Autoinflammation: translating mechanism to therapy

Review discusses autoinflammatory diseases as a new classification of immune disorders due to dysregulated innate immunity yet sensitive to targeted cytokine therapy.

Abstract

Autoinflammatory syndromes are a clinically heterogeneous collection of diseases characterized by dysregulation of the innate immune system. The hereditary recurrent fever disorders were the first to be defined as autoinflammatory. Several of the responsible genes are now known to encode proteins forming multimeric complexes called inflammasomes, which are intracellular “danger sensors” that respond to a variety of different signals by activating caspase-1, responsible for cleavage and subsequent release of bioactive IL-1β. This discovery of the causative link between autoinflammation and IL-1β maturation has led to a significantly improved understanding of the mechanisms of innate immunity, as well as life-altering treatments for patients. Targeting IL-1β for the treatment of autoinflammatory syndromes is an excellent example of the power of translational research. Given the central role of inflammation in many complex multigenic diseases, these treatments may benefit larger numbers of patients in the future. Here, we review current treatment strategies of autoinflammatory diseases with a focus on IL-1 antagonism.

Introduction

The past 30 years have seen important strides in understanding the intricate processes behind adaptive immune dysregulation, resulting in complex disorders such as allergy and autoimmunity. The pathways leading to development of antibodies to nonharmful antigens or autoantibodies and aberrant T cell populations are under intense study. However, the identification of the molecular basis of a few rare, inherited inflammatory conditions with some clinical features of allergy and autoimmunity led to the newly coined “autoinflammatory” syndromes. These disorders apparently have little contribution from adaptive immunity. Instead, these conditions result primarily from dysregulation of the innate immune system, long considered a static, nonspecific collection of cells and proteins with little role beyond providing an initiating boost to the adaptive arm.

Autoinflammation is a systemic disease affecting a large number of organs, including the skin, joints, and nervous system. The first clinical descriptions of what are now known as inherited autoinflammatory diseases were reported over 50 years ago [1, 2]; however, the genetic basis of these disorders was not determined until the late 1990s. Mutations in the gene MEFV, encoding pyrin protein, were identified in patients with FMF, the most common inherited autoinflammatory syndrome [3, 4]. Shortly thereafter, familial Hibernian fever was mapped to mutations in the p55 TNFR and renamed TRAPS [5]. Pyrin was a protein of previously unknown function, and the TNFR had long been studied as an important immune signal transducer. Interestingly, the identification of a third autoinflammatory gene established a hitherto unsuspected link between cholesterol metabolism and inflammation when mutations in MVK were found in recurrent fever patients with HIDS [6, 7].

Despite these exciting discoveries, our understanding of autoinflammatory disease pathogenesis and its treatment remained limited. In 2001, a significant breakthrough occurred when researchers demonstrated that mutations in the gene NLRP3 (cold-induced autoinflammatory syndrome 1), encoding NLRP3 (NALP3, cryopyrin) protein, are responsible for FCAS and MWS [8]. NLRP3 was later shown to nucleate a multiprotein complex called the inflammasome, essential for the release of bioactive IL-1β in response to various cytosolic “danger signals” [9, 10]. Taken together, these two findings provided a clear link between disease and a cellular pathway leading to overproduction of a specific inflammatory mediator.

The influence of NLRP3 expanded further the following year, when mutations were found in patients with NOMID, a condition with similarities to FCAS and MWS but with severe neurological sequelae. FCAS, MWS, and NOMID are now collectively called the CAPS, a disease continuum with extensive genetic and pathophysiologic overlap. This review will cover aspects of inflammasome biology and treatment of autoinflammatory syndromes, with a focus on IL-1 antagonism in CAPS. Additional, in-depth reviews of inflammasome biology can be found elsewhere [11, 12].

INNATE IMMUNITY—LEUKOCYTES AND CYTOKINES

The hallmark of innate immunity is rapid production and release of proinflammatory cytokines, including TNF-α and IL-1β, in response to danger signals such as microbial PAMPs, hypoxia, and toxins [13]. The pathways that allow these responses to occur are thought to be the oldest and most “hard-wired” in the immune repertoire. Originally thought to provide only a first, nonspecific defense while allowing the adaptive arm to fully mobilize, the innate immune system is now considered highly complex, with essential functions lasting long beyond the initial phase.

The principle cells commencing the innate response include tissue macrophages, DCs, and nonhematopoietic cells such as epithelial and endothelial cells at the pathogen/host interface. It is likely that these cell populations are responsible for directing the bouts of sterile inflammation experienced by patients with autoinflammatory disease. Indeed, peripheral blood monocytes from CAPS patients, but not normal controls, secrete high levels of IL-1β constitutively or in response to low concentrations of inflammatory stimuli.

Other cell types, such as γ-δ T cells [14], NKT cells [15], NK cells [16], and B1 cells [17], are additional, important contributors, although their role in autoinflammation, if any, has not been elucidated. Mast cells can also function as innate immune cells [18] and may play a role in IL-1β-driven cutaneous inflammation [19].

High numbers of neutrophils infiltrate the tissues of patients with autoinflammatory diseases and mutant mice engineered to carry related NLRP3 mutations [20], possibly recruited by one or more of the cell types listed above. IL-1β increases expression of adhesion molecules on the endothelium and release of chemoattractants such as MIP-2, which along with other cytokines and chemokines, attracts neutrophils to tissues. Once in situ, neutrophils can induce tissue injury and are involved in subsequent repair. The roles of other cell types in CAPS are unclear. IL-1β is essential for proper development of a Th17-polarized response [21], and one study on mice expressing a mutation homologous to a human MWS mutation demonstrated an IL-17-positive T cell population that could contribute to neutrophilic responses [22]. However, experiments using NLRP3 mutant mice on a B and T cell-deficient genetic background suggest that adaptive immunity is not required for the murine CAPS phenotype [20].

The cytokines produced by innate immune cells are diverse and include the IL-1 family (IL-1, IL-18, IL-33), TNF family (TNF-α, LT-α), IL-6 family (IL-6, IL-11), IL-17 family (IL-17A, IL-25), and type 1 IFNs (IFN-α, IFN-β), among others. These mediators are rapidly released from a wide array of cell types and under specific conditions such as viral infections (type 1 IFNs) or allergic triggers (IL-25). In many scenarios, these cytokines play redundant roles, suggesting that therapy targeting only one member will not adequately treat inflammatory disorders. However, the notable improvement seen in large cohorts of rheumatoid arthritis patients receiving TNF-α blocking therapy indicates that targeting one cytokine, even in a polygenic, complex inflammatory disorder, can be beneficial [23]. As will be discussed in detail, the blockade of IL-1β in many autoinflammatory disorders has further strengthened the rationale of this therapeutic strategy.

NLRP3 INFLAMMASOME

NLRP3 is a member of the NLR family, an array of danger-sensing proteins that act as a cytosolic counterpart to the membrane-associated TLRs. Structurally, NLRs are characterized by a central nucleotide-binding motif called the NAIP, CIITA, HET-E, and TP1 domain (or NOD domain), an N-terminal PYD [or CARD domain in related proteins NLRC1 and NLRC2 (NOD1 and NOD2)], and a series of LRRs at the C terminus (Fig. 1) [24]. NLRP3 binds two adaptor proteins: ASC and CARDINAL, as well as chaperone proteins (HSP90 and SGT1). Both adaptors contain CARD domains that recruit multiple molecules of caspase-1. It should be noted that there is no murine homologue to CARDINAL, and knockdown studies in human cells have revealed no effect on subsequent caspase-1 activation, suggesting that the CARDINAL component may not be required for NLRP3 function [25]. Upon activation of NLRP3, the multimeric inflammasome forms, allowing proximity-induced autocleavage of caspase-1 and activation. Active caspase-1 then cleaves pro-IL-1β to IL-1β (Fig. 2). Caspase-1 is also responsible for maturation of IL-18 and possibly IL-33, although recent reports have suggested that IL-33 may be cleaved by other means and actually inactivated by the inflammasome [26].

Figure 1. Protein domain structure of NLRs, including CARD, PYD, nucleotide-binding domains (NBD), LRR domains, and a domain with function to find (FIIND).

Figure 2. Targeting IL-1β-mediated inflammation.

The inflammasome is an intracellular protein complex that is activated by numerous danger signals including danger-associated molecular patterns (DAMPs) and PAMPs. This activation involves several hypothesized mechanisms, including K⁺ efflux secondary to ATP-gated channels, ROS, MSU, and membrane perturbation. Activation of the inflammasome leads to the cleavage and activation of caspase-1 and subsequent cleavage of pro-IL-1β to its mature, active form, which is then released from the cell. Once released, IL-1β binds to the IL-1R, leading to downstream signaling and a cascade of inflammation involving other proinflammatory cytokines. TLR triggering and autocrine IL-1R activation lead to pro-IL-1β transcription. Targets that may inhibit IL-1-mediated inflammation (depicted by red bulls eyes) include specific inflammasome triggers, activation mechanisms, specific components of the inflammasome, caspase-1, IL-1β release, binding of IL-1β to the IL-1R, IL-1R signaling transduction, and downstream proinflammatory cytokines. O₂^–, Superoxide ion.

A broad array of danger signals has been shown or hypothesized to activate the NLRP3 inflammasome, leading to processing of IL-1β. These include endogenous and exogenous signals, such as bacterial and viral microbial products, ATP, MSU crystals, oxygen radical species, K⁺ efflux, and membrane perturbation [27–30] (Fig. 2). Just how so many heterogeneous signals impact a single protein is not understood and may translate through an as-yet unidentified common activator upstream of NLRP3. This funneling of varied danger signals through a single sensor implicates a role for the inflammasome in many inflammatory disorders, even certain common conditions recently linked to innate immune pathology, such as type 2 diabetes and atherosclerosis.

There are various theories regarding common pathways of NLRP3 activation. It has been proposed that all of the activators result in the generation of ROS that are sensed by and activate NLRP3. In support of this hypothesis, ROS scavengers, small interfering RNA knockdown or pharmacologic inhibition of the p22^phox subunit of NADPH oxidase, required for ROS generation, can suppress NLRP3 activation induced by crystals (MSU, asbestos, silica) [27, 28], ATP [31], and microbes such as Candida species [32]. Although the list of signals shown to activate NLRP3 continues to lengthen, by no means does it equal the vast array of ROS-generating stimuli; thus, some additional sensing component must exist to provide specificity.

Although intriguing, the ROS hypothesis is only one of several feasible explanations. It is also possible that the inflammasome becomes activated when lysosomal damage causes release of proteases into the cytosol. Known NLRP3 activators, such as silica or β-amyloid, can disrupt lysosomes. Further, pharmacologic disruption of lysosomes induces NLRP3 activation, in part, through proteases such as cathepsin B [33, 34]. K⁺ efflux has consistently been demonstrated to be crucial for NLRP3 activation and may contribute to the above models or function in an independent manner. Multiple signals are likely needed for full activation, as evidenced by recent work showing that TNF-α or other NF-κB signals are critical for inflammasome priming before activation can ensue [35, 36]. This “two signal” model of innate immunity appears to be one of the early steps of IL-1β regulation.

The large numbers of infiltrating neutrophils in murine and human CAPS increase the likelihood of inflammasome-independent IL-1β cleavage. Neutrophil proteases such as proteinase 3 are capable of cleaving caspase-1 in the absence of NLRP3 activation [37]. This finding and the pleiotropic effects of IL-1β on the induction of fever and generation of acute-phase cytokines such as IL-6 necessitate further regulation beyond the inflammasome. An endogenous IL-1Ra, which competes with IL-1α and IL-1β for receptor binding, is one of the proteins induced by IL-1β-induced NF-κB signaling, providing a negative-feedback loop to control inflammation. Anakinra, a recombinant form of IL-1Ra, was the first drug developed to antagonize IL-1 [38].

IL-1 production in innate cells can also be dampened by interaction with certain adaptive immune cells. Effector and memory T cells were recently shown to reduce macrophage IL-1β production via cell contact through TNFR family member ligation [39]. In this model, negative feedback by the adaptive immune arm may limit uncontrolled, innate IL-1β production and thus, lead to appropriate degrees of inflammatory responses.

AUTOINFLAMMATORY SYNDROMES

The emergence of syndromes characterized by dysregulated innate immunity or “autoinflammation” has redefined classical paradigms of inflammatory disease. The autoinflammatory syndromes are heterogeneous with varying clinical manifestations and severity, but can be separated arbitrarily based on genetics. The monogenetic disorders are rare, inherited diseases and include CAPS, FMF, HIDS, TRAPS, and others listed in Table 1.

Table 1. Inherited Autoinflammatory Disease and Current Therapies.

Disease	Mutation	Inheritance	Onset	Organs affected	Treatments	Target	Efficacy
Hereditary fever disorders
FMF	MEFV/pyrin	AR	80% before 20 years old	Skin, joints, peritoneum, pleura	Colchicine	Microtubule inhibitor, neutrophil chemotaxis	Episode number, inflammatory markers
					Anakinra	IL-1R	All symptoms, markers
					Rilonacept	IL-1β	In progress
TRAPS	TNFRSF1A/TNFRSF1A, TNFR1, p55	AD	Median: 3 years old	Skin, eyes, joints, peritoneum, pleura	Etanercept	TNF-α/LT-α	Steroid dose, inflammatory markers
					Anakinra	IL-1R	Episode number, inflammatory markers
HIDS	MVK/mevalonate kinase (MK)	AR	Median: 6 months	Skin, eyes, joints, serosa, prominent LNs	Anakinra	IL-1R	All symptoms, markers
CAPS
FCAS	NALP3/cryopyrin	AD	<6 months	Skin, eyes, joints	Anakinra	IL-1R	All symptoms, markers
					Rilonacept Canakinumab	IL-1β	All symptoms, markers
MWS	NALP3/cryopyrin	AD	Infancy to adolescence	Skin, eyes, joints, ears, meninges	Anakinra	IL-1R	Most symptoms, markers
					Rilonacept Canakinumab	IL-1β	Most symptoms, markers
NOMID	NALP3/cryopyrin	AD/sporadic	Neonatal/infancy	Skin, eyes, joints, ears, meninges, bones	Anakinra	IL-1R	Most symptoms, markers
					Canakinumab	IL-1β	In progress
Other hereditary disorders
Pyogenic arthritis, pyoderma gangrenosum, and acne	CD2BP1 or PSTPIP1	AD	Early childhood	Skin, joints	Anakinra	IL-1R	Skin symptoms
Blau syndrome	NOD2/CARD15	AD/de novo	Early childhood	Skin, eyes, joints	Anakinra	IL-1R	Symptoms in most
Deficiency of IL-1 Ra	IL1RN/IL-1Ra	AR	Neonatal/infancy	Skin, bones, lungs, vascular	Anakinra	IL-1R	All symptoms, markers

AR, Autosomal-recessive; TNFRSF1A, TNFR superfamily 1A; AD, autosomal-dominant; CD2BP1, CD2-binding protein 1; PSTPIP1, proline-serine-threonine phosphatase-interacting protein 1.

Complex or polygenic autoinflammatory disorders include systemic-onset juvenile idiopathic arthritis, adult-onset Still disease, the syndrome of periodic fever with aphthous stomatitis, pharyngitis, and cervical adenitis, Behçet's disease, and chronic recurrent multifocal osteomyelitis. Additionally, since the discovery that MSU and CPPD crystals activate the inflammasome [29], gout (gouty arthritis) and pseudogout are now largely considered autoinflammatory diseases. Finally, a possible pathogenic role for the IL-1/NLRP3 pathway has been demonstrated recently in type 2 diabetes, although this disease is not routinely classified as autoinflammatory [40].

CAPS

The CAPS clinical spectrum includes FCAS, MWS, and NOMID, as well as overlap syndromes falling between specific phenotypes. All CAPS diagnoses follow an autosomal-dominant inheritance pattern, although NOMID tends to be sporadic as a result of reduced fitness in affected individuals. CAPS are a true disease continuum, ranging from FCAS at the mild end to NOMID at the most severe, with some overlap reported in individual patents [41]. Common features of CAPS, regardless of the specific clinical entity, include fever, urticarial-like rash, conjunctivitis, bone and joint symptoms, and elevated inflammatory markers such as CRP.

Familial cold urticaria was described as early as 1940 [2]. Some 60 years later, the disease was renamed FCAS when the specific genetic lesion was identified [8]. FCAS is characterized by cold-induced episodes of urticarial-like rash, fever, conjunctival inflammation, and joint/limb pain, which commonly last less than 24 h and typically begin in the first 6 months of life [42]. Although patients have significant morbidity related to these attacks, long-term prognosis is favorable.

MWS patients tend to have somewhat longer febrile episodes without the association with cold. In addition, patients tend to develop inner-ear inflammation that eventually leads to deafness. Other long-term sequelae include the development of amyloidosis, a widespread deposition of amyloid proteins in multiple organs resulting from chronic inflammation. The kidneys are often most severely affected, resulting in renal failure and the need for dialysis and transplant [43, 44].

The most severe form of CAPS is NOMID, characterized by neonatal or infantile onset of urticarial rash and chronic aseptic meningitis that can result in cerebral atrophy, seizures, hearing and vision loss, and mental retardation. In addition to CNS pathology, NOMID patients develop a pathognomonic arthropathy as a result of overgrowth of the epiphyses of the long bones. Similar to MWS, amyloidosis is a known risk. Prior to administration of IL-1 antagonistic therapy, the prognosis for patients with NOMID was poor [45].

FMF

The most prevalent and well-described autoinflammatory syndrome is FMF, which is an autosomal-recessive disease characterized by recurrent attacks of fever associated with sterile inflammation of the peritoneum, pleura, joints, skin, and occasionally, pericardium, lasting 1–3 days. Similar to CAPS, SAA amyloidosis is a frequent complication of FMF, leading to kidney failure. Colchicine was reported in the 1970s and 1980s as effective in reducing the number of attacks and preventing amyloidosis. It remains a mainstay of treatment [46, 47].

Although mutations in pyrin protein were identified in FMF patients in 1997 [3, 4], it wasn't until the discovery of NLRP3 that a pathogenic role for the inflammasome and IL-1β was realized for FMF. In vitro experiments suggest pyrin can bind to ASC or caspase-1 directly. Additionally, pyrin may compete with NLRP3 and caspase-1 for binding to ASC, thereby inhibiting inflammasome activity. Therefore, FMF mutations may activate the inflammasome or attenuate the anti-inflammatory properties of pyrin. In support of these hypotheses, FMF-mutated pyrin led to increased IL-1β release from cells in tissue culture [48–50]. Perhaps most compelling, the use of the IL-1 pathway blockade has been successful in some FMF patients [51–53].

TRAPS

TRAPS is an autosomal-dominant disease caused by mutations in the p55 component of the TNFR [5]. The syndrome is characterized by febrile episodes associated with peritonitis, pleuritis, and arthritis, which last anywhere from 1 to 6 weeks. Other features of TRAPS are migratory lymphedema and periorbital edema, and patients have a lifetime risk of renal failure as a result of amyloidosis. Potential mechanisms for uncontrolled inflammation in TRAPS include aberrant TNFR1 signaling or impaired shedding of TNFR1, leading to reduced levels of soluble receptor to inhibit TNF binding to the cell surface [54, 55]. TNF blockade with the soluble receptor fusion protein etanercept has proved efficacious for patients by reducing inflammatory markers and allowing a reduction in corticosteroid dose. Interestingly, IL-1 blockade also ameliorates clinical symptoms and lowers inflammatory markers, particularly in patients refractory to the TNF blockade [56–58]. Thus, IL-1β may play an additional and/or redundant role in a TNF-mediated disease.

HIDS

Recurrent fever HIDS is an autosomal-recessive disease characterized by 3–7 day long episodes of fever, abdominal pain, rash, lymphadenopathy, and joint pain that begin early in life [59]. As the name suggests, patients usually have elevated serum IgD levels, although this finding does not appear to be pathogenic. Interestingly, immunizations can precipitate in inflammatory attacks in about half of the patients. In 1999, mutations in the gene MVK were identified in patients with HIDS [6, 7].

Currently, treatment options for HIDS patients are limited, but one promising treatment for these patients has been anakinra, with resultant attenuation of nearly all symptoms [60, 61]. Consistent with this response to IL-1 blockade, elevated IL-1β production has been found in HIDS PBMCs [62]. One possible mechanism may be related to decreased production of nonsterol isoprenoid end-products in the setting of MVK deficiency. These end-products, such as geranylgeranyl pyrophosphate, have been shown in vitro to inhibit the excessive IL-1β production in PBMCs from HIDS patients [62]. Further, inhibition of HMG-CoA reductase, upstream of MVK, led to caspase-1 activation and increased IL-1β production in a Rac1/PI3K-dependent manner, and inhibition of Rac1 resulted in reduced IL-1β secretion [63]. However, contrary to these results, unstimulated PBMCs from coronary artery disease patients taking HMG-CoA reductase inhibitors (statins) for 6 months had decreased IL-1β secretion [64]. Perhaps this is related to a separate drug effect in vivo, but it implies that nonsterol isoprenoid reduction may be only part of the explanation for increased IL-1β production in HIDS.

DIRA and others

Recently, another IL-1-mediated, inherited disease was described known as DIRA [65, 66]. Patients developed skin pustules in early life, along with periostitis, multifocal osteomyelitis, oral mucosal lesions, and pain with movement. Treatment with anakinra has resulted in dramatic improvement and considered potentially lifesaving if the disease is recognized early. The genetic causes of other inherited autoinflammatory syndromes, such as pyogenic arthritis, pyoderma gangrenosum, and acne, and Blau syndrome have been identified, and anti-IL-1 therapy was reported to improve symptoms in some of these patients as well (Table 1).

TARGETING IL-1 IN AUTOINFLAMMATORY SYNDROMES

Initially, IL-1 was targeted in septic shock, an acute, multiorgan, inflammatory process that occurs in response to microbial infection and can result in death or severe lifelong impairment. Unfortunately, large trials failed to validate the efficacy of anakinra for use in this complex disease, despite earlier studies showing some promise [67]. In 2001, anakinra was approved for use in rheumatoid arthritis, although the efficacy in this heterogeneous group of patients was less than other available biologics. The results from these initial studies with IL-1 blockade in human disease were somewhat disappointing given that IL-1β was regarded as a critical mediator in a wide array of inflammatory conditions. Despite this, the rationale for targeting IL-1 in CAPS was met with enthusiasm when it was discovered that NLRP3 is a critical part of the inflammasome, and macrophages purified from CAPS patients spontaneously secreted IL-1β [9, 10, 68]. The combined efforts of scientists in various fields, including genetics, genomics, immunology, and molecular biology, allowed for the translation of basic laboratory research findings into treatments for patients.

Anakinra in CAPS

Anakinra is an injectable rIL-1Ra that competes with IL-1α and IL-1β for binding to IL-1R, given s.c. daily with a half-life of 4–6 h (Fig. 3). The initial reports of CAPS patients treated with anakinra revealed remarkable clinical improvements and confirmed the critical role of IL-1 in the pathogenesis of CAPS. The first report described two patients with MWS, who achieved a lasting remission in inflammatory symptoms within hours of the first dose of anakinra as long as treatment was continued [69]. Additionally, SAA protein levels were reduced, and amyloid-induced kidney changes improved. Shortly thereafter, a report of anakinra administered to patients with FCAS before cold-room challenge revealed that inflammatory symptoms could be prevented and serum IL-6 reduced [70]. The most dramatic clinical response to the IL-1 blockade was seen in NOMID, the most severe form of CAPS [71, 72]. In addition to reduction in fever, rash, and inflammatory markers, improvements were observed in the chronic leptomeningeal sequelae associated with NOMID. Many studies using anakinra in CAPS have been performed since these initial therapeutic reports, confirming the rapid and profound improvement in symptoms for these patients [73, 74]. Unfortunately, there are some unfavorable characteristics of anakinra, including daily dosing and painful injection-site reactions. Anakinra is also not yet FDA-approved for CAPS, despite the impressive efficacy seen in trials.

Figure 3. Current mechanisms of IL-1-targeted therapy.

Anakinra, a recombinant form of IL-1Ra, targets IL-1R, and rilonacept and canakinumab target IL-1β.

Novel IL-1 pathway inhibitors in CAPS (rilonacept and canakinumab)

Rilonacept and canakinumab are two new IL-1 pathway inhibitors recently FDA-approved for use in CAPS (Fig. 3). Their approval is a testament to the combined effort of physicians treating CAPS patients, industry, and the FDA orphan drug-approval process to fast-track such therapies for rare diseases.

Rilonacept (also known as IL-1 TRAP) is a fusion protein that includes portions of the IL-1R and IL-1Ra protein fused to the Fc portion of IgG. Rilonacept has a high affinity for IL-1β, but also binds to IL-1α and IL-1Ra. The half-life of rilonacept is 8.6 days, allowing for weekly dosing and minimal associated injection-site reactions, both desirable features for patients when compared with anakinra. The initial trials with rilonacept in FCAS and MWS revealed efficacy with reductions in symptoms, inflammatory markers, and SAA levels [75, 76]. In 2008, the FDA approved rilonacept for use in MWS and FCAS patients over age 12. There was a trend toward increased respiratory infections in one of the trials and two deaths in the treated group in an open label extension, one as a result of pneumococcal meningitis and the other as a result of coronary disease. Certainly, an increased risk of infection is conceivable with IL-1-blocking therapy, but other potential side-effects, such as cardiac events, are limited to a few reports at this time, and any relationship to the therapy is unknown [77]. In general, rilonacept is well-tolerated. Data using rilonacept in NOMID patients are not yet available but would be expected to be promising given the response to anakinra.

Canakinumab is a humanized mAb to IL-1β with a half-life of 28 days, allowing for dosing every 2 months as a s.c. injection (Fig. 3). Initially, canakinumab was dosed more frequently based on its pharmacokinetics, but a study using a mathematical model, correlated with measured IL-1β antibody complex levels, predicted that the constitutive rate of IL-1β in treated CAPS patients remained low long after dosing [78]. An important conclusion of this report was that IL-1β appears to regulate production of itself in patients with CAPS, creating a positive-feedback loop (Fig. 2). The 8-week dosing schedule with canakinumab was used in a pivotal trial, which found complete or near-complete remission in 97% of treated CAPS patients during the study period [79]. The side-effects reported included increased suspected infections, one hospitalization for a severe urinary tract infection, and severe vertigo in a few patients. In 2009, the FDA approved canakinumab for use in FCAS and MWS patients older than 4 years of age.

IL-1 pathway blockade in other autoinflammatory diseases

Successful blockade of IL-1β has moved well beyond patients with CAPS (Table 1). Not surprisingly, patients with the recently described DIRA demonstrate near-complete remission when treated with anakinra [65, 66]. Use of anakinra may also be warranted in the treatment of the other monogenic-inherited, autoinflammatory diseases; however, the outcomes in case reports are variable [56, 57, 61, 80–92]. IL-1 blockade has been beneficial in the treatment of other rare, noninherited conditions affecting the skin and joints, including Schnitzler's syndrome, systemic-onset juvenile idiopathic arthritis, and adult-onset Still's disease (Table 2) [93–110]. Taken together, these findings highlight the strong contribution of IL-1β to a broad spectrum of autoinflammatory diseases regardless of heritability or specific mutation.

Table 2. Complex Autoinflammatory Diseases and Current Therapies.

Complex disorders	Onset	Organs affected	Treatments	Target	Efficacy
Schnitzler syndrome	Adulthood (often >40 years old)	Skin, joints, bones	Anakinra	IL-1R	Skin symptoms
Systemic-onset juvenile idiopathic arthritis	Childhood	Joints	Anakinra	IL-1R	Most symptoms, steroid sparing
Adult-onset Still disease	Adulthood	Skin, joints, pleura/pericardium	Anakinra	IL-1R	Most symptoms, markers
Acute gout	Adulthood	Joints	Anakinra	IL-1R	Symptoms in most
			Canakinumab	IL-1β
Chronic gout	Adulthood	Joints, skin	Rilonacept	IL-1β	Symptoms in most
Pseudogout (can be inherited)	Adulthood	Joints	Anakinra	IL-1R	Symptoms in a few
Type 2 diabetes	Adolescence/Adult	Kidney, eyes, heart, vascular, neuro	Anakinra	IL-1R	Improved glucose control, β cell function

Most of the conditions described thus far are relatively rare but give insight into the possibility that IL-1β is central to the pathogenesis of more common inflammatory disorders. Gout affects at least 1% of men (8:1 male:female) in Western countries and is characterized by recurrent attacks of acute, painful arthritis, and some patients develop chronic disease and tophi in the skin [111]. Excess uric acid (MSU) is deposited in the joints, triggering an inflammatory cascade. A report in 2006 revealed that MSU crystals can trigger the NLRP3 inflammasome (Fig. 2), raising the possibility that IL-1 inhibition could be an effective therapeutic strategy [29]. In a pilot study of 10 patients with refractory acute gout, a 3-day course of anakinra resolved all inflammatory symptoms completely and rapidly [112]. More recently, rilonacept was administered to patients with chronic gout in a proof-of-concept, placebo-controlled trial [113], and canakinumab was studied versus corticosteroids in acute gout in a phase II study [114]. Pain scores improved significantly in both studies. These reports have paved the way for larger trials to evaluate IL-1 blockade in gout. Given the initial promising results, patients with severe gout may now have better options for treatment of this debilitating disease.

Pseudogout is a similar crystalline-joint disease brought on by deposition of CPPD crystals, which like MSU crystals, can trigger the NLRP3 inflammasome [29]. Case reports of a few patients with pseudogout treated with anakinra showed remarkable improvements in symptoms [115, 116].

IL-1-mediated inflammation also appears to contribute to the pathogenesis of type 2 diabetes, a major health problem in developed countries characterized by chronic insulin resistance and complications that include kidney failure and heart and cerebrovascular disease. Although initially considered solely a metabolic disorder, type 2 diabetes is now recognized to have an inflammatory component as well. One study indicated elevated IL-1β levels were a risk factor for the development of type 2 diabetes [117], and a placebo-controlled trial of 34 diabetic patients demonstrated improvements in pancreatic β cell function, hyperglycemia, and IL-6 levels following 13 weeks of anakinra treatment [118]. CRP and IL-6 levels remained low even 39 weeks following anakinra withdrawal [119]. Interestingly, a subset of patients with genetically low IL-1Ra levels maintained improved pancreatic islet β cell function as well.

A link between glucose homeostasis and NLRP3 was suggested in a recent report, showing that TXNIP can directly activate NLRP3 in a ROS-dependent manner, although a second study refutes this finding [120]. TXNIP is induced by hyperglycemia and regulates islet IL-1β production [40]. Interestingly, TXNIP has been implicated previously in insulin resistance, and the observation that NLRP3-deficient mice are more glucose-tolerant compared with controls may be relevant to the human disease. Although the available treatments for type 2 diabetes can control disease in most patients who follow lifestyle-modification regimens, there may be a subset of difficult-to-control or high-risk individuals who would benefit from IL-1 blockade.

Other treatments of autoinflammation

Colchicine has been a mainstay of treatment for inflammatory disorders since the first century, originally in the form of an extract made from meadow saffron [121]. Several studies demonstrated the efficacy of colchicine in preventing FMF attacks as well as amyloidosis [47, 122–126]. By far, however, its most common use is to treat gout. Although its mechanism of action is still unclear, the microtubule-destabilizing properties of colchicine are thought to affect many cells types, including neutrophils. Colchicine may also impact inflammation by altering adhesion molecule expression, chemotaxis, and ROS generation [127]. Used appropriately, colchicine is generally tolerated but has well-known gastrointestinal, hematologic, and neuromuscular side-effects.

The synthetic, soluble TNFR, etanercept, binds to TNF-α and LT-α and is efficacious at ameliorating TRAPS symptoms, allowing a reduction in corticosteroid use [128, 129]. Surprisingly, other TNF inhibitors, such as the mAb infliximab, have been reported to worsen disease in some TRAPS patients [130]. The reason for this discrepancy is unclear and may lie in the pharmacologic differences between agents, such as the ability of etanercept to bind LT-α in addition to TNF-α [131]. Although there are case reports of TNF-α blockade in other autoinflammatory diseases, such as FMF, IL-1 antagonism and colchicine appear to have greater efficacy [132].

Future treatments and possible targets of autoinflammatory diseases

The IL-1 activation pathway is regulated at many steps, providing several potential targets for therapeutic intervention (Fig. 2). Targeting the inflammasome directly might be an ideal approach. The inflammasome chaperone HSP90 has been shown to regulate NLRP3 activity, and in one report, chemical inhibition of HSP90 reduced features of gout-like arthritis in an animal model [133]. Caspase-1 is a possible target from evidence in vitro [134], but potential side-effects of inhibitors and dosing will need to be addressed prior to in vivo use. Additionally, caspase-1 may inactivate the innate, Th2-inducing cytokine IL-33, resulting in off-target, immune-modulating effects [26].

As there are many triggers of the inflammasome, including ATP, K⁺ efflux, ROS, and MSU, there are also opportunities to target these specific pathways. Purine receptor (P2X7) and K⁺ channel inhibitors, as well as antioxidants that reduce ROS, may be of use in the future for specific inflammatory disorders. Disease-targeted therapy, such as blockade of TXNIP in type 2 diabetes, offers exciting, new approaches for the treatment of common disorders. As the potential exists for new therapeutic angles, caution is prudent, given the important role of the NLRP3 inflammasome in controlling some infections such as influenza in mouse models [135]. Future studies using inflammasome-directed therapy will undoubtedly reveal redundant and unique pathways of innate immunity that dictate the viability of treatments for human disease.

SUMMARY

Autoinflammatory diseases represent a newly classified group of disorders characterized by defects in the innate immune system. The discovery of mutations in rare, inherited syndromes and collaborative scientific investigation into mechanisms underlying the pathogenesis of autoinflammation have led to treatments for patients and an enhanced understanding of innate immunity. The importance of studying rare conditions is underscored as the role of the inflammasome is extended to common diseases and maintenance of health. Importantly, the research success story of autoinflammatory disease exemplifies the power of translational investigation and should be used as a model for the study of other human diseases.