Abstract T046

Functional characterization of STAT6 mutations as oncogenic drivers in classical Hodgkin lymphoma (cHL)

Background: cHL is largely curable but a subset of patients still dies of this lymphoma, underscoring the need to better define its drivers for enabling targeted therapies. Mutations in the DNA-binding domain of the STAT6 transcription factor occur in 30% of cHL cases (Tiacci et al. Blood 2018), yet how they contribute to cHL pathogenesis is unclear.

Methods: We studied the most frequent STAT6 mutations (N417Y, D419G/H/N, and double mutants, Fig. A) in cHL cell lines expressing endogenous or exogenous wild-type (WT) or mutant STAT6 through an integrated multi-omics approach combining chromatin binding (ChIP-seq), chromatin accessibility and regulatory profiling (ATAC-seq; Cut&Run for H3K27ac), 3D genome organization (Hi-C-seq), and transcriptomic analyses (RNA-seq).

Results: Compared to WT, each STAT6 mutant showed stronger and expanded genomic occupancy, at least in part through neomorphic binding to non-canonical (“alternative”) STAT6 DNA motifs (Fig. B), leading to increased chromatin accessibility, activation of regulatory elements including super-enhancers (SEs), and increased long-range chromatin interactions at multiple loci. These alterations established aberrant transcriptional programs (Fig. C) promoting proliferation, metabolic adaptation, immune evasion (including upregulation of PD-L1/PD-L2) and lineage plasticity, while repressing interferon signaling and tumor suppressors.

Notably, mutant STAT6 newly generated a SE overlapping the locus of the myeloid receptor tyrosine kinase CSF1R, leading to expression of a novel truncated isoform lacking the ligand-binding domain but retaining the kinase domain, consistent with potential constitutive activity. Indeed, viability of STAT6-mutant cHL cells was impaired by the clinical-grade CSF1R-specific inhibitor pimicotinib (Fig. D). Consistently, knock-out of endogenous mutant STAT6 induced cHL cell apoptosis, which was rescued only by exogenous mutant (but not WT) STAT6 (Fig. E), indicating oncogenic addiction to mutant-specific programs. Finally, a young patient with STAT6-mutant cHL not responding to 9 prior therapies (including nivolumab) achieved a rapid and marked clinical response to treatment with the JAK/STAT inhibitor ruxolitinib combined with nivolumab.

Conclusions: We functionally dissected the role of STAT6 mutations in cHL pathogenesis and provided a mechanistic rationale for therapeutic strategies targeting the JAK/STAT axis and key downstream effectors like CSF1R.