Primary myelofibrosis (PMF) is a myeloproliferative neoplasm, characterized amongst others by stem-cell derived clonal myeloproliferation, bone marrow fibrosis, anemia, splenomegaly, constitutional symptoms and leukemic progression. Diagnosis is based in most cases on cytomorphology/histology demonstrating fibrosis as well as on mutations in JAK2 or MPL. The Dynamic Prognostic Scoring System (DIPSS)-plus is the current base for prognostication using different clinical parameters including karyotype. Furthermore, molecular genetic alterations are currently addressed to provide additional prognostic information. Recently, besides JAK2 and MPL further gene mutations have been described in a limited number of patients, including ASXL1 and SRSF2.


To analyze in a large cohort the frequency of SRSF2 and ASXL1 mutations in PMF patients, and to identify their prognostic impact in the context of other previously described gene mutations.

Patients and Methods

Diagnosis was done according to WHO classification. The cohort comprised 82 female and 131 male patients. In all cases a BCR-ABL rearrangement was excluded by RT-PCR or fluorescence in situ hybridization. JAK2V617F mutation was analyzed in all cases by melting curve analysis, MPLW515 mutation was subsequently analyzed in JAK2V617 wild type (wt) patients. In addition, we analyzed all patients for SRSF2 mutations by Sanger sequencing of the mutational hot spot region coding for amino acid Pro95. Cytogenetics was available in 139 patients. Patients were grouped in favorable (n=121) and unfavorable (n=18) karyotypes based on the DIPSS-plus-scoring system. Based on the previously described correlation of SRSF2mut with ASXL1mut and SETBP1mut in other myeloid entities, SRSF2 mutated cases were also analyzed for mutations in ASXL1 and SETBP1by Sanger sequencing. Follow-up data was available for 136 patients.


56% (120/213) of the patients showed JAK2V617F mutations and 18.0% (16/89) of JAK2wt patients carried a mutation in MPLW515 summing up to 65.1% of patients with an already established molecular marker. Of note, SRSF2 was mutated in 12.7% (27/213) of all PMF patients. Patients with SRSF2 mutation had higher white blood cell counts in comparison to SRSF2wt patients (20.00x109/L vs. 7.35x109/L; p=0.005), but there was no correlation to gender, age, hemoglobin level, platelet count or % of myeloblasts in the peripheral blood. In 17 SRSF2mut cases the karyotype was available, 12 were normal karyotype, while two cases showed an unfavorable karyotype according to DIPSS-plus with +8 and i(17)(q10), respectively. The remaining three aberrations belong to the favorable aberration group (del(20q), del(13q), and der(14)). There was no correlation of SRSF2 mutations to the cytogenetic subgroups normal karyotype (n=91) or DIPSS categories favorable and unfavorable aberrations. SRSF2 mutations were also equally distributed between both JAK2V617 or MPLW515 mutated and wild type cases. 18/27 SRSF2mut cases carried also either a JAK2 or MPL mutation, while 9 cases showed no additional JAK2 or MPL mutation. Therefore 30.6% patients remain that carry no mutation in at least one of the three genes investigated first. Interestingly, ASXL1 was frequently mutated in SRSF2 mutated patients (16/23 analyzed SRSF2mut patients) while none of the 24 analyzed SRSF2 mutated cases showed a mutation in SETBP1. To evaluate a potential influence of gene mutations on clinical outcome the overall survival (OS) was calculated. We could confirm that JAK2V617F had no prognostic impact. The same was true for MPLW515 mutations. In contrast to other studies we could not find any impact of SRSF2mutations on OS. Only cytogenetics, i.e. the normal karyotype showed a trend to a prognostic relevance: the median 3 year OS was 70.8% in patients with normal karyotype (n=56) but 58.8% in patients with cytogenetic aberrations (n=29; p=0.153).


1) SRSF2 is mutated in 13% of PMF patients. 2) SRSF2 mutated patients show frequently an additional ASXL1 mutation but no coincidence with SETBP1. 3) The prognostic relevance of cytogenetic aberrations was confirmed, while the molecular marker SRSF2 shows no impact on prognosis.


Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Alpermann:MLL Munich Leukemia Laboratory: Employment. Eder:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

Author notes


Asterisk with author names denotes non-ASH members.