Mevalonate Kinase Deficiency: A Survey of 50 Patients
OBJECTIVE: The goal of this study was to describe the spectrum of clinical signs of mevalonate kinase deficiency (MKD).
METHODS: This was a retrospective French and Belgian study of patients identified on the basis of MKD gene mutations.
RESULTS: Fifty patients from 38 different families were identified, including 1 asymptomatic patient. Symptoms began during the first 6 months of life in 30 patients (60%) and before the age of 5 years in 46 patients (92%). Symptoms consisted of febrile diarrhea and/or rash in 23 of 35 patients (66%). Febrile attacks were mostly associated with lymphadenopathy (71%), diarrhea (69%), joint pain (67%), skin lesions (67%), abdominal pain (63%), and splenomegaly (63%). In addition to febrile attacks, 27 patients presented with inflammatory bowel disease, erosive polyarthritis, Sjögren syndrome, and other chronic neurologic, renal, pulmonary, endocrine, cutaneous, hematologic, or ocular symptoms. Recurrent and/or severe infections were observed in 13 patients, hypogammaglobulinemia in 3 patients, and renal angiomyolipoma in 3 patients. Twenty-nine genomic mutations were identified; the p.Val377Ile mutation was the most frequently found (29 of 38 families). Three patients died of causes related to MKD. The disease remained highly active in 17 of the 31 surviving symptomatic patients followed up for >5 years, whereas disease activity decreased over time in the other 14 patients. Interleukin 1 antagonists were the most effective biological agents tested, leading to complete or partial remission in 9 of 11 patients.
CONCLUSION: MKD is not only an autoinflammatory syndrome but also a multisystemic inflammatory disorder, a possible immunodeficiency disorder, and a condition that predisposes patients to the development of renal angiomyolipoma.
WHAT'S KNOWN ON THIS SUBJECT:
The previously described clinical spectrum of mevalonate kinase deficiency ranges from recurring febrile attacks (hyperimmunoglobulin D syndrome) to a more severe form (mevalonic aciduria), which also includes psychomotor retardation, facial dysmorphia, cataracts, and failure to thrive.
WHAT THIS STUDY ADDS:
This series of 50 patients with mevalonate kinase deficiency widens the known clinical spectrum of this disorder, revealing it to be not only an autoinflammatory syndrome but also a multisystemic inflammatory disorder, a possible immunodeficiency disorder, and a condition that may predispose patients to the development of renal angiomyolipoma.
Mevalonate kinase deficiency (MKD) is a rare autosomal recessive autoinflammatory syndrome caused by mutations in the gene encoding mevalonate kinase (MK), an enzyme involved in the biosynthesis of cholesterol and isoprenoids (MVK) (Online Mendelian Inheritance in Man No. 251170). It results in a continuous spectrum of clinical signs, ranging from recurrent febrile attacks, known as hyperimmunoglobulin D syndrome (HIDS) (Online Mendelian Inheritance in Man No. 260920), to a more severe form, known as mevalonic aciduria (MA) (Online Mendelian Inheritance in Man No. 610377), with psychomotor retardation, facial dysmorphia, cataracts, and failure to thrive (in addition to the febrile episodes).1,2 The MA and HIDS phenotypes cannot be distinguished on the basis of biochemical or genetic features, and it has been suggested that the term “mevalonate kinase deficiency” should replace HIDS as the name of this metabolic autoinflammatory syndrome.3 The main clinical characteristics of the febrile attacks were reported in a single large series focusing on patients with HIDS, excluding patients with the most severe phenotype of MKD.4 We retrospectively reviewed the charts of 50 French patients with documented MVK mutations, regardless of their initial phenotype, to describe the spectrum of clinical and genetic manifestations of MKD.
The inclusion criteria for this retrospective study were: (1) the presence of at least 1 mutation in the MVK gene in the proband or the proband's siblings, associated with recurrent attacks of fever; or (2) the presence of 2 mutations in the MVK gene and any phenotype. All patients followed up at French and Belgian pediatric centers (Société Francophone pour la Rhumatologie et les Maladies Inflammatoires en Pédiatrie) and at French adult rheumatology or internal medicine centers (Club Rhumatismes et Inflammations) in whom a mutation of the MVK gene had been demonstrated between January 1999 and December 2009 were eligible for inclusion. The referring physicians were invited, by mail, to participate in this survey by completing a questionnaire sent online with the support of the Société Francophone pour la Rhumatologie et les Maladies Inflammatoires en Pédiatrie (www.sofremip.org) and Club Rhumatismes et Inflammations (http://cri-net.com). We recorded demographic data (age at onset and diagnosis, gender, origin of both parents, and family history), clinical manifestations (type and length of febrile episodes, organ involvement, and infections), laboratory evaluation results, and treatment efficacy and tolerance. Laboratory tests included determinations of inflammatory markers (blood cell counts, C-reactive protein concentration, erythrocyte sedimentation rate), serum immunoglobulin (Ig) G, A, M, and D concentrations, liver, and kidney function, MA, lymphocyte and/or fibroblast MK activity (with respect to control values), and mutational analysis. The principal investigators (Drs Bader-Meunier and Florkin) also reviewed detailed medical histories that had been rendered anonymous, to take into account details not included in the database when possible. The physician's global subjective assessment of disease activity during the year preceding the last visit was used to assess disease activity (0 = inactive disease, 1 = mild, and 2 = severe) in patients who had not received anti–interleukin-1 (IL-1) or anti–tumor necrosis factor agents for >6 months. In patients who had been treated with these agents during the preceding 6 months, disease activity before treatment initiation was considered. Complete remission was defined as the absence of all clinical signs and the normalization of C-reactive protein concentrations and erythrocyte sedimentation rate during follow-up. Partial remission was defined as a decrease in the mean yearly frequency of febrile attacks by at least 50%, in the absence of complete regression, and/or by the absence of persistent inflammatory disease during follow-up. Informed consent for genetic analysis was obtained from all patients or their parents. In accordance with French regulations, no other ethical committee approval was required for this retrospective study. The cutoff date for follow-up was June 2010.
Extraction of Genomic DNA, Analysis of MVK Gene Mutations, and Measurement of MK Activity
Genomic DNA was extracted from whole blood or Epstein-Barr virus–immortalized lymphoblasts from the patients, and all the coding exons and intron/exon junctions of the MVK gene were amplified and sequenced, as previously described by 2 of us (Drs Cuisset and Touitou).5 MK enzyme activity was determined in peripheral blood lymphocytes or fibroblasts, with a radiometric assay based on quantification of the phosphorylated products of 14C-mevalonic acid, as previously described.6
We used nonparametric statistics (Mann-Whitney U test) and Fisher's exact test to compare patients with severe and nonsevere disease courses.
Fifty patients (29 female patients, 21 male patients) from 38 families, 3 of whom were consanguineous (families 1, 4, and 21), were included from 17 French and 2 Belgian pediatric (n = 17), rheumatology (n = 1), and internal medicine (n = 1) centers. Five of these patients were previously described in a study of an international cohort in 2008 or as case reports (patients 1, 8, 13, 26, and 32).7,–,10 The demographic and clinical features of the patients are shown in Table 1. The cohort included 49 symptomatic patients and 1 asymptomatic 40-year-old patient (patient 49). Febrile episodes were reported for 48 patients, and age at onset ranged from 1 day to 20 years (median age: 4 months) in the 48 symptomatic patients for whom data were available. The first attack occurred within the first 6 months of life in 30 patients (60%) and before the age of 5 years in all but 3 patients.
Clinical Features During Febrile Attacks
Febrile episodes were reported in 48 of the 50 patients and lasted from 1 to 10 days (mean: 3.7 days). Thirty-eight patients experienced at least 1 febrile attack per month at the onset of MKD. Triggering factors (infections and/or immunization) were documented in 20 of the 48 patients. Onset and cumulative features are presented in Fig 1. Initial nonfebrile signs were documented in 35 patients and consisted of rash and/or diarrhea in 23 of these 35 patients (66%). Only 2 patients presented with isolated fever as the first symptom. During the course of the disease, at least 2 of the following features were found in addition to fever in the 48 patients with febrile attacks: lymphadenopathy, abdominal pain, vomiting, diarrhea, skin lesions, or musculoskeletal symptoms. Macrophage activation syndrome was observed in 6 patients.
The laboratory test results obtained during attacks are shown in Table 2. All 33 patients tested during febrile attacks had urinary mevalonic acid concentrations above the upper limit of normal range. Serum IgD concentrations were normal in 12% of the patients tested (4 of 34), including 2 patients with severe multisystemic disorder who died of MKD (patients 24 and 25). Serum IgA concentration exceeded 3 g/L in 17 of the 30 patients tested (57%). Twenty-nine genomic mutations were identified in the homozygous (8 families), heterozygous (4 families), or compound heterozygous (26 families) state in the proband (Table 1). Nineteen of these mutations had not been described before: 14 missense mutations, 2 nonsense mutations, 1 heterozygous point mutation, and 2 deletions of a few nucleotides. The p.Val377Ile mutation was found in 29 of 38 families (76%), in the compound heterozygous (21 families), heterozygous (4 families), and homozygous (4 families) states. The 6 patients homozygous for p.Val377Ile did not have a uniform phenotype.
Three patients died, between the ages of 2 and 3 years, of staphylococcal sepsis associated with macrophage activation syndrome (patient 25) or multiorgan failure (patients 19 and 24). Four patients were lost to follow-up. Mean disease duration, from initial onset to most recent assessment, was 24 years (range: 1–55 years) in the remaining 42 symptomatic patients.
Thirty-three patients developed associated inflammatory and autoimmune features, infections, or tumors, in addition to the febrile attacks (Table 1). Twenty-seven patients had chronic neurologic, abdominal, renal, pulmonary, endocrine, cutaneous, ocular, or hematologic involvement; erosive polyarthritis; and/or Sjögren syndrome. Infections were observed in 13 of 49 (27%) children and adults, and consisted of otitis, sinusitis, and pneumonitis, including at least 3 severe pneumococcal infections in patients 8, 31, and 46. Low serum IgG concentrations, from 4 to 5.5 g/L, were observed in 3 patients (patients 11, 29, and 41) who had not been treated with steroids or immunosuppressive drugs. Patient 25, who eventually died of MKD, had 4 episodes of septicemia caused by Staphylococcus epidermidis, Salmonella species, and Citrobacter species. Three patients developed renal angiomyolipoma.
Fourteen of the 31 surviving symptomatic patients followed up for >5 years became asymptomatic or had only mildly active disease and displayed a significant decrease in the frequency of attacks (disease activity score of 0 or 1) in the absence of long-term treatment. The disease remained highly active in 17 patients, as evidenced by frequent febrile attacks or persistent organ involvement (disease activity score of 2). No significant difference was found between the group with persistently high levels of disease activity or death and the group in which disease activity decreased over time, in terms of mean age at disease onset (8.2 vs 12.2 months) or the percentage of patients carrying the p.Val377Ile mutation (70% vs 57%). One 40-year-old patient (patient 49) carrying the same homozygous substitution (V377I) as her sister was asymptomatic despite low levels of MK activity (Table 1).
Various treatments were administered for the treatment and/or prevention of attacks (Table 3). Nonsteroidal antiinflammatory drugs were the most frequently administered and had a partial and transient effect. Anti–IL-1 and anti–tumor necrosis factor agents were tested in 16 of the 49 symptomatic patients. Eleven patients received anakinra to treat either frequent febrile attacks or erosive polyarthritis (patient 22) over a period of 5 to 36 months. Anakinra resulted in complete or partial remission in 9 of the 11 patients, including 3 patients displaying no response to etanercept. Anakinra was discontinued in 2 patients because of severe neutropenia (patient 13) and an increase in the frequency of infection (patient 22). Canakinumab resulted in partial or complete remission in 3 patients, including 1 patient who did not respond to anakinra (patient 10) and another who developed neutropenia (patient 13). We previously reported a successful allogenic bone marrow transplant in a 3-year-old boy with severe HIDS (patient 1) from an HLA-identical sister.7 Renal transplant was required in another patient with previously reported crescentic glomerulonephritis (patient 32).8
Our findings for this large cohort of patients with MKD reveal that the clinical and genetic spectrum of the disease is wider than previously reported. These results show that MKD is a disorder with multiple organ involvement that may be associated with severe and/or recurrent bacterial infections, autoimmune cytopenia, macrophage activation syndrome, and renal angiomyolipoma. The p.Val377Ile mutation is associated with both asymptomatic and severe phenotypes.
Lymphadenopathy and gastrointestinal and joint symptoms were the more common manifestations during febrile attacks, as previously reported.4 Our study also confirms that IgD concentrations, which were in the normal range in 12% of the patients, is not very relevant for the diagnosis of HIDS,3 whereas the level of MA during febrile attacks is very sensitive for screening. We also found that macrophage activation syndrome, which occurred in 6% of the patients, may complicate the febrile attacks in the absence of infectious or toxic agents, as occurred in 5 of 6 patients. Macrophage activation syndrome has previously been reported in only 1 patient with MKD.9 The clinical signs of MKD may be confused with those of this syndrome, which should be considered in MKD patients with persistent fever.
To the best of our knowledge, this is the first study showing that MKD may affect many organs and tissues, even in patients with low levels of disease activity. Our findings also expand the spectrum of neurologic, digestive, hepatic, renal, pulmonary, endocrine, and hematologic involvement. In addition, we report here the first case of MKD in an adult with no previous history of fever. The neurologic phenotype of the disease classically includes mild-to-severe mental retardation and cerebellar ataxia.10 Fatal ischemic stroke and cerebral hemorrhage of unknown origin have also been reported in adults with MKD.4,10 Other types of chronic involvement included cholestasis, thrombocytopenia, glomerular disease (crescentic glomerulonephritis and membranoproliferative glomerulonephritis), each reported in 1 or 2 patients.8,11,12 Conversely, the incidence of AA amyloidosis was lower than in other autoinflammatory syndromes,3,13,14 and this condition was not observed in our series.
The occurrence of renal angiomyolipoma in 3 patients with MKD is an intriguing feature for which we have no clear explanation. Renal angiomyolipoma is a large renal tumor consisting of fat, muscle, and blood vessels, with a frequency estimated at 0.02% to 0.29% in autopsy or imaging studies.15 It has previously been reported to be associated with only 2 conditions affecting other organ systems: tuberous sclerosis complex and sporadic lymphangioleiomyomatosis. Its occurrence in 6% of patients with MKD in this series raises the possibility of a nonfortuitous association.
The occurrence of recurrent and/or severe pulmonary or ear, nose, and throat diseases observed in one-quarter of these patients suggests that susceptibility to bacterial infections, including those caused by Streptococcus pneumoniae in particular, may be a feature of MKD. The reports of a death from pneumococcal sepsis in 1 patient from the cohort of van der Hilst et al,4 recurrent pneumonia and fatal meningococcal sepsis in 2 of 15 Italian patients with MKD,16 and neonatal death caused by recurrent septicemia in 2 siblings17 are consistent with a probable link between MKD and immunodeficiency. The mevalonate pathway is essential for the survival of S pneumoniae in the lungs and serum,18 and the increase in plasma mevalonate content due to MKD may favor the persistence of S pneumoniae in patients with MKD. In addition, hypogammaglobulinemia, which occurred in 3 of our patients and in 2 previously reported brothers with MKD,19 may be a component of the spectrum of the disease, potentially accounting for the susceptibility to pneumococcal infections. Immunization against S pneumoniae should therefore be recommended in patients with MKD.
In this series of 38 families, we identified 29 different mutations, 19 of which have not been reported before. The p.Val377Ile mutation, which was present in at least 1 allele in 74% of the families, was the most common genetic defect responsible for MKD deficiency. It was found in white patients, as previously reported,4 but also in patients of Arab descent (both parents). Most of the patients were compound heterozygous or homozygous, but a few patients carried the mutation in the heterozygous state. We revealed in a previous study,20 and in 1 heterozygous patient from this series (not shown), that these patients, who had very low levels of MK activity, do not produce messenger RNA from the unmutated allele; this finding suggests that this allele was either repressed or gave rise to an unstable messenger RNA. It was initially proposed that patients carrying a homozygous p.Val377Ile mutation would present a milder phenotype or would be asymptomatic,21 but symptomatic homozygous carriers were subsequently found in an Italian cohort.16 In our French cohort, the 6 patients homozygous for the p.Val377Ile mutation included asymptomatic patients and patients with a mild or severe phenotype. No relationship between genotype and phenotype severity was found in 2 series restricted to less severe phenotypes, including no neurologic impairment.4,5
Biological agents were required in more than one-third of the patients. We and others have previously highlighted the possible efficacy of an IL-1 receptor antagonist (anakinra)4,9,22 and a tumor necrosis factor inhibitor (etanercept)4,23,24 for the treatment and prevention of attacks in some patients with MKD. Anti–IL-1 agents, including canakinumab (a fully human anti–IL-1β monoclonal antibody) and anakinra, were possibly the most efficient treatments at preventing febrile attacks in our small retrospective study, and these drugs also resulted in the remission of erosive polyarthritis. However, the lack of a control group of patients not given anti–IL-1 agents made it impossible to draw definitive conclusions, and anakinra treatment had to be stopped in 2 of 11 patients because of the development of neutropenia and recurrent infections. Furthermore, MVK is an enzyme of the geranylgeranyl group pathway,1 the impairment of which leads to the activation of caspase 1, the enzyme responsible for processing pro–IL-1β into its active and secreted form.25,26 Thus, the increase in IL-1β production in patients with MKD constitutes a possible link between MKD and inflammation,27 a hypothesis supported by the efficacy of anti–IL-1 agents against MKD-related symptoms.
Our study is subject to several limitations. First, some data were missing, particularly for initial clinical signs, because of the retrospective design of the study. However, the study design included a review of detailed medical histories in addition to chart reviews. These steps made it possible to review the medical history of most patients in detail. The initial signs of the disease were documented in >30 patients, providing valuable tools for a diagnostic approach. Second, we cannot exclude the possibility of some selection bias: some mild cases were missed that were eventually undiagnosed or misdiagnosed and not referred to a rheumatology or pediatric center. Third, the frequency of some subclinical signs and of hypogammaglobulinemia may have been underestimated because these aspects were not investigated in some patients.
This study enlarges the clinical spectrum of MKD by showing that this condition is not only an autoinflammatory syndrome but also a multisystemic inflammatory disorder, a possible immunodeficiency disorder, and a condition predisposing patients to the development of renal angiomyolipoma.
We thank the following clinicians, for referring patients: Achille Aouba (Paris, France), Marc Brouillard (Arras, France), Daniele Bruno (Marseille, France), Brigitte Chabrol (Marseille, France), Alain Fouilhoux (Lyon, France), Marie-Hélène Girard-Madoux (Lyon, France), Nathalie Guffon (Lyon, France), Vincent Guigonis (Limoges, France), Olivier Hermine (Paris, France), Claudie Kerouedan (Lisieux, France), Laurence Lepage (Amiens, France), Guy Leverger (Paris, France), Karin Mazodier (Marseille, France), Michel Moutschen (Liège, Belgium), Martine Munzer (Reims, France), Béatrice Pellegrino (Paris, France), Pascal Pillet (Bordeaux, France), Michel Tsimaratos (Marseille, France), and Julie Zyss (Paris, France).
We also thank Séverine Guillaume (Paris, France) and Valentine Brousse (Paris, France), who were involved in the initial collection of data for patients, and Catherine Villeneuve (Paris, France), for assistance in the preparation of the manuscript.
- Accepted March 16, 2011.
- Address correspondence to Brigitte Bader-Meunier, MD, Department of Pediatric Immunology and Rheumatology, Necker Hospital, 75743 Paris Cedex 15, France. E-mail:
Drs Bader-Meunier and Florkin had full access to all the study data and take responsibility for the integrity of the data and the accuracy of data analysis. Drs Florkin, Prieur, Quartier, and Bader-Meunier assisted with the study conception and design; Drs Florkin, Bader-Meunier, Sibilia, Hachulla, Grateau, Richer, Farber, Fischbach, Hentgen, Jego, Laroche, Neven, Lequerré, Mathian, and Pellier contributed to the acquisition of data; Drs Cuisset and Touitou provided the genetic analysis; Dr Acquaviva conducted the measurement of MK activity; Drs Acquaviva and Rabier conducted the MA measurements; and Drs Bader-Meunier and Florkin analyzed and interpreted the data. All of the authors were involved in drafting the article or revising it critically for its intellectual content, and all of the authors approved the final version for publication.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
- MKD —
- mevalonate kinase deficiency
- MK —
- mevalonate kinase
- HIDS —
- hyperimmunoglobulin D syndrome
- MA —
- mevalonic aciduria
- Ig —
- IL —
- Ammouri W,
- Cuisset L,
- Rouaghe S,
- et al
- Simon A,
- Kremer HP,
- Wevers RA,
- et al
- Nevyjel M,
- Pontillo A,
- Calligaris L,
- et al
- Raupp P,
- Varady E,
- Duran M,
- et al
- Wilding EI,
- Brown JR,
- Bryant AP,
- et al
- Lequerré T,
- Vittecoq O,
- Pouplin S,
- et al
- Kuijk LM,
- Beekman JM,
- Koster J,
- Waterham HR,
- Frenkel J,
- Coffer PJ
- Copyright © 2011 by the American Academy of Pediatrics