Targeted deep sequencing identifies mosaicism in patients with immune dysregulation

Background: Identifying genetic mechanisms of inborn errors of immunity (IEI) is important for diagnosis and treatment of patients, yet most patients with suspected IEI have negative genetic testing results. Genetic mosaicism is an emerging mechanism of IEI, but it is challenging to identify. Objective: We used a discovery-based approach to identify mosaic variants in genes relevant to immune dysregulation in patient and healthy cohorts. Methods: We developed custom panels for high-depth sequencing of genes known or hypothesized to cause dominant immune dysregulation. Samples from 452 patients with immune dysregulation (affected) and 154 currently healthy were sequenced using 71- or 101-gene targeted panels. Results: We identified mosaic variants in 9.5% of undiagnosed patients and 7.8% of healthy individuals. Using a strategy to predict pathogenicity of variants in IEI, 33% of variants identified in patients were predicted to be likely pathogenic or pathogenic, while no mosaic variants in healthy individuals were predicted to be pathogenic. Genes with mosaicism in >1 affected undiagnosed patients included FAS, STAT3, CARD11, CARD14, NRAS, TNFAIP3, NLRP3, and IKZF2. Four patients had variants in FAS with allele fractions <5% in blood but highly enriched in double-negative T cells, diagnostic for somatic FAS autoimmune lymphoproliferative syndrome. Conclusion: These findings establish the utility of a high-depth sequencing panel to identify mosaic variants and demonstrate that mosaicism in immune-relevant genes is present in healthy individuals.