Article Text

Original research
Anti-interleukin 1 agents for the treatment of recurrent pericarditis: a systematic review and meta-analysis
  1. Massimo Imazio1,2,
  2. Alessandro Andreis3,
  3. Francesco Piroli3,
  4. George Lazaros4,
  5. Marco Gattorno5,
  6. Martin Lewinter6,
  7. Allan L Klein7,
  8. Antonio Brucato8
  1. 1 University Cardiology, Cardiovascular and Thoracic Department, AOU Città della Salute e della Scienza di Torino, Torino, Italy
  2. 2 Department of Public Health and Pediatrics, University of Turin, Torino, Italy
  3. 3 Department of Medical Sciences, Division of Cardiology, AOU Città della Salute e della Scienza di Torino, University of Turin, Turin, Italy
  4. 4 Cardiology, National and Kapodistrian University of Athens, School of Medicine, Hippokration General Hospital, Athens, Greece
  5. 5 Istituto Pediatrico di Ricovero e Cura a Carattere Scientifico, Giannina Gaslini, Genova, Liguria, Italy
  6. 6 Larner College of Medicine, University of Vermont, Burlington, Vermont, USA
  7. 7 Department of Cardiovascular Medicine, Center for the Diagnosis and Treatment of Pericardial Diseases, Section of Cardiovascular Imaging, Heart, Vascular, and Thoracic Institute, Cleveland Clinic, Cleveland, Ohio, USA
  8. 8 Department of Biomedical and Clinical Sciences, Fatebenefratelli Hospital, University of Milan, Milano, Italy
  1. Correspondence to Professor Massimo Imazio, University Cardiology, Cardiovascular and Thoracic Department, AOU Città della Salute e della Scienza di Torino, 10126 Torino, Italy; massimo_imazio{at}yahoo.it

Abstract

Aims Corticosteroid-dependent and colchicine-resistant recurrent pericarditis (RP) is a challenging management problem, in which conventional anti-inflammatory therapy (nonsteroidal anti-inflammatory drugs, colchicine, corticosteroids) is unable to control the disease. Recent data suggest a potential role for anti-interleukin-1 (IL-1) agents for this condition. This study was designed to assess the safety and efficacy of anti-IL-1 agents in this setting.

Methods We performed a systematic review and meta-analysis of randomised controlled trials and observational studies assessing pericarditis recurrences and drug-related adverse events in patients receiving anti-IL-1 drugs for pericarditis.

Results The meta-analysis assessed 7 studies including 397 pooled patients with RP. The median age was 42 years, 60% were women and the aetiology was idiopathic in 87%. After a median follow-up of 14 months (IQR,12–39), patients receiving anti-IL-1 agents (anakinra or rilonacept) had a significantly reduction in pericarditis recurrences (incidence rate ratio 0.06, 95% CI 0.03 to 0.14), compared with placebo and/or standard medical therapy. Anti-IL-1 agents were associated with increased risk of adverse events compared with placebo (risk ratio (RR) 5.38, 95% CI 2.08 to 13.92): injection-site reactions occurred in 15/41 (36.6%) vs none (RR 14.98, 95% CI 2.09 to 107.09), infections occurred in 13/51 (25.5%) vs 3/41 (7.3%; RR 3.65, 95% CI 1.23 to 10.85). Anti-IL-1 agents were not associated with increased risk of severe adverse events.

Conclusions In patients with RP, anti-IL-1 agents (anakinra and rilonacept) are efficacious for prevention of recurrences, without severe adverse events.

  • pericarditis

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

INTRODUCTION

Recurrence is the most frequent complication of idiopathic and other forms of pericarditis, with an incidence ranging from 15% to 50% (depending on the underlying aetiology).1–3 Conventional anti-inflammatory therapy includes non-steroidal anti-inflammatory drugs (NSAIDs), colchicine and corticosteroids. In about 5% of patients with recurrent pericarditis (RP), incomplete resolution of symptoms occurs, with unsuccessful attempts at corticosteroid tapering and frequent relapses, leading to a clinical entity termed corticosteroid-dependent and colchicine-resistant refractory pericarditis.4–6 These patients have a severely impaired quality of life, severe side effects related to corticosteroids and very challenging clinical management due to poor efficacy of conventional anti-inflammatory drugs.7 Other therapeutic approaches (immunosuppressive agents, immunoglobulins, pericardiectomy)8–10 have uncertain efficacy and limited published evidence. Interleukin (IL)-1 induces a complex network of proinflammatory cytokines and regulates and initiates inflammatory responses by expression of integrins on leucocytes and endothelial cells. Additional effects of IL-1 include the induction of fever on the thermoregulatory centre in the hypothalamus, induction of hyperalgesia (increased pain sensitivity), vasodilatation and hypotension. IL-1 α and β, that remain the most studied members of the family and possess the stronger proinflammatory effect. IL-1 α can be released by damaged/inflamed pericardial cells triggering the release of IL-1 β by inflammatory cells. Inhibition of both IL-1 α and β represents a new therapeutic target for pericarditis.11 On this basis, anti-IL-1 agents have recently emerged as a possible new and safe therapeutic option, based on recent randomised controlled trials (RCTs) and observational studies. The present study was designed to systematically assess the efficacy and safety of anti-IL-1 agents in patients with RP, based on published RCTs and observational studies.

Methods

This meta-analysis was performed according to the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement,12 including all published RCTs and observational studies of patients treated with IL-1 inhibitors, in order to assess the role of anti-IL-1 drugs in the prevention of RP.

We searched through electronic databases (PubMed, MEDLINE, Scopus, Embase and the Cochrane Library) using the terms ‘pericarditis’ and ‘anti-IL1 agents’ OR ‘interleukin-1 inhibitor’ OR ‘interleukin-1 antagonist’ OR ‘anakinra’ OR ‘rilonacept’ OR ‘canakinumab’. This research was conducted up to 5 December 2020. Three independent reviewers (MI, AA, FP) screened titles and abstracts of all studies and potentially eligible ones were appraised in full text. The most relevant articles quoted in the studies were also screened. Discrepancies were resolved by consensus.

We excluded studies that were not performed on humans, not published in the English language or without outcomes of interest. To be eligible for inclusion, studies had to report baseline characteristics of the patients, procedural features and at least one of the outcomes of interest. If multiple reports described data from the same trial, only the study with the most recent and complete data was included. Data regarding study design, sample size, inclusion/exclusion criteria, clinical presentation, length of follow-up and outcomes of interest were extracted from the selected studies (tables 1 and 2, online supplemental tables 1 and 2).

Supplemental material

Table 1

Main features of included studies

Table 2

Outcomes definitions of included trials

The primary end-point was the incidence of RP, assessed as incidence rate ratio (IRR). The secondary end-points were drug-related adverse events (AEs) occurrence, including: any AE, cutaneous event (skin reaction) and infections.

The quality of included RCTs was independently assessed by three reviewers (MI, AA, FP) using the Cochrane quality assessment tool, considering six criteria: random sequence generation (selection bias), allocation sequence concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessment (detection bias), selective outcome reporting (reporting bias), incomplete outcome data (attrition bias). The risk of bias was judged to be low, medium, high or unclear (online supplemental table 3). Observational studies bias assessment was performed with the National Institutes of Health (NIH) quality assessment tool (see online supplemental table 4). Discrepancies were resolved by consensus.

The primary endpoint was quantified by computing the IRR of RP in patients receiving IL-1 inhibitors versus standard of care/placebo group. The incidence rate of RP was calculated as the number of events divided by the person-time at risk. For secondary endpoints, a pooled analysis of risk ratio (RR) of developing AEs was performed only based on studies comparing IL-1 inhibitors versus placebo. AEs were defined as adverse events that started or increased in severity between the first dose of study drug and 6 weeks after the last dose of study drug.

Statistical pooling was performed according to a random-effect model with generic inverse-variance weighting, computing risk estimates with 95% CI, using Review Manager 5.3 (The Cochrane Collaboration, the Nordic Cochrane Centre, Copenhagen, Denmark). Standard hypothesis testing was set at the two-tailed 0.05 level. Heterogeneity between studies was assessed by measuring inconsistency using Cochrane I2 index,12 which describes the percentage of total variation across the studies that is due to heterogeneity rather than chance. I2 values of 25%, 50% and 75%, were attributed to small, moderate and large amounts of heterogeneity, respectively. A prespecified subgroup analysis for type of study (RCTs versus observational) was performed to assess whether the loss of randomisation due to the design of the studies would have changed the results.

Patient and public involvement

This research was done without patient involvement. Patients were not invited to comment on the study design and were not consulted to develop patient-relevant outcomes or interpret the results. Patients were not invited to contribute to the writing or editing of this document for readability or accuracy.

Results

After research and selection of eligible studies (see figure 1 and online supplemental figure 1 – PRISMA Flow diagram), a total of 7 studies and 397 pooled patients were included in the present analysis. Among the seven studies collected, two assessed rilonacept and five assessed anakinra. No study of interest using Canakinumab was found to be eligible.

Figure 1

PRISMA flow diagram. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Regarding study design, two were RCTs: the RHAPSODY trial,6 a double-blind, multicentre study comparing rilonacept versus placebo and the AIRTRIP trial,4 a double-blind, multicentre study comparing anakinra versus placebo. The remaining five studies were observational studies comparing the incidence of RP before and after anakinra administration.5 13–16

The sample size ranged from 12 to 224 patients. The median age was 42 years (IQR, 14–46), 60% of patients were females and 87% of patients were considered to have idiopathic pericarditis. The median follow-up was 14 months (IQR,12–39). Patients baseline characteristics are reported in online supplemental table 1.

Primary outcome: pericarditis recurrence

After a median follow-up of 14 months (IQR 12–39), patients receiving anti-IL-1 agents (anakinra or rilonacept) had a significant lower IRR for RP (IRR 0.06, 95% CI 0.03 to 0.14, I2=95%) compared with patients receiving placebo and/or standard medical therapy, as shown in figure 2. The consistency of the results was confirmed through subgroup analyses according to study design. Anti-IL-1 agents reduced the risk of RP in both RCTs (IRR 0.04, 95% CI 0.01 to 0.12, I2=0%) and observational studies (IRR 0.07, 95% CI 0.03 to 0.17, I2=95%), despite great heterogeneity in this latter subgroup.

Figure 2

Recurrent pericarditis, assessed as IRR in patients treated with or without anti-IL-1 agents. IL-1, interleukin-1; IRR, incidence rate ratio.

Secondary outcome: drug-related adverse events (AEs)

Anti-IL-1 agents were associated with increased risk of any AEs compared with placebo (RR 5.38, 95% CI 2.08 to 13.92, I2=0%): injection-site reactions occurred in 15/41 (36.6%) vs none, RR 14.98, 95% CI 2.09 to 107.09, I2=0%; infections occurred in 13/51 (25.5%) vs 3/41 (7.3%), RR 3.65, 95% CI 1.23 to 10.85, I2=0% (figure 3). All events were mild or moderate in severity according to the MedDRA System Organ Class of infections and infestations. Injection site reactions were the most common AEs. Permanent discontinuation of drug due to AEs was unusual in RCTs, but occurred in four patients in the RHAPSODY trial during the run-in period before randomisation. There were no deaths during follow-up.

Figure 3

Drug-related AEs in patients treated with or without anti IL-1 agents. AEs, adverse events; IL-1, interleukin-1.

Discussion

Corticosteroid-dependent and colchicine-resistant refractory RP is a clinically challenging scenario, due to the paucity of therapeutic options and their high incidence of AEs, especially for corticosteroids. The pathogenesis of RP is complex and still not fully understood. In predisposed individuals, pericardial inflammation is likely due to an amplified and self-sustained autoinflammatory and/or autoimmune process, triggered by a wide number of endogenous or exogenous stimuli.17 18 The autoinflammatory hypothesis is supported by the fact that RP shares several clinical features (ie, recurrent fever with CRP elevation) with autoinflammatory diseases such as familial Mediterranean fever (FMF), and tumour necrosis factor receptor-associated periodic syndrome (TRAPS) as well as a striking response to anti-IL-1 agents. The nucleotide-binding oligomerization domain-like receptor pyrin containing domain 3 inflammasome (NLRP3) is the key element in these diseases: its activation, due to gene mutation (Mediterranean fever (MEFV) in FMF, TNFRSF1A in TRAPS), is associated with increased production of IL-1, the ‘endogenous pyrogen’ responsible for inflammatory cascade amplification and recruitment of neutrophils and macrophages.17 18 The autoimmune hypothesis, on the other hand, is supported by the demonstration of anti-intercalated disk and antinuclear autoantibodies in the serum of patients with RP19 20 and by the presence of high concentrations of proinflammatory cytokines (interferon-γ, IL-6, IL-8) in pericardial fluid, which are conventionally associated with autoimmune processes.21 Moreover, the dramatic clinical response to corticosteroids and other more specific immunosuppressive drugs also supports this mechanism.22 23

These two pathogenetic hypotheses are both valuable and supported by evidence; therefore, it cannot be excluded that a synergistic interaction of these two components is present simultaneously with a predominant autoinflammatory response (based on IL-1 overproduction) in patients with idiopathic RP with fever and CRP elevation (inflammatory phenotype) and a predominant autoimmune response in patients with underlying autoimmune disease (non-inflammatory phenotype).18 In this scenario, anti-IL-1 agents appear to be a major advance in medical therapy of RP and represent a potential paradigm shift, especially in the setting of recurrent episodes caused by a predominant autoinflammatory response.

The 2015 ESC Guidelines for the diagnosis and management of pericardial diseases24 suggest IL-1 inhibitors only as a third-line option. Besides, being based on case series and expert opinion, this recommendation had a low level of evidence (IIb C). The recently published RCTs (the RHAPSODY trial6 of Rilonacept and the AIRTRIP trial4 of Anakinra) along with multicentre observational studies with larger cohorts (ie, the IRAP registry,5 support this newer approach with stronger evidence.25–28 This is the first meta-analysis assessing the role of anti-IL-1 agents in term of efficacy and safety for RP.

The main finding of this meta-analysis confirms that anakinra and rilonacept: (1) are both effective in reducing the rate of RP, showing a >90% reduced risk of developing episodes of RP compared with subjects on standard medical therapy; (2) are both not associated with severe AEs, although their use increases the risk of cutaneous reactions (RR 14.98, 95% CI 2.09 to 107.09) and infections (RR 3.93, 95% CI 1.33 to 11.66). The consistency of these results was confirmed in the prespecified subgroup analyses according to study design. Regarding the between-study heterogeneity, its origin could be explained by study design, unavoidable when comparing data from RCTs together with data derived from observational studies. Indeed, it is noteworthy that in the subanalysis considering only RCTs the heterogeneity assessed with I2 test showed a value of 0%.

The most significant reduction of recurrences was achieved in patients with an idiopathic aetiology (87%), confirming once again the key role of IL-1 in patients with inflammatory phenotype. This is important since other aetiologies (neoplastic, autoimmune, infective and so on) were often excluded in the studies that we included. Furthermore, this is in line with the assessment by both the AIRTRIP and RHAPSODY authors who emphasised the importance, in terms of efficacy endpoints, of patient selection before starting IL-1 inhibitors, since patients with an inflammatory phenotype are likely better candidates than those without evidence of systemic inflammation by C reactive protein elevation.4–6

Regarding the safety profile of anti-IL-1 agents, most importantly, no severe adverse events were recorded during administration of anakinra and rilonacept, even with the longest follow-up (60 months). The most frequent adverse events were transient injection-site reactions (RR 14.98, 95% CI 2.09 to 107.09). Such lesions are caused by a local hypersensitivity reaction that can be easily treated with short courses of topic and/or oral antihistamines and corticosteroids.29 Furthermore, this meta-analysis shows an increased risk of infections with anti-IL-1 agents. However, it should be observed that among the two RCTs considered (figure 3),4 6 this association was observed only with rilonacept in the RHAPSODY trial (in which 7/12 the events were caused by upper respiratory tract infections).6 On the other hand in the AIRTRIP trial, the risk of infection was not increased by anti-IL-1 agents (one single event described, in a patient with Herpes Zoster during treatment with anakinra).4 Indeed in the large IRAP study on 224 patients receiving anakinra, infections occurred in 6/224 (3%) patients (mostly respiratory and soft-tissue infections).5 Other smaller observational studies included in this meta-analysis did not report infectious events with anti-IL-1 agents, except for a study on anakinra in pericardial constriction (1/39, 2.5%)15 and another assessing rilonacept (1/12, 8.3%). Serious infections occurred in 1/224 (0.4%) patient in the IRAP and 1/12 (8.3%) patient in a study on rilonacept.11 In both cases, skin or soft-tissue infections were observed and anti-IL-1 agents were discontinued. In previous studies, significant infections were rarely described during anti-IL-1 agents use (mostly upper tract respiratory and soft-tissue infections), with an incidence of severe events in less than 2% of cases.30

The safety profile of anti-IL agents is reassuring. Indeed, no severe adverse events were reported in RCTs. This side effect profile compares favourably with those associated with chronic therapy with corticosteroids and/or high doses of NSAIDs.

Anakinra has different approved indications: Cryopyrin-associated periodic syndromes (CAPS) and rheumatoid arthritis for the Food and Drug Administration (FDA) and CAPS, systemic juvenile idiopathic arthritis, adult-onset Still disease and rheumatoid arthritis for European Medicines Agency (EMA). Ongoing evidence is now supporting its use in the first episode of pericarditis,w1 other cardiovascular indications, such as myocardial infarction,w2 remodelling and heart failure,w3 and diabetes.w4

Limitations

The limitations of this meta-analysis are due to the paucity of RCTs in the literature; the analyses were derived from data pooled with observational studies. This may partly affect the reliability of the results, which shows high heterogeneity in the subgroup of observational studies (I2=95%). Furthermore, while in RCTs anti-IL-1 agents were compared with placebo, while in some observational studies, a pre-post comparison after IL-1 therapy was reported. Among the strengths of our study is that this is the first meta-analysis assessing the efficacy and safety profile of anti-IL-1 agents for RP management and the overall quality of the RCT studies was very high.

Conclusion

Anti-IL-1 drugs (anakinra, rilonacept) are a valuable option in patients with RP, especially in patients with an ‘inflammatory phenotype’. Further studies are warranted to assess their possible use as an early treatment option, before corticosteroids and even before recurrences, in selected patients.

Key messages

What is already known on this subject?

  • Corticosteroid-dependent and colchicine-resistant recurrent pericarditis is a major challenge in the management of pericardial diseases.

  • Anti-interleukin-1 (IL-1) agents (eg, anakinra and rilonacept) have been investigated as new potential therapeutic options for recurrent pericarditis with evidence of C reactive protein elevation.

What might this study add?

  • In this systematic review, after a median follow-up of 14 months (IQR 12–39), patients receiving anti-IL-1 agents (anakinra or rilonacept) had a significant lower incidence rate ratio for recurrent pericarditis (incidence rate ratio 0.06, 95% CI 0.03 to 0.14) compared with patients receiving placebo and/or standard medical therapy.

  • Anti-IL-1 agents were associated with increased risk of any adverse events compared with placebo (risk ratio (RR) 5.38, 95% CI 2.08 to 13.92), including injection-site reactions (RR 14.98, 95% CI 2.09 to 107.09) and infections (RR 3.93. 95% CI 1.33 to 11.66).

How might this impact on clinical practice?

  • Anti-IL-1 agents represent a major advance in the medical therapy of recurrent pericarditis and a potential paradigm shift in the therapy of this disease.

  • Anti-IL-1 agents are efficacious for the treatment of refractory recurrent pericarditis, mostly in cases with C reactive protein elevation.

  • The safety profile of these drugs is reassuring, particularly when compared with those of corticosteroids and high doses of non-steroidal anti-inflammatory drugs; nevertheless, careful monitoring is warranted due to an increased risk of local skin reactions and infections.

Supplemental material

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • Contributors All authors contributed to the planning, conduct, drafting and reporting of the work. The manuscript was revised and approved by all authors. MI, AA and FP are responsible for the overall content as guarantors.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests MI, MG, ML, ALK and AB have been Advisory Board members for KINIKSA and SOBI.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Author note Additional references w1–4 can be found in online supplemental file 2.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.

Linked Articles