Article Text

The 2021 European Alliance of Associations for Rheumatology/American College of Rheumatology points to consider for diagnosis and management of autoinflammatory type I interferonopathies: CANDLE/PRAAS, SAVI and AGS
Free
  1. Kader Cetin Gedik1,
  2. Lovro Lamot2,
  3. Micol Romano3,
  4. Erkan Demirkaya3,
  5. David Piskin4,5,
  6. Sofia Torreggiani1,6,
  7. Laura A Adang7,
  8. Thais Armangue8,
  9. Kathe Barchus9,
  10. Devon R Cordova10,
  11. Yanick J Crow11,12,
  12. Russell C Dale13,
  13. Karen L Durrant9,14,
  14. Despina Eleftheriou15,
  15. Elisa M Fazzi16,
  16. Marco Gattorno17,
  17. Francesco Gavazzi7,18,
  18. Eric P Hanson19,
  19. Min Ae Lee-Kirsch20,
  20. Gina A Montealegre Sanchez21,
  21. Bénédicte Neven22,
  22. Simona Orcesi23,24,
  23. Seza Ozen25,
  24. M Cecilia Poli26,
  25. Elliot Schumacher9,
  26. Davide Tonduti27,
  27. Katsiaryna Uss1,
  28. Daniel Aletaha28,
  29. Brian M Feldman29,30,
  30. Adeline Vanderver7,31,
  31. Paul A Brogan15,
  32. Raphaela Goldbach-Mansky1
  1. 1 Translational Autoinflammatory Diseases Section, Laboratory of Clinical Immunology and Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA
  2. 2 Department of Pediatrics, University Hospital Centre Zagreb, University of Zagreb School of Medicine, Zagreb, Croatia
  3. 3 Division of Paediatric Rheumatology, Department of Paediatrics, Schulich School of Medicine and Dentistry, University of Western Ontario, London, Ontario, Canada
  4. 4 Department of Epidemiology and Biostatistics, Schulich School of Medicine & Dentistry, University of Western Ontario, London, Ontario, Canada
  5. 5 London Health Sciences Center, Lawson Health Research Institute, London, Ontario, Canada
  6. 6 UOC Pediatria a Media Intensità di Cura, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milano, Lombardia, Italy
  7. 7 Division of Neurology, The Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
  8. 8 Pediatric Neuroimmunology Unit, Neurology Service, Sant Joan de Deu Children's Hospital, and IDIBAPS-Hospital Clinic, University of Barcelona, Barcelona, Spain
  9. 9 Autoinflammatory Alliance, San Francisco, California, USA
  10. 10 Aicardi-Goutieres Syndrome Americas Association, Manhattan Beach, California, USA
  11. 11 Centre for Genomic and Experimental Medicine, MRC Institute of Genetics and Molecular Medicine, University of Edinburg, Edinburg, UK
  12. 12 Laboratory of Neurogenetics and Neuroinflammation, Institut Imagine, Université de Paris, Paris, Île-de-France, France
  13. 13 Kids Neuroscience Centre, Faculty of Medicine and Health, The University of Sydney, Sydney, New South Wales, Australia
  14. 14 Kaiser San Francisco Hospital, San Francisco, California, USA
  15. 15 Great Ormond Street Institute of Child Health, University College London, London, UK
  16. 16 Child Neurology and Psychiatry Unit, Department of Clinical and Experimental Sciences ASST Civil Hospital, University of Brescia, Brescia, Italy
  17. 17 Center for Autoinflammatory diseases and Immunodeficiencies, IRCCS Istituto Giannina Gaslini, Genoa, Italy
  18. 18 Department of Molecular and Translational Medicine, University of Brescia, Brescia, Italy
  19. 19 Riley Hospital for Children, Indiana University School of Medicine, Indianapolis, Indiana, USA
  20. 20 Department of Pediatrics, Medical Faculty Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany
  21. 21 Intramural Clinical Management and Operations Branch (ICMOB), Division of Clinical Research, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA
  22. 22 Necker Children’s Hospital, Assistance Publique-Hôpitaux de Paris, Université de Paris, Institut Imagine Institut des Maladies Genetiques, Paris, Île-de-France, France
  23. 23 Child Neurology and Psychiatry Unit, IRCCS Mondino Foundation, Pavia, Italy, Italy
  24. 24 Department of Brain and Behavioral Sciences, University of Pavia, Pavia, Lombardia, Italy
  25. 25 Pediatric Rheumatology, Hacettepe University, Ankara, Turkey
  26. 26 Department of Pediatrics, Facultad de Medicina Clinica Alemana Universidad del Desarrollo, Santiago, Chile
  27. 27 Child Neurology Unit, COALA (Center for Diagnosis and Treatment of Leukodystrophies), V. Buzzi Children’s Hospital, Milano, Italy
  28. 28 Department of Rheumatology, Medical University of Vienna, Vienna, Austria
  29. 29 Division of Rheumatology, Hospital for Sick Children, Toronto, Ontario, Canada
  30. 30 Department of Pediatrics, Faculty of Medicine, University of Toronto Institute of Health Policy Management and Evaluation, Toronto, Ontario, Canada
  31. 31 Department of Neurology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, USA
  1. Correspondence to Dr Raphaela Goldbach-Mansky, Translational Autoinflammatory Diseases Section, Laboratory of Clinical Immunology and Microbiology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA; goldbacr{at}mail.nih.gov

Abstract

Objective Autoinflammatory type I interferonopathies, chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature/proteasome-associated autoinflammatory syndrome (CANDLE/PRAAS), stimulator of interferon genes (STING)-associated vasculopathy with onset in infancy (SAVI) and Aicardi-Goutières syndrome (AGS) are rare and clinically complex immunodysregulatory diseases. With emerging knowledge of genetic causes and targeted treatments, a Task Force was charged with the development of ‘points to consider’ to improve diagnosis, treatment and long-term monitoring of patients with these rare diseases.

Methods Members of a Task Force consisting of rheumatologists, neurologists, an immunologist, geneticists, patient advocates and an allied healthcare professional formulated research questions for a systematic literature review. Then, based on literature, Delphi questionnaires and consensus methodology, ‘points to consider’ to guide patient management were developed.

Results The Task Force devised consensus and evidence-based guidance of 4 overarching principles and 17 points to consider regarding the diagnosis, treatment and long-term monitoring of patients with the autoinflammatory interferonopathies, CANDLE/PRAAS, SAVI and AGS.

Conclusion These points to consider represent state-of-the-art knowledge to guide diagnostic evaluation, treatment and management of patients with CANDLE/PRAAS, SAVI and AGS and aim to standardise and improve care, quality of life and disease outcomes.

  • inflammation
  • polymorphism
  • genetic
  • immune system diseases

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

Autoinflammatory type I interferonopathies are genetically defined (monogenic or digenic) immunodysregulatory disorders characterised by the presence of a type I interferon (IFN) signature in peripheral blood and variable systemic inflammation.1–3 In this expanding group of ultra-rare diseases, chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature/proteasome-associated autoinflammatory syndrome (CANDLE/PRAAS), stimulator of interferon genes (STING)-associated vasculopathy with onset in infancy (SAVI) and Aicardi-Goutières syndrome (AGS) are the most common.

Patients with type I interferonopathies present early in life often within the first week of life; prenatal onset has been reported in patients with AGS; however, late-onset cases presenting at ages 14, 18 and 5.6 years with CANDLE/PRAAS, SAVI and AGS, respectively, have been reported.4–11 Despite CANDLE/PRAAS, SAVI and AGS having distinct clinical phenotypes of varying disease severity, the individual clinical manifestations of these diseases can overlap, and all are associated with high morbidity and mortality if untreated.4 12 Recent advances in the genetic description of these disorders permit better characterisation of disease-specific clinical manifestations, and provide evidence supporting the pathogenic role of type I IFN signalling.1 2 12 13 These developments prompted the Task Force lead by the steering committee (two convenors (PAB, RG-M), a neurologist (AV), two methodologists (BMF, ED) and three paediatric rheumatologists/EULAR fellows (KCG, LL, MR) and a rheumatologist (ST)) to review the existing data and develop consensus statements, with the aim of formulating state-of-the-art guidance on the diagnosis, treatment and long-term monitoring of patients with these rare diseases.

Thus, the objective of this project was to develop points to consider for the diagnosis, treatment and long-term monitoring of patients with CANDLE/PRAAS, SAVI and AGS.

The Task Force targets their guidance to paediatricians, internists and subspecialists involved in the care of patients with autoinflammatory type I interferonopathies and to patients and caregivers. These points to consider were developed not only to provide a resource for physicians to facilitate management but also for policy makers governing who have a role in authorising patients’ access to various diagnostic tools and treatment options; all with the ultimate goal to harmonise the level of care and to improve quality of life and disease outcomes in this patient population.

Methods

The European Alliance of Associations for Rheumatology (EULAR)14 and the American College of Rheumatology (ACR) standardised operating procedures (SOPs) were followed during the project period (see online supplementary methods). With approval from the EULAR and ACR Executive Committees, an international Task Force consisting of worldwide recognised experts from North America, South America, Europe and Australia convened to develop points to consider for the diagnosis, treatment and long-term monitoring of three type I interferonopathies: CANDLE/PRAAS, SAVI and AGS. The Task Force members were selected based on expertise in treatment and care of these patients.

A face-to-face meeting in August 2019 defined the goal of the project and the target population. Then, the Task Force developed research questions related to diagnosis, treatment and long-term monitoring of these diseases using the Population, Intervention, Comparison, Outcome (PICO) format. Search terms were derived from PICO questions and a systematic literature review (SLR) was performed by three research fellows (KCG, MR, LL), with support from a librarian and an epidemiologist (DH and DP), and a senior methodologist (ED) to identify relevant literature published before September 2020.

Two rounds of pre-consensus meeting questionnaires, using the Delphi technique,15 included questions pertaining to diagnosis, treatment and long-term monitoring were sent to all Task Force members to indicate their agreement with each question or statement with yes/no using the Delphi technique; the Delphi questionnaire was sent to 28 Task Force members, of whom 22 were voting members. The Task Force members were asked to indicate their agreement with each statement, and a free text option was provided to capture every member’s comment for each statement. Draft statements and items in questions with 80% or higher agreement were retained for voting at the consensus meetings. Statements and items in questions that did not reach a greater than 80% consensus were reviewed and reworded and sent out in a second round of the Delphi questionnaire. The original and the revised/modified draft statements with the previously achieved level of agreement and the participants’ comments were included in the second survey. A free text option to capture comments and additional items was again included. Draft statements with 80% or higher agreement were retained for voting at the consensus meetings, and statements, which did not achieve 80% agreement, were marked for further discussion and refinement at the two consensus meetings. Responses were anonymous.

Based on the SLR findings and two pre-consensus meeting Delphi questionnaires, draft statements were refined by the steering group and were sent to the voting members prior to the consensus meetings. These draft statements were reviewed, discussed, revised and voted on in two consensus meetings that were held online in October 2020 due to the COVID-19 pandemic, one for CANDLE/PRAAS and SAVI, and one for AGS.

Two conveners (RGM, PAB), three methodologists (BMF, ED, DA), three fellows, an allied health professional and three disease experts attended both consensus meetings and, otherwise, participation was based on disease-specific expertise. The voting panel included 19 experts, 1 allied health professional and 1 patient representative for each disease. The joint statements addressing all three interferonopathies were voted on by the entire voting panel; CANDLE/SAVI-specific statements were voted on by 10 experts, 1 allied health professional, 1 SAVI and 1 CANDLE/PRAAS patient presentative, and AGS specific statements were voted on by 14 experts, 1 allied health professional and 1 AGS patient representative. During the meetings, statements that achieved at least 80% agreement were accepted; statements with <80% were discussed a final time in a Nominal Groups round robin discussion (https://www.cdc.gov/healthyyouth/evaluation/pdf/brief7.pdf) and were only accepted if the revised statement reached an 80% agreement.

The Oxford Levels of Evidence (LoE) were applied to each point to consider.16 The strength of each statement ranged from A (directly based on level I evidence) to D (directly based on level IV evidence or extrapolated recommendations from level I, II or III evidence).16 Finally, the finalised statements were circulated in a post-consensus meeting Delphi questionnaire to determine level of agreement (LoA). Members of the Task Force were asked to provide their final LoA for each point to consider using a scale of 0 (completely disagree) to 10 (completely agree), which is reported in the tables below.

Results

Systematic literature review

A summary of the literature search strategy and results are provided as supplementary material (online supplementary methods). Based on SLR and consensus conferences, 4 overarching principles and 17 disease-specific points to consider pertaining to the genetically defined interferonopathies (table 1) with their respective LoE, grade of recommendation (GoR) and LoA were generated.17

Table 1

Points to consider for the diagnosis, treatment and long-term monitoring of patients with type I interferonopathies, CANDLE/PRAAS, SAVI and AGS

Overarching principles guiding the management of patients with CANDLE/PRAAS, SAVI and AGS

The systemic inflammatory multiorgan involvement in patients with CANDLE/PRAAS, SAVI or AGS can ultimately result in progressive organ injury and early mortality.4 Damage accrues over time, often manifesting later in life, thus highlighting the importance of early diagnosis and treatment.1 12

Autoinflammatory syndromes may present with phenotypic overlap early in life, which poses diagnostic challenges.12 In addition, mutations in individuals genes may be associated with considerable phenotypic heterogeneity and variable disease severity.18 19 Genetic confirmation is thus essential for making a precise diagnosis which then facilitates targeted therapy and initiation of genetic counselling with the goal of achieving better clinical outcomes. Patients, their parents and siblings should have access to formal genetic counselling. Genetic counselling can initiate the risk assessment process depending on the type of inheritance for specific disease-causing mutation and help patients understand their test results, including the medical implications for themselves, their reproductive health concerns and impact on their relatives. Patients with clinical symptoms of CANDLE/PRAAS, SAVI or AGS who do not harbour any of the disease-causing mutations described here should be referred to specialty/research centres that can guide further workup and treatment. There is no cure for type I interferonopathies. Current treatment options therefore aim to prevent development or progression of end organ damage by controlling systemic and organ inflammation,20 21 to improve quality of life and to improve disease outcomes.1 Given the paucity of long-term outcome data on newly available treatments, monitoring of disease activity, and development of organ-specific and treatment-related complications is essential.1 22 23 A multidisciplinary team is required to provide optimal care in the context of multiorgan system involvement.24 25

Points to consider 1–8: diagnostic evaluation focuses on raising an early suspicion and on facilitating genetic testing, appropriate clinical and laboratory workup and early treatment

Diagnostic evaluation

The presence of a chronically elevated peripheral blood IFN signature is a common finding in patients with the type I interferonopathies CANDLE/PRAAS, SAVI and AGS. In contrast, traditional inflammatory markers such as C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) are typically elevated in CANDLE/PRAAS and SAVI but rarely in patients with AGS.2 7 12 18 26–30 A peripheral blood IFN signature may be measured using different methodologies, including a 28-gene IFN scoring system using NanoString technology or by quantitative reverse transcriptase (RT) PCR methods of gene subsets should be measured repeatedly to establish chronic elevation.13 Scores may be negative in the diagnostic phase in patients with milder disease; or in response to glucocorticoid treatment. In addition, patients with AGS with RNASEH2B mutations may have a negative IFN signature even with active disease.31 A practical barrier is the limited number of centres with the ability to check an IFN signature. Thus, a chronically elevated peripheral blood IFN signature is not required for diagnosis but can be very useful in raising the suspicion of an interferonopathy. For most IFN signatures, sensitivity and specificity data are not available. However, in a retrospective study, the IFN signature at a set cut-off score was helpful in differentiated patients with an interferonopathy from healthy controls and from patients with a cryopyrin-associated periodic syndrome (an interleukin-1-mediated autoinflammatory disease). The IFN signature demonstrated an area under the receiver operator characteristic (ROC) curve of 0.98, with sensitivity and specificity exceeding 0.8.12 Currently, the IFN signature should be interpreted in the context of normal values of the laboratory that conducts the test, since no internationally standardised methodologies or reference ranges are currently available.

Genetic evaluation

As there can be significant overlap of clinical features across several autoinflammatory disorders, a confirmed genetic diagnosis is critical to facilitating a precision medicine approach and targeted therapy. Next-generation sequencing (eg, targeted gene panel, whole exome or whole genome sequencing) to screen for pathogenic variants rather than single gene Sanger sequencing is recommended. Sanger sequencing of individual genes may still be cost effective in patients with known familial disease; and may be the only available option if next-generation sequencing is not yet available to the patient. However, this increasingly outdated ‘gene by gene’ approach ultimately may result in diagnostic delay and may not be cost-effective.32 In addition to the known disease-causing genes1 2 5 7 12 18 31 33–39 (table 1), screening should be considered for diseases that can mimic one of these disorders; their genetic causes8 12 40–45 are listed in table 2. Allelic, monogenic or digenic, double heterozygous mutations in genes encoding proteasome or immunoproteasome subunits are the cause for CANDLE/PRAAS, with biallelic pathogenic PSMB8 variants being the most common cause. Digenic disease causing mutations including PSMB8, PSMA3, PSMB4 and PSMB9,1 2 26 compound heterozygous mutations including PSMB4, PSMB8 and PSMG2 2 12 and autosomal dominant loss-of-function mutations in POMP 2 also cause CANDLE/PRAAS but are rarer. However, novel disease-causing genes are being added as causes for CANDLE/PRAAS. All proteasome genes should be specifically assessed in a patient with a suggestive clinical phenotype. Both parents may need to be tested to confirm digenic inheritance. The inheritance of SAVI is mostly autosomal dominant, and most patients harbour a de novo heterozygous missense mutations in the STING1 gene that confers a gain-of-function by increasing TANK-binding kinase 1-mediated IRF3 phosphorylation and IFNB1 transcription.7 46 Liu et al also reported somatic mosaic mutations in one patient (OMIM-615934). So far only additive STING1 gain-of-function mutations in p.R284W require homozygosity to confer disease.47 Furthermore, mostly loss-of-function mutations in genes encoding proteins that regulate nucleic acid metabolism or signalling cause AGS.34 These include biallelic null mutations in TREX1 and SAMHD1; biallelic null mutations in the disease-causing genes, RNASEH2A, RNASEH2B, RNASEH2C or ADAR1 have not been reported. Disease-causing IFIH1 variants are all heterozygous gain-of-function mutations that increase type I IFN signalling.34 Recently, biallelic mutations in LSM11 and RNU7-1, which encode components of the replication-dependent histone pre-mRNA–processing complex extend defects in nucleic acid metabolism to histone mRNAs.48 It is important to note that large deletions, such as deletions in AGS-related genes including SAMHD1, may be missed on exome sequencing and need to be reviewed using other testing modalities.31 49 50 If following routine genetic workup, a molecular diagnosis is not established in a patient with suggestive phenotypic features, referral to a research centre of excellence for further evaluation should be considered.

Table 2

List of genetically defined disease and genes that should be considered in the differential diagnosis of CANDLE/PRAAS, SAVI and AGS

Clinical evaluation

In patients with undifferentiated autoinflammatory diseases or otherwise unexplained systemic inflammation, certain clinical features are suggestive of CANDLE/PRAAS, SAVI or AGS (tables 1 and 3).

Table 3

Clinical features suggestive of CANDLE/PRAAS, SAVI and AGS

The following clinical features are relevant to the workup of patients with suspected interferonopathies:

Cutaneous manifestations

Inflammatory skin lesions are present in all three diseases; however, the nature of the rash differs. Nodular rashes or violaceous annular rashes should prompt a diagnostic workup for CANDLE/PRAAS. Another specific cutaneous finding for CANDLE/PRAAS is panniculitis (particularly neutrophilic panniculitis) and panniculitis-induced lipodystrophy, which are hallmarks of the disease.1 2 9 12 18 36 37 51

The presence of vasculopathic skin lesions such as pernio (‘chilblain lesions’) or acral ischaemia presenting as Raynaud’s phenomenon, and/or ‘purple toes’ is suggestive of SAVI7 44 47 and AGS,33 52–55 the development of gangrene with prolonged ischaemic attacks is a feature of SAVI1 7 44 (table 3). Skin involvement is the most common symptom in patients with SAVI at presentation1 7 56–59 but some patients can present with severe lung disease and only minimal skin involvement.8 46 60 61

In addition to chilblain-like lesions and acrocyanosis, other skin manifestations such as periungual erythema, or necrotic lesions of the toes, fingers and outer helix, can be seen in patients with AGS.33 52–55 Moreover, some patients with AGS can have panniculitis as well.34 Finally, some patients with AGS have recurrent oral ulcers.50 62

Lesional skin biopsies in areas that can safely be biopsied can be beneficial in revealing the neutrophilic dermatosis, small vessel vasculitis (from necrotic area), fasciitis57 and granulomatous nodular dermatitis,59 thus supporting the diagnosis of SAVI while in AGS specifically, a lesional biopsy can demonstrate deposition of immunoglobulin and complement in the walls of small vessels.63

Neurological manifestations

Although CANDLE/PRAAS-affected patients present with headaches and may develop aseptic meningitis,24 neurological findings are most common and severe in AGS and include subacute or acute neurologic decline, unexplained developmental delay, progressive microcephaly, dystonia, spasticity, encephalopathy, irritability and focal motor findings. A lumbar puncture typically shows sterile cerebrospinal fluid (CSF) pleocytosis.11 64 65

Neuroimaging should be performed in individuals with a suspected diagnosis of an interferonopathy in the presence of neurologic symptoms. The initial workup may include MRI of the brain which identifies best white and grey matter changes.41 CT head should be considered when calcium-sensitive modalities on MRI are not available or not able to detect calcifications, since it is more sensitive for the detection of cerebral calcification.66 Risks and benefits of sedating a child for MRI brain should be considered.67 It is useful to have a baseline brain MRI to assess the severity and to monitor disease-associated complications; however, this is not a diagnostic prerequisite, especially for SAVI and CANDLE/PRAAS. Neuroimaging may be particularly helpful in patients with suspected AGS due to the dominant neurological phenotype which should be differentiated from mimickers of interferonopathies.

Basal ganglia or other intracerebral calcifications are overlapping neuroimaging findings for all three diseases68; they are more common, more severe and typically start earlier in life in patients with AGS compared with CANDLE/PRAAS, while calcifications are rare in SAVI.8 41 68 69 In addition, the presence of leukoencephalopathy is suggestive of AGS and typically starts early in life in AGS patients with severe disease; it is unusual in CANDLE/PRAAS or SAVI.11 70 71 Other supportive neuroimaging characteristics for AGS are early and rapid cerebral atrophy with or without calcifications, cerebral white and grey matter changes and Moyamoya disease.12 41 69 70 72–74 Intracerebral large vessel vasculitis or Moyamoya can be seen and is associated with SAMHD1 mutations.49 74–77

Additional workup for neurodegenerative diseases in patients with suspected AGS may also be considered. Lumbar punctures are not required to make the diagnosis of AGS but may support the diagnosis72 and characterise the immunological features of the central nervous system (CNS) inflammation, including the presence of lymphocytosis and raised levels of interferon-alpha (IFN-α), CXCL10 and CCL2 in the CSF.31 54 69 The CSF studies are most beneficial if a molecular diagnosis of AGS is not confirmed by genetic testing and provide support for additional molecular testing.72

Pulmonary manifestations

The presence of early onset interstitial lung disease (ILD) raises suspicion for SAVI, in particular in the context of unexplained systemic inflammation.1 7 46 56 61 Many patients with SAVI are reported to have lung involvement, mostly manifested as ILD, ranging from mild ILD with no respiratory symptoms to lung fibrosis. Also, alveolar haemorrhage is reported as the presenting feature in a few cases with SAVI.47 60 Although ILD is a major concern for patients with SAVI, it is rarely present in patients with CANDLE/PRAAS1 18 51 and not reported in AGS. Low radiation chest CT and pulmonary function tests (PFTs) are recommended modalities to screen for ILD.8 Lung biopsies may distinguish infectious from inflammatory disease but are not required to make the diagnosis of SAVI.7 46 60 61

Another significant pulmonary manifestation is pulmonary hypertension, which is a potentially life threatening and possibly underdiagnosed complication of CANDLE/PRAAS and AGS.1 12 78 While CANDLE/PRAAS and AGS are known to affect the vascular system, the full impact of systemic vasculopathy is currently undercharacterised. All patients with suspected CANDLE/PRAAS and AGS should undergo regular evaluation for pulmonary hypertension; echocardiography is recommended as a screening and monitoring tool.

Hepatic manifestations

Forty to eighty per cent of patients with CANDLE/PRAAS develop metabolic syndrome and hepatic steatosis, often in the first decade of life.1 In addition, patients may develop hepatosplenomegaly which could be due to extensive metabolic disturbance in fat processing.2 5 9 36 37 39 51 In an open-label trial in CANDLE/PRAAS, it is reported that baricitinib did not significantly improve hepatic steatosis in two patients with hepatic steatosis prior to baricitinib treatment nor prevent it in three patients with hyperlipidaemia at baseline pointing to the role of proteasome dysfunction in the aetiology of hepatic steatosis.1

In AGS, hepatosplenomegaly and/or transaminitis can be an initial presentation in the neonatal period when it resembles congenital viral infection.31 33 72 79 Patients can develop autoimmune hepatitis, the presence of liver-specific antibodies has been described.34 62 80

Transaminases should be evaluated at presentation and may be monitored as a marker for hepatic disease activity in patients with type I interferonopathies, although it should be noted they can also be elevated in CANDLE/PRAAS and AGS due to myositis.12

Information about the clinical features of hepatic involvement in patients with SAVI is limited. However, case reports of patients with SAVI presenting with hepatic disease, such as necrotising granulomatous hepatitis, cholestatic hepatitis and cholangitis and multiple biliary cysts are presented.58 81

Metabolic manifestation

Metabolic abnormalities are significant concerns in patients with CANDLE/PRAAS and patients can develop metabolic syndrome defined by Ford et al (presence of at least three of the following five criteria: hypertriglyceridaemia >110 mg/dL, low high-density lipoprotein cholesterol <40 mg/dL, abdominal obesity with waist circumference >90th percentile (sex specific), hyperglycaemia >110 mg/dL, systolic or diastolic blood pressure >90th percentile (age, height, sex specific)).82 In addition, these patients can have increased abdominal girth secondary to intra-abdominal fat deposition.1 51 The workup in CANDLE/PRAAS should include screening for metabolic abnormalities.

Patients with AGS may have hypothyroidism, often requiring replacement therapy, and insulin-dependent diabetes mellitus is reported.34 49 53 54 77 83–85 Other endocrine manifestations include central diabetes insipidus, growth hormone deficiency and adrenal insufficiency.34 83

Musculoskeletal manifestations

Myositis is a common feature of patients with CANDLE/PRAAS. It is usually patchy in distribution and can be demonstrated by muscle MRI.1 39 51 In addition, most patients with CANDLE/PRAAS will develop variable degrees of joint contractures in the hands and feet; these can be severely disabling.1 2 9 37 51 Myopathy is described in individual case reports in AGS.86 In AGS-affected patients, joint involvement can include a lupus-like arthritis, or progressive arthropathy with joint contractures.50 87 88 Articular involvement in SAVI is seen in one-third of the patients.8 Rheumatoid factor (RF) positivity was reported in majority of cases (57%)8 while anti-cyclic citrullinated peptide (anti-CCP) was not common in patients with SAVI but systematic testing has not been performed. Interestingly, the course of the arthritis in SAVI can be destructive, especially in childhood, when associated with RF and anti-CCP antibodies.7 43

Growth and development

Many children with chronic inflammation, including patients with type I interferonopathies, have lengths/heights and bone mineral density (BMD) that are below that of age-matched controls. Height and BMD are further decreased in the context of treatment with glucocorticoids. Weight percentiles can increase sharply with high doses of glucocorticoids, and this should be taken into consideration when evaluating weight.1

In addition to abnormalities in stature, patients with AGS can have significant developmental delay; after a subacute onset most individuals develop profound neurological regression and present with severe impairment in psychomotor development.22 23 34 Patients with AGS and CANDLE/PRAAS may also present with mild developmental delay5 22 51; these delays are not reported in patients with SAVI.8

Haematological manifestations

Cytopenias can occur in all three diseases due to temporary bone marrow suppression or homing changes and may correlate with disease activity.1 12 Cytopenias including autoimmune cytopenias occur more frequently in patients with CANDLE/PRAAS and AGS but are also seen in patients with SAVI.1 8 18 33 50 52 54 60 79 83 89 Thrombocytopenia in patients with AGS can be present during the neonatal period mimicking congenital infection, but also later during the course of the disease associated with other haematological abnormalities such as anaemia and leucopenia.19 79 Complete blood count with differential should be evaluated at presentation and may be monitored as a marker for disease activity in patients with type I interferonopathies.

Ophthalmological manifestations

Patients with type I interferonopathies can develop different types of ophthalmological manifestations. While patients with CANDLE/PRAAS can present with keratitis and/or episcleritis,2 18 51 patients with SAVI and AGS can develop glaucoma.8 54 76 Glaucoma has been reported in 6.3% of patients with AGS (up to 20.8% of patients with SAMHD1 mutations), with most cases presenting in the first 6 months of life, in patients who were not receiving glucocorticoids.34 76 Retinopathy has been described in AGS and SAVI but it remains unclear whether this occurs in the context of secondary mutations.90

Cardiac manifestations

Patients with AGS, especially those with mutations in TREX1, are prone to develop infantile-onset hypertrophic cardiomyopathy.31 34 There is an important risk of cardiac valve calcification in disease related to mutations in IFIH1 and ADAR.91

Other considerations

Immunodeficiency workup

Patients with known type I interferonopathies may have some degree of immunodeficiency, either due to chronic disease and cytopenias or due to treatment with immunosuppressants.92 Early manifestations may overlap with non-type I interferonopathy immunodeficiencies. Therefore, a basic immunologic workup should be considered even in the context of a confirmed diagnosis. The workup should include a history of infections and assessment of lymphocyte subsets and immunoglobulin levels, as a minimum.1 12 93

Infections in patients with CANDLE/PRAAS can be associated with the development of macrophage activation syndrome (MAS). Opportunistic infections in patients with other CANDLE/PRAAS mutations or SAVI and AGS are rare, although pneumocystis infection has been reported in a patient with SAVI who was not on any immunosuppressive treatment.89 Furthermore, defects in maturation of CD8+ cells are identified in patients with CANDLE/PRAAS,2 94 and in some patients with SAVI.8 57 89 Severe infections are reported in two patients with POMP mutations,94 which may be modified by additional genetic variants.

Points to consider 9–12: treatment focus on optimising inflammatory disease control

The goal of treatment is the control of the systemic and organ-specific disease manifestations and to manage complications of existing organ damage that are consequences of untreated disease.

Pharmacological treatment with Janus kinase inhibitors (JAKIs), particularly baricitinib, is widely used to treat patients with type I interferonopathies.1 95–98 The JAKIs are reported to be beneficial in controlling inflammatory symptoms and in preventing progression of end organ damage. Specifically, treatment with baricitinib resulted in a significantly lower daily diary score as well as significant reduction in glucocorticoid use in patients with type I interferonopathies in different open-label trials.1 95 In the study by Sanchez et al, none of the patients had achieved remission before initiating baricitinib treatment, and 50% of patients with CANDLE/PRAAS achieved lasting remission with no clinical symptoms, normalisation of inflammatory markers on baricitinib, all discontinued glucocorticoids. In addition, patients with CANDLE/PRAAS had improvement in myositis and cytopenias (haemoglobin, lymphocyte and platelets). Moreover, significant clinical improvement, including fewer vasculitis flares, prevention of skin involvement/progression of spontaneous amputations/the development of gangrene, and stabilisation of ILD by preserving pulmonary function, was achieved in patients with SAVI.1 However, to date, no patient with SAVI treated with JAKI achieved complete remission. Furthermore, JAKIs reduce IFN-α-mediated STAT-1 phosphorylation in a dose-dependent manner in patients with interferonopathy,26 56 thus demonstrating an in vivo effect of the JAKI on type I IFN signalling. The JAKIs, ruxolitinib and tofacitinib, are also reported as potential treatment options.44 56 59 98 Population pharmacokinetics and pharmacodynamic analyses in children treated with baricitinib showed a substantially shorter half-life in paediatric than in adult populations requiring more frequent dosing, and led to a proposed weight-based and estimated glomerular filtration rate-based dosing regimen to guide dose adjustments in the growing child.26 Doses of JAKI used to treat these conditions that were published are summarised in online supplemental table 4. A beneficial effect of JAKI on inflammatory disease manifestations is also observed in patients with AGS, including in an open-label trial. The treatment led to a decrease in interferon signalling genes expression scores and improvement of AGS-related symptoms, including neurologic disability, crying, sleep disturbances, irritability, seizures, fever and skin inflammation of the trunk, arms and legs.95–97 In all instances, pre-existing organ damage is irreparable (ie, the neurological manifestations) stressing the need for early treatment. In patients with AGS, treatment with HIV-1 reverse-transcriptase inhibitors reduced IFN scores, however, clinical benefit was not demonstrated99 and thus it is unclear if these drugs can be recommended.

Viral reactivation including BK viral reactivation has been reported in type I interferonopathy patients treated with JAKI.1 59 BK polyomavirus reactivation caused by therapeutic immunosuppression a commonly reported complication in renal transplant patients that can result in nephropathy and renal allograft loss. There is no proven treatment for BK nephropathy and management is limited to early detection and to controlling BK viral load by reducing the dose of immunosuppressive medications.100 101 Monitoring for BK viral load in blood and urine and renal function prior to initiation of JAKI, at baseline, and then routinely at each visit is recommended.

Other viral reactivations, such as herpes, are reported in CANDLE/PRAAS and SAVI1; however, there are insufficient data to routinely recommend anti-viral drug prophylaxis for patients with CANDLE/PRAAS and SAVI treated with JAKI. Similarly, in AGS, viral prophylaxis for patients on JAKI is not currently recommended.

Finally, the data from an open-label trial indicated that patients with AGS who are receiving baricitinib should be monitored closely for thrombocytosis, leucopenia and infection, especially those with underlying thrombotic risk factors or those who are receiving systemic glucocorticoids or immunosuppressive regimens,95 while no such events were reported in two other reports.96 97

Glucocorticoids are generally considered useful in CANDLE/PRAAS and SAVI patients with systemic inflammation, although their use is limited by toxicity.1 When used for a prolonged time, glucocorticoids cause serious side effects including growth arrest, truncal obesity, hypertension, glucose intolerance and osteopenia.102 Therefore, the lowest possible dose of glucocorticoids should be targeted for disease control.

There is generally no role for chronic glucocorticoids in AGS, as glucocorticoids do not improve the long-term neurological features nor outcome of AGS. However, short courses of glucocorticoids to treat acute CNS and non-CNS inflammatory manifestations, such as cytopenias and hepatitis, may be beneficial.

Points to consider 13–17: long-term monitoring and management focus on assessing inflammatory organ manifestations, minimising treatment-related toxicities, and encouraging general health measures, including vaccines, and fostering of self-management skills and medical decision-making

A multidisciplinary team approach to regular clinical follow-up is recommended and may include access to medical subspecialists, including a rheumatologist, geneticist, neurologist, ophthalmologist, pulmonologist, cardiologist, hepatologist, gastroenterologist, haematologist, immunologist, dermatologist, endocrinologist, nephrologist, and access to supportive services including a physiatrist, wound care specialist, psychologist, bone health specialist, physical therapist, dental/oral surgeon, dietitian, psychiatrist, rehabilitation care, orthopaedic care and social support services. With current treatment strategies the ultimate treatment goal in inflammatory diseases, namely inflammatory remission, can only be achieved in a subset of patients. Remission is mainly described in patients with CANDLE/PRAAS.1 The current treatment goal is therefore to reduce systemic and organ inflammation and to prevent or limit the development or progression of organ injury/damage. This requires treatment adjustments and close monitoring of disease progression. Table 4 provides general and disease-specific guidance for the monitoring of disease activity and assessment of organ damage. The monitoring should include (1) assessment of the level of systemic inflammation, and of growth and sexual development, (2) the assessment of general and disease-specific clinical signs and symptoms including the use of validated instruments when available,1 22 23 (3) monitoring of disease-specific organ manifestations and (4) monitoring of the development of autoimmune features (see online supplemental table 5 for autoantibody associations with organ-specific autoimmune manifestations in CANDLE/PRAAS, SAVI and AGS), cytopenias, treatment-related complications and infections (immunodeficiencies). Preliminary guidance regarding the monitoring of JAKI treatment (tables 3 and 4) is provided but may need to be adjusted as experience with treatment of interferonopathies grows.

Table 4

Evaluation of inflammatory disease manifestations and organ involvement with proposed interval monitoring

All patients should be evaluated at each visit for the presence of disease-specific symptoms and presence of systemic inflammation (table 4).

Chronic inflammation and chronic glucocorticoid treatment negatively affect bone health (eg, osteoporosis), growth (stunting) and development.1 These parameters should be monitored regularly, as well as cardiac (eg, hypertension) and ophthalmologic complications of chronic glucocorticoid use.

Patients with CANDLE/PRAAS should also be monitored for headaches, skin and musculoskeletal disease, development of metabolic syndrome (hypertension, hyperglycaemic and hepatic steatosis) and for development of primary pulmonary hypertension. Pulmonary hypertension can be insidious in onset. Although ILD is rare, it should be screened for at baseline and monitored as indicated by PFTs and low radiation chest CT. Ophthalmologic and dental assessment may be required in patients with eye inflammation and hypodontia and tooth eruption problems.1 2 5 9 18 36 37 39 51

Patients with SAVI may require wound care (including wound culture as necessary), and close assessment of ILD and the development of secondary pulmonary hypertension. Patients should be screened for systemic hypertension, otolaryngology, ophthalmology and dental disease at baseline and be followed as indicated. Patients should be instructed in self-care, including keeping peripheries warm, and in emergency management of acute ischaemic digits (eg, with, but not limited to, intravenous fluids, pentoxyphylline or intravenous vasodilators), prompt use of antibiotics if infection is suspected, and meticulous wound care.1 8 103

Patients with AGS are monitored for progression of neurological disease including gross and fine motor function and cognitive function using validated scales when available.22 23 Patients with SAMHD1 mutations require yearly MRI and MR angiography studies to screen for intracerebral artery disease (eg, Moyamoya).49 74 77 Patients should be monitored for the development of systemic hypertension, pulmonary hypertension and cardiomyopathy.78 Other complications include autoimmune hepatitis25 83 and autoimmune endocrinopathies, most frequently hypothyroidism.34 Other manifestations that can develop insidiously include glaucoma and epilepsy, and should be monitored as clinically indicated.76 104 Neurological tone abnormalities in non-ambulatory patients can lead to joint dislocation and scoliosis and should be monitored. Families should be instructed in prevention of skin complications, physical therapy, management of disturbed sleep–wake patterns and irritability commonly seen in AGS. Families can also participate in home stretching programmes, and appropriate positioning of children with tone abnormalities.

The heightened type I interferon-mediated autoimmune response contributes to the development of autoantibodies and autoimmune diseases105 (see online supplemental table 5). Antinuclear antibodies (ANA) are seen in up to 62.5% of patients with SAVI,8 in up to 42% of patients with CANDLE/PRAAS1 2 5 9 18 39 51 93 and 23% of patients with AGS.62 Moreover, antiphospholipid antibodies are present in patients with CANDLE/PRAAS, SAVI and AGS.1 7 62 Antineutrophil cytoplasmic antibodies (ANCA) are, intermittently, elevated in up to 71% of patients with SAVI and 18% of patients with AGS8 62; and RF positivity is reported in patients with SAVI (see above). Urinalysis for kidney dysfunction and screening for autoimmunity based on the disease symptoms are recommended as kidney disease is reported mostly in patients with AGS50 62 79 and SAVI.8 106 107 Antibodies associated with specific autoimmune diseases including autoimmune arthritis, pauci-immune glomerulonephritis, autoimmune cytopenias, thyroiditis and/or hepatitis have been described in CANDLE/PRAAS, SAVI or AGS with variable frequencies (online supplemental table 5). As it remains difficult to diagnose these diseases based on clinical symptoms, regular screening for autoantibodies as outlined in table 4 is currently recommended. Renal pathology prior to treatment with JAKI should be assessed by a baseline renal ultrasound and urine protein/creatinine ratio (or albumin/creatinine ratio).

All patients and families should have access to formal genetic counselling and may require social and other support. Supportive care, including adaptive equipment (eg, orthoses, walkers, wheelchairs, seating equipment), may be required.

Treatment during infections including COVID-19

Disease flares and progression can occur if immunosuppressive treatment is held108 and disease can flare in the context of an infection. Thus, any patient who develops an acute infection (or other complications) may require adjustment of immunosuppressive treatment (and/or institution of other supportive treatment), which should be conducted only under expert supervision. In line with these suggestions, recently published ACR guidance recommends continuing or initiating immunosuppressants when indicated in patients with paediatric rheumatic diseases in the context of exposure to SARS-CoV-2 or if experiencing asymptomatic SARS-CoV-2 infection. Immunosuppressants may be temporarily delayed or withheld if a patient has symptomatic COVID-19.109

Vaccination

Whether vaccination may trigger disease flares in interferonopathies is an important and currently unanswered question. There are no data suggesting that patients with CANDLE/PRAAS and SAVI develop disease flares to routine childhood vaccinations and the Task Force therefore recommended compliance with local regulations when patients are not treated with immunosuppressive treatments or glucocorticoids. No such consensus was achieved for AGS: the safety of vaccines in this population is not fully evaluated, and anecdotal reports of vaccine-induced neurological regression were concerns debated by the Task Force. No specific recommendation on vaccination for AGS was therefore possible. In line with the general EULAR guidance, the Task Force recommends avoiding live vaccines in patients with CANDLE/PRAAS, SAVI and AGS while on treatment with JAKI or other immunosuppressive medications.110 Treatment discontinuation can result in withdrawal flares. In general, we suggest following recommendations for other autoimmune and inflammatory rheumatic diseases,110 111 we however currently do not advise treatment adjustments for treatments recommended for the type I interferonopathies including JAKI.

RNA-based SARS-CoV-2 vaccines are not live vaccines, suggesting that they may be safe for immunosuppressed patients. Whether vaccines against COVID-19 have the potential to provoke a disease flare is unknown, theoretical concerns about disease flare in type I interferonopathies caused by RNA vaccines exist. There are currently no data to back specific recommendations.

Conclusion

The aim of these points to consider is to address the unmet need to provide guidance for healthcare professionals involved in the care of patients with the recently characterised type I interferonopathies, CANDLE/PRAAS, SAVI and AGS. A lack of high-level evidence is a limitation to these points to consider and reflect the challenges of studying novel, ultra-rare diseases. To address these challenges, the Task Force generated guidance statements based on results from a thorough SLR and on specialists’/experts’ opinions where evidence was lacking or was insufficient. The Task Force included various specialists with broad expertise in relevant clinical areas and representing different regions, disease interests and practice environments.

Important areas of future research are outlined in box 1. The cost and availability of genetic testing, interferon signature assays and JAKI treatment are substantial barriers that currently prevent optimised care for patients with interferonopathies. Furthermore, patients with the autoinflammatory interferonopathies CANDLE/PRAAS, SAVI and AGS live in many different countries and are managed in different healthcare systems. These points to consider address the multiple challenges of managing patients with these ultrarare diseases, by providing guidance on improving clinical recognition, support for decision-making on genetic testing as well as treatment and long-term management. These points to consider were developed to increase awareness of these diseases, and to standardise the level of care by characterising the diagnostic and therapeutic tools that can improve care.

Box 1

Research agenda

  • To define autoinflammatory disease outcomes, including:

    • Develop validated remission criteria for each disease including patient reported outcome measures.

    • Develop minimal disease activity criteria.

    • Validate sensitive biomarkers of progression of organ disease (including central nervous system).

  • To further assess efficacy of Janus kinase inhibitors (JAKI) and other type I IFN targeted therapies.

  • To assess long-term safety with treatment of JAKI.

    • Assess long-term effect of chronic BK viral reactivation.

    • Recommend monitoring guidance including frequency of BK viral loads measurements and management of BK viraemia.

  • To assess requirement of viral prophylaxis on JAKI.

  • To identify novel therapeutic targets and better treatments.

  • To validate an interferon signature to diagnose and monitor patients (eg, number of interferon response genes to include, sensitivity and specificity of score).

  • To evaluate the effect of vaccination in triggering or exacerbating disease activity in patients with type I interferonopathies while on or off treatments with immunosuppressive medications and/or glucocorticoids.

  • To identify new genetic causes for interferonopathies.

Ethics statements

Patient consent for publication

Ethics approval

This study does not involve human participants.

Acknowledgments

The Task Force gratefully thanks the librarian Darren Hamilton (London Health Sciences Center, London, Ontario, Canada) for his contribution to the systematic literature search, Brian Feldman, Hayyah Clairman and Natasha Naraidoo for their support of conducting the Delphi questionnaires using the Redcap platform and the European Alliance of Associations for Rheumatology and the American College of Rheumatology for financial and logistical support. This project is part of a series of 'points to consider' consensus efforts to standardise the diagnosis, treatment, and long-term monitoring of patients with the 3 major groups of known autoinflammatory diseases including (1) the IL-1-mediated diseases CAPS, TRAPS, MKD and DIRA, (2) the autoinflammatory interferonopathies CANDLE/PRAAS, SAVI and AGS and (3) the on early diagnosis, treatment, and long-term monitoring of inflammatory conditions with the potential progression to HLH/MAS. This research was supported in part by the intramural research program of the NIH institutes, NIAID (KCG, RGM), NHGRI and NIAMS and by the Deutsche Forschungsgemeinschaft, grant CRC237 369799452/B21 to MAL-K. We would like to acknowledge and are grateful for the generous and invaluable financial and organisational support from the Autoinflammatory Alliance and the systemic JIA foundation. The Autoinflammatory Alliance substantially contributed to an international meeting and workgroup organisation in August 2019 that developed the outline of the points to consider project. The funds for this preceding international meeting were largely due to patient fundraisers, online fundraising and the work of countless volunteers who made this project possible.

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

  • Handling editor Désirée van der Heijde

  • Twitter @drsezaozen

  • KCG, LL and MR contributed equally.

  • Correction notice This article has been corrected since it published Online First. The title has been corrected.

  • Contributors All authors contributed to the formulation of the points to consider. In details, the steering committee of the Task Force (RG-M, PB, AV, BF, ED) defined the research questions for the SLR. A systematic literature review was conducted by KCG, MR, LL with support from a librarian and epidemiologist (DH and DP) under supervision of a senior methodologist (ED). KCG, LL, MR extracted the data. RGM, PB and AV synthesised the results form SLR and Delphi questionnaires and generated draft statements. The manuscript was drafted by KCG, LL and MR and revised by RG-M, PB, AV, ED and BF. DA oversaw the proceedings and provided advice of this points to consider project as EULAR methodologist. All other authors participated in the Task Force meetings, in two pre-meeting Delphi questionnaires, suggested and agreed upon the research questions, read the final statements prior to the manuscript, discussed results and made contributions to the text. All authors approved the final version of the manuscript.

  • Funding This work was funded by European Alliance of Rheumatology Associations/American College of Rheumatology.

  • Competing interests EMF received NIH Grant “Clinical Outcomes in Aicardi Goutières Syndrome” (Grant number 5U01NS106845-02) and participated in an advisory board of Biogen. EPH: Spouse employed by Eli Lilly and received stock options in 2019 and 2020. SO1: NIH Grant “Clinical Outcomes in Aicardi Goutières Syndrome” (Grant number 5U01NS106845-02) and participated in advisory board of Biogen. CP received consulting and lecture fees from Novartis. BMF is associate editor of Arthritis and Rheumatology and member of the ACR guidance document committee. AV received grant support from Eli Lilly. PAB received grants from Roche and consulting and lecture fees and travel support from Roche, SOBI and Novartis. RG-M received study support under government CRADAs from Eli Lilly, IFM and SOBI. The consensus meetings were sponsored as EULAR/ACR projects.

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

  • Author note This article is published simultaneously in Arthritis & Rheumatology.

  • 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.