Follicular Lymphoma (FL) is the second most common non-Hodgkin lymphoma and remains an incurable disease for the majority of patients who present with advanced stage disease. Virtually all patients subsequently acquire treatment resistance and a third of patients ultimately develop histologically transformed disease with aggressive clinical course and poor prognosis. Thus, there is an unmet clinical need to decipher the underlying biology of this disease, and identify dysregulated pathways that may serve as therapeutic targets.

In a cohort of 258 FL specimens, we identified 33 cases with STAT6 mutations (12.8%), including 15 mutations at D419. Overall, 34 out of 35 (97.1%) of STAT6 mutations were within the DNA-binding domain. MutSigCV analysis (Lawrence, Nature 2013) indicated that STAT6 was significantly mutated in FL, i.e., more commonly mutated than expected based on background mutation rate.

Previously, STAT6 mutations, including D419 hotspot mutations, have been reported in other lymphomas, including primary mediastinal B-cell lymphoma (Ritz et al., Blood 2009) and FL (Yildiz et al., Blood 2015). STAT6 proteins are transcription factors that are mainly activated by interleukin-4 (IL-4). A recent study identified a subset of IL-4 producing follicular helper T cells to be involved in the survival of FL B cells (Ame-Thomas et al., Blood 2015).

We compared the expression levels of selected STAT6 target genes in FL patient samples with or without STAT6 mutations. Gene expression data (nCounter, Nanostring) was available for 138 patient samples with known STAT6 mutation status. Thereof, 13 cases had STAT6 mutations, including six at D419. Among the STAT6 target genes were CD23 (and its soluble form, sCD23) and SOCS1. Both showed significantly higher expression (P<0.0001 and P=0.0016, respectively) in STAT6 mutated lymphomas. CD23 had previously been reported to be involved in B-cell growth and survival and was thus selected for further studies.

CD23 immunohistochemistry was performed on 13 patient samples with known STAT6 mutation status. Five patient samples with wild-type STAT6 stained negative for CD23. In contrast, four out of eight cases with STAT6 mutations stained positive for CD23.

To test if STAT6 mutations are directly linked to enhanced CD23 expression, we stably expressed wild-type (wt) STAT6, mutant STAT6 (D419G, N421K, and D519V), or empty vector in two t(14;18) positive B-lymphoma cell lines (OCI-Ly1, OCI-Ly8). Incubation of OCI-Ly1 cell lines with IL-4 (10 ng/mL for 24h) resulted in enhanced CD23 surface expression by flow cytometry (3.2-fold for D419G, 3.1-fold for N421K, and 2.3-fold for D519V), compared to STAT6 wt (N=3), whereas no significant difference was seen in the absence of IL-4. Similar results were observed in OCI-Ly8 cells.

To investigate if these findings result from enhanced transcription, we performed quantitative PCR (SybrGreen). Briefly, Ct values were normalized to GAPDH reference gene and Ct values from untreated controls were subtracted from IL-4 treated samples (ΔΔCt method). These experiments showed increased CD23 transcriptional levels for OCI-Ly1 expressing mutant STAT6 (5.6-fold for D419G, 5.0-fold for N421K, and 8.4-fold for D519V), compared to STAT6 wt (N=3). For OCI-Ly8 cells, CD23 transcription was increased 4.6-fold for D419G, 5.2-fold for N421K and 3.0-fold for D519V, compared to STAT6 wt (N=3).

In addition to membrane-bound CD23, the soluble CD23 (sCD23) in the microenvironment might further contribute to FL biology. To address this, we established a sCD23 ELISA using OCI-Ly1 cells stably expressing wild type or mutant STAT6. Supernatants of IL-4 stimulated cells expressing mutant STAT6 resulted in a 4.1-fold (D419G), 1.5-fold (N421K), and 1.3-fold (D519V) increase of sCD23 levels, as compared to STAT6 wt.

We conclude that STAT6 is recurrently and significantly mutated in FL including hotspot mutations at D419. STAT6 mutations are gain-of-function and may promote disease progression by enhancing IL-4 induced expression of surface and soluble CD23.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution