Systematic genetic analysis of pediatric patients with autoinflammatory diseases

Abstract

Monogenic autoinflammatory diseases (AID) encompass a growing group of inborn errors of the innate immune system causing unprovoked or exaggerated systemic inflammation. Diagnosis of monogenic AID requires an accurate description of the patients' phenotype, and the identification of highly penetrant genetic variants in single genes is pivotal. We performed whole exome sequencing (WES) of 125 pediatric patients with suspected monogenic AID in a routine genetic diagnostic setting. Datasets were analyzed in a step-wise approach to identify the most feasible diagnostic strategy. First, we analyzed a virtual gene panel including 13 genes associated with known AID and, if no genetic diagnosis was established, we then analyzed a virtual panel including 542 genes published by the International Union of Immunological Societies associated including all known inborn error of immunity (IEI). Subsequently, WES data was analyzed without pre-filtering for known AID/IEI genes. Analyzing 13 genes yielded a definite diagnosis in 16.0% (n = 20). The diagnostic yield was increased by analyzing 542 genes to 20.8% (n = 26). Importantly, expanding the analysis to WES data did not increase the diagnostic yield in our cohort, neither in single WES analysis, nor in trio-WES analysis. The study highlights that the cost- and time-saving analysis of virtual gene panels is sufficient to rapidly confirm the differential diagnosis in pediatric patients with AID. WES data or trio-WES data analysis as a first-tier diagnostic analysis in patients with suspected monogenic AID is of limited benefit.

Keywords: FMF; autoinflammatory diseases; genetic diagnostics; inborn errors of immunity (IEI); whole exome sequencing (WES).

FIGURE 1

Genetic workflow displaying step-by-step diagnostic strategy. First, WES data from all patients were screened using virtual vPANEL_1 (13 genes); in case of non-confirmatory findings, virtual vPANEL_2 was analyzed (542 genes). In a third step, WES data (WES_3) were filtered to detect the most likely deleterious variants highly suspicious for AID. Additionally, HPO terms were used according to the patient’s phenotype. If applicable, trio WES were performed; PV = pathogenic, or LPV = likely pathogenic variant found through genetic diagnostics.

FIGURE 2

Final diagnosis of pediatric AID cohort after genetic testing. The pie chart is presenting the final diagnosis of 125 pediatric patients with AID categorized into the following subclasses: Genetic autoinflammation (AID), juvenile idiopathic arthritis (JIA), PFAPA, patients with transient symptoms, unclassified SAID, Mobus Behçet, other defined disease of the immune system (no genetic cause identified) and Inborn Errors of Immunity (IEI) (other than autoinflammation). AID was subdivided into FMF, VAIHS-syndrome, CAPS and CRIA syndrome. n = amount of patients.

FIGURE 3

Phenotypes in FMF. Radar chart visualizing the symptoms of biallelic and monoallelic carriers of pathogenic MEFV variants depending on zygosity. Turquoise line: amount of patients with monoallelic variants in %, orange line: amount of patients with biallelic variants in %.

FIGURE 4

Serum levels of inflammatory biomarkers measured in patients with pathogenic MEFV variants a) serum calprotectin levels of monoallelic and biallelic carriers of pathogenic MEFV variants; b) serum amyloid A levels of monoallelic and biallelic carriers of pathogenic MEFV variant. Each dot and each horizontal extreme line corresponds to a value in one patient (n = 9); cross symbolizes the mean value of all patients; red line drawn at reference serum level cut-off point; the horizontal line in the middle of a box plot displays the median.

FIGURE 5

Flowchart for diagnosis, treatment and genetic evaluation of patients with suspected AID. AID should be suspected if patients present with fever episodes and auto-inflammatory signs and symptoms. If CBC, CRP, ESR, S100A8/A9 (Serum calprotectin) and Serum-Amyloid A are increased the further approach depends on the age of the patient. For patients younger than 1 year or older than 6 years WES should be requested immediately. According to clinical guidelines patients between one and 6 years PFAPA syndrome should be ruled out in a first step. If inflammation markers in fever-free intervals are not elevated and cyclic neutropenia is excluded in patients with oral ulcers, PFAPA syndrome is most likely. A response to a steroid test confirms the diagnosis. A non-responsive steroid test indicates a PFAPA-mimicking monogenetic disease and WES should be performed. For all children identification of pathogenic variant(s) in tPANEL entails clinical validation and treatment according to the disease. For negative sequencing results an extended scientific analysis (e.g. Trio- WES, Genome, analysis of genes of uncertain significance) is recommended. Variants of unknown significance can be further evaluated by segregation analysis. Confirmation of variants as having arisen de novo may result in reclassification under ACMG guidelines. AID, autoinflammatory disease, CBC, complete blood count, CRP, C-reactive protein, ESR, erythrocyte sedimentation rate, WES, whole exome sequencing, PFAPA, Periodic Fever, Aphthous Stomatitis, Pharyngitis, Adenitis, pV, pathogenic variant, VUS, Variant of unknown significance.