Germline genetic testing uncovers a high frequency of inborn error of immunity diagnoses in children with single and multi-lineage immune cytopenias in a large US cohort

Abstract Background Immune cytopenias including Immune Thrombocytopenia (ITP), Autoimmune Hemolytic Anemia (AIHA), or multilineage immune cytopenias/Evans syndrome (ES) can have variable clinical courses ranging from self-resolving to chronic or relapsing. They may occur in isolation or as the initial presentation of an underlying inborn error of immunity (IEI). Recent studies have found a high frequency of disease-causing variants in immune-related genes in children with ES. However, little is known about variants associated with single-lineage cytopenias. Aim To identify the prevalence of pathogenic (P) and likely pathogenic (LP) variants identified via clinical genetic testing in children with immune cytopenias. Methods This ITP Consortium of North America (ICON)-led retrospective study included children evaluated for immune cytopenias at centers across the United States who underwent clinical genetic testing through a commercial clinical laboratory (Invitae, now part of Labcorp). Included patients were age 0-21 years at time of genetic testing and had a diagnosis of ITP, AIHA, or ES as determined by clinician-provided ICD-10 code and/or recorded indication for genetic testing. Patients with isolated neutropenia were excluded from this study. Patients were included if their clinical genetic testing included any of the following targeted next generation sequencing (NGS) panels performed: Inborn Errors of Immunity and Cytopenias panel, Primary Immunodeficiency Panel, Autoimmune Lymphoproliferative Disorders (ALPS) Panel, Phagocytic Disorders Including Neutropenia Panel, Autoinflammatory and Autoimmunity Syndromes Panel, and/or Common Variable Immunodeficiency Panel. Variants identified as P or LP (P/LP), as determined by a semiquantitative, validated variant classification framework were evaluated in this study. Variants of uncertain significance were not included. Molecular IEI diagnosis was defined as presence of a single P/LP variant in monoallelic/autosomal dominant gene (or X-linked for individuals who are XY) or presence of two P/LP variants in a biallelic/autosomal recessive gene, depending on the gene and International Union of Immunological Societies (IUIS) classification. Results In total, there were 715 patients with immune cytopenias identified who had targeted NGS panels: 430 (60.1%) with ITP, 119 (16.6%) with AIHA, and 166 (23.2%) with ES. Median age at time of testing was 12 years (range 0-21) and 47.4% (n=339) were female. The frequency of patients with P/LP variants was 30.1% (215/715) in the total cohort, including 30.9% (133/430) in patients with ITP, 30.3% (36/119) in patients with AIHA, and 27.7% (46/166) in patients with ES. The most common variants identified were in genes related to diseases of immune dysregulation (37 patients), combined immunodeficiencies with associated syndromic features (33 patients), and immunodeficiencies affecting cellular and humoral immunity (29 patients). However, variants were spread across many IUIS categories at lower frequencies. Molecular IEI diagnoses were found in a subset of patients with P/LP variants. The frequency of molecular IEI diagnoses in the entire cohort was 7.7% (55/715), including 6.5% (28/430) in patients with ITP, 9.2% (11/119) in patients with AIHA, and 9.6% (16/166) in patients with ES. Median age of patients with IEI diagnoses was 13 years (range 2-20) and 30.9% (n=17) were female. The most common genetically identified conditions were related to dominantly inherited variants in TBX1 (n=8), CTLA4 (n=7), and STAT3 (n=4), associated with DiGeorge syndrome, CTLA4 haploinsufficiency, and STAT3 gain-of-function, respectively. Additionally, heterozygous TNFRSF13B variants were identified in 16 patients. Monoallelic variants in TNFRSF13B have been associated with increased risk of CVID with autoimmunity. Conclusions Genetically defined IEI were identified in approximately 7.7% of children with immune cytopenias who underwent genetic testing, and with similar frequency in children with single- and multi-lineage cytopenias. The high rate of genetic diagnoses in patients with single and multi-lineage immune cytopenias who were tested supports the clinical utility of genetic testing in these patients. Establishing these diagnoses informs selection of targeted therapies and potential need for stem cell transplant. Molecular diagnosis offers the opportunity for rational selection of targeted therapies rather than empiric broad immunosuppression.