Tumor necrosis factor receptor–associated periodic syndrome (TRAPS; MIM no. 142680), the most common autosomal-dominant recurrent fever, is caused by mutations in the gene encoding tumor necrosis factor receptor superfamily 1A (TNFRSF1A) on chromosome 12p13 (1). It is characterized by periodic fever, arthralgia, myalgia, abdominal pain, serositis, and skin rash. These symptoms are partially explained by defective shedding of TNFR and structural changes in the extracellular domain of the receptor leading to uncontrolled proinflammatory signaling (1–3). Here, we describe the first known case of rapidly progressive glomerulonephritis during an autoinflammatory episode in a patient with TRAPS.
The patient, a 9-year-old boy, was hospitalized in our unit for renal failure. He had a history of recurrent episodes of unexplained fever, abdominal pain, myalgia, and arthralgia since infancy. No prior investigations had been made in relation to these symptoms, nor had any treatment been administered. The boy is the first child of unrelated parents of Caucasian origin, and it is of note that his mother had had similar symptoms until the age of 15 years.
Upon admission, the patient had severe growth retardation (−3 SD for height), daily fevers ranging from 39°C to 40°C, gross hematuria, paleness, migratory skin rash, and diffuse myalgia. Laboratory tests revealed hematuria, nephrotic-range proteinuria (4 gm/day), severe renal failure (blood urea nitrogen concentration 49 mmole/liter, serum creatinine level 760 μmole/liter), a hemoglobin concentration of 50 gm/liter, and inflammatory syndrome (C-reactive protein level 250 mg/liter). A renal biopsy was performed (Figure 1). No deposits of amyloidosis were found. Light microscopy (16 glomeruli) showed diffuse endocapillary and extracapillary proliferation, with the presence of crescents in 50% of glomeruli, and small-vessel vasculitis.
An immunofluorescence study showed mild glomerular deposits of C3 complement only, and no dense deposit was observed by electron microscopy (Figure 1). Neither antinuclear antibodies nor anti–glomerular basement membrane or antineutrophil cytoplasmic circulating antibodies were detected. Serum complement and serum immunoglobulin concentrations were normal. Results of bacterial cultures or viral serologic analysis remained negative. This prompted us to search for hereditary recurrent inflammatory diseases. No mutations were found in genes known to be associated with recurrent fevers (CIAS1, MEVK, MEFV). However, we identified a heterozygote C29S mutation in TNFRSF1A, confirming the diagnosis of TRAPS. C29S is a previously unreported mutation in the extracellular region of TNFRI. Two other mutations in this cystein-encoding sequence (C29F and C29Y), also resulting in TRAPS, have already been referenced (http://fmf.igh.cnrs.fr). The same mutation was detected in the mother and the younger sibling of the patient.
The patient was initially treated with 3 methylprednisolone pulses (1 gm/1.73 m2), followed by maintenance prednisolone (2 mg/kg/day for 1 month, followed by progressive tapering) in association with an angiotensin-converting enzyme inhibitor (enalapril, 10 mg/day). Extrarenal symptoms disappeared within a few days, and the serum creatinine concentration normalized within 1 month. Six months later, the boy was asymptomatic, the hemoglobin concentration was 120 gm/liter, and the serum creatinine level was 45 μmole/liter; however, pathologic microalbuminuria (100 mg/day) persisted.
More than 1 year after the episode of acute renal failure, the patient is still doing well and is currently being treated with enalapril and a TNFα blocker (etanercept, 0.4 mg/kg twice weekly), which was started upon the diagnosis of TRAPS, allowing successful steroid cessation and disappearance of microalbuminuria 9 months after the onset of the renal symptoms. Renal amyloidosis, which has been observed in 14% of patients with TRAPS (3), is the most important complication of this disease. To our knowledge, glomerulonephritis in TRAPS has not been reported so far. We hypothesize that there was a causal relationship between TRAPS and the crescentic glomerulonephritis in our patient, because it occurred during an acute episode, and no other cause of nephritis was found. Our observation is consistent with previous works that demonstrated an increased production of TNFα in human glomerulonephritis by activated infiltrating mononuclear cells and intrinsic renal cells and effectiveness of TNF blockade in experimental models of crescentic glomerulonephritis (4, 5). Moreover, Lamprecht et al reported the case of a patient with small-vessel vasculitis and TRAPS (6). Because impaired TNFα regulation has been implicated in the pathogenesis of systemic vasculitis (7) as well as in glomerular inflammation, we speculate that proliferative glomerulonephritis can be a feature of TRAPS. Our report suggests that clinicians should consider glomerulonephritis as part of the growing range of TRAPS manifestations.