The Landscape of Myeloid Neoplasms with Isochromosome 17q Discloses a Specific Mutation Profile and Is Characterized By an Accumulation of Prognostically Adverse Molecular Markers

Introduction: Isochromosome 17 (i(17q)) is a rare cytogenetic abnormality resulting in the loss of the short arm and the duplication of the long arm of chromosome 17. i(17q) has been reported in different myeloid neoplasms like acute myeloid leukemia (AML), chronic myeloid leukemia (CML), myelodysplastic syndromes (MDS), myeloproliferative neoplasms (MPN), MDS/MPN overlap, as well as in Hodgkin- and non-Hodgkin-lymphoma. i(17q) has been described both as primary and as secondary chromosomal aberration. In myeloid neoplasms i(17q) as sole abnormality is suggested to define a distinctive clinicopathological entity with a high risk to leukemic progression and poor prognosis. So far, however, it is only briefly mentioned in the WHO classification.

Aim: To comprehensively characterize the molecular features of patients with myeloid neoplasms and i(17q).

Patients and Methods: Patients were selected by the presence of i(17q) and diagnosis of a myeloid neoplasm. Philadelphia positive CML were excluded. The cohort comprised 62 cases, 47 males and 15 females with a median age of 69 years (range: 30 - 87 years). Classification of all cases was performed by cytomorphology on peripheral blood and/or bone marrow smears according to the WHO. Chromosome banding and FISH analysis were performed in all cases. 19/27 cases with sole i(17q) were additionally analyzed by array CGH. All 62 samples were analyzed by next generation sequencing using a 29-gene panel targeting ASXL1, BCOR, BRAF, CALR, CBL, CSF3R, DNMT3A, ETV6, EZH2, FLT3-TKD, GATA1, GATA2, IDH1, IDH2, JAK2, KIT, KRAS, MPL, NPM1, NRAS, PTPN11, RUNX1, SETBP1, SF3B1, SRSF2, TET2, TP53, U2AF1, and WT1. Variants of unknown significance were excluded from statistical analyses (n=14).

Results: Following WHO classification four of the 62 patients were diagnosed as MPN, 13 as MDS/MPN overlap, 24 as MDS, and 21 as AML. 27 cases showed i(17q) as sole abnormality, while 23 cases showed additional chromosome aberrations, and eight even a complex karyotype (>3 aberrations). Further four cases had two independent cell clones, with one harboring the sole i(17q) abnormality. Array CGH revealed that in 15/19 cases i(17q) was the only abnormality, while four patients showed additional aberrations (1-3 per patient): loss of 7p, 7q, 12p, gain of 13q, and CN-LOH 19p and 22q (n=2). The comprehensive mutational analyses revealed only 3/62 patients carrying no mutation, while a median of 3 mutations per patient was observed (range 0-6). The three most frequently mutated genes were ASXL1 (66%, 41/62), SRSF2 (65%, 40/62), and SETBP1 (48%, 30/62) with no association to any WHO entity, indicating the presence of this genetic profile also across entities beyond the expected overlap between different neoplasms. Following genes showed mutation frequencies >10%: TET2 (24%), ETV6 (16%), CBL (13%), TP53 (15%), RUNX1 (11%), and NRAS (10%), all genes known for adverse prognostic impact. Interestingly, mutations in the three most frequently mutated genes ASXL1, SRSF2, and SETBP1 often co-occurred (n=21) and ASXL1 and SRSF2 were rarely mutated alone (n=9; n=5), while SETBP1 was even never mutated solely, indicating acquirement of SETBP1 mutations during disease course. Therefore, SETBP1 mutations associated significantly with mutations in ASXL1 as well as SRSF2 (24/41 vs 6/21 in ASXL1 wild type (wt), p=0.033; 27/40 vs 3/22 in SRSF2 wt, p<0.001). Furthermore, mutations in ASXL1 associated significantly with i(17q) sole in comparison to cases with additional chromosomal aberrations (22/27 vs 19/35, p=0.032). Therefore, also cases harboring mutations in all three genes ASXL1, SRSF2, and SETBP1 associated with sole i(17q) (13/27 vs 8/35, p=0.058), indicating that these three mutations might be drivers of disease pathogenesis in this cytogenetic background. Reviewing the bone marrow morphology showed characteristic pseudo-Pelger-Huet anomaly in 36 of 59 (61%) analyzed smears. These changes were not associated with the cytogenetic profile, but showed a trend towards co-occurrence with ASXL1 mutations (27/39 vs 9/20 ASXL1 wt p=0.094).

Conclusion: 1) Myeloid neoplasms with i(17q) show a distinct molecular mutation pattern, accumulating prognostically adverse mutations. 2) Patients with sole i(17q) show co-occurring mutations in ASXL1, SRSF2, and SETBP1. 3) Frequency and co-existence of ASXL1, SRSF2, and SETBP1 mutations predispose these as driver mutations.