Article Text

Download PDFPDF

Consensus proposal for taxonomy and definition of the autoinflammatory diseases (AIDs): a Delphi study
  1. Eldad Ben-Chetrit1,
  2. Marco Gattorno2,
  3. Ahmet Gul3,
  4. Daniel L Kastner4,
  5. Helen J Lachmann5,
  6. Isabelle Touitou6,
  7. Nicolino Ruperto7
  8. On behalf of the Paediatric Rheumatology International Trials Organisation (PRINTO) and the AIDs Delphi study participants
    1. 1 Rheumatology Unit, Hadassah Hebrew University Medical Center, Jerusalem, Israel
    2. 2 UOSD Centro Malattie Autoinfiammatorie e Immunodeficienze, Istituto Giannina Gaslini, Genoa, Italy
    3. 3 Division of Rheumatology, Department of Internal Medicine, Istanbul Faculty of Medicine, Istanbul University, Istanbul, Turkey
    4. 4 Inflammatory Disease Section, Metabolic, Cardiovascular and Inflammatory Disease Genomics Branch, National Human Genome Research Institute, Bethesda, Maryland, USA
    5. 5 Division of Medicine, Centre for Amyloidosis & Acute Phase Proteins, University College London Medical School, London, UK
    6. 6 CEREMAIA, CHU Montpellier, INSERM U1183, Université de Montpellier, Montpellier, France
    7. 7 Clinica Pediatrica e Reumatologia–PRINTO, Istituto Giannina Gaslini, Genoa, Italy
    1. Correspondence to Eldad Ben-Chetrit, Rheumatology Unit, Hadassah Hebrew University Medical Center, Jerusalem, Israel; eldad{at}


    Autoinflammatory diseases (AIDs) are a relatively new family of disorders, defined about 19 years ago. Some of them are hereditary and some are not. The names given to these diseases do not follow any systematic guidelines, and sometimes the same disorder carries several names. The aim of this study is to refine the definition of AIDs and to provide some conventions for their naming. We focused mainly on monogenetic AIDs. Delphi technique, which enables consensus among a group of experts through internet and mail communication and questionnaires, was employed. After achieving 100% consensus among six members of a steering committee, the questionnaire containing AID definitions and the agreed-upon conventions were sent to 26 physicians and researchers working in the field of AIDs in order to gain broader support for the committee’s proposals. The committee proposed the following definition for AIDs: “Autoinflammatory diseases are clinical disorders caused by defect(s) or dysregulation of the innate immune system, characterized by recurrent or continuous inflammation (elevated acute phase reactants-APR) and the lack of a primary pathogenic role for the adaptive immune system (autoreactive T-cells or autoantibody production).” Several rules were defined for guiding the naming of these diseases among which are: abandoning eponyms and preferring the name of the gene over its encoded protein. The new definition for AIDs allows inclusion of clinical disorders mainly associated with defects in the innate immune system. The new conventions propose names with clinical meaning and in some cases even clues for treatment.

    • autoimmune diseases
    • fever syndromes
    • familial mediterranean fever
    • inflammation

    Statistics from

    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.

    Taxonomy is the science of naming. It is relevant to all fields of biology in which we name plants, animals, objects and diseases. In medicine, naming of diseases or syndromes has a special importance since it can give some clue about the nature of the clinical condition, its clinical features, pathogenesis and sometimes even an approach to treatment. Naming is also important for accurate and effective communication among different health disciplines. However, medical disorders have not been named in a standard way.1 Physicians, who treat patients with a particular disorder or face a new clinical condition, are often the first to propose a name for the disease. Expert working groups may later revise the names to improve their usefulness.

    Names of medical disorders are often derived from one or a combination of the following sources: genetic basis or biochemical defect; geographic spread or by eponyms. The main drawback of many names is the lack of a clinical meaning that could help the novice to understand the origin of the disease or recognise its clinical characteristics.

    The autoinflammatory diseases (AIDs) are a group of medical disorders, derived from defects or dysregulation of the innate immune system.2 This family of diseases was established in 1999 following the identification of the genes underlying two recurrent fever syndromes: familial Mediterranean fever (FMF)3 4 and TNF-receptor-associated periodic syndrome (TRAPS).5 Over the last 19 years, more and more diseases have been classified among this group of disorders, some of which may not fit well with the classical definition of the AIDs. Moreover, many of them were given names with no systematic guidelines or rules. In some cases, the same disease carries several names (table 1).6–48 This has led to a chaotic situation in naming these clinical disorders and has called for a better standardisation of this field. This need is accentuated by recent progress in next generation sequencing techniques, which have led to an increasing capability to identify new genes and new syndromes, expanding the spectrum of AIDs.

    Table 1

    Current name of the disorder (in bold) and additional names (normal characters) derived from the literature

    Indeed, following the International Society for Systemic Autoinflammatory Diseases (ISSAID) meeting in Lausanne, in 2013, a mandate was given to one of us (E B-C) to undertake a preliminary consensus-based exercise for the following aims: (1) to refine the definition of the ‘autoinflammatory diseases’; (2) to provide some rules and new proposals for naming this current group of clinical conditions and those that will be identified in the future.


    In order to find the different definitions proposed for AIDs over the years, we searched the MEDLINE/PubMed Central (PMC) from 1998 to January 2016, using the MESH search term: ‘autoinflammatory diseases’ (online supplementary figure S1). In order to find the names used for each AID, we took one of their current names as depicted in table 1 and searched for papers where they were first reported. Then, we searched for reviews on these items to find additional synonymous names. Table 1 is based on a list of AIDs published by one of the authors (IT),49 properly integrated and updated during the consensus process and finally approved by all the steering committee members. It focused—mainly—on monogenic disorders.

    Supplementary file 1

    For choosing the best definition for AIDs and the most appropriate name for each AID, we have used the Delphi technique, which enables consensus among a group of experts through mail communication.50 The Delphi method is essentially a series of questionnaires involving several steps, each of which is based on the results of the previous step. The process stops when consensus of at least 80% of the participants on each item is reached.51

    An ad hoc steering committee of six clinicians and researchers from six different countries who are working in the field of autoinflammation was established.

    The first Delphi questionnaire was built through sending broad and open-ended questions in order to elicit different opinions from the panellists about the current definitions and names of AIDs.

    Once received, the replies from the panellists were analysed to generate a series of statements that were employed as the basis for follow-up questionnaires that were sent back to the individual participants. In each subsequent questionnaire, the panellists were also provided with the overall results (responses) of the previous questionnaire from all the members. After achieving 100% consensus among the steering committee members, the questionnaire containing the AIDs definitions and the agreed-on names of AIDs were sent to 26 physicians and researchers working in the field of AIDs around the world. They were identified in the Paediatric Rheumatology International Trials Organization (PRINTO) mailing list through their high active participation in the Eurofever registry.52 53 The aim of this step was to gain broader support for the committee’s proposals and to consider changes once a name was rejected by or was not acceptable to more than 80% of the participants of the large group of AIDs experts. Delphi survey implementation was conducted by PRINTO.54


    AIDs proposed definition

    The literature review disclosed 536 papers of which only 7 specifically dealt with the definitions of AIDs5 49 55–59 (online supplementary figure S1). The first definition for AIDs was proposed by the NIH group in 1999.5 This definition was as follows: ‘The autoinflammatory syndromes are systemic disorders characterised by apparently unprovoked inflammation in the absence of high-titre autoantibodies or antigen-specific T lymphocytes’. This definition was based mainly on the two diseases whose related genes had then been identified: FMF and TRAPS.3–5 Since in both diseases the flares appeared mostly spontaneous, the definition included the word ‘unprovoked’. The definition stresses the lack of involvement of the adaptive immune system in these disorders, since no autoantibodies or autoreactive T-cells were involved.

    Seven years later McGonagle and McDermott suggested another definition: ‘AIDs are characterised by self-directed inflammation, whereby local factors at sites predisposed to disease lead to activation of innate immune cells, including macrophages and neutrophils, with resultant target tissue damage. For example, disturbed homeostasis of canonical cytokine cascades (as in the periodic fevers), aberrant bacterial sensing (as in Crohn’s disease), and tissue microdamage predispose one to site-specific inflammation that is independent of adaptive immune responses’.55 The authors proposed that immunological diseases ought to be conceived as a continuum with ‘pure monogenic autoinflammatory diseases’ at one end and ‘pure monogenic autoimmune diseases’ at the other. This definition is relatively complex, but explicitly invokes innate immunity and widens the spectrum of AIDs.

    Later, several other definition or refinement were proposed.49 56–58 In a recent study, de Jesus et al provide an outstanding classification of AIDs strictly based on their pathophysiology.59 However, the authors do not propose a new definition for the AIDs.

    Given the proliferation of AID definitions, with sometimes conflicting concepts, the steering committee agreed to adopt the first and original definition with minor modifications: ‘Autoinflammatory diseases are clinical disorders caused by defect(s) or dysregulation of the innate immune system, characterised by recurrent or continuous inflammation (elevated acute phase reactants (APR)) and the lack of a primary pathogenic role for the adaptive immune system (autoreactive T-cells or autoantibody production)’.

    This definition emphasises the essential fact that the disorders are caused by defects in the innate immune system and are continuous or recurrent. The word ‘unprovoked’ has been deleted since in many cases there is a trigger for the acute flares.

    The steering committee is aware that diseases such as PLCG2-associated antibody deficiency and immune dysregulation (PLAID) or Heme-oxidised IRP2 ubiquitin ligase1 (HOIL-1) deficiency, traditionally included among the AIDs, will not be part of this group, because they may contain components of the adaptive immune system such as autoantibodies.60 The ‘Interferonopathies’ include some disorders also manifesting autoantibodies. However, the consensus seemed to be that for disorders like Aicardi-Goutières syndrome in which nucleic acid sensing is primarily intracellular, autoantibodies usually play a minor role in disease pathogenesis, and thus the autoinflammatory designation may still be appropriate. In their recent review, Rodero and Crow propose that ‘type I interferonopathies can reasonably be considered as autoinflammatory in origin, with ‘spillover’ into autoimmunity in some cases’.61 The group of ‘typical’ autoimmune diseases includes disorders affecting primarily or only the adaptive system such as systemic lupus erythematosus, Hashimoto thyroiditis, DNAse deficiencies and autoimmune lymphoproliferative syndrome.

    AIDs proposed nomenclature

    The current names for AIDs bring several problems and issues, which called for a new approach and nomenclature modification; many AIDs possess more than a single name (FMF—seven different names, TRAPS—three and so on) (table 1 and online supplementary table S1); different clinical presentations are associated with similar sequence alterations in the same gene, for example, Muckle-Wells syndrome (MWS), familial cold autoinflammatory syndrome (FCAS) and neonatal onset multisystem inflammatory disease (NOMID) are associated with NLRP3 gene whereas FMF and pyrin-associated autoinflammation with neutrophilic dermatosis (PAAND) are associated with MEFV gene. In addition, discussion arose about several topics briefly summarised herein: In naming AIDs, should we use the name of the gene or that of the encoded protein (MEFV or pyrin)? Should we include typical clinical features or just genetic attributes? Should historical names be retained?

    Supplementary file 2

    Following more than six cycles of Delphi questionnaires and oral discussions among the six members of the steering committee with further involvement of the 26 AIDs experts around the world—a consensus of at least 80% was reached for the nomenclature of the diseases shown in tables 1 and 2.

    Table 2

    Results from the Delphi questionnaires for consensus on nomenclature

    General conventions

    The proposed names for AIDs have been established according to the rules and suggestions outlined in box 1.

    Box 1

    Recommendations for naming autoinflammatory diseases

    1. Try not to change wherever the name is appropriate.

    2. Avoid names of persons or geographical spread of disease (eponyms).

    3. Include the genetic basis (name of the gene) of the disease where it is known (prefer the name of the gene over the name of the encoded protein unless the name of the gene is not accurate or meaningless).

    4. Include key clinical features where appropriate.

    5. Shorten the name as much as possible.

    6. Choose a name that is as clear as possible.

    7. In diseases where our knowledge about the pathogenesis is still limited, leave the previous name (periodic fever, aphthous stomatitis, pharyngitis and adenitis).

    8. In diseases with different phenotypes but mutations in the same gene, use a general ‘roof’ name with subtypes (pyrin-associated autoinflammatory disease, NOD2).

    9. When the clinical features seemed to be ‘continuous’ give a general name (‘roof’ name) and classify the various presentations according to their phenotypic severity (NLRP3-associated autoinflammatory disease, mevalonate kinase deficiency).

    In many diseases, the course of the disease is episodic with frequent attacks and attack-free intervals. When the frequency of the attacks is relatively regular (as is the case with Periodic fever, aphthous stomatitis, pharyngitis and adenitis (PFAPA) and sometimes with mevalonate kinase deficiency (MKD), we preferred the term ‘periodic’. When the attacks do not have a regular pattern, we suggested the word ‘recurrent’.

    In the past, both terms, ‘periodic’ and ‘recurrent’, have been used interchangeably but now the term ‘periodic’ remained in the names of three conditions only: cryopyrin-associated periodic fever syndrome (CAPS), TRAPS and PFAPA. In TRAPS, we decided to keep the original name ‘periodic’, although its flares are recurrent rather than periodic. In CAPS, we propose a new name (NLRP3-AID) which does not contain the word ‘periodic since the attacks are not periodic’. Thus, we strongly suggest using the more appropriate terms in naming disorders in the future.

    As a general rule, we tried to use names containing aetiopathological (genetic) features of the disease and where appropriate or possible, to add a significant clinical characteristic of the syndrome. Thus, we left the name TRAPS without change, since it consists of its genetic aetiology (mutations in TNFRSF1A gene) and characteristic clinical features (periodic (recurrent) fever). On the other hand, the name hyper IgD syndrome (HIDS) was abandoned since it is an absolutely inaccurate name: serum IgD is not always elevated in these patients while it may be elevated in other AIDs. Therefore, this name was replaced by MKD based on our knowledge of the gene involved, mevalonate kinase (MVK). In this way, a physician or researcher who approaches these names for the first time may have immediately a basic understanding of the disorder and sometimes even a clue to the potential treatment.

    In cases where the choice was between using the name of the gene associated with the disease or the protein encoded by the gene, we preferred the name of the gene over that of the protein unless the former was meaningless. A typical example is the choice of NLRP3 gene over cryopyrin despite the tendency of some clinicians to stay with the former term CAPS. Fortunately, in many cases, the name of the gene and the encoded protein are the same (MK, NOD2) making the choice easier. However, this was not the case with the MEFV gene and pyrin where the name of the protein was chosen, as will be discussed later.

    In our proposals for new taxonomy of AIDs, we tried to avoid the use of names of persons (such as Nakajo Nishimura syndrome) or geographical distribution of the disease (such as Guadeloupe fever) or names with unusual meaning (such as ‘Cherubism’).

    Specific names

    In the case of CAPS—which encompasses three clinical entities (FCAS, MWS, NOMID/CINCA), the committee has proposed using a single name; NLRP3-associated autoinflammatory disease (NLRP3-AID). Since the various disorders reflect different levels of phenotypic severity of the same disease, it was suggested to add the adjectives: mild, moderate and severe phenotypes, instead of using the historical names FCAS, MWS and CINCA/NOMID, respectively (table 1 and online supplementary table S1).

    In familial cold autoinflammatory syndrome 2 (FCAS2) (Guadeloupe fever), different families present with different phenotypes.36 Since the gene associated with the disease (NLRP12) is known, the committee decided to name this syndrome NLRP12-associated autoinflammatory disease (NLRP12-AID).

    In the case of MEFV-associated diseases, the question raised was as follows: should we use the old name FMF or ‘atypical FMF’ for all syndromes associated with mutations in the MEFV gene even if they have totally different clinical manifestations? Alternatively, should we find a different way to classify these disorders? The committee chose to use a general name (as a ‘roof‘) ‘pyrin-associated autoinflammatory diseases’ (PAAD) which includes all diseases associated with pyrin defects or MEFV mutations. Under this general term, there are subtypes of disorders with different names, according to their clinical presentation or genetic features, such as PAAND, FMF and so on62 (figure 1). Although it is preferred using the name of the gene over the name of the encoded protein, in the case of FMF, the protein pyrin was chosen rather than the MEFV gene. One of the reasons was that the name MEFV, which was coined to denote its association with FMF, is no longer accurate, since it may lead to totally different AIDs, such as PAAND and CRMO-like disorder. In addition, we did not change the name of FMF, although sometimes it is neither familial nor restricted to the Mediterranean basin and in rare cases, it may even be without a documented fever. Most members of the steering committee thought that FMF is a well known and defined entity and that changing the name would cause discomfort and confusion among the AID clinical community. The name FMF remained under the ‘roof’ of ‘pyrin-associated autoinflammatory diseases’ (PAAD) as a clinical entity which is restricted mainly to Middle Eastern patients or to patients elsewhere, whose disease is associated with exon 10 mutations.63

    Figure 1

    The group of diseases associated with MEFV sequence alterations. The ‘roof’ name is a general name whereas the subtypes are more specific and meaningful. AIDs, autoinflammatory diseases; CNO, chronic non-bacterial osteomyelitis; FMF, familial Mediterranean fever.

    Regarding Mevalonate kinase disorders the committee suggested leaving MKD as a general name with the option of adding ‘mild’ for those with HIDS and ‘severe’ for those with mevalonic aciduria.64 In rare cases, where the patient with MKD has also retinitis pigmentosa or porokeratosis, it is suggested to mention these manifestations in addition to MKD (table 1 and online supplementary table S1).

    The name NOD2-associated granulomatous disease was chosen by the committee for the three phenotypes: Blau syndrome, familial sarcoidosis and familial Crohn’s disease. Since all these syndromes are characterised by granulomas, this feature was included in the name. Nevertheless, an option was offered to add inflammatory bowel disease (IBD) in cases where the intestines are the main site of involvement for example, NOD2-associated granulomatous IBD (formerly called familial Crohn’s disease).

    The name for CRMO was replaced by the name chronic non-bacterial osteomyelitis (CNO). The reason for that was the presence of many cases where the disease was neither recurrent nor multifocal. Furthermore, the new name emphasises the main feature of the disease, non-bacterial osteomyelitis. Since this clinical entity may be associated with mutations in various genes, it is optional to add the name of the gene when it is known. For example, in case the gene involved is LPIN2, it can be marked as LPIN2-CNO (previously known as Majeed’s syndrome). In adults, patients with sporadic CNO are usually diagnosed with SAPHO, a symptom complex of synovitis, acne, pustulosis, hyperostosis and osteitis.65

    Chronic atypical neutrophilic dermatosis with lipodystrophy and elevated temperature syndrome (CANDLE syndrome) gained a new name: PSMB8-PRAAS—where PSMB8 stands for Proteasome Subunit Beta 8 and PRAAS for PRoteasome-Associated Autoinflammatory Syndrome. This name replaces also the eponym Nakajo Nishimura syndrome, and JMP which stands for Joint contractures, Muscle atrophy, microcytic anaemia and Panniculitis-induced lipodystrophy. The name PSMB8-PRAAS consists of the genetic aetiopathology of the disorder but does not include any clinical feature of the disease.

    CARD14-associated disease is usually characterised by psoriasis with or without pustulosis. Therefore, the name was refined to be CARD14-associated psoriasis.

    Since the three variants of IL-10 deficiency are always associated with inflammatory bowel disease, the committee proposed a single name as IL-10 deficiency-associated IBD.

    The names, deficiency of the interleukin 1 receptor antagonist (DIRA), deficiency of the interleukin 36 receptor antagonist (DITRA), pyogenic arthritis, pyoderma gangrenosum and acne (PAPA) and PFAPA remained unchanged since they already conform to our naming conventions. However, the letter P in the abbreviation ‘PAPA’ now stands for the name of the gene PSTPIP1 rather than ‘Pyogenic’ and, therefore, the new name is ‘PSTPIP1-associated arthritis, pyoderma gangrenosum and acne’ (PAPA).

    The name ‘Cherubism’ was derived from the Biblical ‘cherub’ (plural cherubim) who has four faces of different species and several pairs of wings,. For most physicians, this name does not mean much and therefore, the committee proposed the name SH3BP2 deficiency with multilocular cystic disease of the mandibles (SDCM). This name gives the aetiopathological basis of the syndrome with the main clinical feature of fibrous dysplasia of the mandibles.

    Finally, the name ‘Schnitzler syndrome’ also remained as an historical one, since its pathogenesis is still obscure and its relationship with NLRP3 mutations has not been established.66 A proposal to convert the name of the syndrome to a clinical description: ‘late onset gammopathy with recurrent urticaria and fever’ did not gain support from most of the committee members.


    The definition of AIDs has changed over the years in order to accommodate the new diseases discovered since 1999—the year the term was first proposed.2 5 49 55–58 Widening the scope and spectrum of definition of AIDs resulted in the inclusion of disorders with additional defects in the adaptive immune system such as PLAID or HOIL-1 deficiency. Most defects in the immune system may affect primarily either the innate or the adaptive arm. However, it is becoming increasingly obvious that the innate immune system almost always has an effect on the adaptive system. This leads to the situation that there are disorders that do not fit neatly into the ‘pure’ autoinflammatory or autoimmune categories and reside actually in a ‘grey zone’ between these two groups. In order to include these disorders with the typical AIDs under the same ‘rafter’, Peckham et al offered the term ‘autoinflammatory-immune diseases’.67 We believe that this new name may lead to confusion since all the disorders caused or related to defects in the immune system can be classified under this wide term with no clear categorisation. The interferonopathies are clinical disorders caused by defects in the innate response, leading to inflammation after DNA sensing. The activation of cells of the adaptive immunity is a secondary effect of this condition and seems to play a minor role in their pathogenesis. Therefore, they may create the bridge which fits the concept that the AIDs, and the autoimmune diseases are actually in the same spectrum of immune disorders. This continuum model is further supported by the recent discovery of the innate lymphoid cells. These cells are defined by differential expression of cell-surface markers and are activated by neuropeptides, cytokines and other alarmins.68 Their specialised distribution in lymphoid and non-lymphoid tissues, coupled with their functional heterogeneity, has provoked a fundamental reassessment of how they integrate innate and adaptive immune responses.

    As already mentioned, many of the current names of AIDs were not appropriate, inaccurate or lack any clinical meaning. Therefore, an attempt to establish new conventions for naming them was really needed.

    The conventions (box 1), and the ensuing proposals (table 1 and online supplementary table S1), call for using the name of the gene associated with the disease when it is known rather than the encoded protein. In these cases, demonstration of functional significance of the identified sequence alteration is mandatory. The main advantage of using the name of the gene is that such a name gives the physician a clue about the pathogenesis of the disease and sometimes even about a potential treatment. Moreover, it may allow definite diagnosis using genetic testing. However, it should be borne in mind that including the gene in the name of the disease may pose a problem in cases where the clinical features of the patient are compatible with a certain diagnosis while no expected sequence alteration is found. Thus, the main drawback of using the name of the gene is that definite diagnosis can be made only by genetic testing.

    In cases where the clinical features and the genetic testing results are in accord, the name is appropriate and the diagnosis is correct and definite. When there is a clearly pathogenic mutation but the clinical features are completely incompatible with the expected diagnosis, one should consider a different disease with a different name. This situation is illustrated by the case of the MEFV mutation S242R, which causes neutrophilic dermatosis. The name of this disease is not ‘FMF’ or ‘atypical FMF’ despite the fact that there are MEFV mutations—but ‘PAAND’. Similar approach may be applied in the case of PSTPIP1 with the new mutation and different clinical presentation.69 We suggest here a ‘roof’ name: PSTPIP1-associated AIDs with two subtypes: PAPA and PAMI (PSTPIP1-associated myeloid-related proteinaemia inflammatory syndrome). However, we cannot add this approach to table 1 since it was not discussed in the Delphi questionnaires among the large group of participants.

    When the clinical features are typical for a certain disease (eg, FMF) and yet no genetic support for this diagnosis is found, one can denote this medical condition as an FMF-like disease. However, a better choice would be to leave the case as an undefined AID until mutations in other genes are found or additional explanations for the disease are given. The reason is that clinical features typical for one AID may be associated with mutations in different genes. For example, in a recent report, Karacan et al described two Turkish families in whom four patients presented with typical clinical features of FMF.70 Genetic analysis performed in these patients failed to show MEFV mutations. However, total exon sequencing revealed that two patients were homozygous for mutations in MVK and the two other patients carried mutations in the TNFRSF1A gene. These cases illustrate the difficulties in making a diagnosis of AID based on clinical features only and justify the proposal to use the gene in naming AIDs wherever it is known.

    The way we proposed naming CAPS and FCAS 2, namely NLRP3-AID and NLRP12-AID, respectively, may pave the way for naming future disorders to be discovered or identified among the other members of the large family of NOD-like receptors. Similarly, PSMB8-PRAAS, the name which was proposed to replace CANDLE syndrome, JMP and NNS, may also serve as an example for naming additional proteasome associated diseases to be discovered, just by changing their number. In fact, Brehm et al recently described several cases that carry mutations in PSMA3 (encodes α7), PSMB4 (encodes β7), PSMB9 (encodes β1i) and proteasome maturation protein (POMP).71

    Unfortunately, the current study did not include many other monogenic AIDs such as those associated with ADA2, NLRC4, NLRP1 genes or X linked inhibitor of apoptosis deficiency and (SLAM)-associated protein deficiency.72 The reason is that we limited ourselves mainly to the basic list reported by Touitou et al.49 However, we hope that the conventions we propose herein may help modifying names of additional diseases—old and new—when, they do not follow the rules suggested.

    For this project, we used the Delphi technique which allowed discussion via an ad hoc web-based system developed by the PRINTO staff under the supervision of NR who has an extensive expertise in consensus formation methodologies. The PRINTO system allowed remote interaction between the participants who had the possibility to share written comments with the other participants in a transparent and traceable way. A limitation of the current work was that for lack of funding we could not conduct a formal nominal group technique which is a guided face-to-face discussion and interaction, among small groups of experts. However, the additional discussion of the ad hoc steering committee consensus proposal by another group of 26 worldwide experts in the field of AIDs further strengthens these proposals.

    In conclusion, the currently proposed rules for nomenclatures of AIDs are expected to allow a better organisation of these groups of immune diseases. However, taxonomy is a dynamic process and some of the proposed names may be changed in the future as we gain a better knowledge about their pathogenesis. The proposed taxonomy may gain a broader consensus following an effective communication with other societies such as the International Union of Immunological Societies Expert Committee.


    We thank the staff from the PRINTO office (Eugenia Mosci, Luca Villa and Roberto Cavanna).



    • Handling editor Josef S Smolen

    • Contributors All authors contributed equally to the planning and conduct of the study. Their placement in the authors’ list is dictated by the alphabetic order of their family names. The first version of the present manuscript was written by EB-C, MG and NR and then revised critically by all the remaining coauthors (AG, DLK, HJL, IT).

    • Funding The development and coordination of the Delphi survey has been funded with the research budgets of Istituto Giannina Gaslini; no external entity such as pharmaceutical companies has been involved at any stage of the project.

    • Competing interests For EB-C, MG, AG, DLK, HJL, IT: Nothing to disclose for this manuscript. NR received honoraria for consultancy of speaker’s bureau from the following pharmaceutical companies since last 5 years: Abbott, AbbVie, Amgen, Biogenidec, Astellas, Alter, AstraZeneca, Baxalta Biosimilars, Boehringer, BMS, CD Pharma, Celgene, CrescendoBio, EMD Serono, Hoffman La Roche, Italfarmaco, Janssen, MedImmune, Medac, Novartis, Novo Nordisk, Pfizer, Rewind Arms, R Pharma, Sanofi Aventis, Servier, Sinergie, Takeda, Vertex, UCB Biosciences GmbH. The Gaslini Hospital, which is the public Hospital where NR works as full time public employee, has received contributions from the following industries: Abbott, BMS, ‘Francesco Angelini’, GlaxoSmithKline (GSK), Hoffman La Roche, Italfarmaco, Janssen, Novartis, Pfizer, Sanofi Aventis, Schwarz Biosciences, Sobi, Xoma, Wyeth. This money has been reinvested for the research activities of the hospital in a fully independent manner without any commitment with third parties.

    • Patient consent Not required.

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

    • Data sharing statement There are no data to be shared besides the manuscript.

    • Collaborators List of additional participants to the Delphi study: Ivona Aksentijevich, Bethesda, USA; Jordi Anton, Barcelona, Spain; Juan I Arostegui, Barcelona, Spain; Karyl S Barron, Bethesda, USA; Luca Cantarini, Siena, Italy; Fatma Dedeoglu, Boston, Massachusetts , USA; Erkan Demirkaya, London, Ontario, Canada; Dirk Foell, Muenster, Germany; Joost Frenkel, Utrecht, Netherlands; Philip J Hashkes, Jerusalem, Israel; Veronique Hentgen, Paris, France; Michael Hofer, Lausanne, Switzerland; Tilmann Kallinich, Berlin, Germany; Isabelle Koné-Paut, Paris, France; Jasmin B Kuemmerle-Deschner, Tuebingen, Germany; Ronald Laxer, Toronto, Ontario, Canada; Avi Livneh, Ramat Gan, Israel; Alberto Martini, Genoa, Italy; Laura Obici, Pavia, Italy; Seza Ozen, Ankara, Turkey; Dorota Rowczenio, London, UK; Ricardo Russo, Buenos Aires, Argentina; Yael Shinar, Tel Hashomer, Israel; Natasa Toplak, Ljubljana, Slovenia; Yosef Uziel, Kfar Saba, Israel; Marielle van Gijn, Utrecht, The Netherlands.