STAT3 Mutations in Large Granular Lymphocytic Leukemia

Abstract Abstract 1606 Introduction: Large granular lymphocytic leukemia (LGL leukemia) is a rare lymphoproliferative disorder characterized by the presence of increased numbers of LGL cells in the peripheral blood. Discrimination of LGL leukemia and reactive T-cell processes is difficult and a matter of controversial discussion. Very recently, somatic STAT3 mutations located in the Src homology 2 domain (SH2 domain) have been described with a high frequency of 40% in T-cell LGL (T-LGL) leukemia (Koskela et al., 2012) and 30% in chronic NK-cell lymphoproliferative disorder (CLPD-NK) (Jerez et al., Blood 2012) enabling the distinction of truly malignant lymphoproliferative diseases involving T- and NK- cells from reactive processes. Aim: Analysis of frequency and prognostic impact of STAT3 mutations in a cohort of 65 patients with LGL leukemia in comparison to 88 cases with other T-cell malignancies (70 non-LGL T-cell lymphoma and 18 with T-ALL) and 47 cases with non detectable lymphoma as a validation study for the usefulness of this new marker in diagnostics. Patients and Methods: Screening for STAT3 mutations was performed by direct Sanger sequencing of SH2 domain containing exon 21 in a total cohort of 65 patients with LGL leukemia (T-LGL leukemia: n=62; CLPD-NK: n=3). The detection limit of this analysis was 1:10 allowing the detection of 1 mutated in a background of 10 wildtype alleles. T-cell receptor (TCR) rearrangements were assessed by multiplex PCR with subsequent fragment analysis. Diagnosis of T-LGL leukemia was based on the following WHO criteria: a monoclonal T-cell receptor (TCR)-Rearrangement, the presence of an abnormal CTL population with expression of CD3, CD8 and CD57 detected by flow cytometry and LGL count by peripheral blood smear of >2×109/L (however, cases with LGL < 2×109/L but which met all other criteria were counted as consistent with the diagnosis T-LGL leukemia). Each patient must have met at least two of these criteria to be included into the study. Diagnosis of CLPD-NK was equally based on WHO criteria. The cohort was composed of 28 males and 37 females. Median age was 65.5 years (range: 38.2 to 89.7 years). For comparison, we analyzed 88 cases with other T-cell malignancies (70 with non-LGL T-cell lymphoma and 18 cases with T-ALL) and 47 cases with non detectable lymphoma. Results: Overall, in 39/62 T-LGL leukemia cases (62.9%) STAT3 mutations were detected. All mutations were point mutations. 7 different mutations were observed, with Tyr640Phe (n=16) and Asp661Tyr (n=12) accounting for 71.8% of all mutations detected. One patient harbored two mutations (Asn647Ile and Tyr640Phe). There was no association of STAT3 mutations with age, gender, leukocyte count, hemoglobin level and platelet count in T-LGL leukemia cases. In 50 T-LGL leukemia cases immunophenotyping data was available and cases with STAT3 mutations (n=31) had a higher expression of CD3 (75±20% vs. 50±37%, p<0.0015) and TCRαβ (68±18% vs. 44±19%, p<0.001) as well as lower expression of CD56 (7±8% vs. 25±26%, p=0.009) compared to STAT3 wildtype cases. Of note, all STAT3 mutated cases had a detectable TCR rearrangement at molecular level. Furthermore, of three analyzed CLPD-NK cases, two patients harbored STAT3 mutations (66.7%). Interestingly, one patient had an insertion of one aa (Tyr657dup), a mutation not observed in our T-LGL leukemias. However, due to the limited CLPD-NK cases available, the validity of this data is constricted. Of note, no STAT3 mutations were identified in the cohort of 135 cases diagnosed with non-LGL T-cell lymphoma, T-ALL or non detectable lymphoma. We were also interested whether STAT3 mutations were of any prognostic impact for T-LGL leukemia patients. However, due to limited follow-up information and the generally good prognosis of LGL leukemia, survival analysis was not informative yet. Conclusions: We were able to confirm the presence of STAT3 mutations in T-LGL as published by Koskela et al., 2012, with an even higher frequency of 62.9%. The discovery of STAT3 mutations in T-LGL leukemia has significant diagnostic value since it allows with high specificity to distinguish LGL leukemia from other T-cell lymphomas and reactive conditions. This can strongly support immunophenotyping and morphology in diagnosing T-LGL leukemia. Regarding CLPD-NK STAT3 mutations significantly improve diagnostic capabilities, which have been hampered by the scarcity of genetic markers in this entity. Disclosures: Fasan: MLL Munich Leukemia Laboratory: Employment. Kern:MLL Munich Leukemia Laboratory: Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Equity Ownership.