Abstract
Novel therapies are needed for recurrent pericarditis, particularly when corticosteroid dependent and colchicine resistant. Based on limited data, interleukin-1 blockade with anakinra may be beneficial. The aim of this multicentre registry was to evaluate the broader effectiveness and safety of anakinra in a ‘real world’ population.
This registry enrolled consecutive patients with recurrent pericarditis who were corticosteroid dependent and colchicine resistant and treated with anakinra. The primary outcome was the pericarditis recurrence rate after treatment. Secondary outcomes included emergency department visits, hospitalisations, corticosteroid use and adverse events. Among 224 patients (46 ± 14 years old, 63% women, 75% idiopathic), the median duration of disease was 17 months (interquartile range 9–33). Most patients had elevated C-reactive protein (91%) and pericardial effusion (88%). After a median treatment of 6 months (3–12), pericarditis recurrences were reduced six-fold (2.33–0.39 per patient per year), emergency department admissions were reduced 11-fold (1.08–0.10 per patient per year), hospitalisations were reduced seven-fold (0.99–0.13 per patient per year). Corticosteroid use was decreased by anakinra (respectively from 80% to 27%; P < 0.001). No serious adverse events occurred; adverse events consisted mostly of transient skin reactions (38%) at the injection site. Adverse events led to discontinuation in 3%. A full-dose treatment duration of over 3 months followed by a tapering period of over 3 months were the therapeutic schemes associated with a lower risk of recurrence.
In patients with recurrent pericarditis, anakinra appears efficacious and safe in reducing recurrences, emergency department admissions and hospitalisations.
Introduction
Recurrent pericarditis (RP) affects up to 30% of patients after a first episode, and up to 50% of patients with recurrences if not treated with colchicine.1–5
Some patients may have an inadequate response to standard therapy with non-steroidal anti-inflammatory drugs (NSAIDs), colchicine and corticosteroids, typically with unsuccessful attempts to taper corticosteroids. These patients with corticosteroid-dependent and colchicine-resistant refractory RP may have a severely impaired quality of life due to an inability to control symptoms and severe side effects from chronic glucocorticoid therapy.5,6
Although the pathogenesis has not been fully elucidated, a putative hypothesis in recurrent idiopathic pericarditis is an amplified and self-sustained autoinflammatory and/or autoimmune response following exogenous or endogenous triggers.7–11 Interleukin-1 blockade with anakinra is of particular interest because it may interfere with this self-sustained pathway and dramatically reduce recurrences.11–14 However, the published literature on anakinra in refractory RP is limited to case reports,12,15–24 small case series25–28 and a single randomised controlled trial including 21 patients.29
Due to the single-centre nature of observational studies and the narrowly defined patient population in the small randomised controlled trial, the generalisability of these data is unknown. This international, all-comers, multicentre registry was consequently designed to address this deficit in external validity, and the specific aim of the current study was to evaluate the safety and efficacy of anakinra in a ‘real-world’ population of patients with refractory RP.
Methods
Study design
The International, all-comers, multicentre Registry of Anakinra for Pericarditis (IRAP) is a multicentre observational cohort study including 14 referral centres for pericardial diseases across six different countries (see Supplementary Appendix A). The present study complied with the Declaration of Helsinki and subsequent modifications. The study was approved by the ethics committee or institutional review board at each participating centre. All patients provided written informed consent, or an exemption for individual patient consent was provided by the local ethics committee or institutional review board. Consecutive adult patients (>18 years old), referred for refractory RP at each clinical site between 2014 and 2018, were enrolled at the time of a pericarditis recurrence if they were eligible to receive anakinra according to institutional and international guidelines.
Inclusion and exclusion criteria
All patients eligible for inclusion had corticosteroid-dependent and colchicine-resistant refractory RP, defined as a first episode of acute pericarditis followed by a minimum of two recurrences despite guideline-based traditional medical treatment (NSAIDs, colchicine and/or corticosteroids). Corticosteroid dependence was defined as the inability to taper glucocorticoids without recurrence. The first episode of pericarditis was diagnosed in the presence of at least two of the following criteria: typical pericarditic chest pain (sharp, pleuritic, improved by sitting up and leaning forward), pericardial friction rub, ST-segment elevation or PR-segment depression on electrocardiogram and new or worsening pericardial effusion. A recurrence was diagnosed when pericarditic chest pain reoccurred along with one or more of the following signs: fever, friction rubs, electrocardiographic changes, new or worsening pericardial effusion, or C-reactive protein (CRP) elevation.1–5
Exclusion criteria were specific contraindications to anakinra treatment including hypersensitivity to the active substance or to Escherichia coli-derived proteins, neutropenia (absolute neutrophil count <1.5 × 109/L), active tuberculosis and active malignancy.
Study protocol
All enrolled patients received anakinra 100 mg once daily by subcutaneous (sc) injection. Concomitant medical treatment for pericarditis (NSAIDs, colchicine, corticosteroids) was maintained, tapered, or discontinued according to clinical evaluation. Clinical, laboratory testing, electrocardiographic and echocardiographic assessments were performed in all patients at the time of enrollment, according to local practice and in accordance with established guidelines.5
Study end-points
The primary end-point was pericarditis recurrence after the initiation of anakinra. The secondary end-points included emergency department (ED) admissions, hospitalisations and corticosteroid use after initiation of anakinra. The occurrence of any adverse event (AE) and AE-related drug discontinuation were assessed as safety end-points.
Statistical analysis
Blinded and independent analysis of data was performed in the Coordinating Centre (Torino, Italy). This cohort study followed the recommendations of the STROBE statement.30 A minimum sample size of 24 patients was estimated to be sufficient to detect a difference in recurrence rate from 80% before anakinra to 10% after anakinra, based on previous evidence in the literature, with a power of 90% at a confidence level of 95%.10,29 Due to the observational real-world purpose of the registry, a greater number of patients were subsequently included during the enrollment phase. Continuous variables, presented as means and standard deviations or medians and interquartile range, were compared by non-parametric tests: Mann–Whitney’s test was used for independent data and Wilcoxon’s signed-rank test was usedfor paired data (pre-post evaluations). Categorical variables, presented as counts and percentages, were compared using the chi-square test with Yates’ correction or Fisher’s exact test as appropriate. The survival probability and the freedom from AEs were evaluated with Kaplan–Meier curves, compared by the Mantel–Cox test. A time varying covariate analysis was performed to evaluate recurrences according to treatment duration. Cox proportional hazard models were also used to establish independent associations with recurrence. All analyses were performed using SPSS version 18.0 (SPSS, Inc., Chicago, IL, USA) and a two-sided significance level less than 0.05 was considered statistically significant.
Results
Population characteristics
A total of 224 patients was included in the present registry. Enrollment ranged from one to 53 patients, based on centre volume and local practice (see Supplementary Appendix A). Each centre enrolled consecutive patients in order to reproduce real-life practice and minimise selection bias. Baseline population characteristics are reported in Table 1.
Participant characteristics at enrollment.
| Characteristics at enrollment | Population (n = 224) |
|---|---|
| Age, years | 46 ± 14 |
| Female gender | 140 (63%) |
| Pericardial disease duration, months | 17 (9–33) |
| Previous pericarditis recurrences | 4 (3–6) |
| Previous ED admissions | 2 (1–3) |
| Previous hospitalisations | 2 (1–3) |
| Aetiology | |
| Idiopathic | 167 (75%) |
| Post-cardiac injury syndrome | 28 (13%) |
| Autoimmune diseasea | 21 (9%) |
| Autoinflammatory disease | 5 (2%) |
| Radiation | 2 (0.7%) |
| Traumatic | 1 (0.3%) |
| CRP elevation | |
| Pericardial effusion | 203 (91%) |
| Mild | 196 (88%) |
| Moderate | 101 (45%) |
| Severe | 48 (21%) |
| Tamponade | 23 (10%) |
| Therapy | 24 (11%) |
| NSAIDs | 170 (76%) |
| Colchicine | 198 (88%) |
| Corticosteroids | 180 (80%) |
| Triple therapy (NSAIDs + colchicine + corticosteroids) | 141 (63%) |
| Follow-up observation after anakinra initiation, months | 13 (7–25) |
| Characteristics at enrollment | Population (n = 224) |
|---|---|
| Age, years | 46 ± 14 |
| Female gender | 140 (63%) |
| Pericardial disease duration, months | 17 (9–33) |
| Previous pericarditis recurrences | 4 (3–6) |
| Previous ED admissions | 2 (1–3) |
| Previous hospitalisations | 2 (1–3) |
| Aetiology | |
| Idiopathic | 167 (75%) |
| Post-cardiac injury syndrome | 28 (13%) |
| Autoimmune diseasea | 21 (9%) |
| Autoinflammatory disease | 5 (2%) |
| Radiation | 2 (0.7%) |
| Traumatic | 1 (0.3%) |
| CRP elevation | |
| Pericardial effusion | 203 (91%) |
| Mild | 196 (88%) |
| Moderate | 101 (45%) |
| Severe | 48 (21%) |
| Tamponade | 23 (10%) |
| Therapy | 24 (11%) |
| NSAIDs | 170 (76%) |
| Colchicine | 198 (88%) |
| Corticosteroids | 180 (80%) |
| Triple therapy (NSAIDs + colchicine + corticosteroids) | 141 (63%) |
| Follow-up observation after anakinra initiation, months | 13 (7–25) |
Values are presented as no. (%) or mean ± standard deviation or median (interquartile range).
Presumed in patients with a defined diagnosis of: systemic lupus erythematosus, rheumatoid arthritis, axial spondyloarthropathy, Behcet syndrome, Sjogren syndrome, sarcoidosis.
ED: emergency department; NSAIDs: non-steroidal anti-inflammatory drugs; CRP: C-reactive protein.
Participant characteristics at enrollment.
| Characteristics at enrollment | Population (n = 224) |
|---|---|
| Age, years | 46 ± 14 |
| Female gender | 140 (63%) |
| Pericardial disease duration, months | 17 (9–33) |
| Previous pericarditis recurrences | 4 (3–6) |
| Previous ED admissions | 2 (1–3) |
| Previous hospitalisations | 2 (1–3) |
| Aetiology | |
| Idiopathic | 167 (75%) |
| Post-cardiac injury syndrome | 28 (13%) |
| Autoimmune diseasea | 21 (9%) |
| Autoinflammatory disease | 5 (2%) |
| Radiation | 2 (0.7%) |
| Traumatic | 1 (0.3%) |
| CRP elevation | |
| Pericardial effusion | 203 (91%) |
| Mild | 196 (88%) |
| Moderate | 101 (45%) |
| Severe | 48 (21%) |
| Tamponade | 23 (10%) |
| Therapy | 24 (11%) |
| NSAIDs | 170 (76%) |
| Colchicine | 198 (88%) |
| Corticosteroids | 180 (80%) |
| Triple therapy (NSAIDs + colchicine + corticosteroids) | 141 (63%) |
| Follow-up observation after anakinra initiation, months | 13 (7–25) |
| Characteristics at enrollment | Population (n = 224) |
|---|---|
| Age, years | 46 ± 14 |
| Female gender | 140 (63%) |
| Pericardial disease duration, months | 17 (9–33) |
| Previous pericarditis recurrences | 4 (3–6) |
| Previous ED admissions | 2 (1–3) |
| Previous hospitalisations | 2 (1–3) |
| Aetiology | |
| Idiopathic | 167 (75%) |
| Post-cardiac injury syndrome | 28 (13%) |
| Autoimmune diseasea | 21 (9%) |
| Autoinflammatory disease | 5 (2%) |
| Radiation | 2 (0.7%) |
| Traumatic | 1 (0.3%) |
| CRP elevation | |
| Pericardial effusion | 203 (91%) |
| Mild | 196 (88%) |
| Moderate | 101 (45%) |
| Severe | 48 (21%) |
| Tamponade | 23 (10%) |
| Therapy | 24 (11%) |
| NSAIDs | 170 (76%) |
| Colchicine | 198 (88%) |
| Corticosteroids | 180 (80%) |
| Triple therapy (NSAIDs + colchicine + corticosteroids) | 141 (63%) |
| Follow-up observation after anakinra initiation, months | 13 (7–25) |
Values are presented as no. (%) or mean ± standard deviation or median (interquartile range).
Presumed in patients with a defined diagnosis of: systemic lupus erythematosus, rheumatoid arthritis, axial spondyloarthropathy, Behcet syndrome, Sjogren syndrome, sarcoidosis.
ED: emergency department; NSAIDs: non-steroidal anti-inflammatory drugs; CRP: C-reactive protein.
Patients had a mean age of 46 years (range 18–84) and 140 (63%) were women. At baseline, the median duration of disease was 17 months (interquartile range (IQR) 9–33) with a median of four prior recurrences (IQR 3–6), two prior ED admissions (IQR 1–3) and two prior hospitalisations (IQR 1–3). Before initiating anakinra, the recurrence rate was 2.33 flares per patient per year, a mean of one recurrence every 157 days (Figure 1). Likewise, the prior ED admission rate was 1.08 admissions per patient per year with 0.99 hospitalisations per patient per year. All patients were diagnosed with recurrent pericarditis at the time of enrollment. Most patients had idiopathic aetiology (167 patients, 75%), CRP elevation (203 patients, 91%) and echocardiographic evidence of pericardial effusion (196 patients, 88%). At the time of enrollment, medical treatment included NSAIDs in 170 patients (76%), colchicine in 198 patients (88%) and corticosteroids in 180 patients (80%). Forty-four patients (20%) previously on corticosteroids were not on treatment at the time of enrollment because of drug intolerance or drug tapering and discontinuation shortly before the index recurrence.
Recurrences, emergency department admissions, hospital admissions before and after anakinra (number of events/year per patient and in the population of the registry). Corticosteroid use before and after anakinra.
Pericarditis recurrences after anakinra
During a follow-up of 3889 patient-months, one or more pericarditis recurrences were observed in 78 patients, with a median flare-free time of 10 months (IQR 5–18). Multiple recurrences were observed in 26 patients.
After treatment with anakinra, a median of zero recurrences (IQR 0–1) with a mean of 0.6 ± 1 per patient occurred. The recurrence rate after starting anakinra was 0.39 flares per patient per year, implying a mean of one recurrence every 939 days. Overall, a 83% reduction in recurrence rate was observed (P < 0.001, rate ratio 0.17, 95% confidence interval (CI) 0.14–0.20), equivalent to 1.94 fewer flares per patient per year.
At 36 months after anakinra initiation, 72% patients experienced none or at most one recurrence: respectively, 43% patients were in stable remission, while 29% patients had a single recurrence over a period of 36 months. The remaining 28% patients had two or more recurrences during the follow-up (ranging from a minimum of two to a maximum of seven).
ED admissions and hospitalisations after anakinra
During follow-up, a median of zero ED admissions (IQR 0–0) with a mean of 0.1 ± 0.5 per patient and a median of zero hospitalisations (IQR 0–0) with a mean of 0.2 ± 0.5 per patient occurred, corresponding to 0.10 ED admissions per patient per year and 0.14 hospitalisations per patient per year, respectively (Figure 1). Compared to the period prior to anakinra treatment, there was a reduction of 91% for ED admissions (P < 0.001, rate ratio 0.09, 95% CI 0.06–0.13) and 86% for hospitalisations (P < 0.001, rate ratio 0.14, 95% CI 0.11–0.19).
During follow-up, after 10 ± 9 months (range 1.8–38), 20 patients (8.9%) were admitted and underwent pericardectomy with subsequent discontinuation of anakinra.
Corticosteroid dependence after anakinra
Although all patients had been corticosteroid dependent, at the time of enrollment 180 patients (80%) were still on active treatment with glucocorticoids. After treatment with anakinra, corticosteroids were successfully tapered in most patients without recurrence. In fact, only 61 patients (27%) remained on glucocorticoid therapy (Figure 1).
Similarly, NSAIDs were discontinued in most patients. Only 54 patients (24%) remained on NSAIDs. Conversely, the majority of patients continued on colchicine. During follow-up, 131 patients (58%) remained on colchicine.
Adverse events
AEs were observed in 99 patients (44%), as described in Table 2. The most frequent AE was a transient skin reaction at the injection site (associated with erythema, pruritus and/or pain), which occurred in 86 patients (38%), 13 ± 16 days after the beginning of treatment. Three patients (1.3%) required permanent drug discontinuation because of intolerable injection-site reactions. Arthralgias and myalgias were reported by 13 patients (6%) during the treatment, and one of these patients (0.4%) required drug withdrawal. Seven patients (3%) had a mild transient transaminase elevations. A transient neutropenia was observed in three patients (1%), not associated with clinical events, with subsequent resolution. Other AEs, with questionable association with anakinra treatment, were reported by single patients, including hypereosinophilia, mild fever, hot flushes and sweating, perforated diverticulitis and optic neuritis.
Adverse events.
| Adverse events | Population (n = 224) |
|---|---|
| Occurrence of one or more AE | 99 (44%) |
| Skin reaction at injection site | 86 (38%) |
| Arthralgias and myalgias | 13 (6%) |
| Transaminase elevation | 7 (3%) |
| Infections | 6 (3%) |
| Neutropenia (<1.5 × 109/L) | 3 (1%) |
| Hypereosinophilia | 1 (0.4%) |
| Mild evening fever | 1 (0.4%) |
| Hot flushes and sweating | 1 (0.4%) |
| Perforated diverticulitis | 1 (0.4%) |
| Optic Neuritis | 1 (0.4%) |
| AE requiring permanent anakinra discontinuation | 7 (3%) |
| Skin reaction | 3 |
| Infection (skin abscess) | 1 |
| Perforated diverticulitis | 1 |
| Arthralgias and myalgias | 1 |
| Multiple side effects (skin reaction, arthralgias and myalgias, transaminases elevation, neutropenia) | 1 |
| Adverse events | Population (n = 224) |
|---|---|
| Occurrence of one or more AE | 99 (44%) |
| Skin reaction at injection site | 86 (38%) |
| Arthralgias and myalgias | 13 (6%) |
| Transaminase elevation | 7 (3%) |
| Infections | 6 (3%) |
| Neutropenia (<1.5 × 109/L) | 3 (1%) |
| Hypereosinophilia | 1 (0.4%) |
| Mild evening fever | 1 (0.4%) |
| Hot flushes and sweating | 1 (0.4%) |
| Perforated diverticulitis | 1 (0.4%) |
| Optic Neuritis | 1 (0.4%) |
| AE requiring permanent anakinra discontinuation | 7 (3%) |
| Skin reaction | 3 |
| Infection (skin abscess) | 1 |
| Perforated diverticulitis | 1 |
| Arthralgias and myalgias | 1 |
| Multiple side effects (skin reaction, arthralgias and myalgias, transaminases elevation, neutropenia) | 1 |
Values are presented as no. (%).
AE: adverse event.
Adverse events.
| Adverse events | Population (n = 224) |
|---|---|
| Occurrence of one or more AE | 99 (44%) |
| Skin reaction at injection site | 86 (38%) |
| Arthralgias and myalgias | 13 (6%) |
| Transaminase elevation | 7 (3%) |
| Infections | 6 (3%) |
| Neutropenia (<1.5 × 109/L) | 3 (1%) |
| Hypereosinophilia | 1 (0.4%) |
| Mild evening fever | 1 (0.4%) |
| Hot flushes and sweating | 1 (0.4%) |
| Perforated diverticulitis | 1 (0.4%) |
| Optic Neuritis | 1 (0.4%) |
| AE requiring permanent anakinra discontinuation | 7 (3%) |
| Skin reaction | 3 |
| Infection (skin abscess) | 1 |
| Perforated diverticulitis | 1 |
| Arthralgias and myalgias | 1 |
| Multiple side effects (skin reaction, arthralgias and myalgias, transaminases elevation, neutropenia) | 1 |
| Adverse events | Population (n = 224) |
|---|---|
| Occurrence of one or more AE | 99 (44%) |
| Skin reaction at injection site | 86 (38%) |
| Arthralgias and myalgias | 13 (6%) |
| Transaminase elevation | 7 (3%) |
| Infections | 6 (3%) |
| Neutropenia (<1.5 × 109/L) | 3 (1%) |
| Hypereosinophilia | 1 (0.4%) |
| Mild evening fever | 1 (0.4%) |
| Hot flushes and sweating | 1 (0.4%) |
| Perforated diverticulitis | 1 (0.4%) |
| Optic Neuritis | 1 (0.4%) |
| AE requiring permanent anakinra discontinuation | 7 (3%) |
| Skin reaction | 3 |
| Infection (skin abscess) | 1 |
| Perforated diverticulitis | 1 |
| Arthralgias and myalgias | 1 |
| Multiple side effects (skin reaction, arthralgias and myalgias, transaminases elevation, neutropenia) | 1 |
Values are presented as no. (%).
AE: adverse event.
Infections were observed in six patients (3%) during follow-up: two respiratory infections (bronchopneumonia), as well as four skin and soft-tissue infections (skin abscess, Cryptococcus neoformans necrotising cellulitis, lymphocele infection and pelvic cyst infections). In three cases, anakinra was temporarily suspended, and one patient permanently discontinued therapy (patient with skin abscess). In all patients, infections successfully resolved with appropriate treatment. Overall, as mentioned, seven patients (3%) permanently discontinued anakinra due to an AE.
Treatment duration and tapering
Patients were treated with full-dose anakinra for a median duration of 6 months (IQR 3–12), followed by a tapering period with a median duration of 3 months (IQR 0–6). Patients who received a full-dose treatment for more than 3 months had a greater freedom from recurrences at 36 months (46% vs. 26%, P = 0.002; Figure 2). A time varying covariate analysis confirmed that a full dose treatment for more than 3 months was independently associated with a lower risk of recurrences (hazard ratio (HR) 0.96, 95% CI 0.94–0.99).
Recurrences according to treatment duration (36-month Kaplan–Meier survival analysis).
A further analysis was performed on the subgroup of 135 patients (60%) who permanently discontinued the active treatment with anakinra. At 18 months after discontinuation, 74% patients were still free from recurrences. Among patients who discontinued anakinra, 87 (64%) gradually suspended the active treatment, while the other 36% of patients stopped it abruptly. Among patients in whom anakinra was gradually withdrawn, those treated with a reduced dose regimen for at least 3 months before discontinuation had a greater freedom from recurrence at 18 months after drug interruption (89% vs. 50%, P = 0.002). A lower freedom from recurrence was observed in patients undergoing a more rapid drug withdrawal (3 months or less) or those in whom anakinra was abruptly discontinued (Figure 3).
Recurrences according to tapering duration (18-month Kaplan–Meier survival analysis).
Baseline and treatment factors associated with pericarditis recurrence
In order to investigate further the association of baseline clinical data and treatment-related variables with pericarditis recurrence, a logistic regression analysis was performed, showing that: pericardial effusion at baseline, permanent discontinuation of anakinra, full-dose treatment duration and tapering duration were significantly correlated with the risk of recurrence.
In a Cox regression model (Table 3), a longer full-dose treatment (HR 0.96 per one-month increase, P = 0.023) and a longer tapering were both associated with lower recurrence (HR 0.84 per one-month increase, P < 0.001).
Cox regression analysis of factors associated with pericarditis recurrence during a 36-month follow-up of patients treated with anakinra.
| Factors | P value | HR | 95% CI |
|---|---|---|---|
| Pericardial effusion at anakinra start | 0.297 | 1.75 | 0.61–4.98 |
| Permanent discontinuation of anakinra treatment (for any reason, included treatment completion) | 0.362 | 0.50 | 0.11–2.21 |
| Full-dose treatment (per one-month increase) | 0.023 | 0.96 | 0.93–0.99 |
| Tapering duration (per one-month increase) | <0.001 | 0.84 | 0.77–0.91 |
| Factors | P value | HR | 95% CI |
|---|---|---|---|
| Pericardial effusion at anakinra start | 0.297 | 1.75 | 0.61–4.98 |
| Permanent discontinuation of anakinra treatment (for any reason, included treatment completion) | 0.362 | 0.50 | 0.11–2.21 |
| Full-dose treatment (per one-month increase) | 0.023 | 0.96 | 0.93–0.99 |
| Tapering duration (per one-month increase) | <0.001 | 0.84 | 0.77–0.91 |
HR: hazard ratio; CI: confidence interval.
Cox regression analysis of factors associated with pericarditis recurrence during a 36-month follow-up of patients treated with anakinra.
| Factors | P value | HR | 95% CI |
|---|---|---|---|
| Pericardial effusion at anakinra start | 0.297 | 1.75 | 0.61–4.98 |
| Permanent discontinuation of anakinra treatment (for any reason, included treatment completion) | 0.362 | 0.50 | 0.11–2.21 |
| Full-dose treatment (per one-month increase) | 0.023 | 0.96 | 0.93–0.99 |
| Tapering duration (per one-month increase) | <0.001 | 0.84 | 0.77–0.91 |
| Factors | P value | HR | 95% CI |
|---|---|---|---|
| Pericardial effusion at anakinra start | 0.297 | 1.75 | 0.61–4.98 |
| Permanent discontinuation of anakinra treatment (for any reason, included treatment completion) | 0.362 | 0.50 | 0.11–2.21 |
| Full-dose treatment (per one-month increase) | 0.023 | 0.96 | 0.93–0.99 |
| Tapering duration (per one-month increase) | <0.001 | 0.84 | 0.77–0.91 |
HR: hazard ratio; CI: confidence interval.
Discussion
This is the first international registry on the use of anakinra in patients with corticosteroid-dependent and colchicine-resistant refractory RP.
The main findings of this registry are that anakinra dramatically decreased pericarditis recurrence rate (six-fold reduction, from 2.33 flares per patient per year to 0.39 flares per patient per year), ED admissions (11-fold reduction, from 1.08 admissions per patient per year to 0.10 admissions per patient per year), hospitalisations (seven-fold reduction, from 0.99 hospitalisations per patient per year to 0.13 hospitalisations per patient per year) and corticosteroid dependence. This observation is particularly important because side effects from corticosteroids can be severe and affect up to 25% of chronically treated patients.6
Overall, our results correspond to a reduced pericarditis recurrence burden from once every 157 days to once every 939 days. Moreover, during follow-up of this high-risk population, 43% of patients were in stable remission without recurrence, and another 29% had only a single recurrence. Compared with the AIRTRIP trial, results are similar with an incidence rate of 0.11 flares per patient per year in the AIRTRIP and 0.39 flares per patient per year in this real-world registry.
The safety profile of anakinra is also reassuring in this registry. Indeed, despite a high incidence of AEs (44%), there were no life-threatening AEs. Most AEs were mild, predominantly related to transient (some weeks) local skin reactions. Skin reactions had an incidence (38%) similar to previous studies (44%).12,29 They can be treated with oral and/or topical antihistamines and local corticosteroids. Furthermore, their occurrence can be decreased by changing the injection site every time and warming the anakinra syringe to room temperature before use, along with the application of a cold pack to the area of injection for some minutes before and after drug administration.31 Patients should be informed in advance about the potential for such reactions to prevent drug discontinuation. Other AEs included arthralgias and myalgias (6%), transaminase elevations (3%) and neutropenia (1%). These two latter AEs had previously been reported in the literature with an incidence of, respectively, 14% and 1.5%.29 Significant infections during treatment with IL-1 antagonists do occur, but are rare. In a review of clinical trials,32 significant infections occurred in 1.7% of patients, mostly respiratory and soft-tissue infections. In our registry, we observed six infections (3%), predominantly involving the respiratory system and soft tissues, all of which resolved with proper treatment.
Chronic inflammation underlying recurrent pericarditis is likely to be determined by an amplified and self-sustained autoinflammatory and/or autoimmune response following different exogenous or endogenous triggers; autoinflammatory and autoimmune mechanisms are not alternative, because they are often in overlap in most immune-mediated diseases. Anakinra, a recombinant IL-1 receptor antagonist which inhibits IL-1 action, has recently shown promise in recurrent pericarditis, but has been used for more than 15 years in the treatment of rheumatoid arthritis and various monogenic and polygenic systemic autoinflammatory disease (such as cryopyrin-associated periodic syndromes, familial Mediterranean fever and Still’s disease).8,9 Our results support the hypothesis that IL-1 plays a critical role in the pathogenesis of refractory RP.
Recent reviews investigated the current evidence and future challenges with anakinra treatment for refractory RP, and emphasised that the main unresolved issues relate to the duration of initial treatment and the tapering protocol.8,9,12 To address this issue, the analysis of this registry showed that both a full-dose treatment duration of at least 3 months as well as a cautious tapering of at least 3 months were associated with a lower risk of recurrence. A longer full-dose treatment (HR 0.97 per one-month increase) and a longer tapering were both associated with lower recurrence (HR 0.84 per one-month increase).
A limitation of this study, that reports data from an international registry, is the observational design and inclusion of patients only treated with anakinra, lacking a control group. Our results provide evidence for larger randomised controlled trials of anakira in patients with refractory RP. Even though refractory RP without baseline CRP elevation and without idiopathic aetiology is relatively rare as shown in this registry, further studies specifically targeted at the assessment of anti-IL-1 efficacy in these subgroups of patients could be useful to identify the most appropriate treatment for each patient.
In addition, our registry provided ‘real world’ data regarding the efficacy and safety of anakinra.
Conclusions
Anakinra appears to be a safe and efficacious option for corticosteroid-dependent and colchicine-resistant refractory RP despite optimal anti-inflammatory therapy (NSAIDs, colchicine and corticosteroids), allowing not only a dramatic reduction of recurrences but also of corticosteroid use. A full-dose treatment duration of at least 3 months followed by a tapering period of at least 3 months were the therapeutic approaches associated with a lower risk of recurrences.
Author contributions
MI and AA: study conception and design, data acquisition and analysis, manuscript drafting, critical revision. AK, AB and YA: study conception and design, data acquisition, critical revision. AA and EP: statistical analysis. GMDF, PC, VM, SM, SAL, GL, GE, DL, RM, GL, MDB, LC, LD, ACC, BV, LB, ET, FI, DP, ALC, DV, DT, GDL, CG, MR and JO: data acquisition, critical revision. All authors gave final approval and agree to be accountable for all aspects of the work ensuring integrity and accuracy.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: MI, PC, AK, AB, SAL and LD have participated as advisory board members for SOBI (Swedish Orphan Biovitrum AB, Stockholm, Sweden). LD received institutional research grants from SOBI. SAL has received consulting fees from SOBI.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
References
Author notes
The first two authors contributed equally to this paper
The final four authors contributed equally to this paper as last senior co-authors
Comments