Opinion: Why paediatric rheumatologists need to understand inborn errors of immunity

Abstract

Tables

Table 1. The “open book of medicine”: timeline to diagnosis, with evolving & various features over 14 years [patient No 6; table S5 and figure S5-A,B; Zhou et al. NEJM 2014;370:911–20 (54)].

Age	Features/Dg	Investigations	Treatment	Complications
6 months	Recurrent (2–6 weeks), transient asymptomatic crops of erythematous dermal nodules/plaques	Skin bx - NLCH/GEH
18 months	Recurrent, transient stroke episodes	Brain MRI – thalamic infarct	Aspirin (transient.)
		Extensive investing. (all N)
4 years	Evans Sy.	DCT + ve	Steroids
		Extensive immunological	Rituximab
		Investigations (all N)
6–8 years	Hypersplenism	Liver bx - HNRH	Splenectomy
	Portal hypertension		Spleno-renal shunt	Arterial Hypertension
	Oesophageal varices		HSCT discussed	IVIg replacement
	? complex MSAID
9–10 years	Hepato-pulmonary Sy.		Orthoptic Liver transplant	Shunt patency problems/TIPSS
	? Immune dysregulatory disease			Liver VOD
	? Pro-thrombotic (hypercoagulability) state			IVC obstruction/Pleural effusions
	? Endothelial disorder			Anaemia/CMV reactivation
				Heart failure/RC
12–13 years	Relapse Hepato-pulmonary Sy.		? Palliation/2nd OLT/HSCT
14 years	ADA2 deficiency (novel disease)			Died

Full text

Introduction

Paediatric rheumatologists are referred a wide variety of patients affected by diseases characterised by chronic or recurrent inflammation of the musculoskeletal system and connective tissues. The “classic” rheumatological diseases include rheumatic fever, juvenile idiopathic arthritis (JIA), the spondyloarthropathies, systemic lupus erythematosus (SLE), juvenile dermatomyositis (JDM), scleroderma, and the vasculitis syndromes. However, a significant number of children present with unexplained fevers, rashes, and various musculoskeletal complaints which do not fit any defined pattern, thus “making the matter of accurate differential diagnosis extremely important” (1).

Continued paediatric rheumatology education is of utmost importance, primarily for the benefit of patients under their care (2, 3). To that effect we aim to raise the awareness of, and to bring the field and the “nomenclature” of various immunodeficiency states closer to the attention of paediatric rheumatologists.

Lessons in history

Rheumatology, including paediatric rheumatology, was recognised as a subspecialty in the 1950's and has always been closely linked to the field of immunology, not least because rheumatic diseases are often loosely referred to as “autoimmune” conditions (1, 4). Indeed, alongside host defence against invading microorganisms, maintaining the immunologic tolerance to self-antigens is the other major function of the immune system. Preventing autoimmune tissue damage from an “overactive” immune system requires closely integrated and coordinated function of the various components of both the innate [complement system, myeloid cells, innate lymphoid cells (ILCs) - including natural killer (NK) cells] and the adaptive immune system (T and B lymphocytes), linked by numerous messenger (e.g., cytokine) pathways.

Today we understand that many of the clinical features that patients referred to paediatric rheumatology services present with, are due to underlying genetic defects of the immune system functions (5). In the initial report of the rare diseases “resulting from a failure to produce the effectors of the immune response - i.e., antibodies and sensitized lymphocytes” - termed primary immunologic deficiency (PID), the authors draw attention to the “associated disorders” - malignancies and autoimmunity. They stated that “the high incidence of autoantibodies, with or without autoimmune disease, in patients with immunodeficiency has been well documented, but the reason for the association is not known”. They speculated that it might directly relate to a genetic abnormality or to effects of the immunodeficiency, such as latent virus infection (6). Initially, autoantibodies were observed to be mainly directed to antigenic determinants of the formed elements of the blood such as red blood cells, platelets, and neutrophils, and were thought to be only rarely organ-specific (7). An “unidentified infectious agent” was suggested as a plausible explanation to the “baffling propensity” of polyarthritis occurring in some patients with agammaglobulinaemia (8), which was vividly highlighted by the Journal of Pediatrics in its “50 Years Ago” vignette (9). In the context of immunodeficiency, autoimmunity has been perceived not as a classical breakdown of tolerance to self-antigens but rather as tissue damage incurred as the host attempts to rid itself of foreign immunogens. Inability of the host's inherently defective immune system to completely eradicate microbial pathogens and their antigens through the usual immune pathways has been proposed to result in a compensatory, often exaggerated and protracted inflammatory response by alternative and less effective immune pathways damaging not only infected cells but also the surrounding tissues (10, 11). A seminal report confirmed that autoimmune and inflammatory manifestations frequently complicated PIDs in a large cohort of patients from the French national registry, including autoimmune cytopaenias (31%), gastrointestinal (24%), skin (14%), and rheumatic disorders (13%) (12). The authors proposed various triggers for autoimmunity and/or inflammation that included exacerbated production of type I interferon (IFN) and/or interleukin (IL)-1 production, defective negative selection of self-reactive T and B lymphocytes with high affinity, defective editing of the B-cell receptor in the periphery, defective peripheral (self)-antigen–induced cell death, gain of function (GOF) of B- or T-cell activation/effector molecules, defective regulatory T (T-reg) cells, homeostatic expansion of self-reactive T lymphocytes, an intrinsic defect in cis regulatory molecules, and defective clearance of immune complexes and apoptotic cell bodies. These can operate individually or as “a la carte” combination of misaligned mechanisms.

In addition to the well-reviewed associations between conventional PIDs (i.e., due to defects in complement, phagocytes, B and T lymphocytes, where affected patients manifest unusual susceptibility to various infectious agents) and rheumatic diseases (13), a novel group of disorders of immune dysregulation (or primary immune regulatory disorders, PIRD) became evident, characterized predominantly by autoimmunity, inflammation, or other immune-mediated pathology (14–16). This led the way to broadening the spectrum of inborn immune system “failures” associated with rheumatic diseases (17, 18) including the monogenic autoimmune and autoinflammatory disorders, eventually named as inborn errors of immunity (IEI) (19).

By the end of 1990's, the first three described monogenic autoimmune diseases were formally classified as PIDs - autoimmune lymphoproliferative syndrome (ALPS) caused by mutations in CD95/FAS gene (20), autoimmune polyendocrinopathy-candidiasis-ectodermal dystrophy (APECED) or autoimmune polyendocrinopathy syndrome (APS) type I caused by mutations in AIRE gene (21), and immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) caused by mutations of FOXP3 gene (22). However, as far back as 1970s, young children suffering from SLE-like disorders were reported with hereditary complement component C1q deficiency (23, 24). Resulting deficient classical complement pathway activity was postulated to underpin a failure to clear apoptotic cells leading to exaggerated inflammatory responses. This formed the basis of the “waste disposal” hypothesis to explain the pathogenesis of SLE (25, 26).

In parallel, an entirely novel group of diseases related to periodic fever syndromes initially described in 1950s (27, 28) was added to the canon, heralded by the finding of MEFV gene mutations as a cause of familial Mediterranean fever (FMF) (29–31) and of TNFR1 gene mutations defining a family of dominantly inherited autoinflammatory syndromes (32).The ever-expanding spectrum of the monogenic autoinflammatory diseases refers to the inborn disorders of the innate immune system, with presentations characterized by episodes of systemic inflammation that are mediated largely by myeloid cells and by lack of autoantibodies and/or autoreactive T lymphocytes, thus distinguishing them from autoimmune diseases (33–35). Fascinatingly, C1q deficiency is currently understood and classified as an interferonopathy, one of the subcategories of autoinflammatory diseases (36).

Semantics and nomenclature

The term “primary immunologic deficiency (PID)”, as per the first report by the World Health Organisation (WHO) Committee on classification of PIDs referred to a group of diseases caused by a failure of either B lymphocyte/humoral immunity and/or of thymus-dependent T lymphocyte immunity, where affected individuals presented with markedly increased susceptibility to various infectious agents manifesting usually (but not exclusively) in early childhood (6). Although primary complement and phagocytic cell defects were not initially included, over the next 5 decades the term PID expanded to comprise disorders resulting from mutations in an ever-increasing number of genes involved in immune host defence (both of innate and adaptive immunity) and/or immunoregulation (central/thymic and peripheral tolerance) as evidenced by numerous WHO and later International Union of Immunological Societies (IUIS) Committees updates. However, as stated in the 2017 update “this terminology does limit the conceptualization of disorders to those in which susceptibility to infection is the main manifestation. The improving recognition of immune dysregulation diseases, including the growing field of autoinflammatory disorders and interferonopathies, has mandated that a more encompassing terminology be used”, i.e., IEI (19).

Changes to accepted nomenclature are continuously being debated (37–41) confirming the astute prediction published at the dawn of PIDs coming into existence (42) by one of the pioneers in the field: “Like all hypothetical schemes, based on incomplete knowledge and imperfect interpretation, it must be considered tentative and useful mainly in establishing hypotheses to be demolished by future investigation” (43).

The latest update in the IUIS Committee classification lists 508 different genes known to cause IEIs that can present clinically with various manifestations, from increased susceptibility to infections, autoimmunity, autoinflammation, allergy, bone marrow failure, and/or malignancy (44). Importantly for understanding the underlying pathophysiology leading to the phenotypic (clinical) manifestations of these disorders, immunodeficiency (e.g., susceptibility to viral infections) and immune dysregulation caused by different mutation(s) of the same gene (e.g., STAT2) often overlap (45). Closer to the everyday paediatric rheumatology practice, this dichotomy is well illustrated by IL-1 mediated disorders that may present as bacterial infections if the downstream signalling pathway (IRAK4/MyD88) is non-functional (i.e., deficiency) or as cryopyrin-associated periodic fever syndrome (CAPS) due to overproduction of IL-1 in GOF mutations in NLPR3 gene (46, 47). Drawing from this knowledge into the modern era management, it should be expected that biologic therapies targeting particular (specific) immunologically/pro-inflammatory active molecules and/or pathways are capable of mimicking the naturally occurring IEIs (often deficiency, but occasionally pro-inflammatory) and causing potentially serious side effects (48–51).

Discussion – or what has it all to do with paediatric rheumatology?

It is within this maze of presentations and overlapping phenotypes that paediatric rheumatologists of the day must navigate, explore and eventually succeed in assessing and diagnosing the patient correctly (5, 52). Often, the problem is a “novel” disorder (although most likely unknowingly encountered previously), eluding the traditional evidence-based approach to clinical decision-making (53). One memorable patient of ours strikingly illustrates the evolving pattern and myriad variations in clinical presentations which characterise many of these disorders, as well as the anguish of the patient, parents, and the medical team while “searching for the unknown” (Table 1).

The unprecedented advances in development of modern genetic analysis allowed for an ever-increasing number of novel monogenic disease entities to be defined in the last two decades and there is no reason to doubt that this is to continue (55, 56). The fact that almost half of the newly reported genes underlying IEIs in the previous two years refer to either autoinflammatory or immune dysregulation disorders is a strong proof that the field of IEIs is inseparably intertwined with rheumatology (5, 44).

Therefore, the main practice implications for paediatric rheumatologists are as follows:

The most important clinical acumen is to discern the “red flags” which may signal the possibility of an underlying monogenic IEI as a cause for rheumatological presentation. Broadly, these are features that are unusual, atypical (in pattern, severity, age of onset) and evolving in multisystem involvement often requiring very specific “precision and personalised” therapies (Box 1).

Box 1"Red Flags" for suspecting a monogenic IEI paediatric rheumatology. Adapted from (85).

• ○(Very) early onset with a severe course - unusual
• ○Constellation of multiple autoimmune disorder - unusual
• ○Evolution of autoimmunity with accrual of different organ system involvement (multisystem involvement) - a pattern
• ○Family history (similar disorder; sex-linked/male-only) and parental consanguinity (autosomal recessive trait) - a pattern
• ○Predisposition to infection (usually of early onset, frequent and severe) - unusual (pattern)

The modern-day paediatric rheumatologist then requires ready access to current diagnostic tools, including functional immunology tests, cytokine profiling, interferon-stimulated gene (ISG) expression, and whole genome sequencing (WGS). These are powerful tools in the diagnosis and management of rheumatological conditions (informing “precision treatments”) although not without caveats (e.g., the accurate evaluation and interpretation of results) of which practicing paediatric rheumatologists ought to be well aware (57–62).

They benefit from collaborative working with their colleagues in paediatric immunology and from managed clinical networks through which experience and learning gleaned from the care of small numbers of children with very rare conditions can be shared more widely. In this way, the ongoing dialogue between immunology and rheumatology continues to enrich both specialities and drives improvements in the care of their patients. Many patients under our care benefited from multidisciplinary team (MDT) approach via combined paediatric rheumatology and immunology clinics (initially set up as “periodic fever clinics”, later combined with other specialities), as well as by national (63–66) and international collaboration (67–71). Our team contributed to identification of several novel IEI disorders with significant “rheumatological” phenotypes that were part of their clinical presentation (72–78). These are rare diseases, often with no more than a single or handful of patients initially observed by astute clinicians; yet in the following years it transpired that some (e.g., deficiency of adenosine deaminase 2, DADA2) are among the most common monogenic immunodysregulatory and/or autoinflammatory diseases (79).

Today's options for “personalised” treatments are unparalleled in comparison even to those of yesterday, let alone to the historical “therapeutic pyramid” approach (80–82). However, as some patients with monogenic immune dysregulation and/or autoinflammatory disorders are resistant even to the very specific and targeted treatments, there is a valid role for haematopoietic stem cell transplantation (HSCT) in their management (83) and prospect for gene therapy and/or editing in the near future are promising (84).

Summary

Advances in immunology over the last 30 years have revolutionised the field of rheumatology. Today many of the clinical phenotypes presenting to rheumatology clinics have clearly defined genetic aetiologies through which disruptions in carefully regulated immune pathways lead to autoimmunity, autoinflammation and other immune-mediated pathology, which are understood with ever more clarity. Dozens of newly-recognised primary immune regulatory disorders have taken their place alongside the classical PIDs of old, in a list of more than 500 monogenic inborn errors of immunity, as described in the latest IUIS Classification (44, 56).