Primary atopic disorders (PADs) are monogenic inborn errors of immunity marked by severe allergic disease. Heterozygous gain-of-function (GOF) variants in STAT6 have recently been recognized as a PAD associated with severe atopy, gastrointestinal involvement, and, in some cases, lymphoma. The phenotypic spectrum remains incompletely defined. As genetic testing expands, more patients with atopy are found to carry variants of uncertain significance (VUS) in STAT6, and existing prediction tools often fail to identify GOF effects.
In previous work, we assembled an international STAT6-GOF cohort and performed functional testing using transient overexpression in HEK cells, IL-4 stimulation, and quantification of STAT6 phosphorylation as readout of pathway activation 1. Only recently, we evaluated the impact of other possible but so far not reported missense substitutions—particularly within previously reported GOF “hotspot” regions—using general pathogenicity tools (e.g., Combined Annotation Dependent Depletion [CADD]) and structure-informed prediction (Dynamut2, Alpha missense).
Beyond published STAT6-GOF cases, we identified two VUS: p.Ala390Ser in a Japanese patient with clinical phenotype and p.Asp419Glu in Saudi individuals located within a recurrent GOF hotspot but without clinical manifestations. Multiple prediction tools—including structure-aware models—flagged both variants as likely pathogenic. However, functional testing for p.Ala390Ser and the absence of a phenotype for p.Asp419Glu argue against a GOF effect. Integrating both published and novel variants, we find that predictions tend to be stronger when substitutions alter amino acid residue charge and when variants appear in COSMIC (a catalog of somatic mutations found in tumor samples) but not in gnomAD (a database of germline variation from large-scale sequencing projects). It is important to keep in mind that not all somatic hotspot substitutions confer GOF, and in silico evidence alone is insufficient. Therefore, functional validation remains essential.
Our findings reinforce the notion that mechanistic validation remains essential for STAT6 monoallelic VUS, even within canonical GOF hotspots. Current computational tools have limited accuracy for predicting GOF apart from haploinsufficiency. We advocate a workflow that integrates targeted functional assays with clinical and genetic phenotyping to determine variant significance. Ongoing efforts within an international consortium to conduct an in-depth clinical evaluation aim to refine genotype–phenotype correlations and inform personalized management for patients with suspected STAT6-GOF.
