In this report we describe a case of severe chronic infantile neurologic, cutaneous, articular (CINCA) syndrome with a novel G307V cryopyrin mutation and all of the characteristic clinical and laboratory features of this autoinflammatory disease. There was no clear response to standard therapies, including human interleukin-1 (IL-1) receptor antagonist (anakinra) and soluble tumor necrosis factor receptor (etanercept). The patient finally had a partial clinical response (reduction in fever and irritability) and complete laboratory response (improved C-reactive protein and serum amyloid A levels) to humanized anti–IL-6 receptor antibody (MRA), but died from congestive heart failure and interstitial pneumonia 2 months after initiation of therapy. We serially measured the serum cytokine levels and expression of NF-κB activation in the patient's peripheral blood mononuclear cells before and during consecutive therapies. Pathologic examination of autopsy specimens was also performed. This case illustrates the continued difficulty in management of patients with CINCA syndrome and the complexity of the inflammatory pathways in this disorder.
Chronic infantile neurologic, cutaneous, articular (CINCA) syndrome is an autoinflammatory disease characterized by persistent rash, central nervous system impairment, joint symptoms, and recurrent fever (1–3). Mutations in exon 3 of the CIAS1 gene have been identified in approximately half of the patients with CINCA syndrome (4). The CIAS1 gene codes for cryopyrin (NALP3/PYPAF1) and is expressed in polymorphonuclear leukocytes, monocytes, chondrocytes, and activated T cells (5). Cryopyrin is involved in the activation of NF-κB, a pivotal transcription factor for genes that encode proinflammatory cytokines, and caspase 1, an enzyme responsible for the processing of interleukin-1β (IL-1β) and IL-18 (6). However, pathogenesis of the cryopyrin-associated diseases has not been fully elucidated.
We herein describe a case of severe CINCA syndrome in a male infant. To clarify disease pathogenesis, we serially measured the serum levels of several cytokines and the expression of NF-κB activation in the patient's peripheral blood mononuclear cells (PBMCs). In an attempt to control the symptoms and laboratory evidence of inflammation, we treated the patient with consecutive anticytokine therapies, including human IL-1 receptor antagonist (IL-1Ra) (anakinra), soluble tumor necrosis factor receptor (sTNFR) (etanercept), and humanized anti–IL-6 receptor antibody (MRA).
The patient, a boy born at 37 weeks' gestation with a birth weight of 2,818 grams, developed fever and rash on the lower extremities within 3 hours after birth, which gradually became generalized. Despite treatment with antibiotics, antifungal agents, antiviral drugs, intravenous immunoglobulin, and corticosteroids, the fever and rash persisted. The possibility of an infection was excluded, and the patient was discharged at the age of 2 months.
At the age of 10 months, the patient was referred to the Yamaguchi University School of Medicine for further evaluation and treatment. He had persistent high fever, arthralgia, and typical generalized urticarial lesions. The clinical findings included delayed psychomotor development, bilateral papilledema, moderate perceptive deafness, and osteolysis with remodeling of the tibial metaphysis. He did not have lymphadenopathy or hepatosplenomegaly.
The white blood cell (WBC) count was 28,700/μl with a differential profile of 2% band neutrophils, 76% segmented neutrophils, and 10% lymphocytes. The level of C-reactive protein (CRP) was 5.57 mg/dl and the level of serum amyloid A (SAA) was 62.4 mg/ml (normal range <8 mg/dl). The cerebrospinal fluid cell count was 32/μl (47% neutrophils and 53% mononuclear cells). Cultures were negative for human immunodeficiency virus antibody and Epstein-Barr virus antibodies, and the results of polymerase chain reaction (PCR) were negative. Cytomegalovirus (CMV) IgG antibodies were found to be positive (titer of 64.1) but IgM antibodies were negative (titer of 0.41), and PCR for detection of CMV antigen yielded negative results. Tibial metaphyseal enlargement was noted on radiologic examination, and computed tomography (CT) of the brain revealed brain atrophy. With the use of automated DNA sequencing, as previously described (7), we confirmed the diagnosis of CINCA syndrome and identified a novel single-nucleotide mutation of G920T in exon 3 of the CIAS1 gene that resulted in an amino acid substitution of G307V.
Although auscultation of the chest was normal, the patient had chest retractions and clubbing of the fingers. Radiologic and CT scans of the chest revealed organized pneumonia and/or atelectasis at the hilum of the lung without cardiomegaly. A baseline electrocardiogram (ECG) and echocardiogram showed no abnormalities. The findings revealed on chest radiograph progressed to chronic interstitial pneumonia over the next 3 months. In addition, serum levels of KL-6, a serum marker of interstitial lung disease, increased to 778 units/ml (age-matched normal range <250 units/ml).
The patient's body temperature persisted at or higher than 39°C and seldom dropped below 38°C (Figure 1). Due to failure of several immunosuppressive treatment strategies, we initiated the use of biologic agents after obtaining informed consent from the family and receiving approval from the institutional review board of Yamaguchi University Hospital. At the age of 17 months, the patient began receiving anakinra subcutaneously at 0.8–1.6 mg/kg/day. The patient's rash would subside ∼3 hours after administration of anakinra; however, despite the use of increasing doses, a generalized rash consistently reappeared within 20 hours after administration. The frequency of high fever was reduced, but complete defervescence did not occur during treatment with anakinra. Narrowing of retinal vessels on physical examination and mild worsening of the articular lesions on radiographs were noted. The serum levels of CRP and SAA as well as WBC counts remained elevated (Table 1). Therefore, anakinra was discontinued after 5 months.
Table 1. Markers of acute inflammation, serum cytokines, and NF-κB activation in the patient's peripheral blood during treatment with biologic agents*
Human IL-1 receptor antagonist (anakinra)
Soluble TNF receptor (etanercept)
Anti–IL-6 receptor antibody (MRA)
Values are the mean (range) of samples obtained weekly before and during anticytokine therapy. IL-1 = interleukin-1; TNF = tumor necrosis factor; WBC = white blood cell; CRP = C-reactive protein; SAA = serum amyloid A.
Parameter was measured only once.
The percentages of cells with NF-κB activation in CD4+, CD8+, or CD14+ cells.
Because of the lack of complete response to anakinra, 0.8 mg/kg etanercept (administered subcutaneously once a week) was initiated in the patient at the age of 23 months. There was slight improvement observed in the tibial metaphyseal lesions on radiographs, but no improvement in the fever, rash, or ocular lesions occurred. Moreover, the SAA and CRP levels and WBC counts increased gradually during treatment with etanercept, and therefore this agent was discontinued after only 6 weeks (Table 1).
At the age of 25 months, treatment with MRA (8 mg/kg once a week, intravenous infusion) was initiated in the patient. Within 12 hours of initiation of MRA, the patient's body temperature dropped below 38°C and remained below 38°C for ∼4 days of the week (Figure 1). The patient also had significantly less irritability; however, there was no measurable improvement in the ocular, joint, or skin manifestations on physical examination, nor were there any improvements in the findings on bone radiographs or head CT. The CRP and SAA levels as well as WBC counts dramatically decreased 6 days after the first dose of MRA and remained below the level of detection for the duration of MRA treatment (Table 1).
Two months thereafter, the patient developed sudden-onset tachypnea and tachycardia. The chest radiograph showed cardiomegaly, and echocardiography showed a low ejection fraction of 41%. Since ECG and chest radiography showed no cardiac abnormalities immediately prior to MRA treatment, a diagnosis of acute congestive heart failure (CHF) was made. Within 24 hours, the patient died without responding to aggressive treatment for heart failure. A limited necropsy (restricted to the lung and heart) was performed after receiving the consent of the parents.
Microscopic examination of hematoxylin and eosin–stained tissue sections obtained from the patient at autopsy revealed swelling of the alveolar epithelium and mild infiltration of inflammatory cells, mainly lymphocytes, in the alveolar septa of the lung (Figures 2a and b). Immunohistochemical staining demonstrated the presence of T cells, with a predominance of CD8+ T cells. The alveolar septa showed some CD14+ macrophages but not CD19+ B cells.
Histologic investigation of the heart revealed slight edema between the cardiac muscle fibers. Infiltration of inflammatory cells was absent, and the cardiac muscle fibers had no evidence of hypertrophy (Figures 2c and d). Apoptotic cells were not detected by the TUNEL method. Amyloid deposits were not detected and cardiac vessels were normal. Electron microscopy showed enlarged myocardial cells with slight edema of the cytoplasm (Figures 2e and f). Deformation of the nuclei and mitochondria in the myocardial cells was not observed.
Serum inflammatory cytokine levels.
We measured the levels of several cytokines and soluble cytokine receptors in the patient's serum immediately after diagnosis. The serum levels of IL-1β, sIL-2R, monocyte chemoattractant protein 1 (MCP-1), macrophage colony-stimulating factor (M-CSF), transforming growth factor β (TGFβ), and sTNFR were determined by enzyme-linked immunosorbent assay with commercialized kits (R&D Systems, Minneapolis, MN and Bender MedSystems, Vienna, Austria). The concentrations of IL-2, IL-4, IL-6, IL-10, IL-12, interferon-γ (IFNγ), and TNFα were measured by flow cytometry with a cytometric bead array kit (BD PharMingen, San Diego, CA).
The serum levels of IL-1β, IL-6, IL-12, sTNFR, sIL-2R, MCP-1, and M-CSF were increased in the patient at diagnosis, when compared with the normal pediatric ranges. The level of IL-1β was 8.8 pg/ml (normal range <3.9 pg/ml), IL-6 was 75.6 pg/ml (normal range <19.9 pg/ml), IL-12 was 142 pg/ml (normal range <26.5 pg/ml), sTNFR was 5,000 pg/ml (normal range 1,470–4,160 pg/ml), sIL-2R was 7,436 pg/ml (normal range 676–2,132 pg/ml), MCP-1 was 1,926 pg/ml (normal range 220–722 pg/ml), and M-CSF was 2,280 pg/ml (normal range 257–1,715 pg/ml). There was no increase in the serum levels of TNFα, IL-10, TGFβ, IFNγ, IL-4, and IL-2.
Levels of the proinflammatory cytokines IL-1β and IL-6 were measured sequentially once every week before and during anticytokine therapy (Table 1). IL-1β levels increased and IL-6 levels decreased during anakinra treatment. However, the levels of IL-6 progressively increased after the discontinuation of anakinra, consistent with the dependence of serum IL-6 levels on IL-1β receptor activation. The levels of IL-6 again decreased after treatment with etanercept, as did the levels of IL-1β. With MRA treatment, the levels of IL-1β remained low, but IL-6 levels significantly increased, apparently in response to the drug.
Activation of NF-κB in PBMCs.
The expression of NF-κB in the PBMCs of the patient was measured by flow cytometry, as previously reported (8), by labeling PBMCs with mouse anti–NF-κB antibodies (nuclear-localized signal) (Chemicon, Temecula, CA), which recognize an epitope overlapping the nuclear location signal of NF-κB-p65. Prior to the administration of any biologic agents, the activation of NF-κB in the PBMCs was observed in CD4+ T cells and CD8+ T cells but not in CD14+ monocytes/macrophages (Table 1). Anakinra administration did not influence the NF-κB activation in CD4+ T cells or CD8+ T cells; however, it significantly increased the NF-κB activation in CD14+ monocytes/macrophages. With etanercept treatment, the NF-κB activation in CD4+ T cells and CD8+ T cells decreased as compared with that in cells at pretreatment. The NF-κB activation in CD14+ monocytes/macrophages decreased to low levels after treatment with etanercept, consistent with the levels at baseline. With MRA treatment, the NF-κB activation remained high in CD4+ T cells and CD8+ T cells and remained low in CD14+ monocytes/macrophages.
We have described a case of severe CINCA syndrome with a novel single-nucleotide missense mutation in exon 3 of the CIAS1 gene (5, 6). Over 40 mutations in CIAS1 have been reported in patients with CINCA syndrome, Muckle-Wells syndrome, and familial cold autoinflammatory syndrome. The clinical presentations of these disorders vary in severity and specific symptoms, but all are characterized by neonatal rash, fever, and joint symptoms. Elevated serum cytokine levels and increased cytokine release from cultured peripheral blood cells have been described in each of these disorders (4, 9), which is consistent with the proposed role of cryopyrin in the regulation of cytokines at the transcriptional and posttranslational level. Aksentijevich et al (4) reported evidence of increased serum levels of IL-1β as well as of TNFα, IL-3, IL-5, and IL-6 in a CIAS1 mutation–positive patient. We also found increased IL-1β and IL-6 levels in the sera of our patient.
We are the first to report increased NF-κB activation in peripheral blood cells from a patient with a cryopyrin-associated disease. NF-κB activation was noted in peripheral blood CD4+ T cells and CD8+ T cells at baseline, prior to anticytokine therapy. Surprisingly, significant NF-κB activation was not observed in the peripheral blood CD14+ monocytes/macrophages, which are the primary producers of IL-1β and IL-6. Possible explanations for this phenomenon are that 1) peripheral blood monocytes may not be the main source of cytokine production, 2) activated monocytes may be withdrawn from the peripheral circulation, or 3) massive activation of monocytes may result in apoptosis.
Because our patient had a severe case of CINCA syndrome and did not respond to any previous immunosuppressive therapy, we attempted treatment with biologic agents as anticytokine therapy. Some of these agents have recently been used to treat pediatric patients with rheumatic diseases (10). There have been case reports in which anakinra (11,12) and etanercept (13) were found to be effective in the treatment of patients with CINCA syndrome. MRA treatment has been shown to significantly reduce disease activity in adult rheumatoid arthritis (RA) (14–16), Castleman's disease (17), and juvenile idiopathic arthritis (18), but has not been previously used in CINCA syndrome. IL-6 is elevated in the serum of patients with each of the cryopyrin-associated disorders, and therefore MRA was a reasonable therapeutic choice for our patient.
Anakinra was effective in reducing the rash temporarily and mildly improving the laboratory evidence of inflammation. Etanercept was effective in slightly improving tibial lesions, but was not effective in reducing the fever or systemic inflammation measured by laboratory markers. MRA was most effective in improving both the clinical symptoms and the laboratory findings. However, the patient still had moderate fever, rash, and arthralgia. The biologic agents had different effects on the serum cytokine levels. Serum IL-1β levels fluctuated and tended to be increased after anakinra treatment, consistent with its mechanism of action as a competitive receptor antagonist. Etanercept and MRA resulted in low serum IL-1β levels, reflecting improvement of clinical symptoms and reduction of inflammation. The serum levels of IL-6 increased during MRA treatment, possibly because of a compensatory increase in IL-6 production, as previously suggested in a study of RA patients (14). However, it was difficult to interpret the effect of MRA treatment based on the results of NF-κB expression in PBMCs or changes in the serum cytokine levels.
Unfortunately, the patient died suddenly, in the absence of obvious infection. There were no cardiac abnormalities detected by ECG and chest radiography 2 months prior to the development of CHF. The apparent cause of death was CHF with interstitial pneumonia. The limited histologic investigations showed interstitial pneumonia and damage to myocardial cells without apoptosis. We speculate that there were 3 possible mechanisms of the CHF in this patient. 1) The cardiac muscle change may have been a result of the underlying chronic inflammation, although this has not been reported previously in CINCA syndrome. 2) The long duration of interstitial pneumonia of unknown etiology may have resulted in CHF. 3) Finally, the use of biologic agents may have contributed to CHF. The issues surrounding the use of TNF blockers and CHF were reviewed at a US Food and Drug Administration Arthritis Advisory meeting (19). It was reported that TNF blockers might induce new-onset CHF or exacerbate existing disease (20). However, other conflicting reports indicate that the risk of CHF may not be significant in most RA patients treated with TNF blockers (19). There was 1 patient with preexisting ischemic heart disease who died from myocardial ischemia in a clinical trial of MRA for RA (15). There are no published reports of CHF associated with anakinra or MRA treatment.
While anakinra has been reported, in multiple case reports (11,12), to be extremely effective in patients with CINCA syndrome, the treatment elicited only a partial and temporary improvement response in our patient. Etanercept has been shown to have clinical efficacy only rarely in CINCA syndrome but was used because of the incomplete response to anakinra (13). This is the first report of the use of MRA in a patient with CINCA syndrome, and the clinical and laboratory responses were promising; however, 2 months of MRA therapy in a single patient is not adequate to truly assess its efficacy. The etiology of interstitial lung disease and CHF in our patient remains unclear. Despite remarkable improvement in the therapy for CINCA syndrome in the last few years, particularly with biologic agents, these patients continue to be a challenge for the clinician and require careful monitoring of clinical signs and laboratory data.
The authors thank Yoshiko Ueno (Yamaguchi University) for her help with the experiments and James Mueller (University of California at San Diego School of Medicine) for his help in DNA sequencing.