion. Our data demonstrate favorable clinical and laboratory outcomes in patients with SLE at 6 months across all racial and ethnic groups, with similar improvement seen among patients with childhood-onset SLE.
Systemic lupus erythematosus (SLE) is a chronic, multisystem, autoimmune disease of unknown etiology with a heterogeneous range of clinical and serological manifestations. Adding to the complexity of SLE is the unpredictability of flares, or periods of increased inflammatory disease activity, as well as the accrual of organ damage1,2. “Standard of care” therapy for SLE includes antimalarials, corticosteroids, immunosuppressives, and cytotoxic agents. These treatments often result in undesired longterm adverse effects and organ damage that are unavoidable because of the essential role of these medications in SLE treatment.
Although the etiology of SLE remains largely unknown, an increased understanding of its immunopathogenesis, in particular B cell–specific mechanisms, has allowed for a focus on targeted immunotherapy. This led to the development and testing of belimumab for SLE and its subsequent approval in 2011 by the US Food and Drug Administration (FDA) for the treatment of adults with active SLE3. Belimumab is a fully human, recombinant inhibitor of soluble B lymphocyte stimulator and neutralizes a key cytokine that plays a major role in B cell differentiation, proliferation, and survival4,5,6. Two randomized, double-blind, placebo-controlled, multicenter phase III trials (BLISS-52 and BLISS-76) demonstrated improvement in disease activity in patients treated with belimumab and standard-of-care SLE therapy compared to those treated with placebo and standard therapy alone7,8.
Since the initial belimumab clinical trials, the post-approval clinical experience of patients receiving belimumab at 10 mg/kg/dose every 4 weeks along with standard SLE therapy as well as the practice patterns of academic physicians in clinical settings have received increased attention. Even so, the data are scarce on the use of belimumab in patients with childhood-onset SLE. Childhood-onset is diagnosed prior to the patient’s 18th birthday and accounts for up to 20% of all patients with SLE9. When compared to adult-onset SLE, childhood-onset SLE has increased disease activity, morbidity, and mortality, often requiring earlier and more aggressive treatment with immunosuppressive medications for disease control10,11,12,13,14,15,16.
Because of the multitude of challenging characteristics of successful trials, including identifying meaningful outcomes and appropriate patient populations, postmarketing observational studies are often useful in providing supplementary information to trials. The objective of this multicenter study was to analyze the experience with belimumab in academic clinical practices.
MATERIALS AND METHODS
Study design
This is a prospective, multicenter, observational study conducted in 10 large academic SLE clinical practices, 9 of which were in the United States. Approvals to conduct this study were obtained from the institutional review boards of the respective institutions. Sixteen adult rheumatologists experienced in phase III SLE clinical trials were invited to participate, 10 of whom accepted the invitation and completed repeat assessments of patients with SLE treated with belimumab as part of their standard of care. Assessments consisted of a single-page questionnaire per patient completed every 3 months. All investigators who agreed to participate were located in countries where belimumab has received regulatory agency approval (United States and Sweden).
Inclusion and exclusion criteria
All patients were required to have a diagnosis of SLE, have met at least 4 of 11 American College of Rheumatology (ACR) or Systemic Lupus International Collaborative Clinics classification criteria17,18, and have started treatment with belimumab after approval by regulatory agencies. Exclusion criteria included patients who had previously participated in the belimumab clinical trials, and those who met fewer than 4 ACR classification criteria. We also excluded patients who had severe renal or neuropsychiatric involvement.
Questionnaire
The single-page questionnaire included basic demographic information (e.g., age, sex, race/ethnicity), SLE data [including disease duration, year of diagnosis, serologies, clinical manifestation(s), and concomitant medications], and belimumab information (including start date, clinical response, laboratory response, side effects, and reasons for discontinuation, if appropriate). The questionnaires were completed every 3 months by the treating physician to record longterm clinical patterns of belimumab.
Outcomes
All investigators adhered to the guidelines for 50% improvement as defined by the Responder Index for Lupus Erythematosus and British Isles Lupus Assessment Group, including clinical improvement and at least intent to decrease therapy19,20 without worsening in any organ systems. This was further defined as the treating physician’s impression of a ≥ 50% improvement in the initial manifestation(s) being treated and the ability to taper existing steroid dose by at least 25% of the initial dose for those patients who were taking steroids at the initiation of belimumab. Anti-dsDNA positivity was defined as having an anti-dsDNA level above the normal range; similarly, low complement was defined as having a C3 and/or C4 level below the normal range. Laboratory response was defined as a ≥ 25% improvement in the levels of C3, C4, and/or a 25% decrease in anti-dsDNA antibody levels. Responders were those patients who achieved the outcome measures as defined above; nonresponders were those who did not. Moreover, to be classified as a responder, there could be no worsening in other organ systems.
Data analysis
Descriptive statistics were used to evaluate the baseline characteristics of enrolled patients using Graphpad Prism 6. Every 3 months, clinical and laboratory response data were evaluated using t tests and chi-squared tests where appropriate.
RESULTS
Demographics
A total of 195 patients with SLE treated with belimumab were enrolled, most of whom were female (Table 1), with 29% black and 11% Hispanic. The average disease duration at initiation of belimumab was 11.9 ± 8.1 years among the 151 patients with data available regarding the date of diagnosis. Four patients were antinuclear antibody–negative.
For all patients with SLE: baseline demographics, concomitant medications, and clinical manifestations leading to treatment with belimumab.
Concomitant medications
Of 195 patients, 7 with adult-onset SLE were not taking any background SLE medications at the time of belimumab initiation (Table 1), while 73% were taking antimalarials, and 74% were taking corticosteroids. Of those taking corticosteroids at initiation of belimumab, the mean daily prednisone equivalent dose was 12.2 mg/day (range 5 to 50 mg/day). The majority of patients (66%) were taking immunosuppressive medications, with 21% taking azathioprine and 34% mycophenolate mofetil. No patients were concurrently receiving cyclophosphamide or rituximab.
Clinical manifestations
The breakdown of clinical manifestations requiring treatment at the initiation of belimumab is detailed in Table 1. The main clinical manifestations driving treatment were musculoskeletal and mucocutaneous. Proteinuria was present in 11% of patients at the time of initiation of belimumab.
Belimumab was used off-label in 7 patients who were not taking any background SLE medications at the time of initiation, although 4 were taking daily oral corticosteroids. The clinical manifestations resulting in treatment were rash, arthritis, and serositis. One of the 7 patients began taking belimumab for constitutional symptoms.
Fifteen patients were taking only antimalarial treatment at the time of initiation of belimumab. The main clinical manifestations driving treatment were rash, arthritis, and serositis. Four patients had hematological abnormalities, and 2 had constitutional symptoms.
Response to belimumab
At 3 months of belimumab treatment, 52% of patients clinically responded with improvement in the clinical manifestations that led to initiation of belimumab (Figure 1). At 3 months, 61% of patients who started belimumab for arthritis, 43% of patients with rash, and 78% of patients with renal manifestations responded to therapy. At 6 months, data about 120 patients were available for analysis and showed 51% with clinical response. Responders included 46% of patients with arthritis, 52% of patients with rash, and 57% of patients with renal manifestations. The mean daily prednisone equivalent dose of corticosteroids decreased to 9.3 mg/day at 6 months from 12.2 mg/day at baseline [p = not significant (NS)].
Response to belimumab in all patients with SLE at 3 and 6 months after initiation of belimumab. Data were available on 158 patients at 3 months and 120 patients at 6 months of treatment. Clinical response was defined as the treating physician’s impression of ≥ 50% improvement in the initial manifestation(s) being treated and no worsening in other organ systems. Improvement in serology included at least 25% decrease in anti-dsDNA level from baseline. SLE: systemic lupus erythematosus.
At 3 months after initiation of belimumab, 68 of 103 patients (66%) had at least a 25% increase in C3 values, and 35 of 72 patients (51%) had a 25% decrease in anti-dsDNA values (Figure 1). This was maintained at 6 months, with 67% of patients demonstrating at least a 25% decrease in anti-dsDNA values. No significant differences were found in either clinical or laboratory response rates in patients on different background immunosuppressive medications (data not shown).
Elderly patients
Four patients were over 65 years of age at the time of initiation of belimumab therapy, and it was initiated off-label. The clinical manifestations driving treatment were arthritis in all patients and serositis in 2 patients. All patients were taking oral prednisone daily at the time of initiation of therapy, with a range of 3 to 11 mg/day. Only 1 patient was taking another immunosuppressive (mycophenolate mofetil), and no patients were taking hydroxychloroquine. Two patients had elevated levels of anti-dsDNA antibodies, but no patients had hypocomplementemia. At 6 months, only 1 patient responded to therapy with resolution of arthritis and serositis, and all patients continued to take the same baseline dose of prednisone.
Black patients
Data on blacks, although limited, show a similar pattern of improvement. Of 57 black patients, 4 were receiving no background medications at the start of belimumab. There were 17 patients taking corticosteroids. The major clinical manifestations for initiation of belimumab were arthritis, inability to taper steroids, and rash. When compared to the non-black patients in the cohort, black patients showed a higher clinical response rate to belimumab at 3 months (82% vs 45%, p = 0.0001). Data were available for 21 black patients taking belimumab for 6 months; 67% responded to belimumab.
Patients with lupus nephritis
Twelve patients in the cohort had a history of lupus nephritis and proteinuria > 1000 mg/24 h at the time of initiation of belimumab. Followup data were available on 6 of these patients; 3 patients had > 50% improvement in proteinuria after initiation of belimumab.
Childhood-onset SLE
A subset of 39 patients treated with belimumab, most of whom were female, were diagnosed with SLE prior to their 18th birthday (Table 2), with a mean disease duration of 12 ± 8 years at initiation of belimumab. Four patients were under 18 years of age at the time of initiation of belimumab therapy, and it was initiated off-label. All patients with childhood-onset SLE were taking background medications at the start of belimumab, including 82% who took corticosteroids with a mean daily prednisone equivalent dose of 17 mg, compared to 72% of patients with adult-onset SLE, who took corticosteroids with a mean daily dose of 11 mg (p < 0.01). The main clinical manifestations treated in childhood-onset SLE were musculoskeletal and mucocutaneous ones, along with the inability to taper corticosteroids. In the 4 patients under 18 years of age who were treated off-label, the clinical manifestations leading to the decision to treat were rash in 2 patients, thrombocytopenia in 1, and inability to taper steroids in all 4. Clinical improvement was detected as early as 3 months after initiation of therapy in 41% of patients with childhood-onset SLE. At 6 months of treatment with belimumab, 65% of patients with childhood-onset SLE had clinically responded compared to 45% of patients with adult-onset SLE (p = NS; Figure 2). In addition, 18% of patients with childhood-onset SLE who took belimumab had at least a 25% improvement in C3 levels, and 44% had at least a 25% decrease in anti-dsDNA values 3 months after starting belimumab (p = 0.0001), and this was maintained at 6 months.
Response to belimumab in patients with childhood-onset SLE at 3 and 6 months after initiation of belimumab. Data were available on 35 patients at 3 months and 26 patients at 6 months of treatment. Clinical response was defined as the treating physician’s impression of ≥ 50% improvement in the initial manifestation(s) being treated and no worsening in other organ systems. Improvement in serology included at least 25% decrease in anti-dsDNA level from baseline. SLE: systemic lupus erythematosus.
Childhood-onset vs adult-onset patients with SLE: baseline demographics, concomitant medications, and clinical manifestations leading to treatment with belimumab.
Moreover, 6 months after initiation of belimumab, the mean prednisone equivalent dose decreased from 17 mg/day in those with childhood-onset SLE to 11 mg/day. Steroids were discontinued in 35% of patients with childhood-onset SLE 6 months after belimumab initiation, as compared with 11% of patients with adult-onset SLE (p = 0.002).
Safety
Data on adverse events and discontinuation of belimumab are presented in Table 3. No deaths were reported and 32 patients (16%) discontinued belimumab after an average of about 9 months across all sites. The most common adverse effects and reasons for discontinuation were infection and lack of response to belimumab. The most serious infection was Group A streptococcal bacteremia, which occurred in 1 patient. That patient had an elective cervical lymph node biopsy performed 5 days prior to development of bacteremia. The lack of response to belimumab included patients who demonstrated no improvement and/or worsening clinical manifestations of disease, or developed new organ system involvement, as judged by the treating physicians.
Adverse events and reasons for discontinuation of belimumab.
Of note, 4 patients developed features of neuropsychiatric SLE (NPSLE) while taking belimumab (1 with stroke, 1 with psychosis, 1 with severe depression, and 1 with new-onset seizures), and 2 patients who had existing NPSLE experienced worsening of disease while taking belimumab (1 with worsening depression, the other without specifics reported). There were also 3 patients who developed or experienced worsening of lupus nephritis over a year after initiation of therapy.
A total of 9 black patients discontinued belimumab: 2 owing to exacerbation of renal disease, 2 for development of or worsening neuropsychiatric disease, 1 because of arthritis flare, and 1 owing to worsening myositis. Discontinuation reasons not related to SLE included 1 infusion reaction, 1 who lost her insurance, and 1 who self-discontinued.
DISCUSSION
Belimumab is the first FDA-approved drug for the treatment of SLE in several decades and marks a breakthrough in drug development in SLE21. Our study evaluates the efficacy and safety of belimumab in 10 large academic clinical practices, and is the first report, to our knowledge, of the efficacy and safety of belimumab in childhood-onset SLE.
Posthoc analyses of the phase II belimumab trial data resulted in the development of a novel SLE responder index (SRI), which served as a primary endpoint in the phase III program22; however, the stringency of this index precludes its practical use in routine clinical settings. Therefore, the definition of clinical response we used is a simplified version of the SRI; clinical response in our study was constructed around the treating physician’s subjective assessment of at least 50% improvement in clinical manifestations, in line with the Systemic Lupus Erythematosus Disease Activity Index SRI-5023,24. Our data suggest that clinical improvement in response to belimumab may be detected as early as 3 months after initiation of therapy, and 51% of patients show a response at 6 months. Improvements in complement and dsDNA antibody levels were seen as early as 3 months after initiation of belimumab therapy and continued through the 6-month followup.
Since the results of the phase III trials, postmarketing analyses have consistently demonstrated arthritis and mucocutaneous disease as common indications for the initiation of belimumab therapy25. Interestingly, < 5% of the cohort as a whole were being treated with methotrexate prior to initiation of belimumab. Our data in a subset of patients with childhood-onset SLE demonstrate similar trends; however, our data also show that over 80% of patients with childhood-onset SLE were given belimumab because of an inability to taper existing steroids for treatment of SLE. Patients with childhood-onset SLE may respond as quickly as 3 months after initiation of belimumab (41%) and a greater proportion responded to belimumab (65%) than did patients with adult-onset SLE (45%) at 6 months (p = NS). The ability to taper and often discontinue corticosteroids in patients with childhood-onset SLE while taking belimumab treatment may, in part, reflect a degree of noncompliance to oral immunosuppressive medications in these patients.
Despite the small sample sizes in our study, serologically positive patients with SLE responded favorably to treatment with belimumab plus standard-of-care therapy. This is consistent with the phase III belimumab trials7,8,26. Moreover, given the higher prevalence of SLE in black patients27,28, it is of interest to note that 67% of a subset of black patients responded to belimumab therapy at 6 months. Subset analyses in phase III trials yielded lower response rates of black patients receiving belimumab versus those receiving placebo. Our data demonstrate a better clinical response rate in blacks that is significantly higher than those of other racial or ethnic subsets. Thus, results from clinical trials currently under way would be of utmost importance in individually tailoring therapy for black patients.
Further studies in a larger population and for a longer time will also make it possible to examine longterm outcomes and delineate adverse events. Ginzler, et al29 recently noted that there was a reduced risk of SLE flare and continued improvement in clinical and laboratory markers 7 years after initiation of belimumab therapy. Studies in patients with lupus nephritis and/or NPSLE would also help define the role for belimumab and describe the target population for optimal response. Although patients with NPSLE have been excluded from trials to date, there are ongoing trials using belimumab for patients with lupus nephritis that will provide more definitive answers.
Although data are limited, belimumab appears to be well tolerated; 16% of patients experienced an adverse event. This is a much lower rate of occurrence than that reported by Wallace, et al30 in a posthoc analysis of the phase III trials. In the phase III trials, 6% of patients developed depression, and 0.1% had suicidal ideation while taking belimumab therapy. In our cohort, there were 6 patients who developed or had an exacerbation of existing NPSLE; belimumab was discontinued in all those patients. Wallace, et al30 also noted a small percentage of renal disease flare in patients taking belimumab therapy. In addition, Dooley, et al31 demonstrated in a posthoc analysis of the phase III trial patients that of the 267 patients with renal disease at the time of initiation of belimumab therapy, those taking baseline immunosuppressive medications or with serologic activity at baseline had greater renal improvement with belimumab than those with placebo. The effect on renal disease was mixed: Dooley, et al noted that both renal flares and reduction of renal symptomatology such as proteinuria and serologic activity appeared to occur more frequently in patients taking belimumab, but the data were not statistically significant when compared with placebo. Our data show that 3 patients developed or had an exacerbation of lupus nephritis; all were patients with childhood-onset SLE. However, in those patients, noncompliance with background standard-of-care SLE immunosuppressive medications may have also played a role in worsening lupus nephritis.
Moreover, given that we defined the lack of response to be no improvement and/or worsening disease in any involved organ system, and/or development of disease in another previously uninvolved organ system, it is possible that the inclusion of these patients in calculating our response rate may reflect a lower rate of response than could be expected. Other limitations include the short time of followup (6 months) in our study, and the lack of comparison to a control group. This may have led us to overstate the efficacy of belimumab in our population.
Our study further substantiates the safety and efficacy of belimumab in the treatment of SLE. The higher clinical response rate among black patients, with comparable safety, is reassuring and should help alleviate concerns regarding the use of belimumab in this racial group. Overall, belimumab appears to be well tolerated; however, caution is advised with its use in patients with SLE who have active NPSLE and/or renal disease until more data are available. Moreover, our data suggest a larger role for belimumab as a corticosteroid-sparing agent in childhood-onset SLE, given that 35% of patients were able to discontinue steroid treatment after the addition of belimumab to standard-of-care SLE treatments. The results of the clinical trials currently under way in pediatric patients with SLE (under 18 years of age) will be of interest in demonstrating the efficacy and safety of belimumab in this population.
Acknowledgment
We thank the patients for their participation in our study.
Footnotes
Supported by an investigator-initiated award to the Systemic Lupus Erythematosus International Collaborating Clinics from GlaxoSmithKline. Dr. Hui-Yuen is the recipient of a T32 Medical Genetics training grant from the US National Institutes of Health. Dr. Askanase is a consultant for GlaxoSmithKline (< $10,000 in the past 12 months), and Dr. Petri has been a consultant for GlaxoSmithKline.
- Accepted for publication July 28, 2015.
