Abstract
Objective To report real-world experience on the use of anifrolumab (ANI) in refractory systemic lupus erythematosus (SLE).
Methods The present study is a multicenter, retrospective study involving 9 Italian SLE referral centers participating in a compassionate use program for the use of ANI in adult patients with active SLE in whom all the available treatment choices failed, were not tolerated, or were contraindicated. At baseline and 1, 3, 6, 9, and 12 months of treatment, overall and organ-specific disease activity, flares, daily glucocorticoid (GC) dose, and adverse events were recorded.
Results A total of 26 patients were enrolled. At 4 weeks after starting ANI, a significant decrease in the Systemic Lupus Erythematosus Disease Activity Index 2000 (P = 0.01), Systemic Lupus Erythematosus–Disease Activity Score (P = 0.01), and physician global assessment (P = 0.001) was recorded, and the same trend was maintained over time. A significant reduction in Cutaneous Lupus Erythematosus Disease Area and Severity Index-activity (P < 0.001) and in tender (P = 0.03) and swollen (P = 0.02) joint counts was also recorded. At 3 months of follow-up, 33% of patients already achieved a remission state, whereas 46% were in Lupus Low Disease Activity State (LLDAS); at 6 months, 50% were in remission and 80% were in LLDAS. A significant reduction in the mean GC daily dose was observed, starting from week 4 (P = 0.04). A total of 4 disease flares according to the Safety of Estrogens in Lupus Erythematosus National Assessment–Systemic Lupus Erythematosus Disease Activity Index were recorded (3 mild-moderate and 1 severe). Overall, 4/20 patients with at least 24 weeks of follow-up (20%) were considered nonresponders.
Conclusion This study provides real-world experience on the use of ANI in patients with refractory SLE, confirming its rapid effectiveness and an overall acceptable safety profile.
In recent years, the type I interferon pathway has gained a central role in systemic lupus erythematosus (SLE) pathogenesis, and it has become an important treatment target.1
Anifrolumab (ANI), a fully human monoclonal antibody against the type I interferon receptor, has been recently approved in several countries for the treatment of moderate-to-severe SLE as an add-on treatment to standard therapy.
Data from phase II and III clinical trials demonstrated the efficacy of ANI against global disease activity measures, as well as in reaching a number of other clinically relevant endpoints, including oral glucocorticoid (GC) tapering, flare rates, cutaneous responses, and joint counts.2-4
At the time of publication, real-world data on ANI’s use are limited to small case series and case reports, mainly focused on cutaneous SLE manifestations. Overall, these real-world data confirmed the rapid efficacy on refractory cutaneous manifestations, with little attention to systemic disease involvement.5-10
In this study, we report a multicenter, real-world experience of the use of ANI in refractory SLE provided through a compassionate use program.
METHODS
This is a multicenter, retrospective study involving 9 Italian referral centers for SLE (Pisa, Padova, Milano, Roma, Cagliari, Napoli, Udine, Firenze, and Bari) participating in a compassionate use program for the use of ANI in adult patients with active SLE in whom all the available treatment choices failed, were not tolerated, or were contraindicated. Consecutive patients fulfilling available classification criteria for SLE and meeting criteria for compassionate use of ANI were enrolled in the study. Demographic data, clinical and treatment history, and Systemic Lupus International Collaborating Clinics Damage Index (SDI) were retrieved from clinical charts.
At baseline, 1 month, and then every 3 months of treatment, Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K), Systemic Lupus Erythematosus–Disease Activity Score (SLE-DAS), daily prednisone dose, tender joint count (TJC) and swollen joint count (SJC), Cutaneous Lupus Erythematosus Disease Area and Severity Index-activity (CLASI-A) and CLASI-damage, and physician global assessment (PGA) were recorded.
Remission state (according to the 2021 definition of remission in SLE [DORIS]) and low disease activity state (according to the Lupus Low Disease Activity State [LLDAS] definition) were assessed at 3, 6, and 12 months11,12 among patients with at least 6 months of follow-up. Nonresponders were defined as patients who did not achieve at least LLDAS at the last follow-up visit or withdrew the treatment because of flare or persistently active disease.
Adverse events (AEs) were recorded at each visit. From the sixth month of ANI treatment onward, the occurrence of disease flare was evaluated according to the Safety of Estrogens in Lupus Erythematosus National Assessment (SELENA)-SLEDAI Flare Index (SFI) every 3 months.
Ethical approval. The study was approved by the Comitato Etico Regionale per la Sperimentazione Clinica della Regione Toscana (approval number 24198), and Comitato Etico Area Vasta Nord Ovest. All patients gave consent for their medical information to be published.
Statistical analysis. Continuous variables were described as median (IQR) or as mean (SD), as appropriate. Categorical variables were reported as proportions. Cross-tabulated data were analyzed using paired samples Wilcoxon rank-sum test, whereas intergroup comparisons were performed using the Mann-Whitney U test. Statistical analysis was performed by using the SPSS Statistics (version 29.0) statistic software package (IBM Corp.). P values < 0.05 were considered statistically significant.
RESULTS
Baseline data. Twenty-six patients were enrolled in the analysis. Demographics and clinical characteristics are summarized in Table 1.
Characteristics of the cohort at baseline (N = 26).
Briefly, patients were predominantly female (n = 24, 92%) and White (n = 24, 92%), with a median age of 48 (range 25-69) years and a median disease duration of 13 (range 5-34) years at enrollment.
Ninety-six percent of patients had a history of articular involvement, 92% mucocutaneous manifestations, 62% hematological manifestations, 35% lupus nephritis (of which 7 cases were biopsy-proven and the remaining 2 were defined on the basis of persistent proteinuria in the absence of other reasonable causes), 27% serositis, and 12% neuropsychiatric involvement. Twenty-three percent had concomitant antiphospholipid syndrome and 8% had overlapping Sjögren disease. At baseline, 54% had an SDI > 0 (median 2 [range 1-6]).
All patients had been previously treated with immunosuppressants with a median number of previous immunosuppressive therapies of 4 (range 1-8); the median cumulative dose of GC (prednisone equivalent) was variable, ranging from 1.2 g to 90 g (median 13 g). Eighty percent of patients had failed belimumab (n = 21) and 19% rituximab (n = 5); all patients who had not responded to rituximab were also nonresponders to belimumab.
Clinical manifestations and ongoing therapies at the time of ANI initiation are summarized in Table 1. Briefly, all patients had active disease at enrollment and reasons for adding ANI to the background therapies were a persistently active disease in 17 patients (65%) or a disease flare in 9 patients (35%). A detailed description of the characteristics of the 2 patient groups at baseline is reported in Supplementary Table S1 (available with the online version of this article). The most frequent active manifestations at enrollment were mucocutaneous (73%), articular (46%), and hematological (27%). Twenty-one patients (81%) presented with active serology at baseline (anti-dsDNA positivity and/or hypocomplementemia).
Regarding concomitant treatment, the majority of patients were receiving conventional immunosuppressive drugs (n = 23, 89%), GCs (n = 24, 92%), and antimalarials (n = 21, 81%). The median daily dosage of oral GC at enrollment was 7.3 (range 0-25) mg and no GC pulses were used. Of note, 3 patients were only on antimalarials or GCs because of failure or intolerance to several previous immunosuppressive therapies (Table 1).
Efficacy evaluation. Patients were followed for a median of 36 (IQR 4-72) weeks. During follow-up, at 1 month after starting ANI, a significant decrease in both SLEDAI-2K and SLE-DAS global activity indices was recorded (baseline mean SLEDAI-2K 8.0 [SD 4.1] vs 6.2 [SD 3.4], P = 0.01 and baseline mean SLE-DAS 9.3 [SD 5.4] vs 6.3 [SD 4.2], P = 0.01). The same trend was maintained over time (Figure 1A and Table 2). In parallel, PGA also significantly decreased from a baseline mean of 1.3 (SD 0.5) to 1.0 (SD 0.5) after 1 month (P = 0.001).
(A) Global activity indices over time. * P = 0.01 between T0 and T4. § P = 0.002 between T0 and T36. (B) CLASI-A over time. * P < 0.001 between T0 and T4. § P = 0.01 between T0 and T36. (C) Joint counts over time. * P = 0.03 for Tj and P = 0.02 for Sj between T0 and T4. § P = 0.01 for Tj and P = 0.04 for Sj between T0 and T36. (D) Mean GC daily dose (prednisone equivalent). * P = 0.045 between T0 and T4. § P = NS between T0 and T36. CLASI: Cutaneous Lupus Erythematosus Disease Area and Severity Index; GC: glucocorticoid; NS: not significant; Sj: swollen joint count; SLEDAI-2K: Systemic Lupus Erythematosus Disease Activity Index 2000; SLE-DAS: Systemic Lupus Erythematosus–Disease Activity Score; T: week; Tj: tender joint count.
Disease activity indices and daily dosage of prednisone equivalent over time.
In terms of active skin and joint involvement, a significant reduction in the CLASI-A and in TJC and SJC was also recorded, as summarized in Figures 1B,C and Table 2. The improvement was maintained through the entire follow-up. Baseline hematological manifestations improved in 3/6 patients (50%) by week 12 and this improvement was maintained over the follow-up period.
Interestingly, the percentage of patients with active serology slightly decreased over time starting at week 24 of treatment (63% at week 24, 54% at week 36, and 57% at week 48) and a complete normalization of serology was observed in 43% of patients at the last follow-up.
At 3 months of follow-up, 8/24 patients (33%) had already achieved a remission state whereas 11/24 (46%) were in LLDAS; at 6 months, 10/20 (50%) were in remission and 16/20 (80%) were in LLDAS. Among the patients who completed 12 months of follow-up, 6/7 (86%) were in remission at this timepoint, with 1 patient not achieving remission because of a daily GC dose > 5 mg. A decrease in the mean daily dose of GC was observed, which was statistically significant at 1 month (7.6 [SD 5.0] mg/day vs 6.4 [SD 4.9] mg/day, P = 0.045); thereafter, the difference lost its significance after 36 weeks (Figure 1D and Table 2). Interestingly, at 12 weeks of follow-up, 63% of patients were receiving < 5 mg of prednisone per day.
A subgroup analysis showed no significant differences in treatment efficacy between patients with SDI = 0 and SDI > 0 at baseline. The detailed trend of the efficacy measures over time in the 2 patient groups is shown in Supplementary Table S2 (available with the online version of this article).
During the follow-up period, a total of 4 disease flares according to the SFI were recorded (3 mild-moderate and 1 severe). One flare occurred at week 4 (a new-onset pericarditis in a patient with no previous history of serositis), 2 flares at week 24 (1 with leukopenia and thrombocytopenia and 1 with new-onset proteinuria and a diagnosis of lupus nephritis), and 1 flare at week 36 (with arthritis and thrombocytopenia). Two of 4 flares, specifically pericarditis and lupus nephritis, required treatment discontinuation. In particular, the patient with new-onset proteinuria, who had no previous history of renal involvement, underwent a kidney biopsy that proved a class IV lupus nephritis; consequently, ANI treatment was discontinued in favor of induction therapy with cyclophosphamide.
Overall, 4/20 patients with at least 24 weeks of follow-up (20%) were defined nonresponders.
Safety. A total of 27 AEs were recorded during the entire study period; of these, 23 (85%) were infections, generally mild, including 3 cases of oligosymptomatic coronavirus disease 2019 (COVID-19), 1 COVID-19 case that required hospitalization, and 1 case of multimetameric varicella zoster reactivation in a patient who later also developed cellulitis of a lower limb. Recurrent severe infections were the reason for treatment discontinuation in this patient. In some other cases, a temporary treatment suspension was required.
All patients had previously been vaccinated for SARS-CoV-2 (median number of vaccine doses 3 [IQR 2-3]) and only 10/26 (39%) had received the herpes zoster vaccine. Of note, the patient who developed multimetameric herpes zoster was among the unvaccinated patients. Among the other AEs, 1 case of pulmonary thromboembolism affecting a 66-year-old male patient was registered. Details of AEs and timing are summarized in Table 3.
AEs.
DISCUSSION
In this study, we describe real-world experience using ANI in patients with active SLE enrolled in a compassionate use program for refractory cases.
As expected, according to the compassionate use requirements, patients included in this study displayed active disease despite previous treatments, including biologics. Overall, the study population was characterized by a long mean disease duration and significant organ damage at ANI initiation, with more than half of patients exhibiting an SDI > 0 at baseline, despite a young mean age.
In this analysis, we showed a significant improvement in all the efficacy measures, including global activity scores (SLEDAI-2K, SLE-DAS, PGA) and organ-specific measures (CLASI-A, joint counts). Interestingly, the first significant improvement was observed as early as the first month of therapy and was maintained over time in most of the cases despite concomitant tapering of GCs.
Indeed, after 6 months of treatment, 50% and 80% of patients were in remission and LLDAS, respectively. Some early signs of efficacy could be intercepted even before, as already at the 3-month follow-up, 33% of patients were in remission and 46% in LLDAS. Four flares were recorded during follow-up, one of which was severe and required treatment discontinuation. Moreover, overall, 20% of patients of this series could be considered nonresponders. These data are in line with what has been reported in clinical trials3,4 and also confirm the real-world rapid treatment response and ANI effectiveness on a range of clinical manifestations.
In a previous post hoc analysis of pooled data from the Treatment of Uncontrolled Lupus From the Interferon Pathway (TULIP-1 and TULIP-2) trials, from as early as week 8, the proportion of patients with a British Isles Lupus Assessment Group-based Composite Lupus Assessment (BICLA) and CLASI-A response was significantly greater with ANI than placebo, and a greater reduction in GC dosage compared with placebo by week 20 was also reported.13
Another interesting aspect of this current analysis relies on the fact that the introduction of ANI to reduce disease activity was associated with the reduction in oral GCs dosage that was already evident as early as the first month of treatment. Moreover, the rapid response allowed a further GC reduction over time; indeed, at 12 weeks of follow-up, 63% of patients were receiving < 5 mg of prednisone per day. This is a very important aspect that might suggest the use of the drug to rapidly control disease activity, thereby avoiding high doses of GCs.
The GC-sparing effects of ANI have been clearly demonstrated in clinical trials.14 The present data further reinforce this result from a real-world perspective and might encourage physicians to minimize the GC dosage in their patients with active disease when prescribing ANI.
To the best of our knowledge, this is the first report on the real-world use of ANI in a cohort of patients with SLE following a well-defined data collection approach. Moreover, although several case reports and case series have described the effectiveness of ANI on a wide range of muco-cutaneous manifestations, to date, less data are available on the real-world use of the drug in other disease manifestations. In our case series, hematological manifestations and arthritis were present at enrollment in 27% and 46% of patients, respectively. Thus, this patient sample is highly representative of the complexity and the variability of the clinical picture of the disease and very much illustrative of the type of patient for whom ANI is indicated.
Several AEs were recorded, with few being moderate or severe, and treatment discontinuation has been rare. Similarly to what has been observed in clinical trials, infections were the most frequent AEs. This is not surprising given that the patients included in this analysis had active SLE, with long disease duration and significant organ damage, and were on concomitant treatment with GC and conventional immunosuppressants, all of which are considered important risk factors for infections in SLE.15
In conclusion, this study provides real-world experience on the use of ANI in patients with refractory SLE, confirming its effectiveness and overall acceptable safety profile. Moreover, these data also provide some new and interesting insights to be further developed and tested in larger studies.
Footnotes
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication June 4, 2024.
- Copyright © 2024 by the Journal of Rheumatology
