I am grateful for the interest of Drs. Rigante and Cantarini in my recent review of the role of IL-1 in systemic JIA. They highlight, very correctly, the abundant evidence for innate immune dysfunction in this condition and suggest that systemic JIA should be recognized as part of the growing family of autoinflammatory diseases.
This is clearly a very legitimate position and places Drs. Rigante and Cantarini in the company of other authorities who share this view. JIA is an umbrella term that includes a diverse range of clinical phenotypes, and systemic JIA is without doubt the most distinctive of these. As a form of idiopathic childhood arthritis, systemic JIA seems to me to fit comfortably enough under the umbrella, but other ways of grouping these diseases are certainly reasonable.
However, it seems to me that we should not—yet—become too comfortable with the conclusion that systemic JIA is an autoinflammatory disease. In its most precise use, this term encompasses inflammatory diseases that arise through aberrant antigen-independent activation of the immune system. Paradigmatic examples include familial Mediterranean fever (FMF) and the cryopyrinopathies. Like these conditions, systemic JIA is characterized by fever and responds briskly to IL-1 antagonism, although there are also differences. Garden-variety systemic JIA is almost never familial, and monozygotic twins can be discordant for disease; it almost never presents in early infancy; it frequently evolves into an afebrile chronic arthritis; and in many cases systemic JIA resolves completely over time, never to return. These differences should give us pause about classifying systemic JIA together with the monogenic autoinflammatory diseases.
One peril entailed in making premature conclusions about the biology of systemic JIA is that important pathogenic mechanisms may be overlooked. Autoinflammatory diseases are commonly, albeit perhaps imprecisely, regarded as diseases of innate immunity. From this point of view, T cells and B cells might be assumed to be irrelevant. However, IL-1 is a critical modulator of lymphocytic immunity, including Th17 cell differentiation and Treg cell function. The goal of the review was to raise the possibility that IL-1 and other cytokines might engender T cell–driven pathology in systemic JIA, taking a cue from mice deficient in IL-1 receptor antagonist in which T cell–mediated arthritis develops (). Indeed, the largest genome-wide association study in systemic JIA, which is still published only in abstract form, identifies a clear if relatively weak association of systemic JIA with the HLA class II locus, a hallmark of antigen-driven T cell autoimmunity ().
A further complication in assigning systemic JIA to the autoinflammatory family is that excessive immunity and immunodeficiency are sometimes hard to tell apart. This point is illustrated by the innate immune–sensing protein nucleotide-binding oligomerization domain–containing protein 2 (NOD-2). Gain-of-function mutations affecting NOD-2 result in the autoinflammatory disease Blau syndrome. Loss-of-function mutations can result in inflammatory bowel disease, potentially through failure to properly defend the intestinal barrier ([3, 4]). From this point of view, it is interesting that patients with systemic JIA and macrophage activation syndrome often bear mutations that result in defective cell–cell killing. Such mutations are postulated to impair control of activated macrophages, thereby leading to enhanced inflammation. Cell–cell killing is also a key mechanism for control of viruses, and it is legitimate to question whether mishandling of viral infections (i.e., immunodeficiency) might represent an important early step in the pathogenesis of systemic JIA. If this is the case, one could debate whether systemic JIA is really a primary autoinflammatory disease.
Finally, it is by now well recognized that the “autoinflammatory” label is not typically all or none. Even in diseases with relatively clear antigen-driven autoimmunity, such as rheumatoid arthritis (RA) and lupus, innate immune mechanisms including neutrophils and complement represent important mediators of tissue injury. It is therefore to be expected that variation in innate immune function might affect the incidence and severity of diseases of many types. Indeed, in parts of the world where FMF is common, heterozygous carriers of MEFV mutations appear to exhibit a greater predilection for JIA, a higher incidence of Henoch-Schönlein purpura and other vasculitides, and more severe RA ([5-7]). In fact, most inflammatory diseases should probably be conceptualized as residing in an autoinflammatory–autoimmune continuum (). Systemic JIA is no exception, and I share with Drs. Rigante and Cantarini the opinion that systemic JIA probably lies closer to the autoinflammatory end of the spectrum than most other subtypes of JIA, although enthesitis-related arthritis (perhaps driven by HLA–B27 misfolding) might make a competing claim (). Only further research will tell for sure.